Finally a definitive study.
First of all, this is not a study. It's a paper published with analysis of the results of other studies that have been previously completed...that's called a "meta-analysis." Not a study. Did you read the paper (or just the brief article about
the paper which you linked to)? And, do you understand the major shortcomings which were identified by the authors in the paper?
In my view, the three most important shortcomings were (i) differences in ages of the RP vs RT patients, (ii) age of the original studies and (iii) absence of the isolation of salvage therapies.
(i) It's no myth that radiation is more widely used by older men. Treatment for each and every case should be individualized, and the rigors of surgery can be inappropriately risky for some men as they age and develop other co-morbidities...leading to RT being increasingly common as men age. But the bottom line for the discussion at-hand, in each and every one of the studies analyzed the median age of the RT patients is older than the surgery patients...sometimes by quite a few years. This is going to have an influence on mortality.
(ii) As sheepguy mentioned, RT procedures have improved over time...but so have surgical techniques, so that in of itself is not necessarily significant. However, the RT studies analyzed included brachytherapy, EBRT and IMRT, but most studied the efficacy of EBRT, and today IMRT has largely replaced EBRT, and brachytherapy is on the decline...so certainly the meta-analysis results do not necessarily reflect the latest and greatest.
(iii) Lastly, the results do not include any isolation of salvage therapies. The studies all focused on primary treatments, but men who have failed surgery typically follow-up with SRT, and then systemic therapies, and follow-up for failed RT goes typically directly to a systemic therapy. No accounting for these even though some of the studies included--and others excluded--men with salvage therapy; even though the sole objective of salvage treatments is to extend life. Confounded results.
The most important take-away is the solidification of the understanding that no two cases are the same, and each require individual assessment of the patient's case characteristics before deciding treatment mode, or non-treatment, pros and cons. The analysis did break down the HR for different risk categories, but 80% of the men were in the low-risk category...and today most of those men and many men with intermediate-risk cases would follow an Active Surveillance protocol rather than immediate aggressive treatment.
While this report is NOT "definitive," there is a study in process now that will be quite definitive for men with clinically localized PC: the ProtecT trial. In fact, even the report on this meta-analysis specifically call this out, stating: "Implications for future research assessing the comparative efficacy of surgery and radiotherapy in prostate cancer will largely depend on the results of the upcoming randomized ProtecT trial."
Data from ProcecT should be available in 2016, where consenting participants were randomly assigned to AS, RP or RT, with the primary endpoint of PC mortality at a median 10-year follow-up.
This brings me to address sheepguy's questions on hazard ratio. First of all, HR is the standard metric for this type of study and it stands alone. HR is an absolute measure of the difference in hazard rates between two groups, but the details sheepguy asks of--the % mortality for each study from which the HR is generated--are in the report...I'll leave it to you to actually read the report instead of having to "wonder
Post Edited (JackH) : 12/24/2015 12:35:23 PM (GMT-7)