Select Excerpts from the Interview

Tracks 3-4

DR LOVE: Can you summarize the recently released efficacy data from the BCIRG 006 adjuvant trastuzumab trial?

DR SLAMON: The efficacy data are based on the first interim analysis of a three-arm trial with 300 events, and we recognize we’re walking a fine line, but even so, both arms crossed their efficacy boundaries. We’ve known all along that trastuzumab was the critical molecule. The relevant question is: How does the TCH arm — the nonanthracycline arm — look relative to the anthracycline-containing arm? We have a lot of safety information now, and that’s got to be weighed against the efficacy information.

What we can say now is that the risk reduction in the TCH arm is 0.39 (2.1). The risk reduction in the ACTH arm is 0.51, which is almost identical to what was seen in the trials reported at ASCO for that kind of combination (Perez 2005; Piccart-Gebhart 2005a; Romond 2005a). There are very few event differences between the two trastuzumab arms. The two confidence intervals completely overlap, and the statisticians have said there is no statistical difference between the two arms.

DR LOVE: Oncologists have to make practical decisions with the information available. Based on these numbers, many may be thinking TCH will not be as effective as the anthracycline arm. What are your thoughts on that conclusion?

DR SLAMON: My thoughts are consistent with what the data show: There is a numerical difference and it’s statistically insignificant. We need to wait until the data mature, and it’s not going to take a long period of time. Physicians should do what they feel most comfortable with at this point. If they feel more comfortable with the ACTH data, based on those relative differences despite the fact they are not different statistically, they should go with that arm, recognizing what everyone has said all along: Those patients are going to have to be watched very closely for cardiotoxicity.

DR LOVE: What is the statistical likelihood that the TCH arm will be superior to the ACTH arm?

DR SLAMON: I have no idea at this point. All I can tell you, without giving away information that will be presented at San Antonio, is that the results are based on very few event differences.

DR LOVE: Do you think adjuvant TCH is a reasonable alternative in the clinical setting at this time?

DR SLAMON: Based on what we know, yes. TCH has been around a while in the metastatic setting, and a lot of data have been presented, even randomized data.

DR LOVE: Do you have an adjuvant protocol available to you right now?

DR SLAMON: We do not.

DR LOVE: When you see a younger patient with a node-positive breast cancer, which adjuvant therapies do you consider and what’s your usual recommendation?

DR SLAMON: At this point we try, whenever possible, to avoid anthracycline-containing regimens because of the known interaction of trastuzumab with anthracyclines. However, we’re not restricted to TCH. There are a number of different drugs that interact very well with trastuzumab, including vinorelbine, for which we have published data in the metastatic setting (Burstein 2003).

If the trastuzumab story has told us anything, it’s that what we see in the metastatic setting gets even better in the adjuvant setting when we take it forward. I know oncologists in practice probably always go with what looks like the best number, and that makes sense, but I look at the composite picture.

We presented the data on cardiac safety from 006 (2.2), and the results are profoundly different from TCH, so it’ll all depend on the weight of the efficacy data when we have sufficient numbers and how that stands up against the safety data.

If you look at left ventricular dysfunction progressively over time, both anthracycline- containing arms do worse. The information has been around a while that nontrastuzumab/anthracycline regimens do make an impact, and we have learned from the adjuvant trials with trastuzumab that we make an even bigger impact than we previously thought.

We thought we were out of the woods with the anthracycline doses we’re using, and now we’ve found, as Chuck Geyer pointed out that the incidence is much higher than we thought, even for the standard arm, and this concerns me.

That has to be weighed against efficacy, and if the efficacy is strongly and statistically significantly different, then I think that has to be taken into consideration when you treat patients. We don’t have data sufficient to speak to that at this point, but we will soon and when that data is available, it needs to be made public.

Track 5

DR LOVE: How would you treat a woman in her fifties with nodepo-sitive, HER2-positive disease?

DR SLAMON: We use TCH and will continue to do so until we see that it is inferior, and the safety profile doesn’t make up for that inferiority.

DR LOVE: Can we anticipate that the relative risk reduction of trastuzumab chemotherapy regimens will extend to patients who have node-negative disease, particularly with smaller tumors?

DR SLAMON: I would think so. Based on the biology of the disease, we know that when the HER2 alteration is present, it’s a more aggressive disease. Now if you have a patient who has a smaller tumor, it has to be weighed against the likelihood of cure from the initial therapy — surgery and radiation therapy.

We also know there’s a subpopulation of patients who are HER2-positive, who aren’t cured, even if they have very small tumors. Treating with trastuzumab in the metastatic versus the adjuvant setting clearly isn’t as advantageous.

My feeling is that we’re dealing with a relatively benign agent in trastuzumab, when used correctly. It’s an enormously expensive drug, and that has to be put into context, but we have the two goalposts set. We know the drug is effective in the adjuvant setting and in metastatic disease. Between those two goalposts, I think it can be very effective.

