Dan Spratt: Thanks so much, Jeff. It's a great pleasure to be on with you. And yeah, I'm excited to present the results from a trial you were a part of, NRG GU006. The NRG GU006 or the BALANCE trial is a double blinded, placebo-controlled biomarker stratified randomized trial of apalutamide and radiation for recurrent prostate cancer. These are disclosures.
And so, as many of you know, surgery or radical prostatectomies are treatment options for appropriately selected men with localized prostate cancer. But recurrence rates are variable, but they do range up to 80% in clinical trials for men with high-risk disease, and we now prefer the use of what's called early secondary. It used to be called salvage radiation for these patients.
Now, one other way that we can improve outcomes is through inhibiting androgen receptor signaling through hormone therapy, but this treatment has side effects that, in men, often drive a decline in their quality of life. And we don't have any prospectively validated predictive biomarkers yet to guide hormone therapy use. And while there are variable benefits of hormone therapy with postoperative radiation, it doesn't seem when given early at a low PSA that this seems to improve overall survival. And this was shown recently in this meta-analysis.
So, back in about a decade ago, myself, Felix Feng and George Zhao, shown here, developed a modification of the breast cancer PAM50 signature and adapted to prostate cancer. And we showed retrospectively that the luminal B patients seemed to drive greater benefit from hormone therapy as shown on the left, whereas the non-luminal B or basal patients did not seem to derive that benefit.
So, we designed this trial with that hypothesis in mind that the luminal B patients will drive greater benefit from apalutamide, which is a potent oral inhibitor of the androgen receptor in men with a recurrent prostate cancer receiving secondary radiotherapy.
So, we enrolled men post-surgery with detectable PSA is between 0.1 and 1 with no evidence of metastatic disease. They had prospective biomarker testing and were stratified by their margin status, their pre-radiotherapy PSA, and the PAM50 subtype, and were randomized in a double-blinded fashion to secondary radiotherapy to six months, both of an oral placebo or apalutamide.
So, the design here is a biomarker stratified design as we first evaluate the benefit in biomarker positive or luminal B patients, and if there is benefit, then we would move on to test in the biomarker negative patients or the non-luminal B patients. However, if there was no benefit, we would just pool everything together and analyze it like we would in normal trial.
The primary endpoint is biochemical progression-free survival, which is a composite endpoint of biochemical recurrence, local, regional, or distant failure or death, and some key secondary endpoints shown as well. Now, just in terms of some details on the statistical design, you can see these here, but it was well-powered for the luminal B subset, a one-sided log rank at a 0.012 alpha with 95% confidence intervals utilized.
So, we estimated based upon our predicted distribution of these subtypes, you can see the estimated enrollment numbers there. And so, this trial we're very fortunate, thanks to all the support and NRG Oncology on the member sites, it accrued very rapidly three times ahead of schedule. And we ended up accruing 295 patients. It was well-balanced and we're presenting results at a median follow-up of five years.
As intentional to keep this at kind of an early secondary radiotherapy trial, you can see about 85% of patients had a low PSA below 0.5. About 20% of patients had Gleason eight to 10, about half had positive surgical margins and half had pathological T3 disease. And molecularly about 42%, 43% of patients had luminal B. The second most common was basal at 38%. And a lot of patients had a high Decipher score, so about a little over half of patients had a high or very high Decipher score on this trial.
So, the primary endpoint is shown here for the luminal B patients, and apalutamide significantly improved biochemical progression-free survival by 18% at five years. And you can see the hazard ratio at 0.45 with a significant P value. And as predicted, non-luminal B patients did not derive a benefit. The five-year point estimate was essentially identical around 70%. The hazard ratio also at 0.95. And very importantly in prostate cancer is the endpoint term metastasis free survival, which is developing metastatic disease or death.
