ASCO 2016: What is the optimal chemotherapy dosing regimen for prostate cancer?
Prof Charles Ryan – UCSF School of Medicine, San Francisco, USA
Prof Nick James – Warwick Medical School Cancer Research, University of Warwick, UK
NJ: Hi, I’m Nick James, I’m Professor of Clinical Oncology at Queen Elizabeth Hospital, Birmingham, and at the University of Warwick and I’m here with Professor Charles Ryan from UCSF in San Francisco and we’re going to talk about the prostate oral abstract session which we’ve just left now. So, Charles, what was your highlight of the session?
CR: Well I think the highlight of the session for those of us who treat metastatic castration resistant prostate cancer is a lot of clarification around the use of chemotherapy, answering some lingering questions about the efficacy of cabazitaxel, answering questions about the dose of cabazitaxel and a little bit about the sequencing and timing of the chemotherapy as well.
NJ: Yes, I think there were a number of questions in there, really, weren’t there? One that, as you say, is the dose and the other is the head to head comparisons, maybe we’ll start with that.
CR: One of the things to say at the outset is that when cabazitaxel was approved in the post-docetaxel setting the United States Food and Drug Administration, as a condition of that approval, mandated that these studies be done. They knew what they were doing, is an important point, because there was a concern that clinicians might look at cabazitaxel and say, ‘Because it is better in the post-docetaxel setting than mitoxantrone it must be better than docetaxel,’ and just make that assumption and begin to use it. So in mandating that study it answers a very important toxicity question, obviously an efficacy question and a cost economics or medical economics question as well. So the FIRSTANA study, which was a nice name, they compared the dose of 20mg of cabazitaxel and 25mg of cabazitaxel to the standard dose of 75mg of docetaxel and essentially saw no difference with respect to overall survival. There were some nuances in the data but the bottom line is we really didn’t see any signal that cabazitaxel at any dose was favourable over docetaxel.
NJ: I entirely agree with all of that; on the headline things they’re clearly the same. So there’s really no justification for flipping the order of the taxanes.
CR: I don’t think so. One of the conversations I had briefly afterwards was some data suggested that cabazitaxel is associated with a slightly higher PSA response rate and what I didn’t see is whether there were subgroup analyses that really seemed to suggest that there was a group that was more favourably benefitted by the cabazitaxel. But basically I don’t think it’s going to change practice.
NJ: No, and I think Derek Raghavan’s discussion at the end was very apt when he said, yes, we know there’s a link between PSA response rates and outcome but it’s not strong enough to say a better PSA response rate means anything very much and the study really shows that very clearly.
NJ: I think the other thing, just in terms of the fine detail of the toxicity, and it’s certainly my impression in clinical practice, is that it does appear that you can give cabazitaxel to patients with neuropathy from their previous taxane whereas you really wouldn’t want to give them more docetaxel. So that was a plus and it is obviously also more myelotoxic. So the other study, the 20 versus 25 study, I thought was very interesting because obviously that was specifically set up to address those sorts of issues and the outcomes.
CR: Right, so that’s the PROSELICA study and it was a two arm study in the post-docetaxel setting looking at 20mg and 25mg of cabazitaxel. One slight nuance between the two studies was in the FIRSTANA study, the front line study, there were almost no patients who had had prior abiraterone or enzalutamide, which I thought was really interesting. It speaks to when the study was started and where it was conducted, relative to the approvals of those two drugs. So we really can’t make any observations there. However, in the PROSELICA study there were a fairly substantial number of patients, I think it was about 45% of patients, who had received either abiraterone or enzalutamide. And one piece of data that came out, so the headline data was 20mg is just as good as 25mg with respect to overall survival, a little bit less toxicity in the form of neutropenic sepsis and diarrhoea.
NJ: Quite significantly less, yes.
CR: But anyway there was an observation that perhaps patients who had had prior abiraterone might have benefited more from the 25mg but a pretty wide confidence interval so I’m not sure we’re going to move forward with that as anything that is going to change practice.
