MR image-guided P-31 MR spectroscopy in the evaluation of brain tumor treatment

Zero-Field NMR J-Spectroscopy of Organophosphorus Compounds

Organophosphorus compounds are a wide and diverse class of chemicals playing a crucial role in living organisms. This aspect has been often investigated using nuclear magnetic resonance (NMR), which provides information about molecular structure and function. In this paper, we report the results of theoretical and experimental studies on basic organophosphorus compounds using zero-field NMR, where spin dynamics are investigated in the absence of a magnetic field with the dominant heteronuclear J-coupling. We demonstrate that the zero-field NMR enables distinguishing the chemicals owing to their unique electronic environment even though their spin systems have the same alphabetic designation. Such information can be obtained just in a single measurement, while amplitudes and widths of observed low-field NMR resonances enable the study of processes affecting spin dynamics. An excellent agreement between simulations and measurements of the spectra, particularly in the largest frequency J-couplings range ever reported in zero-field NMR, is demonstrated.

In Vivo 1H Spectroscopy of the Human Brain at 1.5 Tesla; Preliminary Experience at a Clinical Installation

In vivo localized water suppressed proton spectroscopy of human brain was carried out on 15 healthy volunteers and 2 patients suffering from a brain tumour and an infarction, respectively. The measurements were performed on a whole body MR system, operating at 1.5 tesla using the stimulated echo technique. Our preliminary results indicate that it is possible to detect a number of metabolites in the brain within a total measurement time of one hour. The dominant peaks in the spectra from healthy volunteers are N-acetyl aspartate, choline and creatine/phosphocreatine. The spectra obtained from the brain tumour and the infarct, respectively, differed very much from those obtained in healthy brain tissue. Our preliminary results indicate that localized proton spectroscopy may contribute to non-invasive brain tumour classification and possibly also to the differentiation between tumours and infarcts in clinically doubtful cases.

Mr Spectroscopy in Clinical Research

MR spectroscopy (MRS) offers unique possibilities for non-invasive evaluation of biochemistry in vivo. During recent years there has been a growing body of evidence from clinical research studies on human beings using 31P and 1H MRS. The results indicate that it is possible to evaluate phosphorous energy metabolism, loss of neurones, and lactate production in a large number of brain diseases. Furthermore, 31P and 1H MRS may be particularly clinically useful in evaluation of various disorders in skeletal muscle. In the heart 31P MRS seems at the moment the most suitable for evaluation of global affections of the myocardium. In the liver 31P MRS appears to be rather insensitive and non-specific, but absolute quantification of metabolite concentrations and using metabolic “stress models” may prove useful in the future. The clinical role of MRS in oncology is still unclear, but it may be useful for noninvasive follow-up of treatment.

Taken together, the evidence obtained so far certainly shows some trends for clinical applications of MRS. Methods are now available for the clinical research necessary for establishing routine clinical MRS examinations.

Phospholipid metabolites in recurrent glioblastoma: in vivo markers detect different tumor phenotypes before and under antiangiogenic therapy

Purpose

Metabolic changes upon antiangiogenic therapy of recurrent glioblastomas (rGBMs) may provide new biomarkers for treatment efficacy. Since in vitro models showed that phospholipid membrane metabolism provides specific information on tumor growth we employed in-vivo MR-spectroscopic imaging (MRSI) of human rGBMs before and under bevacizumab (BVZ) to measure concentrations of phosphocholine (PCho), phosphoethanolamine (PEth), glycerophosphocholine (GPC), and glyceroethanolamine (GPE).

Methods

¹H and ³¹P MRSI was prospectively performed in 32 patients with rGBMs before and under BVZ therapy at 8 weeks intervals until tumor progression. Patients were dichotomized into subjects with long overall survival (OS) (>median OS) and short OS (<median OS) survival time from BVZ-onset. Metabolite concentrations from tumor tissue and their ratios were compared to contralateral normal-appearing tissue (control).

Results

Before BVZ, ¹H-detectable choline signals (total GPC and PCho) in rGBMs were elevated but significance failed after dichotomizing. For metabolite ratios obtained by ³¹P MRSI, the short-OS group showed higher PCho/GPC (p = 0.004) in rGBMs compared to control tissue before BVZ while PEth/GPE was elevated in rGBMs of both groups (long-OS p = 0.04; short-OS p = 0.003). Under BVZ, PCho/GPC and PEth/GPE in the tumor initially decreased (p = 0.04) but only PCho/GPC re-increased upon tumor progression (p = 0.02). Intriguingly, in normal-appearing tissue an initial PEth/GPE decrease (p = 0.047) was followed by an increase at the time of tumor progression (p = 0.031).

Conclusion

An elevated PCho/GPC ratio in the short-OS group suggests that it is a negative predictive marker for BVZ efficacy. These gliomas may represent a malignant phenotype even growing under anti-VEGF treatment. Elevated PEth/GPE may represent an in-vivo biomarker more sensitive to GBM infiltration than MRI.