And again, at five years there was a 13% improvement in MFS with the use of apalutamide with a hazard ratio of 0.27. And the non-luminal B patients, these curves were overlapping again, similar five-year point estimates and a hazard ratio of 1.06. This was overall a well-tolerated treatment. We're just showing grade 3 or higher adverse events regardless of any attribution. But you can see there's probably about a 9% increase with use of apalutamide that seems to be driven mainly by rash, which is a known potential adverse event from apalutamide that's transient, and hypertension.
Fortunately, things like cardiac disorders and more serious things were very rare in both arms. And given the use of sort of more modern image-guided radiation therapy, you can see the radiation-related side effects are extremely low here. And so, overall, in summary, PAM50 represents actually the first and only prospectively validated predictive biomarker to guide the use of hormone therapy. We demonstrate that the most benefit seems to be in luminal B tumors, and this is included in metastasis-free survival from the addition of apalutamide to radiotherapy. And we did not identify or observe these benefits in non-luminal B tumors.
And so, six months of apalutamide with secondary radiotherapy was well tolerated and is an effective treatment option for these patients with recurrent prostate cancer after surgery. And lots of people to thank, all the patients' families, all the NRG participating sites and trial staff, the core committee, George Zhao, a big shout-out to Felix Feng who is the co-PI of this trial and our grants and sponsors. Thank you.
Jeff Michalski: Great, Dan. Thanks. I enjoyed the presentation as much this time as I did the first time in San Francisco. So, you mentioned in one of your early slides some of the trials that have used salvage radiation therapy with or without hormone therapy like RTOG 9601, 0534, the French trial and the RADICALS trial. So, specifically RTOG studies, I'm interested, how does this patient population differ from those? And why is that important in your conclusions?
Dan Spratt: Yeah, it's a great question. So, I think first, one of the first trials that have tested the use of hormone therapy in the setting was RTOG 9601. Obviously, the patients in the mid-1990s are very different than what we're seeing today. A lot of the patients, we gave what's called late salvage radiation. I think the median PSA was about 0.7, so it's much higher now and went up as high as 4.
And by today, as you know, Jeff, is with a PSMA PET scan, probably many of these may have nodal or even distant disease. And so, it makes sense in that trial why there was such a, I mean, very large survival benefit seen in those with high PSAs. When you move to more modern trials like RTOG 0534 or the French, the GETUG trial, the RADICALS trial.
Overall, I would say, they're generally similar in that most of the patients on these trials had lower PSAs like in RTOG GU006, most were Gleason 7, some had Gleason 8 to 10. A lot had T3 and positive margins. So, I would say, there's definitely a decent amount of overlap in these studies. And what was not shown in the presentation is that, if you were to pool both the luminal B and non-luminal B together, you do start seeing the same trends we observe in things like RTOG 0534 and the GETUG trial where you start seeing trends of some of these harder endpoints of benefit. But I think what we're seeing is that probably this is driven by that luminal B subset.
Jeff Michalski: A lot of people are going to ask, "Well, why did you pick apalutamide?" And going forward, can I assume that I can use just six months of ADT for those luminal B positive patients? And if that's okay, what type of ADT does it have to be a short-acting antiandrogen like apalutamide? Could it be a short-acting LHRH antagonist like Relugolix? Or can we use one month injection six times? So, kind of curious thoughts.
Dan Spratt: Yup. No, it's an excellent question, Jeff. Regarding the choice of apalutamide, it was around the time certain trials were being designed or ongoing, even trials like the EMBARK trial that was using enzalutamide monotherapy. For years, we've sort of known that there's some potential benefits regarding the way the patients feel with bicalutamide monotherapy, but the efficacy didn't seem quite as good.
And so, the thought was, "Could we now with a more potent next-generation hormone therapy use apalutamide monotherapy instead?" So, that was sort of the genesis of seeing, could we get maybe patients tolerating it well, but still get that bang for your buck of having testosterone still circulating. But I think that's the big question is this fully translatable to just ADT, just any form. And I'll say that we do have some preliminary data that's postop from some other trials that it seems that this phenomenon is going to hold at least in the post-operative setting, whether it's just ADT or whether it's even there's a trial done with Phuoc Tran with that used enzalutamide. It's a much smaller trial, but those same signals seem to hold.
