SUO 2017: From Biopsy, AS vs Rx: Stratifying Risk in Localized PCa

Washington, DC ( Dr. Lin explored the broad landscape of management for low-risk (LR) prostate cancer (PCa) patients in the current era. As opposed to 1-2 decades ago, the new approach to LR PCa management is to avoid overtreatment. The use of active surveillance (AS) has grown widely over the past decade, as has been shown in recent large publications including the Michigan MUSIC cohort. AS has now grown to be used in over 50% of low-risk patients nationally. 

This shift in understanding has been reflected in the AUA guidelines, which defines AS as the most appropriate primary option for men with LR PCa. Given the widespread adoption, many clinicians rightfully worry that under-treatment may now be a risk in some LR PCa men. Hence, strategies involving biomarkers and newer imaging modalities have come under increased scrutiny as a way to better risk-stratify patients in this cohort.

To date, biomarkers have not been endorsed by any society guidelines for this space. There are several reasons for this hesitation. There are no validated studies for biomarkers in the AS space; all biomarkers have so far been tested within treatment paradigms and are not easily translated for use in AS patients. Furthermore, LR PCa patients tend to live for a long time, regardless if treatment or AS is chosen. This has been shown in multiple large prospective cohorts including PIVOT, ProtecT, etc… Making a measurable impact on outcomes in a patient population that already does very well is understandably difficult.

The largest AS trials to date (from Johns Hopkins, UCSF, University of Toronto, Canary PASS) have shown that although AS is safe, at least 50% of patients are reclassified to more aggressive pathology between 5-10 years after initiating AS. It is clearly important to better categorize patients who are at risk for reclassification and may need future treatment, as this could dramatically change the discussion regarding treatment options. 

Biomarkers hold great potential in helping to risk-stratify these patients, but more data is needed. Dr. Lin revisited what we would like in an ideal biomarker: One that is highly sensitive/specific, correlates well with disease outcome, is low cost, and is quick & easy. 

Three commercial biomarker tests are currently available and are widely used. Prolaris is a 31 gene assay used on biopsy specimens with an endpoint of disease-specific mortality and disease aggressiveness. Oncotype DX is a RNA transcript profile of 17 genes whose endpoint is adverse pathology at treatment. Decipher is the most widely-used assay that utilizes a 22 gene multi-pathway signature from biopsy tissue whose endpoints include high-grade disease and risk of metastasis. Additionally, it is validated as a predictor of BCR, metastasis, and PCa-specific-mortality after radical prostatectomy.

As we move forward with validating these, and other, tests in the AS population, it may be important to redefine the important endpoints of biomarker effectiveness. Dr. Lin argues that more appropriate endpoints would include outcomes such as likelihood of remaining on AS and likelihood of BCR after initial treatment choice. 

Additionally, the quality of imaging with multiparametric MRI is rapidly improving. However, MRI suffers from high interobserver variability; and for men on AS, studies show that up to 30% of progression/recurrences are found on TRUS biopsy and are missed by MRI. Due to this limitation in sensitivity, most guidelines warn against using MRI to replace standard prostate biopsy in the initial evaluation of patients. Combinations of imaging and biomarkers may be the key to improving the test characteristics of both modalities in the future.

I would argue that biomarkers are probably most effectively used in LR PCa patients with a higher probability of reclassification (e.g. those with higher PSA density or greater extent of cancer). However, there remains the need to redefine appropriate endpoints, rigorously test biomarkers under AS conditions, and to incorporate biomarkers into validated multinomial nomograms. 

Presented by: Daniel W. Lin, MD. Chief of Urologic Oncology at the University of Washington Department of Urology.  Urologist specializing in genitourinary oncologic surgery.

Written by: Shreyas Joshi, MD, @ssjoshimd Fox Chase Cancer Center, Philadelphia, PA at the 18th Annual Meeting of the Society of Urologic Oncology, November 20-December 1, 2017 – Washington, DC
email news signup