DR NEIL LOVE: Dr Gradishar referred to another poster presentation at San Antonio by Dr David Miles following up on the capecitabine-docetaxel trial presented at last year's meeting. A few months after that presentation, when I interviewed the principle investigator of the study, Joyce O'Shaughnessy, she noted that a frequent question she was receiving was whether a sequential approach to using these agents might have yielded the same result. While she noted that the early separation of the progression-free and survival curves in the capecitabine-docetaxel arm suggested that the biochemical synergy of these two agents favored a combination approach, she believed it would be worthwhile to investigate in detail, the therapies and outcomes in women randomized in the study to docetaxel alone. These data were the focal point of the poster presented by Dr Miles, who began our discussion by reviewing the background to the study.

DR DAVID MILES: In many ways, the development of docetaxel in advance disease and then docetaxel with Xeloda has had a resonance with us because the original studies looking at docetaxel in Phase III compared it with mitomycin c-vinblastine, which rightly or wrongly, was very much our standard anthracycline-failed regimen. So, when the Nabholtz data came out showing an advantage there, that rang bells with us, and then, when the Xeloda trial came along we were interested in contributing to that, which we did. Clearly, the survival benefit has been demonstrated. So, while survival benefit of advanced chemotherapy is not something that we talk about a great deal, we talk about the palliative benefits of chemotherapy. But if you look over the last few years, we've gone from median survival of 9 to 11 1/2 to now 14 1/2 months with this combination. This poster we've got here really is about the treatment after progression as part of that Phase III trial.

DR. LOVE: I guess one of the questions about that trial was that there wasn't a structured crossover.

DR MILES: Sure. I think that's increasingly the case these days. You look at treatments like Herceptin and you can't mandate beyond crossover really. That has difficulties for then interpreting the attributable benefit of giving a treatment up front. In this case, a lot of patients went on to have other chemotherapies. As you might expect, those patients who were randomized to single-agent Taxotere (docetaxel) originally, quite a lot of those patients went on to Xeloda as second-line therapy, if you will. What I think is potentially interesting about the data, actually, is the fact that those patients who went onto Xeloda, actually did better than patients that went onto other therapies. So, if you look at the odds ratio and the hazard ratio of dying, if you had Xeloda as salvage therapy, it was about half the risk in patients that received other chemotherapies.

DR. LOVE: The most interesting comparison from a practical point of view in terms of the kinds of things that would commonly be done would have been Xeloda vs. vinorelbine. Sure you can go back to another taxane, you can go back to anthracyclines again, but I think one common choice in that setting would be Xeloda vs. vinorelbine.

DR MILES: I guess so. I think the difficulty with vinorelbine is that it is essentially a spindle poison, same as docetaxel. I think there are a couple of fairly structured Phase II studies of vinorelbine following taxoid failure. One shows not a lot of response, and one shows some response. The fact is, by the time you fail anthracycline and fail taxoid - conventional therapy - there aren't a heck of a lot of options left. When you're talking about going back on taxoids even in those who failed, I guess it is looking at different schedule, say weekly Taxol or weekly Taxotere where the three-weekly regimen has failed. I think a lot of people use vinorelbine, it's used, I guess, more commonly in Europe. It's something that people do use, I guess particularly in the more elderly patient population perhaps. I think the point is, when you look at vinorelbine compared to other chemotherapies following failure of Taxotere in the context of this Phase III study, vinorelbine isn't associated with any difference in hazard ratio of death, whereas Xeloda is.

DR. LOVE: Compared to?

DR MILES: Compared to any other therapies. If you take vinorelbine out and compare it to everything else, vinorelbine has no greater advantage over anything else, where Xeloda appears to be.

DR. LOVE: So, basically you're just looking at the survival of patients who initially got Taxotere, based on what subsequent therapy they got?

DR MILES: Correct.

DR. LOVE: I guess one question would be: If all the patients had received capecitabine, what do you think you would have seen?

DR MILES: That in a way is the crucial question. Whether or not combination therapy is actually any better than sequential therapy. This was not something that was ever addressed by this trial, and maybe that could have been addressed by this trial. I think the thing that makes people concerned about not using the combination in a patient that is fit enough to receive it is, when you look at time to progression and survival from the original Phase III study that Joyce O'Shaughnessy has reported, the progression-free and overall survival curve separate very early. I think there is a problem there, that you may potentially run a risk of losing the advantage of the combination if you don't use them together. I think the message of this poster presentation on post-progression therapy is Xeloda is probably the most potent agent there is, at least when you look at it in survival terms compared to other therapies.
If you have a patient in whom you're concerned about using the combination because of points such as toxicity, then it may be reasonable to use them sequentially rather than in combination. I think that until we have a trial that looks at combination vs. sequential, we really won't know to what extent the combination is better than sequential.