So, I would say for right now, I think it's given the other data we have from all these trials with LHRH agonists, I think it's very reasonable to do six months of an LHRH agonist at this point, or if you prefer an antagonist like Relugolix, I think that's very reasonable. I'll be very curious once the paper's published, what's the response from the community and experts like yourself, should apalutamide be an option or not? And obviously, there's different cost implications and things like that, but it will be interesting to hear sort of the broader community's thoughts.
Jeff Michalski: Yeah, it'll be interesting to see how this is applied in the real-life world. So, moving on to my next favorite topic, and that is pelvic radiation. So, obviously, the SPPORT trial randomized patients received pelvic radiation, yes or no. And in this trial, I didn't see how frequently that was used. It was allowed, but I didn't see how frequently it was used. So, I'm kind of curious if that was common or uncommon. And what are your thoughts about, do you use whole pelvic radiation using ADT for patients in the pelvic setting?
Dan Spratt: No, that's an excellent question, Jeff. So, I think, and hopefully, I'm not misquoting, I believe it's somewhere between 25% to 40%, and for some reason my brain, because I just requested those data for the study. I think it's more like 25%, but it might be 40%. But it is the minority, it's not the majority had nodal radiation therapy and it was well-balanced between arms. But that's an excellent point.
And so, I think that, for me, predating these trial results, I typically did not use a lot of whole pelvic radiation for low PSA patients with a good nodal dissection. I definitely, once I got to PSAs closer to around 0.5 where I really am thinking of adding ADT, I was really thinking of adding nodal radiation based on the SPPORT trial, knowing that there's still a lot of unanswered questions in the era of PSMA and how extensive that dissection is.
So, I would say that hopefully this... I'm not sure, 006 really changes my impression of pelvic radiotherapy in this setting and that if you are someone who feels that there's a benefit, especially if you're using a hormone therapy, I don't think this should sway you away from it. But yeah, I'm not sure. It is definitely something worth, and we have profiled 0534 with Decipher. And so, hopefully, there'll be insights. We can learn that maybe, for example, do the basal patients maybe do we need to be giving nodal radiation, right? Because if some people will say, "Well, this is so great, Dan, you have a predictive biomarker that these luminal B patients benefit, but what do you do for the non-luminal B?"
And so, I think a lot needs to be done in that subset for sure. I'm curious, Jeff, for you, what is your practice of nodal radiation in this setting?
Jeff Michalski: Yeah. Well, I was trained with my mentor, Carlos Perez, who was a pelvic node radiator. And so, I tend to treat the lymph nodes. And I might be a bit more selective in patients who had a very thorough lymph node dissection. These days, I tell my patients that even though it says lymph node dissection, it's really that much more than a lymph node sampling. And so, if I see zero or one or two lymph nodes in the dissection, I don't know that we can confidently say that it's pathologic N zero.
Dan Spratt: Yeah, I agree with it.
Jeff Michalski: I'll look at nomograms, what was the predicted risk relative to what we actually removed. But I tend to radiate the lymph nodes. But the other trial that we talked about at the ASTRO meeting that I think will be important in the future discussions is the trial that Mack Roach presented for patients with unfavorable and high-risk prostate cancer, lymph node radiation, yes or no, although follow-up is short in the big scheme of things, seven years versus 18 years that you need to have for prostate cancer, the trial at this point was negative.
So, I'm rethinking how often I need to electively radiate the lymph nodes. So, let me ask you, in this trial, 40% of patients had luminal B, which is higher than I think most reports, which is around a third or so. So, I'm kind of curious, what do we do about, and you alluded to it just a minute ago, what do we do for the non-luminal patients? Because their diagnosis is, well, it's better with salvage, but it's still not great.
