Radiation Therapy Options for Prostate Cancer - Paul Nguyen
August 13, 2019
They also discuss disease progression, patients who are treated with prostatectomy upfront, and then experience biochemical recurrence, rising PSA, and thinking about salvage radiotherapy. And lastly, they discuss stratifying patients to appropriately target and optimize systemic therapies.
Paul Nguyen, MD, Associate Professor, Radiation Oncology, Harvard Medical School, Director of Prostate Brachytherapy and Clinical Trials, Dana-Farber Cancer institute
Alicia Morgans, MD, MPH, Associate Professor of Medicine in the Division of Hematology/Oncology at the Northwestern University Feinberg School of Medicine in Chicago, Illinois.
Watch: Management of Men with PSA Recurrence After Radical Local Radiation Therapy Presentation - Felix Feng
Read: Management of Men with PSA Recurrence after Radical Local Radiation Therapy
Read: How to Treat Men with Newly Diagnosed cN1 cM0 Prostate Cancer - Radiation Therapy Options
Alicia Morgans: Hi! I'm thrilled to have here with me today Dr. Paul Nguyen, who is an Associate Professor of Radiation Oncology at Harvard, as well as the GU Program Lead in Radiation Oncology at the Dana-Farber. So wonderful to have you here.
Paul Nguyen: Great to be here.
Alicia Morgans: I really wanted to pick your brain on some hot topics in radiation oncology and prostate cancer, and there are many. So, I'd love to hear your advice so that you can help sort out some clinical conundrums that we face, actually, pretty commonly in the care of men with prostate cancer.
And issue number one is, how do you think about radiating the primary in low-volume metastatic hormone-sensitive disease? What do you think about that?
Paul Nguyen: Yeah. I think the data is extremely exciting, that we can offer a new treatment paradigm now for our patients with low-volume metastatic disease. As you know, before the STAMPEDE data, we only had the Horad data, which looked at radiating the primary for these patients and it wasn't a positive study. It didn't show a survival benefit, but there was an interesting subgroup analysis in the Horad data where they looked at patients who had fewer than five metastases. And they actually found the hazard ratio was .68, which was pretty encouraging, and that it seemed like it favored radiating those patients.
But it wasn't statistically significant. And then, this year we had the STAMPEDE data, which was so exciting because it looked at sort of a similar metastatic population and when they compared ADT alone versus ADT plus radiating the primary, there was no difference overall, but in that subgroup that you talked about, the CHAARTED definition of low-volume metastatic patients, there was all of a sudden this same hazard ratio, .68, now statistically significant, suggesting a real 32% improvement in overall survival.
And so, for us, I found this very exciting. I do now offer radiation basically to all the patients that meet the low-volume criteria representing with metastatic disease. Now, of course, the caveat is that we have now abiraterone and other drugs, as well, and that STAMPEDE data wasn't done in the setting of abiraterone. We don't technically know whether it benefits patients to radiate them in the setting of abiraterone, but in my view right now, I'm still doing it where we can.
Alicia Morgans: Absolutely! And it sounds like you're probably combining that with abiraterone?
Paul Nguyen: Yes.
Alicia Morgans: With your medical oncology colleagues.
Paul Nguyen: Yeah. If our medical oncology colleagues are giving it, then I feel fine adding the radiation, as well.
Alicia Morgans: Yeah. And actually, we're doing a lot of the same, and one of the other things that end up coming up in these low-volume metastatic hormone-sensitive patients, some of them bordering on the sort of oligometastatic setting, whatever that definition is, whichever study you're reading, radiating the metastatic sites is also something that I think is one of the big controversies in hormone-sensitive disease.
So, what do you think about that, and how do we answer those questions? Because I don't know that they're answered yet, in terms of...
Paul Nguyen: Right. No, I agree with you. Definitely, the questions aren't answered yet, and everything we have right now is just sort of Phase 2 studies that are randomized but extremely intriguing. Of course, we have the STOMP data previously in the JCO that showed that radiating oligomets with SBRT could delay the time to hormone therapy from 13 to 21 months, which is certainly a benefit for patients.
I think particularly intriguing this year that came out at ASCO as a plenary was the COMET data, and this looked at patients with all kinds of cancers with fewer than five metastases and found that... It was a Phase 2 randomized study, so they did SBRT to half the patients, and what they had found was there was actually an improvement in overall survival.