31P magnetic resonance spectroscopic imaging with polarisation transfer of phosphomono- and diesters at 3 T in the human brain: relation with age and spatial differences

Tissue levels of the compounds phosphocholine (PC), phosphoethanolamine (PE), glycerophosphocholine (GPC) and glycerophosphoethanolamine (GPE) can be studied by in vivo 31P MRS. However, the detection of the signals of these compounds suffers from low sensitivity and contamination by underlying broad resonances of other phosphorylated compounds. Improved sensitivity without this contamination can be achieved with a method for optimal polarisation transfer of 1H to 31P spins in these molecules, called selective refocused insensitive nuclei-enhanced polarisation transfer (sRINEPT). The aim of this study was to implement a three-dimensional magnetic resonance spectroscopic imaging (MRSI) version of sRINEPT on a clinical 3 T magnetic resonance system to obtain spatially resolved relative levels of PC, PE, GPC and GPE in the human brain as a function of age, which could be used as a reference dataset for clinical applications. Good signal-to-noise ratios were obtained from voxels of 17 cm(3) of the parietal and occipital lobes of the brain within a clinically acceptable measurement time of 17 min. Eighteen healthy subjects of different ages (16-70 years) were examined with this method. A strong inverse relation of the PE/GPE and PC/GPC ratios with age was found. Spatial resolution was sufficient to detect differences in metabolite ratios between white and grey matter. Moreover, we showed the feasibility of this method for clinical use in a pilot study of patients with brain tumours. The sRINEPT MRSI technique enables the exploration of phospholipid metabolism in brain diseases with a better sensitivity than was possible with earlier 31P MRS methods.

A metabolomics perspective of human brain tumours

During the past decade or so, a wealth of information about metabolites in various human brain tumour preparations (cultured cells, tissue specimens, tumours in vivo) has been accumulated by global profiling tools. Such holistic approaches to cellular biochemistry have been termed metabolomics. Inherent and specific metabolic profiles of major brain tumour cell types, as determined by proton nuclear magnetic resonance spectroscopy ((1)H MRS), have also been used to define metabolite phenotypes in tumours in vivo. This minireview examines the recent advances in the field of human brain tumour metabolomics research, including advances in MRS and mass spectrometry technologies, and data analysis.

Proton magnetic resonance spectroscopy in immunocompetent patients with primary central nervous system lymphoma

BACKGROUND

Magnetic resonance spectroscopy imaging (MRSI) non-invasively evaluates the metabolic profile of normal and abnormal brain tissue. Primary central nervous system lymphoma (PCNSL) is a highly aggressive tumor responsive to high-dose methotrexate based regimens. Patients often have complete responses but relapses are common. We characterized the MR spectra of PCNSL patients, correlated MRSI with MRI and evaluated whether early recurrence could be detected by MRSI.

METHODS

Patients with PCNSL had multi-voxel MRSI before, during, and after treatment. The region of interest was defined using axial FLAIR images. Metabolites assessed were N-acetyl-aspartate (NAA), choline (Cho), creatine (Cr), lipid, and lactate. Ratios of Cho/Cr, NAA/Cho, and NAA/Cr were calculated and correlated with MRI. Overall survival (OS), progression free survival (PFS), and relative risks of each of the ratios were determined.

RESULTS

MRSI was performed on 11 men and seven women; median age of 59. Sixty-seven MRSI studies were performed, 17 baseline and 48 follow-up studies. Median ratios in 16 pretreated patients were Cho/Cr-1.90, NAA/Cho-0.39, and NAA/Cr-1.27. Two patients had lipid at baseline, five had lactate and two had both. MRSI correlated with tumor response or progression on MRI; in three patients MRSI suggested disease progression prior to changes on MRI. Univariate analysis of metabolite ratios, lipid, and lactate revealed that none significantly affected PFS or OS. Kaplan-Meier analysis of the presence or absence of lipid, lactate or both revealed a trend for increased PFS.

CONCLUSION

MRSI and MRI correlate with tumor response or progression and may allow early detection of disease recurrence. The presence or absence of lipid and/or lactate may have prognostic significance. Further research using MRSI needs to be done to validate our findings and determine the role of MRSI in PCNSL.

Prediction of treatment response of head and neck cancers with P-31 MR spectroscopy from pretreatment relative phosphomonoester levels

RATIONALE AND OBJECTIVES

Combinations of chemotherapy and fractionated radiation therapy are the currently preferred nonsurgical treatment methods for squamous cell carcinoma of the head and neck, but to the authors' knowledge there is no reliable marker for predicting therapeutic response. Early identification of nonresponders would allow prompt replacement of ineffective, toxic therapy by alternative, potentially more effective procedures. Frequent regional node involvement facilitates surface coil investigation with phosphorus-31 magnetic resonance spectroscopy.

MATERIALS AND METHODS

P-31 magnetic resonance spectra were acquired from 12 patients before radiation therapy or chemotherapy. In vivo three-dimensional localized P-31 nuclear magnetic resonance chemical shift imaging was performed with a 1.5-T clinical imager and a dual-tuned H-1/P-31 surface coil. Proton decoupling and nuclear Overhauser enhancement were used to improve sensitivity and resolve overlapping signals in the phosphomonoester region of the spectrum.

RESULTS

The average pretreatment ratio of phosphomonoester to beta-nucleoside triphosphate was significantly smaller in complete responders (n = 4) than in incomplete responders (partial responders plus nonresponders, n = 8) (0.0 +/- 0.0 vs 1.22 +/- 0.17 [P = .004]).

CONCLUSION

Results of this preliminary study suggest that H-1-decoupled P-31 magnetic resonance spectroscopy may prove to be a useful predictor of therapeutic response in head and neck cancers.

Three-dimensional magnetic resonance spectroscopic imaging of brain and prostate cancer

Clinical applications of magnetic resonance spectroscopic imaging (MRSI) for the study of brain and prostate cancer have expanded significantly over the past 10 years. Proton MRSI studies of the brain and prostate have demonstrated the feasibility of noninvasively assessing human cancers based on metabolite levels before and after therapy in a clinically reasonable amount of time. MRSI provides a unique biochemical "window" to study cellular metabolism noninvasively. MRSI studies have demonstrated dramatic spectral differences between normal brain tissue (low choline and high N-acetyl aspartate, NAA) and prostate (low choline and high citrate) compared to brain (low NAA, high choline) and prostate (low citrate, high choline) tumors. The presence of edema and necrosis in both the prostate and brain was reflected by a reduction of the intensity of all resonances due to reduced cell density. MRSI was able to discriminate necrosis (absence of all metabolites, except lipids and lactate) from viable normal tissue and cancer following therapy. The results of current MRSI studies also provide evidence that the magnitude of metabolic changes in regions of cancer before therapy as well as the magnitude and time course of metabolic changes after therapy can improve our understanding of cancer aggressiveness and mechanisms of therapeutic response. Clinically, combined MRI/MRSI has already demonstrated the potential for improved diagnosis, staging and treatment planning of brain and prostate cancer. Additionally, studies are under way to determine the accuracy of anatomic and metabolic parameters in providing an objective quantitative basis for assessing disease progression and response to therapy.

