Magnetic Resonance Spectroscopy Imaging, a Diagnostic Tool in Prostate Cancer

By Ryan C. Handzlik, MS, and David A. Corral, MD

The human prostate remains one of the most frequently diseased organs.1 Prostate cancer has been identified as the most common malignancy in males. Treating prostate cancer at an early stage may be the most effective means of achieving long-term survival, and a number of diagnostic methods have been used to detect and monitor early-stage malignancies. The previously available noninvasive techniques that have been used to diagnose this disease do not adequately differentiate prostate cancer from benign prostatic hyperplasia (BPH) and normal prostatic tissues.2 The only definitive measure for distinguishing cancer of the prostate from BPH and normal tissue remains a histological analysis of biopsy samples.

Magnetic resonance spectroscopic imaging (MRSI) has been proposed as a noninvasive, diagnostic approach for defining the presence and spatial extent of prostate cancer.3 The technique potentially can differentiate prostatic adenocarcinoma from BPH and normal prostatic zonal anatomy based on observable metabolite levels present in the prostate. MSRI can be used to measure the distribution of choline and citrate throughout the normal and diseased prostate in an effort to define the presence, extent, and orientation of prostate cancer.2


The ability to differentiate areas of prostate cancer from BPH and normal prostatic tissue via a noninvasive approach is important in terms of every aspect of patient care, early cancer staging, and follow-up. Studies using animal models and tissue extracts have identified low levels of citrate and high levels of choline in regions of prostate cancer.2 Levels of citrate and choline in regions of normal peripheral zone and BPH were used as a baseline for tracking metabolite levels. The differences in metabolite levels in cancer, BPH, and the normal peripheral zone are the focus of the investigations, as 68% of all prostatic cancers occur in the peripheral zone.

The combined use of both localized MRSI and high-resolution MRI of the in situ human prostate has demonstrated lower mean levels of citrate within regions that have been identified as cancerous. It has been proposed that malignant epithelial cells have a diminished capacity for synthesizing and secreting citrate and developing glandular ducts for citrate storage.2

Prior to the study by Kurhanewicz and colleagues, changes in choline levels in prostatic cancers were never measured. Choline remains an important metabolite because choline compounds have been implicated in cell membrane synthesis and degradation.4 Elevations of choline peaks in MR spectra have been reported in a number of human cancers to date. However, the range of metabolite levels associated with normal prostate anatomy, BPH, and cancers of varying grades have not been studied. MRSI has the advantage of providing MR spectra from more than one region, as well as stipulating the exact size and spatial position of the area under investigation. The technique, along with MRI, allows the investigator to measure the distribution of choline and citrate throughout the normal and diseased prostate.

Magnetic Resonance Spectroscopic Imaging of the in situ Human Prostate

In an investigation by Kurhanewicz et al, 3D MRSI, in combination with endorectal MRI techniques, was used to determine the spatial extent of prostate cancer by generating metabolite images and comparing metabolite ratios to normal peripheral zone values present in the prostate. The study was composed of nine healthy volunteers younger than age 40 (ages 28-36), five patients with BPH (ages 63-77), and 85 patients with biopsy-positive prostate cancer (ages 55-75). The results were collected by aligning 3D MRSI data with MRI data, then comparing the results with the pathological findings from biopsy tissue samples. The results of the study identified higher levels of choline and lower levels of citrate in regions of cancer when compared to BPH and normal tissues.2 Upon completion of MRSI in the 85 prostate cancer patients, significantly lower mean levels of citrate (P = 0.0001), and higher mean levels of choline (P = 0.001) were detected in regions of cancer when compared to the normal peripheral zone in the same patient.2

Hahn and coworkers performed a similar study that displayed potential for the possibility of an in vivo analysis of the prostate using MRSI.5 They ultimately identified six spectral subregions as having diagnostic potential in prostate tissue. However, in the investigation they found it difficult to integrate only choline resonance in its spectral region because peak overlaps were present in this region of the spectrum. Particularly, this spectral region includes, among other metabolites, creatine resonance. Citrate levels between cancer and BPH were found to be statistically significant. Several hypotheses have been proposed for these findings that have identified an increased amount of the enzyme aconitrase in cancer cells, which contributes to the break down of citrate.5 In addition, increased levels of the enzyme ATP-citrate lyase in malignant cells is proposed to contribute to citrate breakdown during lipid synthesis.

