Magnetic Resonance in Oncology [1st ed.] 978-3-540-51194-6;978-3-642-74706-9

370 8 28MB

English Pages XIII, 176 [177] Year 1990

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Magnetic Resonance in Oncology [1st ed.]
 978-3-540-51194-6;978-3-642-74706-9

Table of contents :
Front Matter ....Pages I-XIII
Magnetic Resonance Imaging in Oncology (A. Breit)....Pages 1-2
Physical Principles and Signal Behaviour in Magnetic Resonance Imaging (H. Kett, C. Prüll)....Pages 3-14
Magnetic Resonance Imaging of the Brain in Oncology (G. Sze)....Pages 15-28
Magnetic Resonance Imaging of the Orbit in Oncology (G. Sze, G. Wilms)....Pages 29-39
Magnetic Resonance Imaging of the Spine in Oncology (G. Sze)....Pages 41-54
Magnetic Resonance Imaging of Tumours of the Head and Neck (P. Held, N. Obletter)....Pages 55-67
Magnetic Resonance Imaging of the Chest (R. Langer, R. Felix)....Pages 69-73
Magnetic Resonance Imaging of Breast Tumours: Techniques, Indications (W. A. Kaiser)....Pages 75-85
The Value of Magnetic Resonance Imaging in Abdominal Tumours (R. Langer, R. Felix)....Pages 87-95
Magnetic Resonance Imaging of the Retroperitoneum (R. Musumeci, L. Balzarini, E. Ceglia, R. Petrillo, J. D. Tesoro Tess)....Pages 97-103
Magnetic Resonance Imaging of the Urinary Tract and Male Genitalia (R. Musumeci, L. Balzarini, E. Ceglia, R. Petrillo, J. D. Tesoro Tess, C. Massari)....Pages 105-116
Magnetic Resonance Imaging of the Pathological Female Pelvis: Assessment of 24 Patients with High-Field Magnetic Resonance Imaging (P. Demaerel, A. L. Baert, P. Ide, N. Obletter, G. Marchal, P. Van Hecke et al.)....Pages 117-124
Magnetic Resonance Imaging in Soft Tissue Tumours (J. Gielen, A. L. Baert, G. Marchal, W. Wouters, L. Vanfraeyenhofen, P. Van Hecke)....Pages 125-135
Magnetic Resonance Imaging of Primary Bone Tumours (J. Gielen, A. L. Baert, G. Marchal, P. Demaerel, L. Vanfraeyenhoven, P. Van Hecke)....Pages 137-150
The Value of Magnetic Resonance Imaging in Three-Dimensional Treatment Planning in Radiation Therapy (P. Kneschaurek, H. Kett, C. Prüll, A. Breit)....Pages 151-155
In Vivo Magnetic Resonance Spectroscopy: Basic Principles and Clinical Applications in Oncology (W. Semmler)....Pages 157-173
Back Matter ....Pages 175-176

Citation preview

••••• •• : •• • Monographs Series Editor: U.Veronesi

A. Breit (Editor-in-Chief)

Magnetic Resonance in Oncology A.L. Baert, R. Felix, R. Musumeci, W. Semmler and G. Sze (Co-Editors)

With 147 Figures and 7 Tables

Springer-Verlag Berlin Heidelberg GmbH

Professor Dr. ALFRED BREIT

Professor Dr. RENATO MUSUMECI

Institut und Poliklinik fOr Strahlentherapie

Radiodiagnostica

und Radiologische Onkologie

Sezione RMN - Urologia Linfografia

der Technischen Universitat MOnchen,

Istituto Nazionale Tumori

Klinikum rechts der Isar

Via Venezian 1

Ismaninger StraBe 15

20133 Milano, Italy

8000 MOnchen 80, FRG Dr. WOLFHARD SEMMLER

Professor Dr. ALBERT L. BAERT

Deutsches Krebsforschungszentrum

Department of Radiology

Heidelberg

Universitaire Ziekenhuizen

Institut fOr Nuklearmedizin

Herestraat 49

1m Neuenheimer Feld 280

3000 Leuven, Belgium

6900 Heidelberg 1, FRG

Professor Dr. ROLAND FELIX

Professor Dr. GORDON SZE

Abteilung Radiologie Universitatsklinikum Rudolf Virchow, Standort Charlottenburg,

Department of Diagnostic Radiology Yale University School of Medicine 333 Cedar Street, P.O. Box 3333

Freie Universitat Berlin

New Haven, CT 06510, USA

Spandauer Damm 130 1000 Berlin 19, FRG

The European School of Oncology gratefully acknowledges sponsorship for the Task Force "Nuclear Magnetic Resonance in Oncology" chaired by A. Breit (Munich), received from Siemens S.p.A. Medical Division - Milan.

ISBN 978-3-642-74708-3

ISBN 978-3-642-74706-9 (eBook)

DOI 10.1007/978-3-642-74706-9

Library of Congress Cataloging· in-Publication Data Magnetic resonance in oncology I A. Breit, editor-in-chief ; A. L. Baert ... let al.), co-eds. - (Monographs I European School of Oncology) Results of study group meetings conducted by the European School of Oncology. Includes bibliographical references. ISBN 0-387-51054-0 (U.S. : alk. paper) 1. Cancer - Magnetic resonance imaging - Congresses. I. Breit, Alfred. II. Baert, A. L. (Albert L.), 1931- . III. European School of Oncology. IV. Series: Monographs (European School of Oncology) RC270.3.M33M34 1990 616.99'207548 - dc20 89-21774 CIP This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, reCitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only permitted under the provisions of the German Copyright Law of September 9, 1965, in its version of June 24, 1985, and a copyright fee must always be paid. Violations fall under the prosecution act of the German Copyright Law. © Springer-Verlag Berlin Heidelberg 1990 Originally published by Springer-Verlag Berlin Heidelberg New York in 1990 Softcover reprint of the hardcover 1st edition 1990 The use of general descriptive names, registered names, trademarks, 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. Product Liability: The publisher can give no guarantee for information about drug dosage and application thereof contained in this book. In every individual case the respective user must check its accuracy by consulting other pharmaceutical literature. Typesetting and Printing: Druckhaus Beltz, Hemsbach/Bergstr.; Bookbinding: J. Schaffer GmbH & Co. KG, GrOnstadt 2123/3145-543210 - Printed on acid-free paper

Foreword

The European School of Oncology came into existence to respond to a need for information, education and training in the field of the diagnosis and treatment of cancer. There are two main reasons why such an initiative was considered necessary. Firstly, the teaching of oncology requires a rigorously multidisciplinary approach which is difficult for the Universities to put into practice since their system is mainly disciplinary orientated. Secondly, the rate of technological development that impinges on the diagnosis and treatment of cancer has been so rapid that it is not an easy task for medical faculties to adapt their curricula flexibly. With its residential courses for organ pathologies and the seminars on new techniques (laser, monoclonal antibodies, imaging techniques etc.) or on the principal therapeutic controversies (conservative or mutilating surgery, primary or adjuvant chemotherapy, radiotherapy alone or integrated), it is the ambition of the European School of Oncology to fill a cultural and scientific gap and, thereby, create a bridge between the University and Industry and between these two and daily medical practice. One of the more recent initiatives of ESC has been the institution of permanent study groups, also called task forces, where a limited number of leading experts are invited to meet once a year with the aim of defining the state of the art and possibly reaching a consensus on future developments in specific fields of oncology. The ESC Monograph series was designed with the specific purpose of disseminating the results of these study group meetings, and providing concise and updated reviews of the topic discussed. It was decided to keep the layout relatively simple, in order to restrict the costs and make the monographs available in the shortest possible time, thus overcoming a common problem in medical literature: that of the material being outdated even before publication.