The P-value was .09, and so, technically, it actually was statistically significant because it was a Phase 2, and they'd pre-specified a P-value of .2, so technically positive, but still. A phase two study, so really, hypothesis-generating right now and I think we really need Phase 3 data to say that this is standard of care. But I think it's encouraging, and we do have studies open now, essentially, where we're following that paradigm where we're treating aggressively the primary for these oligometastatic patients, and we're doing SBRT to those oligomets, as well.
Alicia Morgans: I agree. Really exciting, actually, and certainly hypothesis-generating, and also I think something that's probably really going to be tolerable to our patients. If we find that there's a benefit in terms of disease control and prostate cancer-specific outcomes, I still think it's going to be beneficial also potentially in a tolerable way for these patients.
Paul Nguyen: Yeah, I think so. I mean, I've certainly been very impressed. The patients that we've treated, they seem to do very well. Essentially, we're radiating very small areas, and they tolerate it pretty well.
Alicia Morgans: And over how many days do you usually give these SBRT or whatever you're using for the metastatic sites?
Paul Nguyen: It's usually a short number of fractions, so typically you're looking at three to five, or as few as one fraction, depending on the size.
Alicia Morgans: Okay. So, really, both tolerable in terms of side effects but also in terms of feasibility for patients to actually take the time-
Paul Nguyen: That's right.
Alicia Morgans: To do these kind of things.
Paul Nguyen: Yeah, the time isn't too bad.
Alicia Morgans: And I think there's actually a study here, a partnership, it originated between some of the investigators here at the Farber as well as those investigators at Memorial, putting together this trial that we're lucky enough to participate in, as well, the Metacure trial.
Paul Nguyen: Yeah.
Alicia Morgans: Can you speak about that a little bit?
Paul Nguyen: Yeah. Metacure, it's a great trial. It's started by Howard Scher, Mary Ellen Taplin and also Sean McBride down at Sloan Kettering, and we've participated here. And it's basically looking at either a high-risk or this oligometastatic population, and it's treating them with next-generation hormone therapy as well as, for the oligometastatic patients, surgery to the primary, and SBRT to the oligomets.
Really, kind of building off of this paradigm that we were talking about of aggressive treatment to the primary and aggressive treatment to the oligomets. And using next-generation hormone therapy, so it's a really strong combination of therapies for these patients with oligometastatic disease.
In addition, part of the trial is recommending adjuvant therapy using adjuvant radiation after the surgery. This is really sort of comprehensive, aggressive, local and systemic therapy for these patients with oligometastatic disease. We're pretty excited about the study.
Alicia Morgans: I'm excited, too, and I think that, as I think about, "Why is this coming along now when prior therapies, or prior studies looking at these combinations of different therapies, didn't necessarily seem to be successful?" We have better techniques, I think. We have better techniques, I'm sure, in SBRT. We have better techniques at least in terms of the systemic therapies, and we can learn a lot from the data gathered from removing these specimens at prostatectomy, and then evaluating them.
I just think that the time is right for us to really engage, and we have enough data to suggest that these local therapies and intensive systemic therapies and these metastases-directed approaches are really going to be combinatorial. So, I'm very excited about this trial.
Paul Nguyen: Yeah. I think so. I think that as systemic therapy has gotten better, we could actually see a benefit to local therapy now.
Alicia Morgans: Yeah.
Paul Nguyen: Whereas before, when systemic therapy wasn't that great, it wasn't quite as promising.
Alicia Morgans: Absolutely.
Paul Nguyen: I think they're working together very synergistically now.
Alicia Morgans: Wonderful. One of the other big controversies that I wanted to pick your brain about is, what happens when the disease comes back? When we have patients who are treated with prostatectomy upfront, and then they have a biochemical recurrence, rising PSA, and we think about salvage RT, what's the best way to approach that?
We have the data from 9601 suggesting that hormonal therapy coupled with RT could improve overall survival, but there's some caveats there, and that was bicalutamide therapy, right? How do you interpret that data, and how are you applying it in clinical practice when we have patients with biochemical recurrence?
