Amar Kishan: Thank you. Thank you very much for that introduction and your kind words. I'm happy to talk more about the study. So we know that in the context of definitive radiotherapy, adding hormone therapy to radiation improves overall survival and also extending its duration improves overall survival. But in the context of postoperative radiation after prior prostatectomy, the results have been a lot less clear in the randomized studies, and there have been about six randomized comparisons that have been published that have been asking a permutation of this question. So last year there was a very great study that was a study-level meta-analysis called the DADSPORT meta-analysis came out in European Urology. And it showed that there wasn't a clear statistically significant overall survival benefit and maybe the absolute effect was 2% in terms of overall survival improvement with hormone therapy, but they didn't have the individual-patient data, which sometimes that helps us looking at subgroup effects, interaction effects, and non-linear associations. So the premise behind our study, which is called the POSEIDON Meta-analysis, was to obtain individual-patient data from each of these different cooperative groups that ran the practice-changing randomized trials so that we could try to delve a little bit deeper into some of these specific questions. So that was kind of the overall goal. Our primary outcome was overall survival, and the main question was, does adding hormone therapy improve overall survival?
So we were able to get the data from these six randomized comparisons that asked plus or minus short term, plus or minus long term or taking short-term to long-term hormone therapy. And long story short, happy to delve into details, but with the 6,057 patients overall that we had a median follow-up of nine years, looking at plus or minus any hormone therapy, what we found was that the 10-year overall survival benefit was only 0.7%, which is exceedingly small. Turned out to also be not statistically significant. The hazard ratio was 0.87, the P value was 0.06, but again, it's that absolute impact was very small. We then did some subgroup analyses and tests for interaction. We found that the duration did not have an interaction effect, meaning it didn't help survival more to do long-term hormone therapy versus short-term. But the pre-radiation PSA had a big interaction effect, particularly if we did a multivariate interaction test and did a cut point of 0.5 or more versus less than or equal to 0.5, then we saw an overall survival benefit in patients with the elevated PSA. We then looked at some different things like non-linear associations. Ultimately, what we found was that for patients with a PSA less than or equal to 0.5, there did not appear to be an overall survival benefit. So it's the patients with a PSA greater than 0.5 that really get the overall survival benefit from adding hormone therapy. And if you're going to add hormone therapy, it seems like short term, which is four to six months, is likely sufficient for the vast majority of patients, no need to go all the way to 24 months.
Neeraj Agarwal: It's interesting, cutoff for 0.5. How did you come up with this cutoff of 0.5 for determining whether radiation therapy is helped by adding androgen deprivation therapy?
Amar Kishan: Yeah, that's a great question. So when we did the analysis, we pre-specified for PSA strata just to try to capture as much of the data as possible as less than 0.25, 0.25 to 0.5, 0.5 to one and greater than one. And when we looked in those four strata, the P value for interaction was 0.08. And in the groups that had a PSA less than 0.5, there was no OS benefit. And in the patients who had a PSA greater than 0.5, we did see an OS benefit. So we decided to then look maybe at the binary cutoff of less than or equal to 0.5 versus greater than 0.5. It also happened to be the median PSA in patients that were in the plus or minus long-term hormone therapy randomization.
Neeraj Agarwal: And to me, as a medical oncologist, it makes sense. I personally think most of the patients who we send for salvage radiation therapy have a PSA of 0.5. I'm sure it may be reflective of your own practice and patients who have a PSA level of more than 0.5, they likely have metastatic disease. So any comments on that?
Amar Kishan: I think that's exactly right. I think there's maybe two effects that we could be seeing. One is there is a radiosensitizing aspect to hormone therapy and maybe the patients who don't have metastatic disease but have a PSA greater than 0.5, they have more bulk of disease in the prostate bed itself. So adding the hormone therapy might help from that perspective. But I completely agree that likely what we're picking up is the benefit of hormone therapies maybe from controlling micrometastatic disease in these patients with very elevated PSAs. In fact, we did something called a spline analysis, and it looked like the biggest benefit to adding hormone therapy was in patients with really high PSAs, like greater than 1.6, which almost certainly modern era, if you did a PSMA PET, you would see something likely outside the pelvis. So I overall agree.
Neeraj Agarwal: And especially given the emergence of our awareness, increasing awareness of long-term side effects of androgen deprivation therapy, especially in these patients who are going to live for many years, I think it's personally for my practice is a very exciting news that these patients who have a low PSA level of less than 0.5 nanogram per milliliter, they do not need radiation therapy. They do not need androgen deprivation therapy when they're getting salvage radiation therapy, and we can potentially protect these patients from long-term cardiovascular side effects of androgen deprivation therapy.
Amar Kishan: Absolutely agree. And that was one of the major driving focuses of this. You could say that maybe there are some patients with very aggressive disease features, rapid doubling time or something that falls under 0.5 and maybe it's a nuanced discussion with those patients. But I think yes, we can take a look at the POSEIDON data and say for the vast majority of patients, it's probably overtreatment to add hormone therapy to salvage radiation. For the reasons that you mentioned that nobody wants to be on hormone therapy if they don't have to be in terms of an OS benefit.
Neeraj Agarwal: So for our worldwide audience today, this is definitely, as I said, a practice-informing study. Any message for our colleagues across the world, when they are seeing patients for salvage radiation therapy, for biochemical recurrence, what they should be doing from androgen deprivation therapy perspective?
Amar Kishan: I think the two main key takeaways would be that for patients with a PSA greater than 0.5, they may have this benefit from adding hormone therapy in terms of an OS for patients who are getting hormone therapy, short term is likely sufficient. And maybe a third one, which is ultimately it's always a nuanced discussion. If it's someone with, again, at least a nine, rapid doubling time, it's a risk-benefit ratio for adding the hormone therapy, but hopefully for patients being referred for early salvage radiation therapy less than 0.5, we can spare a lot of them from hormone therapy.
Neeraj Agarwal: Thank you so much for summarizing these data so clearly for our colleagues out there.
Amar Kishan: Thank you for the invitation.