Post prostatectomy case, impact of Decipher?

[communitydoc13]

Healthy 67 yo s/p RRP to negative margins for Gleason 4+3=7, pT3aN0 adenocarcinoma with tertiary pattern 5. PSA remains undetectable 8 months out.

Got a Decipher score. Comes back high risk (0.72).

Using MSKCC nomogram, if I wait till PSA gets to 0.2 and pick moderate doubling time, 6 year biochemical salvage 55% w/o ADT and 75% with.

What would you do?

1. Rec adjuvant XRT now without ADT?
2. Rec adjuvant XRT now with ADT?
3. Counsel that either adjuvant or salvage appropriate but would rec concurrent ADT in either circumstance?
3. Counsel that either appropriate but would forego concurrent ADT if treated adjuvantly vs. recommending ADT if waiting for salvage?

---

Members don't see this ad.

 

[Chartreuse Wombat]

Early salvage (two detectable levels, not necessarily 0.2) plus STADT.

 

[communitydoc13]

Early salvage (two detectable levels, not necessarily 0.2) plus STADT.

I like this answer. This is also the "you shouldn't have ordered the Decipher in the first place" answer.

Is anyone convinced based on this:

https://ascopubs.org/doi/abs/10.1200/jco.2014.59.0026

that Decipher is meaningful and should drive adjuvant XRT without ADT in the high risk Decipher group?

Otherwise, maybe no more Decipher ordering in this setting.

 

[TheWallnerus]

I don’t know that anyone knows how to decipher Decipher yet. Not ready for prime time?

OTOH may start offering adjuvant instead of salvage RT for more prostate patients such as your example (which I guess would call T3 almost Gleason 9)…

https://ascopubs.org/doi/full/10.1200/JCO.20.03714

 

[ramsesthenice]

https://ascopubs.org/doi/full/10.1200/JCO.20.03714

Oh boy. I can see me using this in TB to start arguing for adjuvant therapy. "Hi urology friends. I want you to ignore any of the BS retrospective studies you have suggesting we should operate and do post-op RT for very high risk people. But we should use this retrospective study to start offering adjuvant radiotherapy to men with high risk disease and negative post-op PSA." Im sure there could be something to this but its probably not a good idea to ignore the fact reliance on retrospective studies has long had carried undo weight in prostate cancer management.

I actually find myself in the situation where urology seems more apt to use Decipher to guide radiation than I do. Not all of them, but some of our urologists are getting decipher or patients are coming for evaluation with decipher in hand from outside centers and our urologists are asking me to consider treating for high scores in the absence of a detectable PSA. At least adjuvant radiation is an acceptable option irrespective of decipher scoring. I don't love using an unproven technology (I believe it is proven to predict risk, but early intervention based on this scoring is not proven to improve outcomes) to guide treatment but I REALLY don't love using an unproven technology to do something outside of standard of care.

I personally see 2 possible management questions using Decipher. The first is can a high decipher score help better predict who may benefit from any RT. The second is can it help predict who may benefit from nodal RT and/or ADT. All possible. And currently unproven.

 

[communitydoc13]

I personally see 2 possible management questions using Decipher. The first is can a high decipher score help better predict who may benefit from any RT. The second is can it help predict who may benefit from nodal RT and/or ADT. All possible. And currently unproven.

The other complicating issue to me is that if adjuvant XRT is recommended, do you still get a benefit from short course ADT. This would be a good reason to recommend Decipher from a value standpoint.

Prospective study to define the clinical utility and benefit of Decipher testing in men following prostatectomy - PubMed

The use of GC substantially altered treatment decision-making, with a number needed to test of only 3. Implementing best practices to routinely recommend ART for genomic-high patients led to larger than expected improvements in early biochemical endpoints, without jeopardizing outcomes for...

pubmed.ncbi.nlm.nih.gov

Seems like short course ADT rarely recommended with a recommendation for adjuvant and not salvage therapy.

I have had referrals from academic places where this is how they treat their high risk post-prostatectomy patients. If Decipher high, adjuvant XRT without ADT.

 

[TheWallnerus]

Oh boy. I can see me using this in TB to start arguing for adjuvant therapy. "Hi urology friends. I want you to ignore any of the BS retrospective studies you have suggesting we should operate and do post-op RT for very high risk people. But we should use this retrospective study to start offering adjuvant radiotherapy to men with high risk disease and negative post-op PSA." Im sure there could be something to this but its probably not a good idea to ignore the fact reliance on retrospective studies has long had carried undo weight in prostate cancer management.