The classification of BPH vs. cancer in the Hahn et al study provided MRSI diagnoses that are consistent with histopathology results. These results were directly indicative of the high sensitivity of the technique employed in the study.5 The results ultimately show that MRSI produces high sensitivity and specificity that can reliably be used for distinguishing between benign and malignant prostatic tissue (100% sensitivity, 95.5% specificity).5

Approaches to Prostate Cancer Using Magnetic Resonance Spectroscopy Imaging

Zaider and colleagues describe a method that correlates MRSI data to intraoperatively-obtained ultrasound images and incorporates these data into a treatment planning system for brachytherapy.6 After MRSI data are obtained, regions of high risk for cancer cells were identified based upon elevated peaks in the MR spectrum. These peaks were then mapped on a spatial grid covering the entire prostate tissue. This technique is followed by an integer-programming procedure in which optimal radioactive seed distribution is determined, and tissue is then implanted within the prostate. This technique is potentially effective, as it may spare surrounding healthy tissues from radioactive dose escalations employed during the course of treatment.

Current treatment-planning algorithms have been developed to determine the ideal placement of radioactive seeds in the prostate. However, uncertainties about tumor position force the delivery of a maximum dose of radiation to the entire prostate gland, which may result in unnecessary dose escalations to the urethra. In the end, the side effects following dose escalation can have an overall effect on the patients’ quality of life and adversely result in urinary side effects.6 MRSI mapping of citrate and choline levels within the prostate can identify regions associated with prostate carcinoma. Ultimately, this approach facilitates the localization of tumors to specific regions within the prostate, and radioactive seeds preferentially can be placed in areas identified with tumor, thus minimizing radiation doses to normal regions.

The results of the study by Zaider et al indicate that the incorporation of clinical data from MRSI into a brachytherapy treatment-planning optimization system is feasible.6 The findings revealed that dose escalation is critical for improved outcome. The treatment-planning model may lead to local control. In essence, the MRSI-guided treatments direct an increased dose of 125I radioactive seeds to individual target areas in the prostate without adversely affecting healthy tissue surrounding the prostate and the urethra.


The MRSI techniques described enable identification of more sites of carcinoma of the prostate than does prostate biopsy. These results indicate that a larger volume of cancer normally is present upon diagnosis than is indicated by biopsy alone. In patients with detectable elevated prostate specific antigen (PSA), MRSI identified, location-for-location, all foci of prostate carcinoma and benign prostatic tissue that were identified on prostate biopsy.7 Ultimately, MRSI is superior to transrectal ultrasound and MRI for differentiating carcinoma of the prostate from BPH.7 In the future, the combined use of endorectal magnetic resonance imaging and MRSI will serve as an invaluable diagnostic tool for differentiating normal from carcinomatous prostate. (Dr. Corral is Editor-in-Chief of Cancer Research Alert and is in the Department of Urologic Oncology, Roswell Park Cancer Institute; Mr. Handzlik is a Graduate Student in the Natural Sciences Program at Roswell Park Cancer Institute, Buffalo, NY.)


1. Greenlee RT, Hill-Harmon MB, Murray T, et al. Cancer Statistics, 2001. Cancer J Clin 2001;51:15-36.

2. Kurhanewicz J, Vigneron DB, Hricak H, et al. Three-dimensional H-1 MR spectroscopic imaging of the in situ human prostate with high (0.24-0.7-cm3) spatial resolution. Radiology 1996;198:795-805.

3. Kurhanewicz J, Vigneron DB, Hricak H, et al. Prostate cancer: Metabolic response to cryosurgery as detected with 3D H-1 MR spectroscopic imaging. Radiology 1996;200:489-496.

4. Hara T, Kosaka N, Kishi H. PET imaging of prostate cancer using carbon-11-choline. Nucl Oncol 1998;39: 990-996.

5. Hahn P, Smith I, Leboldus L, et al. The classification of benign and malignant human prostate tissue by multivariate analysis of 1H magnetic resonance spectra. Cancer Res 1997;57:3398-3401.

6. Zaider M, Zelefsky MJ, Lee EK, et al. Treatment planning for prostate implants using magnetic-resonance spectroscopy imaging. Int J Radiat Oncol Biol Phys 2000;47:1085-1096.

7. Parivar F, Hricak H, Shinohara K, et al. Detection of locally recurrent prostate cancer after cryosurgery: Evaluation by transrectal ultrasound, magnetic resonance imaging, and three-dimensional proton magnetic resonance spectroscopy. Urology 1996;48:594-599.

8. Pickett B, Vigneault E, Kurhanewicz J, et al. Static field intensity modulation to treat a dominant intra-prostatic lesion to 90 GY compared to seven field 3-dimensional radiotherapy. Int J Radiat Oncol Biol Phys 1999;43:921-929.