Paul Nguyen: Right. Yeah, it's a great question, and it still remains pretty confusing. Maybe I'll tackle the type and length of the hormone therapy first.
Alicia Morgans: Perfect.
Paul Nguyen: And then we can talk about the subgroup analyses. As you said, the 9601 showed an improvement in overall survival when you added two years of bicalutamide 150 to salvage radiation, but that's not a very common therapy, and who really wants to be on two years of bicalutamide? It's the huge gynecomastia issue.
And so, for our patients, I've been favoring six months of the GNRH agonist instead of the two years of bicalutamide. And the way I reason that is I look at the benefit from 9601, and that had an overall survival hazard ratio of 0.77, so a 23% reduction in overall death.
And I compare that to the GETUG study, which also looked at salvage radiation but here, it used six months of GNRH agonist instead of the two years of bicalutamide. And in the GETUG study, which had only five years of follow-up, the hazard ratio of biochemical recurrence was exactly the same. It was about .5, so it cut the risk of recurrence in half.
But the overall survival hazard ratio was 0.7, so even a little bit stronger than the 9601. And it's not quite statistically significant yet; the P-value is only 0.18, but there's only five years of follow-up. In my view, if we're to follow the GETUG study out, it's probably going to be positive. We'll probably see that survival benefit, and the hazard ratio is pretty promising.
I think we're better off giving patients right now six months of the GNRH agonist, rather than doing the two years of Bicalutamide 150.
Alicia Morgans: I think that sounds great. That's what we're doing in our practice, as well. The 150 of Bicalutamide is really difficult to tolerate for a number of reasons, as you mentioned. Plus, two years of therapy is another challenge in a population that we're hoping actually to be treated with curative intent. We'd love to be able to kind of curtail that, just cut it short, keep people off therapy, and really get that upfront benefit.
But you alluded to one other thing that I want to hear more about, and this is still, I think, kind of controversial. These subgroup analyses that suggested that at certain PSA levels, when we start in radiation, that we would want to use GNRH agonist or antagonist therapy, and certain ones that we would not. Can you tell us a little bit about those subgroup analyses?
Paul Nguyen: Yeah. That remains, I think, a big puzzle for the field. The 9601 was stratified by a PSA greater than 1.5 at entry versus less than or equal to 1.5.
And then, there was a subgroup analysis performed that suggested that if your PSA was less than .7, which is not something that was pre-stratified, but if your PSA was less than .7 and there did not seem to be an overall survival benefit, there was a benefit, it seemed, in biochemical recurrence and there was sort of a trend at least in the right direction towards a metastases free survival benefit, but not an overall survival benefit.
And so, people have struggled with what to do with that. On the one hand, you could read the study and say, "Well, the overall trial was positive. It wasn't stratified by a PSA of .7, so that's just sort of a hypothesis-generating analysis, and so, we should treat all patients with hormone therapy."
And that has tended to be the approach that we've favored here, so at least at the Farber, within our group, we've tended to give hormone therapy for most patients in the salvage setting because ultimately, most of the patients we see these days have a PSA much less than .7. Typically, we're trying to salvage them at a PSA of .1, and so, for the most part, we've kind of leaned towards giving hormones for all or at least the vast majority.
But, you know, there's a strong sentiment out there, which I think is a reasonable view, that, look, if these patients, if we don't know for 100% sure that these patients are having a survival benefit, then why expose them to the toxicity of hormone therapy?
And I think the field is kind of split on it. And we've kind of tended towards giving hormones for all, but I know that there are very reasonable groups out there that are looking at this subgroup analysis and saying, "Well, we'll try to avoid hormones for as many of these patients as we can."
And there are a couple of different studies looking at these different approaches. Dan Spratt at the University of Michigan has a study through the NRG, so it's NRGG 006; he's doing this with Felix Feng. And they're looking at salvaged patients, and they're giving them either salvage alone or salvage plus apalutamide.
And they're stratifying them by the basal luminal subtypes, by the PAM50 score. That's going to be an interesting study. It's going to help to see whether some of those patients benefit from apalutamide or not.
Alicia Morgans: Are these only those patients who have a PSA that's low, or...
Paul Nguyen: It tends to be more favorable patients.
Alicia Morgans: Okay.
Paul Nguyen: I don't remember if they exactly picked out a PSA less than .7, but it tends to be more favorable patients.
