Comparative Effectiveness of Radical Prostatectomy vs EBRT Plus Brachytherapy in High-Risk Localized Prostate Cancer

[snoraste]

This is a study done on "healthy" adults. There are limitations to this study to be sure. It's a population study, and not addressing the advanced disease. Nonetheless, another data point to keep in mind.

practiceupdate.com/C/76714/...

"TAKE-HOME MESSAGE

The comparative efficacy of radical prostatectomy (RP) vs radiation (RT) for high-risk prostate cancer remains a source of debate. This study used data from the National Cancer Data Base to compare overall survival with external beam radiation therapy plus brachytherapy (EBRT + BT) versus RP in comparatively young (≤65 years) and healthy men with localized prostate cancer. Median follow-up was roughly 8 years, and nearly 14,000 men were studied. All patients had high-risk disease, and the study did not account for androgen-deprivation therapy. The patients treated with initial EBRT + BT had a higher risk of all-cause mortality (HR, 1.22) compared with those treated with initial RP.

These data indicate that young and healthy men presenting with high-risk, localized prostate cancer have better survival with RP compared with EBRT + BT, challenging prior research suggesting better or equivalent outcomes with radiotherapy."

[pjoshea13]

Virtually all of the study results I have read on RP versus radiation show better survival with RP. This is why I opted for RP 14 years ago.

The rebuttal, of course, is that RP remains RP, but radiation continues to be refined. "Median follow-up was roughly 8 years" then becomes a weakness (old data), rather than a strength.

From the full text [1]:

"Radical prostatectomy (RP) and external beam radiation therapy (EBRT) with androgen deprivation therapy (ADT) are standard of care options for high-risk localized prostate cancer (PCa) [1]. The ASCENDE-RT trial showed that brachytherapy (BT) as an adjunct to EBRT is associated with a recurrence-free survival benefit"

True. However:

"Although biochemical failure was associated with increased mortality and randomization to DE-EBRT doubled the rate of biochemical failure, no significant overall survival difference was observed between the treatment arms" [2]

Typically, RP does not increase the risk of death from non-RP causes. There is a significant downside to aggressive radiation.

-Patrick

[1] europeanurology.com/article...

[2] ncbi.nlm.nih.gov/pubmed/282...

[Fairwind]

I suppose the younger you are, the better your chances with RP over RT. Younger men will tolerate the surgery better and heal faster..RT at a younger age exposes the patient to a longer period for the radiation side-effects to manifest themselves..In older men where surgery becomes more risky, radiation might be the better choice as many will die from old age (or something else) before the radiation side-effects kick in....

[Dayatatime in reply to Fairwind]

Very sensible way to look at it. I feel men in their 40's and 50's having this disease are often left sorting out a wide array of statistical data that doesn't apply when researching. The data out there is limited because in the big picture there just isn't enough to target trials to that specific cohort of men. There really is a lot to consider when choosing treatment options considering the toxicity of today's aggressive treatment.

Ron

[Tall_Allen]

I've studied this closely. You can't use the NCDB to draw conclusions about comparisons. There is way too much selection bias. In general, men who get radiation are about 10 years older than men who get surgery. When they limit the age group, as they did in this flawed study, they are looking at a population of men who were too sick to undergo surgery for some reason. The NCDB lacks the data to correct for this. This is the kind of fatal flaw that statistics can't correct for (it violates basic assumptions of the statistical model they used).

For a discussion of the kind of flawed analysis you cited, see:

ascopubs.org/doi/10.1200/JC...

Also, see my discussion of another flawed study inappropriately using the NCDB by Ennis et al.

pcnrv.blogspot.com/2018/03/...

A much better analysis that challenged whether brachy boost is better than RP+ADT+SRT (called "Max RP") was the one discussed here:

pcnrv.blogspot.com/2018/11/...

[snoraste]

Allen,

I read R Chen's article you linked and he makes many valid points. The data problem clearly goes beyond just one database (NCDB), there is no centralized database in existence that has everything you need to perform a robust, unbiased analysis without some sort of assumptions (e.g., health and comorbidities). I would not use the term "inappropriate" for Ennis's approach. They appear to have used a robust statistical model and rational, consistent methods for their assumptions. Even R. Chen proposes that:

"..the main result of this study that surgery and RT have similar survival outcomes in high-risk prostate cancer is likely close to reality."

What I found more interesting in his paper was his claim on biases for RCTs, and that we may never have a truly unbiased RCT in this particular debate:

"it is worth noting that some important clinical questions may never be answerable by randomized trials. Prostate cancer randomized trials have been especially difficult to conduct because of large patient and physician biases, which often preclude treatment determined by randomization. In high-risk prostate cancer, it is possible that there will never be a completed randomized trial comparing surgery versus RT"

and then again:

"a common limitation of randomized trials is a lack of representativeness of enrolled patients; thus, trial results may not be generalizable to certain patient subgroups. Analyses of cancer registry data can provide insight into whether treatment benefits may apply to patients not included in the trials (eg, older patients) and thus play a complementary role to the trials."

I don't know if the institutions are doing a better job of collecting data nowadays. I hope they are.