DR. LOVE: The question that's starting to come in my mind, is what drug should come first?

DR MILES: Well, I think that is a really good point. I guess without the studies, we are not going to know. I wouldn't disagree that Xeloda itself is a very active drug. Whether people would ever feel comfortable using it before a taxoid, I'm not really sure. Certainly in raw response rates terms, it may not be quite as good as using docetaxel, but I wouldn't disagree that there are a group of patients where you might want to use it as a single agent in whom a taxoid is not suitable.
I think that there are two main points I would make. The fact is, you can't mandate post-progression therapy, but in fact, 27% of patients who had any chemotherapy following relapse of single-agent docetaxel had Xeloda. You might argue that the survival benefit that was reported from the Phase III trial may have been underestimated. There is a significant amount of crossover. I think that's the first point to make. That three-month benefit that you're seeing, how often do we see improvements in survival in metastatic disease? Not often. There is a benefit there, and it may have been underestimated.
The other point that is worth making is the fact that the curves separate early. So, I think there is an anxiety that the survival benefit that you are beginning to see, you may miss out on that if you do use them sequentially. I guess that until we've got sequential data, you've got data that says the combination is better than docetaxel on its own.

DR. LOVE: How are you taking these data, right now, into your practice and managing patients with metastatic disease off-protocol?

DR MILES: It probably suggests to us, that in patients who are young, fit and have visceral disease where survival is a serious issue for the patient, rather than necessarily palliation, the combination should be reasonably considered. I think if you've got patients who are less fit, perhaps an older age group, then you may not feel that way. The survival data are there. I think it is a choice that you would then sit down with a patient and make together.

DR. LOVE: The other issue is, you have the response rate data and that may be just as important in terms of getting control of the tumor.

DR MILES: I think that's right. That's reflected obviously in the progression-free survival data as well. So that they split really very early. I guess as you say that's down to differences in the response rate.

DR. LOVE: How much toxicity do you think adding in capecitabine adds to Taxotere?

DR MILES: They are actually non-overlapping toxicities. That's a phrase we often use when using combination therapy. It's a phrase that actually applies relatively rarely in truth. But at least with Xeloda, really the myelosuppression is negligible. That is fine except for the fact that you do have GI toxicity and the hand-foot syndrome. So, sure they're not overlapping, but it's kind of like piling on toxicities of a different kind. I think at 1250 that did prove quite difficult. Clearly, handling Xeloda is very much a function of renal capacity. So, in the over 60's starting at 1000 is really mandatory, and that might well be so for those younger patients with docetaxel as well. I think docetaxel, as we said earlier, is a drug that we have - I guess we can understand the data of docetaxel a lot better than perhaps some of the other taxoids because it was originally compared to a regimen that we regarded as standard. However, when you look at the MTD's of docetaxel around the world, basically on different continents, they are very different. A hundred of docetaxel, we find not always easy to get in, would often then start at 75. Because then coming in at 75 with a completely different drug, we've added the different toxicities. I think it is a cause for concern as evidenced by the number of patients who underwent dose reductions in the Phase III trial. But, despite the dose reductions, you're still seeing survival. I think that's what still sticks in the mind very much.
The stock phrase that oncologists use is that toxicity is acceptable. That is not something I like to say a great deal. I think that toxicity, in this instance, is manageable. With the dose reduction that occurred in the trial, and despite that reduction, you are still seeing survival benefit. I think it is a very real option. Is it the new standard of care? I think people are anxious about it because of the toxicity element, but if you're talking about pushing out the boundaries of survival in this group, then I think it has to be considered the standard of care.

DR. LOVE: One of the things I hear from people is the excitement that this study has generated in terms of taking this combination into the adjuvant and neoadjuvant setting. It looks like that is happening a lot right now.

DR MILES: Yes. I suppose in a way Xeloda is what everyone had hoped for. People in the UK, obviously Ian Smith at the Royal Marsden, did a lot of work with infusional 5-Fluorouracil with ECF regimen with very good-looking response rates. Here you've got an agent that essentially gives you continuous infusion 5-Fluorouracil but without the Hickman line.

DR. LOVE: The other thing is, as I've been thinking about it more and talking to people more, it seems to be a lot more than just 5FU because of the intratumoral concentration. If you buy into that, it seems like a legitimate phenomenon.