Alicia Morgans: Okay.
Paul Nguyen: And then, we have a study that's open that's a multi-center study that we call FORMULA 509, and it's basically looking at a higher-risk population. They have to have at least one high-risk feature, and they're getting salvage radiation. And we randomize them to salvage radiation plus six months of GNRH and bicalutamide, versus salvage radiation, six months of GNRH, and abiraterone and apalutamide.
It's trying to intensify treatment using six months of GNRH as the baseline.
Alicia Morgans: Great. Well, I know you're also trying to de-escalate. I know you're working with Oliver Sartor on studies, as well, and one of the interests in that study is to sort of reducing where we can, and stratify patients to appropriately target and optimize systemic therapies.
Can you tell me a little bit about that?
Paul Nguyen: Yeah, absolutely. This is for patients with de novo high-risk prostate cancer who are going to be treated with radiation. And it's always been a struggle for us, I think, to find the right duration of hormones for these patients.
Alicia Morgans: Yes.
Paul Nguyen: And one of the key things is that these patients are all different. When we look at high-risk disease, those with Gleason 8-to-10, or PSA greater than 20, or clinical T3a rate, they could have very different risks of disease on a disease spectrum. Some of these patients might have Gleason 9 and T3band a PSA of 100; some of these patients might be like a patient with a T1c, PSA 5, and one core of Gleason 8.
And you can imagine the disease spectrum is quite wide there, and yet, we treat all these patients as a one-size-fits-all right now for high-risk disease. And so, the trial that we're doing that we're kind of planning through the NRG with Oliver and Felix Feng, is trying to do exactly what you're saying which is trying to kind of tease out these patients a little bit.
Dan Spratt did a nice analysis in the JCO looking at using the Decipher test to basically stratify different D'amico risk classifications of patients and found that you can create these other subgroups that basically add to the standard D'amico classifications.
Kind of building off that idea, we're going to look at high-risk patients using classical high-risk definition, and we're going to run the Decipher test on all these patients. And then for the patients with the upper third of Decipher risk, those patients have a pretty high risk of death. And those patients, I don't think we want to cut back on anything.
So, we're going to randomize those patients to either radiation and standard of care hormones, or radiation, standard of care hormones plus additional drug. In this case, probably abiraterone and apalutamide.
But then for those patients who fall into the lower two-thirds of genomic risk, I think here's where we have an opportunity to try to de-escalate therapy for these patients. So, instead of giving them 24 months or maybe 18 months, we're going to either randomize them to that duration, or an intermediate duration, which is either 12 months or 9 months of hormones. We're going to actually ratchet it down.
And I think it's very likely that we can get away with shortening the duration for these patients. I don't think that it's a one-size-fits-all, and ultimately, the trials that established two to three years of hormone therapy had patients that were pretty locally advanced.
Alicia Morgans: Yeah.
Paul Nguyen: And 70% of those patients had T3 or T4 disease. They're not like the patients we're seeing today, the T1c's with the low PSA and maybe a little bit of Gleason 8. Those are the patients where I think we can do better, and we can kind of think about personalizing and de-escalating therapy for those patients because we all know the side effects of hormone therapy, and it's just something that we just want to be able to shorten as much as possible for those patients.
Alicia Morgans: I agree. And I think what they didn't have, too, in those studies were designed was the Decipher risk classifier, as well. You're really adding in some molecular stratification techniques that I think empower us to have more than just those clinical features, even though, as you described those two patients, I mean, they clearly are very distinct in what we expect from their disease outcome.
If you had a patient right now, and you did not have access to this trial that you're developing, what would you do for high-risk patients in terms of the duration of hormonal therapy? Because that's one of the other big controversies in radiation, the duration of therapy, there's been some data; 18 months, 36 months, 24 months.
Where do you fall in that?
Paul Nguyen: Yeah. It's such a mess. I mean, there's the 12 months, 18 months, 24, 28, 36 months.
Alicia Morgans: Yes.
Paul Nguyen: And so, there's so much out there. I would say now for my standard high-risk patient, I try to use the best radiation possible and try to do the best hormone therapy duration possible. And so, for my typical patient, I want to give them a brachytherapy boost if I can, and that's based on the ASCENDE RT data, plus data out of Peter Hoskins' group that looked at brachytherapy boosts and it showed that it basically cut the risk of recurrence in half for these patients.
