Racial Disparities in Prostate Cancer Treatment and Outcomes - Susan Halabi

February 12, 2019

Susan Halabi discusses new findings in the evaluation of racial disparities in progression-free survival across nine trials.   


Susan Halabi Ph.D Professor of Biostatistics and Bioinformatics. Duke 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.

Read the Full Video Transcript

Alicia Morgans: I have here with me today Dr. Susan Halabi, who is a Professor of Biostatistics and Bioinformatics at Duke University, and actually the first biostatistician fellow of ASCO. So a true pleasure to have you here with us today Susan.

Susan Halabi: Thank you Alicia, it's always a pleasure to talk with you. 

Alicia Morgans: Wonderful. Well, we had a great conversation at ASCO this past year about a really intriguing analysis that you did on studies of metastatic castration resistant prostate cancer (mCRPC) that had used docetaxel as their treatment arm. And you looked at racial differences in survival outcomes, and I'd just love to hear what updates you have on that analysis, what further data you collected.

Susan Halabi: Thank you Alicia. So, what we have presented during ASCO was a big meta-analysis of nearly 9,000 men who were randomized on nine phase three clinical trials where patients were randomized to either docetaxel/prednisone or docetaxel/prednisone and an experimental agent. And it's interesting to note that none of the trials were positive for overall survival. 

So, based on the extensive published literature in terms of how disparity, we were interested to answer really this basic question; is this, have disparity due to biology or is it due to access to care. So obviously, we have data from nine clinical trials and we have the endpoints collected. And our hypothesis based on the extensive data was that African-American men would have worse clinical outcomes than Caucasian men. 

So we first hypothesized that African-American men will have a shorter duration of survival than Caucasian men. In addition, we hypothesized that African-American men will have a shorter duration of progression free survival than Caucasian men. And we also looked at the third hypothesis, has to do with radiographic response and PSA decline. 

So what we have done in terms of overall survival, we looked at the data and our amazement, we found no differences in unadjusted analysis. The median survival time was 21 months in both African-American and Caucasian men. However, when we adjusted for important prognostic factors that were common across the nine clinical trials, we found that the hazard ratio of death was 0.8 in favor of African-American men.

So this was reported at ASCO last year. The next step was to look at other clinical outcomes. So we looked at progression free survival, which was the really a very important clinical endpoint across the nine trials. And what's interesting looking at those nine trials is, that all the trials define progression free survival in different ways. 

So despite that, we pooled the data across the nine trials and we conducted another meta-analysis with progression free survival as the endpoint. And when we did that, what we noticed in unadjusted analysis that the median progression free survival was eight months in both African-American men and Caucasian men. Then what we did is, we went ahead and run adjusted analysis and multi-variable analysis within each trial, and then we combined all these hazard ratios across the nine trials to come up with a weighted pooled hazard ratio.

And when we did that, we found that really there's no differences in terms of progression free survival between African-American and Caucasian men who were treated on these trials. The hazard ratio was 1.0 with a P value of 0.461. So that was again amazing that we found no differences with progression free survival. So one limitation though of this analysis could be that the PFS endpoint was not harmonized across the nine trials. 

And the next step would be to go back to the sponsor and get individual components of the progression free survival endpoint, because unfortunately, I don't have access to that. 

Alicia Morgans: Of course, but are you looking at radiographic-so you look then I guess at radiographic events versus PSA events?

Susan Halabi: Exactly. So, for instance, if we look at CALGB 90401, progression free survival was defined as time to first event, whether the event could be from PSA or it could be from objective progression or bone progression or death, whichever occurs first. Now, if you look at another study, let's say SWOG 0421, they defined PFS totally different. So their component included pain, bone progression, but it did not include PSA. 

So as an investigator, I have only access to the PFS as defined per protocol, and this is what the analysis was based on. So it's gonna be very challenging both statistically and logistically to take the components and harmonize them retrospectively across the trials. But I think nevertheless, the results were very intriguing. Basically, it appears that men, African-American men who have enrolled on these trials, have equal if not more favorable outcomes than Caucasian men who were treated with docetaxel and prednisone.

Alicia Morgans: That's really interesting. So the other thing that I would hope that you can look at is prostate cancer specific survival. And I don't know that you can, given the databases that you have. But, it might be that Caucasian men are developing other complications from cardiovascular disease, from skeletal events, whatever it is that might be increasing their mortality such that although progression free survival; and again, however it's defined is the same, they still have an increased mortality because they may be dying of other things. But that may not be something you're gonna have access to.

Susan Halabi: Right. This is a very good point Alicia, because unfortunately most of these men die of prostate cancer. Looking at some specific data sets at least within the NCTN, we know that most of the men are dying of prostate cancer. And I don't have necessary access to the cause of death, but this is something that I'm pursuing with the sponsor. 

I think the overall picture and the overall message that I wanna give out is, because someone just asked me, would you enroll more African-American men in clinical trials? And I would say, "Absolutely. I will go out of my way, because we know African-American men are doing better on clinical trials. Now of course, it may be a selected patient population and there's biases, but at the end of the day, everyone else is benefiting including African-American men from participating in clinical trials. 

Alicia Morgans: Absolutely. So I think unless we enroll populations across the board unless we get a representative amount of people, of these men, we're not gonna be able to answer the question of whether they have a differential response. At a minimum, it looks like they have at least the same response, which is important for us and we may be able to target things like access to care as we're trying to improve outcomes for our patients more aggressively.

If they are having the same responses to therapy, but still having increased mortality in the overall population, then we need to target multiple things in order to improve mortality. And I sincerely commend you for this work that is adding to a body of literature that suggests that these patients actually can do quite well if we give them the opportunity to have the treatments that can help them.

Susan Halabi: Yeah. Thank you Alicia, and I think it is important to note that at the population level that these differences are real and they exist, but we're not seeing that in clinical trials. And no matter what, I think we need to keep doing this type of work, and we need to target and enroll African-American and other minorities so we understand if they're biologically responding differently to treatment or not.

Alicia Morgans: Absolutely. Well thank you so much for your time and your insights, and I look forward to talking to you again soon. 

Susan Halabi: Okay, thank you, my pleasure.