NJ: Absolutely. And so the PROSELICA study, as you say, is more the current real world setting because it was cabazitaxel in the second line chemo setting in patients who had had prior exposure to an AR targeting therapy of the new generation ones as well. I think the other thing that is very interesting, just going back to the FDA and the regulators mandating the study, it shows that the regulators are acting very much against the pharmaceutical companies’ interests there because that study must have cost them a lot of money. And selling the drug at a lower dose will presumably cost them some money as well. But the other thing was that Johann De Bono’s comments in the beginning of his presentation, that he’d obviously been involved with the study with the drug for a long period of time, going back to the phase I trials. He kind of hinted that it looked to him like 20mg might have been the right dose all along, which would imply an awful lot of time and energy and money has been wasted on nailing a dosing question that really could have been nailed a very long time ago and at much lower cost.
CR: Certainly tens of millions of dollars, if not more than that. That is true; I think that when we look at these population studies and we look at several hundred patients treated and we look at Kaplan-Meier plots we tend to make a broad determination as to the efficacy and safety of things. But one of the things that we aren’t doing on a broad enough scale that we should be doing is looking at individual pharmacokinetics, and this came up in the discussion. There are probably patients who have a lower plasma level of a cabazitaxel type drug for the 20mg dose and could and should be treated with 25mg. I would hope that we don’t give up the opportunity to leverage the fully amount of benefit from a chemotherapeutic by assuming that all patients across the world metabolise the drug at the same rate and that there is a single dose. So I would hope that there are some studies underway or planned that would look at dose titration and those types of things.
NJ: Interestingly, both Oliver Sartor and Johann De Bono hinted that they actually will start at a lower dose and escalate, rather along the lines you said. Slightly tangentially I was giving a talk in Tokyo about three months ago and I got a very strong sense, it was Japanese urologists, that they thought that prostate cancer behaved differently in Japanese men to Caucasian men and that, considering the dosing, obviously for things like the oral agents you use a fixed dose. They pointed out the average American is 80, 90, 100kg…
CR: And growing.
NJ: Yes, and getting bigger. The average Japanese is 60-65kg and with different distributions of fat and muscle and so on. So it is very likely that metabolism is going to be different in populations as different as that. And I don’t think we pay enough attention to it, you’re right.
CR: Absolutely, and I’m sure your practice is similar to mine in that I have some very young patients who are very fit with horrible prostate cancer and could probably take a higher dose and would prefer a higher dose if I could give it to them. So I think that we do need to do this. The other interesting point is we can go back over a decade now, almost twenty years, and look at docetaxel. We had two phase III trials with docetaxel, one at 75mg/m2 and one at 60mg/m2, they both demonstrated a survival benefit over mitoxantrone and that’s a fairly significant difference in dose when you think about it. So I’ve always wondered if we really can say that those are equivalent doses or whether the standard dose of 75mg/m2 is something that should be rethought. And that came up in the conversation today as well.
NJ: It did, yes. Yes, a sort of one size fits all dosing is, for sure, simplistic. The other studies that were there, there was the intermittent versus continuous therapy study which obviously had failed to accrue to its target, did meet one of its endpoints but you could question whether the endpoint was the right endpoint – one year survival. It was done in a pre-abi/enza era. Do you think we can take anything away from that? Do you think we should be considering intermittent taxane therapy?
CR: I’m not sure. We’ve done intermittent chemotherapy and I think that it’s a practical consideration in that it allows you to deliver more drug over time. Of course, in breast cancer it has become… there are different dosing schemes that allow you to take breaks and things like that, that allow you to deliver more drugs. So I would say that my centre, for example, in the use of docetaxel intermittent is almost our standard way of doing things. Typically what we will do is if a patient is identified as having a very good response we might give them a break after six cycles unless they don’t want to or don’t need it. So I think it answers a practical question around this topic as well.
NJ: Yes, we published a paper ourselves, actually, it was quite a long time ago now, pre-abi/enza, essentially showing that if you stopped while patients were still responding almost all of them would respond to re-challenge.
CR: Yes, well there’s two points. We showed that as well. So what we also noticed is that the re-challenge occurred relatively quickly. In other words, when you have a patient’s disease under control with chemotherapy and you stop it, they might only get two cycles of a break. It’s not like intermittent hormone therapy where you take them off for a year, you’re talking about sparing them maybe one or two or three or four cycles of docetaxel.
