By: Dan Sperling, MD

About a third of patients diagnosed with localized prostate cancer choose some type of external beam radiation therapy (EBRT) to treat their cancer. There are many reasons why EBRT is the choice, including not wanting or qualifying for surgery, and not being a candidate for a focal treatment. As with all treatments, there is a risk of recurrence (cancer returning). On average, biochemical recurrence rates (rising PSA) after primary treatment with EBRT are reported as anywhere from 20-60%.

Typically, three successive rises in PSA will lead to a DRE, TRUS biopsy, CT and bone scanning. In the effort to identify localized recurrence (still contained in the prostate), there is a potential margin for error with PSA, DRE and TRUS biopsy. The PSA test itself cannot accurately predict for any given individual whether the recurrence is localized or has already begun to infiltrate beyond the prostate capsule. The digital rectal exam may be inconclusive due to the fibrous and irregular nature of the radiated gland. The ultrasound-guided biopsy may miss or underestimate localized recurrence because ultrasound is blind to tissue differences within the gland.

An article published in 2010 in the journal Investigative Radiology explains the research done by a Dutch team regarding the merits of 3T MRI and MRI-guided biopsy to identify localized recurrence after EBRT.[i] To the best knowledge of the team, no previous studies on this use of MRI had been published. Because of the acknowledged difficulty of imaging irradiated prostate glands due to the radiation-induced shrinkage and tissue alterations, this can be said to be a landmark study.

Twenty-four patients with three successive rises in PSA were enrolled in the study. The imaging protocol is described in technical detail in the article. Essentially, the protocol involved capturing T2-weighted and T1-weighted images before and after the administration of a contrast agent. The sequence and positioning of the image capture was such that a 3-dimensional model of each patient’s prostate was constructed. Computer colorization was used to highlight the dynamic contrast enhancement (DCE-MRI) once the contrast agent was injected. Two experienced readers interpreted the images, and tumor suspicious areas (TSR) were identified.

Following tumor localization based on the images, each patient underwent an MRI-guided targeted biopsy, supervised by a radiologist with four years experience in this method. Needles were directed into the TSRs. The report stated that tissue samples were successfully obtained from all patients, the procedure was well-tolerated, and there were no complications. All samples were examined by a specialized pathologist with 18 years of experience. Samples were deemed negative if there was no evidence of living cancer. All positive biopsies were given a Gleason score.

Based on MR imaging, four patients were determined to have metastatic disease (to bone or lymph nodes) and were placed on hormone ablation. In the remaining 20 patients, a total of 38 TSRs were found by imaging and subsequently biopsied with MRI-guided targeted sampling of the TSRs. Most TSRs were found using the combination of T2W and DCE-MRI, but in 5 patients DCE-MRI revealed a total of 8 TSRs that were not detected with the T2W sequences.

Fifteen patients were found to have biopsy-proven local recurrence. They were treated with either salvage cryosurgery (5 patients), salvage RP (1 patient), watchful waiting (1 patient), hormonal therapy (2 patients), were lost to follow-up (4 patients) or died (2 patients). Of the 38 different TSRs detected by imaging, 26 contained biopsy-proven recurrence (67%), 8 showed radiotherapy-incuded atypical areas, 1 contained residual indeterminate prostate cancer with severe radiation changes (3 %) and the remaining three had fibrosis (8%), similar to scar tissue from the radiation.

On a per-patient basis, the MRI-guided biopsy into TSRs had a positive predictive value of 75% (15/20), meaning it correctly predicted cancer; on a per-TSR basis, the positive predictive value was 68% (26/38). Gleason score 10 was present in 2/26 (8%), Gleason score 9 in 8/26 (31%), Gleason score 8 in 3/26 (12%), Gleason score 7 in 9/26 (35%), and Gleason score 6 in 4/26 (15%) TSRs.

Although TRUS-guided biopsy has limitations in that it can miss cancer and/or underestimate the aggression level when it’s found, it continues to be widely used because it is more accessible than MRI-guided biopsy. Still, the authors demonstrated the feasibility of detecting and diagnosis localized radiation-recurrent cancer using multiparametric MRI. The value for clinicians and patients alike is the efficiency and accuracy of diagnosis that facilitates treatment planning. This is especially important if patients are to be considered candidates for a local salvage therapy, which is potentially curative, unlike placing them on hormone therapy.