The Correlation Between Alkaline Phosphatase Decline and Overall Survival in Men with Metastatic Castration-Resistant Prostate Cancer in the REASSURE Study - Nicholas James

January 5, 2023

In this discussion, Nicholas James and Alicia Morgans discuss the relationship between alkaline phosphatase decline and overall survival in men with metastatic castration-resistant prostate cancer (mCRPC) and symptomatic bone metastases treated with radium-223 (Ra-223) in the global real-world REASSURE study. REASSURE is evaluating the long-term safety of Radium 223 in routine clinical practice in patients with mCRPC over a 7-year follow-up period.


Professor Nicholas James, MBBS, FRCP, FRCR, Ph.D., Professor of Clinical Oncology at the Institute of Cancer Research at Royal Marsden Hospital, London

Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, Massachusetts

Read the Full Video Transcript

Alicia Morgans: Hi, I'm so excited to talk today with Professor Nick James, who's joining me to talk about the REASSURE study, which is a cohort study, looking at patients treated with radium-223. Thank you so much for joining me to talk about this today, after your presentation at ASCO 2022.

Nick James: It's a pleasure. Always a pleasure chatting.

Alicia Morgans: Wonderful. So can you tell us a little bit about what you and the team presented, related to the REASSURE study at ASCO 2022?

Nick James: So prior to the REASSURE study, I was one of the authors and lead recruiters to the ALSYMPCA trial, which was the licensing trial for radium. And, that trial was obviously carried out in a rather different era. No abiraterone, and sipuleucel, any of these other drugs, which had shifted the order of treatment around. And one of the other things that came out from the ALSYMPCA trial, was that although patients had a survival benefit in that trial, that didn't correlate that well with changes in PSA.

So people struggled with how best to make treatment decisions for patients on radium. For a number of reasons, it's just a registry study, documenting what happens to patients having radium-223 was set up. And as you said, it's a large study. There's 1400 or so people in the thing together. And one of the things that clinicians treating with radium have observed, and I know we were chatting about it before this recording started, is that changed in alkaline phosphatase appear to correlate well with patients who are doing well. So, we set out to do this analysis, to see whether our clinical impression that changes in alk-phos actually did quantifiably correlate with a good or a bad outcome, basically. So that was the underlying rationale for this.

Alicia Morgans: Well, and I think that's so important, because as you said, PSA can be difficult to follow and some patients clearly are clinically benefiting, but there are a lot of patients where it can be less clear whether things are going in the right direction or not. And so what exactly did you analyze and find in terms alk phos?

Nick James: So the hypothesis was that if you get a fall in your alkaline phosphatase, that that would indicate patients having a good response. And furthermore, we split the patients up into patients with a normal alkaline phosphatase at baseline and patients with a raised alkaline phosphatase at baseline. Now, clearly the extent to which you've got an alkaline phos is likely to relate largely to bone disease. I mean, we assume that liver disease doesn't change very much during, or liver disease won't change very much. It's not a hepatotoxic drug. So therefore if you get increases or decreases in alk phos, it's likely to reflect differences in the bone outcomes. But also we hypothesize that your alk phos per se would be prognostic at baseline. So there's two separate things. One, is alk phos a baseline prognostic? And the second is, are changes in alk phos from baseline also prognostic? So one's a static, one's a dynamic biomarker, if you like.

Alicia Morgans: Great. And, and what exactly did you find when you looked at this within that 1400 patient cohort?

Nick James: So what we found was that we had to restrict analysis obviously to patients who had a PS, an alk phos, not a PSA, had an alk phos baseline, an alk phos at 12 weeks. And so that allowed us to split, identify around 700, well, 779 out of 1400 patients. Roughly half of the patients have those readings allowed us to split it up. And they split roughly down the middle. 440 or so, and 330 with a raised alk phos. So looking at the baseline readings, it was very clearly prognostic. If you had a raised alk phos, you had a worse outcome. But if you then had the second alk phos at 12 weeks, that further split the things in half. So basically, if you had a raised alk phos and no fall in your alk phos at 12 weeks, you had a very short median survival... You were destined to do very badly.