Track 7

DR LOVE: What are your thoughts about the use of adjuvant trastuzumab monotherapy without chemotherapy in patients with comorbidities or of advanced age?

DR SLAMON: The orthodox answer is “no” — not outside the context of a clinical trial. My answer, and take it as my personal answer, is that the drug is effective, and if you’re dealing with a patient not on a clinical trial, it’s an individual sitting in front of you and the art of medicine still applies. If they’ve got significant comorbid disease, I would not withhold an effective therapy (Vogel 2002) because they don’t meet some protocol. I would administer single-agent trastuzumab.

Track 8

DR LOVE: Most physicians seem to be using adjuvant trastuzumab for one year. Is that what you’re doing?

DR SLAMON: Yes, but there are no clinical data to tell us how long to administer the drug. The one-year duration came from sort of an empiric extension of preclinical data, and my sense, based on those data, is that the optimal duration will be somewhere between six and 12 months.

Also, based on that same preclinical data, I don’t think there’ll be a big difference between one and two years. One good thing about the HERA trial is that it’s asking that question so we’ll get that information (2.3).

DR LOVE: It’s a little difficult to interpret the data from the HERA trial as opposed to the NCCTG trial with regard to sequential versus concurrent therapy. The HERA study showed approximately a 50 percent risk reduction (Piccart-Gebhart 2005b; [2.4]), as was seen in the combined analysis (Romond 2005b). However, in the NCCTG trial there’s a 13 percent nonstatistically significant reduction in the sequential arm (Perez 2005; [2.5]). How do you interpret those data?

DR SLAMON: I think it’s far too early to make a call, and I think that’s both the safe as well as the right answer. The HERA investigators have openly conceded the point that their data are more immature, vis-à-vis trastuzumab therapy, given the fact that they had an even longer period before the patients were randomized to trastuzumab, so their follow-up time is short.

Now they make up for that, in part, by large numbers, because they enrolled a lot of patients on the trial. But I think we need one more look at those data in terms of an update before we’re really able to say that they’re seeing the same kind of impact that we are seeing with the two cooperative group trials in the United States.

DR LOVE: The sequential versus concurrent arms in the NCCTG trial don’t have very many events at this point, so could it be that that’s the trial that’s misleading?

DR SLAMON: It’s true that the study doesn’t have many events, but it has longer follow-up.

Track 9

DR LOVE: What about the issue of delayed trastuzumab? Should the patient with a HER2-positive tumor who’s now six months, 12 months, or a couple of years after initial diagnosis be offered trastuzumab?

There’s the orthodox answer, and there’s what I think is the biologic answer. The orthodox answer is, there are no data that address whether giving it out further will be beneficial, and there’s no trial with any data to speak to that.

The biologic answer is very similar to what I said earlier. You know trastuzumab works in the adjuvant and in the metastatic setting. The patient you’re talking about sits between those two goalposts. If she has a disease that has escaped her primary site and it’s a HER2-positive tumor, the likelihood of her receiving benefit from trastuzumab is there. So my sense is, biologically, it should be beneficial.

DR LOVE: In the clinical use of delayed aromatase inhibitor therapy, there’s the mindset that the decision should be determined by the patient’s risk at that point, and assuming therapy would significantly decrease that risk. Do you believe that approach makes sense in terms of initiating delayed adjuvant trastuzumab?

DR SLAMON: I think that’s absolutely the way to go, but I don’t think we have sufficient parameters or parametrics to measure that at this time.

Track 11

DR LOVE: How should patients who received prior adjuvant trastuzumab be managed at relapse?

DR SLAMON: The thinking on treatment after progression following adjuvant trastuzumab falls into two camps: Whether it makes sense to try something entirely different or to continue trastuzumab and add something new. The data we’re seeing, based on preclinical information, would indicate that the latter makes sense. However, there are no clinical data to speak to that, and it’s going to be almost impossible to do a clinical trial to address it because of the half-life of the antibody.

We know that trastuzumab sticks around for a long period of time, so you’d be hard pressed to find a patient or an oncologist willing to wait until all the trastuzumab washes out before starting therapy in the face of progressive disease.

So by definition, if you switch to a different chemotherapy and you randomize to no continuation of trastuzumab versus continuation of trastuzumab, at week three when you add that chemotherapy, you’re still doing a combination study. That’s what makes it impossible to ask the question.

DR LOVE: Then what is your conclusion on this issue? How do you treat these patients?

DR SLAMON: The only conclusion I have is that if you’ve had a response to trastuzumab to begin with, I would continue trastuzumab therapy. The only place where I don’t do that — based on no data, just sort of my gut feeling and what we know about the biology — is when the patient progresses within a few months of having stopped their chemotherapy, I’m less likely to continue trastuzumab. However, if the patient is slowly progressing through the trastuzumab therapy, there still may be an effect of the antibody, and we can add something different to it.

