Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease: A Focus on Early Phase Clinical Drug Development [2nd ed.] 978-3-030-11747-4;978-3-030-11748-1

Table of contents :
Front Matter ....Pages i-xi
Front Matter ....Pages 1-1
Quantification of Insulin Action in Human Subjects (Andrew J. Krentz, Christian Weyer, Marcus Hompesch)....Pages 3-35
Assessment of Islet Alpha- and Beta-Cell Function (Sten Madsbad, Jens J. Holst)....Pages 37-74
Pharmacokinetic and Pharmacodynamic Assessment of Novel and Biosimilar Insulins (Andrew J. Krentz, Christian Weyer, Marcus Hompesch)....Pages 75-100
Measurement of Energy Expenditure (Klaas R. Westerterp)....Pages 101-119
Quantifying Appetite and Satiety (Catherine Gibbons, John E. Blundell)....Pages 121-140
Non-invasive Quantitative Magnetic Resonance Imaging and Spectroscopic Biomarkers in Nonalcoholic Fatty Liver Disease and Other Cardiometabolic Diseases Associated with Ectopic Fat Deposition (Gavin Hamilton, Michael S. Middleton, Elhamy R. Heba, Claude B. Sirlin)....Pages 141-160
Structural and Functional Imaging of Muscle, Heart, Endocrine Pancreas and Kidneys in Cardiometabolic Drug Development (Olof Eriksson, Paul Hockings, Edvin Johansson, Lars Johansson, Joel Kullberg)....Pages 161-189
Positron Emission Tomography and Computed Tomography Measurement of Brown Fat Thermal Activation: Key Tool for Developing Novel Pharmacotherapeutics for Obesity and Diabetes (Monte S. Buchsbaum, Alex DeCastro)....Pages 191-210
Isotopic Tracers for the Measurement of Metabolic Flux Rates (Carine Beysen, Thomas E. Angel, Marc K. Hellerstein, Scott M. Turner)....Pages 211-243
Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other Metabolic Disorders (Andrew J. Krentz, Pierre Bedossa)....Pages 245-274
Utility of Invasive and Non-invasive Cardiovascular Research Methodologies in Drug Development for Diabetes, Obesity and NAFLD/NASH (Gerardo Rodriguez-Araujo, Andrew J. Krentz)....Pages 275-308
Omics: Potential Role in Early Phase Drug Development (Harald Grallert, Carola S. Marzi, Stefanie M. Hauck, Christian Gieger)....Pages 309-347
Front Matter ....Pages 349-349
Peptide Drug Design for Diabetes and Related Metabolic Diseases (Niels C. Kaarsholm)....Pages 351-368
Animal Models of Type 2 Diabetes, Obesity and Nonalcoholic Steatohepatitis – Clinical Translatability and Applicability in Preclinical Drug Development (Henrik H. Hansen, Gitte Hansen, Thomas Secher, Michael Feigh, Sanne S. Veidal, Keld Fosgerau et al.)....Pages 369-403
Drug Development for Diabetes Mellitus: Beyond Hemoglobin A1c (Fernando Bril, Marta Iruarrizaga-Lejarreta, Cristina Alonso)....Pages 405-421
Emerging Circulating Biomarkers for The Diagnosis and Assessment of Treatment Responses in Patients with Hepatic Fat Accumulation, Nash and Liver Fibrosis (Marta Iruarrizaga-Lejarreta, Fernando Bril, Mazen Noureddin, Pablo Ortiz, Shelly C. Lu, José M. Mato et al.)....Pages 423-448
Quantitative Approaches in Translational Cardiometabolic Research: An Overview (Farzaneh Maleki, Puneet Gaitonde, Shannon Miller, Mirjam N. Trame, Paul M. Coen, Parag Garhyan et al.)....Pages 449-466
Transitioning from Preclinical to Clinical Drug Development (Geoffrey A. Walford, S. Aubrey Stoch)....Pages 467-486
Regulatory Considerations for Early Clinical Development of Drugs for Diabetes, Obesity, Nonalcoholic Steatohepatitis (NASH) and Other Cardiometabolic Disorders (G. Alexander Fleming, Brian E. Harvey)....Pages 487-515
Early Phase Metabolic Research with Reference to Special Populations (Linda A. Morrow, Andrew J. Krentz)....Pages 517-538
Back Matter ....Pages 539-556

Citation preview

Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease A Focus on Early Phase Clinical Drug Development Andrew J. Krentz Christian Weyer Marcus Hompesch  Editors Second Edition

123

Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease

Andrew J. Krentz Christian Weyer • Marcus Hompesch Editors

Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease A Focus on Early Phase Clinical Drug Development Second Edition

Editors Andrew J. Krentz ProSciento Chula Vista, CA USA

Christian Weyer ProSciento Chula Vista, CA USA

Marcus Hompesch ProSciento Chula Vista, CA USA

ISBN 978-3-030-11747-4    ISBN 978-3-030-11748-1 (eBook) https://doi.org/10.1007/978-3-030-11748-1 Library of Congress Control Number: 2019934762 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Foreword

Obesity, type 2 diabetes, and associated metabolic diseases such as nonalcoholic fatty liver disease (NAFLD)/nonalcoholic steatohepatitis (NASH), have reached epidemic proportions on a global scale. Insulin resistance is a fundamental etiologic defect in type 2 diabetes, and obesity is the most common cause of insulin resistance in man. There is a great deal of phenotypic overlap between obesity, type 2 diabetes, and NAFLD/NASH, and the great majority of patients who ultimately develop NASH are obese and insulin resistant. Taken together, these disorders represent one of the greatest areas of unmet medical need, creating the opportunity to discover and test new potential drugs. The chapters in this book provide an important contribution to our knowledge on the scientific and regulatory issues related to preclinical studies, with a major emphasis on the early stages of clinical development. As such, the material in this book includes a discussion of the latest methodologies in metabolic research, focusing on early clinical proof of mechanism, early-stage indicators of drug efficacy, biomarkers, and safety. In today’s landscape of drug development, a major goal is to collect information allowing the quickest possible decision on whether a drug candidate operates through the expected mechanism with the desired pharmaceutical properties of target engagement, pharmacokinetics, safety, and the necessary degree of efficacy. Hopefully, this allows informed go/no-go decisions, which enable biopharmaceutical companies to intensify their efforts on the most promising drug candidates. In this second edition of the book, the spectrum of NAFLD/NASH/cirrhosis garners much attention since this is a very active area of metabolic disease drug discovery and development, with many active clinical programs. By using the most cutting-edge in vivo methodologies to assess insulin sensitivity, insulin secretion, thermogenesis, and metabolomics, much can be learned in the early stages of clinical development which has not been possible in the past. In addition, modern imaging techniques allow assessments of hepatic fat content and liver elasticity, providing noninvasive measures across the spectrum of NAFLD/NASH. With all of these approaches available, carefully designed proof of mechanism studies will allow earlier and cleaner decisions on development programs, making resource allocations far more efficient and informative. The editors of this book, Andrew Krentz, Christian Weyer, and Marcus Hompesch, are all highly experienced experts in academic research and drug

v

Foreword

vi

development. They have recruited an expanded list of world leaders in metabolic research to cover a wider range of topics for this second edition. The new edition of the textbook retains the division into two main sections: the first section presents an even more comprehensive review of clinical investigative techniques used in early-phase clinical drug development for diabetes and NAFLD/NASH.  This section employs a structured approach that was successfully pioneered in the first edition. Notable new chapters in this section include the assessment of islet α- and β-cell function, the quantification of appetite and satiety, and the role of tissue biopsy in drug development. Imaging of NAFLD/NASH and other disorders characterized by ectopic fat deposition has been updated, and a new chapter covers state-of-­­ the-art functional imaging of key organs including muscle, heart, pancreas, and kidneys. The second section expands the perspective to preclinical drug development and transitioning to clinical studies. Included in this section are new chapters on drug design focusing on peptide drugs, biomarkers for NAFLD/ NASH, and clinical trial endpoints that lie beyond reducing glycated hemoglobin concentrations. The chapter on regulatory considerations has been expanded with a new emphasis on emerging therapies for NAFLD/NASH. The audience for this textbook includes scientists and clinicians in the biopharmaceutical industry involved in the design and implementation of first-in-man proof of mechanism and efficacy studies. Academic scientists engaged in metabolic research will also find this book to be an important resource. Lastly, this book will be beneficial to a broader audience including students and fellows who are at the early stages of their careers in this field. Jerrold M. Olefsky Department of Medicine University of California, San Diego La Jolla, CA, USA

Preface

We are very pleased to present the second edition of this textbook. The positive reception to the first edition was most gratifying for the authors and editors. The popularity of the textbook provided support for the view that we had achieved our objective of providing a useful guide for investigators involved in early-phase drug development for diabetes, obesity, and related disorders. In this second edition, we are expanding the scope by placing further emphasis on the spectrum of nonalcoholic fatty liver disease (NAFLD). This decision reflects the increasing global impact of this highly prevalent disorder, its serious health implications, and the present unmet need for the development of effective pharmacotherapies. These considerations are currently driving an intense drug development effort aimed primarily at the clinically important subtype of nonalcoholic steatohepatitis (NASH) which carries an increased risk of fibrosis, cirrhosis, and hepatocellular carcinoma. Relevant new chapters in the second edition cover invasive (liver biopsy) and emerging noninvasive imaging and circulating diagnostic and pharmacodynamic biomarkers for NAFLD/NASH. Other new additions include cardiovascular research methodologies, assessment of appetite and satiety, and transitioning from preclinical to clinical research. Furthermore, every chapter that appeared in the first edition has been revised and updated based on scientific advances in the field. The focus of the first part of the book remains on the selection of the most appropriate means of achieving the key objectives of early-phase drug development trials. The pros and cons of established and emerging clinical research methodologies are carefully considered and presented in a balanced and accessible format. The remainder of the book covers topics that include biomarkers for diabetes and insulin resistance, aspects of drug design for diabetes and related metabolic diseases, quantitative approaches to drug safety and efficacy, regulatory considerations, and the challenges of identifying appropriate subjects for clinical trials. The concepts of personalized and precision medicine are well represented throughout the book. It has been our pleasure and honor to work with the distinguished authors who have made important contributions in bringing this second edition to fruition. We are immensely grateful for the time and effort that the authors – all acknowledged leaders in their fields – have invested. The shared objective

vii

viii

to create a state-of-the-art textbook of value to clinicians and scientists has been evident at all stages. We also thank Melissa Morton, Prakash Jagannathan, and their colleagues at Springer for the support and encouragement. Feedback from readers that will help inform future editions of the book is most welcome. Chula Vista, CA, USA Andrew J.Krentz Christian Weyer March 2019 Marcus Hompesch

Preface

Contents

Part I Review of Clinical Investigative Methods 1 Quantification of Insulin Action in Human Subjects��������������������   3 Andrew J. Krentz, Christian Weyer, and Marcus Hompesch 2 Assessment of Islet Alpha- and Beta-Cell Function����������������������   37 Sten Madsbad and Jens J. Holst 3 Pharmacokinetic and Pharmacodynamic Assessment of Novel and Biosimilar Insulins��������������������������������   75 Andrew J. Krentz, Christian Weyer, and Marcus Hompesch 4 Measurement of Energy Expenditure�������������������������������������������� 101 Klaas R. Westerterp 5 Quantifying Appetite and Satiety �������������������������������������������������� 121 Catherine Gibbons and John E. Blundell 6 Non-invasive Quantitative Magnetic Resonance Imaging and Spectroscopic Biomarkers in Nonalcoholic Fatty Liver Disease and Other Cardiometabolic Diseases Associated with Ectopic Fat Deposition ���������������������������������������� 141 Gavin Hamilton, Michael S. Middleton, Elhamy R. Heba, and Claude B. Sirlin 7 Structural and Functional Imaging of Muscle, Heart, Endocrine Pancreas and Kidneys in Cardiometabolic Drug Development �������������������������������������������������������������������������� 161 Olof Eriksson, Paul Hockings, Edvin Johansson, Lars Johansson, and Joel Kullberg 8 Positron Emission Tomography and Computed Tomography Measurement of Brown Fat Thermal Activation: Key Tool for Developing Novel Pharmacotherapeutics for Obesity and Diabetes ���������������������������������������������������������������������� 191 Monte S. Buchsbaum and Alex DeCastro 9 Isotopic Tracers for the Measurement of Metabolic Flux Rates ���������������������������������������������������������������������� 211 Carine Beysen, Thomas E. Angel, Marc K. Hellerstein, and Scott M. Turner ix

x

10 Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other Metabolic Disorders������������������������������������������������������ 245 Andrew J. Krentz and Pierre Bedossa 11 Utility of Invasive and Non-­invasive Cardiovascular Research Methodologies in Drug Development for Diabetes, Obesity and NAFLD/NASH ������������������������������������������ 275 Gerardo Rodriguez-Araujo and Andrew J. Krentz 12 Omics: Potential Role in Early Phase Drug Development ���������� 309 Harald Grallert, Carola S. Marzi, Stefanie M. Hauck, and Christian Gieger Part II Preclinical Drug Development and Transitioning to Clinical Studies 13 Peptide Drug Design for Diabetes and Related Metabolic Diseases ������������������������������������������������������������ 351 Niels C. Kaarsholm 14 Animal Models of Type 2 Diabetes, Obesity and Nonalcoholic Steatohepatitis – Clinical Translatability and Applicability in Preclinical Drug Development �������������������� 369 Henrik H. Hansen, Gitte Hansen, Thomas Secher, Michael Feigh, Sanne S. Veidal, Keld Fosgerau, Jacob Jelsing, and Niels Vrang 15 Drug Development for Diabetes Mellitus: Beyond Hemoglobin A1c ������������������������������������������������������������������ 405 Fernando Bril, Marta Iruarrizaga-Lejarreta, and Cristina Alonso 16 Emerging Circulating Biomarkers for The Diagnosis and Assessment of Treatment Responses in Patients with Hepatic Fat Accumulation, Nash and Liver Fibrosis�������������������� 423 Marta Iruarrizaga-Lejarreta, Fernando Bril, Mazen Noureddin, Pablo Ortiz, Shelly C. Lu, José M. Mato, and Cristina Alonso 17 Quantitative Approaches in Translational Cardiometabolic Research: An Overview�������������������������������������� 449 Farzaneh Maleki, Puneet Gaitonde, Shannon Miller, Mirjam N. Trame, Paul M. Coen, Parag Garhyan, and Stephan Schmidt 18 Transitioning from Preclinical to Clinical Drug Development ���������� 467 Geoffrey A. Walford and S. Aubrey Stoch

Contents

Contents

xi

19 Regulatory Considerations for Early Clinical Development of Drugs for Diabetes, Obesity, Nonalcoholic Steatohepatitis (NASH) and Other Cardiometabolic Disorders������������������������������������������ 487 G. Alexander Fleming and Brian E. Harvey 20 Early Phase Metabolic Research with Reference to Special Populations �������������������������������������������������������������������������������������� 517 Linda A. Morrow and Andrew J. Krentz Index���������������������������������������������������������������������������������������������������������� 539

Part I Review of Clinical Investigative Methods

1

Quantification of Insulin Action in Human Subjects Andrew J. Krentz, Christian Weyer, and Marcus Hompesch

Summary Background Insulin resistance is a characteristic pathological hallmark of obesity, type 2 diabetes and ­non­alcoholic fatty liver disease. Reducing adiposity through non-pharmacological or pharmacological interventions improves whole-body insulin sensitivity. As an adjunct to lifestyle measures, insulin-sensitizing drugs, such as thiazolidinediones and the biguanide metformin, have a well-established role in the treatment of type 2 diabetes. In the context of drug development, insulin action is primarily focused on the assessment of whole-body glucose metabolism. Additional considerations include the contribution of major organ systems, i.e., liver, adipose tissue, muscle, brain, and the regulation by insu-

lin of lipid, protein and amino acid metabolism. Improving insulin sensitivity with thiazolidinediones improves glycemic control and may have protective effects on the cardiovascular system. However, unwanted effects include weight gain, fluid retention and an increased risk of fractures.

Key Methods Accurate and reproducible measurement of insulin sensitivity is required to evaluate new drugs with insulin-sensitizing properties. Methods for quantifying insulin action may be usefully classified according to whether the physiological feedback loop between the islet β-cells and insulin-sensitive target tissues is maintained (closedloop) or interrupted through pharmacological manipulation (open-loop).

A. J. Krentz (*) · C. Weyer · M. Hompesch ProSciento, Inc, Chula Vista, CA, USA e-mail: [email protected] © Springer Nature Switzerland AG 2019 A. J. Krentz et al. (eds.), Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease, https://doi.org/10.1007/978-3-030-11748-1_1

3

A. J. Krentz et al.

4

(a) Closed-loop methods Method Venous serum insulin and plasma glucose are measured in blood samples drawn after an 8–12 h overnight fast

Measure Insulin sensitivity (%S)

Advantages Technically simple; relatively inexpensive; provides an indication of insulin sensitivity in the basal state

Disadvantages Indirect assessment of insulin action; only assesses metabolism in basal (nonstimulated) state; degree of insulin resistance may be underestimated in the presence of hyperglycemia

Mixed meal tolerance test (MMTT)

Plasma glucose and serum insulin responses at defined intervals to a standardized meal

Area under the curve (AUC) for insulin; mathematical models of insulin and glucose responses (e.g. Matsuda Index; Stumvoll Index)

Indirect assessment of insulin action; issues of intraindividual and between-individual variability; may be affected by gastric emptying rate

Oral glucose tolerance test (OGTT)

Glucose, insulin and/or C-peptide responses to 75 g oral glucose

AUC for insulin; mathematical models of insulin and glucose responses

Provides data of relevance to human physiology; flexible, i.e. nutrient components can be adjusted; assesses integrity of incretin axis Simple to perform; reference methods for diagnosing diabetes and impaired glucose tolerance; large existing scientific literature

Insulin tolerance test (ITT)

Response of blood glucose to an intravenous bolus of glucose

Glucose disposal rate (KITT)

Fasting serum insulin and glucose; mathematic models include HOMAa, HOMA2, iHOMA2 and QUICKIb

Homeostasis model assessment Quantitative insulin sensitivity check

a

b

Technically straightforward

Indirect assessment of insulin action; issues of intrasubject and inter-subject variability; β-cell response to secretagogues other than glucose is not assessed; may be affected by gastric emptying rate Risk of hypoglycemia; cannot partition insulin action between insulinmediated glucose disposal and suppression of hepatic glucose production

Value in drug development decisions May provide useful exploratory data in early phase studies; results should be confirmed with a dynamic test of insulin sensitivity; iHOMA2 permits modelling of physiology and drug treatment effects May be of value in providing early signal of effects of drugs on insulin action

May be of value in providing early signal of effects of drugs on insulin action

Limited role in diabetes drug development

1  Quantification of Insulin Action in Human Subjects

5

(b) Open-loop methods Somatostatin is infused to suppress endogenous insulin secretion; exogenous insulin and glucose are infused intravenously to achieve steadystate plasma glucose

Steady-state plasma glucose (SSPG)

Reproducible steady-state method which eliminates endogenous insulin secretion and assesses insulinmediated glucose disposal

Frequently sampled intravenous glucose tolerance test (FSIVGTT)

Glucose, insulin and C-peptide responses to an intravenous bolus of glucose; minimal model analysis of data

Provides dynamic data; widely used in clinical metabolic research

Insulin sensitivity clamp (two-step euglycaemic hyperinsulinaemic clamp)

Insulin is infused to provide steady-state hyperinsulinemia at pre-determined insulin concentrations; variable rate hypertonic glucose is infused to maintain euglycemia

Minimal model yields insulin sensitivity index (SI) and glucose effectiveness (SG) Glucose disposal rate (M); M/I; Insulin sensitivity index(SIclamp)

Conclusions In the context of early phase drug development of diabetes drugs with insulin-sensitizing properties, selection of the most appropriate method for determining insulin action will be influenced by considerations including putative mechanism of action of the investigational product, budgetary considerations, and expertise of the clinical investigators. Careful selection of study subjects

Direct measure of insulin-mediated glucose disposal; reproducible; low co-efficient of variation; can be combined with complementary techniques, e.g. isotopic determination of glucose turnover, indirect calorimetry; automatic clamps using the Biostator or equivalent devices offer certain advantages over manual clamps

Indirect assessment of insulin action on glucose metabolism; labour intensive; relatively inflexible; hepatic insulin sensitivity cannot be determined Indirect integrated assessment of glucose metabolism; questionable relevance to physiology Labour intensive; requires skilled technical staff

Limited role in diabetes drug development decisions

Limited value in diabetes drug development decisions

Generally regarded as the reference method for determining insulin sensitivity

and standardization of dietary intake and physical activity are important considerations in study design. Whole-body insulin sensitivity  – which is a composite of hepatic and peripheral insulin sensitivity – is most robustly measured by experienced investigators using the euglycaemic hyperinsulinaemic clamp technique. Advantages of the euglycaemic hyperinsulinaemic clamp include its sensitivity, reproducibility and adaptability. The technique can be readily combined with com-

A. J. Krentz et al.

6

plementary assessments such as isotopic glucose turnover (e.g. capturing rates of endogenous glucose production), indirect ­calorimetry (facilitating assessment of oxidative and non-oxidative glucose utilization), or studies of tissue protein expression and enzyme activity. Set against these advantages are the resource- and labor-intensive aspects of glucose clamp studies.

Table 1.1  Physiological and pathological states associated with whole-body insulin resistance Physiological states  Adolescence  Pregnancy (Second and third trimesters)  Luteal phase of the menstrual cycle  Post-menopausea  Aginga Common pathological conditions  Obesityb  Glucose intolerance  Type 2 diabetes  Metabolic syndromec  Sedentary lifestyle (vs. regular physical activity, incapacity)  Nonalcoholic fatty liver disease/steatohepatitis

Introduction Insulin is a pivotal hormone that regulates multiple aspects of human metabolism at cellular, organ and whole-body level. Impaired action of insulin – insulin resistance – is a well-documented feature of many physiological and pathological states (Table  1.1) [1–4]. These range from (a) temporary subclinical decrements in insulin action, e.g. during prolonged bedrest or during the second and third trimesters of pregnancy to (b) chronic and clinically apparent disturbances of insulin action as manifest in type 2 diabetes to (c) rare genetic clinical syndromes of severe insulin resistance [5]. Accordingly, there is a wide spectrum of clinical implications of insulin resistance. For example, for a woman with pre-existing or gestational diabetes, pregnancy-related insulin resistance can usually be managed through initiation of insulin treatment and/or a temporary increase in insulin dosage to control hyperglycemia. In contrast, inherited syndromes of severe insulin resistance, e.g. partial or total lipodystophies, may present major management challenges requiring novel therapeutic interventions [5, 6]. Insulin resistance is a major modifiable factor implicated in the pathogenesis of glucose intolerance and type 2 diabetes. From a global public health perspective, excess adiposity – manifested clinically as overweight and obesity  – is the most common acquired cause of impaired insulin action. Insulin resistance is closely associated with an adverse profile of major risk factors for cardiovascular disease including hyperglycemia, dyslipidemia and hypertension [7].

Modified from Krentz [4] The evidence for direct effects of these physiological processes on insulin sensitivity is inconsistent. Changes in body composition and other factors may, at least in part, explain the reduced insulin action reported in some studies b Includes lesser degrees of overweight. Abdominal adiposity, which is characteristic of males, is more closely associated with whole-body insulin resistance than gynoid subcutaneous fat deposition. Ethnicity is an important modifier of the metabolic effects of adiposity; non-white populations including East and South Asians develop adverse cardiometabolic profiles at lower levels of body mass index compared with counterparts of white European ancestry. Ectopic fat in skeletal muscle and liver are closely correlated with impaired whole-body insulin action. Ectopic fat may also be deposited in the pancreas and the heart and vascular system with detrimental effects on organ function c Various definitions of the metabolic syndrome have been proposed. The main features are abdominal adiposity, glucose intolerance, hypertriglyceridemia, low levels of high-density lipoprotein cholesterol, hypertension in variable combinations and in association with insulin resistance and hyperinsulinemia a

I nsulin Physiology and Metabolic Regulation: Therapeutic Implications The majority of human cells express surface insulin receptors. Hepatocytes, skeletal myocytes, and white adipocytes are considered classic insulinresponsive tissues [8]. Insulin regulates glucose metabolism both through direct actions [9] and by influencing inter-organ c­ross-talk pathways e.g. via altered synthesis and secretion of lipids and adipocytokines (Table 1.2) [10].

1  Quantification of Insulin Action in Human Subjects Table 1.2  Classic metabolic actions of insulin Skeletal muscle* Liver

Adipose tissue

Direct effects ↑ Glucose uptake ↑ Glucose oxidation ↑ Glycogen synthesis ↓ Glucose output   ↓ Glycogenolysis   ↓ Gluconeogenesis ↑ Glycogen synthesis ↑ Glycolysis ↑ Lipogenesis ↑ Glucose uptake ↑ Lipogenesis

Indirect effects ↓ NEFA availability and oxidation ↓ NEFA availability and oxidation

↓ Lipolysis Regulation of adipocytokines

Adapted from Konrad et al. [9] *Cardiomyocyte metabolism is also regulated by insulin NEFA non-esterified fatty acids

Insulin signaling within the brain regulates appetite, reproductive function, body temperature, adiposity, hepatic glucose output, and protective responses to hypoglycemia [11, 12]. Insulin levels in the cerebrospinal fluid are ∼25% of those in the blood and increase proportionally after meals or with peripheral infusion of insulin. Certain regions of the brain, such as the hypothalamus, appear to lack an effective barrier to insulin [12]. Other non-classic target tissues for insulin include the heart [13], skeleton [14], brown adipocytes [15], ovaries [16], and the vascular endothelium [17]. Insulin may also exert direct autocrine effects on β-cells [18, 19]. In the setting of insulin resistance, organ function may be affected in either favorable or unfavorable directions by insulin-sensitizing pharmacotherapies such as thiazolidinediones [20]. For example, pioglitazone reduces the risk of stroke or myocardial infarction after thromboembolic stroke or transient ischemic attack in insulin-resistant subjects but is associated with increased risks of weight gain, heart failure and skeletal fractures [21]. The range of physiological actions of insulin has expanded beyond regulation of carbohydrates and other macronutrients to include anti-oxidant, anti-inflammatory and vascular effects (Table 1.3) [22, 23]. Impaired insulin action may mediate long-term organ damage in part through dysregulation of these pathways [24].

7 Table 1.3  Non-classic actions of insulin Anti-oxidant   ↓ Reactive oxygen species Anti-inflammatory   ↓ NFκB ↓ C-reactive protein Anti-thrombotic   ↓ Tissue factor ↓ platelet activation Pro-fibrinolytic   ↓ Plasminogen activator inhibitor-1 Vasodilatory  NO-mediated improvements in endothelial function  Large vessel compliance Lipid regulation   ↓ Hepatic production of very-low density lipoproteins Sympathetic nervous system   ↑ Activation Anti-atherosclerotic actions  Demonstrated in the Apo E null mouse, IRS-1 null mouse Brain  Regulation of appetite, reproductive function, body temperature, white fat mass, hepatic glucose output, response to hypoglycemia. Adapted from Yki-Järvinen and Westerbacka [22]; Dandona et al. [23]; Kleinridders et al. [12] NO nitric oxide, Apo E apoliporotein E, IRS-1 insulin receptor substrate-1, NFκB nuclear factor kappa-lightchain-enhancer of activated B cells

 ellular Insulin Signaling C In peripheral tissues such as muscle and fat, insulin must first leave the intravascular compartment to traverse the interstitial space before interacting with cellular insulin receptors [25]. The insulin receptor is a transmembrane heterodimer comprising two α- and two β-subunits (Fig. 1.1). Binding of insulin to the α-subunit induces a conformational change resulting in release of the inhibitory effect of this subunit and autophosphorylation of tyrosine residues in the β-subunit [26]. Insulin receptor tyrosine kinase phosphorylates and recruits substrate adaptors such as the insulin receptor substrate (IRS) family of proteins. In turn, this initiates events that result in translocation of the facilitative glucose transporter (GLUT4) from the cytosolic vesicles to the cell membrane [27, 28]. Fusion of GLUT4 with the cell membrane results in transport of glucose into the cell where it is phosphorylated to glucose-6-phosphate. A post-binding

A. J. Krentz et al.

8

a Insulin receptor

GLUT 4 Plasma membrane IRS-1

p85

p 110

Glucose

Akt

Pl-3-kinase

NO

S

-1 IRS

Artery Glucose metabolism

b Insulin

Insulin receptor

Glucose GLUT 4 Plasma membrane IRS-1

p85

p 110

Akt

Pl-3-kinase

NO SHC

S

1 RS-

I

P MA ase n i k

Artery Inflammation Cell proliferation Atherosclerosis

Fig. 1.1 (a) Insulin signal transduction system in individuals with normal glucose tolerance. NOS, nitric oxide synthase. (b) In patients with type 2 diabetes insulin signalling is impaired at the level of insulin receptor substrate (IRS)-1 leading to decreased glucose transport/phosphorylation/metabolism and impaired nitric oxide synthase activation/endothelial function. At the same time, insulin signalling through the mitogen activated protein (MAP)

kinase pathway is normally sensitive to insulin. The compensatory hyperinsulinaemia (due to insulin resistance in the IRS-1/phosphatidylinositol-3 (PI-3) kinase pathway results in excessive stimulation of the former pathway, which is involved in inflammation, cell proliferation and atherogenesis. SHC, Src homology collagen. (Reproduced with permission from DeFronzo [62])

cascade of phosphorylation/dephosphorylation reactions activates key enzymes including glycogen synthase and pyruvate dehydrogenase. In the presence of hyperinsulinemia glucose-6-phosphate is mainly (approximately 70%) polymerized to form glycogen, the remainder enters the glycolytic pathway and is either oxidized or converted to lactate [29].

In addition to these acute metabolic effects, insulin is a potent growth factor that exerts transcriptional effects on cell growth and differentiation [30], via the mitogen-activated protein (MAP) kinase pathway [31]. Other actions of insulin include regulation of protein and amino acid metabolism [32] and aspects of cellular ion transport [33].

1  Quantification of Insulin Action in Human Subjects

Insulin Resistance Impaired insulin action is recognized as an important factor in the development and progression of a number of adverse clinical outcomes [34]. Insulin resistance may be defined as a state in which normal concentrations of insulin produce a less than normal biological response [35]. While this definition encompasses all aspects of insulin action the focus in terms of diabetes drug development is, by definition, primarily on glucose metabolism. The assessment of insulin sensitivity in humans is challenging; accurate quantification of insulin action is possible only within a specialized clinical research setting. Clinically useful biomarkers of insulin resistance that could be used to inform therapeutic decisions, e.g. when to initiate an insulin-sensitizing drug in patients with type 2 diabetes, remain elusive [36]. The molecular basis for impaired metabolic actions of insulin in response to nutritional excess remains incompletely understood. Cellautonomous mechanisms and inter-organ communications are both implicated [37]. While much attention has focused on defects in the early steps in the insulin signaling cascade it is apparent that many of the intracellular defects of insulin resistance are elicited further downstream [37]. Increasing levels of obesity are generally correlated with greater insulin resistance [38–40]. However, body mass index (BMI) has limitations as an indicator of whole-body insulin sensitivity [41]. In a study of 465 healthy volunteers, 16% of those in the most insulin-resistant tertile as assessed using the insulin suppression test (see below) were of normal weight (BMI 24 h prior to metabolic studies Miscellaneous  Shift work; sleep deprivation, sleep-disordered breathing, may be associated with reduced insulin sensitivity  Tobacco use (associated with insulin resistance; smoking activates sympathetic nervous system) Central adiposity; glucose intolerance; hypertriglyceridemia; low levels of HDL-cholesterol; hypertension

a

1  Quantification of Insulin Action in Human Subjects

affect glucose metabolism or body weight, either positively or negatively. In countries with well-developed healthcare systems it may be difficult to recruit drug naïve patients with type 2 diabetes for early phase clinical studies of new diabetes drugs as metformin therapy is recommended at an early stage after diagnosis [66]. Monotherapy with metformin or a sulfonylurea may be withheld from selected patients with type 2 diabetes in preparation for such studies. Pre-specified upper limits of hyperglycemia should be not be exceeded when glucose-lowering therapy is temporarily withheld in treatment wash-off studies in patients with type 2 diabetes. The period required for the metabolic effects of prior treatment to dissipate also needs to be considered. Limited data suggest that the glucose-lowering effects of metformin on glucose metabolism have largely dissipated 1  week following withdrawal of the drug. For thiazolidinediones [121] or certain long-acting glucagon-like peptide (GLP)-1 receptor agonists [122] which have effects on insulin sensitivity and body weight, a longer period of withdrawal, e.g. 2–3 months, is recommended to ensure a new metabolic steady state.

Closed-Loop Assessments As mentioned, closed-loop measures of insulin action depend on maintenance of the physiological feedback loop between glucose metabolism and islet β-cell function. The tests may be divided between those that reflect insulin action in the fasting (basal) state, and tests that are based on a glucose challenge, either oral or intravenous.

 losed-Loop Tests of Basal Insulin C Sensitivity Fasting insulin levels  In the fasting state hyperinsulinemia in the presence of normo- or hyperglycemia indicates insulin resistance. Homeostasis model assessment is a mathematical model that estimates insulin sensitivity as a percentage of a normal population based on fasting serum insulin and blood glucose concentrations.

15

HOMA-IR  This test was originally described by Professor David Matthews and colleagues at the University of Oxford [123]. Both the original HOMA approach and an updated version (HOMA2) (https://www.dtu.ox.ac.uk/homacalculator/) [124] assume the presence of a feedback loop between the liver and β–cells. HOMA-IR is the reciprocal of %S (100/%S).

