2011 SABCS, San Antonio Breast Cancer Symposium, 6-10 December, San Antonio, USA

The latest for advanced breast cancer

Dr Hope Rugo – University of San Francisco, California, USA

I think that this year has been a very exciting year for the treatment of advanced breast cancer. A lot of the most important information was presented at this meeting in San Antonio this year which is somewhat unusual. Looking back over what was presented at ASCO, what’s been published this year and then what we’ve just heard about at the San Antonio breast cancer meetings, I think we’ve really ended the year on a very positive note about where we’re going in the future. I think that we’ve seen important advances in a number of areas and then I think that from the rather spectacular disappointments in a couple of areas we end the year on a much more hopeful note about where we can go to better use agents that in an unselected population didn’t show as much benefit as we would have liked. I think that if we break down the trials and the directions that we are going in, probably the most important thing that we have learned is that we need to better understand sensitivity and resistance, prognostic and predictive factors, and we have been talking about that for a long time but I think that now we really have a better directive to focus on disease elements that are very specific. So instead of treating breast cancer as being HER2 positive and HER2 negative and ER positive and ER negative, we need to look within those clinical subsets at the important biologic changes that occur in cancer that help us to target it more effectively.

So I think in that in the hormone therapy area, so ER positive disease, we have learned a lot about the differences between more proliferative and less proliferative ER positive disease, HER2 positivity and negativity in ER positive disease. Obviously at this meeting the results of BOLERO-2 are the most striking and then I think also quite interesting are the data form SWOG-S0226. So BOLERO-2 we’ve learned that if we target patients who have progressed on a non-steroidal aromatase inhibitor, and probably that feature is very important to the results that we saw in BOLERO-2, that a cancer that progresses on the non-steroidal aromatase inhibitors actually up-regulates pathways that are important in our targeting when we add in a new agent. So you couldn’t just, for example, treat everybody in the same way and that was our problem for a number of decades. So adding everolimus targets this downstream pathway that is activated in more resistant ER positive disease and led to the really remarkable results shown today actually by Gabriel Hortobagyi, and by Baselga at ESMO and now published in The New England Journal of Medicine. And that trial, which I am on the steering committee for and participated in, I think it surprised us with its results and again I think gives us a lot of enthusiasm about where to go in the future, first in new treatment for patients with advanced breast cancer that I think is exciting and really makes a meaningful clinical difference, and second, a new energy to move forward in trying to reverse resistance in hormone receptor positive disease.

There was a phase II trial with a histone deacetylase inhibitor and hormone therapy in progressing ER positive breast cancer presented at ASCO Breast and that showed some encouraging results, so that is another area that will be pursued and I think there’s a lot of interest in combining both mTOR inhibitors and inhibitors of another pathway that sometimes allows an escape from mTOR, the insulin-like growth factor receptor. In pre-clinical settings when you combine those two inhibitors you can see a synergy and a better anti-tumour effect, and then in a phase II trial, also done actually by José Baselga, there were responses actually in patients who had luminal B or more proliferative type ER positive breast cancer who progressed on hormone therapy and were just treated with that combination of biologic agents. So that’s led to an on-going randomised phase II effort with Merck that I’m participating in, where we are looking at their two inhibitors of mTOR and IGFR and hopefully we’ll be able to in the future compare that potentially to everolimus and exemestane, and add exemestane to that combination to see if we can further improve outcome. So we want to control toxicity as well at the same time obviously, and learning more about these drugs we’ll be better equipped to manage toxicity.

In the area of HER2 positive disease we of course have a wealth of riches. There has been a lot of discussion at this meeting about different pathways to target and how we might be able to again combine targeted agents to better overcome resistance. It’s kind of funny actually that potentially the best data that we will see in HER2 positive breast cancer for some time is the data from the CLEOPATRA study that is looking at adding an antibody that targets the exact same receptor; it is not combining different targeted agents. These obviously work by different mechanisms otherwise they wouldn’t be so additive but you are using antibodies, so the same kind of class of agent, that both target the same receptor family. So it is curious. We have seen that antibodies result in, so far, the least toxicity of targeted agents compared to small molecule inhibitors and that for some reason combined targeting of the HER2 receptor with bortezomib and trastuzumab, as we have seen in other settings in the neo-adjuvant trials presented last year looking at the oral tyrosine kinase inhibitor lapatinib, trastuzumab is a very, very effective way of treating HER2 positive disease.

