Adding ADT to external beam radiation only benefits unfavorable risk patients

[Tall_Allen]

"Favorable intermediate risk" patients are sometimes overtreated by doctors who give them adjuvant ADT. This proves it does them no good.

prostatecancer.news/2020/09...

[Thetoad]

Thanks. As I've messaged you in the past, I'm Favourable Intermediate with no hormone treatment, in fact I have Sustanon injections to keep me at a normal level. Which may or may not be unusual. PSA has dropped slowly after the initial large drop following rt. Next test in a week and of course I'm getting anxious. I asked about bounces in another post and whether they can occur about 20 months after rt or maybe later is more common? Or any time. Thanks for all your help on this site!

[Tall_Allen in reply to Thetoad]

Here's what I know about bounces:

prostatecancer.news/2018/03...

[Thetoad in reply to Thetoad]

It didn't bounce, in fact it dropped from 2.7 to 1.5. A much bigger drop than the two previous ones of .2 each time. A six month wait again then maybe yearly? We shall see ...

[cesanon]

Sort of interesting.

The studiy's conclustion it nonintuitive.

But once you think about it, it becomes logical.

If you are favorable risk, then the radiation treatment works just fine.

If you are unfavorable risk, maybe the radiation treatment isn't quite enough to do the job, and you need something more.

[Jmr11820]

Thanks for sharing that. Just confirming, this is for primary treatment and and not salvage or early adjuvant RT, correct?

[Tall_Allen in reply to Jmr11820]

Correct. The NCCN risk stratification system is only used for primary treatment of localized PC.

[Stevecavill]

Always need to be careful assigning too much value to retrospective studies.

[Tall_Allen]

This wasn't retrospective. It was a secondary analysis of a randomized clinical trial.

[Stevecavill]

By definition, an unplanned analysis is retrospective. As the authors themselves noted “Multiple limitations of this study warrant discussion, including that it is an unplanned secondary analysis“. The major problem with retrospective studies is selection bias, because the stratification variable was not included in the randomisation.

It doesn’t mean the results aren’t useful, but they aren’t as reliable as a planned prospective study.

[Tall_Allen in reply to Stevecavill]

No. All analyses are done after the data are in. "Retrospective" as used in research refers to a post hoc analysis of a group of patients who are not randomly assigned to one group or another. For example, an analysis of the SEER database is always retrospective, and is subject to huge selection biases. In this case, all intermediate risk patients were prospectively randomly assigned to ADT or no ADT, and the two groups had similar patient characteristics. Since the entire group was randomly assigned, a moment's thought shows that selection bias within those two subgroups (FIR and UIR) was impossible.

In fact, what the authors correctly write is this:

"Although previous studies have also suggested this, they have been retrospective studies limited by selection bias in usage of ADT and short follow-up. Our study, using a large multi-institutional cohort from a cooperative group trial with randomized ADT use and nearly 18 years of follow-up, overcomes these limitations."

It is an unplanned secondary analysis. The problem is not selection bias. One problem can be that the sample size is not powered for the subgroups. In this case, with biopsy data on 377 FIR patients and 513 UIR patients, and 15 years of f/u, that does not seem to be a problem. The other problem with unplanned secondary analyses is that the authors may be hunting for those variables that are statistically significant (sometimes called "p hacking"), and those variables may only randomly be significant. But the authors are not on a hunting expedition here. They purposely only looked at a single stratification.

[doc1947g]

I am 73 y.o. with PSA = 20,40 μg/L. PCa G(4+3=7) Grade 3(4). My RO treated me as a Grade 4, but due to my fantastic response to ADT, he stopped it after 6 monts.

I am one of those UIR patient and I did received VMAT(20X3Gy) and ADT(Lupron Depot 22.5mr/12 weeks X 2.

I do not feel over-treated.

Now with a PSA = 0,03 μg/L and T < 0.2 nmol/L or < 5.7684 ng/dL.

God bless you.

[doc1947g]

Even with my second PSA test = PSA = 0.03 and with my prostate.

My RO was impressed with that result which usualy belong to post-RRP.