Tumour phospholipid metabolism

Following the impetus of early clinical and experimental investigations, in vivo and in vitro MRS studies of tumours pointed in the eighties to the possible significance of signals arising from phospholipid (PL) precursors and catabolites as novel biochemical indicators of in vivo tumour progression and response to therapy. In the present decade, MRS analyses of individual components contributing to the 31P PME (phosphomonoester) and PDE (phosphodiester) resonances, as well as to the 1H 'choline peak', have reinforced some of these expectations. Moreover, the absolute quantification of these signals provided the basis for addressing more specific (although still open) questions on the biochemical mechanisms responsible for the formation of intracellular pools of PL derivatives in tumours, under different conditions of cell proliferative status and/or malignancy level. This article is aimed at providing an overview on: (a) quantitative MRS measurements on the contents of phosphocholine (PCho), phosphoethanolamine (PEtn) and their glycerol derivatives ģlycerol 3-phosphocholine (GPC) and glycerol 3-phosphoethanolamine (GPE)[ in human tumours and cells (with particular attention to breast and brain cancer and lymphomas), as well as in normal mammalian tissues (including developing organs and rapidly proliferating tissues); (b) possible correlations of MRS parameters like PEtn/PCho and PCho/GPC ratios with in vitro cell growth status and/or cell tumorigenicity; and (c) current and new hypotheses on the role and interplay of biosynthetic and catabolic pathways of the choline and ethanolamine cycles in modulating the intracellular sizes of PCho and PEtn pools, either in response to mitogenic stimuli or in relation to malignant transformation.

In vivo proton magnetic resonance spectroscopy of diseased prostate: spectroscopic features of malignant versus benign pathology

In vivo Proton Magnetic Resonance Spectroscopy appears potentially useful for non-invasive discrimination between benign prostatic hyperplasia (BPH) and prostate carcinoma (PC). Aiming to delimit the range within which spectra from one or the other pathology should occur, and establish extreme spectroscopic features of malignant versus benign prostate disease, we performed endorectal proton MR spectroscopy on 20 patients severely affected of either benign prostatic hyperplasia (BPH) (n = 10) or prostate cancer (PC) (n = 10). They were selected on the basis of the large volume and homogeneity of their lesions, which were histologically confirmed after spectroscopy. Consequently, high-quality short-TE proton spectra with well-resolved metabolite signals, and practically free of volume averaging issues were obtained in all cases. Apart from the typical citrate, creatine, and choline signals of prostate spectra, both BPH and PC spectra showed a peak centered at 3.6 ppm which was assigned to myo-inositol. The intensity of this contribution was found significantly increased in PC cases compared to BPH. Possible relationships between neoplastic transformation and the metabolic pathways in which myo-inositol participates are discussed. Average spectroscopic profiles were calculated for both advanced pathologies, and showed obvious differentiated features. In quantitative terms, the ratio of citrate to choline peak areas as well as that of creatine to myo-inositol appeared as the most convenient to discriminate between advanced PC cases (both ratios below 1.0) and advanced BPH cases (both ratios above 1.0).

Phosphorus-31 chemical shift imaging of metastatic tumors located in the spine region

RATIONALE AND OBJECTIVES

Phosphorus-31 magnetic resonance spectroscopy was used in 14 cases to examine metastases of known malignant tumors located in the spine region. The purpose was to test the hypothesis that tumor phosphomonoester (PME) is elevated.

METHODS

Two-dimensional chemical shift imaging was used in combination with a slice-select gradient in the third dimension to obtain true three-dimensional localization.

RESULTS

The spectral maps revealed PME signals increased up to 10 times in voxels containing contrast-enhancing metastatic spine lesions compared with adjacent areas and peripheral muscle voxels. Phosphomonoester increase was significant for all tumors combined (8.6 +/- 5.3 arbitrary units versus 2.4 +/- 0.5 and 2.2 +/- 0.8 arbitrary units in unaffected myelum and corpora; P < 0.001), though smaller than 2 standard deviations in 5 of 14 cases. The latter shared high proportions of phosphocreatine, phosphocreatine > 30% of total phosphate, indicating substantial amounts of muscle tissue included in the tumor voxels (partial volume effect).

CONCLUSIONS

Phosphorus-31 MR spectroscopy can be of value in the recognition of malignant vertebral column abnormalities. Malignant tumor is marked by drastic PME increases-fourfold to tenfold, provided that partial volume effects are small.