HOMA-IR - G b ´ Ib / k

Gb and Ib are, respectively, basal (fasting) glucose (in mmol/L) and insulin (in mU/L) with k = 22.5. Since physiological insulin secretion from the β-cells is pulsatile the mean of three samples taken at 5-min intervals to compute HOMA is more considered more accurate than a single sample; in practice, however, a single glucose measurement is often used. For large subject sample sizes this compromise provides comparable data [125]. As insulin action is not measured directly HOMA-IR may be more appropriately considered a surrogate of insulin resistance. The quantitative insulin sensitivity check (QUICKI) is the reciprocal of the logarithm of HOMA-IR with k assigned a value of 1 [126]. HOMA has been compared with a variety of investigative methods for assessing insulin sensitivity such as the glucose clamp [125] with somewhat inconsistent conclusions [127]. It has been proposed that HOMA-IR provides data primarily relevant to basal hepatic glucose metabolism since the assessment is made in the fasting state. If glucose uptake in skeletal muscle is of interest consideration should be given to using a technique that raises circulating insulin concentrations to levels that stimulate glucose disposal. While HOMA2 also permits assessment of steady-state β-cell function (%B) this is not possible in patients treated with exogenous insulin. HOMA-IR has the advantages of being technically straightforward, quantitative, and relatively inexpensive. The accuracy of HOMA is dependent in part on the precision of the insulin assay as well as the type of sample, i.e. serum vs. plasma [128]. The absence of an international standardized assay for insulin [129] precludes use of HOMA-IR to define universal cut-off points

A. J. Krentz et al.

16

for insulin resistance. For subjects without diabetes the correlation between HOMA-IR and fasting insulin concentration is close to unity [130]. HOMA-IR is well suited to large-scale studies where its deployment is perhaps most appropriately positioned [131]. The relatively high intra-individual variability renders HOMA-IR less suitable for determining the impact of interventions aimed at improving insulin sensitivity [132]. The characteristics of the study population is also relevant. The correlations between HOMA and a more sophisticated measure of insulinmediated glucose uptake (insulin suppression test – see below) were less good in normal weight individuals than in overweight or obese subjects (correlation coefficients 0.36, 0.55, and 0.60, respectively [130]). Similar (inverse) trends were observed for QUICKI [130]. With these caveats in mind, HOMA-IR may nonetheless be useful as an exploratory endpoint in studies in which a change in insulin sensitivity is postulated. For example, a study of the effects of niacin/laropiprant in women with ovary syndrome showed an impairment of glucose regulation despite beneficial effects on blood lipid profiles. The adverse effect on glucose metabolism was accompanied by an increase in HOMA-IR (3.8 vs. 2.2; p  =  0.02) [133]. In a meta-analysis of studies in non-diabetic subjects, angiotensin receptor blockers improved HOMA-IR relative to calcium channel blockers for a similar degree of blood pressure lowering [134]. Divergent effects on insulin sensitivity may be of relevance to the hierarchy of risk of new-onset type 2 diabetes associated with different classes of antihypertensive agents [135]. The Oxford investigators have extended the HOMA2 model in a 24-variable model  – iHOMA2 (http://www.ihoma2.co.uk). The iHOMA2 model addresses some limitations of the invariant model to permit analytic and predictive modes of use [136]. The iHOMA2 model has been tested against data derived from clinical studies of (a) the insulin-sensitizing thiazolidinedione pioglitazone and (b) the sodium-glucose cotransporter-2 (SGLT-2) inhibitor (dapagliflozin) which lowers plasma glucose through renal mechanisms independent of insulin sensitivity. Results

for pioglitazone were considered to be consistent with glucose clamp data indicating that the principal mechanism of action of the drug in the fasting state is mediated via effects on hepatic gluconeogenesis [136]. For dapagliflozin, manipulation of iHOMA2 renal excretion threshold for glucose produced results in close agreement with published data from other sources. The investigators suggest that iHOMA2 could be of value in diabetes drug development by predicting effects on fasting glucose and insulin levels and on insulin sensitivity and ß-cell function. For SGLT-2 inhibitors – which may improve insulin sensitivity by alleviating glucotoxicity – iHOMA2 could be used to examine dose-response and to provide illustrative comparisons between different SGLT-2 inhibitors [136].

Closed-Loop Dynamic Tests In response to oral glucose, the release of endogenous insulin from β-cells is stimulated both directly by glucose and in part indirectly via the activation of the incretin system [137]. The stimulated insulin response to an oral or mixed meal challenge may be used to provide an estimate of insulin sensitivity; the higher the insulin levels, the greater the impairment of insulin action. However, from the foregoing discussion, it will be readily appreciated that the presence of β-cell dysfunction compromises the validity of this approach for assessing insulin sensitivity: if β-cell compensation is incomplete the degree of hyperinsulinemia for a specified level of hyperglycemia will underestimate the level of insulin resistance. If compensatory endogenous insulin secretion was restored, e.g. using a classic insulin secretagogue such as a sulfonylurea, then the resulting hyperinsulinemia would return elevated blood glucose levels back towards normality. However, in the case of sulfonylurea therapy the dynamics of insulin secretion are not normalized in an important sense: insulin secretion is not regulated according to prevailing glucose concentrations [67]. Sulfonylureas continue to stimulate insulin secretion even when plasma glucose has been restored to normal levels; this

1  Quantification of Insulin Action in Human Subjects

property accounts for the well-recognized risk of hypoglycemia. This risk varies according to the particular sulfonylurea, reflecting differences in pharmacokinetics and pharmacodynamics [138]. The risk of such inappropriate hyperinsulinemia, with suppression of hepatic glucose production, is largely avoided with drugs that act via the incretin system, i.e. glucagon-like peptides (GLP-1)-analogs and dipeptidyl peptidase-4 (DPP-4) inhibitors [67, 139]. Drugs in the latter classes promote insulin secretion only in the presence of hyperglycemia [139]. Intravenous glucose tolerance test  In this technique, in which the aforementioned caveat concerning impaired β-cell function also applies, has been widely applied in clinical metabolic research. A bolus of glucose (0.3 g/kg) is administered via an indwelling venous catheter after an overnight fast. Venous blood is sampled frequently from the contralateral arm for insulin and glucose over a period of 3 h [140]. The resulting peak, and subsequent decline, in blood glucose reflects the influence of both the endogenous insulin response and whole-body insulin sensitivity. Modifications of the frequently sampled intravenous glucose tolerance test (FSIVGTT) include co-administration of a sulfonylurea (tolbutamide) to enhance insulin secretion [141] or a bolus of exogenous insulin (0.5  U/kg) [25]. Richard Bergman and colleagues advanced the analysis of data derived from the FSIVGTT and modified FSIVGTT with the introduction of the so-called minimal model [142, 143]. In this approach, the insulin sensitivity index, SI, is calculated from two differential equations. SI, which reflects composite insulin action on muscle and adipose tissue, reportedly correlates well with the M value obtained using the euglycaemic hyperinsulinaemic clamp technique, at least in relatively insulin sensitive and glucose tolerant subjects [144]. In addition, the minimal model permits an assessment of the ability of glucose to promote its own disappearance from the circulation (and inhibit its endogenous appearance) independently of insulin; this value is known as glucose effectiveness or SG [145]. The FSIVGTT avoids potential variations in gastric emptying that may

17

affect the metabolic response to an oral glucose challenge in patients with diabetic autonomic neuropathy [146]. However, this advantage, which in any case is less likely to be relevant in patients with recently diagnosed diabetes, comes at the expense of losing the influence of the incretin effect on insulin secretion [147]. Concerns have been expressed about aspects of validity of the minimal-model and its limited correlation with more direct methods for assessing insulin action [148–150]. Nonetheless, using the theoretical framework of the minimalmodel Michael Schwartz and colleagues at the University of Washington have proposed novel diabetes therapies that target a putative braincentered glucoregulatory system that modulates insulin-independent mechanisms (SG) [151]. Oral glucose tolerance test  The oral route of glucose delivery is self-evidently closer to normal physiology than an intravenous glucose challenge. However, the 75 g oral glucose tolerance test (OGTT) cannot be regarded as a physiologic assessment of glucose metabolism as is sometimes implied in the scientific literature. Other limitations of the OGTT include relatively low day-to-day reproducibility [152]. Factors contributing to intra-individual variability include inconstant rates of glucose absorption and splanchnic glucose uptake, variations in gastric emptying [153] and the modulating effects of gut-derived incretin hormones. A longer-term influence on glucose tolerance comes from the macronutrient composition of the diet, e.g. percentage of calories derived from carbohydrate vs. fat and the proportion of monounsaturated fat [154]. Less well-documented dietary factors that can affect oral glucose tolerance include micronutrient status [155] and the influence of the intestinal microbiome [156]. The 75  g OGTT is widely used in clinical practice to diagnose, or exclude, defined categories of glucose intolerance and type 2 diabetes [157]. The procedure is as follows: after overnight fast, venous blood samples for glucose and insulin concentrations are taken at baseline

18

and then every 30  min until 120  min following a standard oral glucose load (75 g); a standardized meal or meal substitute, e.g. Ensure®, may be used for clinical research purposes. However, the diagnosis of impaired glucose tolerance and diabetes require a 75 g glucose challenge and is based solely on the baseline and 120 min glucose levels [157]. Care has to be taken that glucose is measured using an appropriate method with good accuracy. Note that the classification thresholds differ according to the glucose sample. Glucose should be dissolved in water so that the maximum glucose concentration in the beverage is 25 g/100 mL. The drink, which may be made more palatable with a non-calorie flavor additive, should be consumed within 5 min with the subject sitting quietly throughout the test. For children, the glucose load is calculated according to body weight, i.e. 1.75 g/kg of weight to a maximum of 75  g. It has long been recognized that dietary carbohydrate restriction in the days preceding an OGTT may temporarily impair glucose tolerance. Accordingly, 100–150  g/day carbohydrates should be consumed as part of the diet for 3 days prior to the scheduled OGTT. In parentheses, an example of the clinical impact of prior carbohydrate intake on insulin action is provided by the benefits of carbohydrate loading on surgical outcomes [158]. Several whole-body insulin sensitivity indexes based on measurements derived from the OGTT are available. These include the Matsuda index [159] and the Stumvoll index [160]. The latter have been validated against the glucose clamp technique in subjects with a range of metabolic states and are widely regarded as providing reliable indictors of insulin sensitivity. Others include the OGIS120 index of Mari et al. [161] and an oral glucose minimal model enabling measurement of insulin sensitivity that has been developed and validated against multi-tracer and euglycemichyperinsulinemic glucose clamp protocols [150, 162]. In general, these measures of insulin sensitivity are more accurate in non-diabetic individuals who have normal β-cell function. Moving beyond measures of insulin and glucose, Beysen et al. developed the deuterated-glucose disposal test (2H-GDT). The stable isotope [6,6-2H2]glu-

A. J. Krentz et al.

cose is administered in a 75 g oral glucose load in order to determine whole-body glycolysis. Glycolytic disposal of the deuterated glucose generates 2H2O from which an index of insulin sensitivity is calculated using the insulin exposure resulting from the glucose challenge [163]. The investigators report close correlations with measures of insulin resistance using the hyperinsulinemic euglycemic glucose clamp and the insulin suppression test (see below) across a range of insulin sensitivities [163]. The rise in blood glucose following oral loading is determined in part by the degree of suppression of hepatic glucose production in addition to the absorption and disposal of the oral glucose load [164]. Radiolabeled tracers may be used to ascertain the metabolic fate of an intestinal glucose load and quantify the contribution of endogenous glucose production to the post-challenge blood glucose concentration. The merits of the various tracer options, which involve use of two or more ingested and infused tracers, have recently been reviewed by Dube et  al. [165]. As an example of the application of tracer methodology in a proof-of-mechanism study Polidori et al. used a mixed meal tolerance test with a dual tracer (3H-glucose, 14C-glucose) method to examine the effects of canagliflozin, a sodium glucose co-transporter (SGLT)-2 inhibitor which also has activity at the intestinal SGLT-1 receptor. Canagliflozin reduced postprandial plasma glucose and insulin levels in healthy subjects by increasing urinary glucose excretion (via renal SGLT-2 inhibition) and delaying the rate of appearance of oral glucose (RaO), the latter being attributed to intestinal SGLT-1 inhibition [166]. In this study, plasma insulin level was reduced as a secondary consequence of reduced intestinal glucose absorption, underscoring the potential limitations of the oral glucose challenge as an indicator of insulin action [167]. This study provides an illustration of the well-recognized phenomenon whereby an improvement in insulin sensitivity, indicated by lower insulin levels in conjunction with improved post-challenge glucose tolerance, may occur secondary to a ­reduction in hyperglycemia. Any intervention that lowers blood glu-

1  Quantification of Insulin Action in Human Subjects

cose, whether dietary modifications or exercise, along with a range of non-insulin sensitizing drugs, can reduce insulin resistance by relieving the negative metabolic impact of hyperglycemia on cellular insulin action, i.e. glucotoxicity [65]. It follows that demonstrating that a drug has a primary insulin-sensitizing effect requires direct evidence of improved insulin action using appropriate investigative techniques. Of these, the euglycaemic hyperinsulinaemic clamp is regarded as the reference method (see below). Insulin tolerance test (ITT)  This involves an intravenous injection of a bolus of exogenous insulin (typically 0.1  U/kg) in the fasting state. The response of blood glucose reflects the effects of the injected insulin on hepatic and peripheral insulin-sensitive tissues. Due to the dynamic rise and subsequent fall in serum insulin concentrations the contribution to the decline in blood glucose by reduced hepatic glucose production and insulin-stimulated glucose uptake, respectively, will vary during the test according to relative dose-response characteristics. An insulin sensitivity index may be calculated from the ratio of the change in blood glucose to the basal glucose concentration. The glucose disposal rate (KITT) may be calculated from the slope of the regression line of the logarithm of blood glucose against time during the first 3–15 min [168]. The ITT has major limitations, notably the risk of inducing a harmful hypoglycemia. Clinical hypoglycemia is usually unpleasant and potentially hazardous in certain subjects, e.g. those with ischemic heart disease. Furthermore, the release of the counterregulatory hormones, primarily glucagon and catecholamines in response to hypoglycemia, antagonize the actions of insulin thereby confounding the test results. A mean coefficient of variation of 30% was reported in a study using a 0.05 U/kg ITT insulin healthy volunteers [168]. However, more acceptable reproducibility has been reported using the same insulin dose by other investigators [169]. The ITT is not widely used in clinical metabolic research into the pathophysiology of type 2 diabetes. Furthermore, the ITT cannot be regarded as meeting the rigorous requirements of drug development studies.

19

Open-Loop Methods Insulin suppression test (IST)  This method, introduced by Gerald Reaven and colleagues at Stanford University in 1970, was the first to offer direct measurement of insulin action [150, 170]. In its original inception, an intravenous infusion of epinephrine was used to suppress insulin secretion. The non-selective β-adrenergic blocker propranolol was co-infused to counter the metabolic and hemodynamic effects of the epinephrine. However, complete blockade of the adrenergic receptors could not be guaranteed and the hazard of cardiac arrhythmias led to a modification of the approach to interrupt the glucose-βcell feedback loop [171]. In the latter approach, an intravenous infusion of somatostatin or the somatostatin analog octreotide [172] is used to suppress endogenous insulin and glucagon secretion from the pancreatic β- and α-cells, respectively [172]. Simultaneous infusions of insulin (25  mU/m2/min) and glucose (240  mg/m2/min) are delivered for 3 h. Blood samples for the determination of plasma glucose and insulin are drawn from the contralateral arm every 30 min for 2.5 h and then at 10-min intervals out to 180  min. Under these controlled conditions the steadystate plasma glucose (SSPG) concentration between 150 and 180 min reflects the net effect of the achieved hyperinsulinemia on insulin-sensitive tissues. Since SSPG is inversely related to insulin sensitivity it is predicted that the SSPG concentration will be higher in more insulinresistant subjects, and vice versa. Assumptions inherent in the interpretation of the IST include the complete suppression of endogenous insulin and glucagon secretion, similar effects on growth hormone secretion (which is also suppressed by somatostatin), reproducible direct effects of somatostatin on splanchnic blood flow and peripheral glucose metabolism, and attainment of steady-state hyperinsulinemia. A two-stage variant of the IST that comprises insulin infusions to achieve basal (approximately 15  mU/L; 90 pmol/L) and physiologically elevated (approximately 80 mU/L; 480 pmol/L) steady-state insulin concentrations has been described [173]. This approach permits insulin action on suppression

A. J. Krentz et al.

20

Insulin sensitivity clamp  This technique  – euglycaemic hyperinsulinaemic clamp  – is widely regarded as the ‘gold standard’ among methods for quantifying insulin action. However, while this accolade attests to particular advantages of the technique it should not be taken to imply that the clamp offers insights into insulin action that fundamentally reflect physiology. Rather, the euglycaemic hyperinsulinaemic clamp technique is a pharmacological perturbation of homeostatic metabolic mechanisms that provides quantifiable and reproducible data with which to test hypotheses about the role of insulin resistance in human disease. The hyperinsulinemic euglycemic clamp establishes a temporary state of sustained hyperinsulinemia at a pre-specified level. This permits quantification of a key aspect of glucose metabolism, i.e. insulin-mediated glucose disposal in skeletal muscle. One of the advantages of the technique is that it be can be readily adapted to provide a broader perspective of human metabolism, including hepatic glucose production and adipocyte lipolysis. To achieve this flexibility, the dose-response curves for each of these processes have to be considered. The inhibitory effect of insulin on hepatic glucose production occurs at much lower plasma insulin concentrations than those required to maximally stimulate glucose uptake in skeletal muscle (Fig.  1.3) [143, 174, 175]. In a one-step hyperinsulinemic euglycemic glucose clamp study performed in healthy volunteers, Rizza et al. observed half-maximal suppression of hepatic glucose production (determined using [3-3H]glucose as a tracer) at a mean (±SD) plasma insulin concentration of 29  ±  2  mU/L (174 ± 12 pmol/L). In contrast, the insulin concentration required for half-maximal stimulation of glucose utilization was almost twice as high at

55 ± 7 mU/L (330 ± 36 pmol/L) [176]. Maximal glucose utilization occurred at pharmacological insulin concentrations of 220–700 mU/L (1320– 4200 pmol/L). Other groups have reported similar findings. As discussed above, impaired insulin-mediated suppression of hepatic glucose production, i.e. hepatic insulin resistance, is the main driver of pathological states of fasting hyperglycemia. Following an overnight fast, plasma glucose concentrations are maintained entirely from endogenous sources. The liver and the kidney are the only two organs in the human body with sufficient gluconeogenic enzyme activity and glucose-6-phosphatase activity to release glucose into the circulation as a result of gluconeogenesis. Glucose production by the liver accounts for approximately 80% with renal glucose production accounting for the remainder [177, 178]. When insulin sensitivity in glucose metabolism is impaired defects in other aspects of metabolism are usually detectable using a­ ppropriately sensitive techniques [112, 179]. Adipocyte lipolysis is especially sensitive to inhibition by insulin, with only small increments above fasting levels being sufficient to restrain hydrolysis of triglycerides (Fig.  1.3) [112, 175, 179].

Stimulate glucose uptake Relative amount of insulin

of lipolysis, i.e. NEFA and glycerol, and stimulation of glucose disposal to be assessed at relevant insulin concentrations, i.e., first-stage for NEFA concentrations and second-stage for glucose concentrations [173]. While the IST has been used extensively in clinical metabolic research it has not been widely applied in diabetes drug development.

Suppress HGP Suppress lipolysis

Fig. 1.3 Approximate relative amounts of insulin required for maximal effects on major metabolic processes in  vivo. Dose-response relationships may alter in the presence of insulin resistance to preferentially affect one or more processes. HPG  =  hepatic glucose production

1  Quantification of Insulin Action in Human Subjects

Circulating NEFA concentrations are elevated in subjects with obesity, impaired glucose tolerance, and subjects with type 2 diabetes and fail to suppress normally in response to insulin [112]. These defects are indicative of insulin resistance at the level of the adipocyte with impaired suppression of lipolysis. Two-step hyperinsulinemic euglycemic clamp  This variant of the technique can be used to assess suppression of NEFA by insulin and other hormones, e.g. insulin-like growth factor-1, at low and high insulin concentrations [180]. The two-step hyperinsulinemic glucose clamp technique may also be combined with isotopic determination of glycerol turnover to provide a dynamic assessment of the sensitivity of adipocyte lipolysis [181]. The twostep hyperinsulinemic euglycemic clamp is suitable for determining dose-response suppression of endogenous glucose production [182, 183]. A basal hepatic insulin resistance index (endogenous glucose production × fasting plasma insulin) has been described [184]. The hyperinsulinemic euglycemic clamp is suitable for pairing with various complementary methods that permit insulin action to be studied under controlled conditions at the whole body, regional, or tissue level (Table  1.6). For example, a clamp study may be combined with indirect calorimetry to quantify whole-body substrate oxidation. While combining these two methods is conceptually intriguing and if done properly can generate useful data, managing this rather sophisticated experimental set up is demanding and requires a high level of expertise and experience. Using a combination of these approaches a study of the effects of pioglitazone in insulin-resistant women with polycystic ovary syndrome demonstrated that increases in circulating levels of the insulin-sensitizing cytokine adiponectin correlated closely with improvements in glucose and lipid oxidation as well as inversely with changes in fasting NEFA concentrations [185]. The investigators considered that their observations provided support for the hypothesis that improvements in multiple aspects of

21 Table 1.6  Complementary investigative methods that may be combined with the hyperinsulinemic euglycaemic clamp technique Method Isotopic glucose tracer Functional imaging, e.g. PET/MRI Indirect calorimetry Magnetic resonance spectroscopy Positron emission tomography Venous occlusion plethysmography Isotopic glycerol tracer Tissue biopsy (muscle, fat) Microdialysis

Measurement of interest Glucose turnover/endogenous glucose production and glucose disposal (RaG, RdG) Regional tissue-specific insulin action Substrate oxidation Intramyocellular lipid; hepatic lipid content Regional brain/heart glucose metabolism Endothelial function Lipolysis Insulin-responsive enzyme expression Adipose tissue substrate metabolism

PET positron emission tomography, MRI magnetic resonance imaging

insulin sensitivity with pioglitazone were at least partly explained by an increase in adiponectin levels [185]. The flexibility of the hyperinsulinemic glucose clamp also permits the design and execution of studies in which multiple target glucose levels can be achieved as required. This approach has been used, for example, to test new glucose sensors [186]. Assessment of regional tissue-specific insulinmediated glucose metabolism is possible by combining the hyperinsulinemic clamp with position emission tomography (PET) and magnetic resonance imaging (MRI) in an integrated approach [187].

Procedure for the Hyperinsulinemic Euglycemic Clamp This hyperinsulinemic euglycemic clamp, whether comprising a single step or two steps of steady-state hyperinsulinemia, is conducted in the morning after an overnight fast. An arm vein is cannulated for the infusion of insulin and glucose. In the contralateral forearm another cannula is placed for sampling, the hand being enclosed in a thermostat-controlled warming unit to open

22

arterio-venous channels and so provide ‘arterialized’ blood that approximates the arterial supply; this avoids the reduction in plasma glucose due to extraction during on passage of blood through insulin sensitive tissues and so avoids potential overestimation of apparent insulin action [148]. However, hand warming may induce peripheral vasodilatation resulting in a rise in heart rate and changes in blood pressure [188]. Soluble insulin or a rapid-acting insulin analog is infused using a precision pump at a rate calculated to elevate serum insulin concentration from basal levels to a pre-defined target within the euglycemic range, i.e. to approximately 100 mg/dL (approximately 5.6  mmol/L). A typical insulin infusion dose is 1.0  mU/kg/min (6  pmol/kg/min) or 40  mU/ min/m2 body surface area (0.24  nmol/min/m2). Insulin doses based on surface area are preferred for obese subjects (with body mass index >30 kg/m2) in order to avoid over-insulinization [148]. Infusing insulin at a rate of 1.0  mU/kg/ min (6 pmol/kg/min) generally provides plasma insulin levels in the high-physiological range, albeit with a wide range reported in the literature. Adsorption of insulin to the plastic surfaces of the infusion tubing can be avoided by adding 2 mL of the subject’s blood to the insulin/saline solution used for infusion [189]. The key outcome measure, i.e. glucose disposal rate (also known as the M-value), is approximately linear over the physiologic range of plasma insulin concentrations (Fig. 1.2). It has been suggested that very high doses of insulin, i.e. those producing pharmacological insulin concentrations of 200  mU/L (1200  pmol/L) or more, carry the risk of saturating the physiological clearance of insulin leading to unpredictable hyperinsulinemia. However, dose-response effects have been described using doses as high as 5.0  U/kg/min (30  pmol/kg/min) in insulinresistant subjects [190]. Some investigators advocate the use of a primed-continuous infusion of insulin to achieve the desired level of hyperinsulinemia more rapidly, accepting a temporary overshoot above target. While published algorithms for calculating glucose infusion rates are available [105] many investigators prefer to rely on their own judg-

A. J. Krentz et al.

ment based on experience – the so-called manual clamp. An alternative to the manual clamp is provided by automated or semi-automated clamp. technology. In brief, these methods utilize an algorithm to calculate the rate of glucose infusion [191]. A detailed discussion of manual vs. automated clamps is presented in Chap. 3.

Experimental Conditions for Hyperinsulinemic Euglycemic Clamp Studies If strictly defined, a true steady state of glucose infusion rates will usually not be achieved during a typical 120  min hyperinsulinemic euglycemic clamp [192]. When hyperinsulinemia extends beyond this time period insulin-mediated glucose uptake increases [193, 194]. However, for practical purposes data from the final 30–60 min of a 2–4-h glucose clamp study are generally accepted as being sufficiently stable for the determination of insulin-mediated glucose uptake [148, 189]. A more rigorous approach is to define steady state as a co-efficient of variation (CV) in blood glucose of band2=c(.1, .2, .25, .30, .35, .40, .45, .5, .6, .7, .8, .9) > quantile(prune, band2) 10% 20% 25% 30% 35% 40% 45% 50% 60% 70% 80% 90% -105 -95 -91 -87 -83 -80 -76 -73 -67 -60 -54 -47 Hounsfield unit levels at each decile are shown between −120 and −40 indicating that about half of the voxels fell between the 20th percentile (−95) and the 70th percentile (−60)

M. S. Buchsbaum and A. DeCastro

204 Mean plot of multiple variables Clusterspreadsheet1b.sta 45v*11c Median; whisker: non-outlier range 1.2

Glucose metabolic rate cold minus warm

1.0

0.8

0.6 Median Non-outlier range Raw data

0.4

0.2

0.0

–0.2

–0.4

Lcold-warm

Rcold-warm

Fig. 8.7  Cold minus warm metabolic rate in Hounsfield unit clusters in left and right supraclavicular area. Note that a value greater than 0 is cold activation. Seven of 11

subjects show cold activation, and 1 subject is an outlier in that direction. The median value is above the 0 line

priate unit estimation (Fig. 8.7). Next the number of mm3 between −120 and −40 Hounsfield units was evaluated and found to be 669,372 or about 19% of the volume to be in the fat range. Variability of methods, including Hounsfield unit windows, use of FDG-PET SUV thresholds, and lack of separation of results by thorax level make comparison to other reports uncertain.

mation to the legs as an approximation. Based on the location of the edges of the tibia and fibula on the CT, we stereotaxically located (Fig. 8.8) four spherical regions of interest (tibialis anterior, soleus, gastrocnemius, and subcutaneous fat) 8 mm in diameter (calling the R neuroim subroutine RegionSphere (FDG-PET-inputfile-name. nii, xyz-location, radius-mm)). These regions were chosen to differ in fast twitch (tibialis anterior), mixed (gastrocnemius), and slow twitch (soleus). Leg muscles were significantly more active in the warm condition than the cold condition (environmental temperature main effect F = 10.6, df = 1, 7, p = 0.014 in temperature condition by tissue type two-way repeated measures ANOVA). Muscles and subcutaneous fat had differing metabolic rates (two-way repeated measures ANOVA, main effect of tissue type F = 23.9, df = 3, 21, p 20% enrichment in cellular water in an organism), but these levels are more than an order of magnitude greater than is required or achieved for flux rate measurements in people and would never be achieved in a clinical setting [12]; in general the body water enrichments (% 2H2O) used range from 0.5% to 2% in clinical studies and 95% in skeletal muscle, can be identified in plasma in rodents and humans and that its labeling kinetics in plasma after heavy water administration are essentially identical to CK-M labeling from skeletal muscle sample [109]. Moreover, the FSR of CK-M in plasma or muscle tissue correlates closely with FSRs of other cytosolic and structural proteins from muscle,

235

Accordingly, measuring the labeling kinetics of plasma CK-M provides a minimally invasive “virtual biopsy” of skeletal muscle protein kinetics [110, 111]. Similarly, two proteins that are part of the tissue fibrogenic pathway  – lumican, which is involved in collagen fibril formation, and transforming growth factor-beta induced protein, which is upregulated with collagen synthesis in fibrogenic states – can be detected and measured kinetically by LC/MS in plasma in human subjects after heavy water labeling. In patients with chronic liver disease due to hepatitis C infection or with NAFLD/NASH who had concurrent liver biopsies the FSRs of these two proteins in blood correlates closely with the FSR of liver collagens. It is likely that other proteins in body fluids will also be identifiable, that derive from and thereby reflect the protein dynamics in an inaccessible tissue of interest, as many proteins appear to leak from tissues into the circulation in trace amounts.

 ass Isotopologue Measurements M Advances in high resolution mass spectrometry are also allowing the development of new analytical approaches that offer the potential for rapid and sensitive protein turnover quantification. Monitoring isotope incorporation at the amino acid fragment (immonium ion) isotopologue level, for example, may offer potential benefits including increased dynamic range for quantification of isotope label incorporation as well as increased sensitivity and specificity. The immonium ion is composed of the amino acid variable side chain and can be generated following peptide fragmentation in the gas phase during LC-MS/MS peptide analysis. The measurement of immonium ion isotopologue abundances generated from peptide fragments provides a potentially more sensitive means for quantifying protein fractional synthesis with less total isotope incorporation. The increase in sensitivity when measuring tandem mass spectrometry (MS) fragment immonium ion isotopologue spectrum as compared to peptide level MS spectra is largely due to low baseline isotopologue signal in unlabeled samples and a much larger changes in nor-

C. Beysen et al.

236

malized isotopologue abundances relative to changes expected and observed in isotopomer abundances (Fig.  9.7a and b). Isotopologue abundances of immonium ions exhibit much greater change in normalized abundance compared to matched isotopomer abundance profile following metabolic label incorporation (Fig. 9.8). Importantly, the 15N, 13C, and 2H containing immonium ion isotopologues (in the M1 mass range) comprise a limited number of molecular species that differ in mass by ~3 to ~6 millidaltons (Fig. 9.7a). In comparison, when monitoring peptide isotopomers it is typical to normalize a

abundances among a cluster of 4–5 isotopomers for each peptide. Error in measured peptide isotopomer abundances (poor signal to noise for any of the 4–5 isotopomers) coupled with differences between peptide enrichment models, leads to increased uncertainty in the resultant mean peptide level measurements representing a proteins fractional synthesis. In contrast, all immonium ions isotopologues derived from the same amino acid are represented by the same isotopologue distribution model, regardless of which peptide they are derived from. There are a limited number of immonium ion isotopologue abundance models generated and used to interpret measured

isotopologue

isotopomer

H

H N

H + H N

C

H

C

O

)

C

C

O

n

R

M1 M1

M2 M2

15N

M3

13C 2H 6 mDa 3 mDa

+1 neutron

0.6 0.5 0.4 0.3 0.2 0.1 1.00%

1.50%

2.00%

Deuterium Enrichment

Fig. 9.7  General model of the amino acid immonium ion and mass spectrum showing the M1 isotopologues and peptide isotopomer mass spectrum abundance profile (a). Modeled proline immonium ion isotopologue and isotopomer abundances reveal at 2% deuterium oxide enrich-

Change in isotopomer abundance (EMx)

0.7

0.50%

M4

400 - 1500 m/z

immonium ion Isotopologue Model (proline) Change in isotopologue abundance

N

O

M0

M0

0.0 0.00%

H

H

R

b

(

H

0.3

Peptide Isotopomer model (A*)

0.2 0.1 0.0

0.50%

1.00%

1.50%

2.00%

-0.1 -0.2 -0.3 -0.4 -0.5

Deuterium Enrichment

ment in the total body water the theoretical maximum change in M0 isotopomer from unlabeled baseline is a reduction of ~23% and the increase in 2H isotopologue abundance reaches ~60% from a baseline of ~2% (30 × change from baseline) (b)

9  Isotopic Tracers for the Measurement of Metabolic Flux Rates

immonium ion isotopologue abundances. The reduction in the number of models and the ability to use the same enrichment models across all peptide fragments from the same protein and for all proteins present in the sample represents a significant advantage for data analytics and interpretation when monitoring protein fractional synthesis. Collectively this approach may enable monitoring of protein turnover with less deuterium oxide exposure and reduction in total label incorporation and therefore reduce labeling times for clinical protocols. Monitoring protein turnover measuring isotopologue abundance provides an additional analytical advantage. The immonium ion 15N to 13C isotopologue ratio is predictable and constant regardless of 2H incorporation and represents a measurement reflecting instrumental accuracy on a scan by scan basis. Deviation from theoretical 15 N/13C indicates unexpected abundance in either the 13C or the 15N isotopologue signal due to poor signal to noise, peak coalescence or presence of contaminating molecular species. Given there is an equal likelihood of error being derived from 13 C or 15N abundances removing scans based on this error analysis represents a strategy for data quality control and management. The 15N/13C

ratio can thus be used as a built-in known reference signal for spectral quality control and this is unique for this approach when analyzing stable isotope labeled samples by mass spectrometry. Quantification of protein turnover monitoring immonium ion isotopologue abundance enables reduced time of stable isotope tracer exposure, as compared to peptide level isotopomer abundances (Fig. 9.8). This approach addresses a fundamental limitation and challenges when utilizing deuterium oxide as a universal stable isotope precursor. The heavy water labeling protocol when measuring isotopologue abundances is not different than when measuring mass isotopmer abundances.

