SIU 2017: Active Surveillance: Biopsy vs MRI for Follow Up

Lisbon, Portugal ( Dr. Villers followed in this session on localized prostate cancer with a talk assessing the role of biopsy vs. MRI in active surveillance follow-up. Ideally, in AS, the goal is to eventually replace invasive biopsies with noninvasive testing. While prostate MRI had lots of initial promise, it does not appear that it can yet safely replace prostate biopsy.

In this talk, Dr. Villers specifically asks whether we can avoid confirmatory biopsy in men who are being considered for active surveillance. Towards this effort, he cites data that the MRI has a negative predictive value (NPV) of approximately 95% at his center for intermediate or high-risk disease. As the risk of stage progression is <5%  in the first year, he questions the need for confirmatory biopsy at 1 year.

He reports a preliminary study. Two groups were compared, wherein one group receiving a confirmatory biopsy and the other one did not. Primary objective was curative rates. Secondary objectives were: 2-year tumor progression rate, PSA velocity (PSAV) and PSA doubling time (PSADT), and histologic results at prostatectomy (RP).

Retrospective series between 2007 and 2016 (9 years). 138 consecutive patients treated for AS – all with a non suspicious MRI at entry. Group 1 (78 patients) – treated with per protocol with MRI and confirmatory biopsy at 1 year. Group 2 (60 patients) did not undergo biopsy at 1 year. Both groups were allowed to undergo MRI + biopsies for specific triggers – PSADT < 3 years or PSAV >= 0.75/ng/mL/year. Progression was defined as visible tumor at MRI, Gleason > 3+3, >3 cores positive. All biopsies were transrectal systematic biopsies + targeted biopsies if needed.

Median follow-up was similar in both groups: 28 – 35 months. In terms of outcomes: 25% of Group 1 and 18% Group 2 were off AS at the end of follow-up. 21% tumor progression in Group 1, but only 11% in Group 2. No significant difference in treatment free survival (KM curve). Again, no difference in the 2-year progression free survival (KM curve). 

In the group with confirmatory biopsy, looking at PSA changes, PSAV >- 0.75 ng/mL/year accurately picked up progression (specifically size and Gleason score progression) – AUC 0.92. PSADT was a little less effective – AUC 0.83.


Obviously, this study had significant limitations. There is no gold standard to identify progression in the patients who never had confirmatory biopsy. Their lower rates of progression are because they were never tested for it.

The follow-up is also much too short to pick up any real differences in cancer outcomes. Anecdotally (brought up as a question) – MRI still misses ~20% of high grade cancers.  EAU guidelines panel – NPV was a function of the underlying risk. Could be as low as 65%. NPV may be a lot lower in the general population.

Prospective studies are required. 

Presented by: Arnauld Villers

Written by: Thenappan Chandrasekar, MD, Clinical Fellow, University of Toronto, Twitter: @tchandra_uromd at the 37th Congress of Société Internationale d’Urologie - October 19-22, 2017- Lisbon, Portugal
email news signup