DR MILES: I think that's right as we said a little bit earlier. If you look at the number of patients, 3/4 of the patients who had already had 5FU who went into this study, then I think it attests to the fact that there is something over and above this being just another 5FU. As you know, thymidine phosphorylase is higher in tumor than it is in normal tissue, and probably also induced by drugs like docetaxel, paclitaxel, other chemotherapy drugs.
There is some very interesting work in recent months coming out and looking at the scheduling of docetaxel and Xeloda in animal models. Whereas here we had a situation where docetaxel was given on day one, Xeloda days 1-14. I think it was a Japanese group that looked at various timings of the two drugs. In fact giving docetaxel on day eight, and Xeloda days 1-14 seemed to be better than giving it day one. In that respect, while the thymidine phosphorylase story is quite a nice neat story, these scheduling data which suggest actually coming in with the docetaxel a little later. Presumably if it is not upregulating TP till later, then maybe the TP isn't the whole story and there is something else going on. Certainly, I agree entirely, when you look at the amount of 5FU pretreatment that a lot of these patients had and you look at Xeloda as a single agent in patients who've had 5FU in the adjuvant setting, there is some added value with Xeloda compared to 5FU alone.

DR. LOVE: Are you involved with, or going to be involved with, any studies in the adjuvant or neoadjuvant setting looking at this combination?

DR MILES: I think we will. In the UK, again Ian Smith, who did some of the original infusional fluorouracil work, showed very nicely that ECF where 5FU was given by infusion was better than AC. The second trial they went onto was one that was perhaps less interesting. It was looking at Navelbine and less toxic approaches. I think, all the while it was in the back of Ian Smith's mind that when the oral fluoropyrimidines came along then that was something very much they would go back to as a more convenient schedule. Obviously groups like the EORTC, with which we have been involved with in the past, and also looking at all oral combinations: cyclo, idarubicin, Xeloda clearly then GI toxicity may become something of an issue. I don't think it's - although the studies all say, patients prefer to take oral therapy, perhaps when you start piling on multi-agent oral therapy, it may not be quite as clear cut. But, I think these are exciting times for those orally bioavailable drugs.

DR. LOVE: What do you see as the potential role of capecitabine as a single agent, before receiving a taxane? Let's say in a patient who has had adjuvant CA or an anthracycline-containing regimen. Are there patients that you think that capecitabine as a single agent might be used before a taxane?

DR MILES: I think it could be. Perhaps in the older patient, the patient with a poorer performance status, I think that's a reasonable thing to do. I guess the other group of patients would be those who have what one might describe as non-life threatening disease. As we said, when you look at response rates, you made the very good point about the difference in response rate. If you got someone who is symptomatic or in whom their liver function is now deranged or their pulmonary disease is symptomatic, we probably want to maximize the response rate and get their responses as quickly as possible. In patients where that isn't so much of an issue, they might have failed that hormonal therapy, or you could conceive of moving them onto another oral therapy, which just happens to be a cytotoxic agent.

DR. LOVE: Actually now that I think about it, maybe it would have been difficult because it would have required more patients, but it might have been interesting to set this up as Xeloda-Taxotere one arm, Taxotere in the second arm and then Xeloda in the third arm. Where your single-agent arms then crossover to the other agent, because I would be really curious to know what the survival difference would be between starting with Xeloda and then following it with Taxotere compared with the reverse.

DR MILES: I guess as long as the patient is fairly closely monitored, then it may well be that there is not a great impact on survival. I guess the concern is, when you look at these data and you've got patients, not all patients went onto other therapy after the drugs that failed in their particular instance in this trial. I think then you've always got the concern that as patients get less fit, as the disease advances, then you loose those opportunities. So that mandating the crossover seems like a good thing to do, but there is still quite a high failure in crossing over at all.

DR. LOVE: Absolutely, and if there were a study like the one I described, it wouldn't surprise me at all if the best arm would have turned out to be the combination arm. That's kind of what we think right now anyhow. But, the second issue, though, to me that would have been very interesting would have been again, if you aren't going to use the combination, does it matter which one of these you start first.

DR MILES: I think that's a good question. Maybe it is the kind of physician sponsored studies that we should be doing with all the drug company sponsored studies that we need to do to keep our departments afloat.

DR. LOVE: I think the assumption going in at that point was probably that Taxotere was going to be a more active agent than Xeloda. I guess that's the case, but they've never really been compared head-to-head.

DR MILES: No they haven't. I think we have in our heads, I even have in my head when talking to patients, the docetaxel data from Peter Ravdin and others. Lets remember that this drug got a license from Phase II studies with a response rate in the mid 40's. As ever, when that broadens out to the Phase III studies and less selected populations the response rate now is 30 possibly a little less, particularly when you start to take dose reductions into account. I think it is a good point. We are very wedded to the idea that more toxic necessarily means better and that may not be the case at all.

Home · Search