UMBERTO VERONESI

Chairman, Scientific Committee European School of Oncology

Preface

This monograph has been devoted to magnetic resonance (MR) in oncology and provides a concise overview of magnetic resonance tomography. Further aspects of this method, as far as we considered them to be relevant, have also been presented. Because the field is complex, this collection of papers can only be general in character. Up until now oncological aspects of MR have only been dealt with in broad-based monographs or in discussims of individual organs. A general synopsis was missing from the literature. Thus, the aim of the present volume was first to show MR's sensitivity in all regions of the body. In doing so, regions such as the abdomen and retroperitoneum in which MR is not yet considered to be the preferred method have been covered. In the chapter on radiotherapy planning particular importance has been placed on MR's new role in the staging programme as a whole, where CT is the current standard. When available, however, MR is becoming the preferred method because multiplanar images with clearly superior tissue resolution can be provided. The chapter on technique which discusses three-dimensional volume modes with post-processing demonstrates this. Physical radiotherapy planning now calls for a three-dimensional programme. As soon as a tumour is discovered and the results of histological examination are available, the clinician can decide on further therapeutic strategies based on the MR images. In order to do this, particular imaging procedures, mainly three-dimensional are necessary in each tumour location. In a book of this kind spectroscopic methods must be included since in the future they will definitely improve specificity and, in particular, provide information about response to radiochemotherapy. Both chemotherapeutic and radiotherapeutic techniques will benefit from this. I hope that this book will be a guide to a sensible and positive use of this new and promising technique, particularly for all clinicians working in oncology.

ALFRED BREIT

Contents

Magnetic Resonance Imaging in Oncology

A. BREIT . . . . . . . . . . . . . . . . . . Physical Principles and Signal Behaviour in Magnetic Resonance Imaging H. KETT and C. PRULL . . . . . . . . . . . . . . . . . . . . . . . . . . .

3

Magnetic Resonance Imaging ofthe Brain in Oncology G. SZE . . . . . . . . . . . . . . . . . . . . . . . . .

15

Magnetic Resonance Imaging ofthe Orbit in Oncology G. SZE and G. WILMS . . . . . . . . . . . . . . . . .

29

Magnetic Resonance Imaging ofthe Spine in Oncology G. SZE . . . . . . . . . . . . . . . . . . . . . . . . .

41

Magnetic Resonance Imaging ofTumours ofthe Head and Neck P. HELD and N. OSLETTER . . . . . . . . . . . . . . . . . . . .

55

Magnetic Resonance Imaging ofthe Chest R. LANGER and R. FELIx . . . . . . . . . .

69

Magnetic Resonance Imaging of BreastTumours: Techniques, Indications W. A. KAISER . . . . . . . . . . . . . . . . . . . . . . . . . . . .

75

The Value of Magnetic Resonance Imaging in Abdominal Tumours R. LANGER and R. FELIX . . . . . . . . . . . . . . . . . . . . . .

87

Magnetic Resonance Imaging of the Retroperitoneum R. MUSUMECI, L. BALZARINI, E. CEGLlA, R. PETRILLO and J. D. TESORO TESS.

97

Magnetic Resonance Imaging of the Urinary Tract and Male Genitalia R. MUSUMECI, L. BALZARINI, E. CEGLlA, R. PETRILLO, J.D. TESORO TESS and C. MASSARI. 105 Magnetic Resonance Imaging ofthe Pathological Female Pelvis: Assessment of 24 Patients with High-Field Magnetic Resonance Imaging P. DEMAEREL, A. L. BAERT, P.IDE, N. OSLETTER, G. MARCHAL, P. VAN HECKE and J. BONTE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 117 Magnetic Resonance Imaging in SoftTissueTumours

J. GIELEN, A. L. BAERT, G. MARCHAL, W. WOUTERS, L. VAN FRAEYENHOVEN and

P. VAN HECKE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

X Contents Magnetic Resonance Imaging of Primary Bone Tumours

J. GIELEN,A. L. BAERT, G. MARCHAL, P. DEMAEREL, L. VAN FRAEYENHOVEN and P. VAN HECKE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

. . . . 137

The Value of Magnetic Resonance Imaging in Three-Dimensional Treatment Planning in Radiation Therapy P. KNESCHAUREK, H. KETT, C. PRULL and A. BREIT . . . . . . . . . . . . . . . . . . . 151 In Vivo Magnetic Resonance Spectroscopy: Basic Principles and Clinical Applications in Oncology W. SEMMLER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Contributors

A.L.BAERT

117,125,137

L. BALZARINI 97, 105 J. BONTE 117 A.

BREIT

E.

CEGLIA

1, 151 97, 105 P. DEMAEREL 117, 137 R. FELIX 69,87 J. GIELEN 125, 137 P. HELD 55 P. IDE 117 WA. KAISER 75 H. KETT 3, 151 P. KNESCHAUREK 151 R. LANGER 69, 87

117, 125, 137 105 R. MUSUMECI 97, 105 N.OBLETTER 55, 117 R. PETRILLO 97, 105 C. PROLL 3, 151 W SEMMLER 157 G. SZE 15, 29, 41 J. D. TESORO TESS 97, 105 L. VAN FRAEYENHOVEN 125, 137 P. VAN HECKE 117, 125, 137 G. WILMS 29 W WOUTERS 125 G. MARCHAL C. MASSARI

You will find the addresses at the beginning of the respective contribution

Abbreviations

CE FAST CE FFE CNR CT DRESS EPI FAST FFE FID FISP FLASH FROGS FT G

'Y GE GRASS HFI IR ISIS ME MR(I) MRS w PIXEL PSIF RARE Q

RF SE SNR SSFP STIR

T1 T2 T2* TE TI TR VOXEL

Flip angle Contrast-enhanced FAST Contrast-enhanced FFE Contrastto noise ratio Computed tomography Depth-resolved surface coil spectroscopy Echo planar imaging Fourier acquired steady-state Fast field echo Free induction decay Fast imaging with steady precession Fast low angle shot Fast rotating gradient spectroscopy Fourier tr~nsformation Field gradient (measured inT/m) Gyromagnetic ratio Gradient echo Gradient-recalled acquisition in the steady state Half Fourier imaging Inversion recovery Image-selected in vivo spectroscopy Multi echo Magnetic resonance (imaging) Magnetic resonance spectroscopy (Angular) frequency Picture Element Reversed FISP Rapid acquisition with relaxation enhancement Spin density Radio frequency Spin echo Signal-to-noise ratio Steady-state free precession Short tau inversion recovery Spin-lattice relaxation time Spin-spin relaxation time Effective spin-spin relaxation time Echo time Inversion time Repetition time Volume element

Magnetic Resonance Imaging in Oncology A.

BREIT

Institut und Poliklinik fOr Strahlentherapie und Radiologische Onkologie der Technischen Universitat Munchen Klinikum rechts der lsar, Ismaninger StraBe 15, 8000 Munchen 80, Federal Republic of Germany ,

Introduction Tumors and tumor tissue were already being examined using magnetic resonance as early as 1971 (Damadian 1971, Lauterbur 1973). By 1973 the first images had been published (Margulis et al. 1983). From the beginning this method gave promise to play an outstanding role in oncological diagnosis since it provides anatomical information in arbitrary imaging planes or as a threedimensional display and allows tissue to be characterized, a previously unknown possibility. About 7 years have passed since magnetic resonance imaging was introduced as a routine examination method in clinical tumor diagnosis. (Margulis et al. 1983). In the first years the neurocranium and the spinal cord were the primary areas of concern. Now almost all body regions can be examined, as a result of the extraordinarily rapid developments in hardware and software. The demands placed on physicians and diagnosticians were and are extremely high, both in examination techniques and MR image evaluation. In the field of oncology extensive experience has already been gained with conventional spin-echo procedures. New acquisition modes such as gradient echo sequences and volume scanning techniques - although image contrasts sometimes Significantly differ from those obtained with spin-echo techniques - must now be tested in the daily routine. This is also required for dynamic examinations of the vessel achitecture in tumors (cine modes). Special attention has been given to the indications and results of MR examinations with paramagnetic contrast agents like Gadolinium DTPA. At present it is essential for the Eurpean School of Oncology to make a statement about the current possibilities and limitations of MR tumor diagnosis and spectroscopy. As with all current publications on MR imaging, it is difficult to critically