Advantages/Disadvantages The turnover rates of both untargeted proteins across the proteome and targeted proteins can be determined using deuterated water labeling coupled with LC-MS/MS sample analysis. The protocol for patients is straightforward and non-invasive. When using deuterated water for protein synthesis measurements either at the peptide level (MS) or peptide fragment level (MS/

Isotopomer signal M0

M2

M1

237

M1 Isotopologue signal M3

13C

2H

Relative Abundance

Day 1

Day 3

Day 7

Day 14

Day 21 m/z

Fig. 9.8  Shown are the isotopomer (left) and M1 13C and 2 H proline immonium ion isotopologue abundance profiles (right) for a selected peptide with sequence VCLLHEKTPVSEHVTK.  The isotopologue spectra (right) provides a clear demonstration of larger normal-

m/z

ized signal and greater sensitivity for quantification of isotope labelling when compared to the changes in M0 isotopomer abundance observed reflecting new protein synthesis

C. Beysen et al.

238

MS), however, it is important to correct for any residual label in the molecule of interests when repeated testing is performed [39].

 ranslating Results into Drug T Development and Clinical Practice The synthesis and degradation of proteins is critical to maintain cell viability, and alterations in protein homeostasis have been implicated in many diseases as well as the normal decline in cellular functions with age. Much of the early work studying protein turnover focused on protein balance in the whole body, with an emphasis on the assessment of adequate protein intake. Because of the ease of administration of deuterated water for long periods of time and the small physical sample requirements for LC-MS/MS analyses, large-scale studies evaluating changes in proteome turnover in cardiometabolic diseases are now possible. The application of LC-MS/MS to tracer studies of protein metabolism has several important applications to the study of metabolic disease and development of new therapeutics. Loss of skeletal muscle mass with aging, for example, is a significant contributor to morbidity and mortality, particularly in patients with metabolic disease. The continued synthesis of skeletal muscle proteins is critical for maintenance of skeletal muscle mass; moreover, interventions aimed at improving metabolic health have significant beneficial impact on muscle mass and function. Examination of the dynamics of muscle proteome with deuterated labeling and LC-MS/MS can determine the fractional synthesis rate of glycolytic, mitochondrial, contractile, and endoplasmic reticulum proteins from a single 10-mg sample [24, 25]. The fractional synthesis rate of muscle proteins have been shown to be responsive to sprint-interval exercise in human [112] and has been reported to be predictive of changes in muscle mass several weeks later in a pre-­ clinical setting [113]. Additionally, the fractional synthesis rate of plasma creatine kinase M-type (CK-M) and carbonic anhydrase 3

(CA-3) have been validated as minimally invasive translational biomarkers for skeletal muscle protein synthesis rates in a clinical setting [112]. A similar virtual biopsy approach has been described for measuring hepatic fibrogenesis in blood by measuring the fractional synthesis rates of extracellular matrix– related proteins (such as lumican or TGFβ–induced protein) that escape from liver into the circulation [114]. In patients with NASH, for example, plasma lumican turnover correlates with hepatic collagen fractional synthesis, suggesting its applications as a biomarker for the measurement of hepatic fibrogenesis in this clinical setting [114]. LC-MS/MS analysis of isotope enrichment in proteins also allows the simultaneous assessment of protein composition and turnover rates on lipoprotein particles [115, 116]. It is now well established that HDL, for example, can carry more than 100 different proteins which presumably impact its function as an anti-atherosclerotic particle [117–119]. Improved understanding of the relationship between the HDL-associated proteins and atherosclerotic risk will be valuable. Advances outlined above in high resolution mass spectrometry are enabling the development of an analytical approach, that offer the potential for rapid and simplified protein turnover quantification. Monitoring isotope incorporation at the amino acid fragment (immonium ion) isotopologue level offers potential benefits including increased dynamic range for quantification of isotope label incorporation as well as increased sensitivity and specificity.

Conclusions While still early in their development, the use of proteomic platforms with isotope labeling and the use of virtual biopsies has the potential to significantly impact drug development by better understanding the changes in protein fluxes which are induced by disease and restored with effective therapies.

9  Isotopic Tracers for the Measurement of Metabolic Flux Rates

References 1. Buchanan JM. Biochemistry during the life and times of Hans Krebs and Fritz Lipmann. J Biol Chem. 2002;277(37):33531–6. 2. Waterlow JC, editor. Protein turnover [Internet]. Wallingford: CABI; 2006 [cited 2018 May 26]. Available from: http://www.cabi.org/cabebooks/ ebook/20063115785. 3. Wolfe RR, Chinkes DL. Isotope tracers in metabolic research: principles and practice of kinetic analysis. 2nd ed. Hoboken: Wiley-Liss; 2004. p. 488. 4. Schoenheimer R, Rittenberg D. The study of intermediary metabolism of animals with the aid of isotopes. Physiol Rev. 1940;20(2):218–48. 5. Hellerstein MK, Murphy E. Stable isotope-mass spectrometric measurements of molecular fluxes in  vivo: emerging applications in drug development. Curr Opin Mol Ther. 2004;6(3):249–64. 6. Turner SM, Hellerstein MK.  Emerging applications of kinetic biomarkers in preclinical and clinical drug development. Curr Opin Drug Discov Devel. 2005;8(1):115–26. 7. Klein PD, Klein ER.  Stable isotopes: origins and safety. J Clin Pharmacol. 1986;26(6):378–82. 8. Koletzko B, Sauerwald T, Demmelmair H.  Safety of stable isotope use. Eur J Pediatr. 1997;156(Suppl 1):S12–7. 9. Koletzko B, Demmelmair H, Hartl W, Kindermann A, Koletzko S, Sauerwald T, et  al. The use of stable isotope techniques for nutritional and metabolic research in paediatrics. Early Hum Dev. 1998;53(Suppl):S77–97. 10. Dufner D, Previs SF. Measuring in vivo metabolism using heavy water. Curr Opin Clin Nutr Metab Care. 2003;6(5):511–7. 11. Rachdaoui N, Austin L, Kramer E, Previs MJ, Anderson VE, Kasumov T, et  al. Measuring proteome dynamics in  vivo: as easy as adding water? Mol Cell Proteomics. 2009;8(12):2653–63. 12. Jones PJ, Leatherdale ST.  Stable isotopes in clinical research: safety reaffirmed. Clin Sci. 1991;80(4):277–80. 13. Di Buono M, Jones PJ, Beaumier L, Wykes LJ.  Comparison of deuterium incorporation and mass isotopomer distribution analysis for measurement of human cholesterol biosynthesis. J Lipid Res. 2000;41(9):1516–23. 14. Beysen C, Murphy EJ, McLaughlin T, Riiff T, Lamendola C, Turner HC, et  al. Whole-body glycolysis measured by the deuterated-glucose disposal test correlates highly with insulin resistance in vivo. Diabetes Care. 2007;30(5):1143–9. 15. Beysen C, Murphy EJ, Deines K, Chan M, Tsang E, Glass A, et al. Effect of bile acid sequestrants on glucose metabolism, hepatic de novo lipogenesis, and cholesterol and bile acid kinetics in type 2 diabetes: a randomised controlled study. Diabetologia. 2012;55(2):432–42.

239

16. Beysen C, Murphy EJ, Nagaraja H, Decaris M, Riiff T, Fong A, et al. A pilot study of the effects of pioglitazone and rosiglitazone on de novo lipogenesis in type 2 diabetes. J Lipid Res. 2008;49(12):2657–63. 17. Borén J, Taskinen M-R, Adiels M.  Kinetic studies to investigate lipoprotein metabolism. J Intern Med. 2012;271(2):166–73. 18. Strawford A, Antelo F, Christiansen M, Hellerstein MK.  Adipose tissue triglyceride turnover, de novo lipogenesis, and cell proliferation in humans measured with 2H2O. Am J Physiol Endocrinol Metab. 2004;286(4):E577–88. 19. Previs SF, McLaren DG, Wang S-P, Stout SJ, Zhou H, Herath K, et  al. New methodologies for studying lipid synthesis and turnover: looking backwards to enable moving forwards. Biochim Biophys Acta. 2014;1842(3):402–13. 20. Barrett PHR, Chan DC, Watts GF. Thematic review series: patient-oriented research. Design and analysis of lipoprotein tracer kinetics studies in humans. J Lipid Res. 2006;47(8):1607–19. 21. Chan DC, Barrett PHR, Watts GF.  Recent studies of lipoprotein kinetics in the metabolic syndrome and related disorders. Curr Opin Lipidol. 2006;17(1):28–36. 22. White UA, Fitch MD, Beyl RA, Hellerstein MK, Ravussin E.  Association of in  vivo adipose tissue cellular kinetics with markers of metabolic health in humans. J Clin Endocrinol Metab. 2017;102(7):2171–8. 23. White UA, Fitch MD, Beyl RA, Hellerstein MK, Ravussin E. Differences in in vivo cellular kinetics in abdominal and femoral subcutaneous adipose tissue in women. Diabetes. 2016;65(6):1642–7. 24. Gasier HG, Fluckey JD, Previs SF. The application of 2H2O to measure skeletal muscle protein synthesis. Nutr Metab (Lond). 2010;7:31. 25. Robinson MM, Turner SM, Hellerstein MK, Hamilton KL, Miller BF. Long-term synthesis rates of skeletal muscle DNA and protein are higher during aerobic training in older humans than in sedentary young subjects but are not altered by protein supplementation. FASEB J. 2011;25(9):3240–9. 26. Scalzo RL, Peltonen GL, Binns SE, Shankaran M, Giordano GR, Hartley DA, et al. Greater muscle protein synthesis and mitochondrial biogenesis in males compared with females during sprint interval training. FASEB J. 2014;28(6):2705–14. 27. Harwood HJ.  Treating the metabolic syndrome: acetyl-­ CoA carboxylase inhibition. Expert Opin Ther Targets. 2005;9(2):267–81. 28. Hellerstein MK, Schwarz JM, Neese RA. Regulation of hepatic de novo lipogenesis in humans. Annu Rev Nutr. 1996;16:523–57. 29. Lambert JE, Ramos-Roman MA, Browning JD, Parks EJ. Increased de novo lipogenesis is a distinct characteristic of individuals with nonalcoholic fatty liver disease. Gastroenterology. 2014;146(3):726–35. 30. Tuvdendorj D, Chandalia M, Batbayar T, Saraf M, Beysen C, Murphy EJ, et  al. Altered subcutaneous

240 abdominal adipose tissue lipid synthesis in obese, insulin-resistant humans. Am J Physiol Endocrinol Metab. 2013;305(8):E999–1006. 31. Hudgins LC, Parker TS, Levine DM, Hellerstein MK. A dual sugar challenge test for lipogenic sensitivity to dietary fructose. J Clin Endocrinol Metab. 2011;96(3):861–8. 32. Stanhope KL, Griffen SC, Bremer AA, Vink RG, Schaefer EJ, Nakajima K, et al. Metabolic responses to prolonged consumption of glucose- and fructose-­ sweetened beverages are not associated with postprandial or 24-h glucose and insulin excursions. Am J Clin Nutr. 2011;94(1):112–9. 33. Cramer CT, Goetz B, Hopson KLM, Fici GJ, Ackermann RM, Brown SC, et al. Effects of a novel dual lipid synthesis inhibitor and its potential utility in treating dyslipidemia and metabolic syndrome. J Lipid Res. 2004;45(7):1289–301. 34. Leitch CA, Jones PJ. Measurement of human lipogenesis using deuterium incorporation. J Lipid Res. 1993;34(1):157–63. 35. Hellerstein MK, Christiansen M, Kaempfer S, Kletke C, Wu K, Reid JS, et al. Measurement of de novo hepatic lipogenesis in humans using stable isotopes. J Clin Invest. 1991;87(5):1841–52. 36. Schoenheimer R.  The dynamic state of body constituents [Internet]. 2nd ed. Cambridge: Mass Harvard University Press; 1946. [cited 2018 May 26]. Available from: https://trove.nla.gov.au/ work/14374148. 37. Busch R, Neese RA, Awada M, Hayes GM, Hellerstein MK.  Measurement of cell proliferation by heavy water labeling. Nat Protoc. 2007;2(12):3045–57. 38. Price JC, Holmes WE, Li KW, Floreani NA, Neese RA, Turner SM, et  al. Measurement of human plasma proteome dynamics with (2)H(2)O and liquid chromatography tandem mass spectrometry. Anal Biochem. 2012;420(1):73–83. 39. Holmes WE, Angel TE, Li KW, Hellerstein MK.  Dynamic proteomics: in  vivo proteome-wide measurement of protein kinetics using metabolic labeling. Methods Enzymol. 2015;561:219–76. 40. Lawitz EJ, Coste A, Poordad F, Alkhouri N, Loo N, McColgan BJ, et al. Acetyl-CoA carboxylase inhibitor GS-0976 for 12 weeks reduces hepatic De novo lipogenesis and steatosis in patients with nonalcoholic steatohepatitis. Clin Gastroenterol Hepatol. 2018;16(12):1983–91. 41. Hellerstein MK, Neese RA.  Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations. Am J Phys. 1999;276(6 Pt 1):E1146–70. 42. Chinkes DL, Aarsland A, Rosenblatt J, Wolfe RR. Comparison of mass isotopomer dilution methods used to compute VLDL production in vivo. Am J Phys. 1996;271(2 Pt 1):E373–83. 43. Bederman IR, Reszko AE, Kasumov T, David F, Wasserman DH, Kelleher JK, et al. Zonation of labeling of lipogenic acetyl-CoA across the liver: implications for studies of lipogenesis by mass isotopomer analysis. J Biol Chem. 2004;279(41):43207–16.

C. Beysen et al. 44. Vedala A, Wang W, Neese RA, Christiansen MP, Hellerstein MK. Delayed secretory pathway contributions to VLDL-triglycerides from plasma NEFA, diet, and de novo lipogenesis in humans. J Lipid Res. 2006;47(11):2562–74. 45. Faix D, Neese R, Kletke C, Wolden S, Cesar D, Coutlangus M, et  al. Quantification of menstrual and diurnal periodicities in rates of cholesterol and fat synthesis in humans. J Lipid Res. 1993;34(12):2063–75. 46. Siler SQ, Neese RA, Hellerstein MK. De novo lipogenesis, lipid kinetics, and whole-body lipid balances in humans after acute alcohol consumption. Am J Clin Nutr. 1999;70(5):928–36. 47. Beysen C, Ruddy M, Stoch A, Mixson LA, Rosko K, Riiff T, et al. Dose-dependent quantitative effects of acute fructose administration on hepatic de novo lipogenesis in healthy humans. Am J Physiol Endocrinol Metab. 2018;315(1):E126–32. 48. Stiede K, Miao W, Blanchette HS, Beysen C, Harriman G, Harwood HJ, et  al. Acetyl-coenzyme a carboxylase inhibition reduces de novo lipogenesis in overweight male subjects: a randomized, double-blind, crossover study. Hepatology. 2017;66(2):324–34. 49. Griffith DA, Kung DW, Esler WP, Amor PA, Bagley SW, Beysen C, et  al. Decreasing the rate of metabolic ketone reduction in the discovery of a clinical acetyl-CoA carboxylase inhibitor for the treatment of diabetes. J Med Chem. 2014;57(24):10512–26. 50. Stanhope KL, Schwarz JM, Keim NL, Griffen SC, Bremer AA, Graham JL, et al. Consuming fructose-­ sweetened, not glucose-sweetened, beverages increases visceral adiposity and lipids and decreases insulin sensitivity in overweight/obese humans. J Clin Invest. 2009;119(5):1322–34. 51. Flannery C, Dufour S, Rabøl R, Shulman GI, Petersen KF.  Skeletal muscle insulin resistance promotes increased hepatic de novo lipogenesis, hyperlipidemia, and hepatic steatosis in the elderly. Diabetes. 2012;61(11):2711–7. 52. Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ. Sources of fatty acids stored in liver and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest. 2005;115(5):1343–51. 53. Kim C-W, Addy C, Kusunoki J, Anderson NN, Deja S, Fu X, et  al. Acetyl CoA carboxylase inhibition reduces hepatic steatosis but elevates plasma triglycerides in mice and humans: a bedside to bench investigation. Cell Metab. 2017;26(2):394–406.e6. 54. Goedeke L, Bates J, Vatner DF, Perry RJ, Wang T, Ramirez R, et al. Acetyl-CoA Carboxylase inhibition reverses nafld and hepatic insulin resistance but promotes hypertriglyceridemia in rodents. Hepatology. 2018;68(6):2197–211. 55. Freckmann G, Hagenlocher S, Baumstark A, Jendrike N, Gillen RC, Rössner K, et al. Continuous glucose profiles in healthy subjects under everyday life conditions and after different meals. J Diabetes Sci Technol. 2007;1(5):695–703.

9  Isotopic Tracers for the Measurement of Metabolic Flux Rates 56. Steele R, Wall JS, De Bodo RC, Altszuler N.  Measurement of size and turnover rate of body glucose pool by the isotope dilution method. Am J Phys. 1956;187(1):15–24. 57. Gerich JE, Meyer C, Woerle HJ, Stumvoll M. Renal gluconeogenesis: its importance in human glucose homeostasis. Diabetes Care. 2001;24(2):382–91. 58. Katz J, Rognstad R. Futile cycles in the metabolism of glucose. Curr Top Cell Regul. 1976;10:237–89. 59. Rigalleau V, Beylot M, Laville M, Guillot C, Deleris G, Aubertin J, et al. Measurement of post-absorptive glucose kinetics in non-insulin-dependent diabetic patients: methodological aspects. Eur J Clin Investig. 1996;26(3):231–6. 60. Glauber H, Wallace P, Brechtel G. Effects of fasting on plasma glucose and prolonged tracer measurement of hepatic glucose output in NIDDM. Diabetes. 1987;36(10):1187–94. 61. Chen YD, Swislocki AL, Jeng CY, Juang JH, Reaven GM.  Effect of time on measurement of hepatic glucose production. J Clin Endocrinol Metab. 1988;67(5):1084–8. 62. Hovorka R, Eckland DJ, Halliday D, Lettis S, Robinson CE, Bannister P, et  al. Constant infusion and bolus injection of stable-label tracer give reproducible and comparable fasting HGO.  Am J Phys. 1997;273(1 Pt 1):E192–201. 63. Barrett PH, Bell BM, Cobelli C, Golde H, Schumitzky A, Vicini P, et al. SAAM II: simulation, analysis, and modeling software for tracer and pharmacokinetic studies. Metab Clin Exp. 1998;47(4):484–92. 64. Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care. 1999;22(9):1462–70. 65. Hellerstein MK, Neese RA, Linfoot P, Christiansen M, Turner S, Letscher A.  Hepatic gluconeogenic fluxes and glycogen turnover during fasting in humans. A stable isotope study. J Clin Invest. 1997;100(5):1305–19. 66. Neese RA, Schwarz J-M, Faix D, Turner S, Letscher A, Vu D, et  al. Gluconeogenesis and intrahepatic triose phosphate flux in response to fasting or substrate loads. Application of the mass isotopomer distribution analysis technique with testing of assumptions and potential problems. J Biol Chem. 1995;270(24):14452–63. 67. Hellerstein MK, Neese RA.  Mass isotopomer distribution analysis: a technique for measuring biosynthesis and turnover of polymers. Am J Phys. 1992;263(5 Pt 1):E988–1001. 68. Hellerstein MK, Kaempfer S, Reid JS, Wu K, Shackleton CH.  Rate of glucose entry into hepatic uridine diphosphoglucose by the direct pathway in fasted and fed states in normal humans. Metab Clin Exp. 1995;44(2):172–82. 69. Basu R, Basu A, Johnson CM, Schwenk WF, Rizza RA.  Insulin dose-response curves for stimulation of splanchnic glucose uptake and suppression of endogenous glucose production differ in nondiabetic humans and are abnormal in people with type 2 diabetes. Diabetes. 2004;53(8):2042–50.

241

70. Steele R, Bishop JS, Dunn A, Altszuler N, Rathbeb I, Debodo RC. Inhibition by insulin of hepatic glucose production in the normal dog. Am J Phys. 1965;208:301–6. 71. Cowan JS, Hetenyi G Jr. Glucoregulatory responses in normal and diabetic dogs recorded by a new tracer method. Metab Clin Exp. 1971;20(4):360–72. 72. Hother-Nielsen O.  On the appropriate use of the primed-constant tracer infusion technique. Diabete Metab. 1994;20(6):568–70. 73. Staehr P, Hojlund K, Hother-Nielsen O, Holst JJ, Beck-Nielsen H.  Does overnight normalization of plasma glucose by insulin infusion affect assessment of glucose metabolism in type 2 diabetes? Diabet Med. 2003;20(10):816–22. 74. Steele R, Bjerknes C, Rathgeb I, Altszuler N.  Glucose uptake and production during the oral glucose tolerance test. Diabetes. 1968;17(7):415–21. 75. Mari A, Wahren J, DeFronzo RA, Ferrannini E.  Glucose absorption and production following oral glucose: comparison of compartmental and arteriovenous-difference methods. Metab Clin Exp. 1994;43(11):1419–25. 76. Dalla Man C, Caumo A, Basu R, Rizza R, Toffolo G, Cobelli C.  Minimal model estimation of glucose absorption and insulin sensitivity from oral test: validation with a tracer method. Am J Physiol Endocrinol Metab. 2004;287(4):E637–43. 77. Taylor R, Magnusson I, Rothman DL, Cline GW, Caumo A, Cobelli C, et  al. Direct assessment of liver glycogen storage by 13C nuclear magnetic resonance spectroscopy and regulation of glucose homeostasis after a mixed meal in normal subjects. J Clin Invest. 1996;97(1):126–32. 78. Basu R, Di Camillo B, Toffolo G, Basu A, Shah P, Vella A, et al. Use of a novel triple-tracer approach to assess postprandial glucose metabolism. Am J Physiol Endocrinol Metab. 2003;284(1):E55–69. 79. Haidar A, Elleri D, Allen JM, Harris J, Kumareswaran K, Nodale M, et  al. Validity of triple- and dual-tracer techniques to estimate glucose appearance. Am J Physiol Endocrinol Metab. 2012;302(12):E1493–501. 80. Rossetti L, Giaccari A. Relative contribution of glycogen synthesis and glycolysis to insulin-mediated glucose uptake. A dose-response euglycemic clamp study in normal and diabetic rats. J Clin Invest. 1990;85(6):1785–92. 81. Woerle HJ, Meyer C, Dostou JM, Gosmanov NR, Islam N, Popa E, et  al. Pathways for glucose disposal after meal ingestion in humans. Am J Physiol Endocrinol Metab. 2003;284(4):E716–25. 82. Del Prato S, Bonadonna RC, Bonora E, Gulli G, Solini A, Shank M, et  al. Characterization of cellular defects of insulin action in type 2 (non-­ insulin-­dependent) diabetes mellitus. J Clin Invest. 1993;91(2):484–94. 83. Galgani JE, Ravussin E.  Postprandial whole-body glycolysis is similar in insulin-resistant and insulin-­ sensitive non-diabetic humans. Diabetologia. 2012;55(3):737–42.

242 84. Christiansen MP, Linfoot PA, Neese RA, Hellerstein MK. Effect of dietary energy restriction on glucose production and substrate utilization in type 2 diabetes. Diabetes. 2000;49(10):1691–9. 85. Sherifali D, Nerenberg K, Pullenayegum E, Cheng JE, Gerstein HC. The effect of oral antidiabetic agents on A1C levels. Diabetes Care. 2010;33(8):1859–64. 86. Goldberg RB, Kendall DM, Deeg MA, Buse JB, Zagar AJ, Pinaire JA, et  al. A comparison of lipid and glycemic effects of pioglitazone and rosiglitazone in patients with type 2 diabetes and dyslipidemia. Diabetes Care. 2005;28(7):1547–54. 87. Nissen SE, Wolski K.  Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med. 2007;356(24):2457–71. 88. Erdmann E, Dormandy JA, Charbonnel B, Massi-­ Benedetti M, Moules IK, Skene AM, et  al. The effect of pioglitazone on recurrent myocardial infarction in 2,445 patients with type 2 diabetes and previous myocardial infarction: results from the PROactive (PROactive 05) study. J Am Coll Cardiol. 2007;49(17):1772–80. 89. Merovci A, Solis-Herrera C, Daniele G, Eldor R, Fiorentino TV, Tripathy D, et  al. Dapagliflozin improves muscle insulin sensitivity but enhances endogenous glucose production. J Clin Invest. 2014;124(2):509–14. 90. Ferrannini E, Muscelli E, Frascerra S, Baldi S, Mari A, Heise T, et  al. Metabolic response to sodium-­ glucose cotransporter 2 inhibition in type 2 diabetic patients. J Clin Invest. 2014;124(2):499–508. 91. Foster DM, Barrett PH, Toffolo G, Beltz WF, Cobelli C. Estimating the fractional synthetic rate of plasma apolipoproteins and lipids from stable isotope data. J Lipid Res. 1993;34(12):2193–205. 92. Kasumov T, Willard B, Li L, Li M, Conger H, Buffa JA, et al. 2H2O-based high-density lipoprotein turnover method for the assessment of dynamic high-­ density lipoprotein function in mice. Arterioscler Thromb Vasc Biol. 2013;33(8):1994–2003. 93. Welty FK, Lichtenstein AH, Barrett PH, Dolnikowski GG, Schaefer EJ.  Human apolipoprotein (Apo) B-48 and ApoB-100 kinetics with stable isotopes. Arterioscler Thromb Vasc Biol. 1999;19(12):2966–74. 94. Lichtenstein AH, Cohn JS, Hachey DL, Millar JS, Ordovas JM, Schaefer EJ.  Comparison of deuterated leucine, valine, and lysine in the measurement of human apolipoprotein A-I and B-100 kinetics. J Lipid Res. 1990;31(9):1693–701. 95. Wong ATY, Chan DC, Pang J, Watts GF, Barrett PHR. Plasma apolipoprotein B-48 transport in obese men: a new tracer kinetic study in the postprandial state. J Clin Endocrinol Metab. 2014;99(1):E122–6. 96. Berthold HK, Mertens J, Birnbaum J, Brämswig S, Sudhop T, Barrett PHR, et  al. Influence of simvastatin on apoB-100 secretion in non-obese subjects with mild hypercholesterolemia. Lipids. 2010;45(6):491–500.

C. Beysen et al. 97. Berglund L, Witztum JL, Galeano NF, Khouw AS, Ginsberg HN, Ramakrishnan R.  Three-fold effect of lovastatin treatment on low density lipoprotein metabolism in subjects with hyperlipidemia: increase in receptor activity, decrease in apoB production, and decrease in particle affinity for the receptor. Results from a novel triple-tracer approach. J Lipid Res. 1998;39(4):913–24. 98. Parhofer KG, Barrett PHR. Thematic review series: patient-oriented research. What we have learned about VLDL and LDL metabolism from human kinetics studies. J Lipid Res. 2006;47(8):1620–30. 99. Telford DE, Sutherland BG, Edwards JY, Andrews JD, Barrett PHR, Huff MW.  The molecular mechanisms underlying the reduction of LDL apoB-­ 100 by ezetimibe plus simvastatin. J Lipid Res. 2007;48(3):699–708. 100. Ginsberg HN. Changes in lipoprotein kinetics during therapy with fenofibrate and other fibric acid derivatives. Am J Med. 1987;83(5B):66–70. 101. Watts GF, Barrett PHR, Ji J, Serone AP, Chan DC, Croft KD, et  al. Differential regulation of lipoprotein kinetics by atorvastatin and fenofibrate in subjects with the metabolic syndrome. Diabetes. 2003;52(3):803–11. 102. Lamon-Fava S, Diffenderfer MR, Barrett PHR, Buchsbaum A, Nyaku M, Horvath KV, et  al. Extended-release niacin alters the metabolism of plasma apolipoprotein (Apo) A-I and ApoB-­ containing lipoproteins. Arterioscler Thromb Vasc Biol. 2008;28(9):1672–8. 103. Parhofer KG, Hugh P, Barrett R, Bier DM, Schonfeld G. Determination of kinetic parameters of apolipoprotein B metabolism using amino acids labeled with stable isotopes. J Lipid Res. 1991;32(8):1311–23. 104. Reeds PJ, Hachey DL, Patterson BW, Motil KJ, Klein PD. VLDL apolipoprotein B-100, a potential indicator of the isotopic labeling of the hepatic protein synthetic precursor pool in humans: studies with multiple stable isotopically labeled amino acids. J Nutr. 1992;122(3):457–66. 105. Bilheimer DW, Grundy SM, Brown MS, Goldstein JL.  Mevinolin and colestipol stimulate receptor-mediated clearance of low density lipoprotein from plasma in familial hypercholesterolemia heterozygotes. Proc Natl Acad Sci U S A. 1983;80(13):4124–8. 106. Reyes-Soffer G, Moon B, Hernandez-Ono A, Dionizovik-Dimanovski M, Dionizovick-­ Dimanovski M, Jimenez J, et al. Complex effects of inhibiting hepatic apolipoprotein B100 synthesis in humans. Sci Transl Med. 2016;8(323):323ra12. 107. Huff MW, Hegele RA. Apolipoprotein C-III: going back to the future for a lipid drug target. Circ Res. 2013;112(11):1405–8. 108. Graham MJ, Lee RG, Bell TA, Fu W, Mullick AE, Alexander VJ, et  al. Antisense oligonucleotide inhibition of apolipoprotein C-III reduces plasma triglycerides in rodents, nonhuman primates, and humans. Circ Res. 2013;112(11):1479–90.

9  Isotopic Tracers for the Measurement of Metabolic Flux Rates 109. Hellerstein M, Evans W.  Recent advances for measurement of protein synthesis rates, use of the “virtual biopsy” approach, and measurement of muscle mass. Curr Opin Clin Nutr Metab Care. 2017;20(3):191–200. 110. Stimpson SA, Leonard MS, Clifton LG, Poole JC, Turner SM, Shearer TW, et al. Longitudinal changes in total body creatine pool size and skeletal muscle mass using the D3-creatine dilution method. J Cachexia Sarcopenia Muscle. 2013;4(3):217–23. 111. Clark RV, Walker AC, O’Connor-Semmes RL, Leonard MS, Miller RR, Stimpson SA, et al. Total body skeletal muscle mass: estimation by creatine (methyl-d3) dilution in humans. J Appl Physiol. 2014;116(12):1605–13. 112. Shankaran M, King CL, Angel TE, Holmes WE, Li KW, Colangelo M, et al. Circulating protein synthesis rates reveal skeletal muscle proteome dynamics. J Clin Invest. 2016;126(1):288–302. 113. Shankaran M, Shearer TW, Stimpson SA, Turner SM, King C, Wong P-YA, et  al. Proteome-wide muscle protein fractional synthesis rates predict muscle mass gain in response to a selective androgen receptor modulator in rats. Am J Physiol Endocrinol Metab. 2016;310(6):E405–17. 114. Decaris ML, Li KW, Emson CL, Gatmaitan M, Liu S, Wang Y, et al. Identifying nonalcoholic fatty liver disease patients with active fibrosis by measuring

243

extracellular matrix remodeling rates in tissue and blood. Hepatology. 2017;65(1):78–88. 115. Turner S, Holmes W, Czerwieniec G, Tsang E, Dalidd M, Chen M, et  al. Abstract 587: dynamics of the apolipoproteome: turnover of HDLand LDL-associated apolipoproteins by LC/MS/ MS. Arterioscler Thromb Vasc Biol. 2014;34(Suppl 1):A587. 116. Croyal M, Fall F, Ferchaud-Roucher V, Chétiveaux M, Zaïr Y, Ouguerram K, et  al. Multiplexed peptide analysis for kinetic measurements of major human apolipoproteins by LC/MS/MS. J Lipid Res. 2016;57(3):509–15. 117. Heinecke JW. The HDL proteome: a marker – and perhaps mediator  – of coronary artery disease. J Lipid Res. 2009;50(Suppl):S167–71. 118. Vaisar T, Pennathur S, Green PS, Gharib SA, Hoofnagle AN, Cheung MC, et  al. Shotgun proteomics implicates protease inhibition and complement activation in the antiinflammatory properties of HDL.  J Clin Invest. 2007;117(3):746–56. 119. Shah AS, Tan L, Long JL, Davidson WS. Proteomic diversity of high density lipoproteins: our emerging understanding of its importance in lipid transport and beyond. J Lipid Res. 2013;54(10):2575–85.

Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other Metabolic Disorders

10

Andrew J. Krentz and Pierre Bedossa

Summary Background Tissue biopsy permits assessment of histology, disease staging, gene expression and quantification of metabolic processes and so may be a relevant outcome measure to understand tissue-level pharmacokinetic/pharmacodynamic responses to a therapeutic intervention. Biopsy of key metabolically active tissues, i.e. skeletal muscle, adipose tissue and liver, have informed scientific knowledge of the pathophysiology of diabetes and related metabolic diseases. Moreover, the results of tissue biopsy can usefully inform the development of new drugs for diabetes, obesity and nonalcoholic liver disease (NAFLD). Liver biopsy has a central role in the diagnosis and staging of NAFLD. Confirmation of nonalcoholic steatohepatitis (NASH) also rests on

A. J. Krentz (*) ProSciento, Chula Vista, CA, USA Institute for Cardiovascular & Metabolic Research, University of Reading, Reading, UK e-mail: [email protected]

liver biopsy. In terms of assessing pharmacotherapies for NAFLD/NASH, liver histology obtained through paired biopsies is currently the surrogate primary efficacy endpoint required in registrational clinical trials. Liver histology is also the gold standard for non-invasive NAFLD/ NASH biomarker discovery and validation. However, liver biopsy also has well-recognised limitations.

Key Methods Percutaneous liver biopsy should be performed if necessary by an experienced operator under ultrasound (US) guidance. Procedure can be unpleasant. Care must be taken to identify and minimise potential risks. Needle biopsy of skeletal muscle and adipose tissue are generally safe and well-tolerated.

