Get Free Access to Session Slides

Radiation Therapy Options for Men with Newly Diagnosed cN1 cM0 Prostate Cancer - Mack Roach

Mack Roach presents on radiotherapy options for patients with clinically node-positive prostate cancer at the 2019 Advanced Prostate Cancer Consensus Conference (APCCC). Dr. Roach touches on areas such as the optimal candidate for salvage lymph node dissection (SLND) for nodal recurrence of prostate cancer, the role of adding ADT to radiotherapy, and the NRG/RTOG 9413 study which showed whole pelvic radiotherapy plus neoadjuvant hormonal therapy improved progression-free survival in patients with intermediate-risk or high-risk localized prostate cancer compared with prostate only radiotherapy plus neoadjuvant hormonal therapy, whole pelvic radiotherapy plus adjuvant hormonal therapy, and prostate only radiotherapy plus adjuvant hormonal therapy.


Mack Roach III, MD, Professor of Radiation Oncology and Urology, Department of Radiation Oncology, University of California San Francisco, San Francisco, California, USA.

Read the Full Video Transcript

Mark Roach III: I'm greatly honored to participate in this program and as Felix alluded to, node-positive disease is my sweet spot in prostate cancer. The question is what is it? Were we talking about people that have, by this modern imaging PET positive disease or old CT's? Are we talking about patients who are post-prostatectomy versus the definitive patient who walks in the door? Are we talking about patients who have occult node-positive disease, and that's where we get into the issue of whether we need to prophylactically irradiate or do extended radiation.

And then when we talk about benefit, are we talking about survival, PSA control? Are we talking about documenting that those people are controlled because most studies, even our studies at UCSF looking at PSMA, we don't do the PSMA PET, treat the patient and then repeat the study, or do anything to confirm that actually our treatments work.

Are we talking about Level One evidence, retrospective studies? There are many issues that we need to address. With respect to radiotherapy, I'm going to keep it real simple. There's three things. No radiation, local radiation to a presumed positive node or extended-field radiation. So zero field, a small field or a big field. And typically with a small field, today we're talking about hypofractionated or stereotactic treatment, which involves very few treatments.

This surgical series that was recently published is relevant to the issue of radiation. This is a series where these patients underwent a salvage lymph node dissection. The average PSA was around two. These patients all had PET positive nodes. 51% of them only had one positive node on PET imaging. Interesting, at the time of salvage surgery, 9% the nodes were actually false positives. That is, they didn't find disease. And 54% had more than three nodes positive.

So we know that the node dissection is not a great strategy for evaluating and taking care of all the disease because as you see the curves coming down, the vast majority of the patients still have biochemical failure. So this should give us some insight. 70% of the patients still have a detectable PSA postoperatively. There's also some relevant radiation. Thanks, Peter Ost for sending me their study looking at stereotactic treatment versus extended radiotherapy in the setting of oligometastatic disease after local treatment. These patients either had SBRT using five gray fractions up to 10 fractions. And then they had a typical course of whole pelvic radiation around 45 gray. And the bottom line is it's suggested that at least in patients who had a single positive node that extended-field seemed to be beneficial. So that's what this curve is about.

And of course, it's a little bit counterintuitive that if you only had one positive node that you benefited more from extended radiotherapy. And those people that had more than one positive node seemed not to benefit. And that may be relevant to the issue of in patients who had many positive nodes, maybe their fields were bigger and they approached the field size of those with large field or the patients had extensive metastatic disease.

If you look at the pattern of failure in this study, it's suggested that the extended-field reduced the incidence of pelvic failure, so it did do what we hoped it would do. But it still did not appear to be dramatically beneficial. Again, this is a post hoc study, there's selection bias and maybe that's the issue that makes this difficult to interpret.

So again, is it because of the field size? Is it the definition of failure? Is it the endpoint of metastasis-free survival? And they did note increased toxicity with the extended-field radiotherapy. And then the question is if you gave hormone therapy, would that impact the results? I will talk about interactions between volume of irradiation and the benefits of hormone therapy later on, but there is data to suggest that if you're going to radiate pelvic lymph nodes, the benefit is greater if you give hormone therapy before you do it.

There's also data from a French study looking at extended-field versus involved-field radiotherapy. And in the interest of bells and time, the bottom line is that they also suggested that there was a trend for patients who received extended-field radiotherapy to do better than patients that had involved-field radiotherapy only.

Now, this is a post hoc analysis of data from our Phase 3 randomized trial, 9413. This is arm one versus arm two, where arm one was whole pelvic radiation, which is the top curve, and the bottom two curves are from the prostate-only arm.