assess the new developments mentioned above which are now becoming part of the routine for clinical diagnosis. It is important, however, to avoid misleading statements. The volume of the contributions naturally had to be limited. The authors have tried to present the current state of the art by giving a variety of examples and taking specific studies as well as their own experience into account. The selection of cases and anatomical regions was based on the presently known range of indications (NIH Conference) and the partly completed indication catalogues, e.g. in the Federal Republic of Germany (FRG). For the selection of cases special attention was given to the frequency of occurrence. In all the contributions an attempt was made to give an overview about sensitivity and specifity of MR imaging. Other imaging modalities currently used, especially CT, are included to some extent for comparison. As far as the material of the book as a whole is concerned, it was important to present information still up-to-date at the time of publication. A scope and the latest developments in gradientecho techniques are included in the technical section. However, only a short overview can be given. We recommend that the reader also consult more in-depth monographs on specific subjects and questions. Spectroscopy, although still in the initial stages of clinical application, will certainly playa significant role in oncology in the future (responsiveness tests, monitoring of radiochemotherapy). I would like to thank all my co-workers and express my particular gratitude to Prof. Baert, Leuven, who supported me with advice throughout. At the same time I would like to thank Ms. Rutlidge and Dr. Costa for their constant help.

2

A. Breit

References 1 Damadian R (1971) Tumor detection by nuclear magnetic resonance. Science 171 :1151 2 Lauterbur PC (1973) Image foundation by induced local interaction: example employing nuclear magnetic resonance. Nature 243:190 3 Margulis AR, Higgins CB, Kaufmann L, Crooks LE (1983) Clinical magnetic resonance imaging. Radiology Research and Education Foundation, San Francisco

Physical Principles and Signal Behaviour in Magnetic Resonance Imaging H. KErr and C.

PRUll

Abteilung Radiologie, Klinikum Passau, Lehrkrankenhaus der Technischen Universitat Munchen, Bischof-Piligrim-Strasse 1, 8390 Passau, Federal Republic of Germany

Nuclear Magnetic Resonance The absorption and emission of radio frequency energy by proton spins in external magnetic fields are the basis of magnetic resonance imaging (MRI). Fundamental magnetic properties of electrons or protons are described by the concept spin. Spins in external magnetic fields can be compared with spinning microscopic bar magnets and precess around the direction of the magnetic field like a rotating top in the gravitation field. The frequency of precession is proportional to the applied magnetic field Bo (measured in T = Tesla, Bearth = 50 !l

n:

(1 )

wo=yBo

The gyromagnetic ratio y is a physical constant, characteristic for the nucleus, and has the value of 42.58 MHzfTesla for the 1H nucleus (proton). There are two different energy levels E1 and E2 for

E

proton spins in external magnetic fields with an energy difference L\E, which is proportional to the magnetic field. The population of the energy levels in thermal equilibrium as a function of temperature is given by the Boltzmann equation (see also Fig.

1).

n

NiN1 = exp (- L\Elks

(2)

N1, N2: number of spins in the energy level E1, E2 ks: Boltzmann's constant 1.381 10-23 J/K T: absolute temperature The spins align parallel to the applied magnetic field in the lower energy level E1 and anti parallel in the higher level E2. The resulting macroscopic magnetization is proportional to the difference of the number of spins in both levels. At a temperature of 37°C and for a magnetic field of 1 T, only a small fraction of spins (0.0002%) contributes to the magnetization.

Resonance Excitation

N

Fig. 1. Energy levels in an external magnetic field. The population of the energy levels E1 and E2 in a magnetic field Bo is determined by the Boltzmann distribution: the number of spins N1 and N2 in the energy levels E1 and E2 depends on the temperature T and the energy difference ~E = E2 - E" which is proportional to the applied magnetic field Bo

The thermal equilibrium can be disturbed by electromagnetic waves which have exactly the same frequency as the precessing spins. At this resonance frequency wo, the spins in the lower energy level E1 absorb radio frequency (RF) energy and are transferred to the higher level E2. When the magnetic component of the RF field acts in a direction perpendicular (x) to the static field Bo (z), the magnetization vector is tilted by an angle a away from its initial state, and the motion of the spin components in the x-y plane is synchronized. The flip angle a depends on the amplitude and the duration of the RF pulse. Saturation (a = 90°) and inversion (a = 180°) of the spin system are commonly used disturbances of the thermal equilibrium in MRI (Fig. 2). These effects can easily be described in a frame x', y', z rotating with Wo

4

H. Kett and C. Prull

z

z

z b)

a)

c)

y'

y' x' E2

E1

tttft

y'

x'

x'

E2

E2

E1

tttt

E1

ttt

Fig. 2a-c. Spin system in thermal equilibrium and in the excited state. Spin system in a in thermal equilibrium, b following a 90° excitation pulse (saturation), c following a 180°excitation pulse (inversion). The bold arrow represents the net magnetization in the rotating frame x', y', z; the corresponding population of the energy levels El and E2 with spins is shown below

about the field axis z. Spins precessing with a frequency Wo are stationary in this frame of reference. Relaxation

If the RF pulse is turned off, the spin system relaxes at a tissue-dependent rate into the thermal equilibrium, emitting RF waves. There are two different mechanisms of relaxation, the spinlattice relaxation (time constant T1) and the spinspin relaxation (time constant T2). The origin of relaxation is, among other effects, dipolar fields of the surrounding molecules. If these dipolar fields have frequency components with the precession frequency wo, relaxation processes are induced. Consequently, the relaxation rate depends on the mobility of the adjacent molecules and is, therefore, tissue dependent. The relaxation time T1 increases with increasing magnetic field [1], T2 is nearly field independent (Table 1). Table 1. Relative spin density [2] and relaxation times at 1 Tesla [3] Tissue

T1 (in ms)

muscle fat grey matter white matter

730 240 810 680

± 30 ± 70 ± 140 ± 120

T2 (in ms) 47 84 101 92

± 13 ± 36 ± 13 ± 22

Q

0.75 0.90 0.90 0.75

Spin-Lattice Relaxation

The reorientation of the magnetization Mz parallel to the applied magnetic field (longitudinal magnetization), caused by the spin-lattice relaxation, is characterized by the time constant T1 (Fig. 3). Pathological tissues often show changed relaxation times T1 compared to normal tissue, and that is why they can be detected sensitively by MRI. Spin-Spin Relaxation

Following the RF pulse, the excited spins are synchronized or in phase. In the case of a 90° excitation pulse the spin vectors are then parallel to the y' axis of the rotating frame. Magnetic interactions between the spins cause a dephasing of the spins. The decay of the transversal magnetization is characterized by the relaxation time T2 (Fig. 4). Just as T1, T2 is tissue dependent, but always shorter than T1. T2* Relaxation

In practice, magnetic fields are never ideally homogeneous. Therefore, spins at different locations precess with different resonance frequencies and lose the original phase coherence similar to the spin-spin relaxation process. The rate ofthe effective dephasing depends on both effects and

Physical Principles and Signal Behaviour in Magnetic Resonance Imaging

ents of the magnetization M x' and My' perpendicular to the magnetic field can be measured. The decay of these components is characterized by the relaxation time T2* and is called free induction deday (FI D). By means of the spin-echo technique (Fig. 5) the influence of field inhomogeneities is compensated. .

is described by the time constant T2*, which is always shorter than T2:

1!T2*

= 1!T2 + yilB

5

(3)

ilB: inhomogeneity of the magnetic field [4]. In the resonance experiment, only the compon-

MZ(t) = Mz(O) ( 1 - exp (-t/T1) ) Mz(O) (1-1/e)

t

t=O

t=T1

Fig.3. Excitation and spin-lattice relaxation. Following the 90°excitation pulse the net magnetization is tilted away from the z axis; therefore, spin-lattice relaxation results in an increase of the longitudinal magnetization Mz. At the time t = T1 (spin-lattice relaxation time) the magnetization Mz has reached the value Mz(O) (1-1/e) = 0.63 Mz(O), broken lines