P. Bedossa Liverpat, Paris, France Institute of Cellular Medicine, University of Newcastle, Newcastle upon Tyne, UK

© Springer Nature Switzerland AG 2019 A. J. Krentz et al. (eds.), Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease, https://doi.org/10.1007/978-3-030-11748-1_10

245

A. J. Krentz and P. Bedossa

246

Method Percutaneous ultrasound (US)-guided needle biopsy under local anesthesia; alternatives include biopsy during bariatric surgery, laparoscopic and transjugular approaches

Measurement Histological examination of liver samples

Advantages Biopsy – usually percutaneous in the context of clinical trials – is the definitive method for diagnosis and staging of nonalcoholic fatty steatohepatitis (NASH)

Skeletal muscle biopsy

Modified Bergström needle biopsy under local anesthesia or intraoperatively, e.g. during bariatric surgery

Morphology including fiber type; gene expression; enzyme activity; omics; culture of myocytes and myotubes

Adipose tissue biopsy

Subcutaneous percutaneous biopsy or suction needle biopsy under local anesthesia; visceral adipocytes obtained during abdominal surgery

Morphology including adipocyte size, macrophage infiltration; gene expression; omics; cell culture

Quantitative in vitro techniques can be used to assess impact of therapeutic interventions on muscle metabolism; drug-target engagement; procedure is generally safe and well-­tolerated; suitable for repeated studies; Permits in vitro assessment of impact of therapeutic interventions on adipocyte physiology; drug-target engagement; subcutaneous fat biopsy is a safe and well-­tolerated procedure suitable for repeated studies

Liver biopsy

Conclusions Tissue biopsy may be a relevant outcome measure to understand tissue-level pharmacokinetic/ pharmacodynamic responses to a therapeutic intervention. Biopsy of muscle and adipose tissue may provide information that usefully complements whole-body metabolic studies, both in

Disadvantages Invasive with risk of procedureassociated morbidity and mortality, sampling error, observer variability of interpretation of histological findings; cost; contraindicated in some patients Few; relevance of target muscle (e.g. quadriceps vs. deltoid vs. rectus abdominis) to whole- body metabolism should be considered

Few; invasive

Value in drug development decisions High; required for regulatory approval of new therapies for NASH

Limited role; perhaps underutilized in drug development

May be of value in proof-of-­mechanism studies, e.g. thiazolidinediones, and pharmacokinetic/ pharmacodynamic studies, e.g. 11β-hydroxysteroid dehydrogenase type 1 inhibitors

terms of understanding disease processes and assessing therapies directed towards diabetes and related metabolic disorders. In this regard, the techniques of biopsy and culture of skeletal muscle and fat, while illuminating important aspects of the pathophysiology of insulin resistance and type 2 diabetes, have perhaps been underutilized in drug development. Liver biopsy

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

remains the definitive means to diagnose and stage NAFLD/NASH and is currently an essential regulatory requirement for registrational clinical trials of new treatments. Indeed, improvement of liver histology is considered as an intermediate short-term surrogate that is necessary for a drug to obtaining conditional approval. However, drug efficacy must be confirmed by clinical outcome to gain final regulatory approval. Liver histology is also the standard against which non-invasive biomarkers must be evaluated.

Introduction Tissue biopsy of liver, skeletal muscle and adipose tissue permits histological examination of these key metabolically-active tissues and the quantification of biochemical composition, gene and protein expression, engagement of hormones and drugs with cellular receptors, and activity within intracellular metabolic pathways. Such information may usefully inform the development of new drugs for cardiometabolic disorders notably nonalcoholic fatty liver disease/non­ alcoholic steatohepatitis (NAFLD/NASH), obesity, diabetes and certain complications of diabetes, notably chronic kidney disease. In this chapter we consider tissue biopsy techniques in clinical metabolic research as applied to liver, skeletal muscle, and adipose tissue. The main focus is on the role of liver biopsy in the development of new pharmacotherapies for NASH with fibrosis in adults (see Table 10.1 for disease definitions used in this review). We present an overview of the critical role of liver histology in diagnosis/staging with an emphasis on drug development, positioned within the context of current clinical practice guidelines and practical considerations of undertaking liver biopsy. The advantages and recognized limitations of the procedure are reviewed. We consider some recent technological advances in histopathological techniques and offer an overview of the current status of imaging and other biomarkers as non-invasive alternatives to liver biopsy in NAFLD. We position the role of liver biopsy in clinical trials of

247

Table 10.1  NAFLD and related definitions NAFLD Encompasses the entire spectrum of fatty liver disease in individuals without significant alcohol consumption, ranging from fatty liver to steatohepatitis to cirrhosis. NAFL Presence of >5% hepatic steatosis without evidence of significant inflammation or hepatocellular injury in the form of ballooning of hepatocytes or evidence of fibrosis. The risk of progression to cirrhosis is considered minimal. NASH Defined histologically by the presence of >5% hepatis steatosis with lobular inflammation and hepatocyte injury (ballooning) with or without fibrosis. NASH may progress to cirrhosis, liver failure, and liver cancer (rare) Based on Chalasani et al. [33]

new therapies within the broader perspective of recommended approaches to screening, diagnosis, staging, and management of NAFLD in routine clinical practice. The chapter has been written with the needs of clinical researchers involved in drug development in mind.

Liver Background NAFLD (Table  10.1) is the most common liver disorder globally. This clinicopathological spectrum encompasses NAFL (non­ alcoholic fatty liver) the silent and benign form of the disease and NASH, the inflammatory counterpart with risk of progression to advanced fibrosis and clinical outcomes. The estimated prevalence of NAFLD is ~25% (with a range of ~10–40%) of the world’s adult population [1, 2]. Regional variability has been reported with the highest rates across the Middle East and South America and the lowest rates in Africa [3, 4]. A multi-ethnic US study of 400 adult patients recruited from an Army Medical Center in whom liver biopsy was performed when fatty liver was identified by ultrasound (US) the prevalence of NAFLD was 46% [5]. A diagnosis of NASH was confirmed in 40 patients (12.2%). Hispanics had the highest

A. J. Krentz and P. Bedossa

248

prevalence of NAFLD, followed by Caucasians and African Americans; Hispanics also had a higher prevalence of NASH compared with Caucasians. NASH (Table  10.1) can progress to cirrhosis, liver failure, and rarely liver cancer (hepatocellular carcinoma, HCC) [1, 6]. Approximately 30–40% of patients with NASH will develop hepatic fibrosis [7–9]. Fibrosis may also develop in patients with uncomplicated NAFL. In an analysis of studies of paired liver biopsies at least a year apart, Singh et al. reported more rapid progression by 1 stage in fibrosis (7.1 vs 14.3  years) for patients with NAFL and NASH, respectively [9]. Fibrosis may also regress, as demonstrated in a small study (n  =  52) of middle-aged patients in Hong Kong with biopsy-proven NAFLD who had liver biopsies repeated after 36 months. In this series, 27% patients had progression of fibrosis, 48% had static disease, and 25% showed regression of fibrosis [10]. It is estimated that NASH may account for >600,000 cases of cirrhosis in the USA alone and is projected to become the leading cause of liver transplantation in the near future [11, 12].

 ssociation of NAFLD with Other A Diseases There is a strong association between NAFLD and other obesity-related metabolic disorders. A putative bidirectional association exists between NAFLD and components of the metabolic syndrome [13]. In many cases, NAFLD is associated with metabolic comorbidities including insulin resistance, glucose intolerance, type 2 diabetes, and/or dyslipidemia [14, 15]. Not only are features of metabolic syndrome highly prevalent in patients with NAFLD but the components of metabolic syndrome are associated with an increased risk of developing NAFLD [13, 16]. In the aforementioned biopsy prevalence study of Williams et al. NAFLD patients were more likely to be male, older, and to have hypertension and diabetes [5]. Patients with NAFLD in this study also had a higher body mass index (BMI), ate fast food more often, and exercised less than their non-NAFLD counterparts. The Nonalcoholic Steatohepatitis

Clinical Research Network (NASH CRN) enrolled 1266 adults (median age 50 years) 1101 of whom had liver histology data [17]. Most (82%) were white with Hispanics accounting for 12%of the study population. The median BMI was 33 kg/m2, 49% had hypertension, and 31% had type 2 diabetes. On liver biopsy, 57% were judged to have definite NASH with 31% having bridging fibrosis or cirrhosis. In subgroup analysis, patients with definite NASH were more likely to be female and have diabetes, have higher levels of liver transaminases, and homeostasis model assessment of insulin ­resistance (HOMA-IR). Models for predicting histological diagnoses performed modestly for predicting steatohepatitis or ballooning and better for advanced fibrosis. Up to two-thirds of patients with type 2 diabetes and more than 80% of patients undergoing bariatric surgery have evidence of NAFLD [18– 20]. Observational studies suggest that NAFLD is associated with an increased long-term risk of incident chronic kidney disease [21]. Risk factors and conditions associated with NAFLD are presented in Table  10.2. Mechanistic insights into hepatic lipid accumulation, hepatocyte injury, the role of the immune system and fibrosis as well as the role of the gut microbiota are being extensively explored [1]. The leading cause of death among patients with NAFLD is cardiovascular disease [16, 22]. Some studies have suggested that individuals with NAFLD may be at increased risk of cardiovascular disease even after classic risk factors are accounted for. However, it remains uncertain at present whether NAFLD should be regarded as an independent risk factor for cardiovascular disTable 10.2  Conditions associated with NAFLD Common conditions with established association Obesity Type 2 diabetes Dyslipidemia Metabolic syndrome Polycystic ovary syndrome Chronic kidney disease

Other conditions associated with NAFLD Hypothyroidism Obstructive sleep apnea Hypopituitarism Hypogonadism Pancreatoduodenal resection Psoriasis

Based on Chalasani et al. [33]

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

ease [22, 23]. Genetic and epigenetic factors may help explain inter-individual variations in disease phenotype, severity and progression [1]. Genomewide association studies (GWAS) have revealed several single nucleotide polymorphisms (SNPs) associated with the pathology of NAFLD.  Of these, the patatin-like phospholipase domaincontaining 3 (PNPLA3) gene variant I148M (~30–50%) and the transmembrane 6 superfamily member 2 (TM6SF2) E167K variant (~10– 15%) are associated with development and progression of NAFLD, NASH, and NAFLDrelated hepatocellular carcinoma although not of diabetes or cardiovascular disease [2, 13, 24]. To summarise, NAFLD presents a significant public health impact in terms of the worldwide disease burden and associated morbidity and mortality. There is currently an unmet need for safe and effective pharmacotherapy for NALFD/ NASH. However, development of pharmacological treatments has been hindered by issues such as difficulties in enrolment to clinical trials and the inherent challenges of designing clinical trials in such a heterogeneous and slowly progressive disorder.

 istological Features of the NAFLD H Disease Spectrum Liver biopsy technique is an important consideration when interpreting hepatic histology. The robustness of histological assessment is strongly related to the length and width of the liver biopsy. A 20  mm length liver biopsy done with a 16 Gauge needle is considered as an adequate sample providing reliable data. The quality of the tissue processing is also important. The histological features of nonalcoholic and alcoholic fatty liver diseases are regarded as indistinguishable [1]; the distinction between the two conditions rests on reported history of alcohol consumption. As mentioned, the term NAFLD encompasses a wide spectrum of histological lesion from simple steatosis (NAFL) to NASH, a complex constellation of several lesions, each of which is related to different pathophysiological mechanisms. A report by Ludwig et al., now regarded as seminal,

249

based on a small series of patients (n = 20) with non­alcoholic steatohepatitis (NASH) described the main clinical, laboratory and histological features of the disorder [25]. Hepatomegaly and minor abnormalities of liver function were common, the disorder was more common in women (although current data suggest that the prevalence of NAFLD is higher in men than women), most of the patients were moderately obese, and many had obesity-associated diseases such as diabetes mellitus and cholelithiasis. Liver biopsy specimens were characterized by the presence of (a) fatty change with (b) evidence of lobular hepatitis (c) focal necroses with mixed inflammatory infiltrates and (d) Mallory (nowadays referred to as Mallory-Denk) bodies. Fibrosis was found in most liver specimens with evidence of cirrhosis in three patients. Since the report of Ludwig et al. evidence has accumulated that indicates that patients with histologically-proven NASH, especially those with some degree of fibrosis, are at higher risk of adverse outcomes such as cirrhosis, overall and liver-related mortality including hepatocellular carcinoma [26–28]. Steatosis in NAFLD is usually macrovesicular with a single large intracytoplasmic fat droplet or smaller well-defined droplets displacing the hepatocyte nucleus to the cell periphery. It is directly related to intracytoplasmic triglyceride accumulation. Hepatic steatosis is commonly evaluated and reported semi-quantitatively based on percentage of hepatocytes with fat droplets. In simple NAFLD, steatosis may be pure or accompanied by small foci of lobular inflammation, mild portal inflammation, and lipogranulomas. By definition, features of hepatocellular injury and fibrosis, indicating progression to steatohepatitis, are not observed. Most hepatopathologists view minimal criteria for the histological diagnosis of adult NASH to include steatosis, hepatocyte injury, usually in the form of ballooning, and lobular inflammation, typically localized in acinar zone 3. Fibrosis not a required diagnostic feature of NASH but perisinusoidal fibrosis, a characteristic pattern of fibrosis, is often present in NASH [1]. Fibrosis is a major histological feature since all prospective studies have shown that fibrosis

A. J. Krentz and P. Bedossa

250

a

b

c

d

Fig. 10.1  Nonalcoholic fatty liver (steatosis, NAFL). (a) Gross section of fatty liver. Liver is diffusely enlarged and yellowish. (b) Low magnification of fatty liver. Steatosis is located predominantly in zone 3 of the lobule, around the central vein (CV) (Hematein and Eosin, ×25). (c) Fat

vacuole fill entirely the hepatocytes setting aside cytoplasm and nucleus (Hematein and Eosin, ×20). (d) In pure steatosis, fibrosis is absent or make delicate perisinusoidal fibers around central vein. (Sirius Red Hemalun, ×10)

is the only independent predictor of clinical outcomes. In contrast to chronic hepatitis where fibrosis expands from portal tracts, fibrosis starts and extends into the pericentrolobular area (zone 3 of the lobule) (Fig. 10.2). With progression of the disease, fibrosis extent to adjacent vascular structures (portal tract and/or central vein) to delineate fibrous septa. Advanced fibrosis is characterized by numerous fibrous septa that disrupt lobular architecture and precedes the development of cirrhosis. While fibrosis is very common in NASH, it is rarer in NAFL, although pure steatosis with fibrosis may be recognized (steatofibrosis). Figures 10.1 and 10.2 illustrate the histological findings in NAFL and NASH, respectively. Several comprehensive reviews of the histopathology of NAFLD have been published [1, 29–31]. It is recognised that hepatic injury in children attributable to NAFLD may have a different histological pattern to NAFLD in adults with zone 1 inflammation and periportal fibrosis typically more pronounced than in adults [1, 32, 33].

I mportance of Liver Histology in Clinical Trials of New Therapies for NAFLD/NASH In the development of therapeutic agents for NASH trial design must take account of complexities that include the non-linear, heterogeneous nature of disease progression and bidirectionality of disease [34]. The results of clinical trials have been difficult to compare due to inconsistent definitions of relevant disease parameters in patients with NASH.  In recognition of this issue, the Liver Forum (a multi-­ stakeholder collaboration between the Food & Drug Administration (FDA), the European Medicines Agency (EMA), academic investigators, industry stakeholders, and patient representatives that aims to catalyze therapeutic development for NASH by developing potential solutions to barriers to development) summarized the ‘current consensus or lack thereof’ regarding key elements required to diagnose phenotypes of NAFLD for clinical trials [35]. According to the Liver Forum, ‘variable case

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

a

b

251

c

d

Fig. 10.2 Nonalcoholic steatohepatitis (NASH). (a) NASH is characterized by association of hepatocyte ballooning, steatosis and inflammation around central vein (CV). Perisinusoidal fibrosis (in blue) radiates from central vein along hepatocytes within the lobule. (Masson’s Trichrome, ×10). (b) Ballooned hepatocyte (arrow). Hepatocytes are rounded, enlarged with clear cytoplasm

and nuclei in the center of the cells (Hematein and Eosin, ×20). (c) lobular inflammation. Foci of lymphocytes (arrow) around steatosis vacuoles (Hematein and Eosin, × 10). (d) Advanced fibrosis. Fibrosis is common in NASH. In septal fibrosis, bands of collagen fibers (in blue, arrow) disrupt lobular organization and link adjacent central vein and portal tracts. (Masson’s Trichrome, ×2)

definitions, methods for identification of subjects, and assessment of outcomes limit the ability to anchor diagnostics and therapeutic development on a robust model of the disease, impeding assessment of diagnostics and benefit of therapeutic interventions.’ The National Institute of Diabetes and Digestive and Kidney Diseases’ (NIDDK) NASH Clinical Research Network (CRN) system of grading disease activity and staging fibrosis separately is the most widely used and validated for NAFLD.  The stages of fibrosis extend from stage 0 to stage 4 (0: None; 1: 1a, mild, zone 3 perisinusoidal fibrosis, 1b, moderate, zone 3 perisinusoidal ­ fibrosis, 1c, portal/periportal fibrosis only; 2: Zone 3 perisinusoidal fibrosis and portal/periportal fibrosis; 3: Bridging fibrosis; 4: Cirrhosis, probable or definite. The NAFLD Activity Score (NAS) is the unweighted sum of grading of steatosis, lobular inflammation and hepatocellular ballooning (Table 10.3). Other semi-quantitative

Table 10.3  Nonalcoholic steatohepatitis Clinical Research Network system for scoring activity (NAFLD activity score NAS) in nonalcoholic fatty liver disease Steatosis grade (S) 0: 4

Hepatocyte ballooning (B) 0: None 1: Few ballooned cells 2: Many ballooned cells

Adapted from Brunt and Tiniakos [50] a Counted in 20 × fields. Nonalcoholic fatty liver disease Activity Score: S + L + B (range 0–8)

approaches include the Brunt staging method [36] and the steatosis–activity–fibrosis (SAF) [30, 37] score. The Goodman classification combines disease activity and fibrosis but is regarded as the least validated system [38]. In the opinion of the Liver Forum, the natural course of NAFLD ‘has not been rigorously

A. J. Krentz and P. Bedossa

252

Table 10.4  Liver Forum (2018) proposed standardized format for comparison of study populations across NASH clinical trialsa Phenotype Steatosis Steatohepatitis Indeterminate

Disease activity NAS  Steatosis  Lobular inflammation  Ballooning SAF  Steatosis  Lobular inflammation  Ballooning  Fibrosis

Disease stage Fibrosis  Stage 0: No fibrosis  Stage 1a: Mild perisinusoidal  Stage 1b: Moderate perisinusoidal  Stage 1c: Portal/periportal  Stage 2: Perisinusoidal and portal/periportal  Stage 3: Bridging  Stage 4: Cirrhosis

Etiology/associations Insulin resistance Alcohol Lean NASH PNPLA3+ Drugs Inherited disorders (e.g., Weber-Christian, hypobetalipoproteinemia) Lipodystrophy Short bowel TPN Jejunoileal bypass

Modified from Siddiqui et al. [35] Abbreviations: PNPLA3 patatin-like phospholipase domain containing 3, TPN total parenteral nutrition a Many include one or more subsets from each of the columns. This provides a standardized format for comparison of study populations across trials

characterized, particularly with respect to the contributions of underlying obesity, type 2 diabetes, and other comorbidities and the treatments provided for these comorbidities’ and ‘there remains a major unmet need in the field to develop standardized definitions for populations for interventional trials.’ In their report, the Case Definitions Working Group established by The Liver Forum evaluated the validity of case definitions for populations to be included in clinical trials for NASH from a regulatory science perspective. Based on such analyses, specific recommendations were provided noting the strengths and weaknesses of the case definitions along with knowledge gaps that require additional study (Table 10.4) [35]. Determining optimal yet feasible endpoints for clinical trials of pharmacological agents for NAFLD and NASH is complicated by the slow progression of clinically significant outcomes. Moreover, individual drug treatment for NAFLD/ NASH tends to focus on one primary mechanism of action, i.e. metabolic or NASH disease modifier vs. antifibrotic activity (Table  10.5). Drugs may have different impacts on histological endpoints, e.g. improving NASH but worsening fibrosis and vice versa. Thus, NASH and fibrosis

Table 10.5  Histological features of disease and disease activity that have been demonstrated to improve in interventional clinical trials in NASH Steatosis Lobular inflammation and ballooning Grade of steatohepatitis (NAS) Fibrosis Abbreviations: NAS NAFLD activity score

need to be evaluated independently to ensure a beneficial impact on one parameter does not simultaneously result in a negative impact on another endpoint. Resolution of NASH is defined as a complete resolution of hepatocyte ballooning, with lobular inflammation scores of 0 or 1, in addition to no worsening of fibrosis, though there is no specific requirement regarding change in steatosis. Clinical trials focused on fibrosis require that there is no worsening in NASH [34]. The design of each phase of a clinical trial of a therapeutic agent for NAFLD is determined by the drug’s mechanism of action (Fig. 10.3). Phase 2a clinical trials focus on proof of concept and short-term safety. The short time course of early phase studies creates challenges for the selection of relevant endpoints. Pharmacologic

phase IIa

phase IIb

phase III

Adaptive

. Multiple time points for assessment . Interim analysis to consider dropping study arms and re-randomisation

Longer term: years

Intermediate: 24-72 wk

Truncated: 12-24 wk

Timeline

Paired liver biopsy

Paired liver biopsy

Paired liver biopsy

Liver biopsy not required

Endpoints

. MRE, FS kPa

. ∆ biomarkers, MRE and FS kPa early part of trail

. Resolution of NASH w/o worseing of fibrosis . Multiparametric MRI . ∆ liver enzymes and other biomarkers or multiparametric MRI early part of trial . ∆ histology later part of trial

. ∆ hepatic fat via MRI-PDFF or CAP early part of trail . ∆ hepatic fat via histology later part of trail

. Assess clinical outcomes at end of trial

. HCC . Transplantation

. Liver-related mortality

. Progression to cirrhosis . hepatic decompensation . Overall mortality

Clinical

atohepatitis, PDFF protein density fat fraction. Note: Phase 3 trials may involve surrogate endpoints for initial approval (histology) and outcomes endpoints to confirm clinical benefit (for full approval). (Reproduced with permission from Konerman et al. [34])

. ∆ histology later part of trail

. ∆ fibrosis stage w/o worsening of NASH

. ∆ inflammation and ballooning (NAS)

∆ hepatic fat via histology ± MRI-PDFF/CAP

. MRE, FS kPa

. Resolution of NASH w/o worseing of fibrosis . Multiparametric MRI

. MRE, FS kPa

. ∆ biomarkers

Fibrosis

. ∆ fibrosis stage w/o worsening of NASH

. Multiparametric MRI

. ∆ liver enzymes and other biomarkers

Inflammatory

. ∆ inflammation and ballooning (NAS)

∆ hepatic fat via histology ± MRI-PDFF/CAP

∆ hepatic fat via MRI-PDFF or CAP

Metabolic

Fig. 10.3  Clinical trial designs for studies of NAFLD and NASH. Abbreviations: CAP controlled attenuation parameter, FS fibroscan, HCC hepatocellular carcinoma, MRE magnetic resonance elastography, MRI magnetic resonance imaging, NAFLD nonalcoholic fatty liver disease, NAS NAFLD activity score, NASH nonalcoholic ste-

.

. Cover aims of phase II and III trails sequentially

. Longer term safety and efficacy . Clinical outcomes

. Confirm efficacy

. Safety and adverse events . Therapeutic dosing

. Assess efficacy

. Clarify target engagement

. Short term safety

. Proof of concept

Focus

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other… 253

A. J. Krentz and P. Bedossa

254

agents directed against for NASH  – especially fibrosis  – have a low probability of inducing clinically significant improvements in this scenario. In addition, paired liver biopsies performed at short intervals can raise ethical concerns given the potential risk to the patient. In the absence of histology, phase 2a studies should employ appropriate surrogate endpoints aimed at confirming target engagement. While non-invasive markers are currently not approved as surrogates for histology they offer initial proof of concept testing of target engagement [34]. Regulatory authorities require liver biopsy for phase 2b and phase 3 clinical trials. Indeed, change in histology is considered as the only valid short-term primary endpoint (12– 18 months). Also, phase 3 clinical trials require, in addition a long-term follow-up (4–6 years) and evaluation of clinical outcomes. Resolution of the histological lesions of NASH as defined previously and/or regression of fibrosis (at least 1 stage according to the NASH CRN) are the primary end-points which are considered in phase 3 clinical trials. However, studies of placebo treatment groups have demonstrated that resolution of liver injury can occur in the absence of treatment [39–41]. This realization informs the design of interventional clinical trials i.e., a need for adequate statistical power and randomized controlled trials that contain a placebo-treatment arm. Another important consideration is whether improvements in histological features of NASH in response to treatment are maintained beyond the period during which the intervention is applied. For example, a long-term treatment trial with the thiazolidinedione insulin-­sensitizer rosiglitazone showed no additional anti-steatogenic effect with longer treatment duration, despite an increase in whole-body insulin sensitivity [42]. In contrast, bariatric surgery may have durable beneficial effects on liver histology [43, 44]. Of note, not all disease features may improve with a specific intervention. Furthermore, as mentioned there is potential for one or more pathological features to improve in concert with progression or deterioration in others (see below).

 on-invasive NAFLD Biomarkers: N Relation to Liver Biopsy In response to the recognition that liver biopsy is invasive, expensive and prone to sampling error clinical prediction algorithms, blood-based biomarkers and imaging have been developed to identify patients at high risk of NASH and advanced degrees of fibrosis [45]. Clinical trials of sufficient power with paired biopsy designs provide an opportunity to test non-invasive disease surrogates that may eventually supplant histology or provide an early efficacy or futility signal [34]. Current approaches include predictive models or direct measures of inflammation (Table  10.6). Such biomarkers may allow discrimination between NASH patients with or without or advanced fibrosis, predict changes in NASH/fibrosis, and provide long-term prognostic information, albeit with certain caveats, e.g. age and inflammatory activity. Recent efforts have focused on omics to develop and validating novel biomarkers [45]. Non-invasive tests are being validated against liver histology even though liver histology has its own shortcomings [46]. The emerging roles of non-invasive and imaging biomarkers in the development of drugs for NAFLD are considered in more detail in Chaps. 6 and 16. Table 10.6  Examples of risk scores and biomarkers developed as alternatives to liver biopsy to identify patients at high risk of NASH and advanced fibrosis Predictive models  NAFLD fibrosis score  FIB-4 index  BARD score  Aspartate aminotransferase [AST] to platelet ratio index (APRI), Measures of inflammation or fibrosis  Keratin 18 fragments  FibroTest®  ELF™ panela  Pro-C3 – a marker of type III collagen formation See text for details of tests Enhanced Liver Fibrosis test

a

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

 hallenges in Identifying Patients C with NAFLD and Implications for Clinical Trials There are unique challenges in the design of interventional studies in patients with NASH [47]. These include long time interval between the progression from NAFL to NASH to cirrhosis and ultimately liver failure, along with gaps in knowledge regarding disease modifiers. There is an urgent need to develop methods to identify individuals at particular risk of disease progression and to validate endpoints that reflect meaningful changes in health status in this population. The majority of patients with NAFLD are either asymptomatic or have non-specific symptoms such as fatigue which impedes the development of symptom- and functional-based outcomes in the early stages of the disease [47]. Most cases are detected incidentally when clinical chemistry liver function tests or abdominal US are performed for some other reason. Neither serum transaminase levels (alanine aminotransferase, ALT; aspartate aminotransferase, AST) nor imaging tests, e.g. US, computed tomography (CT), or magnetic resonance imaging (MRI), have the capacity to reliably reflect the spectrum of liver histology in patients with NAFLD. So far, non-invasive markers have been mainly used to target patients that might get some benefit from a liver biopsy, thereby avoiding a significant number of inappropriate biopsies. However, noninvasive imaging methods are evolving to the point where they may be able to replace liver biopsy even though this approach represents replacing one surrogate marker of liver disease with another [48, 49]. As mentioned, the importance of distinguishing uncomplicated hepatic steatosis from NASH rests on the generally benign natural history of the former condition that contrasts with the risks of progress to cirrhosis, liver failure, and liver cancer associated with the latter [33]. The differential diagnoses of NAFLD and NASH are wide and include alcoholic liver

255

disease, drug toxicity, e.g. tamoxifen, glucocorticoids, and highly active antiretroviral therapy in patients with human immunodeficiency virus, metabolic diseases including Wilson’s disease, lipodystrophy syndromes, surgical procedures (such as jejunoileal bypass and biliopancreatic diversion), total parenteral nutrition, and malnutrition [50]. Additional complexity comes also from the high prevalence of NAFLD, making association of NAFLD to another chronic liver disease not so rare, the most common association being nonalcoholic and alcoholic liver disease. Published expert clinical guidelines have differed in their recommendations concerning screening for NAFLD.  In 2016, joint clinical practice guidelines for the management of NAFLD were published by the European Association for the Study of the Liver (EASL) European Association for the Study of Diabetes (EASD) and European Association for the Study of Obesity (EASO) [51]. Screening for NAFLD at community level has been questioned given the high direct and indirect costs of testing, the low predictive value of non-invasive tests, the risks of liver biopsy and the lack of effective treatments [51, 52]. However, the progressive form of NAFLD (i.e. NASH), particularly when associated with advanced fibrosis, should be identified in patients at risk (age >50 years, type 2 diabetes mellitus or metabolic syndrome). In subjects with obesity or metabolic syndrome, screening for NAFLD by liver enzymes and/or US should be part of routine work-up with case finding for advanced disease (i.e. NASH with fibrosis) in the aforementioned high-risk groups. Biomarkers and scores of fibrosis, as well as transient elastography, were regarded as acceptable non-­invasive procedures for the identification of cases at low risk of advanced fibrosis/ cirrhosis. The combination of biomarkers/scores and transient elastography might confer additional diagnostic accuracy and might reduce the need for diagnostic liver biopsies. Monitoring of fibrosis progression in clinical practice may rely on a combination of biomarkers/scores and transient elastography, although it was noted that

256

this strategy was thought requires validation. The identification of advanced fibrosis or cirrhosis by serum biomarkers/scores and/or elastography is less accurate and needs to be confirmed by liver biopsy, according to the clinical context. In selected patients at high risk of liver disease progression, monitoring should include a repeat liver biopsy after at least 5 years of follow-up. In contrast, recent UK guidelines from the National Institute for Health and Clinical Excellence (NICE), whilst recognising the higher prevalence of NAFLD in people with type 2 diabetes or the metabolic syndrome, did not recommend screening for NAFLD in adults due to a paucity of evidence [53, 54]. However, in children with type 2 diabetes or metabolic syndrome who do not consume excessive alcoholic the guidance proposed that a diagnostic US examination should be offered. NICE considered that there was no strong evidence to support the use of a specific non-invasive test for identifying NASH in patients with NAFLD, even though severe liver fibrosis has been consistently shown to indicate a poor prognosis. The evidence base for the ELF score in identifying advanced liver fibrosis (stages 3 and 4) in people incidentally diagnosed with NAFLD was considered to be of very low to low quality. Nonetheless, ELF was recommended by NICE as the most cost-effective option, using a cutoff value of >10.51. AASLD guidance recommends consideration of liver biopsy in patients with NAFLD who are at increased risk of having steatohepatitis and/or advanced fibrosis [33]. Relevant factors include: • Presence of the metabolic syndrome, NFS (NAFLD fibrosis score) or FIB-4 index • Liver stiffness measured by VCTE (vibration controlled transient elastography, FibroScan) or MRE (magnetic resonance elastography) According to AASLD guidance, in patients with NAFLD the metabolic syndrome predicts the presence of steatohepatitis. The NFS and FIB-4 are regarded as clinically useful tools for

A. J. Krentz and P. Bedossa

identifying NALFD patients with a higher probability of having bridging fibrosis (stage 3) or cirrhosis (stage 4). The NFS is based on six readily measured variables (age, body mass index, hyperglycemia, platelet count, albumin and ALT/ AST ratio and is calculated using a published formula (http://gihep.com/calculators/hepatology/ nafld-fibrosis-score/). FIB-4 (fibrosis-4) Index is an algorithm based on platelet count, age, AST and ALT (http://gihep.com/calculators/hepatology/fibrosis-4-score/). VCTE, which measures liver stiffness non-invasively, or MRE are clinically useful tools for identifying advanced fibrosis in patients with NALFD. It has been reported that pharmacologic treatment trials in NASH are enrolling patients at a slower pace than treatment trials in other disease states [34]. This may be explained by several potential factors including low disease awareness among patients and providers, as well as barriers to trial enrolment such as complex inclusion/ exclusion criteria including age, BMI, levels of transaminases, glycated hemogobin, and bilirubin. Restrictions on patients with a history of cancer or bariatric surgery, as well as on those taking concomitant medications further serve to deplete the pool from which patients can be recruited. The lack of reliable surrogate endpoints with reliance on histology requires multiple liver biopsies, particularly in more advanced phase trials. The high volume of clinical trials in NAFLD/NASH in recent years has also contributed to slow enrollment [34, 55].

Percutaneous Liver Biopsy It is 60 years since Menghini described the percutaneous suction liver biopsy technique which is credited with revolutionizing hepatology practice [56, 57]. Liver biopsy is the gold standard for characterizing hepatic histological features in patients with NAFLD [58]. While there has been progress in developing non-invasive approaches, liver biopsy remains the only diagnostic procedure to reliably assess the relevant pathological patterns, their related severity, and associations

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

[31, 59]. Indeed, NASH is the only chronic liver disease whose definition is based solely on histological criteria (steatosis, inflammation and ballooning). As discussed, liver biopsy can be used to score the histological patterns of disease (NASH-CRN, SAF scores) and assess the extent of fibrosis. When specific treatments become available liver biopsy might remain useful for choosing the most suitable therapeutic option based on the predominant histological features, i.e. activity, steatosis, fibrosis [31]. Current practice guidance issued by the American Association for the Study of Liver Diseases (AASLD) requires that for a diagnosis of NAFLD to be sustained there must be (1) evidence of hepatic steatosis, determined either by imaging or histology, and (2) lack of secondary causes of hepatic fat accumulation such as significant alcohol consumption, long-term use of a steatogenic medication, or monogenic hereditary disorders [33]. A challenge in terms of enrolling patients into clinical trials is identifying which patients with NAFLD have NASH, particularly those with advanced fibrosis. As discussed earlier, the natural history of NAFLD is recognized to be highly variable and often non-linear in its progression [60]. Disease progression is most dependent on the presence or absence of NASH as determined by biopsy-determined hepatic histology [60]. Recent technological advances include virtual slides (e-slides) that provide high quality images suitable for sharing between histologists for double readings and second opinions, automated image analysis for collagen proportional area (CPA) measurement or dual-photon microscopy [61] to increase the granularity and accuracy of fibrosis evaluation or lipidomic imaging mass spectrometry (LIMS) [62]. These techniques may have relevance for NAFLD biomarker discovery and validation.

