Ran across this article today while looking for something else. It's reasonably current, Nov 2016.
Application of a Prognostic Gleason Grade Grouping System to Assess Distant Prostate Cancer OutcomesIt was published in European Association of Urology. By analyzing the CaPSURE registry, it explores the prognostic Gleason Grade Group system for various outcomes, by various treatments, and so on. There's a lot of information in it. By its nature it is retrospective, so there could be various biases built into it. There are probably better studies for treatment choices and so on, but I thought it interesting for what it is.
It's actually kind of encouraging to see the relatively high survival rates out to 15 years for most types of prostate cancer. (The median follow up was 81 months, so half shorter and half longer. Hard to say how that affects those long-term graphs; maybe one of the smart folks here could clarify that.)
Of note, it mentions the lack of cases in Grade Group 5, which is G9-10. There are always statistical limitations in studies due to the lack of G9-10 cases. Probably fortunately, there just aren't many of us in that range. Out of a total 10,529 men in this analysis, there were only 389 in Grade Group 5! Even further, in Figure 2 they split the G9 5+4 (81 cases) and the G10 (48 cases) from the rest.
That's why we always get lumped into studies with other categories. If you have only 130 cases out of over 10,000 with Primary Gleason Type 5, it becomes a bit tough to make many conclusions.
Finally, I'm not sure how much the Grade Groups system is being used. I don't hear the terms very often. It was an effort to change the perception of risk, particularly for the lower grades that could be handled simply with active surveillance.
Anyway, this is just presented for your weekend reading pleasure!