Mx',y'

MX',y,(t) = Mz(O) exp (-t/T2)

t

t=O

t=T2

z

z

t=o

t>T2

t=T2

y' x'

z

y'

x'

y'

x'

Fig.4. Excitation and spin-spin relaxation. Following the 90° excitation pulse all excited spins are in phase, the transversal magnetization My' is maximal. Spin-lattice and spin-spin relaxation occur simultaneously: the spins lose their phase coherence, they drift apart and the magnetization in the x' -y' plane decreases exponentially. At the time t = T2 (spin-spin relaxation time) the magnetization in x' or y' direction has reached the value 1/e Mz(O) = 0.37 Mz{O)

6

H. Kett and C. PrOIl

z

z

z

y'

y'

z

y'

y' x'

x' t=O

O-

'iii c

....II)

Fig. 3 a, b. 1H MRS. a 1H spectroscopy of muscle tissue in a healthy volunteer using a conventional FlO sequence. Only the water and lipid resonances can be observed. b Water-supressed 1H MR spectroscopy of the brain. The small resonances of choline, creatine, N-acetyl-aspartate (NNA), and lactate are appreciated [8]

.5 "6 c

Cl

en

8.0

7.0

s.o

5.0

4.0

Chemical shift

3.0 [ppm]

2.0

1.0

0.0

-1.0

160

W. Semmler

a

H H -C-O-

I

H -C-O-

I

H -C-OH

120

160

200

80

o

40

Chemical shift [ppm 1

b 1H-decoupled

1H-coupled

i

72.0

,

32.0

72:0

64:0

56:0

48:0

40:0

32.0

2~.O

Chemical shift [ppm 1

Fig. 4a, b. 13C MRS. a Natural abundance 13C MR spectrum ofthe chest of a volunteer. The splitting due to proton coupling is recognized for CH 2 and CH3groUPS [9]. b 13C MR spectrum of enriched methanol: lower spectrum with and upper spectrum without proton decoupling [9]

64 s. The lines of N-acetyl-aspartate and lactate show up very nicely, and even choline and creatine can be observed. 1H MR spectroscopy will be used in the near future for clinical studies. Besides the low natural abundance of 1.1 % 13C spectroscopy has an additional complication. Carbon and hydrogen nuclei are coupled. A symmetrical splitting of the lines is caused by this coupling, and a more complex spectrum is the result. As shown in Fig. 4a, in a 13C spectrum of the volunteer's chest obtained by Starewicz et al.

[9], the symmetrical splitting is very well recognized in the region of the CH 2 group. This splitting can be circumvented by installing additional hardware, a second transmitter channel to decouple hydrogen and carbon nuclei is necessary. In Fig. 4 b such a decoupled spectrum of a phantom filled with 13C-enriched methylene is displayed at the right. The decoupled spectrum has only one line, whereas the coupled spectrum at left with four lines is more complex. In-vivo 13C MRS using whole-body scanners was pioneered by Shulman and co-workers [10, 11]. It seems that 13C MRS is

In Vivo Magnetic Resonance Spectroscopy

only useful for clinical work if enriched drugs are utilized as well. The main drawbacks are the enormous costs of the isotope-enriched 13C compounds and the complications due to labelling. Furthermore, the non-invasive character of the MRS examination is lost. The easiest accessible nucleus for in-vivo MRS is 31 P. It is sufficiently sensitive and has a large chemical shift. Furthermore, the spectrum is not as complicated as in 13C MRS. In-vivo 31p MRS was therefore used for the first clinical studies. Radda and co-workers and Chance and coworkers started with small bore magnets. Later on they used whole-body scanners for the study of muscle diseases and enzyme defects [12, 13]. Because at present most clinical work is done with 31p MRS, we shall take a closer look at it. The phosphorus compounds observed in in-vivo 31 P MRS are mainly the high-energy phosphates adenosine 5' -triphosphate (ATP) and phosphocreatine (PCr). Most ATP is produced in the mitochondrium by oxidative phosphorylation. The energy is stored in the form of phosphocreatine. In other words, phosphocreatine is the battery of the cell. ATP is used for biosynthesis, transport and muscle contraction and disintegrates in inorganic phosphate (Pi) and adenosine-5' -diphosphate (ADP). The latter metabolites may be used in the production cycle again. The observable phosphometabolites and their corresponding lines are shown in Figs. 5 a, 5 b. Inorganic phosphate and phosphocreatine have only one phosphorus atom, but in different chemical environments; therefore, the lines of both compounds show up at different positions in the frequency spectrum, as displayed in Fig. 5 b in the second and third spectrum. Due to the three phosphorus atoms of ATP, three different lines appear, and, in addition, the spin coupling of the three phosphorus nuclei in this molecule causes all three lines to split. This additional splitting of the ATP resonances will not be discussed in this context. Of course, all the metabolites are observed simultaneously in in-vivo examinations, resulting in a sum MR spectrum as shown at the bottom of Fig. 5 b. A 31p MR spectrum observed in tumours is shown in Figs. 6 a and 6 b. The resonances due to the high-energy phosphorus metabolites PCr and NTP are marked in Fig. 6 b. The three resonances labelled NTP contain not only ATP, but also other nucleoside 5' -triphosphates, such as uridine 5'triphosphate ect., and the a- and y-resonances also contain nucleoside 5' -diphosphates.

161

Downfield a broad resonance shows up. This resonance is composed of several constituents and is labelled phosphomonoester (PME) (Fig. 6b). Some constituents originate from compounds of the first step of glycolysis, for instance glucose 6' -phosphate (SP). More relevant, expecially in tumour spectra, are the phosphomonoesters phosphorylcholine (PC) and phosphorylethanolamine (PE; ct. Fig. 15 a). These compounds are intermediates of phospholipid membrane synthesis. Phosphodiesters (POE) show up between Pi and PCr resonance. They are composed of phospholipids, mainly glycerophosphorylcholine (GPC) and glycerophosphorylethanolamine (GPE), as indicated in Fig. 6b (see also Fig. 15a). These compounds are products of membrane catabolism [7]. The pH value of the tissue can be calculated from the chemical shift of the Pi resonance with respect to the PCr resonance, which serves as an internal reference [3, 4]. In summary, with 31 P MRS we can observe the energy metabolism of cells, and we obtain information about the membrane turnover and the pH value of tissue. The main limitation of MRS is the maximal length of time the patient can stay in the MR scanner. With exceptions, this has to be less than 60 min. Therefore, easy-to-handle hardware, short preparation times, high reproducibility, and short measurement times are perequisites for successful examinations. Last but not least, high spatial selectivity is demanded. Different localization methods have been suggested and implemented on whole-body scanners. Only the methods used in clinical trials are briefly mentioned (application of surface coils; depth-resolved surface coil spectroscopy DRESS; fast rotating gradient spectroscopy, FROGS; and image-selected in vivo spectroscopy, ISIS). The simplest method of achieving spatial resolution is the application of surface coils. The principle is shown in Fig. 7 a. A simple FlO sequence yields high signal-to-noise-ratio SNR. However, only superficial volumes of interest can be investigated because the signal dies out in a manner outlined in the depth profile, and therefore, signal contribution from surface layers to the spectra are usually strong. This contamination can be reduced when specially tailored sequences are used [14]. The detectable volume depends on the size of the coil. Localization by surface coils is sufficient when general muscle diseases or the time courses of treatment of superficial tumours

162

W. Semmler

b

a

* ATP (adenosine triphosphate) Adenine

I

~

~

~

I

I



Ribose -0 - P - 0 - P - 0 - p -

a-



I



ATP

* Inorganic phosphate

J. _ _ __

o

II 0·- P - OH I

_1_

Pi



* Phosphocreatine o

a--

Nt!