Limitations of Liver Biopsy Limitations of liver biopsy relevant to clinical trials include specimen size and quality, tissue

257

processing, specimen interpretation, risk of sampling error, and histopathologist observer variability. Since even an adequate biopsy represents only a tiny proportion of the liver high quality biopsy techniques, such as utilization of appropriately sized needles, careful choice of the sampling area (including which lobe of the organ), and subsequent tissue preparation and interpretation are relevant considerations [50]. Some of the difficulties associated with liver biopsy can be avoided if this procedure is performed by an experienced hepatologist (or interventional radiologist) and read by an experienced liver histopathologist [31]. The training and expertise of the pathologist is recognized to be of importance, as is the greater yield of findings associated with multiple readings. Studies of the reproducibility of histological lesions in NAFLD have generally shown agreement for certain features including assessment of steatosis and fibrosis with less concordance between pathologists concerning lobular inflammation. It is therefore meaningful to perform central reading of liver biopsies in clinical trials. Intra-­observer agreement is generally better than that for inter-observer agreement [50]. Detailed reviews of the pros and cons of liver biopsy have been published (Table 10.7).

Table 10.7  Pros and cons of liver biopsy Pros Gold standard for diagnosis and staging of NAFLD Required by regulators for drug approval Standard against which non-invasive biomarkers of NAFLD are evaluated Cons Invasive Inter-intra observer variability Samples ~1/50,000 of organ Morbidity Mortality (rare) Presence of contraindications Paucity of effective treatment options Expense (~£1500 in US) Sources: Cadranel et  al. [65]; Rockey et  al. [64]; Brunt and Tiniakos [50]

258

 ractical Considerations of Liver P Biopsy Liver biopsy techniques require specific training to ensure appropriate specimen retrieval and the lowest rate of complications [63, 64] (AASLD 2009). The procedure is generally safe in experienced hands [63, 65]. Usual practice includes measurement of the complete blood count, including platelet count, and measures of coagulation including prothrombin time (PT)/international normalized ratio (INR), activated partial thromboplastin time, as appropriate. Blood typing ensures that blood can be made available at short notice in case of bleeding. Expert opinion varies in terms of preference as to whether patients should be fasting prior to biopsy [64] (AASLD 2009). Management of anti-platelet, non-steroidal antiinflammatory drugs and/or anticoagulants before and after liver biopsy is an important consideration. Few data are available to guide management about the timing of discontinuation of (or even the need to discontinue) these medications. The consensus appears to be that these medications should be discontinued prior to the procedure. Drugs such as aspirin, dabigatran, prasugrel, and clopidogrel may need to be discontinued ~2  weeks and warfarin ~5 days prior to liver biopsy with INR checked on the day of the procedure. The needle type is also important; various biopsy needles are available. Sectioning biopsy should be preferred to aspiration procedure since aspiration may lead to major fragmentation specially if advanced fibrosis or cirrhosis. Width of biopsy is also a major condition. A 16-gauge needle biopsy (or larger) provides a biopsy of adequate size to assess confidently score of fibrosis. Thinner needles are not recommended in NAFLD since staging of fibrosis is less reliable when performed on thin biopsies.

Peri-Procedure Considerations Most recommendations regarding peri-procedure restrictions lack definitive evidence because few recommendations are based on comparative study

A. J. Krentz and P. Bedossa

data [64]. Usual daily activities may be undertaken up until the day preceding the procedure. Liver biopsy should be performed in a dedicated clinical space that can accommodate operator(s), assistants, and emergency resuscitation equipment, as necessary. The preference for oral or intravenous anxiolytic therapy or conscious sedation is variable. If such medications are utilized any substantial oral intake should be avoided prior to the procedure. Routine placement of an intravenous cannula prior to the procedure is practiced in many facilities as a precaution should there be significant pain and/or bleeding after the procedure. Imaging with US may be used to guide biopsy. The field should be prepared with povidone-iodine solution and sterile drapes placed appropriately. Once the target area has been infiltrated in superficial and deep planes with a local anesthetic agent (lidocaine 1%) the biopsy is taken. The biopsy needle is inserted percutaneously while the patient holds their breath, then withdrawn quickly, removing a 2–4 cm long sample of liver tissue. Pressure is applied to the site, followed by an adhesive bandage. The patient is then asked to lie on the right side so that their body weight compresses the wound.

 omplications of Percutaneous Liver C Biopsy Well-recognized complications of liver biopsy, which may vary in terms of risk and magnitude by patient characteristics and the type of procedure include: • Pain (common; usually straightforward management using analgesia) • Vasovagal episodes • Bleeding (less than 1  in 100 cases; usually immediately after biopsy but delayed haemorrhage is recognized) • Pneumothorax (1 in 500 cases) • Biliary leak (less than 1 in 1000 cases) • Death (very uncommon after percutaneous biopsy and usually due to haemorrhage) Following liver biopsy, patients are moved to a recovery room to rest quietly, particularly if

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

they received sedation and/or opiate analgesia. The patient remains in bed while vital signs are observed for ~4–6  h post-procedure, every 15 min for the first hour, every 30 min during the second hour, and then hourly until discharge. Patients should not drive themselves or travel on their own and should have someone to stay with them overnight. Some physicians recommend that patients spend the first evening within an hour’s driving distance of the hospital where the biopsy is done in case a complication develops. In the absence of complications or significant pain that necessitates use of potent analgesia no restrictions are placed upon returning to work the following day. Patients are discouraged from lifting heavy objects for a minimum of 24  h because of intra-abdominal pressure increases that in theory could facilitate bleeding from the puncture site. Patients should be able to get back to usual activities gradually over a period of a week. In clinical hepatology practice, various approaches may be utilized for obtaining a liver tissue specimen. These include a blind percutaneous approach using digital percussion, biopsy under US or CT guidance, transjugular, and intraabdominal biopsy at laparoscopy or laparotomy, notably during bariatric surgery [66]. Laparoscopic biopsy has the advantage of being a targeted biopsy. Another advantage is that bleeding can be arrested directly. Disadvantages of this method include the need for a general anesthetic. Transjugular liver biopsy involves passing a catheter via a neck vein to the liver veins. There is lower risk of bleeding but a smaller tissue sample is obtained.

259

of data from a randomized controlled trial of adult patients with biopsy-proven NASH suggested that weight reduction of at least 5% over 24  weeks was required in order to observe significant decreases in liver fat (relative decrease of 25.5%) and volume [68]. Alcohol consumption is a risk factor for chronic liver disease and may impair resolution of steatosis and steatohepatitis [69, 70]. Beyond medical nutrition therapy and exercise counselling aimed at achieving target body weight bariatric surgery should be considered in selected patients. A recent systematic review and meta-analysis found that bariatric surgery can lead to resolution of NAFLD in obese patients [71]. However, some patients may develop new or worsening features of NAFLD post-operatively. Attention should be given to modifiable risk factors for cardiovascular disease. Several drugs have been demonstrated to be beneficial for NAFLD/NASH.  However, treatment options remain limited. Current recommendations for pharmacotherapy as determined by the AASLD clinical guidance are presented in Table  10.8 [33]. Besides these agents more than a dozen other miscellaneous agents have been investigated in small, proof-of-concept studies, and their detailed evaluation was considered to lie beyond the scope of the AASLD guidance. Guidance on pharmacotherapy differs in some respects between different expert bodies [55]. The 2016 NICE clinical guidance recommends consideration of pioglitazone or vitamin E for adults with advanced liver fibrosis whether they have diabetes or not. Advanced fibrosis is defined as grade F3 or above using the Kleiner NASH-­CRN system [72] or the steatosis, activCurrent and Emerging Therapeutic ity and fibrosis (SAF) score from the European Options for NAFLD Fatty Liver Inhibition of Progression Consortium The current mainstay of management of NAFLD [37, 73]. This is referred to as bridging fibrosis is lifestyle modification. In the largest paired (the presence of fibrosis linking hepatic veins to biopsy study to date of patients with histologi- portal tracts). Of note, this recommendation cally proven NASH, 12  months of a calorically does not specifically call for a liver biopsy but restricted diet (750 kcal/day) plus recommenda- advises on referral to a hepatologist. Neither piotions to walk 200  min/week resulted in a dose-­ glitazone nor vitamin E has marketing authorisaresponse relationship of weight loss to tion for the treatment of NAFLD and ‘off label’ histopathological improvement in inflammation, use remains at the discretion of the prescribing ballooning, and fibrosis [67]. Secondary analysis physician.

A. J. Krentz and P. Bedossa

260

Table 10.8  Drug therapy that may be considered for patients with NAFLD according to the AASLD Drug or class Biguanides

Thiazolidinediones

Vitamin E

Glucagon-like peptide-1 (GLP-1) agonists

Ursodeoxycholic acid

Omega-3 fatty acids

AASLD Recommendations Metformin improves aminotransferase levels and insulin sensitivity; does not improve liver histology in patients with NAFLD/NASH. Not recommended for adults with NASH. Pioglitazone improves liver histology in patients with or without type 2 diabetes with biopsy-proven NASH and may be used to treat these patients; risks and benefits should be discussed with patients before starting therapy. Until further data support its safety and efficacy, pioglitazone should not be used to treat NAFLD without biopsy-proven NASH. Vitamin E (α-tocopherol) administered at a daily dose of 800 IU/day improves liver histology in nondiabetic adults with biopsy-proven NASH and therefore may be considered for this patient population. Risks and benefits should be discussed with each patient before starting therapy. Until further data supporting its effectiveness become available, vitamin E is not recommended to treat NASH in diabetic patients, NAFLD without liver biopsy, NASH cirrhosis, or cryptogenic cirrhosis. Recent clinical trial evidence from 52 patients with biopsy-­proven NASH showed that once-daily liraglutide was associated with greater resolution of steatohepatitis and less progression of fibrosis over 24 weeks. It is premature to consider GLP-1 agonists to specifically treat liver disease in patients with NAFLD or NASH. Several studies have investigated ursodeoxycholic acid (UDCA; conventional and high doses) to improve aminotransferases and steatosis in patients with NAFLD and liver histology in patients with NASH. All but one study have been proof-of-concept studies with small numbers of participants and/or surrogate endpoints. UCDA is not recommended for the treatment of NAFLD or NASH. The interpretation of human studies of omega-3 fatty acids is limited by small sample size and methodological flaws. Moreover, two recently reported studies failed to show convincing therapeutic benefit for omega-3 fatty acids in patients with NAFLD or NASH. Omega-3 fatty acids should not be used as a specific treatment of NAFLD or NASH, but they may be considered to treat hypertriglyceridemia in patients with NAFLD.

Based on Chalasani et al. [33]

According to NICE, more than 80% of patients with NAFLD have normal routine liver blood tests. Accordingly, there is an urgent need for a simple, accessible, cost-effective, non-invasive test capable of case-finding NAFLD in the huge numbers of people at risk. The prevalence of NAFLD in the UK general population is estimated at 20–30% with 2–3% of the population having NASH. While the gold standard for diagnosis is liver biopsy, it is not feasible to perform liver biopsy in large numbers of at-risk patients, so magnetic resonance (MR) based techniques are increasingly used as the comparison in studies assessing non-invasive tests for NAFLD. These demonstrate high diagnostic accuracy; however, they are impractical or too expensive for large scale case finding.

 rugs in Development for  D NAFLD/NASH The drug development landscape of new therapies for NAFLD/NASH has recently been extensively reviewed elsewhere [33, 34, 55, 74–76]. It has been noted that a common therapeutic strategy is lacking, as evidenced by >30 different targets currently in development. In part, this may reflect deficiencies of disease understanding and limitations of experimental models for identifying and validating targets [77]. While a large number of animal models are currently available there is an ongoing challenge to identify models that closely mirror human pathology to ensure good translation of results into clinical development [78]. Emerging targets for drug development may involve either single agents or

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

261

Table 10.9  Drugs in phase 3 registration trials for NAFLD/NASH in late 2018 Mechanism Medication Elafibranor OCA CVC SEL

PPARα/δ agonist FXR agonist CCR2/CCR5 antagonist ASK1 inhibitor

Phase II efficacy data Resolution of Decrease in NASH fibrosis stage Yes No

RESOLVE-IT

No

Yes

REGENERATE

No

Yes

AURORA

No

Yesa

STELLAR 3 and 4

Phase III RCT

Effective dosage 120 mg/ day 10–25 mg/ day 150 mg/ day 6 and 18 mg/day

Planned interim analysis duration 72 weeks 72 weeks 52weeks 48 weeks

Based on Alkhouri and Scott [76] Abbreviations: OCA obeticholic acid, CVC cenicriviroc, SEL selonsertib, RCT randomised controlled trial. See text for definitions of other abbreviations a Numerically higher rates of fibrosis improvement that did not reach statistical significance. This was a proof-of-concept study that was not powered to detect histological changes in fibrosis stage

combination therapies intended to arrest or reverse disease progression [79]. Novel therapies are currently directed towards improving the metabolic status of the liver, cellular stress, apoptosis, inflammation, or fibrosis. Several agents are now in pivotal trials and there are expectations that the first therapies will be approved within 2–3 years [80]. Noteworthy phase 3 registration trials of agents for NASH include obeticholic acid (REGENERATE, NCT02548351), elafibranor (RESOLVE-IT, NCT02704403), cenicriviroc (AURORA, NCT03028740) and selonertib (STELLAR 3 and 4, NCT03053050 and NCT03053063, respectively) (Table  10.9). These trials were designed to target predominantly NASH (elafibranor), fibrosis (selonsertib), or both (OCA). Obeticholic acid, a farnesoid X receptor agonist with efficacy in preclinical models, improved steatohepatitis and fibrosis over a 72-week period in a large, multicenter, phase 2b trial in a dose of 25 mg daily (FLINT) [81]. The REGENERATE study was designed to evaluate the effect of obeticholic acid (10 mg or 25 mg) compared to placebo on liver histology in non-­cirrhotic NASH subjects with stage 2 or 3 fibrosis by assessing the following primary endpoints (time frame: measurements at baseline and 18  months). These included: the proportion of obeticholic acid treated patients relative to placebo achieving at least one stage of liver fibrosis improvement with no worsening of NASH, or the proportion of

obeticholic acid treated patients relative to placebo achieving NASH resolution with no worsening of liver fibrosis. In February 2019 Intercept Pharmaceuticals announced top-line results demonstrating a statistically significant (p  =  0.0002) improvement in liver fibosis (defined as equal to or greater than one stage) at 18 months with no worsening of NASH with 25 mg daily obeticholic acid. While a numerically greater proportion of obeticholic acid-treated patients in both treatment arms compared with placebo achieved resolution of NASH with no worsening of liver fibrosis this result did not achieve statistical significance. Obeticholic acid was approved in 2016 by the FDA for the treatment of primary biliary cholangitis in combination with ursodeoxycholic acid (UDCA) in adults with an inadequate response to UDCA, or as a single therapy in adults unable to tolerate UDCA. In a recent phase 2 study elafibranor, a dual PPARα/δ agonist, exhibited an efficacy signal for improving NASH without fibrosis worsening over 12-months [82]. In RESOLVE-IT, the primary outcome measure (time frame 72 weeks) is the proportion of elafibranor treated patients with NASH relative to placebo achieving NASH resolution without worsening of fibrosis. The AURORA study of cenicriviroc, an oral, dual CCR2/CCR5 antagonist, has two parts. Part 1 will examine the surrogate endpoint of improvement in fibrosis of at least 1 stage (NASH CRN) and no worsening of steatohepatitis at month 12. Subjects from Part 1 will continue into Part 2 and addi-

A. J. Krentz and P. Bedossa

262

tional subjects will be newly randomized in Part 2 to determine long-term clinical outcomes composed of histopathologic progression to cirrhosis, liver-related clinical outcomes, and all-cause mortality. Selonsertib, an inhibitor of apoptosis signal-­ regulating kinase (ASK)-1 inhibitor involved in stress responses, was studied in the STELLAR program. In early 2019 it was announced that the efficacy of selonsertib in patients with NASH and F4 fibosis had not achieved superiority to placebo in terms of improvement of fibrosis in STELLAR 4. The diverse mechanisms of action of drugs that have reached phase 3 trials reflects the complex and heterogeneous pathology of NAFLD/NASH for which combinations of therapies may be required in treatment strategies similar to those for managing type 2 diabetes [75, 76]. In the context of the increased risk of cardiovascular disease associated with NASH recent studies of glucagon-like peptide-1 (GLP-1) agonists and sodium-glucose co-transporter-2 (SGLT2) inhibitors in NASH are attracting interest. Both classes of glucose-lowering agents, which reduce body weight, have been shown to reduce cardiovascular events in high-risk patients with type 2 diabetes and are now being studied in non-diabetic populations [83]. In a recently published randomized, placebo-controlled trial involving 52 patients with biopsy-proven NASH, liraglutide 1.8  mg administered subcutaneously once-daily for 48  weeks was associated with greater resolution of steatohepatitis and less progression of fibrosis [84]. Liraglutide was associated with greater weight loss than placebo, but also with well-recognized gastrointestinal side effects. In animal models, limited data suggest that SGLT2 inhibitors may have potential to improve steatohepatitis [85]. While no liver biopsy studies of SGLT2 inhibitors have been reported to date empagliflozin (in addition to standard glucoselowering therapy) reduced hepatic fat measured by MRI-derived proton density fat fraction (MRIPDFF) in 50 patients with type 2 diabetes and NAFLD who were treated for 20 weeks [86]. By convention, successful phase 2a proof of concept studies of a NASH therapeutic are followed by phase 2b and phase 3 clinical trials that aim to confirm the therapeutic dose, verify efficacy via histology, and assess long-term clinical out-

comes. After initial approval, subsequent phase 3b studies evaluate longer term safety and efficacy. Adaptive clinical trial designs allow for planned modifications in one or more aspects of the study design based on response in earlier phases [34, 79]. In this scenario, pre-planned adaptations such as discontinuation of study arms, randomisation of additional patients, or premature study termination decisions are informed by interim data analyses. By permitting patients to transition through multiple phases of the study adaptive trials have potential to limit the total sample size of patients over the entire process of drug development. The FDA fast track process is designed to facilitate the development and expedite the review of drugs to treat serious conditions and fill an unmet medical need (see Chap. 19). FDA fast track designation can help speed up the pathway to the approval of therapies for NASH based on the phase 1 as well as preclinical data (https://www. fda.gov/forpatients/approvals/fast/default.htm).

Skeletal Muscle Skeletal muscle is an important organ in metabolic regulation. Under fasting conditions, when plasma insulin levels are low, non-esterified fatty acids (NEFA) serve as the principal fuel source in skeletal muscle [87]. Muscle is a major regulator of glucose metabolism accounting for the majority of insulinstimulated glucose uptake, e.g. during a hyperinsulinaemic euglycaemic clamp [88, 89]. Defects in whole-­body insulin sensitivity in subjects with type 2 diabetes have been attributed to impaired insulin action at the level of skeletal myocytes with reduced activity of glycogen synthase [90] and impaired glycogen synthesis [89, 91]. Diabetes is also associated with disturbances in muscle protein metabolism that may lead to decreased muscle mass [92].

 uscle Biopsy: Insights into M the Pathophysiology of Type 2 Diabetes Clinical studies using skeletal muscle biopsies have revealed alterations in cellular enzyme activity, phosphorylation, gene expression, and

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other… Table 10.10  Defects in muscle glucose metabolism in type 2 diabetes and other insulin-resistant states Insulin signaling

Glucose transport Intracellular glucose metabolism

↓ insulin receptor tyrosine phosphorylation ↓ IRS-1 tyrosine phosphorylation ↓ PI3-kinase activation ↓ GLUT4 translocation ↓ GLUT 12 translocation ↓ glucose phosphorylation ↓ glucose oxidation and glycolysis ↓ glycogen synthase activity

Based on Abdul-Ghani and DeFronzo [87] Abbreviations: IRS-1 insulin receptor substrate-1, PI3-­ kinase phosphatidylinositol 3-kinase, GLUT glucose transporter

protein expression in healthy subjects, patients with type 2 diabetes and individuals at elevated risk of developing type 2 diabetes (Table 10.10). Needle biopsy of skeletal muscle permits assessment of gene expression in disease states, e.g. type 2 diabetes, morphological changes, e.g. in obesity, and responses to interventions such as bariatric surgery or drug therapy.

 uscle Biopsy: Practical M Considerations Bergström introduced the eponymous percutaneous muscle biopsy technique in the 1960s thereby providing a simple and repeatable sampling method [93, 94] which has made skeletal muscle available for direct studies of morphology and metabolic functions. The Bergström needle consists of an outer cannula with a small opening at the side of the tip; an inner trocar contains a cutting blade at the distal end. Details of protocols using the suction-modified Bergström technique under local anesthesia and sample handing, including a step-by-step video guide, have been published [95–97]. The biopsy team consists of a trained operator and assisting technicians [97]. The procedure takes approximately 15–20 min in experienced hands, most of which is spent in preparation for the incision. Skeletal muscle biopsies are considered to be generally safe and well tolerated [97].

263

Complications include local discomfort, bruising, transient bleeding and infection. Subjects can typically resume their activities of daily living immediately afterwards and can fully participate in vigorous physical activity within 48–72  h of biopsy. Heavy resistance exercise should be avoided for 48 h. to reduce the risk of herniation of the muscle through the incision in the fascia [97]. Wounds usually heal with minimal scarring.

 orphological Studies of Skeletal M Muscle in Metabolic Disorders Mammalian skeletal muscles are composed of two major fiber types that differ in size, metabolism and contractile properties. As a rule, slow-­ twitch type I fibers are rich in mitochondria and have greater insulin sensitivity than fast-twitch type II muscle fibers [98, 99]. A lower proportion of type I muscle fibers and a higher proportion of type II, particularly IIb, fibers have been reported in subjects with type 2 diabetes [100, 101]. Fewer type I fibers and more mixed (type IIa) fibers have been reported in subjects with metabolic syndrome [102] which may be modulated by exercise intervention [103]. Palsgaard et al. studied gene expression profiles in skeletal muscle biopsies from Caucasian men with type 2 diabetes, healthy first-degree relatives, and healthy controls. Genes involved in insulin signaling were significantly upregulated in first degree relatives and significantly downregulated in subjects with type 2 diabetes [104]. Insulin stimulates microvascular perfusion – via capillary recruitment – of skeletal muscle (and subcutaneous adipose tissue) to increase blood flow after meals or physical exercise. This increases delivery of insulin and substrates to facilitate glucose disposal [105]. Some histochemical studies have shown a lower capillary density in skeletal muscle from subjects with type 2 diabetes [100]. Exercise training increases skeletal muscle capillary exchange and blood flow capacities. Traininginduced changes in the muscular vasculature and in insulin signaling within muscle fibers and the vasculature augment glucose and insulin delivery as well as glucose uptake [106].

264

Ectopic lipid deposition and sarcopenia are putative mechanisms of insulin resistance in skeletal muscle [107–109]. Inflammation within skeletal muscle is a reported feature of obesity manifested by increased immune cell infiltration and proinflammatory activation [110, 111]. Other pathological interactions between skeletal muscle and adipose tissue are supported by studies that show that fat-derived adipokines including leptin, resistin, visfatin and adiponectin affect skeletal muscle insulin sensitivity via insulin signalling pathways, affecting GLUT4 translocation, and by modulating insulin-mediated skeletal muscle glucose uptake [112]. Non-­ esterified fatty acids (NEFA) impair intracellular insulin-mediated glucose metabolism in muscle [113–115]. Interactions between NEFA and skeletal mitochondrial function have been postulated [116]. However, the therapeutic potential of lowering plasma NEFA concentrations on insulin action and glucose homeostasis remains uncertain. Phielix et al. observed that acipimox, an inhibitor of hormone-sensitive lipase, lowered NEFA concentrations and improved insulin sensitivity in subjects with patients with type 2 diabetes [117]. However, no effect of acipimox was observed on basal or insulin-stimulated muscle mitochondrial oxidative capacity measured in permeabilised fibers and isolated mitochondria. Makimura et al. studied the effects of 6 month’s treatment with acipimox on glucose homeostasis in obese adults [118]. In vivo skeletal mitochondrial function was assessed as post-exercise phosphocreatine recovery on 31P-magnetic resonance spectroscopy. In addition, paired percutaneous muscle needle biopsy was obtained (n = 11) from the lateral gastrocnemius muscle using a Bergström core biopsy needle under local anesthesia. Fasting glucose decreased in acipimox-­treated individuals but did not affect insulin-stimulated glucose uptake, as assessed by hyperinsulinaemic euglycaemic clamp technique. Acipimox significantly increased adiponectin levels but had no effect on mitochondrial function or mitochondrial density as assessed by electron microscopy, or on muscle insulin sensitivity. Abdominal surgical procedures, e.g. bariatric surgery, offer an opportunity to simultaneously

A. J. Krentz and P. Bedossa

biopsy muscle and other tissues to provide pathophysiological insights into complex multi-system disorders. In a cross-sectional study, liver and deltoid muscle biopsies were collected during bariatric surgery in NAFLD patients (n = 51) of whom 43% had NASH [119]. In these morbidly obese subjects, the presence of intramyocellular lipid was associated with NASH and advanced hepatic fibrosis. The investigators concluded that muscle mitochondrial dysfunction did not appear to be a major driving force contributing to either muscle fat accumulation, insulin resistance or liver disease. In an interventional study in subjects with morbid obesity with or without type 2 diabetes, omental visceral and subcutaneous adipose tissue and rectus abdominis muscle intraoperatively were excised [120]. One year after bariatric surgery percutaneous biopsies of subcutaneous fat and muscle were performed under local anaesthesia. For comparison, biopsies were also available for lean non-diabetic subjects undergoing non-bariatric abdominal surgery. Pre-­ surgery in the obese state, subcutaneous and visceral adipose tissue showed signs of fibrosis/ necrosis with small mitochondria, free interstitial lipids, and thickened capillary basement membrane compared with normal weight control subjects. In rectus abdominis, intramyocellular fat was more abundant pre-operatively in the subjects with obesity and type 2 diabetes. Following surgery-induced weight loss, most of the subcutaneous fat and muscle changes resolved and metabolic functions – insulin sensitivity, lipolysis, and β-cell function – improved.

 keletal Muscle as an Endocrine S Organ Skeletal muscle may be regarded as an endocrine organ that produces and releases myokines that exert endocrine effects on itself and other organs [121, 122]. Cachexia and sarcopenia are characterized by marked decreases in muscle protein content, myonuclear number, muscle fiber size, and muscle strength [123]. The discovery of the negative muscle mass regulator myostatin stimulated research activity into pharmacological

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

approaches to defective skeletal muscle structure and function in a range of diseases including age-­ related sarcopenic obesity [124, 125]. The latter condition is associated with frailty, physical disability, impaired insulin signaling, metabolic dsyregulation and premature mortality [125, 126]. Myostatin influences metabolism and pharmacological inhibition of myostatin can attenuate the progression of obesity and diabetes [125, 127]. During the past decade there has been a marked increase in potential targets for the pharmacological treatment of sarcopenia [128]. For example, bimagrumab is a human monoclonal antibody inhibitor of activin type II receptors (ActRII) that exerts anabolic actions on skeletal muscle mass by blocking binding of myostatin and other negative regulators of muscle growth. Metabolic benefits are hypothesized to accompany increased skeletal muscle mass. In a proof of concept study adults (n  =  16) with insulin resistance and/or glucose intolerance a single dose of bimagrumab increased lean mass and reduced fat mass compared with placebo, and had a neutral effect on body weight, reduced glycated hemoglobin, and improved measures of whole-body insulin sensitivity [129].

 yocyte and Myotube Culture M Techniques Human skeletal muscle cultures have been used to examine the contributions of hyperglycemia and hyperinsulinemia to the defective muscle glycogen synthase activity [130]. Human myotubes are primary skeletal muscle cells displaying both morphological and biochemical characteristics of mature skeletal muscle [131, 132]. Satellite cells are isolated from skeletal muscle biopsies, activated to proliferating myoblasts and differentiated into multinuclear myotubes. These cell cultures provide a model for intact human skeletal muscle which allows assessment of interventions [133]. For example, Costford et al. took biopsies from the vastus lateralis in (a) seven obese subjects with or without type 2 diabetes who had completed a standardised weight loss protocol and (b) seven non-diabetic

265

participants using a 5  mm Bergström needle [134]. All participants were euglycemic at the time of study and had similar insulin sensitivity. Satellite cells were isolated from muscle biopsies and differentiated under low or high glucose and insulin conditions. Cells from participants with no history of type 2 diabetes showed increases in mitochondrial content, citrate synthase and cytochrome c oxidase activities when exposed to high levels of glucose and insulin. This increase in oxidative capacity was absent in cells from patients with a history of type 2 diabetes. High glucose and insulin caused increased oxidative damage in cells from the latter subjects, despite higher superoxide dismutase expression. Cells from subjects with a history of type 2 diabetes were unable to decrease mitochondrial membrane potential in response to glucose and insulin. It was concluded that primary myotubes from subjects with a history of type 2 diabetes were unable to adapt to a hyperglycemic–hyperinsulinemic challenge with impaired mitochondrial biogenesis and an inability to manage oxidative stress defining a muscle phenotype at risk of obesity-­associated type 2 diabetes. Whether these phenotypic characteristics were genetically programmed or acquired epigenetically was not determined in this study. Altered profiles of proteins associated with protein dynamics and gene regulation have been reported in myotubes derived from patients with type 2 diabetes from whom biopsies were obtained from the vastus lateralis portion of the quadriceps femoris muscle [135]. Proteins involved in fatty acid and amino acid metabolism, tricarboxylic acid cycle, mitochondrial function, mRNA processing, DNA repair and cell survival showed higher abundance, whereas proteins associated with redox signaling, glutathione metabolism and protein dynamics showed reduced abundance in myotubes derived from subjects with type 2 diabetes versus non-diabetic donors. There have been calls for greater use of in  vitro models based on human muscle, as opposed to rodent muscle, for functional validation of factors such as adipokines on skeletal muscle insulin sensitivity in subjects with insulin resistance or type 2 diabetes. In this context, as in

266

other aspects of drug development, research focused on human tissue may facilitate the translation of basic science into novel pharmacotherapies [112].

A. J. Krentz and P. Bedossa

like structures in the subcutaneous fat of patients with obesity and the metabolic syndrome [143]. Bariatric surgery may reverse both macrophage infiltration and the altered secretory profile observed in the adipose tissue of subjects with severe obesity [144]. However, knowledge Adipose Tissue remains incomplete about aspects of fat physiology including the in vivo kinetics of adipose tisWhite adipose tissue, like skeletal muscle, is an sue and its components. In part this reflects the important and multi-faceted regulator of whole-­ slow turnover rate of adipose tissue and the combody metabolism and energy balance. Obesity, plexity of directly labeling pathway precursors. usually defined by BMI, is a major risk factor for highly prevalent disorders including type 2 diabetes and cardiovascular disease. Adipose Percutaneous Adipose Tissue Biopsy tissue distribution, which is regulated by Techniques numerous factors including sex steroid hormones, is a well-recognized determinant of car- Adipose tissue biopsy offers the clinical investidiometabolic health. Excess intra-abdominal gator an opportunity to identify the metabolic adipose tissue accumulation  – visceral obe- roles of adipocytes within their respective depots sity  – is part of an unhealthy phenotype that and associations with factors such as gender, age, includes dysfunctional subcutaneous adipose ethnicity, and risk of cardiometabolic disorders tissue expansion and ectopic triglyceride stor- [145]. Biopsy of white adipose tissue may also be age closely related to a clustering of cardiomet- of value in the development of pharmacotheraabolic disease risk factors [136–139]. pies for obesity and related cardiometabolic disPhysiological characteristics of abdominal adi- orders particularly with respect to understanding pose tissues, e.g. adipocyte size and number, relevant cellular mechanisms of observed drug lipolytic responsiveness, lipid storage capacity, effects. and inflammatory cytokine production correlate The standard procedure is open biopsy or with, and may determine, the increased cardio- needle aspiration of subcutaneous fat under metabolic risk associated with visceral obesity sterile conditions and local anesthesia. For nee[136]. Adipose tissue is a source of adipokines dle biopsy a vacuum is generated by pulling and numerous other paracrine and endocrine back the syringe plunger. Visceral adipose tisfactors [140, 141]. sue may be obtained during abdominal surgical Adipose tissue mass is determined by procedures along with subcutaneous fat for dynamic changes in the synthesis and break- comparative studies of these depots [146–148]. down of adipocytes and triglycerides/triacylg- Visceral adipose tissue has been shown to be lycerols [142]. Adipose tissue is plastic and relatively resistant to the inhibitory actions of adaptable; expansion accommodates chronic insulin on lipolysis and the stimulatory effects excess energy intake and is characterised by of insulin on re-esterification as well as expressenlargement of existing adipocytes, i.e. hyper- ing more genes encoding pro-­ inflammatory trophy, and by an increase in numbers of pre-­ cytokines compared with subcutaneous adipose adipocyte and adipocytes, i.e. hyperplasia [142]. tissue [148, 149]. Multiple mechanisms mediate the association Analysis of adipose tissue gene expression between adiposity and cardiometabolic disease. provides a means to discover the gene targets relEctopic lipid deposition in the liver and skeletal evant to the understanding the pathogenesis and muscle contributes to the pathogenesis of treatment of obesity. Numerous studies have obesity-­ related disorders [107]. In addition, sought to identify new candidate genes and related there is an increase in macrophages and crown- gene products based on adipose tissue gene

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other… Table 10.11 Example procedure for subcutaneous abdominal tissue biopsy in a phase 1b pharmacotherapeutic study Biopsies will be taken from subcutaneous abdominal tissue, approximately 10 cm lateral to the umbilicus, alternating sides in paired biopsies and using a different site for each biopsy (baseline and after intervention). The skin will be cleaned with iodine (betadine) and a sterile technique will be used. The skin and underlying tissue will be anesthetized using lidocaine 1% or lidocaine 1% with epinephrine (approximately 5–10 ml). An incision will be made in the skin with a sterile scalpel and the tissue will be carefully acquired by an excision technique using a scalpel and a pair of tweezers. The tissue will be acquired and immediately blotted on a nylon mesh and cleaned from blood with saline. This should result in a clean tissue preparation of approximately 1.5–2.0 g. The tissue will be put into a sterile tissue container and will be transported immediately to the laboratory. The area may be closed with stitches, Steristrips, and/or sterile dressing. A cauterization pen may be used to control bleeding. Ecchymosis (bruising) will be minimized by ~20 min of compression of the biopsy site just after the biopsy has been taken.

expression signatures and the influence of modulating factors, e.g. dietary modifications, physical activity, as well as pharmacotherapy. Of note, however, limited data suggest that the biopsy sampling method – needle-aspirated or surgical – may influence gene expression in sampled tissue [150]. A comparative microarray analysis of subcutaneous adipose tissue sampling methods was performed in age-matched lean (n = 19) and obese (n = 18) female subjects. Biopsy technique influenced gene expression relevant to obesity research, e.g. inflammation, extracellular matrix, and metabolism while immunohistochemistry experiments showed that needle-aspirated biopsies (which are easier to obtain) poorly aspirate the fibrotic fraction of subcutaneous adipose tissue, resulting in an underrepresentation of the stroma-vascular fraction. Table 10.11 presents an example from a phase 1b clinical study of an insulin sensitizing drug in obese subjects with hepatic steatosis in which the effects of the medication on subcutaneous adipose tissue were of interest.