I think this analysis might get special dispensation for a few reasons. First, it's a big number of patients. Second, the time period is longer than any of the randomized trials. Third, adjuvant therapy was a standard of care for a while in CaP on the basis of randomized trials which had been done because everyone (urologists) "knew" that adjuvant didn't work. Yet even when those trials came out showing adjuvant was a SOC, adjuvant never really caught on. Thus, fourthly, there was still a fascination with adjuvant vs salvage and those trials came out after the classic German and SWOG adjuvant-is-good trials and showed no benefit of adjuvant over salvage. Fait accompli? No; fifthly, as well expounded upon in this recent JCO mega-retrospective analysis, there *could be* methodological flaws of "immortal time bias" in the randomized salvage vs adjuvant trials which did lead to some mathematically weird results...

It's this last point IMHO where there is room to make some arguments in TB. And sixthly, just from an "onco-logical" perspective, it still doesn't sit right that we should wait for a cancer to become more clinically apparent and attempt treatment then versus adjuvant treatment like we do in almost every other body site (as adjuvant seems to be best in almost every other body site).

 

[Palex80]

I would wait and salvage when PSA rises.
I am not a friend of STADT when performing early salvage, an OS benefit has not been demonstrated in modern trials.
bPFS-improvement by addition of STADT to salvage RT is a no-brainer (you lower the PSA for those who have tumor outside the prostatic fossa, without salvaging them).

I only give ADT in postoperative cases for:
a) pN1
b) PSA >0.8 ng/ml
c) macroscopic residual/recurrent tumor (on MRI or PSMA-PET-CT)

Actually, b) has become obsolete, since >90% of patients with a PSA >0.8 ng/ml will show "something" on PSMA-PET-CT

 

[communitydoc13]

And sixthly, just from an "onco-logical" perspective, it still doesn't sit right that we should wait for a cancer to become more clinically apparent and attempt treatment then versus adjuvant treatment like we do in almost every other body site (as adjuvant seems to be best in almost every other body site).

In CaP salvage caught on for 2 reasons IMO. First, we have PSA, which allows us to designate failure prior to emergence of radiographically evident disease. Second, natural hx of CaP combined with that 7 year less life expectancy for men makes competing risks much more of a big deal and every patient that you avoid meaningless treatment in represents a therapeutic win.

It's clear that you can wait too long and that early salvage better than late salvage. Likewise, on an individual level, adjuvant treatment is never likely to provide worse cancer control outcomes than salvage unless you also de-escalate therapy (like not include ADT).

So to me, the possible benefits of adjuvant RT are 3-fold.
1. Possibly better outcomes due to lowest burden of disease at tx.
2. Possible therapeutic de-escalation (no ADT) (acknowledging that you escalated tx for every pt you could have observed)
3. Relief of anxiety.

bPFS-improvement by addition of STADT to salvage RT is a no-brainer (you lower the PSA for those who have tumor outside the prostatic fossa, without salvaging them

Really? Something else is going on here, I'm pretty sure. These biochemical control rates are at 6 years. You're telling me that 4-6 mos of ADT functionally suppresses biochemical progression of low volume disease over a 6 year period?

Even if the case, is this so bad? Sort of like saying, "Intermittent ADT is going to be much more intermittent over the next 10 years if we give you a little bit now".

The numbers speak for themselves. In this case, nomograms predict a 20% lower chance of indefinite ADT starting within the next 6 years if he takes a short course now.

Back to Decipher. Sound's like I don't know how to use it. If I had gotten a low score, it could have provided reassurance (which has some value) but since I got a high score, I am am just providing anxiety. Seems like not enough data to say treatment necessary now or that the potential biochemical benefit of short course ADT is eliminated relative to very early salvage.

Seems like fair number of academics full in on Decipher. Integrated into NRG trials now.

 

[ramsesthenice]

I think this analysis might get special dispensation for a few reasons. First, it's a big number of patients. Second, the time period is longer than any of the randomized trials. Third, adjuvant therapy was a standard of care for a while in CaP on the basis of randomized trials which had been done because everyone (urologists) "knew" that adjuvant didn't work. Yet even when those trials came out showing adjuvant was a SOC, adjuvant never really caught on. Thus, fourthly, there was still a fascination with adjuvant vs salvage and those trials came out after the classic German and SWOG adjuvant-is-good trials and showed no benefit of adjuvant over salvage. Fait accompli? No; fifthly, as well expounded upon in this recent JCO mega-retrospective analysis, there *could be* methodological flaws of "immortal time bias" in the randomized salvage vs adjuvant trials which did lead to some mathematically weird results...