P

II 112 P- N - C -N-CH-", I I I 2 aO· H CH:J

* Sum spectrum

PCr

Sum

Fig. 5a, b. Scheme of a typical 31 P MR spectrum. a Formula ofthe main phosphorus metabolies ATP, Pj, and PCrobserved in MR spectra; whereas ATP has three phosphorus atoms and three corresponding resonances, PCr and Pi have one phosphorus atom each, but with different electronic environment due to the different chemical bonds. b Calculated spectra of the ATP, Ph and PCr. The spectrum at the bottom is the sum of the three spectra

31p

a

oxidative phosphorylation phosphorylated glycolytic intennediates

Phosphomonoesters (PME)

ATP-AOP + ~ PCr

PCr

~

Pi + energy

NTP (ATP) i

P

i

a

b phosphorylcholine (PC)

glycerophospho rylcholine (GPC)

phosphorylethanolamine (PE)

glycerophospho rylethanolamine (GPE)

Phosphomonoesters Phosphodiesters (PME) (POE)

--------

Fig. 6a, b. 31p MR spectrum of a tumour. The shaded areas indicate relevant resonances. Light shading indicates resonances which are exhibited in both high energy and membrane metabolism. a Resonances corresponding to the high energy metabolism are the nucleoside 5'-triphosphates; and PCr, phosphocreatinine; resonances. Ph inorganic phosphate. b Resonances corresponding to the membrane metabolism. PME, phosphomonoesters; POE, phosphodiesters

In Vivo Magnetic Resonance Spectroscopy

are investigated, as shown by Radda and coworkers [12] and our group [15-18], respectively. Improvement in spatial resolution can be gained when in combination with surface coils a sliceselective sequence (see Fig. 7 b) is applied, as suggested by Bottomley and co-workers (DRESS sequence) [19]. Price of higher spatial resolution is deterioration of spectral quality, which is mainly due to eddy currents produced by the gradient currents. Bottomley et al. used this method in combination with ECG triggering to monitor heart infarction by 31p MRS [20]. FROGS, developed by Sauter et al. [21], is just the inversion of the DRESS method. A series of RF pulses in the presence of a slice-selection gradient saturate the magnetization of the selected surface layer and , hence, when an RF pulse is applied, nuclei of this layer cannot contribute to the signal (see Fig. 7 c). Compared with the DRESS sequence, field gradients are turned on only before the measurement-RF pulse is applied and therefore distortion of the spectra is less significant. However, localization may not be as good. ISIS is a method with high localization accuracy; however, the fact that it is highly complex is a disadvantage [22]. Eight consecutive experiments have to be carried out to get a localized

a

163

spectrum: In a first experiment a non-selective 900 pulse is applied and all nuclei in the volume are excited and contribute to the FID signal. In a second experiment, a selected slice is excited at first by a 1800 pulse and immediately after this pulse a non-selective 900 pulse is again applied to the entire volume. With respect to all other nuclei, the nuclei in the selected slice now have opposite phases. All nuclei from outside the selected slice contribute with the same signal as in the first experiment. However, nuclei within the slice contribute with a signal of an opposite phase. Subtraction of the signals of both experiments results in a cancellation of the signal of the nuclei outside the slice, whereas the signals of the nuclei inside the slice add up. This method can easily extended to three dimensions, and it is possible to select a cube. The main drawbacks of this method are signal distortion caused by eddy currents, resulting in phase distortions, and the fact that eight experiments with identical conditions have to be performed, which can be difficult in routine clinical work. One advantage of the method is the possibility that with more complex pulse sequences an additional reference volume can be measured simultaneously. Furthermore, with ISIS volume can be selected with ease directly in the image [23, 24]. The SNR is not as good as in surface coil experiments, however, the good localization

b

c

Locali zation :

Depth profile :

~!

slice

I~~~~~ Selected slice

h

Fig. 7 a-c. Scheme of surface coil localization techniques. a Application of surface coils without any magnetic field gradient. The depth profile is lined out. A simple RF pulse is used which can be tailored to improve the depth resolution. b DRESS localization. The depth profile and the selected slice are outlined. A slice-selective gradient pulse is used to select the slice of interest parallel to the surface coil. c FROGS localization. The depth profile and the selected slice adjacent to the surface are outlined. The magnetization in the selected slice parallel to the coil is destroyed by spoiler pulses while the magnetic field gradient is applied. After switching off the gradient, a simple 90° pulse is applied to the entire volume and the FlO is detected

164

W. Semmler

in an important single area of medicine". Furthermore, he states, "if MRS does not fulfill these criteria, MRS will not become a routine examination in the daily clinical routine" [26]. One feasable clinical application of 31p MRS is tumour therapy control. The monitoring of tumour response to therapy is a routine oncological procedure and is predominantly performed by the imaging modalities. Tumours size has served as the main indicator of tumour response, however, in most of the patients a reduction of the tumour size cannot be observed within the first week or two after onset of therapy. Changes of relaxation times in MR images are also used to monitor the tumour response. The time range in which changes of tumour size and relaxation times occur is demonstrated by Just and co-workers [27]. These authors show that in a patient with a Ewing sarcoma of the distal femur the reduction of tumour size is obvious about a month after onset of therapy. The same holds for the T2 relaxation times. The increase of T2 relaxation time is observed the second month after onset of therapy, whereas all other MR imaging parameters are constant within the errors (Fig. 8).

allows spectra acquisition without contamination of other tissues.

Clinical Applications Clinically relevant applications of 31 P MRS are still rare. In the past, most effort was directed to biochemical studies, and muscle spectroscopy was the major interest [12, 13, 25]. These MRS studies had a great impact on medicine because of their contribution to the basic understanding of pathophysiology and normal physiology. However, these studies are not routine clinical examinations. As opposed to the sophisticated experimental arrangements and evaluation of data used in biochemistry and in-vivo animal experiments, the demands placed on MRS by the clinician are very simple and completely different from those in biochemical research. As formulated by Peter J. Bore the clinical application of MRS requires that MRS "either find multiple application in a wide variety of conditions in the way that X-ray studies have, or there must emerge a major role for MRS

a

2000

b

Tumor

Muscle

1500

1000 500

(ms)

800

T2

~

500 400

100

200

50

to

to

(ms)

~

0.75 0.5

Rho

0.5

0.25

t

2345578 Length of therapy (months)

t

2345578 Length of therapy (months)

Fig. 8a, b. Tissue parameters (T1, T2, and Q) extracted from MR images as a function of time (in months) after onset of therapy in the a tumour and b in the muscle of a 14-year-old patient with a Ewing sarcoma of the distal femur [27]

In Vivo Magnetic Resonance Spectroscopy

165

Ross and co-workers compared surface coil localization and the chemical shift imaging (CSI) method in patient monitoring after chemotherapy [28]. They compared whole-volume spectroscopy with surface coils and one-dimensional CSI. Spectra of the whole volume are on the top, those localized at the bottom. The PCr is significantly reduced in the CSI spectra, and the increase and decrease of PME is seen very clearly. Even though, muscle contamination is observed in the whole-volume spectra, the increase and decrease of PME is observed as well (Fig. 9). In localized

The succes of therapy - that is to differentiate responders and non-responders - can be assessed only very late with imaging modalities. If MRS is apt to qualify early effects, the resulting time advantage of weeks would provide more effective tumour therapy, and an optimization of therapy may be possible. Up to now most clinical MRS investigations were performed using surface coils, because it is a simple technique and a high SNR is achieved. As long as no absolute quantitation is necessary and only relative measurements are necessary, and as long as T1 relaxation times and the size of tumour do not change significantly, surface coils are sufficient. In a recently published study, Brian

c

b

a

PCr

PCr

ATP

e PME Before trealmenl

Pi

PCr

d

a

ATP

PME

Pi

"2

CI

iii

Chemical shift [ppm]