267

 ulture of Adipocytes and Adipose C Tissue In contrast to adipocyte culture, adipose tissue cultures contain other cells including preadipocytes, endothelial cells, and fibroblasts and various immune cells [145]. The latter is considered to be the preferred method when assessing the long-term regulation of gene expression and adipocyte function. Variability between adipocyte isolation and culture techniques may impact experimental results [145].

 dipocyte Size and Metabolic A Function Weight loss induced by lifestyle changes generally induces preferential mobilization of visceral fat [136]. Furthermore, obese individuals, weight reduction is also associated with a decrease in adipocyte size; this effect is more pronounced in upper-body adipocytes [151]. Drugs including thiazolidinediones, estrogen replacement in postmenopausal women, and testosterone replacement in androgen-deficient men have been shown to favorably modulate body fat distribution and risk of cardiometabolic disease [136]. Interestingly, thiazolidinediones increase wholebody adiposity while improving insulin sensitivity [152]. Clinical data indicate that thiazolidinediones are associated with a favorable shift in fat distribution from visceral to subcutaneous adipose depots [153]. The increase in adipose tissue mass occurs in the context of qualitative changes in adipose tissue, including the remodeling of adipocytes to a smaller size with higher lipid storage potential [154]. Hammarstedt et al. studied non-diabetic subjects (n  =  10) with reduced insulin receptor substrate (IRS)-1 and GLUT-4 protein in adipose cells as markers of insulin resistance [155]. After 3  weeks of treatment with pioglitazone, a hyperinsulinaemic euglycaemic clamp was performed and biopsies were taken from abdominal subcutaneous adipose tissue. Pioglitazone improved whole-body insulin sensitivity in the absence of any changes in circulating NEFA or lipid levels. Markers of adipose

268

cell differentiation, previously shown to be reduced in insulin resistance including adiponectin, adipocyte P 2, uncoupling protein 2, GLUT4 and liver X receptor-α increased, supporting the concept that insulin resistance was associated with impaired terminal differentiation of the adipose cells in the study subjects [155]. In another study, McLaughlin et  al. studied 12 overweight/ obese nondiabetic, insulin-resistant individuals who underwent biopsy of abdominal subcutaneous adipose tissue at baseline and after 12 weeks of pioglitazone treatment [156]. Subcutaneous abdominal fat biopsy was performed after an overnight fast. Under local anesthetic and sterile conditions, biopsy was performed via scalpel incision inferior to the umbilicus. An increase in the ratio of small-to-­large cells as well as a 25% increase in the absolute number of small cells was interpreted as evidence of suggestive of adipogenesis in abdominal subcutaneous adipose tissue. Furthermore, CT scanning demonstrated redistribution of fat from visceral to subcutaneous depots [156]. Adipose tissue may be useful in exploring the correlation between drug-induced gene expression and circulating mediators of metabolic effects. Adiponectin is an adipokine with insulin-­ sensitizing, anti-diabetic, anti-inflammatory, and anti-atherosclerotic properties [157]. Rasouli et al. took adipose tissue biopsies from the lower abdominal wall before and after treatment with pioglitazone in subjects with normal or impaired glucose tolerance (IGT) [158]. The subjects with IGT had lower levels of adiponectin and adiponectin mRNA compared with normal subjects. Plasma levels and secretion of adiponectin increased in response to pioglitazone in concert with improved insulin sensitivity (assessed using an insulin-­ modified intravenous glucose tolerance test) without a concomitant increase in adiponectin adipocyte mRNA expression suggesting post-­transcriptional regulation of adiponectin by the thiazolidinedione. Characterization of dose-response relationships of new therapies at an early stage of development can support rational selection of doses and regimens for subsequent studies. For example, adipose tissue enzyme 11β-hydroxysteroid

A. J. Krentz and P. Bedossa

dehydrogenase type 1 (11β-HSD1) is a target for therapeutic interventions aimed at improving insulin sensitivity in subjects with type 2 diabetes and for the treatment of NAFLD [159, 160]. The enzyme catalyses the intracellular conversion of inert cortisone to physiologically active cortisol, thereby enhancing local cortisol action. Gibbs et  al. examined pharmacokinetic/pharmacodynamic properties of AMG 221, a small molecule selective inhibitor of 11β-HSD1  in ex  vivo adipose tissue samples obtained at open biopsy in healthy volunteers [161]. Using single doses of 3, 30, or 100 mg of oral AMG 221 or placebo the investigators were able to relate plasma drug levels to enzyme inhibition in adipose tissue and to demonstrate that AMG 221 potently blocked 11β-HSD1 activity over a 24-h period.

White, Brown, and Beige Adipocytes Sustained excessive caloric intake results in accumulation and storage of lipid in white adipocytes. In contrast, energy expenditure by fat oxidation predominately occurs in brown adipose tissue (BAT) (see Chap. 8). Recently, the existence of a third type of fat, referred to as beige or brite (brown in white), has been recognized [162]. It has been suggested that white adipocytes can undergo browning in response to stimuli that induce and enhance the expression of thermogenes characteristic of those typically associated with brown fat. Some studies suggest that contracting skeletal muscles release ­myokines that promote browning of white adipocytes [163]. Irisin has been identified as a myokine putative mediator of this effect [164]. Recognition of cross-talk between exercising muscle and the metabolic activity of adipocytes opens the possibility of pharmacotherapeutic approaches that promote browning of white adipocytes as a new potential strategy for the treatment of obesity and obesity-­associated metabolic disorders [165–167]. Adipose tissue biopsy can be expected to play a role in elucidating physiological regulators of browning and effects of target genes, e.g. uncoupling protein-1 [168] and the development of novel metabolic drugs.

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other…

Current research aims to identify cell population in human fat depots that can undergo efficient thermogenic transformation in response to pharmacological stimuli [169]. Even if new therapeutic interventions achieve expansion and activation of BAT the translation of this effect into clinically relevant changes in body weight and improved glucose metabolism remains to be tested. Inducing brown adipocyte biogenesis through pharmacology along with engineered tissue transplantation are other options that are currently being explored [170].

References 1. Brunt EM, Wong VW, Nobili V, et al. Nonalcoholic fatty liver disease. Nat Rev Dis Primers. 2015;1:15080. 2. Stefan N, Haring HU, Cusi K.  Non-alcoholic fatty liver disease: causes, diagnosis, cardiometabolic consequences, and treatment strategies. Lancet Diabetes Endocrinol. 2018. https://doi.org/10.1016/ S2213-8587(18)30154-2. 3. Bellentani S.  The epidemiology of non-alcoholic fatty liver disease. Liver Int. 2017;37(Suppl 1):81–4. 4. Younossi Z, Anstee QM, Marietti M, et  al. Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol. 2018;15(1):11–20. 5. Williams CD, Stengel J, Asike MI, et al. Prevalence of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis among a largely middle-aged population utilizing ultrasound and liver biopsy: a prospective study. Gastroenterology. 2011;140(1):124–31. 6. Stine JG, Wentworth BJ, Zimmet A, et al. Systematic review with meta-analysis: risk of hepatocellular carcinoma in non-alcoholic steatohepatitis without cirrhosis compared to other liver diseases. Aliment Pharmacol Ther. 2018;48(7):696–703. 7. Fassio E, Alvarez E, Dominguez N, Landeira G, Longo C. Natural history of nonalcoholic steatohepatitis: a longitudinal study of repeat liver biopsies. Hepatology. 2004;40(4):820–6. 8. Vernon G, Baranova A, Younossi ZM.  Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults. Aliment Pharmacol Ther. 2011;34(3):274–85. 9. Singh S, Allen AM, Wang Z, et al. Fibrosis progression in nonalcoholic fatty liver vs nonalcoholic steatohepatitis: a systematic review and meta-analysis of paired-biopsy studies. Clin Gastroenterol Hepatol. 2015;13(4):643–54 e1–9; quiz e39–40. 10. Wong VW, Wong GL, Choi PC, et al. Disease progression of non-alcoholic fatty liver disease: a pro-

269

spective study with paired liver biopsies at 3 years. Gut. 2010;59(7):969–74. 11. Koehler E, Watt K, Charlton M. Fatty liver and liver transplantation. Clin Liver Dis. 2009;13(4):621–30. 12. Parikh ND, Marrero WJ, Wang J, et  al. Projected increase in obesity and non-alcoholic-steatohepatitis-­ related liver transplantation waitlist additions in the United States. Hepatology. 2017. https://doi. org/10.1002/hep.29473. 13. Yki-Jarvinen H. Non-alcoholic fatty liver disease as a cause and a consequence of metabolic syndrome. Lancet Diabetes Endocrinol. 2014;2:901–10. 14. Roden M.  Mechanisms of disease: hepatic steatosis in type 2 diabetes--pathogenesis and c­linical relevance. Nat Clin Pract Endocrinol Metab. 2006;2(6):335–48. 15. Katsiki N, Mikhailidis DP, Mantzoros CS.  Non-­ alcoholic fatty liver disease and dyslipidemia: an update. Metabolism. 2016;65(8):1109–23. 16. Byrne CD, Targher G. NAFLD: a multisystem disease. J Hepatol. 2015;62(1 Suppl):S47–64. 17. Neuschwander-Tetri BA, Clark JM, Bass NM, et  al. Clinical, laboratory and histological associations in adults with nonalcoholic fatty liver disease. Hepatology. 2010;52(3):913–24. 18. Leite NC, Salles GF, Araujo AL, Villela-Nogueira CA, Cardoso CR.  Prevalence and associated factors of non-alcoholic fatty liver disease in patients with type-2 diabetes mellitus. Liver Int. 2009;29(1):113–9. 19. Subichin M, Clanton J, Makuszewski M, et  al. Liver disease in the morbidly obese: a review of 1000 consecutive patients undergoing weight loss surgery. Surg Obes Relat Dis. 2015;11(1): 137–41. 20. Radaelli MG, Martucci F, Perra S, et  al. NAFLD/ NASH in patients with type 2 diabetes and related treatment options. J Endocrinol Investig. 2018;41(5):509–21. 21. Mantovani A, Zaza G, Byrne CD, et al. Nonalcoholic fatty liver disease increases risk of incident chronic kidney disease: a systematic review and meta-­ analysis. Metabolism. 2018;79:64–76. 22. Dunn W, Xu R, Wingard DL, et al. Suspected nonalcoholic fatty liver disease and mortality risk in a population-based cohort study. Am J Gastroenterol. 2008;103(9):2263–71. 23. Targher G, Byrne CD, Lonardo A, Zoppini G, Barbui C. Non-alcoholic fatty liver disease and risk of incident cardiovascular disease: a meta-analysis. J Hepatol. 2016;65(3):589–600. 24. Seko Y, Yamaguchi K, Itoh Y.  The genetic backgrounds in nonalcoholic fatty liver disease. Clin J Gastroenterol. 2018;11(2):97–102. 25. Ludwig J, Viggiano TR, McGill DB, Oh BJ. Nonalcoholic steatohepatitis: Mayo Clinic experiences with a hitherto unnamed disease. Mayo Clin Proc. 1980;55(7):434–8. 26. Argo CK, Caldwell SH.  Epidemiology and natural history of non-alcoholic steatohepatitis. Clin Liver Dis. 2009;13(4):511–31.

270 27. Caldwell S, Argo C.  The natural history of non-­ alcoholic fatty liver disease. Dig Dis. 2010;28(1):162–8. 28. Angulo P, Kleiner DE, Dam-Larsen S, et  al. Liver fibrosis, but no other histologic features, is associated with long-term outcomes of patients with nonalcoholic fatty liver disease. Gastroenterology. 2015;149(2):389–97 e10. 29. Hubscher SG.  Histological assessment of non-­ alcoholic fatty liver disease. Histopathology. 2006;49(5):450–65. 30. Kleiner DE.  Histopathology, grading and staging of nonalcoholic fatty liver disease. Minerva Gastroenterol Dietol. 2018;64(1):28–38. 31. Bedossa P.  Diagnosis of non-alcoholic fatty liver disease/non-alcoholic steatohepatitis: why liver biopsy is essential. Liver Int. 2018;38(Suppl 1):64–6. 32. Sanyal AJ, Chalasani N.  Trials and tribulations in drug development for nonalcoholic steatohepatitis. Clin Gastroenterol Hepatol. 2014;12(12):2104–5. 33. Chalasani N, Younossi Z, Lavine JE, et  al. The diagnosis and management of nonalcoholic fatty liver disease: practice guidance from the American Association for the Study of Liver Diseases. Hepatology. 2018;67(1):328–57. 34. Konerman MA, Jones JC, Harrison SA.  Pharmacotherapy for NASH: current and emerging. J Hepatol. 2018;68(2):362–75. 35. Siddiqui MS, Harrison SA, Abdelmalek MF, et  al. Case definitions for inclusion and analysis of endpoints in clinical trials for nonalcoholic steatohepatitis through the lens of regulatory science. Hepatology. 2018;67(5):2001–12. 36. Brunt EM, Janney CG, Di Bisceglie AM, Neuschwander-Tetri BA, Bacon BR.  Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol. 1999;94(9):2467–74. 37. Bedossa P, Poitou C, Veyrie N, et al. Histopathological algorithm and scoring system for evaluation of liver lesions in morbidly obese patients. Hepatology. 2012;56(5):1751–9. 38. Younossi ZM, Stepanova M, Rafiq N, et  al. Pathologic criteria for nonalcoholic steatohepatitis: interprotocol agreement and ability to predict liver-­ related mortality. Hepatology. 2011;53(6):1874–82. 39. Neuschwander-Tetri BA, Brunt EM, Wehmeier KR, Oliver D, Bacon BR.  Improved nonalcoholic steatohepatitis after 48 weeks of treatment with the PPAR-gamma ligand rosiglitazone. Hepatology. 2003;38(4):1008–17. 40. Lindor KD, Kowdley KV, Heathcote EJ, et  al. Ursodeoxycholic acid for treatment of nonalcoholic steatohepatitis: results of a randomized trial. Hepatology. 2004;39(3):770–8. 41. Ratziu V, Giral P, Jacqueminet S, et al. Rosiglitazone for nonalcoholic steatohepatitis: one-year results of the randomized placebo-controlled Fatty Liver

A. J. Krentz and P. Bedossa Improvement with Rosiglitazone Therapy (FLIRT) Trial. Gastroenterology. 2008;135(1):100–10. 42. Lutchman G, Modi A, Kleiner DE, et  al. The effects of discontinuing pioglitazone in patients with nonalcoholic steatohepatitis. Hepatology. 2007;46(2):424–9. 43. Dixon JB, Anderson M, Cameron-Smith D, O’Brien PE. Sustained weight loss in obese subjects has benefits that are independent of attained weight. Obes Res. 2004;12(11):1895–902. 44. Pillai AA, Rinella ME. Non-alcoholic fatty liver disease: is bariatric surgery the answer? Clin Liver Dis. 2009;13(4):689–710. 45. Vilar-Gomez E, Chalasani N.  Non-invasive assessment of non-alcoholic fatty liver disease: ­clinical prediction rules and blood-based biomarkers. J Hepatol. 2018;68(2):305–15. 46. Poynard T, Munteanu M, Charlotte F, et  al. Diagnostic performance of a new noninvasive test for nonalcoholic steatohepatitis using a simplified histological reference. Eur J Gastroenterol Hepatol. 2018;30(5):569–77. 47. Sanyal AJ, Friedman SL, McCullough AJ, et  al. Challenges and opportunities in drug and biomarker development for nonalcoholic steatohepatitis: findings and recommendations from an American Association for the Study of Liver Diseases-­ U.S. Food and Drug Administration Joint Workshop. Hepatology. 2015;61(4):1392–405. 48. Byrne CD, Targher G.  Time to replace assessment of liver histology with MR-based imaging tests to assess efficacy of interventions for nonalcoholic fatty liver disease. Gastroenterology. 2016;150(1):7–10. 49. Li Q, Dhyani M, Grajo JR, Sirlin C, Samir AE. Current status of imaging in nonalcoholic fatty liver disease. World J Hepatol. 2018;10(8):530–42. 50. Brunt EM, Tiniakos DG.  Histopathology of nonalcoholic fatty liver disease. World J Gastroenterol. 2010;16(42):5286–96. 51. European Association for the Study of the Liver, European Association for the Study of Diabetes, European Association for the Study of Obesity. EASL EASD EASO clinical practice guidelines for the management of non-alcoholic fatty liver disease. J Hepatol. 2016;64(6):1388–402. 52. Byrne CD, Targher G.  EASL-EASD-EASO clinical practice guidelines for the management of non-­ alcoholic fatty liver disease: is universal screening appropriate? Diabetologia. 2016;59(6):1141–4. 53. National Institute for Health and Clinical Excellence. Non-alcoholic fatty liver disease (NAFLD): assessment and management. NICE guideline [NG49]. 2016. 54. Glen J, Floros L, Day C, Pryke R, Guideline Development Group. Non-alcoholic fatty liver disease (NAFLD): summary of NICE guidance. BMJ. 2016;354:i4428. 55. Sumida Y, Yoneda M. Current and future pharmacological therapies for NAFLD/NASH. J Gastroenterol. 2018;53(3):362–76.

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other… 56. Menghini G. One-second needle biopsy of the liver. Gastroenterology. 1958;35(2):190–9. 57. Ghent CN. Percutaneous liver biopsy: reflections and refinements. Can J Gastroenterol. 2006;20(2):75–9. 58. Spengler EK, Loomba R.  Recommendations for diagnosis, referral for liver biopsy, and treatment of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. Mayo Clin Proc. 2015;90(9):1233–46. 59. Gunn NT, Shiffman ML. The use of liver biopsy in nonalcoholic fatty liver disease: when to biopsy and in whom. Clin Liver Dis. 2018;22(1):109–19. 60. Hannah WN Jr, Torres DM, Harrison SA.  Nonalcoholic Steatohepatitis and endpoints in clinical trials. Gastroenterol Hepatol (N Y). 2016;12(12):756–63. 61. Sun TL, Liu Y, Sung MC, et  al. Ex vivo imaging and quantification of liver fibrosis using second-­ harmonic generation microscopy. J Biomed Opt. 2010;15(3):036002. 62. Scupakova K, Soons Z, Ertaylan G, et  al. Spatial systems lipidomics reveals nonalcoholic fatty liver disease heterogeneity. Anal Chem. 2018;90(8):5130–8. 63. Pinelo E, Presa J.  Outpatient percutaneous liver biopsy: still a good option. Eur J Intern Med. 2009;20(5):487–9. 64. Rockey DC, Caldwell SH, Goodman ZD, et al. Liver biopsy. Hepatology. 2009;49(3):1017–44. 65. Cadranel JF, Rufat P, Degos F.  Practices of liver biopsy in France: results of a prospective nationwide survey. For the Group of Epidemiology of the French Association for the Study of the Liver (AFEF). Hepatology. 2000;32(3):477–81. 66. Barbois S, Arvieux C, Leroy V, et  al. Benefit-risk of intraoperative liver biopsy during bariatric surgery: review and perspectives. Surg Obes Relat Dis. 2017;13(10):1780–6. 67. Vilar-Gomez E, Martinez-Perez Y, Calzadilla-Bertot L, et  al. Weight loss through lifestyle modification significantly reduces features of nonalcoholic steatohepatitis. Gastroenterology. 2015;149(2):367–78 e5; quiz e14–5. 68. Patel NS, Doycheva I, Peterson MR, et al. Effect of weight loss on magnetic resonance imaging estimation of liver fat and volume in patients with nonalcoholic steatohepatitis. Clin Gastroenterol Hepatol. 2015;13(3):561–568 e1. 69. Boyle M, Masson S, Anstee QM. The bidirectional impacts of alcohol consumption and the metabolic syndrome: cofactors for progressive fatty liver disease. J Hepatol. 2018;68(2):251–67. 70. Ajmera V, Belt P, Wilson LA, et al. Among patients with nonalcoholic fatty liver disease, modest alcohol use is associated with less improvement in histologic steatosis and steatohepatitis. Clin Gastroenterol Hepatol. 2018;16(9):1511–1520 e5. 71. Lee Y, Doumouras AG, Yu J, et al. Complete resolution of nonalcoholic fatty liver disease after

271

bariatric surgery: a systematic review and metaanalysis. Clin Gastroenterol Hepatol. 2018. https:// doi.org/10.1016/j.cgh.2018.10.017. 72. Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005;41(6):1313–21. 73. Bedossa P, Consortium FP.  Utility and appropriateness of the fatty liver inhibition of progression (FLIP) algorithm and steatosis, activity, and fibrosis (SAF) score in the evaluation of biopsies of nonalcoholic fatty liver disease. Hepatology. 2014;60(2):565–75. 74. Banini BA, Sanyal AJ. Current and future pharmacologic treatment of nonalcoholic steatohepatitis. Curr Opin Gastroenterol. 2017;33(3):134–41. 75. Alkhouri N, Poordad F, Lawitz E.  Management of nonalcoholic fatty liver disease: lessons learned from type 2 diabetes. Hepatol Commun. 2018;2(7):778–85. 76. Alkhouri N, Scott A.  An update on the pharmacological treatment of nonalcoholic fatty liver disease: beyond lifestyle modications. Clin Liver Dis. 2018;11:82–6. 77. Oseini AM, Cole BK, Issa D, Feaver RE, Sanyal AJ.  Translating scientific discovery: the need for preclinical models of nonalcoholic steatohepatitis. Hepatol Int. 2018;12(1):6–16. 78. Jahn D, Kircher S, Hermanns HM, Geier A. Animal models of NAFLD from a hepatologist’s point of view. Biochim Biophys Acta Mol basis Dis. 2018. https://doi.org/10.1016/j.bbadis.2018.06.023. 79. Friedman SL, Neuschwander-Tetri BA, Rinella M, Sanyal AJ.  Mechanisms of NAFLD development and therapeutic strategies. Nat Med. 2018;24(7):908–22. 80. Issa D, Patel V, Sanyal AJ. Future therapy for non-­ alcoholic fatty liver disease. Liver Int. 2018;38(Suppl 1):56–63. 81. Neuschwander-Tetri BA, Loomba R, Sanyal AJ, et  al. Farnesoid X nuclear receptor ligand obeticholic acid for non-cirrhotic, non-alcoholic steatohepatitis (FLINT): a multicentre, randomised, placebo-­ controlled trial. Lancet. 2015;385(9972):956–65. 82. Ratziu V, Harrison SA, Francque S, et  al. Elafibranor, an agonist of the peroxisome proliferator-activated receptor-alpha and -delta, induces resolution of nonalcoholic steatohepatitis without fibrosis worsening. Gastroenterology. 2016;150(5):1147–1159 e5. 83. Newman JD, Vani AK, Aleman JO, et al. The changing landscape of diabetes therapy for cardiovascular risk reduction: JACC State-of-the-Art Review. J Am Coll Cardiol. 2018;72(15):1856–69. 84. Armstrong MJ, Gaunt P, Aithal GP, et al. Liraglutide safety and efficacy in patients with non-alcoholic steatohepatitis (LEAN): a multicentre, double-blind, randomised, placebo-controlled phase 2 study. Lancet. 2016;387(10019):679–90.

272 85. Honda Y, Imajo K, Kato T, et  al. The selective SGLT2 inhibitor ipragliflozin has a therapeutic effect on nonalcoholic steatohepatitis in mice. PLoS One. 2016;11(1):e0146337. 86. Kuchay MS, Krishan S, Mishra SK, et al. Effect of empagliflozin on liver fat in patients with type 2 diabetes and nonalcoholic fatty liver disease: a randomized controlled trial (E-LIFT trial). Diabetes Care. 2018;41(8):1801–8. 87. Abdul-Ghani MA, DeFronzo RA.  Pathogenesis of insulin resistance in skeletal muscle. J Biomed Biotechnol. 2010;2010:476279. 88. Bjornholm M, Zierath JR.  Insulin signal transduction in human skeletal muscle: identifying the defects in Type II diabetes. Biochem Soc Trans. 2005;33(Pt 2):354–7. 89. Defronzo RA.  Banting Lecture. From the triumvirate to the ominous octet: a new paradigm for the treatment of type 2 diabetes mellitus. Diabetes. 2009;58(4):773–95. 90. Hojlund K, Beck-Nielsen H.  Impaired glycogen synthase activity and mitochondrial dysfunction in skeletal muscle: markers or mediators of insulin resistance in type 2 diabetes? Curr Diabetes Rev. 2006;2(4):375–95. 91. Hojlund K, Staehr P, Hansen BF, et  al. Increased phosphorylation of skeletal muscle glycogen synthase at NH2-terminal sites during physiological hyperinsulinemia in type 2 diabetes. Diabetes. 2003;52(6):1393–402. 92. Workeneh B, Bajaj M. The regulation of muscle protein turnover in diabetes. Int J Biochem Cell Biol. 2013;45(10):2239–44. 93. Bergström J.  Muscle electrolytes in man, determined by neutron activation analysis on needle biopsy specimens, a study in normal subjects, kidney patients, and patients with chronic diarrhoea. Scand J Clin Lab Invest. 1962;14:1–110. 94. Bergstrom J. Percutaneous needle biopsy of skeletal muscle in physiological and clinical research. Scand J Clin Lab Invest. 1975;35(7):609–16. 95. Hojlund K, Yi Z, Hwang H, et al. Characterization of the human skeletal muscle proteome by one-­ dimensional gel electrophoresis and HPLC-ESI-MS/ MS. Mol Cell Proteomics. 2008;7(2):257–67. 96. Tarnopolsky MA, Pearce E, Smith K, Lach B. Suction-modified Bergstrom muscle biopsy technique: experience with 13,500 procedures. Muscle Nerve. 2011;43(5):717–25. 97. Shanely RA, Zwetsloot KA, Triplett NT, et  al. Human skeletal muscle biopsy procedures using the modified Bergstrom technique. J Vis Exp. 2014;10(91):51812. 98. Duan Y, Li F, Tan B, Yao K, Yin Y.  Metabolic control of myofibers: promising therapeutic target for obesity and type 2 diabetes. Obes Rev. 2017;18(6):647–59. 99. Albers PH, Pedersen AJ, Birk JB, et  al. Human muscle fiber type-specific insulin signaling: impact of obesity and type 2 diabetes. Diabetes. 2015;64(2):485–97.

A. J. Krentz and P. Bedossa 100. Marin P, Andersson B, Krotkiewski M, Bjorntorp P.  Muscle fiber composition and capillary density in women and men with NIDDM.  Diabetes Care. 1994;17(5):382–6. 101. Gaster M, Staehr P, Beck-Nielsen H, Schroder HD, Handberg A.  GLUT4 is reduced in slow muscle fibers of type 2 diabetic patients: is insulin resistance in type 2 diabetes a slow, type 1 fiber disease? Diabetes. 2001;50(6):1324–9. 102. Stuart CA, McCurry MP, Marino A, et  al. Slow-­ twitch fiber proportion in skeletal muscle correlates with insulin responsiveness. J Clin Endocrinol Metab. 2013;98(5):2027–36. 103. Stuart CA, South MA, Lee ML, et  al. Insulin responsiveness in metabolic syndrome after eight weeks of cycle training. Med Sci Sports Exerc. 2013;45(11):2021–9. 104. Palsgaard J, Brons C, Friedrichsen M, et  al. Gene expression in skeletal muscle biopsies from people with type 2 diabetes and relatives: differential regulation of insulin signaling pathways. PLoS One. 2009;4(8):e6575. 105. Lambadiari V, Triantafyllou K, Dimitriadis GD.  Insulin action in muscle and adipose tissue in type 2 diabetes: the significance of blood flow. World J Diabetes. 2015;6(4):626–33. 106. Laughlin MH.  Physical activity-induced remodeling of vasculature in skeletal muscle: role in treatment of type 2 diabetes. J Appl Physiol (1985). 2016;120(1):1–16. 107. Lettner A, Roden M.  Ectopic fat and insulin resistance. Curr Diab Rep. 2008;8(3):185–91. 108. Samuel VT, Shulman GI.  Mechanisms for insulin resistance: common threads and missing links. Cell. 2012;148(5):852–71. 109. Srikanthan P, Hevener AL, Karlamangla AS. Sarcopenia exacerbates obesity-associated insulin resistance and dysglycemia: findings from the National Health and Nutrition Examination Survey III. PLoS One. 2010;5(5):e10805. 110. Varma V, Yao-Borengasser A, Rasouli N, et  al. Muscle inflammatory response and insulin resistance: synergistic interaction between macrophages and fatty acids leads to impaired insulin action. Am J Physiol Endocrinol Metab. 2009;296(6): E1300–10. 111. Glass CK, Olefsky JM.  Inflammation and lipid signaling in the etiology of insulin resistance. Cell Metab. 2012;15(5):635–45. 112. Nicholson T, Church C, Baker DJ, Jones SW.  The role of adipokines in skeletal muscle inflammation and insulin sensitivity. J Inflamm (Lond). 2018; 15:9. 113. Randle PJ, Garland PB, Hales CN, Newsholme EA. The glucose fatty-acid cycle. Its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. Lancet. 1963;1(7285):785–9. 114. Kelley DE, Mokan M, Simoneau JA, Mandarino LJ. Interaction between glucose and free fatty acid metabolism in human skeletal muscle. J Clin Invest. 1993;92(1):91–8.

10  Role of Tissue Biopsy in Drug Development for Nonalcoholic Fatty Liver Disease and Other… 115. Hue L, Taegtmeyer H.  The Randle cycle revisited: a new head for an old hat. Am J Physiol Endocrinol Metab. 2009;297(3):E578–91. 116. Martins AR, Nachbar RT, Gorjao R, et  al. Mechanisms underlying skeletal muscle insulin resistance induced by fatty acids: importance of the mitochondrial function. Lipids Health Dis. 2012;11:30. 117. Phielix E, Jelenik T, Nowotny P, Szendroedi J, Roden M.  Reduction of non-esterified fatty acids improves insulin sensitivity and lowers oxidative stress, but fails to restore oxidative capacity in type 2 diabetes: a randomised clinical trial. Diabetologia. 2014;57(3):572–81. 118. Makimura H, Stanley TL, Suresh C, et al. Metabolic effects of long-term reduction in free fatty acids with acipimox in obesity: a randomized trial. J Clin Endocrinol Metab. 2016;101(3):1123–33. 119. Machado MV, Ferreira DM, Castro RE, et  al. Liver and muscle in morbid obesity: the interplay of fatty liver and insulin resistance. PLoS One. 2012;7(2):e31738. 120. Camastra S, Vitali A, Anselmino M, et  al. Muscle and adipose tissue morphology, insulin sensitivity and beta-cell function in diabetic and nondiabetic obese patients: effects of bariatric surgery. Sci Rep. 2017;7(1):9007. 121. Pedersen BK.  Muscles and their myokines. J Exp Biol. 2011;214(Pt 2):337–46. 122. Delezie J, Handschin C.  Endocrine crosstalk between skeletal muscle and the brain. Front Neurol. 2018;9:698. 123. Sakuma K, Yamaguchi A. Drugs of muscle wasting and their therapeutic targets. Adv Exp Med Biol. 2018;1088:463–81. 124. Bouchonville MF, Villareal DT. Sarcopenic obesity: how do we treat it? Curr Opin Endocrinol Diabetes Obes. 2013;20(5):412–9. 125. Consitt LA, Clark BC.  The vicious cycle of myostatin signaling in sarcopenic obesity: myostatin role in skeletal muscle growth, insulin signaling and implications for clinical trials. J Frailty Aging. 2018;7(1):21–7. 126. Cleasby ME, Jamieson PM, Atherton PJ.  Insulin resistance and sarcopenia: mechanistic links between common co-morbidities. J Endocrinol. 2016;229(2):R67–81. 127. Allen DL, Hittel DS, McPherron AC.  Expression and function of myostatin in obesity, diabetes, and exercise adaptation. Med Sci Sports Exerc. 2011;43(10):1828–35. 128. Morley JE.  Pharmacologic options for the treatment of sarcopenia. Calcif Tissue Int. 2016;98(4):319–33. 129. Garito T, Roubenoff R, Hompesch M, et  al. Bimagrumab improves body composition and insulin sensitivity in insulin-resistant individuals. Diabetes Obes Metab. 2018;20(1):94–102. 130. Nikoulina SE, Ciaraldi TP, Abrams-Carter L, et  al. Regulation of glycogen synthase activity in cultured skeletal muscle cells from subjects with type II dia-

273

betes: role of chronic hyperinsulinemia and hyperglycemia. Diabetes. 1997;46(6):1017–24. 131. Beck-Nielsen H, Vaag A, Poulsen P, Gaster M.  Metabolic and genetic influence on glucose metabolism in type 2 diabetic subjects--experiences from relatives and twin studies. Best Pract Res Clin Endocrinol Metab. 2003;17(3):445–67. 132. Thingholm TE, Bak S, Beck-Nielsen H, Jensen ON, Gaster M.  Characterization of human myotubes from type 2 diabetic and nondiabetic subjects using complementary quantitative mass spectrometric methods. Mol Cell Proteomics. 2011;10(9):M110 006650. 133. Aas V, Bakke SS, Feng YZ, et  al. Are cultured human myotubes far from home? Cell Tissue Res. 2013;354(3):671–82. 134. Costford SR, Crawford SA, Dent R, McPherson R, Harper ME.  Increased susceptibility to oxidative damage in post-diabetic human myotubes. Diabetologia. 2009;52(11):2405–15. 135. Al-Khalili L, de Castro Barbosa T, Östling J, Massart J, Katyama M, Nyström C, Oscarsson J, Zierath JR. Profiling of human myotubes reveals an intrinsic proteomic signature associated with type 2 diabetes. Transl Proteomics. 2014;2:25–38. 136. Tchernof A, Despres JP.  Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013;93(1):359–404. 137. Sattar N, Gill JM.  Type 2 diabetes as a disease of ectopic fat. BMC Med. 2014;12:123. 138. Smith U. Abdominal obesity: a marker of ectopic fat accumulation. J Clin Invest. 2015;125(5):1790–2. 139. Piche ME, Poirier P, Lemieux I, Despres JP. Overview of epidemiology and contribution of obesity and body fat distribution to cardiovascular disease: an update. Prog Cardiovasc Dis. 2018;61(2):103–13. 140. Trujillo ME, Scherer PE.  Adipose tissue-derived factors: impact on health and disease. Endocr Rev. 2006;27(7):762–78. 141. Coelho M, Oliveira T, Fernandes R. Biochemistry of adipose tissue: an endocrine organ. Arch Med Sci. 2013;9(2):191–200. 142. White U, Ravussin E.  Dynamics of adipose tissue turnover in human metabolic health and disease. Diabetologia. 2019;62:17–23. 143. Appari M, Channon KM, McNeill E.  Metabolic regulation of adipose tissue macrophage function in obesity and diabetes. Antioxid Redox Signal. 2018;29(3):297–312. 144. Labrecque J, Laforest S, Michaud A, Biertho L, Tchernof A.  Impact of bariatric surgery on white adipose tissue inflammation. Can J Diabetes. 2017;41(4):407–17. 145. Carswell KA, Lee MJ, Fried SK.  Culture of isolated human adipocytes and isolated adipose tissue. Methods Mol Biol. 2012;806:203–14. 146. Petrus P, Rosqvist F, Edholm D, et  al. Saturated fatty acids in human visceral adipose tissue are associated with increased 11- beta-hydroxysteroid-­ dehydrogenase type 1 expression. Lipids Health Dis. 2015;14:42.