Treatment-induced changes in 31P-MRS (magnetic resonance spectroscopy) spectra of sera from patients with acute leukemia

31P-nuclear magnetic resonance (NMR) spectra were obtained in vitro from sera of 40 healthy volunteers and 30 patients with acute leukemia (AL) at the time of diagnosis and repeated up to 2-13 times during therapy. All spectra consisted of inorganic phosphate (Pi) peak (used as a reference peak) and two peaks from phospholipids (PL): one peak due to phosphatidylethanolamine and sphingomyelin (PE + SM) and second peak due to phosphatidylcholine (PC). Prior to initiation of therapy 31P spectra of sera of patients with acute leukemia differed from spectra of sera of normal individuals. Peak intensities of the PL were low in relation to Pi. During therapy leading to remission, resonance from PL progressively increased approximately to the spectral pattern in normal sera. Contrary to that, in non-responders the intensities of the phospholipids peaks remained unchanged. Long-term follow-up 31P-MRS studies showed not only a good correlation between this 31P-MRS evolution of sera and the response to the therapy but also showed changes in phospholipids' levels in the following days during and after therapy. Moreover, correlations were found between high-density lipoprotein (HDL), cholesterol (CHOL) and low-density lipoprotein (LDL) concentrations measured by conventional techniques and peak intensities of PC and of PE + SM acquired by 31P-MRS.

In vivo 31P MRS evaluation of ganciclovir toxicity in C6 gliomas stably expressing the herpes simplex thymidine kinase gene

Phosphorus MRS was evaluated as a monitor of tumour therapeutic response to the herpes simplex virus thymidine kinase suicide gene therapy paradigm. In vivo 31P spectra were obtained from subcutaneous rat C6 gliomas constitutively expressing the HSVtk gene post treatment with ganciclovir (GCV, 15 mg/kg i.p., twice-daily). Significant regression (p < 0.1) of tumour volume was observed 10 days after beginning GCV administration. However, no changes in tumour pH or energy metabolites from pre-treatment values were observed. High-resolution 31P spectra of tumour extracts revealed a statistically significant reduction in the phosphocholine to phosphoethanolamine ratio six days post-GCV administration. These results indicate that the HSVtk/GCV-induced killing of tumours is not associated with corresponding changes in 31P MRS-observable energy metabolites and pH. The observed reduction in the PE/PC ratio may provide a non-invasive in vivo indicator of therapeutic efficacy.

Functional characterization of brain tumors: an overview of the potential clinical value

Early detection and characterization are still challenging issues in the diagnostic approach to brain tumors. Among functional imaging techniques, a clinical role for positron emission tomography studies with [18F]-fluorodeoxyglucose and for single photon emission computed tomography studies with [201Tl]-thallium-chloride has emerged. The clinical role of magnetic resonance spectroscopy is still being defined, whereas functional magnetic resonance imaging seems able to provide useful data for presurgical localization of critical cortical areas. Integration of morphostructural information provided by computed tomography and magnetic resonance imaging, with functional characterization and cyto-histologic evaluation of biologic markers, may assist in answering the open diagnostic questions concerning brain tumors.

Analysis of brain tumors using 1H magnetic resonance spectroscopy

BACKGROUND

Magnetic resonance imaging (MRI) is superior in delineating anatomic and pathologic information and has subsequently been married to the ability of magnetic resonance spectroscopy (MRS) to provide insight into the biochemical changes underlying pathology. Proton magnetic resonance spectroscopy (1H MRS) allows the non-invasive in vivo collection and measurement of chemical information from a selected volume of tissue (voxel).

METHODS

We conducted a prospective trial in 23 patients with brain mass lesions and 16 normal subjects using proton magnetic resonance spectroscopy (1H MRS). The spectra were analyzed for N-acetyl-aspartate (NAA), choline compounds (Cho), creatine (Cr), and lactate (Lac). The ratios of the compounds in tumors were compared to normals.

RESULTS

The tumors showed significant decreases in the mean peak height ratios of NAA/Cho, NAA/Cr, and significant increases in Cho/Cr when compared to tissue from normal subjects. Cho was elevated in all of the meningiomas and gliomas. In benign tumors, Cho was usually elevated while in metastases Cho was often normal or decreased. The four metastatic tumors showed NAA/Cho, NAA/Cr, and Cho/Cr that were similar to controls. Lac varied with tumor type and was elevated in many malignant primary brain tumors.

CONCLUSIONS

1H MRS is a powerful tool for safe, noninvasive analysis of tissue chemistry in vivo. Analysis of intracranial tumors reveals significant trends that might eventually be used in the classification of tumor histology and evaluation of the efficacy of tumor treatment.

Chemotherapy-associated changes in 31P MRS spectra of sera from patients with multiple myeloma

31P NMR spectra were obtained from sera of 22 healthy volunteers and 20 patients with multiple myeloma at the time of diagnosis and repeated up to five times during therapy. All spectra consisted of a Pi peak (used as a reference peak) and two peaks from phospholipids (PL): one peak due to phosphatidylethanolamine and sphingomyelin (PE + SM) and a second peak due to phosphatidylcholine (PC). Prior to therapy, peak intensities of the phospholipids were low relative to Pi. During therapy leading to remission, the resonance from PL progressively increased to approximate the spectral pattern seen in normal sera. By contrast, in non-responders an opposite trend was noted: the intensities of the phospholipid peaks became progressively reduced or remained unchanged. Long-term follow-up studies showed a good correlation between this 31P MRS evaluation of sera and the response of the disease to the therapy. In addition to the correlation with tumor response, our studies also show significant correlations between area, intensities of peaks of PE + SM, PC, and the concentrations of high-density lipoprotein (HDL) (correlation coefficients 0.46, 0.43, 0.59, respectively; p < 0.001). We found that the concentration of HDL in serum of patients with multiple myeloma was significantly reduced. In individuals responding to therapy HDL levels increased to the point where there were no statistically significant differences between them and healthy volunteers. In patients not responding to therapy, HDL concentration did not increase.(ABSTRACT TRUNCATED AT 250 WORDS)

Mineralizing microangiopathy: CT and MRI

Mineralizing microangiopathy, a distinctive histopathologic process involving the microvasculature of the central nervous system (CNS), is usually seen following combined radiation and chemotherapy for the treatment of CNS neoplasms in childhood. CT typically demonstrates calcification within the basal ganglia and subcortical white matter. The areas of calcification may give paradoxically increased signal on T1-weighted MRI due to a surface-relaxation mechanism, and decreased signal on T2-weighted images.

