DR TERRY MAMOUNAS SUPPLEMENT

DR LOVE: You said that right now your intention is to start a new study that will compare AC followed by Taxotere compared to AC followed by Taxotere and Xeloda. What were your thoughts in terms of setting that trial up?

DR MAMOUNAS: Well, our thoughts were to try to improve upon what we’ve learned from B-27. Initially, we were looking, actually, at combination regimens of the taxanes and anthracyclines, but once the B-27 response data became available we felt compelled to use that as a control group because, essentially, we’ve doubled the pathologic response. So, now we want to see whether with Xeloda, we can go up to, let’s say 35 or 40 percent.

This trial also is a very important trial because we’re going to assess a lot of biomarkers before and after chemotherapy with core biopsies in a sequential fashion. And I think the approach that we’ve taken with this is to see whether we can identify molecular biomarkers with DNA microarrays and high through-put technology that will predict what type of response the patient will get to certain chemotherapy regimens.

Now, of course, the best way of doing that would have been to start different groups of patients either with AC first, Taxotere first, or Taxotere-Xeloda first, and see who responds to which, or even sequentially. We’re not really doing that yet. But that may give us some indication that we may pursue that in the future. Once you’ve isolated which patient responds to a certain chemotherapy drug, then you can set up a trial where you treat those patients with that chemotherapy drug only, and you would expect a very high response rate. So, the response rates, if you can find a pattern of gene expression, can go up to 80 to 90 percent because you know that these particular patients will respond to that regimen.

The other approach that we’ve taken in that trial is also to see whether we can avoid doing some of the extensive surgical procedures, not only in terms of lumpectomies, but more importantly, in terms of sentinel node biopsy alone. Based on the data that we have collected from B-27 and B-18, we feel comfortable allowing sentinel node biopsy alone in patients that achieve a pathologic complete response. So, we will try to assess whether patients have a pathologic complete response where we would expect that the nodes are positive only in about 10 percent of the time and that incidence of 10 percent minimizes the potential error of the sentinel node biopsy. In other words, if you have a 10-percent error on your sentinel node biopsy and a 10-percent chance of having positive nodes, you’re overall error is about one percent. So we feel comfortable, in those patients, to avoid doing an axillary dissection. That fits the whole theme that we have developed now, to use neoadjuvant chemotherapy to further reduce the extent of surgery, not only in the breast, but also in the axilla.

DR LOVE: You mentioned the experience of sentinel node in the preoperative B-27 study. What exactly did you see there in terms of sentinel node?

DR MAMOUNAS: We were putting these patients on trial and we figured why not do a sentinel node in these patients sort of outside of the protocol followed by an axillary dissection, which was required by the protocol. So, in about 400 cases where we were able to do sentinel node first, followed by axillary dissection. It turns out that we can identify the sentinel node about 85 percent of the time. So, our success rate wasn’t stellar, but it was okay. It was actually higher in cases where we used radioisotope and blue dye together. It was very much lower if we used blue dye alone. So, with the blue dye and radioisotope, we have about a 90-percent identification rate.

Our false-negative rate, overall, was about 11 percent of all node-positive patients. This is not all that dissimilar from the early studies of the multi-center trial by Krag, which had about an 8 percent. Obviously, with a very well monitored and set protocol that false-negative rate would drop. It’s very clear that it will because that represents the worst-case scenario. The surgeon didn’t have to find the sentinel node, didn’t have to look hard for it. I mean, you could just open up and if you didn’t see it, you’d abandon it. But the fact is, even with the worst-case scenario, we had a reasonable false-negative rate.

DR LOVE: You mentioned that you’re going to be getting biomarkers on the new neoadjuvant study. And, of course, that’s always really fascinating. I assume one of the things you’re going to want to look at is thymidine phosphorylase in terms of the fact that, theoretically, the synergy between Taxotere and Xeloda, is Taxotere upregulating TP.

DR MAMOUNAS: Yes. Absolutely, both that, as well as thymidylate synthase and all the markers that predict, perhaps, for 5FU sensitivity. We’ll certainly be looking at those. But, as well, we’ll look at all the other markers, the common markers, as well as the microarrays.

DR LOVE: Are you using basically the same doses that were used in the randomized trial of capecitabine and docetaxel or are you decreasing them?

DR MAMOUNAS: We actually are decreasing the dose of capecitabine to about two grams per meter squared because of the severe instances of hand-foot syndrome. Whether that will be something that the NCI will accept or not remains to be seen. I think one of the comments that we had was that we have to justify using the lower dose because perhaps the efficacy is reduced. Although there’s good data, I think, to support that the efficacy is not reduced.

DR LOVE: Well, also, I think a lot, if not a majority of the patients in that study, in the capecitabine-docetaxel study, actually had that dose reduction by the second dose anyhow. So, essentially, that’s what they got.

DR MAMOUNAS: Right. Exactly, that’s what we felt. For the adjuvant setting, I think, that is reasonable. Even aside from all that, for the adjuvant setting, to reduce the dose a little bit of a drug from the maximum tolerated dose, so to speak, I don’t think is very unreasonable. We’ve done it with the Taxol study where we used 225 milligrams per meter squared, where you can easily use 250 or 300 as a three-hour infusion.

DR LOVE: I’m curious whether or not there’s been any discussion within the NSABP about looking at capecitabine as a single agent in adjuvant therapy. Hy Muss has a trial in older women, comparing either CA or CMF to capecitabine. And there’s been discussion about whether or not that might even be appropriate to look at in younger women. Has that been discussed in the NSABP?

