In our quest to understand our prostate cancer we have developed many “measures” of our disease aggressiveness and progression. Many of these measures are generally accepted has valid, but many have never been validated.
One example is the prognostic nature of PSA Doubling Time (PSADT) in men with biochemically recurrent prostate cancer (BRPC or PSA only increases with negative scans). PSADT is used as common intermediate endpoint for assessing the efficacy of therapies. A number of investigational agents appear to favorably effect PSADT, slowing of PSADT has also been observed in placebo-treated men.
Researchers from Johns Hopkins Hospital theorized that PSADT calculations could be influenced by the frequency and duration of PSA measurements and not the therapies.
They performed a retrospective analysis of men with BRPC who chose to not use hormonal therapy (ADT). They evaluated men with PSA values greater than or equal t0 0.2 ng/mL and at least 6 values taken on average 3 mo apart, and whose local prostate cancer therapy was completed greater than or equal to 1 year prior.
To examine the actual influence of PSA testing frequency on the duration on PSAD they calculated median PSADT using different subsets of available PSA values (e.g. each vs every other; and first 3 vs. remaining PSA values).
They found that after a median follow-up of 58 months (range, 6-185 months), 213 men with BRPC had greater than or equal to 6 PSA values and 127 men had great than or equal to greater than or equal to 9 PSA values for analysis.
The men in the study (77% white, 23% black/other) had a median age of 61 years with Gleason score distribution as follows (greater than or equal to 6: 30%; 7: 40%; less than or equal to 8: 18%; NOS: 12%).
For men with greater than or equal to 6 data points: PSADT calculated using earlier values ranged from 13.2 to 16.6 months, compared to 16.6 to 17.8 months, respectively, for the remaining values (within-patient change range: 0.6-1.2 month).
For men with greater than or equal to 9 data points: PSADT calculated using earlier values ranged from 15.3 to 19.9 months compared to 22.1 to 22.3 months, respectively, for the remaining values (within-patient change range: 3.5 to 4.5 mo).
When they examined the frequency of PSADT by using every other value, we found little difference (22.3 vs 22.1 month).
These data confirmed that PSADT appears to increase even in the absence of any type of therapy, and may in fact be influenced by duration of PSA follow up. This finding explains PSADT slowing on placebo arms and calls into question the utility of PSADT as a surrogate endpoint.
The researchers concluded that placebo-controlled trials using the standard clinical endpoints are recommended to screen agents in men with BRPC to mitigate bias because of natural PSADT variability.
This finding calls into question many of the supplements that men with BRPC tend to use because they rely on having seen a positive (or lengthening) of their PSADT. This study also calls into question the use of PSADT for any man with prostate cancer despite their actual disease stage.
J Clin Oncol 30, 2012 (suppl; abstr 4559); Channing Judith Paller, Sherlly Xie, Dare Olatoye, Samuel R. Denmeade, Mario A. Eisenberger, Emmanuel S. Antonarakis, Michael Anthony Carducci, Gary L. Rosner;
A Side Bar: Dr. Paller, the primary author of this paper, has recently received funding from our cancer crowd sourcing funding project, Start A Cure (www.StartACure.com). Start A Cure provides each of us an opportunity to pick research that we believe is important and make an economic contribution. Please go to the site and support the researchers who are making a difference to us.
Joel T. Nowak, M.A., M.S.W.