31P NMR phospholipid characterization of intracranial tumors

Phospholipid extracts from 48 intracranial tumors were analyzed using 31P NMR. Phospholipids commonly identified in the tumor spectra included phosphatidylglycerol (PG), phosphatidic acid (PA), diphosphatidylglycerol (DPG), uncharacterized phospholipid (U), ethanolamine plasmalogen (EPLAS), phosphatidylethanolamine (PE), phosphatidylserine (PS), sphingomyelin (SM), lysophosphatidylcholine (LPC), phosphatidylinositol (PI), a choline phospholipid (CPLIP), and phosphatidylcholine (PC). Differences in the mean relative mole-percentage of phosphorus concentrations of individual phospholipids were used to differentiate among tumors. Neural sheath tumors (neurilemmoma, neurofibroma and fibrosarcoma) were noted to contain significantly elevated levels of SM relative to tumors of neural glial origin and individually, glioblastoma multiforme was noted to contain depressed levels of SM relative to neurilemmoma, neurofibroma and meningioma. Significantly decreased levels of PA were noted for glioblastoma relative to neurilemmoma along with significantly decreased levels of PE relative to meningioma. Elevated levels of LPC and CPLIP were seen in glioblastoma multiforme relative to meningioma. Additional findings included elevated levels of PC for glioblastoma multiforme relative to neurofibroma, and neurilemmoma was differentiated from neurofibroma with elevated levels of PA and depressed levels of PI. 31P NMR phospholipid analysis provides supplemental biochemical information which may be used to improve the interpretation of spectra acquired in vivo, and reveals important tumor-specific biochemical information which may further improve the understanding of the biological behavior of intracranial tumors.

Characterization of a phosphonium analog of choline as a probe in 31P NMR studies of phospholipid metabolism

Tumors and transformed cells have been shown by 31P NMR to contain elevated concentrations of two phosphomonoesters, phosphorylcholine and phosphorylethanolamine, involved in phospholipid metabolism. In order to understand the biochemical basis for these phenomena new methods are needed to allow for analysis of the relevant metabolic pathways in intact cells. One such promising tool may be phosphonium-choline, a 31P NMR-visible analog of choline in which the trimethyl-ammonium group of choline has been replaced with a trimethyl-phosphonium moiety. As shown previously [Sim et al. Biochem. J. 154, 303 (1976)], this compound is non-toxic and readily metabolized by cultured cells into phospholipids. In this paper we describe in greater detail some of the chemical and NMR spectroscopic properties of this material. Most significantly we show here that the chemical shift of phosphonium-choline is sensitive to the phosphorylation state of the analog and that the phosphonium nucleus is NMR-visible even after its incorporation into phospholipid. The unique properties of this analog should make it possible to use high-field 31P NMR to follow the flux of phosphonium-choline through the Kennedy pathway in intact perfused cells cultures.

Phosphorus-31 magnetic resonance spectroscopy and ventricular enlargement in bipolar disorder

Phosphorus-31 magnetic resonance spectroscopy (31P-MRS) was used to examine whether reduced levels of phosphomonoesters (PME) were correlated with ventricular enlargement in 40 patients with bipolar disorder and 60 age-matched normal control subjects. Ventricular enlargement was assessed by magnetic resonance imaging (1H-MRI) using the following three methods: Evans ratio (ER), Huckman number (HN), and minimum distance of caudate nuclei (MDCN). Although MDCN and ER were significantly larger in the patient group, no significant correlations were found between 31P-MRS and 1H-MRI. PME was negatively correlated with age in bipolar disorder. Decreased levels of PME were found only in bipolar I disorder. Intracellular pH was positively correlated with duration of lithium treatment. These results suggest that the observed PME reduction was not related to ventricular enlargement, but the issue should be further studied with volumetric MRI analysis.

In vivo 31P MRS of experimental tumours

More than 50% of cancers fail to respond to any individual treatment and tumour follow-up after treatment plays a major role in routine therapy planning and pharmacological research. Today, MRS is the only technological approach providing non-invasive access to tumour biochemistry. Ten years ago, expectations were raised concerning 31P MRS as an exciting and promising technical approach to the study of tumours. However the expectations have not always come to fruition. How close are we now to seeing routine 31P NMR in clinical oncology? This review of the 127 published papers shows spectroscopy results in more than 150 experimental animal tumour models. These tumour/host/treatment systems provide us with a useful basis to evaluate the current state of the art, summarize the basic knowledge presently available, determine the key points underlying the present disappointment of some clinical oncologists and stimulate new basic research. The information collected concerns the discussion of the reliability of experimental models in oncology, the technical improvement of magnetic resonance technology and the monitoring of bioenergetic status, pH regulation and phospholipid metabolism in treated and untreated tumours. Recent advances (two-thirds of the papers have been published in the last 5 years) seem to provide more optimistic perspectives than those generally accepted a few years ago, in the depressing period following early pioneering work.

New and future advanced imaging techniques

The advances in diagnostic imaging in the past 20 years have been nothing short of phenomenal. This article presents what is available with some of the latest in state-of-the-art techniques. Included are three-dimensional techniques, vascular MRI, high-resolution three-dimensional video animation, positron emission tomography, MR spectroscopy, and other developing methodology in diagnostic imaging.

