As part of our coverage of the American Society of Clinical Oncology’s annual Breast Cancer Symposium, held September 4–6 in San Francisco, we are speaking with William M. Sikov, MD, of the Alpert Medical School at Brown University in Providence, Rhode Island, about neoadjuvant breast cancer therapies. Dr. Sikov gave a talk at the meeting on the clinical meaningfulness of pathologic complete response as a surrogate endpoint in breast cancer clinical trials.
—Interviewed by Anna Azvolinsky
Cancer Network: First, what does the pathologic complete response endpoint measure? Why is it being used specifically for neoadjuvant breast cancer trials and why is it used by the US Food and Drug Administration (FDA) as a way to accelerate drug approvals for breast cancer?
Dr. Sikov: Pathologic complete response is a way of measuring response to neoadjuvant therapy, most often neoadjuvant chemotherapy. So, it’s looked at in patients, in who the doctor has decided to administer treatment—again, usually chemotherapy—before surgery to shrink the tumor in the breast and any involved lymph nodes, to assess any response to that treatment. A pathologic complete response is usually defined as the absence of residual invasive disease in the breast and in the axillary lymph nodes at the completion of the neoadjuvant treatment. We’ve known for some time that patients who achieve a pathologic complete response with neoadjuvant therapy tend to have a very good prognosis—that is, eliminating the visible disease in the breast and lymph nodes correlates with eradication of micrometastatic disease that we can’t see. Thus, patients who achieve a pathologic complete response are at much lower risk for subsequent distant disease recurrence.
This finding from individual studies was confirmed in a large meta-analysis, which was, in essence, commissioned by the FDA, and it confirmed that the patients who achieved a pathologic complete response had a much better prognosis than the patients who had residual invasive disease in the breast or lymph nodes at the end of treatment. It also showed that in patients with more aggressive cancers, such as triple-negative breast cancer or HER2-positive breast cancer—especially HER2-positive, hormone receptor–negative breast cancer—the benefit of achieving a pathologic complete response, in terms of recurrence and survival, is much greater. Also, it gets more and more difficult to do adjuvant chemotherapy trials. The reason for that is that as our treatments improve, the percentage of patients who are going to recur with standard treatment goes down, and thus, to show a benefit to a new, added treatment gets more and more difficult, requiring more and more patients and more and more years of follow-up.
So, what the FDA is saying is, well, maybe we can short circuit that process to some extent, by granting accelerated approval to drugs that show clear increases in pathologic complete response rates in the neoadjuvant setting, although they still expect those drugs to eventually show disease-free survival and overall survival advantages in the postoperative setting.
Cancer Network: You mentioned that pathologic complete response correlates with a better prognosis. What else do we know about how well pathologic complete response translates into long-term clinical benefit? In other words, what is its clinical meaningfulness and is there a subset of patients for whom this is more important?
Dr. Sikov: In the meta-analysis that was done and published earlier this year by Cortazar and her colleagues, patients who achieved a pathologic complete response had half the risk of distant disease recurrence compared with patients who did not achieve a pathologic complete response, and a 36% improvement in overall survival. But again, if you hone down on the more aggressive breast cancer subtypes, such as hormone receptor–negative, HER2-positive cancers and triple-negative cancers, then the benefit was as high as 75%. That is, patients who achieved pathologic complete response had only a quarter of the risk of recurrence compared with patients who did not achieve a pathologic complete response.
So, then the question is, if you do a study where you show a higher pathologic complete response rate, how is that going to translate into long-term benefit—reduction in the risk of recurrence and improvement in survival? The problem is that the Cortazar et al analysis was not really able to show that benefit looking at results of individual studies. And the likely reasons for that are that the studies they included in their meta-analysis were somewhat more heterogeneous in terms of patient population, they had relatively low pathologic complete response rates overall, and the differences between the control and experimental arms were relatively small. Fortunately, we are now seeing more studies where we are getting a larger benefit in terms of increase in pathologic complete response rates, and we are hopeful that that will translate into improvements in disease-free and overall survival.