NJ: But that might be quite important because these are patients in the last couple of years of their lives and there may be an event they want to go to. They don’t want to be up to their eyeballs in chemotoxicity. To me, that intermittent study did seem to confirm what I felt was the case, obviously you do as well for your own practice, that it’s perfectly reasonable to interrupt therapy and there didn’t appear to be any sort of survival penalty of any substantial size from doing it. Even though the study had all sorts of problems I thought it was an important message, really.
CR: There’s also a biological issue, though. I totally agree, practical and quality of life issues make intermittent chemotherapy something that’s attractive. Emmanuel Antonarakis now has shown some data to show that in AR-V7 patients who receive docetaxel, a certain number of them will clear their AR-V7. So, just to reiterate, we think very strongly that the AR-V7 patients don’t respond to abiraterone enzalutamide, they do respond to docetaxel. So if you can show that you could take an AR-V7 positive patient, essentially clear that, then you can go back to the second line hormones.
NJ: Yes, I was there in that session, I thought it was an interesting observation. I had a long chat about this with Gerhardt Attard yesterday as well. The experience generally is that you don’t get fantastic responses to abi or enza post, as you go down the lines the responses get less and less, don’t they. The Antonarakis AR-V7 test, if I’m right, is on CTCs. So if you have a good response to taxane you clear the CTCs so you will clear the AR-V7 but you may not really be clearing the AR-V7. So the reason I was mentioning Gerhardt’s name is that they’ve developed a test based on circulating free DNA which has a less discrete cut-off in the sense that you can detect things lower levels. They’ve got a digital droplet assay that is very sensitive. So we’re designing an AR-V7 study at the moment, actually it’s funded, but we’re going to use that test rather than the CTC one to randomise patients with an AR-V7 mutation. But we’ll be able to track the circulating free DNA. We’ve also within STAMPEDE just initiated a circulating free DNA study as well. So we’ve got all these various patients going through various treatments, we’ll be able to see, and it’s on the DNA not on the CTCs, what the impact is probably at a more sensitive level than the CTC based assays.
CR: That’s great. What it gets to is the bigger picture of the iterative response of the biology of the tumour to what we do to it. So we give abiraterone or enzalutamide or docetaxel and we cause biological reactions within the tumour that make it into a different disease. We’re seeing this with our biopsy programme where we’re seeing these neuroendocrine tumours emerge, small cells, that kind of thing. One of the really interesting questions that I have and I’d love to get your thoughts on is what do we know or think about the biology of the disease in patients who get early chemotherapy in STAMPEDE, and that kind of population, when they recur and go on to subsequent therapy, is the biology of the disease different?
NJ: We have data, we haven’t fully cleaned it up and analysed it yet, but the one observation I would make from the STAMPEDE data, and it’s similar in the CHAARTED data is that the delta on the failure free survival, it’s not just a frame shift. So the failure free survival delta in the metastatic subset in STAMPEDE is about six months, it goes from about a year to about eighteen months failure free survival. The delta on the survival is eighteen months so it would appear that once you’ve relapsed things happen more slowly than they did in the control arm because the gap between the relapse and death is bigger in the docetaxel arm than it is in the control arm.
CR: So either the disease is moving more slowly or the clinicians are observing them for a longer period of time before initiating therapy?
NJ: We know that’s not the case. So we plotted out time to first treatment following failure free survival events and essentially the time to the subsequent treatment from failure, obviously failure is occurring later but the time from the point of failure is exactly the same. The distribution of treatments differed in that patients who’d had docetaxel up front were more likely to get abiraterone as their subsequent treatment whereas patients who had had the control arm were most likely to get docetaxel. That was partly related to the timing of when the study was happening and the agents that were available. If you look at the later cohorts of patients where they could in theory have had abiraterone or docetaxel then the number of patients getting abiraterone is higher. But if you lump all the life-prolonging therapies together, the time to first life-prolonging therapy from failure is the same in the control and the experimental arm. The follow-up schedule was mandated so there doesn’t appear to be any sort of artefact created by differential follow-up in STAMPEDE.
CR: Interesting. I predict that we’ll be seeing more presentations from the STAMPEDE data and others looking at these kinds of questions and further clarifying the sequencing. It’s interesting stuff.
NJ: Thank you very much, I think that concludes our discussion for the time being. It’s been a pleasure talking to you.
CR: It’s been a pleasure.