Read the Full Video Transcript

Alicia Morgans: Hi, my name is Alicia Morgans, and I'm a GU Medical Oncologist at Dana-Farber Cancer Institute. I'm so excited to have here with me today, Dr. Gerhardt Attard, who's a Professor of Oncology at the University College of London. He's here to speak with us about the late-breaking abstract that he recently presented at ESMO, really looking into the non-metastatic high-risk population and the combination of abiraterone and enzalutamide in that population. Thank you so much for being here with us today, professor Attard.

Gerhardt Attard: Thank you for asking me.

Alicia Morgans: Wonderful. Can you tell us a little bit about what you actually did in this study, and really also help the listeners by defining that high-risk M0 population once more?

Gerhardt Attard: I'll start with the second question, Alicia, the patient population. The eligible patients had no evidence of metastasis on bone and CT scan of the pelvis/abdomen x-ray, so M0 by conventional definitions, and were high-risk, either if they had newly diagnosed disease, which was node-positive, or if they were node-negative, they had to have either stage 3 or 4, PSA 40 or greater, or Gleason sum 8 to 10. And if they were relapsing after prior primary treatment, prostatectomy, or radiotherapy, they need to totally be node-positive or have high-risk features. Again, a PSA 4 or greater, with a doubling time of fewer than 6 months or a PSA of 20 or greater. So really, a high-risk population, which probably represents 15% to 20% of men with non-metastatic prostate cancer.

Your first question, what did we do? This is an analysis of new data from the STAMPEDE multi-arm, multi-stage platform trial. In total, we analyzed 1,974 patients who were randomized across two comparisons, either to ADT versus ADT plus abiraterone, of course, with prednisolone on the Swiss sites in the first comparison, and then the second comparison, ADT versus ADT with abiraterone plus enzalutamide. Critically, both comparisons were run in sequence. We had no overlapping controls. They have the same eligibility criteria and they used the same regimen of abiraterone and the same dose of abiraterone.

The trial management group made a decision a couple of years ago before inspecting any efficacy data from the abiraterone enzalutamide comparison to do three things. The first is to report M0 separately from M1. You'll have seen data emerging from STAMPEDE previously where we combined M0 and M1 patients, and more recently, we have split the two, really because the outcomes of high-risk M0 patients have improved so significantly since STAMPEDE started in 2005. So that's the first thing. The second thing, we amended the primary endpoint for M0 patients to metastasis-free survival. This is after the ICECAP consortium reports that MFS, or metastasis-free survival, is a valid surrogate of overall survival.

And thirdly, we amended the plan to combine M0 patients from the abiraterone comparison and the abiraterone and enzalutamide comparison. We did that for a number of reasons, but that was partly driven by data presented by Mike Morris that showed no discernible difference in overall survival when Enza was combined with abiraterone. That data was similar when apalutamide was combined with abiraterone in the ACIS trial for overall survival. And also, the magnitude of benefit from abiraterone when the Enza was combined with abiraterone, that comparison was designed in 2012, we didn't appreciate the magnitude of benefits with abiraterone alone.

So using meta-analysis methods, we have analyzed the two comparisons to test the question, whether patients with high-risk M0 prostate cancer have an improvement in metastasis-free survival, and consequently, overall survival, if treated with 2 years of abiraterone.  I say why we split M0 from M1, and they clearly have different outcomes. As we implemented the abiraterone arms, the way we manage the patients also were different from what we were doing originally. And really, the M0 patients have received a fixed duration of abiraterone, versus abiraterone continuously to progression as we do with metastatic disease. So there is a number of differences, both in competing risks of death from other causes and in the duration of treatment we administer comparing M1 from M0.

Alicia Morgans: I think that's so important and thank you for pointing that out. Because I do think in clinical practice, we are still trying to figure out how much is enough when we are talking about patients who have localized or locally advanced disease, which is essentially what your M0 population is. And of course, some with some relapse features as well. But the goal for those patients is to cure as many as possible, for many of us as clinicians, and using an indefinite duration of abiraterone is counter to that plan. So, I think this analysis is extremely timely and very consistent with what we hope to be able to do in the clinic. What did you find with your primary endpoint, the MFS endpoint?

Gerhardt Attard: I'll just make one point. In the patient population, and maybe we will come to this in the limitations, but we have an under-representation of relapse. So when you asked me to describe the population, those were our eligibility criteria I defined there, but only about 4% of the patient population had relapsed disease. So they are underrepresented. In the end, we will discuss whether that is relevant.

The primary endpoint. The pre-defined statistical analysis plan was written to target a 25% improvement in metastasis-free survival. So a hazard ratio of 0.75, which required 300 events in the ADT control arm. We have observed a hazard ratio of 0.53. So, at 6 years, the improvement in metastasis-free survival is from 69% with ADT alone to 82% with 2 years of abiraterone, or abiraterone and enzalutamide. So a very significant result. A 95% confidence interval of 0.44 to 0.64. The p-value is 10 to the minus 11. So a very significant result.