Cancer Network: Are there examples of drugs that have been approved based on this pathologic complete response surrogate endpoint? Do we need to wait for long-term outcomes before considering using a drug in the neoadjuvant setting? Are there other endpoints that should be considered in conjunction with pathologic complete response?
Dr. Sikov: There is only one drug that has received an accelerated approval in the neoadjuvant setting based on neoadjuvant data, and that is pertuzumab. When given in combination with chemotherapy and trastuzumab, pertuzumab was shown to result in higher pathologic complete response rates in two relatively small phase II studies. Also, it was well tolerated, with relatively little additional toxicity. So, that drug has received accelerated approval, even though we do not have any long-term survival or recurrence data on it. I think one of the reasons why it may have been approved is that when it was given approval, a very large adjuvant trial was nearing its completion. The hope is that the large adjuvant trial will confirm the benefit seen in the smaller neoadjuvant studies. I think there is controversy within the medical oncology community as to whether pertuzumab should be considered as part of a standard neoadjuvant treatment for HER2-positive breast cancer, based just on pathologic complete response with no long-term benefit demonstration. I would say that, personally, I do typically include pertuzumab when I give neoadjuvant chemotherapy and trastuzumab to HER2-positive patients. But I know that not all medical oncologists and academic medical oncologists feel the same way.
In regard to the question of whether we need to wait for long-term recurrence and survival benefits before we consider using these drugs in the neoadjuvant setting with or without FDA approval, depending on whether the drug is available and if it’s approved for other indications, in my opinion, the answer to that is no, as long as we are cognizant of the benefit of the treatment and its potential risk, because any drug when added to a standard regimen is likely to increase toxicity. But we know that pathologic complete response correlates with other endpoints that may be important to the patients. For example, if you have a patient who has an unresectable cancer, one of the first reasons we give neoadjuvant therapy is to make unresectable patients resectable. If you improve the response rate to neoadjuvant treatment for a patient with unresectable locally advanced disease, you will improve the chances that that person is converted from unresectable to resectable.
Similarly, if you have a patient who is not a candidate for breast-conserving surgery but wants it, you want to do the best you can to shrink that tumor to make that person a potential candidate for breast conservation. The addition of a novel agent that has been shown to increase the pathologic complete response rate is likely to also improve the clinical response to treatment, and thus, their ability to undergo breast-conserving surgery. We showed, for example, in the Cancer and Leukemia Group B (CALGB) 40603 study, of which I was study chair, that for the patients who receive the standard neoadjuvant regimen, who were not considered breast conservation candidates prior to chemotherapy, with the standard treatment their likelihood of being converted to a breast conservation candidate was only around 40%, whereas if they also received carboplatin, that number increased to 57%. That again correlated with improvements in pathologic complete response rates with the addition of carboplatin. I think there are some other endpoints that are worth taking into account in individual patient cases in which it might be worth the risk of increased toxicity to increase the likelihood of a good response that will help us reach an endpoint other than disease-free or overall survival.
Cancer Network: Outside of the clinical trial context, what does pathologic complete response mean for breast cancer patients in the clinic? Are there certain patients for whom a pathologic complete response is more likely to translate into a stable, long-term response or even overall survival?
Dr. Sikov: Yes. Again, it is really patients with the more aggressive cancers. The meta-analysis could not show a significant event-free or overall survival benefit for patients with hormone receptor–positive, HER2-negative, low- or intermediate-grade cancers. So, achievement of a pathologic complete response in those patients did not lead to substantial improvements in long-term outcomes. The findings for patients who were both hormone receptor–positive and HER2-positive were also less impressive, but if you look at higher-grade patients—triple-negative, HER2-positive, hormone receptor–negative, and even the high-grade hormone receptor–positive, HER2-negative patients—if they achieve a pathologic complete response, their long-term outcomes are better than if they did not. And I think that is worthwhile. Of course you have the other side of that coin, when a patient does not achieve a pathologic complete response and now they have identified themselves as higher risk. We have some clinical trials where we will be treating those patients with additional therapy in the hopes of improving their outcomes, but for example, if you did surgery first, you would never know that they were not good responders to standard chemotherapy.
Cancer Network: Thank you so much for joining us today, Dr. Sikov.
Dr. Sikov: Thank you very much, my pleasure.