Dan Spratt: Yeah, no, it's a great question. And I think one of the things that we would need a much, much larger study as to most of those non-luminal B were basal patients, but there's a small, we'll call it 15%, were luminal A. And luminal A, as we've learned more and more about it, we're still learning. These are the tumors that a lot of people think of when you think of Gleason 6 prostate cancer, or they're on all Gleason 6 obviously, but they very, very, very rarely metastasize.
And so, I think it'll be in future work is to probably pull those out because they have a very favorable prognosis and really look at just the basal like patients. And this is where I think strategies, we sort of have this unclear role of dose escalation in the post-op setting where we recently, they did the Porto signature showing they could find a subset that benefited. It'd be interesting. Is there any overlap with this basal like subset or is similar to kind of the nodal radiation?
Regarding the systemic therapy, if it's not an androgen-based therapy as you're leading the charge on theranostics, is this, we saw the LUNAR trial presented in a different space, but is this something that maybe we need a more radiopharmaceutical targeted approach? But clearly, work needs to be done.
Jeff Michalski: Yeah. So, this is a trial of just under 300 patients.
Dan Spratt: Yeah.
Jeff Michalski: So, do you consider this to be practice changing? And if not, maybe you can speculate about what the next generation of trials in the post-op space might look like.
Dan Spratt: Yeah, it's interesting in that we were very pleasantly surprised and you designed these trials, as you said, it's 300 patients. When you do these biomarker stratified designs, we obviously hope for a slam dunk, but we were thinking we might see some difference in biochemical recurrence trends to then maybe motivate a larger trial. I think the fact that the biology that we developed retrospectively really panned out that you have these just huge, even MFS differences that reach significance.
I think just speaking with people, and even at the NRG Oncology group, I'm not sure the full equipoise of people to let's say, take luminal B patient and not give them ADT, maybe we could find a perfect sliver, but it's going to be challenging. So, I think for at least this population, I'm not sure they'll be, especially if we validate this in 0534 and some other studies, I'm not sure there'll be a really big drug.
We power it probably for MFS anyways. But I think a very important thing that I hope people don't just take these results and say, "Okay, we can use this in all of prostate cancer to omit hormone therapy." I think this needs to be done in intact prostate cancer dedicated trial. I think whether it's node positive or higher PSAs, I think we need to be very clear the population that was on this trial and not extrapolate too much.
We've been working on as a team in the post-op setting, a very large multi-arm trial to really try to address a lot of the questions in greater detail, "Can we further refine benefit of nodal radiation? Can we further refine fractionation dose reductions?" We've talked a lot. Do we need... We'll call it a 68, 70 gray to the whole bed if it's no evidence of gross disease. So, I think there's a lot and to be using these molecular classifiers as part of that, so it's super exciting.
Jeff Michalski: Right. Well, Dan, I just want to take a moment to congratulate you on this major presentation at our National Society Radiation Oncology, the American Society for Radiation Oncology. I also want to acknowledge the groundwork that was laid by our friend and colleague Felix, who unfortunately passed away just a few weeks before this data was available for him to see.
And I also want to thank you for acknowledging that and bringing awareness to the American Cancer Society Astro Clinical Science Development grant that we've established in his name, so that future clinician scientists can work in this space and develop more tools, more therapies for physicians like us who want to help our patients going forward. So, thank you for doing that. And I'll let you have the last word.
Dan Spratt: Oh, thank you, Jeff. I mean, as just coming off of you being president of ASTRO and the work you did, obviously, this is always a challenge to get two societies together to partner on something that's this scale. This is very unique. And so, I think that it couldn't be done without so many people such as yourself, coming together and supporting this.
And it's a true honor, as you said, I wish Felix was here to see these results, but his legacy will continue on through the work all of us do. So, thank you so much. I really appreciate it.
Jeff Michalski: Yeah, great. Nice chatting with you.
Dan Spratt: You too.