It's this last point IMHO where there is room to make some arguments in TB. And sixthly, just from an "onco-logical" perspective, it still doesn't sit right that we should wait for a cancer to become more clinically apparent and attempt treatment then versus adjuvant treatment like we do in almost every other body site (as adjuvant seems to be best in almost every other body site).

The SWOG trial is probably one of the worst designed trials we have. I don't know how the word adjuvant was used back when the trial was conceived and I was but a wee college student but for anyone who missed it the following statement is simply not compatible with an adjuvant trial as we now use the term: "An undetectable PSA after radical prostatectomy was not required". A third of the patients would now be considered salvage on that trial. The "best" adjuvant trial we have (in terms of outcomes) isn't really an adjuvant trial.

Now Walrus, you made a couple of very key points. Of course adjuvant works in high risk patients. The question is whether it works well enough. Prostate cancer has a slow natural history and ADT is a better temporizing measure than salvage chemo for pretty much any other solid tumor. As a result it will take a long time and a lot of patients to show that adjuvant therapy is beneficial with respect to meaningful (ie non-biochemical) endpoints. By looking at lots of specifically high-risk patients they are probably getting closer to the information we really want to know. Another strength you have to give them is that their prostate cancer specific mortality was high enough it is believable that radiation at least has the potential to realistically improve survival.

 

[TheWallnerus]

The SWOG trial is probably one of the worst designed trials we have. I don't know how the word adjuvant was used back when the trial was conceived and I was but a wee college student

When it appeared, it established adjuvant RT as the standard of care in T3/4 prostate cancer post-RP. It was NCCN level 1 evidence to do adjuvant, not salvage, for a period of time (citation needed). It was the lead plenary session presentation at ASTRO before it went on to get published and was hailed loudly and widely. And at the time I don't recall any rad onc nitpickers. That said, as mentioned, the urologists (aka Those Who Give Us The Patients) were unswayed and IRL practice patterns didn't really change.

Adjuvant RT we hardly knew ye

 

[Chartreuse Wombat]

The figure below is from Felix Feng's paper at the link

Validation of a 22-Gene Genomic Classifier in Recurrent Prostate Cancer

This ancillary study of the NRG/RTOG 9601 randomized clinical trial assesses the validity of a genomic classifier to estimate the risk of distant metastasis after radical prostatectomy in patients with prostate cancer.

jamanetwork.com

I do realize that the ADT was 2 years bicalutamide (not 6 months LHRH) but this secondary analysis suggests that Decipher can be used to determine the magnitude of the benefit of ADT added to early salvage RT.

The problem is that using PSA level above or below a cutpoint gives similar results and PSA is a lot cheaper than Decipher but it could help us with very early salvage.

 

[ramsesthenice]

When it appeared, it established adjuvant RT as the standard of care in T3/4 prostate cancer post-RP. It was NCCN level 1 evidence to do adjuvant, not salvage, for a period of time (citation needed). It was the lead plenary session presentation at ASTRO before it went on to get published and was hailed loudly and widely. And at the time I don't recall any rad onc nitpickers. That said, as mentioned, the urologists (aka Those Who Give Us The Patients) were unswayed and IRL practice patterns didn't really change.

Adjuvant RT we hardly knew ye

Don’t get me wrong. SWOG is a critical study that showed adjuvant/early salvage is far superior to haphazard post op monitoring. From that perspective it was, is, and always will be practice changing. Post-op RT can improve survival. It just isn’t as helpful for how we define the term adjuvant anymore.

 

[evilbooyaa]

I like this answer. This is also the "you shouldn't have ordered the Decipher in the first place" answer.

Is anyone convinced based on this:

https://ascopubs.org/doi/abs/10.1200/jco.2014.59.0026

that Decipher is meaningful and should drive adjuvant XRT without ADT in the high risk Decipher group?

Otherwise, maybe no more Decipher ordering in this setting.

I'm kinda still at the "you shouldn't have ordered Decipher in the first place" answer.