Fig. 9a-f. Sequential examinations during response of B-cell lymphoma. Whole-volume (top) and localized (bottom) spectroscopy were performed on extremity before, and 3 and 9 days after start of therapy [28]. The spectra (a-c) are obtained using a surface coil alone, the spectra (d-f) at the bottom are obtained with the CSI localization technique

166

W. Semmler

spectroscopy lower SNR and additional phase distortion, which are clearly demonstrated in the spectra (d-f) at the bottom of Fig. 9 are the price. These may hamper the quantitative evaluation of such spectra. Selection of patients with superficial tumours and special optimized surface coils [29] and

b IV

II

III

VII

sequences [14] allow good localization of tumour tissue with only little contamination from overlying tissue. This is demonstrated in a 46-year-old patient with a lymph node metastases of a squamous cell carcinoma. The image in Fig. 10a shows no muscle tissue on top of the tumour, and in the 31 P-spectrum (Fig. 10 b) per is very much reduced. The good spectra quality allows a quantification of the data. The high-energy phosphate peaks per, NTP and the inorganic phosphate Pi were fitted by a least-square fit. The smooth solid line in the spectra represents the fit, the histogramme displays the measured data [30]. This example shows that under the right conditions, surface coils are sufficient to monitor therapy in patients, especially since only relative measurements during therapy are required. It is wise to differentiate between long-term and short-term follow-up examinations. Short term follow-up examinations are performed within the first hours after the start of therapy when the patient is positioned in the scanner. This kind of examination allows monitoring the immediate changes of metabolism during infusion of therapeutic drugs, whereas long-term follow-up examinations last several days and sometimes up to weeks. In this type of studies, the changes in cell populations are observed, that is, the necrotic portion of cells may increase during therapy and the aerobic portion decrease.

VI

Short-Term Follow-up Examinations A 6-year-old patient with a Ewing sarcoma of the

i

10

-5 6[ppml

I

-10

I

-15

-20

Fig. 10a, b. Lymph node metastasis of an unknown primary squamous cell carcinoma in a 46-year-old patient [15]. a Transversal spin-echo image (SE(1200/35)) of the neck at the level of the larynx. The position of the surface coil is indicated. Arrow, center of coil; bar, extension of coil. b 31 P MR spectrum of the lymph node metastasis. The smooth line represents a least-square fit to the data, which are displayed as a histogramme. PME and POE resonances are not included in the fits. The assignment of the resonances are the following: I, ~-nucleoside 5' -triphosphate (~-NTP); II, a-nucleoside 5' -triphosphate (a-NTP), a-nucleoside diphosphate (a-NOP), and nicotinamide adenine dinucleotide; III, y-nucleoside 5' -triphosphate (y-NTP) and ~-nucleoside diphosphate (~-NOP); IV, phosphocreatine (PCr); V, phosphodiester (POE); VI, inorganic phosphate (Pi); VII, phosphomonoester (PME)

scapula serves as an example for short-term follow-up [16]. No change in the size and the relaxation behaviour was observed during the 1-h short-term fOllow-up. However, directly after onset of the infusion, in 31 P MRS, a change in spectral parameters is observed as shown in Fig. 11 b. A slight drop of the PME resonance integrals is seen a few minutes after the start of infusion. Furthermore, changes in the ratio of Per/Pi occur. The experience we have from patients examined during infusion of therapeutic drugs can be summarized as follows: (a) changes of spectral parameters occur: PME, PDE and in the ratio of Per/Pi; (b) no significant change in pH values is observed. Whether the spectral changes are due to immediate therapeutic response of the cells or only reflect physiological reactions is still an open question. Further investigations have to be carried out.

In Vivo Magnetic Resonance Spectroscopy

167

a Short term

40 >-

:!::

. .s:

30

VI

c

c c

.2' VI

b

PCrl Pi

-.

e_.

7.4

.

/.--e_e--e_e__ _ _ e

VI C

.s:

pH·

7.2

7.0

10

c

c

.2' VI

PME

PCrl Pi

::i

5

5 ci ~

>-

4

:!:: VI

-

4 ~

3

.s:

3

2

g

Cl VI

o

20

time (minl

40

Fig. 11 a, b. Short-term follow-up in a 5-year-old patient with a Ewing sarcoma of the scapula [16]. a Series of 31p MR spectra obtained before and during infusion of the chemotherapeutic drugs. b Spectral parameters and protocol of the examination

o

20

40

time (minl control 1 2 _ 3 4 _ 5 6_ 7 8 MRS # _

_9 min

Chemotherapy

~

.~I=;~;:::::~;::J

Vincristin (Botus)

Adriamycin (Infusion)

168

W. Semmler

Long-Term Follow-up Examinations

Long-term studies were performed by several groups [15-18, 28, 31-33]. B. Ross and coworkers [28] monitored chemotherapy in a patient with an osteosarcoma in an extremity. The spectra of this investigation are shown in Fig. 12. Examination always started 3 days after administration of chemotherapeutic drugs. In the course of the treatment, within months they observe a decrease in phosphodiesters and nudeosid 5'triphosphate. In a patient with an osteosarcoma we observed similar changes of spectral parameters [15]. Our examinations as well as Ross' demonstrate that changes in tumour metabolism after administration of chemotherapeutic drugs can be detected and monitored by 31 P MRS. Similar observations were made by Ng et al. [33], who observed changes in the POE!ATP ratio after radiation therapy in a patient with a lymphoma (Fig. 13a, b).

Early detection of tumour response to therapy within the time interval of 1 to 3 days is possible, as can be demonstrated with a 58-year-old patient [18]. The spectra of the long-term followup are shown in Fig. 14a; these spectra show elevated PME and POE resonances. PCr intensities are high and may be in part due to muscle contaminations. The spectral parameters are displayed (Fig. 14 b). PCr increases and Pi is constant. POE and PME have initially a steep increase and drop later. The pH value is more or less constant. The tumour diameter was constant during the first 3 days and dropped later on. This example shows that MRS is capable of monitoring metabolic changes after chemotherapy as early as 2 days, or even 1. At present, it is not yet clear whether 31 P MRS can differentiate responders and non-responders. The number of patients in all studies is too small, in particular the patient collectives are not homogeneous with respect to histology and treatment.

6/12

17/11

6111

Control

24/10

After 3 d 17/10

After 3 d 9110

Arter 3 d

Berore treatment

Chemical shift [ppm)

Fig. 12. Therapy monitoring of an osteosarcoma during chemotherapy. 31 P MR spectra were obtained from extremity before treatment (lowest), and 8, 15,28,39, and 57 days after commencment of chemotherapy [28]. Spectra of the control extremity is shown for comparison

In Vivo Magnetic Resonance Spectroscopy

a

169

b

11th irrad. 3-10-86

( F)

2.5

250

2.0

200

(E) 8th irrgd. 3-7-86

.PDE/ATP ~ Tumor mass

7th irrad. 3-6-86

(O)

1.5

-

150

>-

III

E

4th irrad. 3-3-86

c

!c

-0c

Cl

1.0

100

0.5

50

Cl

VI

3 rd irrad. 2-28-86

L{A) I

32

I

24

I

I

1

r

r

16 8 0 -8 -16 Chemical shift [ppm]

r

2nd irrad. 2-27- 86

I

-24 -32

0

I

8 10 12 14 16 1820 Fractionated radiation dose [Gy]

Fig. 13a, b. Longt-term follow-up study of an non-Hodgkin's lymphoma during BOCO therapy [33]. a Series of 31 P MR spectra before and during radiation therapy. b Comparison of POE!ATP ratio (circles) and tumour mass (triangles) of a non-Hodgkin's lymphoma subjected to fractionated BOCo-irradiation

The preliminary results of the long-term follow-up studies are the following: (a) MRS is capable of monitoring remission and progression of tumours; (b) the influence of chemotherapy on tumour metabolism is detectable within 1-2 days;

(c) differences in tumour metabolism after chemotherapy in responders and non-responders are not proven without a doubt by 31p MRS. The results of the study - especially those of the PME and POE time course - would be better if the resonances could be clearly resolved or if the constituents of the PME and POE lines were known. Comparison of in-vivo investigations with in-vitro measurements of excised tumours can elucidate this problem. The in-vivo tumour spectrum of a malignant melanoma is displayed at the bottom, the high resolution spectrum at the top of Fig. 15 a. Most of the short-life high energy phosphates are already disintegrated, however, the PME and POE peaks are well resolved. And some constituents like phosphorylcholine (PC),

glycerophosphorylcholine (GPC) and glycerophosphorylethanolamine (GPE) can be assigned; uridindiphosphat-glucose (UOPG) and diphosphodinucelotide (OPON) are observed as well.

