THE "NEW" PROSTATE CANCER INFOLINK

Monique Roobol and her colleagues continue to mine data from the ERSPC and other major trials in a serious attempt to describe processes that would allow us to identify the vast majority of clinically significant prostate cancers while avoiding the unnecessary diagnosis and over-treatment of indolent disease.

In a presentation at the recent European Association of Urology annual meeting in Vienna (as reported by Evans on UroToday), Roobol et al. used data from the ERSPC trial and additional data from the REDUCE trial (of dutasteride in the chemoprevention of prostate cancer) to create a model of the potential of combining screening and chemoprevention to manage the diagnosis of clinically significant prostate cancer. Readers need to be very clear that this is only a model, and that the model has not been actually tested in the real world.

The theoretical structure of the proposed model was a follows:

• Men would get an initial PSA at age 50 (or perhaps earlier).
• Men with a PSA level of 2.5 to 10.0 ng/ml who were negative for prostate cancer on their initial biopsy would be treated with dutasteride.
• This group of men would then get another PSA test each year and their PSA would be “normalized” based on the idea that dutasteride therapy reduces PSA levels by 40 to 50 percent compared to normal levels.
• Men with a PSA level that was > 40, or > 45, or > 50 percent of the original PSA level (pre dutasteride therapy) would then be re-biopsied at 4 years

Roobol et al. then applied this hypothetical model to data from a cohort of 1,331 participants in the ERSPC trial who had an initial PSA level of between 2.5 and 20.0 ng/ml, had an initial negative biopsy, and who were actually re-biopsied at 4 years. (Remember that the REDUCE trial showed that dutasteride reduced the probability of prostate cancer by 22.8 percent in men compared to a placebo.) The model demonstrated that:

• Using the criterion of a PSA level > 50 percent of the original PSA
• 22/1,331 men would have qualified for re-biopsy at 4 years.
• The positive predictive value of re-biopsy would have been 100 percent.
• 1,309 biopsies would have been saved.
• Re-biopsy would have identified no cases of Gleason 6 disease, all 17 cases of Gleason 7 disease, 5/8 (63 percent) of the cases of Gleason 8 to 10 disease.
• Using the criterion of a PSA level > 45 percent of the original PSA
• 25/1,331 men would have qualified for re-biopsy at 4 years.
• The positive predictive value of re-biopsy would have been 100 percent.
• 1,306 biopsies would have been saved.
• All cases of prostate cancer with a Gleason score of 7 or higher would have been identified.
• Using the criterion of a PSA level > 40 percent of the original PSA
• The positive predictive value of re-biopsy would have been 57.7 percent.
• All cases of prostate cancer with a Gleason score of 7 or higher would have been identified.

The authors conclude that their theoretical model might decrease unwanted side effects (i.e., excess biopsies and consequent over-treatment) of purely PSA-based screening.

Now this is an interesting hypothetical model, but whether it is actually practical in the “real world” is a whole different question. Dutasteride is certainly not going to get approved for chemoprevention of prostate cancer in the USA or in the major European countries on the basis of the REDUCE data, and it seems unlikely that anyone is going to replicate the ERSPC trial based on a hypothetical model like this. So without such data, the hypothesis being put forward by Roobol and her colleagues is likely to remain just that — an hypothesis. On the other hand, however, the concept outlined does offer a potential model for a combined screening/chemoprevention trial if the right chemopreventive agent were to come available.

Filed under: Diagnosis, Risk | Tagged: biopsy, chemoprevention, combination, Diagnosis, dutasteride, risk, screening |