These patients, again this is randomized data, they all got the same dose of radiation. They all got the sequence. They all got the duration of hormone therapy, suggesting that field size is really important. So when you look at the data from the previous study that suggested that the patients with more nodes didn't seem to benefit, maybe it's because the differences in field size were smaller. And so maybe the patients who had more nodes even needed bigger fields. And this just sort of brings it home. If you looked at the distribution of where people tend to have nodal involvement, field size does matter. And if you look for example at the GETUG-01, the slide on my right, the one without the nodes, you see that that was a relatively small field compared to what we call whole pelvic radiation in our RTOG study.

So the field size really does matter when you're talking about the value of radiating the nodes. And the other point, and I didn't speak about the PET data, is that we know from the previous studies that plenty of people will have nodal disease beyond where the PET shows the disease to be.

Now, I think this is an important study that was published many years ago. This is a trial, again a randomized trial, where patients were randomized with high-risk disease between radiotherapy alone or radiotherapy with hormone therapy. And if you look at patients who had node-positive disease, it's clear that the hormone therapy appeared to be associated with a better outcome. So I think that the notion that we should expect or hope that by giving radiotherapy alone that we're going to control these patients, I think that is an unlikely event.

And these are from two large studies looking at hormone therapy alone in patients with high-risk disease. Again, they were not known to be node-positive, but we know statistically they were very likely to have node-positive disease. You can see that adding radiotherapy to hormone therapy improved the results. So the bottom line is radiotherapy alone is inadequate and hormone therapy alone is inadequate, and the two of them need to be used together. And also sequence matters, and I'll talk about that in a bit.

Now we have a new randomized trial to support the value of prophylactic radiation. Again, only people with positive nodes can benefit from prophylactic radiation. This is our randomized trial RTOG 0534. These are post-op patients, rising PSA, randomized between prostate bed only radiation or prostate bed plus short-term hormone therapy or the use of neoadjuvant hormone therapy combined with whole pelvic radiation showing a clear benefit in the salvage setting. So we think that this information is relevant to these node-positive patients that are failing in the post-op setting and in the post-radiation setting. So from my perspective, I think that patients that are undergoing treatment would be best served with an extended-field.

Coming back to our original report from 2003 on our 1,200 patient randomized trial, you can see the top curve is neoadjuvant whole pelvis. The bottom yellow curve is neoadjuvant prostate only, and the other two curves were intermediate because the timing of hormone therapy was later. And we have recently updated this data with 14-year follow-up in surviving patients. And what we see, as we initially reported, is that there are sequence and volume-dependent interactions. And so when you look at whole pelvic radiation, which is on the ... here, you see this is for neoadjuvant versus adjuvant hormone therapy. The control rate is different. If you give it with whole pelvis, it's lower. If you give the hormone therapy after the radiation, it's better if you give the hormone therapy before the radiation.

And it depends on the volume. So if you're only going to radiate the prostate, it doesn't matter. In fact, maybe the patients do better if you give the hormone therapy after the radiation. There are theoretical reasons that might be the case. I don't have time to go into that now. But the bottom line is that with 14-year follow-up in surviving patients, those who receive neoadjuvant plus whole pelvis and those who receive prostate bed only plus adjuvant hormone therapy did the best compared to the other groups. So it is possible that if for some reason you wanted to do stereotactic radiotherapy in patients with node-positive disease, you might want to give the hormone therapy after the radiation.

Now in order to try to answer this question definitively, we chose to take arm one and arm two from RTOG 9413. This is an analysis of the original RTOG 9413 showing the biggest benefit in terms of control rate by doing whole pelvic radiation in a post hoc analysis. We showed that that would result theoretically in the disease-specific survival difference of 13%. Based on that we launched a large Phase 3 randomized trial. We stratified by various factors. And one of the things that we took into account is that if you want to prove something doesn't work, design an underpowered study. We did not want to design an underpowered study, so I made the study as big as I could get away with. That's another way of doing statistics. Do what you can.

So the reason that this is really important is here's a practical sort of walkthrough about what we mean when we talk about a study. And I'll skip the first point. But suppose you had a 1,200-patient study and the incidence of real lymph node involvement is 33%. So you'd have really 400 patients. Your study is really going to be based on 400 patients, not 1,200. And then assume that some of those people have disease beyond the pelvis, say 25%. Now your study's based on 300 patients. And then suppose your treatment has some local recurrences. Say one third. Now your study is based on 200 patients. And then you have competing causes of death in half the patients. Now it's 100 patients. And then you have improved salvage therapy with platinum-based drugs and other kinds of drugs. Your study's going to be kind of small.

So we chose to do a 2,500 patient study. Fortunately, we recruited 2,592 patients, completed our study a year ahead of schedule, and hopefully we will be able to answer the question about prophylactic irradiation, which is relevant to the issue of patients with node-positive disease.

So my conclusions are we're going to need composite endpoints to design the appropriate study. We need to be able to look at PSA kinetics and other things. We need to be prepared for unexpected interactions such as sequence and volume. And we need to understand the doses and techniques. And an underpowered study or a poorly designed study is worse than no study at all. Thank you for your attention.