274 147. Verboven K, Wouters K, Gaens K, et al. Abdominal subcutaneous and visceral adipocyte size, lipolysis and inflammation relate to insulin resistance in male obese humans. Sci Rep. 2018;8(1):4677. 148. Zierath JR, Livingston JN, Thorne A, et  al. Regional difference in insulin inhibition of nonesterified fatty acid release from human adipocytes: relation to insulin receptor phosphorylation and intracellular signalling through the insulin receptor substrate-1 pathway. Diabetologia. 1998;41(11):1343–54. 149. Dusserre E, Moulin P, Vidal H. Differences in mRNA expression of the proteins secreted by the adipocytes in human subcutaneous and visceral adipose tissues. Biochim Biophys Acta. 2000;1500(1):88–96. 150. Mutch DM, Tordjman J, Pelloux V, et al. Needle and surgical biopsy techniques differentially affect adipose tissue gene expression profiles. Am J Clin Nutr. 2009;89(1):51–7. 151. Murphy J, Moullec G, Santosa S.  Factors associated with adipocyte size reduction after weight loss interventions for overweight and obesity: a systematic review and meta-regression. Metabolism. 2017;67:31–40. 152. Yki-Jarvinen H. Thiazolidinediones. N Engl J Med. 2004;351(11):1106–18. 153. Fonseca V.  Effect of thiazolidinediones on body weight in patients with diabetes mellitus. Am J Med. 2003;115(Suppl 8A):42S–8S. 154. de Souza CJ, Eckhardt M, Gagen K, et al. Effects of pioglitazone on adipose tissue remodeling within the setting of obesity and insulin resistance. Diabetes. 2001;50(8):1863–71. 155. Hammarstedt A, Andersson CX, Rotter Sopasakis V, Smith U. The effect of PPARgamma ligands on the adipose tissue in insulin resistance. Prostaglandins Leukot Essent Fatty Acids. 2005;73(1):65–75. 156. McLaughlin TM, Liu T, Yee G, et  al. Pioglitazone increases the proportion of small cells in human abdominal subcutaneous adipose tissue. Obesity (Silver Spring). 2010;18(5):926–31. 157. Trujillo ME, Scherer PE. Adiponectin--journey from an adipocyte secretory protein to biomarker of the metabolic syndrome. J Intern Med. 2005;257(2):167–75. 158. Rasouli N, Yao-Borengasser A, Miles LM, Elbein SC, Kern PA.  Increased plasma adiponectin in response to pioglitazone does not result from increased gene expression. Am J Physiol Endocrinol Metab. 2006;290(1):E42–6.

A. J. Krentz and P. Bedossa 159. Hollis G, Huber R. 11beta-Hydroxysteroid dehydrogenase type 1 inhibition in type 2 diabetes mellitus. Diabetes Obes Metab. 2011;13(1):1–6. 160. Stefan N, Ramsauer M, Jordan P, et  al. Inhibition of 11beta-HSD1 with RO5093151 for non-alcoholic fatty liver disease: a multicentre, randomised, double-­ blind, placebo-controlled trial. Lancet Diabetes Endocrinol. 2014;2(5):406–16. 161. Gibbs JP, Emery MG, McCaffery I, et  al. Population pharmacokinetic/pharmacodynamic model of subcutaneous adipose 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) activity after oral administration of AMG 221, a selective 11beta-HSD1 inhibitor. J Clin Pharmacol. 2011;51(6):830–41. 162. Thyagarajan B, Foster MT. Beiging of white adipose tissue as a therapeutic strategy for weight loss in humans. Horm Mol Biol Clin Investig. 2017. https:// doi.org/10.1515/hmbci-2017-0016. 163. Stanford KI, Goodyear LJ.  Muscle-adipose tissue cross talk. Cold Spring Harb Perspect Med. 2018. https://doi.org/10.1101/cshperspect.a029801. 164. Perakakis N, Triantafyllou GA, Fernandez-Real JM, et al. Physiology and role of irisin in glucose homeostasis. Nat Rev Endocrinol. 2017;13(6):324–37. 165. Kajimura S.  Engineering fat cell fate to fight obesity and metabolic diseases. Keio J Med. 2015; 64(4):65. 166. Kusminski CM, Bickel PE, Scherer PE. Targeting adipose tissue in the treatment of obesityassociated diabetes. Nat Rev Drug Discov. 2016;15(9):639–60. 167. Vargas-Castillo A, Fuentes-Romero R, Rodriguez-­ Lopez LA, Torres N, Tovar AR.  Understanding the biology of thermogenic fat: is browning a new approach to the treatment of obesity? Arch Med Res. 2017;48(5):401–13. 168. Dinas PC, Valente A, Granzotto M, et al. Browning formation markers of subcutaneous adipose tissue in relation to resting energy expenditure, physical activity and diet in humans. Horm Mol Biol Clin Investig. 2017. https://doi.org/10.1515/hmbci-2017-0008. 169. Kiefer FW. The significance of beige and brown fat in humans. Endocr Connect. 2017;6(5):R70–9. 170. Mukherjee J, Baranwal A, Schade KN. Classification of therapeutic and experimental drugs for brown adipose tissue activation: potential treatment strategies for diabetes and obesity. Curr Diabetes Rev. 2016;12(4):414–28.

Utility of Invasive and Non-­ invasive Cardiovascular Research Methodologies in Drug Development for Diabetes, Obesity and NAFLD/NASH

11

Gerardo Rodriguez-Araujo and Andrew J. Krentz

Summary Mounting evidence supports the existence of clinically important mechanistic intersections between prevalent metabolic and cardiovascular disorders. Novel pharmacotherapies that may simultaneously impact metabolism and vascular function are being developed. While medications for diabetes and obesity are closely scrutinized for cardiovascular safety and efficacy, methodologies for the early phase assessment of vascular function remain underutilized. This consideration extends to drugs being developed for nonalcoholic fatty liver disease (NAFLD)/nonalcoholic steatohepatitis (NASH), which is associated with an increased risk of atherothrombotic cardiovascular disease.

Available cardiovascular methodologies may be classified as either invasive or non-invasive. Invasive methodologies are recommended for the characterization of direct interactions with a target vessel or the vasculature more generally. Various invasive methods that are very selective and precise in their targets are suitable for proof of concept or proof of mechanism studies. Non-­ invasive methods tend to be more suitable for large cohorts or for screening activities selecting subpopulations of subjects with cardiovascular disease for participation in clinical studies. Thus, a balance exists between the scalable properties of non-invasive methodologies for screening and patient selection on one hand, and the precision of invasive methods as primary or exploratory endpoints when designing a clinical study on the other.

G. Rodriguez-Araujo (*) ProSciento, Inc., Chula Vista, CA, USA University of Arkansas for Medical Sciences, Graduate School of Medicine, Little Rock, AR, USA e-mail: [email protected] A. J. Krentz ProSciento, Inc., Chula Vista, CA, USA Institute for Cardiovascular and Metabolic Research, University of Reading, Reading, UK © Springer Nature Switzerland AG 2019 A. J. Krentz et al. (eds.), Translational Research Methods in Diabetes, Obesity, and Nonalcoholic Fatty Liver Disease, https://doi.org/10.1007/978-3-030-11748-1_11

275

G. Rodriguez-Araujo and A. J. Krentz

276

Key Methods Invasive Methods

Method

Measurement

Advantages

Disadvantages

Real-time radiographic assessment using contrast infused into the LV

LV anatomy (remodeling changes), ejection fraction (%).

Direct visualization of real timeejection fraction in LV and LV anatomy. Special use for structural defects in LV.

Often need a core lab for image analysis.

Coronary Angiography

Real timeradiographic assessment using contrast infused directly into the lumen of the target coronary artery

% of stenosis; patency; vessel and lesion length (mm); vessel anatomy

Gold standard for the assessment of lesions and vessel anatomy.

Commonly needs core lab for image analysis or have a blinded independent operator reading the images.

Peripheral Artery Angiography

Real timeradiographic assessment using contrast infused directly into the target vessel lumen in limbs. Intravascular introduction of a microultrasound probe into the target vessel and/or healthy vessel

% of stenosis; patency; vessel and lesion length (mm); vessel anatomy

Gold standard for peripheral vessel and lesions anatomy.

Vessel diameter (mm), vessel volume (when used 3D reconstruction), lesion area (mm2), big thrombus or debris presence.

A resolution of 100 μm. Practical and inexpensive materials. Most cardiovascular centers are familiar with this procedure.

Commonly needs for core lab for image analysis or have a blinded independent operator reading the images. Image capturing can be challenging at times, difficult to characterize plaque composition and to visualize thrombus.

Imaging Left ventriculography (LV gram)

Intravascular Ultrasound (IVUS)

Value in drug development decisions Accurate measures of endpoints that include ventricular function such as LVEF and wall motion. Evaluation of response to treatment of drugs that impact cardiac function and remodeling. Accurate and comparable measures of any structural changes in the coronaries with the use of new drugs or devices. Early phase car diovascular (CV) studies frequently include this modality to assess any important effects of the IP in the coronaries. Accurate and comparable measures of any structural changes in peripheral arteries with the use of new drugs or devices. Direct and practical assessment of vessel, plaque and thrombus in larger studies (phase IIb, III). Useful as endpoint for metabolic drugs with potential effects on atherogenesis and ischemic heart disease. Can also be applied to peripheral vessels.

11  Utility of Invasive and Non-invasive Cardiovascular Research Methodologies in Drug Development…

277

Invasive Methods

Optical Coherence Tomography (OCT)

Method Intravascular introduction of a microtomography probe into the target vessel and/or healthy vessel (control)

Measurement Vessel diameter (mm); vessel volume (mm3); lesion area (mm2); lesion characteristics (calcium, fracture); thrombus presence and anatomy

Functional/Proof Concept-Mechanism FFR ratio. Flow Fractional Intracoronary/ Normal reserve Reserve (FFR) intraarterial >0.75. infusion of a vasodilator after Abnormal reserve 0.75 [61]. A value of 80% inhibition of plasma DPP-4 activity for 24 h was achieved at a dose of 100 mg [40]. The prediction that 100 mg would lead to clinical efficacy was confirmed to significantly lower HbA1c in a dose-­ ranging Phase 2b study in patients with type 2 diabetes [41]. Notably, single doses up to 800 mg and multiple doses up to 600 mg were safe and tolerated [40], increasing confidence in the safety profile of sitagliptin at therapeutic levels. The POC study for sitagliptin was initiated ~2 years after the start of the initial clinical studies, a process that generally requires closer to 4 years to accomplish. Several factors helped to accelerate the early clinical development of sita-

18  Transitioning from Preclinical to Clinical Drug Development

gliptin. First, and perhaps most important, was the development of a novel target engagement assay measuring plasma DPP-4 inhibition. The combined analysis of DPP-4 inhibition, the PK profile (i.e., drug concentration), and disease-­ related biomarkers (post-prandial glucose, GLP-­ 1) were applied across the transition from preclinical to clinical development. This approach provided confidence in estimating the doses of sitagliptin that would provide clinical benefit in patients. Because of this, the traditional POC study (Phase 2a) was skipped, and a Phase 2b dose range-finding study in patients was initiated immediately after completion of Phase 1 studies. The safety of repeated doses up to 600  mg increased confidence that longer-term administration of much lower doses would be safe and well-tolerated in patients and reduced the need to establish the dose range with acceptable tolerability in chronic dosing studies prior to the POC study. Second, extensive work had led to a deep understanding of incretin biology in glucose homeostasis. In addition, interventions with GLP-1 receptor agonists had validated the role of incretin mimetics in humans. Therefore, a drug that could increase GLP-1 levels was likely to have clinical benefit in patients with type 2 diabetes. Third, a large unmet medical need exists to combat the growing global epidemic of type 2 diabetes [42]. The majority of patients with type 2 diabetes do not achieve glycemic goal of HbA1c 500 mg/dL; >5.56 mmol/L) of triglyceride as a regulatory endpoint for reduction of high triglyceride levels, but specifically not for reduction of cardiovascular risk [31]. However, all other non-LDL lipid-lowering indications aimed at cardiovascular risk reduction, including triglyceride-lowering agents, require outcome trials [32]. Doubt was raised about LDL cholesterol as a suitable regulatory endpoint, starting with the ENHANCE trial which failed to show a benefit of adding ezetimibe to simvastatin in reducing intima–media thickness [33]. However, the subsequent IMPROVE-IT trial demonstrated that adding ezetimibe to statin therapy resulted in incremental lowering of LDL cholesterol levels and improved cardiovascular outcomes. The trial also showed that lowering LDL cholesterol to levels below previous targets provided additional benefit [34]. Doubt about LDL cholesterol as a reliable indicator of benefit was again raised by the ACCELERATE trial that showed no effect of the cholesteryl ester transfer protein inhibitor (CTEPi) evacetrapib on cardiovascular events among patients with highrisk vascular disease despite favorable effects on established lipid biomarkers, including LDL

G. A. Fleming and B. E. Harvey

cholesterol. Interestingly, despite failure of evacetrapib and dalcetrapib, which increased HDL cholesterol levels but did not reduce the risk of recurrent cardiovascular events [35], and torcetrapib which in the ILLUMINATE trial resulted in an increased risk of mortality and morbidity [36], the most recently report CTEPi trial REVEAL (Randomized Evaluation of the Effects of Anacetrapib through Lipid Modification) showed both improved lipid levels and a reduction in major coronary events [36]. FDA has approved alirocumab and evolocumab, members of the new proprotein convertase subtilisin/kexin type 9 (PCSK9) class on the basis of LDL cholesterol endpoint [37]. However, both approvals were contingent on post-approval confirmation of improved cardiovascular outcomes. Those confirmatory results have since been provided [38]. Despite the historic utility of LDL cholesterol in supporting development of therapies to reduce cardiovascular risk, it appears unlikely that this endpoint alone will continue to be accepted for regulatory approvals. Though these more recent issues pertain to late-stage and post-approval therapies, this understanding is relevant to early-stage investigation for several reasons. DMEP is busy, stressed and perhaps more risk averse, which can have an impact on early IND review. Regulatory endpoints for assessment of efficacy in the metabolic space, such as HbA1c and LDL cholesterol, are also increasingly debated within FDA.  Even at the earliest phases of clinical investigation, sponsors are being pressured to show evidence of both safety and efficacy that is superior to currently marketed products [39]. Early-phase investigators can benefit from understanding these dynamics since difficult benefit/risk judgments that should be made at the approval stage have an impact even on DMEP’s review of initial in-­human studies.

 ow Therapeutic Products Are H Regulated at FDA FDA is a vast organization with regulatory responsibility for more than $1 trillion worth of consumer goods, roughly 20% of consumer

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

expenditures in the U.S. This includes $466 billion in food sales, $275 billion in drugs, $60 billion in cosmetics and $18 billion in dietary supplements [40]. Unlike any other therapeutic regulatory authority in the world, FDA has a large staff of scientific experts and medical specialists. Most of its regulatory review work is performed without the help of outside experts. FDA’s budget in 2017 totaled $5.1 billion [41]. FDA is comprised of Centers with responsibility for regulating each of the major health product areas—drugs, biologics, and devices—as well as dietary supplements and food/nutrition.

Therapeutic Product Jurisdiction The Center for Drug Evaluation and Research (CDER) is most likely to regulate the early clinical research of most investigators, but the Center for Biologics Evaluation and Research (CBER), the Center for Food Safety and Nutrition (CFSAN), and the Center for Devices and Radiologic Health (CDRH) could also be involved. A product application will be assigned to the appropriate center for review but may be reviewed in some circumstances by more than one center. For example, a dermal patch that delivers insulin could be reviewed by both CDER and CDRH. Some products may be more difficult to assign. For example, a gel that swells in the stomach to result in weight loss could be considered a medical device instead of a drug, but a non-absorbed resin that binds cholesterol and bile acids would be considered a drug. Dietary supplements, regulated by CFSAN, are by definition used by consumers without physician supervision and can make no disease treatment claims. Supplements are required to have evidence of only safety and good manufacturing quality. Supporting data are not reviewed prior to marketing and may not ever be reviewed. A widely used dietary supplement becomes a drug under CDER review when it is being investigated for a disease treatment or prevention. However, parallel development of a product as a dietary supplement and drug product rarely occurs. Once a supplement is commercially available, there is less likelihood

493

that any further investment will be made in studies for disease treatment or prevention. Typically, a strategic decision is made early in the developmental pathway to follow a supplement route or a drug route, since these routes are very different in both cost and time to market. The exception to this is omega-3 products derived from fish oil. Omega-3 products are available as dietary supplements when less purified, and approved drug products for treatment of severe hypertriglyceridemia when more purified. Insulin and other smaller peptide hormones are considered drugs and approved by New Drug Applications (NDA). Monoclonal antibody products have been regulated by CDER since 2003 but are approved by Biologics Licensing Applications (BLA). Some other products—cell and gene therapies and some therapeutic vaccines—are licensed by BLAs but reviewed by CBER. Combinations of products, including drug-device, drug-drug, and drug-supplement, may be regulated by one, two or, conceivably, three centers. However, the FDA Office of Combination Products (OCP) was established on Dec. 24, 2002, to help minimize multiple center reviews [42]. The Medical Device User Fee and Modernization Act of 2002 (MDUFMA) of 2002, gave the OCP broad responsibilities covering the regulatory life cycle of combination products, as well as chose a “Lead Center” for combination product review, based upon the “Primary Mode of Action” [43]. These efforts in the area of combination products were also supported by the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (MMA). The agency benefits from having these product decisions made from a centralized perspective, but complexities remain. For example, a diabetic wound-healing drug that is applied in a synthetic matrix could be jointly reviewed by CDRH and CDER as the lead center. The same drug in a cellular matrix could be regulated by CDER and CBER as the lead center. It is relevant here to mention perhaps the most unique regulatory pathway in all of FDA— that of the medical food, which is regulated by CFSAN.  This route is often raised as an attractive commercialization approach since it does not involve an IND and pre-market review and

494

approval. Medical food products must have a food-like quality that addresses a nutritional deficiency or gastrointestinal condition. Specifically, FDA’s definition is “a food which is formulated to be consumed or administered enterally under the supervision of a physician and which is intended for the specific dietary management of a disease or condition for which distinctive nutritional requirements, based on recognized scientific principles, are established by medical evaluation” [44]. These products can be studied without IND review [45], and they can have medical claims for treatment of disease. Ironically, the original law establishing the definition of medical food is the Orphan Drug Act, in part because some of the first medical foods were used for treatment of rare disease and inborn errors of metabolism. Recent modification to the FDA guidance for medical foods has made clear that a medical food is not a drug, nor need it be for an orphan indication. Furthermore, contrary to what has commonly been the case, a medical food label cannot bear an Rx symbol or “by prescription only” statement. Still, the product must be used under supervision of a physician [46]. This regulatory pathway is becoming more scrutinized by FDA as more dietary supplements and non-qualifying food products are sold as prescription products and/or for unaccepted indications.

The IND Review Process The FD&C Act establishes that if an unapproved drug is used in the U.S., it can be considered misbranded and the marketer is therefore subject to prosecution [47]. An exception to this requirement must be made for an experimental drug to be studied. FDA calls this an Investigative New Drug (IND) exemption. In common parlance, IND is used to refer to this exemption. FDA does not approve INDs or IND exemptions—it allows them to proceed.

 re-IND and Other Meetings with FDA P FDA’s role in the development of a new drug begins when the drug’s sponsor (e.g., an academic investigator, a manufacturer, potential marketer), is preparing to conduct clinical studies

G. A. Fleming and B. E. Harvey

in humans. CDER and CBER provide an option for a pre-IND [48] meeting to provide advice in response to questions and preliminary data that the sponsor has compiled. CDRH provides a comparable pre-IDE [49] (Investigation Device Exemption) meeting. Generally, for this meeting the sponsor will present data to FDA that characterize efficacy and safety in animal models and that are sufficient to justify the anticipated initial exposure in humans. In some cases, the sponsor may have clinical data that have been produced in another country. A request for a pre-IND meeting follows a standard outline and includes an initial draft of the questions that the sponsor wishes to have answered by FDA.  In advance of the meeting, a briefing package of relevant data is submitted for review. If the meeting is not granted, the divisions are generally required to provide preliminary written responses to the sponsor prior to the meeting or within about 30 days of submission of the briefing package. In my experience (GAF), DMEP grants face-to-face pre-IND meetings infrequently but does provide detailed written responses to submitted questions. The pre-IND meeting as the first interaction between the FDA and the sponsor sets the tone for an ongoing working relationship. Therefore, careful and thoughtful preparation of questions, background material and proposed product development plan is essential. After receiving final written responses or FDA meeting minutes, the sponsor may engage in some limited written exchanges. Many believe that the critical sponsor meeting with FDA is the End of Phase 2 Meeting (EOP2) [50], during which all non-clinical and clinical Phase 1/Phase 2 data are reviewed and questions are posed to the agency regarding the Phase 3 development program. Official Meeting Minutes are provided within 30 days after the meeting [51]. The less well known “End of Phase 2A (EOP2A)” meeting is also provided by CDER, but infrequently held [52]. This meeting can be of value for early clinical programs to gain feedback from FDA on clinical pharmacology issues. As described in the FDA guidance, “The overall purpose of an EOP2A meeting is to discuss options for trial designs, modeling strategies, and clinical trial simulation scenarios to

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity… improve the quantification of the exposure-­ response information from early drug development. The goal of these meetings is to optimize dose selection for subsequent trials to improve the efficiency of drug development. The exposure-­ response data discussed might be pertinent to evaluation of efficacy outcomes or adverse outcomes. In addition, the meetings would provide opportunities for discussions of complex issues pertaining to drug interactions, trials in special populations defined by genetic characteristics or other biomarkers, and other pharmacokinetic or pharmacokinetic/pharmacodynamic (PK/PD) relationships.”

495

tional drug is administered or dispensed. The physician is also ultimately responsible for meeting all IND requirements, including supervision of study personnel, handling and analysis of data, reporting of adverse events, and annual IND reports. These responsibilities may be delegated to others. Investigator-­ sponsored INDs are typically used to explore the potential of a new molecule or metabolite or to evaluate a hypothesis involving an approved drug for a different indication or In addition to the pre-IND meeting and EOP2 new patient population. FDA has traditionally meeting, FDA regulations provide for another been more flexible in its review and requireformal interaction, the pre-NDA/pre-BLA meetments for investigator INDs, but there is no ing. These are extremely important and valued formal difference in requirements for investilandmark meetings for drug products in later-­ gator- and commercial-sponsored INDs. stage development but are only mentioned here. 2. A Commercial IND is typically sponsored and The FDA guidance on formal meetings provides owned by a company or research organizauseful information [53]. tion, though an individual (including, rarely, a The specific objectives of the EOP2A meeting physician investigator) could be the sponsor. are “to help select the dosing regimens for the next The intent of the commercial IND is to phase (typically phases 2 and 3) of drug developdevelop a new molecule for approval or to ment and to design informative dose-­response tridevelop an approved drug for a new indicaals that will inform later phase clinical trials by tion, formulation, route of administration, best incorporating prior quantitative knowledge.” patient population, and/or combination with In cases in which two or more treatment another drug. All of the responsibilities indications are sought by the sponsor, and these described above and some others pertain to a come under the purview of different divisions, commercial IND the sponsor can request separate pre-IND meet- 3. An Emergency Use IND involves the authoriings with more than one division. In other cases, zation of an experimental drug to be used in representatives from another division can be an emergency situation (life-threatening or requested to address a specific issue involving severely debilitating disease) that does not the consulting division’s expertise. For examallow time for submission of a conventional ple, a meeting involving a diabetes or obesity IND. This provision is generally reserved for drug with potential for causing depression could exceptional circumstances in which a very benefit from participation from the Division of serious condition is involved and approved Psychiatry Products (DPP) [54] to discuss early options have been exhausted. evaluation of mood alteration as a safety issue. 4. The Treatment IND is rarely used, particularly in the metabolic area. FDA provides this option Varieties of INDs for an experimental drug that shows significant FDA provides two major categories of INDs and efficacy in late-stage clinical trials. It enables a several less well known specialized forms. treatment of serious or immediately life-threatening condition to be used more widely prior to 1. The Investigator IND is submitted by a physifinal FDA review and approval. The sponsor cian or a qualified non-physician investigator may also be allowed to charge for the product with a responsible physician to conduct a clinunder the treatment IND prior to FDA clearical trial in humans. The physician conceives, ance for marketing. This regulatory provision designs, conducts the investigation and is is an option only after pivotal phase 3 clinical responsible for directing how the investigatrials have been started. Many FDA staff view

496

this process as labor intensive and burdensome. Some patient advocacy groups have stated that a Treatment IND or any “compassionate use” pathway, is no substitute for FDA approval of the product. 5. The Exploratory IND [55] was developed just over 10 years ago [56], with the aim of streamlining the identification of lead molecules for further development. FDA described in its guidance some additional flexibility and discretion in what is required to conduct a clinical study under an exploratory IND. Such a study is always conducted in phase 1, involves very limited human exposure, and has no therapeutic or diagnostic intent. The exploratory IND may be used to screen a number of related compounds or investigate the PK resulting from a “microdose” exposure. Because of the very small exposures involved, it was hoped that FDA reviewers would waive or modify some of the nonclinical or product manufacturing requirements.

 hen an IND Is Not Required W After receiving many related questions, FDA has issued guidance regarding when an IND does not need to be submitted to FDA [57]. An IND is not required under two major circumstances—investigation of a marketed drug and bioequivalence studies of a marketed drug product and a generic version. A clinical study of a marketed drug is exempt from IND requirements if all of the following criteria are met: • The drug product is lawfully marketed in the United States. • The investigation is not intended to be reported to FDA as a well-controlled study in support of a new indication and there is no intent to use it to support any other significant change in the labeling of the drug. • In the case of a prescription drug, the investigation is not intended to support a significant change in the advertising for the drug. • The investigation does not involve a new route of administration, dose, patient population, or other factor that significantly increases the risk (or decreases the acceptability of the risk) associated with the use of the drug product.

G. A. Fleming and B. E. Harvey

When all of these criteria apply, not only is the IND requirement waived, but review divisions are also directed to decline such IND submissions even if the sponsor would like to have an IND. Dietary supplements intended only to affect the structure or function of the body and not intended for a therapeutic purpose are not considered drugs by FDA.  Products that are not drugs, such as foods or dietary supplements, may not require an IND.  Exceptions include when a health claim is being sought for a food, or a therapeutic or diagnostic use is being sought for a dietary supplement. For example, a study designed to evaluate whether vitamin D may reduce the risk of diabetes would require an IND since no health claim for this substance-disease relationship has been issued.

 ontent of the IND Submission C The required content of INDs is detailed in its rather old (but largely still current) guidance [58]. Additional details and updates are found in specialized guidances that have been more recently issued. One of the most significant recent changes to the IND has been in the format required for submission. As mentioned under the ICH section, the CTD (common technical document) format is now being implemented as the required format for all IND submissions. In summary, the IND must contain information from the following three categories: 1. Nonclinical data, including animal pharmacology, toxicology, and in vitro studies. These are required to support the safety of human subjects in the initial proposed clinical study. In summary, two toxicology studies—one in a rodent and one in a non-rodent species—of equal or greater duration to the proposed clinical study are required along with specialized animal safety studies. Animal studies that support the mechanism and targeted efficacy are desirable. In vitro studies include human hERG (ether-a-go-go-related gene assay) [59], drug metabolism and mutagenicity testing. 2. Manufacturing and analytical information, including descriptions of the manufacturing process, identity, strength, quality and purity of the drug substance and drug product, analytical

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

procedures and specification, stability, consistency among batches and container systems is required. These requirements are detailed in a number of guidances, but enumerated in the description of Current Good Manufacturing Practice for Finished Pharmaceuticals under 21 CFR Part 211 (a) General Provisions (b) Organization and Personnel (c) Buildings and Facilities (d) Equipment (e) Control of Components and Drug Product Containers and Closures (f) Production and Process Controls (g) Packaging and Labeling Controls (h) Holding and Distribution (i) Laboratory Controls (j) Records and Reports (k) Returned and Salvaged Drug Products 3. Clinical protocols and investigator information, including one or more complete protocols, investigator brochure, any clinical data that may be available, qualifications of clinical investigators and commitment to obtain informed consent from the research subjects, commitment to obtain review of the study by an institutional review board (IRB) and adherence to the investigational new drug regulations. The studies designed to be conducted in the following 12 months of the development plan should also be described so that FDA can offer suggestions or concerns that can be used to modify these plans.

I ND Review Process Once the IND is submitted, the sponsor must wait 30 calendar days before initiating the proposed clinical study. During this time, the primary chemistry, nonclinical and clinical FDA reviewers will determine whether any substantial safety issues are involved. The reviewers usually provide concerns and recommendations. They may also ask for further information from the sponsor. The primary goal of IND review is to assure that research subjects will not be exposed to unreasonable risk [60]. This determination is based on having sufficient data and information from the three major review areas—nonclinical, CMC (chemistry, manufacturing, and controls)

497

and clinical—aimed first at assessment of safety and second at achievement of other objectives. If there are significant deficiencies and/or concerns about study participant safety, FDA can place a proposed or ongoing phase 1 clinical trial on full or partial clinical hold or, in extreme cases, even terminate the IND.  A partial clinical hold means that a study may go forward or continue, but with some restrictions on the study design and execution. In some cases of IND review, one study may be allowed while another is placed on clinical hold. A clinical hold is never issued lightly by FDA and for phase 1 studies is only done for reasons of safety. FDA may place a phase 2 or 3 study on clinical hold for a s­ ubstantial design or other non-safety issue, in addition to safety concerns. FDA is required to work with the sponsor to resolve the clinical hold on a timely basis. The sponsor can request a meeting if necessary. Once the IND review has been completed and the IND [exemption] allowed, the sponsor may submit subsequent protocols. FDA does not have a 30-day wait requirement for subsequent protocols submitted to the IND file, though it is prudent to wait for any FDA feedback on the protocol before beginning a study. Additional clinical studies may require supplementary nonclinical data and completion and analysis of data from a preceding clinical study. Communication from FDA may be very limited during the execution of the early-stage development plan until the EOP2 meeting [61]. The regulations allow for a stepwise process of accruing and refining CMC information. The amount of information and data needed increases from phase 1 through phase 3 of the program, depending on the proposed duration of the investigation, the dosage form, and the amount of information otherwise available. For example, although stability data are required in all phases of the IND to demonstrate that the new drug substance and drug product are within acceptable chemical and physical limits for the planned duration of the proposed clinical investigation, if very short-term tests are proposed, the supporting stability data can be correspondingly limited for supporting short-term use and extended in parallel to support longer clinical studies.

498

Ethical Review FDA has devolved primary responsibility for ethical review of clinical studies to Investigational Review Boards (IRB). These boards are often chartered within an academic institution or hospital to provide local ethical review, but boards may also be independent and review studies remotely. The responsibilities of IRBs and investigators/sponsors have grown to include financial interest disclosure, verification of adequate research facilities, and others [62]. Details for these responsibilities are found on FDA’s website [63] and HHS under the Common Rule.4 FDA clinical reviewers take into account ethical considerations as part of the evaluation of The current U.S. system of protection for human research subjects is heavily influenced by the Belmont Report, https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/index.html written in 1979 by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The Belmont Report outlines the basic ethical principles in research involving human subjects. In 1981, with this report as foundational background, HHS and the Food and Drug Administration revised, and made as compatible as possible under their respective statutory authorities, their existing human subjects regulations. The Federal Policy for the Protection of Human Subjects or the “Common Rule” was published in 1991 and codified in separate regulations by 15 Federal departments and agencies, as listed below. The HHS regulations, 45 CFR part 46, include four subparts: subpart A, also known as the Federal Policy or the “Common Rule”; subpart B, additional protections for pregnant women, human fetuses, and neonates; subpart C, additional protections for prisoners; and subpart D, additional protections for children. Each agency includes in its chapter of the Code of Federal Regulations [CFR] section numbers and language that are identical to those of the HHS codification at 45 CFR part 46, subpart A. For all participating departments and agencies, the Common Rule outlines the basic provisions for IRBs, informed consent, and Assurances of Compliance. Human subject research conducted or supported by each federal department/agency is governed by the regulations of that department/agency. The head of that department/ agency retains final judgment as to whether a particular activity it conducts or supports is covered by the Common Rule. If an institution seeks guidance on implementation of the Common Rule and other applicable federal regulations, the institution should contact the department/agency conducting or supporting the research. https://www.hhs. gov/ohrp/regulations-and-policy/regulations/commonrule/index.html

4 

G. A. Fleming and B. E. Harvey

submitted studies. It is not required that informed consent documents be submitted to the IND file, but reviewers have the option of requesting the consent document [64]. An ethical concern with a protocol can be a basis for a clinical hold.