The link you posted did fossa alone RT in all scenarios, which we now know for anything but the lowest PSAs is probably not the best idea (meaning some percentage of those distant metastasis failures after salvage RT might've been saved with nodal RT).

That being said, I wouldn't fault somebody, who ordered a Decipher, to use a high risk score as an indication for adjuvant RT. Just don't think it's mandatory to do so at this time. Early vs Late adopter of a new technology.

I don’t know that anyone knows how to decipher Decipher yet. Not ready for prime time?

OTOH may start offering adjuvant instead of salvage RT for more prostate patients such as your example (which I guess would call T3 almost Gleason 9)…

https://ascopubs.org/doi/full/10.1200/JCO.20.03714

I definitely think there is some rationale to repeating RADICALS/RAVES but with actual high risk disease that could have a DMFS benefit from treating microscopic disease before it mets out, rather than most of it being Gleason 7 patients.

Otherwise, agree with most of the discussion. Prostate cancer is an exception to most when it comes to adjuvant because we have this magical blood test, PSA, which can tell us when there is some level of microscopic prostate cells hanging around.

 

[Palex80]

Really? Something else is going on here, I'm pretty sure. These biochemical control rates are at 6 years. You're telling me that 4-6 mos of ADT functionally suppresses biochemical progression of low volume disease over a 6 year period?

Even if the case, is this so bad? Sort of like saying, "Intermittent ADT is going to be much more intermittent over the next 10 years if we give you a little bit now".

I think giving STADT simply prolongs the time until you need ADT "down the road". Perhaps STADT can kill more cells in the prostatic fossa than SRT alone, but apart from that, we do not have good evidence than STADT actually prevents metastatic spread, it may very well simply delay it.

The numbers speak for themselves. In this case, nomograms predict a 20% lower chance of indefinite ADT starting within the next 6 years if he takes a short course now.

Well, if you and the patient believe this is a good endpoint, then you can give STADT. I am just not convinced.

In my opinion, the benefit of STADT in terms of bPFS mainly comes from treating (not salvaging!) low-volume metastatic disease outside the prostatic fossa, simply by delaying the appearance of metastatic disease. And with the incorporation of better imaging (mainly PSMA-PET-CT), that benefit is likely to become smaller over the coming years.
STADT has yet to prove a true survival benefit in patients receiving salvage radiotherapy and in fact its role in the primary setting for intermediate-risk disease is also questionable in the era of dose-escalated RT.

 

[DoctwoB]

My word! What have we here? Radiation oncologists setting aside 2 beautiful RCTs to consider treating based on their priors and retrospective data? Oh the Humanity!

-Back to the original point. Don't check decipher. Follow PSA. Treat with early salvage if/when indicated. Even if ART is better then SRT in some settings, likely the NNT is very high. There is also a significant QOL advantage in delayed XRT vs. post op XRT (paper in european urology this month).

 

[DoctwoB]

I think this analysis might get special dispensation for a few reasons. First, it's a big number of patients. Second, the time period is longer than any of the randomized trials. Third, adjuvant therapy was a standard of care for a while in CaP on the basis of randomized trials which had been done because everyone (urologists) "knew" that adjuvant didn't work. Yet even when those trials came out showing adjuvant was a SOC, adjuvant never really caught on. Thus, fourthly, there was still a fascination with adjuvant vs salvage and those trials came out after the classic German and SWOG adjuvant-is-good trials and showed no benefit of adjuvant over salvage. Fait accompli? No; fifthly, as well expounded upon in this recent JCO mega-retrospective analysis, there *could be* methodological flaws of "immortal time bias" in the randomized salvage vs adjuvant trials which did lead to some mathematically weird results...

It's this last point IMHO where there is room to make some arguments in TB. And sixthly, just from an "onco-logical" perspective, it still doesn't sit right that we should wait for a cancer to become more clinically apparent and attempt treatment then versus adjuvant treatment like we do in almost every other body site (as adjuvant seems to be best in almost every other body site).

The immortal time bias issue is certainly possible, though unlikely to play a major role here IMO. By the way, this bias is a key component of very trial involving ADT or comparing radiation + ADT with surgery. When you give ADT, you are delaying BCR by a certain amount regardless of long term treatment success.

Their insistance that the HR showing SRT being superior to HRT shows there is bias is likely overstated though. If SRT and HRT are exactly equal, there's a 50% chance you'll see SRT looking better in the trial. That's a much more likely explanation imo.