So the CLEOPATRA data is practice changing and, I think, very, very exciting. Where we are going from there is curious because we have another agent, TDM1 trastuzumab, linked to the toxin derivative of maytansine, a very potent microtubule inhibitor that showed efficacy in a randomised phase II trial presented by Sarah Hurvitz at ESMO ECCO, which looked at docetaxel trastuzumab as the control arm, and that was superior and quality of life was better shown in a poster here at San Antonio; very, I think, intriguing data. So we are waiting for the randomised phase III trials and then the idea, really the challenge test, is going to be how are we going to treat patients? We have so many effective ways of treating patients with HER2 positive disease.  And I think that what we are really hoping is that these agents will move into the adjuvant setting and we will be able to triage treatment based on better understanding of the molecular forces that drive resistance and sensitivity, so we’ll be able to find the patients with HER2 positive disease that need very little treatment and those that need a more aggressive combined therapy option, and potentially reduce the number of patients with advanced disease who are going to die of HER2 positive breast cancer to a very small number. So that is an exciting part of this meeting.

There was an interesting study presented by my Chinese colleague who looked at markers of resistance and response to lapatinib combined with paclitaxel versus paclitaxel alone and suggested, as many have before, that abnormalities in the PI3 kinase and P10 AKT pathway are really what’s critical to resistance. But he found some curious data and I think that needs to be confirmed and looked at in a larger setting and in a peer reviewed setting, but it was in collaboration with GSK and the work was done in a central laboratory and I think quite intriguing that, in their hands at least, that P10 might be predictive of response to lapatinib, P10 deficiency lapatinib and paclitaxel, and that PI3 kinase inhibitors might just predict a worse outcome and not respond to a specific agent, so we would need to try maybe a different combination therapy.

There is a lot of interest in trying to combine multiple targets, as I was mentioning in ER positive disease, and we are going to see studies in HER2 positive disease where you combine a HER2 targeted agent with an mTOR inhibitor with PI3 kinase inhibitors; where you combine TDM1 with pertuzumab or trastuzumab to look at TDM1 versus trastuzumab in chemotherapy. There are going to be a lot of different studies which I think will help us, but we do have a wealth of riches. In contrast, patients who have a HER2 negative disease, whether it’s triple negative or just hormone resistant, we have a much more difficult situation where what we’ve seen this year are two, as I mentioned earlier, spectacular disappointments. And I think that what I have described that as is as the phoenix rising out of the ashes, that in order to move forward sometimes you need to falter and then you need to pick yourself up and move forward and you move forward on a stronger foot than the one that you were entering in with. And I think that’s true both for the use of bevacizumab and the study of PARP inhibitors in BRCA unselected triple negative breast cancer.

So this year we’ve seen in the United States the FDA withdrawing the approval of bevacizumab for the treatment of metastatic HER2 normal breast cancer. We will see that bevacizumab can add to trastuzumab in HER2 positive metastatic breast cancer when combined with docetaxel in the AVRIL trial but, as I mentioned, there are a lot of treatments, that are specific for HER2 positive disease. Pertuzumab adds almost no toxicity, so I don’t see bevacizumab moving forward competitively in that market, it will remain to be seen. But bevacizumab was a big hope in HER2 normal disease and what happened was all three randomised trials in the first line setting and in one randomised trial in the second line setting showed they met their primary end points; they all showed progression free survival benefit but no difference in overall survival. So when you see a finding like that, and it probably isn’t just the treatment they received after progression, it makes you realise you’re just barking up the wrong tree, so to speak; that what’s happening is you are treating everybody when what you really need to be doing is selecting the patients who are most likely to benefit. And curiously, results of the AVRIL trial, I think, in HER2 positive disease are going to get us closer to that end point in HER2 normal disease by trying to identify the features that correlated with the best response. And I think that where we are going to go with that is by looking at a marker in the blood, VEGF-A, and it turns out that by looking at it in a different tube, a different preparation – not citrate but EDTA, seems to be a much more reliable indicator and correlate in other diseases with response versus resistance to adding bevacizumab in a quite striking way. And then SNIPS, or single nucleotide polymorphisms, for not VEGF which has been studied already in ECOG 2100 but the VEGF receptor, may also be predictive. So there is a planned trial called the Meridian Trial by Roche and Genentech, that will look and basically recapitulate ECOG 2100 so paclitaxel versus paclitaxel and bevacizumab, and that will look specifically at these end points to try and see whether or not in the, say, 60% of patients who have an increase in VEGF-A or who have single nucleotide polymorphisms, with the primary endpoint being the VEGF, whether or not those are the patients who really benefit. Because then we could say, OK just like we select for HER2, these are the patients who are most likely to benefit.