Track 15

DR LOVE: Can you comment on the work that your group has done evaluating bevacizumab plus trastuzumab and whether you see that moving into the adjuvant setting?

DR SLAMON: I certainly hope it moves into the adjuvant setting. I think there’s more than enough very strong, compelling biologic and preclinical data indicating that this combination is rational.

There are now early clinical data, with very small numbers, both in the Phase I and II settings, which indicate that this is an active regimen, so I think it’s ready to be moved into a larger trial. My sense is that in short order it should be ready to be moved into the adjuvant setting. We know how trastuzumab performs in the HER2-positive population, and the efficacy of bevacizumab was demonstrated in the ECOG-E2100 study (Miller 2005 [2.6]).

We now know — based not only on preclinical but clinical data — that the HER2 population has a higher VEGF level. All of the dots connect. Now it’s a matter of showing that the combination is safe and showing some clear efficacy data that will allow us to think about launching a larger study.

DR LOVE: Can you talk more specifically about the Phase I and II data on the trastuzumab/bevacizumab combination?

DR SLAMON: The Phase I data were with nine patients, three in one of three groups (Pegram 2004). The study evaluated three different doses of bevacizumab, because we didn’t know the right dose when we launched it, along with standard-dose trastuzumab.

We saw responses in all three groups. Nine patients is a very small number, but the responses were pretty compelling in a first-line metastatic group. In addition, the third cohort was actually allowed beyond first-line treatment.

What we found was, in that total group of nine, there was one complete response, four partial responses, two stable diseases for greater than 11 months, and two patients who progressed. That was pretty exciting data for a small group, but you have to take that with a grain of salt because it’s only with nine patients.

We’ve now expanded it into a Phase II study, looking at the two biologics together in first-line metastatic disease. What we’re seeing up to this point is very similar to what we saw in the Phase I trial. We have 20 patients now in the Phase II, so a total experience with about 29 patients, and we’re seeing the same kinds of response rates. We’re very encouraged by the data. It needs to mature further, but I think it’s a rational regimen, and I think it’ll make its way to the clinic in terms of a big clinical study.

DR LOVE: Do you think this combination is ready to go into the adjuvant setting now?

DR SLAMON: I think it will go to the adjuvant setting. It’s ready, but I tend to be a little more aggressive about looking at these things, maybe, than others. However, I would wait until we finish this Phase II study, which is supposed to accrue 50 patients and look at the response rate and duration in those patients. We’ll know that relatively soon.

DR LOVE: Knowing the dangers of interpreting anecdotal cases or small series, other than the numbers in terms of response rates, is the magnitude of responses seen in the Phase I and II trials greater than you would have expected with trastuzumab alone?

DR SLAMON: Absolutely. I would not have expected it with trastuzumab alone or bevacizumab alone. If we were only basing it on thinking these might work well together versus the fact that the preclinical gene array data directed us, then I’d be a little more concerned, but given the fact that there are strong preclinical data, and now clinical data, I’m comfortable with it.

There’s an enormous amount of preclinical data indicating that the pathways of the two agents are linked, which is compelling and reproducible. You can show that HER2-positive tumors have VEGF levels that are much higher than the general breast cancer population by gene array analysis.

DR LOVE: When you say the VEGF levels are high, is that a direct stimulant in terms of tumor growth?

DR SLAMON: Yes. We believe it’s a direct stimulant. We can’t test that in the clinic, but we can test it preclinically. In the clinic, we know the HER2- positive tumors — age-matched, stage-matched controlled — are more metastatic and grow more rapidly. Now there’s direct growth stimulation of the HER2 pathway itself, and there’s all the ancillary things you need to support faster and more aggressive tumor growth, not the least of which is neoangiogenesis, so it all makes sense.

DR LOVE: Do you believe bevacizumab potentiates chemotherapy by improving delivery and that’s why it would work well with trastuzumab, or do you believe that, at least in HER2-positive breast cancer, there’s something else going on with that combination?

DR SLAMON: I’m more in the camp of the latter, thinking there’s something else going on. There are interesting data, including some data from clinical material, indicating that there may be this gradient phenomenon in terms of better penetration of chemotherapy. However, when you look at the doses achieved within the tissue, even when this is present, it’s well above the IC50s.

The question is: Are higher doses achieved due to increased oncotic pressure, which is thought to be one of the mechanisms of VEGF? It’s clear that the VEGF antibody will tie up circulating VEGF and that VEGF is one component needed for neoangiogenesis for a number of different kinds of tumors, so I think there is a lot of room for other mechanisms of how that would work.

DR LOVE: Do you believe that changing the oncotic pressure or profusion theoretically should be beneficial in terms of trastuzumab delivery?

DR SLAMON: Theoretically, but I would want to see supporting data that are reproducible.

Select publications