The disintegration of high energy phosphates is

due to the way the tumour was excised. Another way of obtaining more information about the PME and POE region is to improve the resolution by increasing the magnetic field strength up to 4.0 Tesla. In Fig. 15 b the tumour spectra of a histiocytoma in the groin at 4.0 Tesla and imager at 1.5 Tesla are displayed (Measurements were performed in cooperation with O. Hentschel, Siemens, Erlangen, FRG). The spectrum at 1.5 Tesla was obtained with a 5-cm surface coil and that at 4.0 Tesla obtained with a 10-cm coil. The 10-cm coil has a larger field of view, and therefore muscle contamination is higher than in the upper spectrum. In the 4.0 Tesla spectrum the splitting of PME and POE peaks is observed. The same holds for Pi in which

170

W. Semmler

Long term

a

10

>-

-. :!: III C

.S

c c .2' If)

5

10

b a.u. 4 3

o

5

-5

15 [ppml

-15

-10

A, ----- _--~

, .,

\

b----Pcrl Pi

6 2

pH

pcr

~P-NTP

1C.r

Pi--------__ - :

~-------. Transverse diameter

7.3

7.2 4 cm

3

2

• o

7

control jt Follow up MRS MRI

Therapy

0

14d I 23

7

0

4

14d (5)

t t

I~

2!

days

c:::::::: Il

Fig. 14a, b. Long-term follow-up in a patient with lymph node metastasis in the neck [18]. a Series of 31p MR spectra before and during chemotherapy. b Spectral parameters and tumour diameter as a function of time after commencement of chemotherapeutic drugs. The increase of the PME and PDE intensities the first two days after onset of therapy is clearly observed

In Vivo Magnetic Resonance Spectroscopy

a

PCr

NAD NAOH

In vivo 1.5 Testa

p----NTPa-------~ ~----NOPa

POE

171

b 3lP-spectrum of a tumor in the groin 1.5 Testa SFC

~5cm

j

PME

d

o C

01

Vi +5

0

GPE GPC

PCr PE PS

Ex vivo 11.7 Tesla (0.32 G.30Min) 4°C

4.0 Tesla SFC ~'Ocm j

d

....'iii>.

NTP

c

I

(\j

~ PME

o

4

o

-4

ppm

8

10

Pi

POE

1

P

o

Chemical shift [ppm]

-10

Fig. 15a, b. Comparison of in vivo (top) and in vitro (bottom) 31p MRS of a malignant melanoma. The PME and POE resonances are resolved in the in-vitro 31 P MR spectrum. The depletion of the high energy phosphates are due to the way the tumour was excised (in-vitro 31p MR spectrum: W. Hull, unpublished data). b Comparison of in-vivo 31p MR spectra of a fibrotic histiocytoma in the groin 1.5 Tesla (top) and 4.0 Tesla (bottom). The PME and POE resonances are not completely resolved in the 4.0-Tesla 31p MR spectrum. The measurement at 4.0 Tesla was made using a 10-cm surface coil, muscle contamination is therefore stronger, resulting in a more intense PCr resonance (4.0-Tesla spectra, O. Hentschel et aI., unpublished data)

the splitting is due to intra- and extracellular pH value. Quantification of PME and POE intensities is much easier, and more reliable data can be obtained at 4.0 Tesla. This would help to improve tumour monitoring.

Conclusion Specifically speaking, immediate effects of chemotherapeutic drugs on the tumour metabolism can be observed. It is yet not clear whether these

are therapeutic effects. Tumour response to therapy can be observed earlier than with conventional methods. Discriminating between responders and non-responders may be possible, however, the number of patients is too low and the changes in the spectral parameter are only moderate, so that this statement has to be proved in a larger patient group. In general 31p MRS may be a suitable tool for monitoring tumour therapy.

172

W. Semmler

References 1. Sostman HD, Armitage 1M, Fischer JJ (1984) NMR in cancer I. High resolution spectroscopy of tumors. Magn Reson Imag 2:265-278 2. Evanochko wr, Ng TC, Lilly MB, Lawson AJ, Corbett TH, Durant JR, Glickson JD (1983) In vivo 31p NMR study of the metabolism of murine mammary 16/C adenocarcinoma and its response to chemotherapy, x-radiation, and hyperthermia. Proc Natl Acad Sci 80:334-338 3. Naruse S, Horikawa Y, Tanaka C, Higuchi T, Ueda S, Hirakawa K, Nishikawa H, Watari H (1985) Observations of energy metabolism in neuroectodermal tumors using in vivo 31P-NMR. Magn Res Imag 3:117-123 4. Ng TC, Evanochko wr, Hiramoto RN, Ghanta VK, Lilly MB, Lawson AJ, Corbett TH, Durant JR, Glickson JD (1982) 31p NMR spectroscopy of in vivo tumors J Magn Reson 49:271-286 5. Sijens PE, Bovee WMMJ, Seijkens D, Los G, Rutgers DH (1986) In vivo 31P-nuclear magnetic resonance study of the response of a murine mammary tumor to different dose of y-radiation. Cancer Res 46:1427-1432 6. Barany M, Glonek T (1984) Identification of diseased states by phosphorus-31 NMR. In: Gorenstein DG (ed) Phosphorus-31 NMR, principles and applications. Academic, Orlando 7. Evanochko wr, Sakai n, Ng TC, Rama Krishna N, Kim HD, Zeidler RB, Ghanta VK, Brockman RW, Schiffer LM, Braunschweiger PG, Glickson JD (1984) NMR study of in vivo RIF-1 tumors. Analysis of perchloric acid extracts and identification of 1H, 31p and 13C resonances. Biochim Biophys Acta 805:104-116 8. Luyten PR, den Hollander JA, Segebarth C, Baleriaux D (1988) Localized 1H NMR spectroscopy and spectroscopic imaging of human brain tumors in situ. 7th annual meeting of the Society of Magnetic Resonance in Medicine (abstracts). 20-26 August 1988, San Francisco, 1:252 9. Starewicz PM, Albright MJ (1986) Adaptation of a 1-meter bore magnetic resonance imaging system to in vivo decoupled carbon-13 spectroscopy. Health Care Instrumentation 1:212-215 10. Shulman RG (1983) NMR spectroscopy of living cells. Sci Amer 248:86 11. Jue T, Rothman DL, Tavitian BA, Shulman RG (1988) 13C NMR hepatic glycogen repletion study in man. 7th annual meeting of the Society of Magnetic Resonance in Medicine (abstracts). 20-26 August 1988, San Francisco, 1:248 12. Radda GK (1985) Control of bioenergetics: from cell to man by phosphorus nuclear-mag netic-resonance spectroscopy (18th CIBA medal lecture). Biochem Soc Trans 14:517-525 13. Chance B, Leigh JS, Kent J, McCully K, Nioka S, Clark BJ, Maris JM, Graham T (1986) Multiple control of oxidative metabolism in living tissue as studied by phosphorus magnetic resonance. Proc Natl Acad Sci 83:9458-9462