In a pre-specified subgroup analysis, we also tested the treatment effect in the two randomization periods separately. So patients treated with abiraterone, and patients treated with abiraterone and enzalutamide. In both comparisons, the hazard ratios are equivalent, so 0.54 and 0.53. The interaction hazard ratio is 1.02. Of course, no significant interaction. We see evidence of the same treatment benefit in both comparisons. We see no evidence for a greater benefit of adding enzalutamide. So our conclusion is that abiraterone for 2 years should become the standard of care for this population of patients and that adding enzalutamide to abiraterone is not justified.

Overall survival, which is a secondary outcome measure, but of course, one of importance, is the hazard ratio of 0.6. So the 95% CI is 0.48 to 0.73. What that equates to in absolute terms at 6 years is this improvement in survival from 77% to 86%. And again, when we look at overall survival by a randomization period, the hazard ratio in the abiraterone comparison is 0.63, and then the abiraterone and enzalutamide comparison to 0.54. So we, again, did not see a difference in survival between the two randomization periods.

Alicia Morgans: That is extremely impressive. And I would have to agree with you that combination seems highly justifiable as a standard of care option for this high-risk locally advanced population. One of the other things that patients are going to want to know is, of course, about the safety data, the adverse event rates, and I'd love to hear your comment on that, particularly if there was a difference in those patients who were getting the double combination of Abi and Enza, versus Abi alone.

Gerhardt Attard: Well, I've alluded to that already. There is increased toxicity with abiraterone and enzalutamide. Significantly more, or let's say discernibly more, grade 3 toxicity involving erectile dysfunction, hypertension, and fatigue, and more grade 3/4 toxicity of liver transaminases, so liver toxicity. It's a higher discontinuation rate of abiraterone when given in combination with enzalutamide. It's about 15% fewer patients completed the planned treatment with abiraterone when the combination was being given.

In terms of toxicity, we are only reporting toxicity in the 2 years of treatment. We have concerns about the accuracy of reporting long-term complications after treatments have been discontinued. The 2-year toxicity is similar to what's been reported previously. Just really a slightly higher reporting of grade 3 or 4 toxicity. So 33% with abiraterone, versus 26% and 30% in the control population. So slightly more hypertension, fatigue, and erectile dysfunction, and more mild transaminitis. So what you'd expect in the 2 years of treatment. The long-term implications here of giving 2 years of abiraterone to this population are uncertain. We clearly know we are prolonging overall survival, but whether we are causing premature deaths, despite that overall survival improvement, and what the impact on quality of life is, is uncertain.

Alicia Morgans: But still, I would agree with the approach to really report those AEs during the period of treatment, as many factors may come into play after that time and really confound your assessment. So that makes a lot of sense and is just another reason when we think about that AE data to not use that ADT plus Abi plus Enza combination, it's not more efficacious, and does have discernibly more AEs. Not to mention the fact that patients would be taking quite a number of pills every day if they were taking both of those. That in itself is a challenge for some.

So, as you think about this and what you are planning, I'm sure you're planning some exciting correlatives, we really do look forward to seeing that information over time. Any thoughts on what might be planned for correlatives that you can share?

Gerhardt Attard: Yeah. We, over the past 5, 6 years, have retrieved the tumor samples from all the patients in this analysis, and also the metastatic patients who were randomized in the previously reported abiraterone comparison, and then, as yet, the unreported abiraterone and enzalutamide comparison. We have tumor samples from upwards of 2,000 patients, and on all those blocks, we have performed transcriptome analyses and a number of other analyses, including Ki67, to, for the latter, ask the question about whether we can identify the bad prognosis M0 patients.

One result I haven't given is prostate cancer-specific survival, which is improved from 85% to 93%. Going forward, I think we still need to continue to develop better drugs and better approaches for this population. There are still men who are relapsing and ultimately sadly dying from prostate cancer, but I think it's going to become increasingly challenging to do this in an unselected population. So we are going to need tests to identify who is likely to relapse and only primarily develop drugs for that group of men.

Alicia Morgans: Absolutely, and then hopefully take it a step farther and even understand which type of drug may help which type of patient who is at the highest risk of relapse. I look forward to hearing your answers on those questions over time too because I know you're looking into that as well. If you had to give a final thought to the listeners, a message from this analysis, and the M0 population in STAMPEDE, what would that message be?

Gerhardt Attard: The conclusion here is very clear. There is a very clear and definite benefit in survival from two years of treatment abiraterone. I think over the next few years, other AR-targeting agents may show similar results, and this is going to lead to a change in the treatment pathway. We discussed the potential long-term implications. I think it is such a very definite survival benefit. This will now be a new standard of care.

Alicia Morgans: I hope it is adopted as quickly as it should be. I agree with you completely. I really sincerely thank you so much for your time, and certainly, of course, congratulate you, the STAMPEDE investigators, and all of the patients who participated in this very meaningful work, which continues to help guide our understanding of how to best care for patients with prostate cancer. Thank you so much for your time and expertise today.

Gerhardt Attard: Great. Thank you for your time.