So median survival was around eight months in that group. If you were within the raised alk phos group, and then the alk phos fell, that survival went up to over a year. Went up to 13 months, basically. So a fallen alk phos predicted a much better survival than no falling alk phos. So I think that... You're not going to make a decision purely on that, but I think in the context of other things, how the patient's doing, pain, general condition and so on, we think it's a helpful thing. Because particularly if you've got soft tissue disease will be present, you're allowed soft tissue disease with radium and it won't be being treated. So it will be making PSA. So lymph nodes, low volume lymph node nets, or the prostate itself. So this is helping you to assess the bone disease, which is what you're, of course, targeting with radium. Now where we thought the association might fall apart is in the patients with a normal alk phos.

But it turned out that the same association held there. So any fall in alk phos at 12 weeks, again, predicted a much better outcome. So the median survival of patients with a normal alk phos than therefore presumably less bone disease was 16 months with no fall. I presume the poor response that goes up to 23 months, if you do get a fall. So you've got a fairly easy way of A, splitting patients up who are likely to have a very poor prognosis from the ones that might have a better prognosis, but also once you've started treatment, independent of other factors, you've got a way of helping you with your decision making. There's a few other things that we looked at in the multi variable analysis in terms of things that might help. So the more, not very surprisingly, but the more prior therapy you had, the worse your prognosis was, but also other things like the higher your PSA at baseline and the lower your hemoglobin at baseline again, very unsurprisingly predicted that you were going to have a worse outcome.

So patients with low hemoglobin and a high PSA and more prior treatments do worse than the reverse of that. So the other prognostic practice behaviors you expected, but the most important thing was that the changes in alk phos based and the baseline alk phos behaved very quantifiably as we predicted they should behave and coinciding with our clinical impressions. I know your clinical impressions as well from chatting pre-interview.

Alicia Morgans: And I think as we talked about, this is I think so important. Certainly we can encourage patients who have a normal alk phos that a decline is actually of value. And I actually distinctly remember a patient's wife saying to me, "Look, his alk phos was here and now it's here." And I said, "Oh, well, it's normal. And it just decreased within the normal range." But now I think it's really interesting to know that actually was a suggestion of that patient having actually quite a good prognosis, as you mentioned from the survival data. So we can really come to patients and help them predict and plan, "What am I to expect?" Of course, nothing is absolute, but really I think useful information as we're talking with patients.

Nick James: So yes, we think this is quite a useful adjunct to decision making in a pretty palliative end of life care sort of setting. And not only is PSA, not very useful in end stage care, but also scans are quite hard to assess. CT and bone scans, very hard to assess what's happening in bone on either of those modalities. So we are using, as other people are, whole body MRI as a way of trying to more objectively of assess changes in bone in patients on trials and stuff. But I think it's not really very widely available outside of the clinical trial setting.

Alicia Morgans: Well, and it's not necessarily either here in the US. I do think that our colleagues at Memorial and some other places are using them more routinely, but generally it's not been as widely used and so many reasons for that. But to your point, if we have a way of adding to our understanding of how patients are doing when they are so challenging to really follow and monitor in terms of response or progression, I think this is really quite a simple way to go.

Nick James: Exactly, yes.

Alicia Morgans: Certainly useful. So if you had to sum up the work and what the final message would be to the audience, what would that be?

Nick James: So two things. The alkaline phosphatase baseline is refract for tumor burden and a higher alk phos equals more tumor equals a worse prognosis. And the second thing, irrespective of what alk phos you start with, is if it comes down with treatment, it's likely to indicate treatment's working better and the patient's prognosis is better, therefore, than if the alk phos goes up after you start treatment. So really simple things. Really easy things to measure and not a lot of difficult stuff to remember.

Alicia Morgans: But really, I think, so clinically applicable and really helpful to patients and to clinicians who are trying to make those choices in everyday practice. Thank you so much for your continued work and for going through this with us today.

Nick James: Absolute pleasure.