0022-5347/91/1461-0066$03.00/0 THE JOURNAL OF UROLOGY Copyright © 1991 by AMERICAN UROLOGICAL ASSOCIATION, INC. Vol. 146,66-74, July 1991 Printed in U.S.A. CHARACTERIZATION OF PROSTATE CANCER, BENIGN PROSTATIC HYPERPLASIA AND NORMAL PROSTATES USING TRANSRECTAL 31PHOSPHORUS MAGNETIC RESONANCE SPECTROSCOPY: A PRELIMINARY REPORT PERINCHERY NARAYAN,* PRAHALAD JAJODIA, JOHN KURHANEWICZ, ALBERT THOMAS, JEFFREY MACDONALD, BRUNO HUBESCH, MARCUS HEDGCOCK, CHARLES M. ANDERSON, THOMAS L. JAMES, EMIL A. TANAGHO and MICHAEL WEINER From the Departments of Urology, Pharmaceutical Chemistry and Radiology, University of California School of Medicine and Veterans Administration Medical Center, San Francisco, California ABSTRACT We assessed the ability of 31phosphorus ( 31 P) 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, trans rectal ultrasonography and magnetic resonance imaging were used to aid in accurate positioning of the trans rectal 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-,B-ATP ratios of 1.1 ± 0.1 and phosphomonoester-to- phosphocreatine ratios of 0.9 ± 0.1. Malignant prostates had phosphocreatine-to-,B-ATP ratios that were lower (0.7 ± 0.1) than those of normal prostates (p <0.02) or prostates with benign hyperplasia (1.1 ± 0.2, p <0.01). Malignant prostates had phosphomonoester-to-,B-ATP ratios (1.8 ± 0.2) that were higher than that of normal prostates (p <0.02). Using the phosphomonoester-to-phosphocrea- tine ratio, it was possible to differentiate metabolically malignant (2.7 ± 0.3) from normal prostates (p <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 differ- ences between the normal and malignant prostate. KEY WORDS: prostate, carcinoma, prostatic hypertrophy, nuclear magnetic resonance, prostatic neoplasms Magnetic resonance imaging (MRI) 1-3 and spectroscopy4-11 are relatively new techniques for the study of tumor character- istics and cell biology. In prostatic cancers MRI techniques using conventional body coils have proved to be comparable if not superior to other modalities, such as computerized tomog- raphy and trans rectal ultrasonography, in determination of the local extent of cancer. 12 - 16 However, the image quality and resolution have not permitted use of this modality for 2 of the indications most needed clinically, namely screening of patients for cancer and differentiation of benign adenomas from can- cerous nodules. 1 - 3 Magnetic resonance spectroscopy studies have even more potential than MRI studies because of the ability of magnetic resonance spectroscopy to assess noninva- sively biochemical parameters of tissues in vivo. In 31phosphorus e 1 p) magnetic resonance spectroscopy studies the increased growth rate 1 ? and metabolic turnover 18 of malig- nant cells have been characterized by altered concentrations of high and low energy phosphorus metabolites as compared to normal tissues. 19 Also, studies in our laboratory on rat prostatic cancer suggest that 31P magnetic resonance spectroscopy can Accepted for publication November 9,1990. Read at annual meeting of American Urological Association, Dallas, Texas, May 7-11, 1989. * Requests for reprints: Department of Urology, U-518, University of California, San Francisco, California 94143-0738. detect differences between hormone-sensitive and hormone- resistant tumors. 20 ,21 An additional observation was that the effects of androgen deprivation were observed earlier (3 days versus 2 weeks) by 31 P magnetic resonance spectroscopy com- pared to conventional changes in size and growth parameters. Magnetic resonance spectroscopy of the human prostate gland using conventional surface coils placed outside the body is technically not feasible. This is because the small size and deep location of the prostate preclude acquisition of good signal- to-noise magnetic resonance spectra free of contamination from surrounding tissues. To overcome these problems we developed a transrectal probe containing an insertable coil capable of obtaining 31p spectra and high resolution proton images. We initially tested this probe in canines 22 and subsequently modi- fied it to perform human studies. Even with the use of trans- rectal surface coils a limitation in human studies is the un- quantifiable contribution of rectal wall muscle and intrapros- tatic tissues [normal and benign prostatic hyperplasia (BPH)] to the tumor spectra. To determine the degree of spectral contamination from the rectal wall to the in vivo 31 P tumor spectra, we performed high resolution magnetic resonance spec- troscopy studies of tissue extracts of BPH and a prostate cancer cell line DU145. 23 Tissue extracts of cancer obtained from radical prostatectomy were not possible, since high energy 31 P metabolites undergo rapid degradation during the time required 66