NMR spectroscopy: current status and future possibilities

Nuclear magnetic resonance (NMR) spectroscopy is now established as a non-invasive method of studying metabolism in living systems, ranging from cellular suspensions to man. With respect to clinical applications, recent developments include the successful implementation of new techniques for spatial localisation, and in particular the acquisition of excellent 1H spectra from selected regions of the human brain. Localised 1H spectroscopy opens the way to monitoring a wide range of compounds that are inaccessible to 31P NMR, and should add considerably to the information that is available from 31P studies. NMR spectroscopy does, however, have its limitations, which arise primarily from the fact that it is an insensitive technique. This lack of sensitivity limits the spatial resolution for metabolic studies, and means that metabolites must be present at fairly high concentrations in order to produce detectable signals. In this article, we illustrate the scope and limitations of NMR spectroscopy by describing a few examples of studies undertaken on animals and humans.

Regional metabolism in experimental brain tumors in cats: relationship with acid/base, water, and electrolyte homeostasis

Experimental brain tumors were produced in cats by xenotransplantation of the rat glioma clone F98 into the white matter of the left hemisphere. One to 4 weeks after implantation, local adenosine triphosphate (ATP), glucose, lactate, and tissue pH were measured via imaging techniques in cryostat sections passing through the center of the tumor and correlated with changes in water and electrolyte content. The tumors exhibited a heterogeneous metabolic pattern, with a tendency for ATP to decrease and lactate to increase during tumor development. Tissue pH was above 7.5 in tumors with high ATP content but it sharply declined at low ATP levels. In peritumoral edema, ATP also decreased and lactate increased but, in contrast to tumor tissue, pH became more alkaline. Metabolic changes were associated with edema formation, as evidenced by the rise in water and sodium content. There was a distinct difference between tumor tissue and peritumoral edema: in tumor tissue, pH declined with increasing water content, whereas in peritumoral edema it increased. These observations are interpreted as follows: 1) in tumor tissue, "lactacidosis" and ATP depletion are attributed to disturbances in blood flow, resulting in metabolic failure and the intracellular "cytotoxic" accumulation of water; 2) in peritumoral edema, "lactalkalosis" is the result of an efflux of (alkaline) lactate salts from the tumor into the expanded extracellular compartment, and the decrease in ATP is the volumetric effect of extracellular "vasogenic" edema fluid and not the result of cellular energy failure. These findings are of importance for the interpretation of volume-selective magnetic resonance spectroscopy and may contribute to the establishment of spectroscopic criteria for the evaluation of therapeutical interventions.

Studies of human tumors by MRS: a review

The literature describing 31P, 1H, 13C, 23Na and 19F MRS in vivo in human cancers is reviewed. Cancers have typical metabolic characteristics in 31P and 1H MRS including high levels of phospholipid metabolites and a cellular pH more alkaline than normal. These alone are not specific for cancer but are diagnostic in appropriate clinical settings. Some metabolic characteristics appear to be prognostic indices and correlation with treatment response is emerging as an important potentially cost-effective use of MRS in oncology. 19F MRS examines pharmacokinetics of 5-fluorouracil and by demonstrating its retention predicts response of a cancer to treatment. Current needs include improvement of diagnostic specificity by use of techniques like multivoxel MRS, proton decoupling of 31P, short echo time and fat-suppressed 1H MRS, 13C MRS direct or via 1H-observe, and statistical analysis of multiple spectral features. Trials in large populations in well defined clinical settings are needed to determine if MRS can provide independent prognostic indices useful in cancer management.

Characterization of prostate cancer, benign prostatic hyperplasia and normal prostates using transrectal 31phosphorus magnetic resonance spectroscopy: a preliminary report

We assessed the ability of 31phosphorus (31P) transrectal magnetic resonance spectroscopy to characterize normal human prostates as well as prostates with benign and malignant neoplasms. With a transrectal probe that we devised for surface coil spectroscopy we studied 15 individuals with normal (5), benign hyperplastic (4) and malignant (6) prostates. Digital rectal examination, transrectal ultrasonography and magnetic resonance imaging were used to aid in accurate positioning of the transrectal probe against the region of interest within the prostate. The major findings of the in vivo studies were that normal prostates had phosphocreatine-to-adenosine triphosphate (ATP) ratios of 1.2 +/- 0.2, phosphomonoester-to-beta-ATP ratios of 1.1 +/- 0.1 and phosphomonoester-to-phosphocreatine ratios of 0.9 +/- 0.1. Malignant prostates had phosphocreatine-to-beta-ATP ratios that were lower (0.7 +/- 0.1) than those of normal prostates (p less than 0.02) or prostates with benign hyperplasia (1.1 +/- 0.2, p less than 0.01). Malignant prostates had phosphomonoester-to-beta-ATP ratios (1.8 +/- 0.2) that were higher than that of normal prostates (p less than 0.02). Using the phosphomonoester-to-phosphocreatine ratio, it was possible to differentiate metabolically malignant (2.7 +/- 0.3) from normal prostates (p less than 0.001), with no overlap of individual ratios. The mean phosphomonoester-to-phosphocreatine ratio (1.5 +/- 0.5) of prostates with benign hyperplasia was midway between the normal and malignant ratios, and there was overlap between individual phosphomonoester-to-phosphocreatine ratios of benign prostatic hyperplasia glands with that of normal and malignant glands. To verify the in vivo results, we performed high resolution magnetic resonance spectroscopy on perchloric acid extracts of benign prostatic hyperplasia tissue obtained at operation and on a human prostatic cancer cell line DU145. The extract results confirmed the differences in metabolite ratios observed in vivo. We conclude that transrectal 31P magnetic resonance spectroscopy can characterize metabolic differences between the normal and malignant prostate.