14. Bachert-Baumann P, Neurohr KJ, Gademann G, Semmler W, Zabel H-J, Lorenz W (1987) Klinische 31P-MR-Spektroskopie am Ganzkerpertomographen: Optimierung der MeBparameter und Lokalisierung. Zentralbl Radiol 134:237 15. Semmler W, Gademann G, Bachert-Baumann P, Zabel H-J, Lorenz W-J, van Kaick G (1988) Monitoring human tumor response to therapy by means of 31 P-MR spectroscopy. Radiology 166:533-539 16. SemmlerW, Gademann G, Bachert-Baumann P, et al. (1988) In vivo 31 Phosphor-Spektroskopie von Tumoren: pra-, intra- und posttherapeutisch. ROFO 149: 369-377 17. Semmler W, Gademann G, Schlag P, BachertBaumann P, et al. (1988) Impact of hyperthermic regional perfusion therapy on cell metabolism of malignant melanoma monitored by 31p MR spectroscopy. Magn Res Imag 6:335-340 18. Semmler W, Gademann G, Bachert-Baumann P, Zabel H-J, Lorenz W-J, van Kaick G (1988) Therapieverlaufskontrolle nach Chemotherapie an Tumoren des Menschen mit Hilfe der in-vivo 31 P-Spektroskopie. In: Schneider GH, Vogler E (eds) Digitale bildgebende Verfahren. Interventionale Verfahren. Integrierte digitale Radiologie. Springer, Berlin Heidelberg New York, pp 15-23 19. Bottomley PA, Foster TH, Darrow RD (1984) Depth resolved surface-coil spectroscopy. J Magn Reson 59:338-342 20. Bottomley PA, Herfkens RJ, Smith LS, Bashore TM (1987) Altered phosphate metabolism in myocardial infarction: P-31 MR spectroscopy. Radiology 165:703-707 21. Sauter R, Muller S, Weber H (1987) Localization in in vivo 31p NMR spectroscopy by combining surface coils and slice-selective saturation. J Magn Reson 75:167-173 22. Ordidge RJ, Conelly A, Lohmann JAB (1986) Imageselected in vivo spectroscopy (ISIS). A new technique for specially selective NMR spectroscopy. J Magn Reson 66:283-294 23. Heindel W, Bunke J, Steinbrich W (1987) Bildgesteuerte lokalisierte 31P-NMR-Spektroskopie des mensch lichen Gehirns bei 1,5 Tesla. ROFO 147:374-378 24. Marson GB, Twieg DB, Karczmar GS, Lawry TJ, Gober JR, Valenza M, Boska MD, Weiner MW (1988) Application of image-guided surface coil P-31 MR spectroscopy to human liver, heart, and kidney. Radiology 169:541-547 25. Itoh M, lio M, Kawai M, Takizawa 0, Yoshikawa K, Minausi M, Otitmo K, Aoki S, Niida T, Watanabe T (1986) 31P-NMR spectroscopy of myopathies: clinical application of whole-body MR. Rad Med 4:41-45 26. Bore PJ (1985) The role of magnetic resonance spectroscopy in clinical medicine. Magn Reson Imag 3:407-413 27. Just M, Gutjahr P, Higer HP, Sterkel S, Rudigier J, Ritter G, Pfannenstiel P (1987) Meglichkeiten der MR Tomographie in der Therapiekontrolle maligner Knochentumoren. ROFO 147:413-419 28. Ross B, Helsper JT, Cox IJ, Young IR, Kempf R, Makepeace A, Pennock J (1987) Osteosarcoma and

In Vivo Magnetic Resonance Spectroscopy other neoplasms of bone. Magnetic resonance spectroscopy to monitor therapy. Arch Surg 122: 1464-1469 29. Zabel H-J, Bader R, Gehrig J, Lorenz WJ (1987) High-quality MR imaging using flexible transmission line resonators. Radiology 165:857-859 30. Semmler W, Menningen M, Bachert-Baumann P, Gademann G, van Kaick G (1987) Quantitative evaluation of in-vivo 31 P spectra using a least square fit procedure. 6th annual meeting of the Society of Magnetic Resonance in Medicine (abstracts). 17-21 August 1987, New York, p 505 31. Northrop J, Evans A, Doulon E, Haris J, Chance B (1986) Stage IV neuroblastoma response to chemo-

173

therapy measured by 31p MRS in situ. 5th annual meeting of the Society of Magnetic Resonance in Medicine (abstracts). 19-22 August 1986, Montreal, 1:31 32. Nidecker AC, Muller S, Aue WP, Seelig J, Fridrich R, Remagen W, Hartweg H, Benz, UF (1985) Extremity bone tumors: Evaluation by P-31 MR spectroscopy. Radiology 157:167-174 33. Ng TC, Vijayakumar S, Majors AW, Thomas FJ, Meaney TF, Baldwin NJ (1987) Response of a NonHodgkin Lymphoma to 60CO therapy monitored by 31p MRS in situ. (Int J Radiat Oncol Bioi Phys) 13:1545-1551

U. Veronesi, Milan (Editor-in-ChieO B. Arnesjo, J. Bum, L. Denis, F. Mazzeo (Co-Editors)

Surgical Oncology A European Handbook Foreword by J.Bum 1989. XVIII, 999 pp. 222 figs. 227 tabs. Hardcover ISBN 3-540-17770-1

Springer-Verlag Berlin Heidelberg New York London Paris Tokyo Hong Kong

The European Society of Surgical Oncology and SpringerVerlag, Heidelberg, are pleased to announce the publication of an interesting and exciting new textbook for all those involved in oncology and related specialties: Surgical Oncology - A European Handbook. The first of its kind in Europe, this handbook seeks to cover most of the aspects of surgical oncology, which with the recent moves toward a multi-disciplinary approach in research and treatment, every surgeon needs to know. This handbook should be the essential feature of every surgeon's library. Dealing as it does with the many facets of the disease from biology of cancer, detection and diagnosis, general concepts, emergencies, rehabilitation and the planning and evaluation of therapy to the treatment of cancer of the different organs, it is a basic guide to the surgical approach to neoplastic disease. Contributions are from the most prestigious names in the European oncological world and will surely be a must in years to come - an invaluable reference for all those who have a specific interest in the field of oncology and the many and varied problems related to patient care and treatment. Distribution rights for Japan: Maruzen Company, Tokyo

•••••• •• • •• •••••• ...

..

F. Cavalli, Bellinzona (Ed.)

ESO Monographs

Endocrine Therapy of Breast Cancer /II

Series Editor: U. Veronesi

1989. VII, 65 pp. 26 figs. 7 tabs. Hardcover ISBN 3-540-50819-8

~ ~

• • • 0naJk)gJ

A.B.Miller, University of Toronto (Ed.)

Diet and the Aetiology of Cancer 1989. VII, 73 pp. 2 figs. Hardcover ISBN 3-540-50681-0

L. Domellof, Orebro (Ed.)

Drug Delivery in Cancer Treatment /I Symptom Control, Cytokines, Chemotherapy

1989. VII, 107 pp. 31 figs. Hardcover ISBN 3-540-51055-9 L. Domellof, Orebro (Ed.)

Drug Delivery in Cancer Treatment 1987. VII, 99 pp. Hardcover ISBN 3-540-18459-7 L.Denis, Antwerp (Ed.)

The Medical Management of Prostate Cancer 1988. IX, 98 pp. 8 figs. Hardcover ISBN 3-540-18627-1 B. Winograd, University of Amsterdam; M. Peckham, London; H. M. Pinedo, University of Amsterdam (Eds.)

Human Tumour Xenografts in Anticancer Drug Development 1988. xv, 143 pp. 37 figs. Hardcover ISBN 3-540-18638-7 J.F. Smyth, Edinburgh (Ed.)

Interferons in Oncology Current Status and Future Directions

1987. VII, 70 pp. Hardcover ISBN 3-540-18019-2 F. Cavalli, Bellinzona (Ed.)

Springer-Verlag Berlin Heidelberg New York London Paris Tokyo Hong Kong

Endocrine Therapy of Breast Cancer Concepts and Strategies

1986. VII, 120 pp. Hardcover ISBN 3-540-16959-8