Special Regulatory Programs Early-phase investigators should be aware of special regulatory provisions and their potential relevance to studies that are done in phases 1 and 2.

 rphan Drug Program O The Orphan Drug Program at FDA has catalyzed the development and commercialization of high market value metabolic products including growth hormone and β-glucocerebrosidase (Cerezyme®). The Orphan Drug Program was designed to encourage development of therapies aimed at small patient populations, but the Orphan Drug Program has evolved into a high market value strategy that allows for smaller studies to achieve FDA approval [65]. Abundant information about the orphan drug program is available on the FDA website [66]. When an orphan indication is targeted, the provisions of the Orphan Drug Act can provide a number of benefits, including grants for clinical investigation. Of relevance to the early investigator is the fact that today’s reviewers at FDA’s Office of Orphan Products are increasingly looking for more clinical evidence of a drug’s promise for treating the orphan condition than has been the case in the past. Before, emphasis was primarily put on insuring that the prevalence of the condition was below 200,000 people in the U.S. Results from an early PK/PD study could be decisive in winning an orphan designation [67]. As mentioned above, orphan therapies have been approved in the broader metabolic area. It should be borne in mind that even for common conditions like diabetes, an orphan indication could be developed for a rare form of diabetes that has a well-understood genetic basis, is identifiable, affects fewer than 200,000 people in the U.S., and is responsive to the therapy. Such an example is FDA’s February 2014 approval of metreleptin (MyaleptTM, Bristol-­ Myers Squibb) as replacement therapy to treat

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

the complications of leptin deficiency in patients with congenital generalized or acquired generalized lipodystrophy [68]. Metreleptin, a synthetic analog of the hormone leptin, had originally been envisioned as a therapy for obesity and type 2 diabetes. The drug’s value in improving the metabolic control in patients with the orphan condition of congenital generalized or acquired generalized lipodystrophy was demonstrated in a three-year study organized by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) at the National Institutes of Health. Metreleptin treatment resulted in substantial improvements in both glycemic control and triglyceride levels [69]. Given the orphan indication, FDA’s approval was based on a relatively very small clinical development program. Europe has a comparable orphan disease program, and FDA and EMA have created a common format for applying for orphan status with both agencies [70].

 ast Track and Breakthrough Therapy F Programs FDA has created several programs for speeding the development of drug and biologics therapies. These are explained in more detail on the FDA website [71]. Fast Track (FT) status and Breakthrough Therapy Designation (BTD) are both available to products that show promise for meeting serious unmet clinical need. Generally, clinical efficacy data are required for consideration, especially for BTD. The FDA review divisions vary significantly in their willingness to consider and grant Fast Track status. DMEP has been among the most parsimonious divisions in granting Fast Track status.

Key Regulatory-Related Considerations in the Design and Conduct of Early Metabolic Studies Therapeutic Indications When filing for an IND (investigator or commercial sponsored), it is necessary to specify

499

a therapeutic indication. Specifying the indication is even more important for a commercial IND.  The stated indication is used for assigning the IND to the appropriate division. For a conventional diabetes or lipid-lowering treatment indication, the review division selection is straightforward. All such indications go to the DMEP.  Diabetic complication indications go instead to the division that focuses on the affected organ system. See Table 19.1 for these assignments. The table also provides descriptions of the wording of approved therapeutic indications, the primary efficacy endpoint currently accepted by FDA and some examples of approved products. As will be discussed below, the primary efficacy endpoint for regulatory approval is of importance to the earliest studies even if it cannot be measured until later studies. To a large extent, the approved therapies define minimal efficacy targets for newer therapies. Pressure is increasingly added to early studies to provide some efficacy data that will allow an estimate of the treatment effect on the regulatory endpoint. Several novel metabolic indications are under consideration or being pursued, and all involved in the metabolic field should be aware of them (Table  19.2). Numerous attempts have been made at developing therapies for new-onset (within 6  months of diagnosis) type 1 diabetes. Four programs entered phase 3, but none has gone forward towards approval [72]. Regulatory and other considerations for early clinical studies are discussed elsewhere [73]. On the other hand, two decades of serious efforts to develop aldose reductase inhibitors for diabetic peripheral neuropathy and microvascular complications have ended. Tantalizing are the prospects for prevention of type 2 diabetes and metabolic syndrome treatment indications. FDA has provided attention in its diabetes drug guidance to these potential indications, including a general sense of what would be required for approval [74]. The growing recognition of nonalcoholic steatohepatitis (NASH) as a major cause of cirrhosis [75] is encouraging some sponsors with agents with efficacy in patients with type 2 diabetes to consider NASH as the lead indication [76–78].

G. A. Fleming and B. E. Harvey

500

Table 19.1  Approved therapeutic indications and regulatory endpoints for metabolic diseases

Indication Type 2 diabetes treatment Insulin products for type 1 and type 2 diabetes Weight loss

Phrasing Improve glycemic control in patients with type 2 diabetes Treatment of patients with diabetes mellitus for the control of hyperglycemia Chronic weight management in adults with an initial BMI of 30 kg/m2 without and 27 kg/m2 with presence of at least one weight-related comorbidities

Lipid lowering— LDL-­ cholesterol Lipid lowering— Triglycerides

To reduce elevated total-C, LDL-C, apo B, and TG levels and to increase HDL-C in patients with primary hypercholesterolemia To reduce triglyceride (TG) levels in adult patients with severe (≥ 500 mg/dL, 5.65 mmol/L) hypertriglyceridemia

Nephropathy

Treatment of diabetic nephropathy with an elevated serum creatinine and proteinuria (urinary albumin to creatinine ratio ≥300 mg/g) in patients with type 2 diabetes Treatment of diabetic macular edema Treatment of lower-extremity diabetic neuropathic ulcers that extend into the subcutaneous tissue Treatment of diabetic gastroparesis

Retinopathy Diabetic wound healing Diabetic gastroparesis

Primary efficacy endpoint HbA1c HbA1c

Body weight— continuous and categorical variables

Serum LDL-­ cholesterol levels

Serum triglyceride levels

Reduction in creatinine doubling timea

Approved therapies Many oral and injected products Many injected products Inhaled insulin Qsymia® (phentermine/ topiramate) Belviq® (lorcaserin HCl) Orlistat (Alli®) Phenteramine Many statins Ezetimibe Nicotinic acid Fenoglide® (fenofibrate) Lovaza® (DHA and EPA esters) Vascepa® (EPA ester) Fenoglide® (fenofibrate Many ACEI and ARBs

Chart reading visual function Complete wound closure

Lucentis® (Ranibizumab injection) Regranex® (becaplermin)

Symptom score, vomiting frequency, nuclear medicine gastric emptying, gastric emptying breath testb

Metoclopramide

FDA review division DMEP DMEP

DMEP

DMEP

DMEP

DcaRP

DTOP DDDP

DGIEP

Abbreviations: ACEI angiotensin-converting enzyme inhibitor, ARB angiotensin receptor blocker, DMEP Division of Metabolism and Endocrinology Products, DCaRP Division of Cardiovascular and Renal Products, DNP Division of Neurology Products, DTOP Division of Transplant and Ophthalmology Products, DGIEP Division of Gastroenterology and Inborn Errors Products, DDDP Division of Dermatology and Dental Products. See text for other abbreviations a March 2018 FDA Conference on Renal Endpoints. https://www.kidney.org/CKDEndpoints b The Cairn 13C-Spirulina Gastric Emptying Breath Test (GEBT). http://cairndiagnostics.com/cairn-gastricemptying-breath-test/

Important Early Study Considerations Phase 1 Studies Glucose-Lowering Products Phase 1 studies for most oral metabolic therapies typically involve the classic ascending single-

(SAD) and multiple-dose (MAD) safety study design with blood sampling for PK as the important secondary objective. Phase 1 studies of insulin products require a means of preventing or controlling hypoglycemia, such as a glucose clamp procedure [46]. Phase 1 studies of insulin products may be undertaken in patients with type 1 or type 2 dia-

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

501

Table 19.2  Potential therapeutic indications and regulatory endpoints for metabolic diseases Indication Cardiovascular risk reduction in patient with diabetes New onset type 1 diabetes Prevention of type 1 diabetes Prevention of type 2 diabetes

Metabolic syndrome

Nonalcoholic steatohepatitis (NASH) Cachexia/sarcopenia Diabetic peripheral neuropathy Diabetic cardiac autonomic neuropathy

Phrasing Decrease the rate of a combined endpoint of CV death, MI, or stroke. Preserve endogenous insulin secretion in newly- diagnosed patients with type 1 diabetes Decrease the rate of type 1 diabetes onset Decrease the rate of type 2 diabetes onset and reduce the rate of a combined endpoint of CV death, MI, or stroke. Improve the clinical components of metabolic syndrome and reduce the rate of a combined endpoint of CV death, MI, or stroke. Reduce the rate of fatty liver progression to hepatic fibrosis, NASH resolution, Increase or preserve physical function and muscle mass To improve signs, symptoms and dysfunction of diabetic peripheral neuropathy Reduce the rate of cardiacrelated death

Speculated efficacy endpoint Major Adverse Cardiac Events (MACE)

Review division DCaRD or DMEP

C-peptide

DMEP

Time to incident diagnosis of type 1 diabetes

DMEP

Co-primaries:  1. Time to onset of type 2 diabetes  2. MACE

DMEP

Co-primaries:  1. Composite of MS components  2. MACE

DMEP

Fibrosis and cirrhosis in biopsied liver specimens “Complete resolution of steatohepatitis and no worsening of liver fibrosis” Composite or co-primaries lead by physical function and lean body mass Composite of nerve conduction velocity and functional endpoints

DGIEP

Composite of cardiac associated death and loss of consciousness

DMEP DNP

DCaRD

Abbreviations: CV cardiovascular, DMEP Division of Metabolism and Endocrinology Products, DNP Division of Neurology Products, DGIEP Division of Gastroenterology and Inborn Errors Products

betes and/or healthy volunteers. Patients with type 1 diabetes provide the most straightforward way of eliminating endogenous insulin secretion as a confounder, but normal subjects are often used (also see Chap. 3). The starting dose and dose range are largely dependent on the results of animal disease model and toxicology studies. Testing some immunotherapies with more substantial risks may not be appropriate in normal subjects. The FDA guidance implies a preference for using healthy normal subjects for initial phase 1 studies aimed at characterizing PK parameters of non-insulin therapeutics, and that PK studies also may “be appropriate in the intended patient population.” It can be argued that PK assessment in normal subjects can be skipped and that the first in-human study can be done in patients. Normal subjects may provide somewhat less variability

in drug handling than patients without diabetic complications, but it is unlikely that presence of disease per se can significantly affect variability of drug levels, PK results, or safety assessment. FDA has allowed initial in-human studies to be done in patients, though I have seen DMEP reviewers discourage this in some cases. The rising multiple-dose study does provide for an opportunity to measure glucodynamic responses. It is therefore reasonable to use patients in favor of normal subjects in this early study to provide a preliminary dose-response with respect to glucose-­lowering activity. These data can help to select the dose range of the next study and, in some cases, provide an early estimate of glucose-­ lowering efficacy. Effects of food on the PK of oral products should be evaluated early—perhaps as soon as

502

the first in-human study. Single-dose administration of the new drug can readily be done under fed and fasting conditions. This early preliminary assessment and a more formal food effect study can be deferred if nonclinical studies have shown evidence of moderately high bioavailability under fasting conditions. A fasted dog study would generally be adequate, but rodent studies with chow-administered drug do not provide such assurance. Often the phase 1 program will not have the benefit of an optimized formulation. Ideally, the formal food effect study would be done with a reasonably advanced formulation. Other specialized clinical studies will be required at some point in phase 2, including drug-drug interactions, “thorough QT,” and renal and hepatic impairment studies, but these are not generally performed without cause in phase 1. Phase 1 studies of injected insulin products represent a special case (see Chap. 1). To avoid methodological and interpretational problems posed by endogenous insulin secretion in both normal subjects and patients with type 2 diabetes and insulin resistance in patients with type 2 diabetes, these studies are frequently done in subjects with type 1 diabetes, but certainly can be done in normal subjects. Glucose clamping reduces variability of the pharmacodynamic measures. Given the expense of these procedures and the crucial data that they provide, using patients with type 1 diabetes is sensible. Distinctions should be made between injected and alternately delivered insulin products. In most cases, oral, pulmonary, topical, and other delivery approaches have been conceived as an add-on therapy for patients with type 2 diabetes. Except for having to manage the hypoglycemic effects, these products can be approached similarly to standard oral product delivery. A carefully designed phase 1 study can serve to evaluate the possible advantages of this approach beyond timeaction profile and convenience.

G. A. Fleming and B. E. Harvey

shortest treatment duration to provide a reliable treatment effect on body weight reduction is about 8  weeks—in part to accommodate the menstrual-cycle-related variability of weight in women. Other approaches are realistic for a phase 1b/2 study to measure treatment effects on food intake, food preference, activity level, and respiratory quotient [79, 80]. Lipid-Lowering Drugs Lipid-lowering studies are the easiest phase 1 studies to execute because of the large population of otherwise healthy people with lipid abnormalities. As discussed below, measurement of the regulatory endpoint—LDL cholesterol or triglycerides—is feasible in the first multiple-dose study. LDL cholesterol levels are generally stable and not substantially affected by meals or time of day. A small phase 1b study with adequate duration in hypercholesterolemia patients could therefore potentially provide reliable efficacy readouts. In contrast, triglycerides are much more variable from hour to hour and substantially affected by meals; therefore, greater attention must be given to dietary stabilization and timing of samplings. In addition, FDA has recommended that several baseline values be obtained to compensate for this greater variability.

Efficacy Endpoints As discussed, there is a spectrum of feasibilities for measuring the regulatory primary and secondary endpoints for the respective metabolic indications in early-phase studies. Advice about the regulatory endpoints for weight loss drugs is found at the FDA website [81]. Efficacy endpoints for lipid-lowering agents is undergoing rethinking among regulators and experts [22]. A few points about regulatory endpoints for diabetes drugs are worth mentioning here. It is well known that HbA1c is currently accepted as the primary regulatory endpoint for assessment of all diabetes therapies [46], though FDA has Weight-Loss Products referred to this measure as a “surrogate measure”. Phase 1 studies of weight-loss products entail Superiority and/or non-inferiority HbA1c treatgenerally straightforward single- and multiple-­ ment effect comparisons are ultimately required of rising-­dose studies. Unlike the case for glucose-­ any glucose-lowering therapy in registration trials. lowering therapies, direct readouts on weight Approval of most therapies for type 2 diabetes prireduction are not feasible until phase 2. The marily involves demonstrating an added glucose-

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

lowering benefit compared to placebo. However, non-inferiority comparisons to an approved therapy are often needed and are frequently performed in early phase 2 studies. Because the HbA1c effect is not reliably measured in early phase 2 studies, continuous glucose monitoring (CGM) has increased the ability to project changes in HbA1c. CGM provides many more data points over a brief period of time compared to the traditional multiple daily glucose measurements [82]. Therapies aimed at improving postprandial hyperglycemia are disadvantaged by the ongoing debate regarding the clinical value or benefit to patients of this specific glucose-lowering effect [83], as well as by the fact that FDA continues to emphasize that postprandial glycemic control is not accepted as a primary efficacy endpoint. Therapies that primarily affect postprandial glycemic control are limited in the HbA1c-lowering effect that they can provide since postprandial glycemia contributes only about one-third to the HbA1c effect [84, 85]. Nonetheless, evaluation of postprandial glycemia with standardized meal tests is an important secondary endpoint and provides an early clinically meaningful evaluation of a glucose-lowering therapy. Measurement of the regulatory endpoint— LDL cholesterol or triglycerides—can easily be done in the first multiple-dose study. Increasing attention is being given in both early and late studies to the characterization of lipoprotein fractions and other biomarkers [86]. These supportive data are considered secondary endpoints and are of interest to regulatory reviewers, but do not currently play a major role in regulatory benefit/risk estimates. As previously mentioned, greater attention has to be paid to the conditions and design approaches for triglyceride-lowering therapies.

Subject Population The issue of using normal vs. diabetic subjects in phase 1 studies has been discussed above and is considered further in Chap. 18. Other important study population considerations include level of glycemic control and the use of treatmentnaïve or untreated patients with type 2 diabe-

503

tes vs. washing off or continuing background glucose-­ lowering therapy. These are actually more thorny issues for later-stage studies but do impact early multiple-dose studies in which glucose efficacy is assessed. The ideal subject for these earlier studies is the treatment-naïve patient with inadequate (but not poor) glycemic control. However, treatment-naïve patients are difficult to find; therefore, a frequent compromise is to find patients who have not recently been on a glucose-­ lowering treatment or to actually withdraw the current treatment (typically metformin and often one other drug). Medication withdrawal has its own ethical and methodological disadvantages, including having to re-establish a metabolic baseline and greater risks of screen failures and dropouts due to patients exceeding maximum permitted fasting and random blood glucose levels (see the FDA diabetes guidance for these details). In phase 1b and 2 studies in which demonstration of an impressive glucose-lowering effect is sought, the temptation is to admit patients with very poor glycemic control to allow for a potentially greater absolute treatment effect. The danger of using poorly controlled patients is that they may prove to be less responsive and/or less reliable subjects in adherence to study protocol. If their poor control involves glucose toxicity and/or substantial β-cell failure, their responsiveness to the investigative therapy may be reduced. Perhaps the largest threat is that such patients are likely to experience a strong study effect, which is the drug-independent improvement in glycemic control that results from participating in a more intensively encouraged and monitored environment (the Hawthorne effect). I have seen a number of cases in which a strong study effect undermines the treatment effect and has jeopardized the entire development program. Thus, a balance should be struck in selecting patients on the basis of glycemic control.

Number of Subjects Adequate statistical powering for efficacy endpoints is important for almost all later-phase studies, but phase 1 studies are usually not sized

504

on the basis of formal statistical considerations. As efficacy evaluation becomes the focus, formal statistical powering based on the primary endpoint is expected. While not a direct concern of the early investigator, diabetes drug development programs now require on the order of 3000 exposed patients for NDA approval, which is substantially greater than expected of other chronically administered drugs [87]. This number can easily double when a cardiovascular safety trial is required.

G. A. Fleming and B. E. Harvey

expiratory volume (FEV1), and cognitive testing, the learning effect is likely to be significant. Latin square or factorial designs are used to support the development of fixed-dose combination products of already approved products. Weight loss products that consist of two additively- or synergistically-acting drugs approved for other indications provide a more interesting example of an early clinical study requiring the use of the factorial design [88, 89]. A factorial parallel design was required to show that the combination of two agents results in greater weight loss activity than either alone. Once additivity of Study Designs each component’s efficacy and the optimal doses were established, the product could be developed Usually, initial in-human studies of oral new in a fixed-dose ratio and compared as one or more molecular entities will involve stepwise, rising-­ strengths to placebo. Such an approach would be dose designs. For safety reasons, each dose group open to novel combinations of approved drugs is usually completed before the next higher dose that, together, would form an efficacious glucoseis started. FDA in rare cases will ask that safety lowering therapy. In this case, the factorial design lab data from the first two or more dose groups could be combined with a crossover design since be evaluated before going on to the higher doses. the treatment periods could be short. The 6-week Some degree of delaying or staggering the start (or more) treatment period required for weight of successively higher dose groups is typically loss evaluation is far too long for a crossover used in phase 1 and 1b studies. design. A crossover design is sometimes used in phase 1 studies, but more typically these are used in phase 2 when an active comparator and/ Selection of Control Group, or multiple treatment permutations are involved. Randomization, and Blinding Crossover designs are generally not appropriate for early multiple-dose safety studies since Placebo comparisons are the rule in phase 1 metasome subjects would be exposed to the high dose bolic studies of non-insulin therapies. Sometimes initially. However, a crossover design has the an active comparator group is included to provide advantage of reducing imbalances in potentially an estimate of relative efficacy of the new agent. confounding covariates since each patient serves Active comparisons are the rule in injected insuas his/her own control. Crossover designs have lin product studies. In some cases, the intent is to some statistical power advantages over paral- show similar PK and PD profiles, as could be the lel designs. The disadvantages to the crossover case for development of a copy of an approved design include the “order” effect, in which the insulin product [90]. In other cases, the objective treatment order may affect efficacy response, or may be to contrast the time-action profile of the bias in adverse event reporting. The “carry-over new insulin product to that of an approved human effect” is another potential problem, particularly insulin product. Randomization is expected of virtually all in studies involving products with prolonged biologic effects. Finally, the possible bias of a phase 1 trials as a fundamental means of reducing “learning effect” is not a problem for efficacy bias. Blinding, which tends to go hand-in-hand assessments involving objective laboratory tests, with randomization, is also expected at some but for adverse event reporting and physical func- level. Double-blind is preferred over single-blind, tion tests such as treadmill performance, forced especially if a non-objective efficacy measure is

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

being used. Even when objective efficacy measures are involved, adverse event identification and reporting can be biased when the investigator knows the treatment assignment. Triple-blinding (involving the sponsor) is preferable, though not required if appropriate safeguards are taken to prevent unblinding the investigator. Blinding is often very challenging for early studies. Placebo or active comparator dose forms identical to the active dose forms may be impractical to produce for initial studies.

 hallenges with Early-Phase C Metabolic Studies With the exception of injected insulin products, there are few, if any, challenges unique to metabolic studies. Adequate documentation, subject recruitment difficulties, dropouts, and missing data are challenges faced across all therapeutic areas. Early in-human, single-, and multiple-dose studies for oral metabolic products are straightforward and usually can be completed in a single qualified center. Challenges increase to the extent that early diabetes and obesity studies involve patients and objectives to assess efficacy. Larger and more ambitious early studies start to encounter the same issues that plague later-stage studies. Documentation, data acquisition, and data management involved in the clinical trial process are going through a transition from paper to being electronically based. As mentioned, all FDA submissions will be required to be in a highly structured electronic form [91]. The standards for electronic data and regulatory requirements for electronic data handling have been established [92]. These developments are putting pressure on early-stage companies, smaller investigation sites, and academic sites in which legacy paper systems are more cost-effective and work perfectly well. All these transitions to electronic systems will provide cost and time savings for FDA and the sponsor over time as the development program progresses and they are intended to improve overall results in the therapeutic development process. With the majority of development programs failing to advance beyond early-phase development,

505

the electronic filing requirements impose a proportionately greater burden on early studies and their sites and sponsors. Many early multiple-dose metabolic studies are on the cusp of requiring that multiple sites enroll adequate numbers of subjects, and that this be completed in a reasonable amount of time. Multiple investigation sites drive up costs and variability. Organizations that specialize in phase 1 metabolic studies have frequently been associated with a large clinical specialty practice or have assembled a comparable means to facilitate patient recruitment. In all cases, care has to be taken regarding the suitability of subjects recruited for these studies. For multiple reasons, those patients who have had poor and/or inconsistent care for whatever reason are generally not good subjects for investigation. They are most likely to fail to comply with the protocol and/or drop out of the study. Given the high per-subject cost of early clinical studies, dropouts must be minimized. Patients with poor glycemic control often present a higher risk of dropout and/or study effects. Early weight loss studies aimed at exploration of anti-obesity effects entail special challenges in terms of demographics, logistics, facilities and biologic limitations [93]. Women are clinical candidates for early trials on weight loss drug therapy or other interventions, but the influence of the menstrual cycle on early efficacy indicators for anti-obesity effects is an additional source of variability. Men can preferentially be utilized to reduce variability and increase statistical power to detect an early treatment effect, but the results may not be reflective of results in the general population, which for weight loss drugs consist predominately of women. Obese subjects require compassionate and sensitive attention during the entire investigative process, starting with screening and baseline testing. Sites specialized for morbidly obese studies are well adapted to the requirements for venous blood access and largesize examination tables, blood pressure cuffs and imaging equipment. Non-specialized sites may not be adequately equipped. The greatest challenge to efficacy assessment of weight loss therapies is the inherent biologic resistance to weight loss and the resulting small, slow rate of

506

weight loss. This effect may be highly clinically meaningful if sustained but is difficult to detect in smaller and shorter studies. Weight loss drugs have been a major contributor to the concern about dropouts and missing data in registration trials [94]. This general concern about registration trials has resulted in a guidance for handling missing data, which has been adopted by FDA [95]. This increased concern has pushed down into early clinical studies. Traditionally, early studies have not been closely scrutinized by FDA for trial execution and interpretation problems since it is the later trials that will be pivotal to approval. However, a high dropout rate in an early study will likely be noticed by FDA reviewers and perhaps reduce confidence in subsequent data. In our experience, insufficient attention to the main contributors of a study effect bias has often led to misleading results of early and late studies. This is not so much an issue for small, in-house studies that involve rigorous protocol designs. However, when subjects are tested as outpatients, the study effect and dropouts become much more of a challenge. Because glycemic control is sensitive to diet and exercise, a patient can unwittingly improve glycemic measures independent of assigned treatment. Patients are embarrassed by starting a study with poor metabolic control and naturally want to improve it. Some patients who claimed compliance with background metformin therapy may start to make good on that claim in the study or even add dietary supplements, forgetting that this is prohibited. Patients who are given glucose monitors or continue to use them in a study are somewhat unblinded and may consciously or unconsciously strive to improve glycemic control. Use of placebo run-in periods to reduce such effects is recommended by FDA, particularly for weight loss studies. It is important even in early studies to avoid any major departure in the patient’s care while the patient is in a study, and to avoid subjects who will require major changes in their clinical care. Finally, sloppiness and fraud should be mentioned as a threat across the entire spectrum of biomedical research. Examples of investigators who intentionally set out to commit fraud are very rare. More often, discovered cases of fraud

G. A. Fleming and B. E. Harvey

that are documented on the FDA website [96] started down the slippery slope of cutting corners and taking on more studies than advisable. The increasing pressure on small studies to deliver more with less requires greater attention to trial monitoring and quality assurance.

 he Over-arching Goals of Early T Metabolic Studies It is clear that the over-arching purpose of early-­ phase metabolic studies is to screen out further development of compounds that fail to achieve preliminary targets and to provide a solid foundation for performing later-phase and much more expensive studies [97]. The makeup of that foundation is more complicated. Timing of specialized studies to address important safety questions involves good judgment even with the availability of regulatory guidances and expert opinions. A compromise on every trial design element is to some extent always required. The choice is less often between the good and the perfect; it is more often between the acceptable and the good. One example of a strategic compromise that is increasingly being made is to perform “proof of concept” (POC) trials as soon as possible for a new molecular entity that has gone through some preliminary vetting of safety and efficacy. For a diabetes product, a POC trial would typically involve 12 weeks’ treatment in a patient population with one or two doses compared to placebo and/or sometimes to an active comparator. With a positive POC study on the primary efficacy endpoint, a sponsor can then go back with more confidence to invest in the earlier studies that might have been delayed or abbreviated. This strategic approach has an impact on the “what” and the “how” of early clinical studies.

 ooking Ahead to the Future of Early L Metabolic Studies Some challenges for early metabolic therapeutic investigation have been summarized. One challenge that is not foreseeable is a day when early

19  Regulatory Considerations for Early Study and Clinical Development of Drugs for Diabetes, Obesity…

clinical studies will be obviated by the technological advances in in silico, in vitro, and genetically modified animal testing. Such advances are likely to complement the clinical trial enterprise and accelerate overall success in meeting unmet clinical need and individualizing therapeutics. It is difficult to imagine that the need for early trial data will decline in any substantial way. Truly exciting are the prospects for new metabolic indications, which can potentially avert metabolic diseases as we now know them and leverage major savings in lives and resources. Glucose lowering by itself is a palliative approach for managing a late-stage manifestation of a metabolic disorder that has reached the threshold for a diabetes diagnosis. Sponsors will no longer invest the roughly $1 billion required to develop a glucose-lowering therapy that does not have promise of providing major clinical outcome benefits [98]. Furthermore, we should not confine our focus to just diabetes, as we have become increasingly aware of the massive the iceberg population that closely underlies the disease. Advances in metabolic research have increased our understandings of the metabolic basis of cancer, dementia, and aging. We should lift our eyes to a horizon beyond that of conventional metabolic disease treatment and seek to address the significant unmet need for preventing metabolic diseases and their complications. Early clinical studies start to provide a means of identifying the promise of metabolic therapies. Promising metabolic compounds now have an expanding list of potential applications.

 pecial Considerations for Therapies S Aimed at Nonalcoholic Fatty Liver Disease (NAFLD) and Nonalcoholic Steatohepatitis (NASH) Nonalcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver disease in the industrialized world [99]. The histologic phenotypes of the disease extend from NAFLD to NASH.  A definitive diagnosis of NASH is currently based upon histological evidence of fat accumulation (steatosis) in hepatocytes, as well

507

as liver-cell injury and death, with the accumulation of inflammatory cells. NASH can progress to liver fibrosis and end stage cirrhosis [100]. NASH is also associated with an increased risk of cardiovascular morbidity and mortality, as well as type 2 diabetes [101]. Cirrhosis associated with NASH increases the risk of primary hepatocellular carcinoma. Currently, liver biopsy is the “gold standard” and the accepted method for the diagnosis of NASH and to accurately assess progression to cirrhosis [102]. There are currently no FDA approved drug or biological treatments for NASH [103].

 urrent Regulatory Environment at C U.S. FDA for NASH and Related Disorders The regulation of NASH and related fatty liver disorder treatments occurs in the Division of Gastroenterology and Inborn Errors Products (DGIEP). There is a dedicated team within the division focused upon liver disease treatments, such as NASH and they work in parallel with teams responsible for inborn errors of metabolism and a team for gastrointestinal disorders, such as Crohn’s disease and ulcerative colitis [104]. Although there is no formal guidance document at the time of this book’s publication for NASH treatment development, FDA has been active at academic liver conferences in providing detailed advice regarding clinical trial designs that will have an increased chance of generating data to support agency approval of a drug or biologic product. At a recent “Liver Forum”, which is a public-private-academic partnership formed to promote the exchange of ideas regarding clinical trials in liver disease [105], an FDA presentation [106] focused upon NASH, with an emphasis on appropriate clinical trial design that could increase the odds for generating data that will meet the FDA threshold for drug or biologic product approval. The importance of selecting endpoints for the proposed pivotal clinical trial was emphasized. Traditional clinical endpoints were discussed as well as intermediate clini-

G. A. Fleming and B. E. Harvey

508

cal endpoints and surrogate endpoints reasonably likely to predict clinical benefit in patients [107]. Examples of traditional clinical endpoints included “Reduction in All-Cause Mortality”, “Prevention of Liver Transplantation” (e.g., Model for end-stage liver disease [MELD] Score Increase to 15 from less/equal to 12 = listing for liver transplant) and “Reduction of Decompensation Events [108]. The FDA examples for potential surrogate endpoints for NASH were “Complete Resolution of Steatohepatitis” and “No Worsening of Liver Fibrosis” and “at least 1-point improvement in fibrosis using the Brunt/Kleiner Scale [109] and perhaps 2-point improvement would be a more durable regulatory endpoint5 and “no worsening of steatohepatitis, defined as no increase in ballooning or inflammation on NAS (NAFLD Activity Score)”. A surrogate reasonably likely to predict clinical benefit could be used as the basis of an Accelerated Approval by FDA. Since 2012, “surrogate and intermediate clinical endpoint” language has been added to the U.S. law.6 The Accelerated Approval must be followed by a Verification Trial with a traditional clinical endpoint. A potential alternative for verification is a “seamless” phase 3/4 design where patients are rolled over at the end of phase 3 into a phase 4 verification trial. Recent FDA advice has included Phase 2b/3/4 design, where some histology-­ based evidence of efficacy is provided in Phase 2b, which acts as a regulatory gateway into the pivotal phase of development.7 The agency has expressed concerns in public forums regarding the exposure of a large number of patients for long periods of time with only minimal data FDA Staff, Personal Communication (2017). “In 2012, Congress passed the Food and Drug Administration Safety Innovations Act (FDASIA). Section 901 of FDASIA amends the Federal Food, Drug, and Cosmetic Act (FD&C Act) to allow the FDA to base accelerated approval for drugs for serious conditions that fill an unmet medical need on whether the drug has an effect on a surrogate or an intermediate clinical endpoint.” https://www.fda.gov/ForPatients/Approvals/Fast/ ucm405447.htm 7  FDA Staff, Personal Communication (2018). 5  6 

supporting efficacy. Although this is not the traditional role of FDA in regulatory oversight of IND trials, it is understandable given the large number of industry sponsors pursuing NASH development programs and the limited number of biopsy diagnosed NASH patients available for clinical trial enrollment. It is also consistent with the agency’s duel mission to both “protect” and “promote” the public health articulated since 1997 [110]. Further advice provided by FDA at the Fifth Liver Forum focused upon when a sponsor is deciding between conducting either one or two pivotal trials to support product approval, the following FDA perspective is important to consider. In order for one phase 3 trial to be persuasive, the following FDA Guidance “Providing Clinical Evidence and Effectiveness for Human Drug and Biologic Products for Human Drug and Biologic Products” was referenced: “Generally limited to situations in which a trial has demonstrated a clinically meaningful effect on mortality, irreversible morbidity, or prevention of a disease with potentially serious outcome and conformation of the results in a second trial would be practically or ethically impossible” [111]. Specifically, there should be a large, multicenter trial, with internal consistency across study sites, with evidence of an effect on multiple endpoints and statistically very persuasive efficacy results (“robust p value”).8 As an alternative, if a sponsor is planning for the Accelerated Approval pathway using two seamless phase 3/4 trials, then the following statistical considerations will be applied by FDA reviewers. The alpha (α) must be controlled at or below 0.05 for each of the Phase 3/4 trials. For example, for α  =  0.01 for phase 3 (liver histology-based endpoints), then α must be less than 0.04 for phase 4 (traditional clinical endpoints). However, if the sponsor is planning only one (1) seamless phase 3/4 trial to submit for FDA Approval, then the overall α less than 0.05 and split between Phase 3 (α