Characterization of astrocytomas, meningiomas, and pituitary adenomas by phosphorus magnetic resonance spectroscopy

Phosphorus magnetic resonance (MR) spectroscopy allows noninvasive measurement of phosphate-containing compounds and pH within brain cells. The authors obtained localized phosphorus MR spectra from 10 normal brains, four low-grade astrocytomas, six glioblastomas, four meningiomas, and three pituitary adenomas and found differences in the spectra of each tumor type. Compared to normal brain, the spectra from low-grade astrocytomas showed a significant reduction of the phosphodiester (PDE) peak. Glioblastomas were characterized by a significant reduction of the PDE peak, elevation of the phosphomonoester (PME) peak, and a relatively alkaline intracellular pH. The spectra from meningiomas and pituitary adenomas were markedly different from the glial tumors. Meningiomas showed significant reductions in phosphocreatine, PDE, and inorganic phosphate, as well as a relatively alkaline pH. Pituitary adenomas resembled meningiomas, but had a much higher PME peak. Although the number of tumors studied was small, there appears to be a characteristic spectrum associated with these different tumor types. The present findings can be useful in the preoperative identification of these tumors and in furthering understanding of their growth and metabolism in vivo.

In vivo 1H-spectroscopy of human intracranial tumors at 1.5 tesla. Preliminary experience at a clinical installation

Magnetic resonance spectroscopy (MRS) may contribute to the characterization of intracranial tumors in vivo. Volume selective water suppressed proton spectroscopy offers the possibility to study a number of metabolites in the brain including choline (CHO), creatinine/phosphocreatinine (CR/PCR), N-acetylaspartate (NAA), and lactate. Using the stimulated echo technique we have studied 17 patients with intracranial tumors. In all cases the tumors were classified based on histologic evaluation. The tumor spectra differed considerably from those obtained in healthy brain tissue. The results indicate a relative decrease in the NAA and CR/PCR content. In many cases a lactate peak could be seen especially in the tumors with malignant growth characteristics. Our preliminary results suggest that proton spectroscopy may contribute to the differentiation of brain tumors with respect to benign or malignant growth. However, further research is warranted before a definite conclusion can be drawn.

New directions in medical imaging of cancer. Magnetic resonance methods and single photon emission computed tomography

Magnetic resonance methods and single photon emission computed tomography (SPECT) are developing technologies that provide both functional and anatomic information. Their role in the diagnosis and monitoring of cancer is the subject of current clinical research. Magnetic resonance imaging (MRI) delineates organs and tissue heterogeneities using differences in the relaxation parameters of water and fat protons; both protons and other nuclei can be imaged or studied by magnetic resonance spectroscopy (MRS) to provide information on the state of naturally occurring or infused molecules. SPECT quantifies the distribution of radiolabeled agents in tissues and organs; labeled monoclonal antibodies provide highly specific imaging of tumors. Spatial resolution is the limiting technologic factor. Proton MRI provides the highest current resolution, better than 1 mm in vivo in deep tissues, whereas the resolution of MRS and SPECT is limited to several cubic centimeters. Recent advances in these technologies have significantly increased their specificity and ability to detect small, deep lesions.

31P NMR spectra of extremity sarcomas: diversity of metabolic profiles and changes in response to chemotherapy

We have used 31P NMR spectroscopy to study 22 patients with suspected sarcomas prior to any treatment. The spectra are characterized by the same peaks noted in murine tumors. The mean pH was 7.14 +/- 0.08 and PCr/Pi was 1.18 +/- 0.83. Comparison of pH and PCr/Pi ratios in human and a murine tumor with a low hypoxic cell fraction revealed no significant differences. Six patients subsequently received chemotherapy and three responded to therapy (based on pathologic examination and/or tumor reduction greater than 50%). The three responding patients were noted to have significantly lower PDE/PME in their pretreatment spectra than the three nonresponding patients. The three responding patients with sarcomas also showed a rise of greater than 100% in PDE/PME during the first cycle of therapy. Two of the responding patients had an increase of 0.37 pH units during this interval, which was not detected in the nonresponding patients. These data suggest that 31P NMR spectroscopy may be a useful prognostic indicator in conjunction with other clinical parameters.

Image-guided 31P magnetic resonance spectroscopy of normal and transplanted human kidneys

Image-guided 31-phosphorus magnetic resonance spectroscopy (MRS) was used to obtain spatially localized 31P spectra of good quality from healthy normal human kidneys and from well-functioning renal allografts. A surface coil of 14 cm diameter was used for acquiring phosphorus signals solely from a volume-of-interest located within the kidney. To determine the effects of kidney transplantation on renal metabolism, patients with well functioning allografts were studied. Little or no phosphocreatine in all spectra verifies the absence of muscle contamination, and is consistent with proper volume localization. The intensity ratio of phosphomonoesters (PME) to adenosine triphosphate (ATP) resonances in transplanted kidneys (PME/ATP = 1.1 +/- 0.4) was slightly elevated (P = 0.2) compared to that of healthy normal kidneys (PME/ATP = 0.8 +/- 0.3). The inorganic phosphate (Pi) to ATP ratio was similar in the two groups (Pi/ATP = 1.1 +/- 0.1 in transplanted kidneys vs. 1.2 +/- 0.6 in normal kidneys). Acid/base status, as evidenced from the chemical shift of Pi, was the same in both normal controls and transplanted kidneys. Despite the practical problems produced by organ depth, respiratory movement, and tissue heterogeneity, these results demonstrate that image-guided 31P MR spectra can reliably be obtained from human kidneys.