And while we don't have a survival benefit yet, I think there's a real benefit to having biochemical recurrence cut in half and saving yourself the additional hormone therapy down the line. So, I try to be aggressive with these patients and minimize the risk for local recurrence.
In terms of the duration of hormone therapy, I'm liking two years right now. That's really what I consider my standard for most patients. The NCCN guidelines came out with two to three years initially, and that was based on the Bolla study, which showed the benefit of 36 months; the RTOG study, which showed the benefit of 28 months, and I think it was too hard to say 28 to 36 months, so people just said two to three years, and now basically most international trials are using two years as a standard.
The STAMPEDE group did that, and we've done that in ENZARAD, and the NRG has been using two years as the standard, as well.
Alicia Morgans: Mm-hmm.
Paul Nguyen: In terms of the 18 months data, you know, I think there's a lot to be excited about the 18-month data, and people read it two different ways. One way to read it is that with the most recent update of the 18-month data, we couldn't prove that 36 was superior to 18 months. The hazard ratio was 1.02, which is pretty close to one. And the confidence interval went up to 1.29.
And so, if you were to apply non-inferiority statistics to that, you would probably say that's a non-inferiority... It is non-inferior. 18 months is really non-inferior to 36 months.
And that's one very reasonable way to read it because we've accepted an upper border of 1.35 as a reasonable non-inferiority boundary, and here, they've made it all the way to 1.29, and it didn’t exceed 1.35.
Alicia Morgans: Mm-hmm.
Paul Nguyen: The downside of reading it that way is plenty of statisticians think that you can't really do that. You can't apply non-inferiority statistics to a trial that wasn't defined-
Alicia Morgans: Defined that way.
Paul Nguyen: That way. So, I see that argument. I would say, when I'm in front of a patient that might be a hard distinction to make for them.
Alicia Morgans: Yes.
Paul Nguyen: You know? To say that we're going to commit them to this extra hormonal therapy because of the statistics around it, that it wasn't purely defined that way and it's hard to apply it post-hoc, so that argument, I think, won't be that compelling to patients.
I do have some personal concerns about the data, though, and over-interpreting it, or sort of giving it too much weight. I think it's a very important study, but I think one thing we should look at, for example, is the dropout rate in the 36-month arm. That it was pretty high. That basically, it was something like 25% of the patients in the 36-month arm stopped at 18 months. They saw their buddies in the 18-month arm were just stopping at 18, so they stopped at 18, too.
And then, only 59% of the patients made it to 36 months. And that tends to bias the results, then, towards the null. You know? It's sort of like, "Well, of course, we didn't see a difference because so many patients dropped out before 36 months." It doesn't mean that there's not a benefit to going longer necessarily, but people dropped out.
The other piece of the data that I think gives me a little bit of pause is that we are applying this data immediately to Gleason 8-to-10 patients. But in that study, 40% of the patients had Gleason 7 disease, and those patients aren't going to benefit from lengthening the duration of hormone therapy.
And when you look at just the Gleason 8-to-10 patients, the hazard ratio is 1.13, so that's a point estimate, and then the 95% confidence interval goes way up again to 1.55. So, it exceeds what we would accept as non-inferiority for the Gleason 8-to-10 patients.
Alicia Morgans: Yes.
Paul Nguyen: So, I don't think we've really proven non-inferiority, especially not for those Gleason 8-to-10 patients.
I understand the interest in 18 months, and I think it's a reasonable place to stop, but I prefer going longer, so I'm sticking to 24 right now.
Alicia Morgans: And I think that makes sense. You know, one thing that is also kind of clear from that is that 36 months is hard for patients. I mean, if all these patients were dropping out, then that's hard. But 24 months for most patients, we can get them through. We can get them to that endpoint, so another reason to sort of go in that direction.
But this has been really enlightening and exciting, and I have really appreciated your time. These were some hot topics in radiation oncology in prostate cancer, and I feel like I've learned a ton. So, thank you so much for sharing this.
Paul Nguyen: Well, I've really enjoyed it. Thank you so much for the opportunity.
Alicia Morgans: Great.