By: Dan Sperling, MD
Considerable dialogue on what constitutes clinically significant prostate cancer, and whether insignificant prostate cancer (PCa) requires immediate or aggressive treatment, is ongoing in major professional societies and institutes. Universal agreement remains elusive, resulting in confusion for both physicians and patients. Two key questions are raised: 1) which clinical factors should determine when Active Surveillance (AS) is a safe strategy for a given patient, and 2) what is the best protocol for monitoring patients on AS.
An important aspect of the dialogue concerns the notion of clinically insignificant PCa. Basically, clinically insignificant PCa is defined as a tumor that is unlikely to become life threatening if left untreated. There is general agreement that low-grade, small volume, organ-confined PCa is clinically insignificant,[i] but differences exist regarding the exact factors such as PSA, Gleason grade, percent of tumor in biopsy needles, etc. by which insignificance is assessed. Nonetheless, patients found to have insignificant PCa are encouraged to consider going on AS as an alternative to immediate radical treatment; or they may be offered a partial or focal ablative therapy to obtain tumor control while reducing the risk of treatment-related side effects.
The challenge is how to monitor AS patients in such a way that a progression in their disease can be efficiently identified as a trigger for treatment. Typically, AS protocols include not just PSA blood tests but also repeat biopsies at specified intervals. TRUS biopsies are an imperfect method because of their blind, random nature. They cannot be counted on to rebiopsy in the precise location(s) of the original diagnostic biopsy. Furthermore, patients may become consciously or unconsciously resistant to complying with repeat biopsies.
Multiparametric MRI offers a promising way to rule out significant PCa and thus avoid repeat biopsies. With accurate imaging, patients gain confidence in the safety of embarking on AS and remaining on, without further biopsies, it until a future MRI might reveal a need for biopsy based on progression toward significance. Two recent papers explore the accuracy of multiparametric MRI (mpMRI) in detecting the absence of clinically significant PCa and correctly identifying significant disease.
For the first paper, an international team of experts conducted a thorough review of three publication databases (Pubmed, Embase, and Cochrane Central Register of Controlled Trials) for studies on the ability of mpMRI to detect or rule out clinically significant PCa.[ii] They surveyed the period from January 1, 2000 to September 30, 2014. Of the 1729 pertinent studies they found, 12 were selected based on high methodologic quality. Cumulatively, the median age of patients was 62-65, the median PSA was 5.1-13.4 ng/ml, and the Gleason scores ranged from 6-10. As expected, definitions of clinical significance varied somewhat, and were mostly based on PSA, Gleason grade, number of positive needle cores, and maximum cancer core length. The mpMRI detected clinically significant PCa in 44% to 87% of men who had either not had a biopsy or whose previous biopsy was negative (in all 12 studies, the reference standard for the imaging accuracy was either a subsequent biopsy or prostatectomy specimen). More importantly, the negative predictive value (correct prediction that there was no clinically significant cancer) was 63% to 98%. The authors wrote, “The negative predictive value of mpMRI is important to the clinician because mpMRI could be used to rule out significant disease.” The practical implications are the elimination of unnecessary biopsies, and the safety of AS as a clinical alternative to aggressive treatment.
The second paper was the result of research collaboration among authors from four Boston medical centers and a university hospital in Aachen, Germany. They compared mpMRI images with whole mount prostate specimens from 30 patients who then underwent prostatectomy.[iii] This study was not directly focused on clinically significant disease but rather on identifying and characterizing the index lesion (defined as the tumor, usually the largest if there is more than one, most likely to harbor an aggressive PCa cell capable of cloning itself). The goal of the study was to assess how accurately the imaging was in terms of tumor location and volume. Following surgery, the histopathology results (slicing, staining and examining the specimens under the microscope) were compared with the mpMRI images for consistency of index tumor localization, tumor volume, and other quantitative parameters from the MRI interpretations. The authors found a “strong correlation” between the histopathology tumor volume and that of the MRI, and accuracy of identifying tumor location.
Thus, these two recent papers add to the assurance that mpMRI can detect PCa tumors that are likely to become dangerous and, conversely, identify their absence as an important factor in avoiding an unnecessary biopsy or bolstering the choice of AS.
[ii] Fütterer JJ, Briganti A, De Visschere P et al. Can Clinically Significant Prostate Cancer Be Detected with Multiparametric Magnetic Resonance Imaging? A Systematic Review of the Literature. Eur Urol. 2015 Feb 2. pii: S0302-2838(15)00036-6. doi: 10.1016/j.eururo.2015.01.013. [Epub ahead of print]
[iii] Fedorov A, Penzkofer T, Hirsch M et al. The role of pathology correlation approach in prostate cancer index lesion detection and quantitative analysis with multiparametric MRI. Acad Radiol. 2015 May;22(5):548-55.