An assessment of the sensitivity of in vivo 31P nuclear magnetic resonance spectroscopy as a means of detecting pH heterogeneity in tumours: a simulation study

The presence of hypoxia and low pH in tumours is known to influence several treatment modalities including radiotherapy, chemotherapy and hyperthermia. Hypoxic and acidic regions have been demonstrated in tumours using pH and pO2 microelectrodes. The technique has shown marked heterogeneity within individual tumours. A non-invasive measure of intracellular pH is provided by 31P nuclear magnetic resonance (NMR) spectroscopy from determination of the chemical shift of inorganic phosphate, and is being increasingly applied to the study of human tumours in vivo. Based upon the assumption that hypoxic cells are also acidic, we have assessed the ability of a whole-body NMR spectrometer to detect acidic subpopulations within a tumour using simulated tumour spectra. In these simulations the size of the acidic subpopulation, assigned pH values from 5.0 to 7.0, was varied between 5% and 50% of the neutral "control" population. Gaussian noise was added to the simulated spectra giving signal-to-noise ratios for the neutral control inorganic phosphate peak of 3, 5 and 7, which are typical of values encountered in human tumour measurements. From the results of the simulations it seems unlikely that human hypoxic, and presumably acidic, cell fractions, typically 1-15%, will be detected by this method in the presence of signal-to-noise levels characteristic of current equipment. Therefore, whole-body in vivo 31P NMR spectroscopy may fail to identify significant pH heterogeneity within human tumours. Improvements in signal-to-noise and line separation owing to improvements in technique and higher field strength instruments may improve sensitivity to heterogeneous populations.

Comparison of 31P MRS and 1H MRI at 1.5 and 2.0 T

The goals of this study were to compare 31P magnetic resonance spectroscopy (MRS) and 1H magnetic resonance imaging (MRI) of human subjects and phantoms at 1.5 and 2.0 T. The 31P signal-to-noise (S/N) ratios in phantom standards and in localized volumes in human brain and liver were compared at 1.5 and 2.0 T. In addition, T1 values for 31P resonances in human brain, 31P linewidths of metabolites in human brain and liver, 1H S/N in a phantom standard, and MR image quality in human head and body were compared at the two field strengths. The results of our study showed that at the higher strength field, (1) in vivo 31P MRS studies benefited from up to 32% improvement in S/N; (2) in vivo 31P MRS studies also benefited from increased spectral dispersion; (3) the quality of MR head images remained comparable; and (4) body images showed some decrease in image quality due to increased chemical shift, and flow and motion artifacts.

Detection of metabolic heterogeneity of human intracranial tumors in vivo by 1H NMR spectroscopic imaging

Patients with intracranial tumors (gliomas) were examined by means of localized water-suppressed 1H NMR single volume spectroscopy and spectroscopic imaging. The 1H NMR spectra of the tumors exhibit signal intensities of the N-acetyl aspartate, choline compounds, and creatine plus phosphocreatine resonance lines that are different from the corresponding intensities observed on normal brain tissue. Also, for 6 out of the 10 patients examined so far, lactate resonance lines were detected in the tumor spectra. For one patient, abnormal 1H NMR spectra were obtained of a hemisphere which appeared normal with 1H NMR imaging. Metabolic heterogeneity of the tumorous regions could be demonstrated with 1H NMR spectroscopic imaging, using a spatial resolution in the order of 1 cm. These results suggest a spectrum of metabolic observations that may ultimately provide an important means for characterizing brain tumors.

Phosphorus-31 magnetic resonance spectroscopy in humans by spectroscopic imaging: localized spectroscopy and metabolite imaging

In in vivo phosphorus magnetic resonance spectroscopy (MRS), spectroscopic imaging (SI) can be used as a flexible localization technique, producing spectra from multiple volumes in a single examination. Presented here are phosphorus SI studies of human organs in which a selective-volume SI reconstruction was used rather than the usual array-format SI reconstruction. A linear predictor technique was used to estimate the initial points of the free induction decay missing because of the delay needed for phase-encoding gradients, significantly reducing the baseline artifacts which commonly complicate interpretation of SI spectra. In studies of heart, brain, liver, and kidney, the performance of SI was found to compare favorably with that of ISIS. SI phosphorus metabolite intensity images from a brain tumor patient were obtained at 2 X 2-cm in-plane resolution (with "slice" thickness of roughly 16 cm, determined by coil sensitivity) in 34 min, demonstrating the feasibility of obtaining clinically useful metabolite images in clinically reasonable examination times.

Phosphodiesters in the liver: the effect of field strength on the 31P signal

31P NMR spectra of the rat liver were recorded in vivo at 2.35 and 8.5 T. There was a large peak in the phosphodiester region of spectra obtained at 2.35 T which was much reduced at 8.5 T. The peak at 2.35 T is unlikely to be primarily from free cytosolic phosphodiesters, which would not be expected to display such a marked field dependence.

1H image-guided localized 31P MR spectroscopy of human brain: quantitative analysis of 31P MR spectra measured on volunteers and on intracranial tumor patients

1H image-guided 31P MR spectra of normal human brain and of intracranial tumors have been analyzed quantitatively. Tumor types examined include prolactinoma, lymphoma, and various grade gliomas. The experimental signals were processed by means of a time-domain least-square fitting procedure, which yields the spectral parameters, as well as a prediction of the standard deviations. Significant spectral variations are observed within both populations of normal brain and of intracranial tumor 31P MR spectra. The metabolic ratios derived from the glioma 31P MR spectra and from corresponding uninfiltrated brain tissue do not differ significantly. Significant differences are, however, observed between the metabolic ratios of prolactinoma and uninfiltrated tissue 31P MR spectra. Alkaline pH values are found for the prolactinoma and the high-grade gliomas. Furthermore, spectral differences are observed between the patient's uninfiltrated tissue 31P MR spectra and those of an unmatched population of volunteers. This underscores the necessity for control measurements on the uninfiltrated tissue of the patient and for controls from a matched population of healthy individuals.