So I feel like it was a big disappointment for all of us and for our patients. There are clearly patients who benefit from the addition of bevacizumab. There is no question about that and some benefit quite strikingly but the problem we’ve had is not being able to identify those patients and diluting the effect too much.

In terms of triple negative breast cancer, we could not have been more disappointed. I would say that was even more of a disappointment than bevacizumab where we were already sort of creeping along, and with iniparib we saw dramatic results of a phase II trial published in The New England Journal in January and two weeks later a press release saying that the phase III trial, which we’d all worked very hard on, and enrolled lots of patients on, had failed to meet its primary endpoint. So what does that mean? Well we have learned since that time a dramatic amount about iniparib itself and the doses used in that trial did not really inhibit PARP. It does have significant anti-tumour effects. So that led us to a current trial that is being run by Sanofi looking at a dose escalation of iniparib. It doesn’t cause any toxicity so why not escalate the dose and change the way you administer the drug to make it more convenient for patients so that we can then move forward with the right dose.

The second thing with iniparib, I think, is that we treated patients who had locally determined triple negative breast cancer. What we have realised since then, and some people already knew in the laboratory, is that a triple negative is an incredibly heterogeneous disease, so it is made up of many different subsets. And in fact some of the patients who were treated as triple negative in the phase III trial didn’t have triple negative breast cancer. But I think it’s more than that; I think it’s the heterogeneity of subsets. Now a lot of the oral PARP inhibitors, almost all of them, are really focusing on patients who have metastatic BRCA1 or 2 associated disease because they know that they have single agent efficacy in that setting. But I actually think that there is phenotypic evidence that triple negative breast cancer can have abnormalities in the BRCA pathway and we have seen that. In fact we saw that in Joyce O’Shaughnessy’s presentation by using the T-gen technology to look at twelve patients with metastatic triple negative breast cancer where the BRCA pathway, BRCA1 pathway, was abnormal in some of those patients and in one in particular who had a dramatic response to iniparib combined with gemcitabine and carboplatin on the open label portion of that phase III trial. So I think that the iniparib disappointment is going to lead us to a much stronger position in treating triple negative breast cancer with agents like iniparib and with potentially oral PARP inhibitors as well. And hopefully we’ll achieve that in the next few years, but at least we will have the background data first to make an educated approach to the future.

The other, I think, very exciting area in this meeting isn’t directly applicable to the clinic but I think it is important. We’ve heard talks from today actually, Thursday, from Lisa Coussens, about macrophages and the host response to cancer and how blocking macrophages may significantly improve sensitivity to chemotherapy. And we are actually launching a phase I/II study with a macrophage inhibitor from Plexxikon that will, I hope, help us better understand that in combination with the chemotherapy agent eribulin.

We heard also from Joe Gray, who won another award here at San Antonio for his work, looking at understanding the pathways that drives specific cancers. So in other words I think we are ready to move past intrinsic sub-types, which was very exciting but it’s not giving us enough information about how to treat patients, certainly not in the metastatic setting. We need to know which targets we should be treating for each specific cancer, and that’s going to get us to a more individualised approach. So what he discussed was very closely related to the sort of idea behind Joyce O’Shaughnessy’s work, which was quite preliminary with T-gen, is that you might be able to map a cancer for a specific patient and then be able to devise a group out of those patients where you can use a specific combination of targeted agents most effectively. And obviously in the long term our idea would be to move that into the early stage setting where we could cure more women with breast cancer; and I think at this meeting we really have seen the first glimmers of hope that this is really feasible moving forward.

Do you have an opinion on the FDA’s stance on bevacizumab?

My opinion is that these studies were positive; they did meet their specified end points. My confusion about the FDA’s approach is that we have many drugs that are approved based on PFS endpoints. That was the primary end point of their clinical trial. They never demonstrated survival endpoints in their drugs with some significant toxicities, more than bevacizumab, but they were approved and that has to do with the mechanism of approval, in other words they got approval right up front for their less than fabulous agents whereas bevacizumab got accelerated approval because they had a co-operative group study that was positive only. And then the other studies, although positive, were not quite as positive. And so the issue that I have with it is that why should we have all the other agents approved and not bevacizumab that clearly benefits a minority? I think that if the objective is to try and find the group that truly benefits that that is very noble, but maybe we should have started that a while ago. It would have been great in my mind if that was the requirement at the time of accelerated approval, but where we are now, I think, is in a less clear place.  And so I guess my thought is that really there have to be very clear guidelines for approval and it can’t be different for a bevacizumab-like drug versus a different target or versus a different chemo agent.