DR MAMOUNAS: We have had several discussions regarding launching a trial in the elderly. We actually had proposed an elderly trial back in 1994 or ’95, with Navelbine at the time. And, actually, the response of the NCI was that, clearly, it wasn’t very enthusiastic about us using an agent that wasn’t even approved for the treatment of metastatic breast cancer as adjuvant in the elderly.

We haven’t crystallized our thoughts yet as to the best approach. My concern with Hy Muss’ trial is that they’re comparing two different chemotherapy regimens. And the biggest reason that we’re not treating patients over 70 with chemotherapy is because oncologists have not been convinced that there’s a benefit. The overview analysis shows a decrease in benefit in relative reductions with every decade of life, and certainly does not have many patients over the age of 70. So, I think the bigger holdup of using chemotherapy in those patients is that oncologists have not been convinced that it works.

Certainly, a lot of those patients now have good performance status. They have no co-morbid conditions, and yet there’s some reluctance, particularly for the ER-positive patients, where everybody puts them on tamoxifen or, now, an aromatase inhibitor, and you’d probably say that they don’t need chemotherapy. I think the concept-proven trial in this population in my mind is to compare, actually, tamoxifen alone to tamoxifen plus chemotherapy.

DR LOVE: Hmm.

DR MAMOUNAS: In essence, redo NSABP B-20 for elderly patients. Because, if that trial shows that the survival is improved, then we can clearly say, now we know chemotherapy works. Let’s find an equivalent chemotherapy regimen with less toxicity.

In the Hy Muss study, if both AC and CMF are equivalent to Xeloda, one would not know for sure how much the Xeloda or the AC and CMF contributed to the overall survival of these patients. And that’s our biggest concern with this design. We’ve reiterated several possible designs: weekly taxane or AC or CMF or methotrexate 5FU/leucovorrin schedule used in B-13, which was a pretty reasonable regimen, epirubicin, as a more gentle Adriamycin. But I think, certainly the concept remains to be proven that chemotherapy works well in these patients and what kind of survival improvement will we get.

DR LOVE: Now, you talked about looking at tamoxifen versus tamoxifen plus chemotherapy, which is very interesting. Why would you use tamoxifen, though, as opposed to anastrozole?

DR MAMOUNAS: Yeah. I mean, maybe that is a bad choice of hormonal therapy. At this point, I think you can use either/or, or perhaps even an aromatase inhibitor. But my whole concept is hormonal therapy with or without chemotherapy, I think that is the question that we need to answer.

In the ER-negative patients in this group, I think most physicians will be comfortable treating with chemotherapy because you have no other choice. And in those cases, of course, you could compare two different chemotherapy regimens, and I think probably the Hy Muss trial would have been a good trial for the ER-negative patients. But, for the ER-positive patients, I think the concept still remains to be proven.

DR LOVE: What other trials are you doing?

DR MAMOUNAS: We also have a node-positive trial for HER2/neu-positive patients evaluating the use of Herceptin as an adjuvant. This is a two-arm trial of AC followed by Taxol with or without Herceptin. Herceptin starts with the first dose of Taxol. This, as you know, is a study where very strict criteria for the evaluation of cardiac toxicity are set because of the potential cardiac toxicity with Herceptin following an anthracycline.

Certainly, we cannot use Herceptin with anthracyclines, as we saw in the metastatic trial, and we’re not quite sure if we can use it sequentially yet. It may be that the short interval following the last dose of AC may not be good enough to offset the potential cardiac toxicity of Herceptin.

So far, our trial is moving along and the data monitoring committee has looked at the interim analysis of cardiac toxicity, and we have not reached our mark, which is to see less than five percent instance of severe cardiac toxicity with the Herceptin arm versus the AC-Taxol arm.

Unfortunately, in the other study that the Intergroup was conducting, the NCCTG 9831, where it was the same sequential approach, either together AC followed by Taxol with Herceptin, or Taxol followed by Herceptin. The concomitant arm was closed, or at least suspended, because of cardiac toxicity. And they have the same criteria that we have. So, for some reason, they saw a little bit more in their study so far. So, that’s still an open issue, whether you can give Herceptin following AC.

DR LOVE: How is accrual going in that trial?

DR MAMOUNAS: Accrual is going well. It’s actually going according to projections. We accrue about 40 to 45 patients a month, and we have over, I think, 800 patients now, out of the 2,700 needed.

DR LOVE: Now, one of the issues of that trial is quality control in terms of HER2 testing. Can you talk a little bit about what you’ve seen so far?

DR MAMOUNAS: Yes. We actually did do an analysis of the first 100 patients, and we found that there was actually some significant inconsistency between the results that we obtained for, particularly, non-reference laboratories and the results that we obtained centrally by doing FISH assay. So, we actually now mandate that the assay, if it’s immunohistochemistry, it should be done by either a reference laboratory or should be done centrally before we accept the patients on trial.

DR LOVE: Now, has there been any problem or inconsistency in terms of FISH results in the community?

DR MAMOUNAS: Not to my knowledge. I may be wrong on that, because I don’t remember all the data that we collected. But the majority of the diagnoses we had were with immunohistochemistry, and certainly there, there will be a lot of variability.

DR LOVE: So, the problem there was there were people being called 3+ – because they had to be 3+ to get in the trial – who weren’t 3+.

DR MAMOUNAS: They were FISH-negative.

DR LOVE: Was the IHC repeated, or they just got FISHed?

DR MAMOUNAS: I think it was repeated, as well, and that also was discordant.

Continue
Page 2 of 4

Home · Search