DR MAMOUNAS: Pathologic complete response, which is essentially eradication of all invasive disease from the breast with negative nodes — there are a couple of definitions. The FDA preferred definition is no invasive disease in the breast, negative nodes — has traditionally been correlated with better outcome. And this has been shown in all the randomized clinical trials and also in the recent overview that the FDA undertook, which included some of the NSABP, German trials, some European trials.
Where it gets a little tricky is that in that overview analysis, we could not demonstrate that in individual trials, increasing pathologic complete response relates — translated to increasing overall survival. And that’s a difficult endpoint to show, because some of the trials have smaller numbers, maybe the power wasn’t there, and they were done in different times with different regimens.

But when I fall back, when I look at this issue of pathologic complete response and outcome correlation, every treatment that we have seen so far that improved overall survival also has been shown to improve pathologic complete response, going back from the days of CMF, with pathologic response rate, 7% to 8%, to the anthracyclines at 15%, to taxanes at about 30%, trastuzumab-based therapy, 40%, 50%, 60%.

So whether that’s a strict correlation or not, the fact is that more effective chemotherapy improves survival but also improves pathologic complete response. And I haven’t yet found a trial that that’s not quite the case. People go back to the NSABP B-27 and say, “We didn’t show a correlation between doubling the pathologic response and improved survival.” Probably, the trial was underpowered. But when you look at, let’s say, the overview analysis, the addition of taxanes to anthracyclines improved survival and indeed improve pathologic complete response.

The same thing, for example, with the ALTTO and the NeoALTTO trial. That’s a little different, because the NeoALTTO wasn’t exactly the ALTTO trial in the neoadjuvant setting, because some of the therapy was given afterwards. So I think the correlation is maintained, or hopefully we’ll prove that with additional trials in the future.

DR LOVE: So Kim, I’m curious. If you think about kind of the 3 major subsets we’re talking about here today, what you might say to a patient even or what your thoughts are in terms of what the long-term prognosis is of a patient who has a path CR as opposed to doesn’t have a path CR.

DR BLACKWELL: Those numbers are a little hard to get at, because what we’ve typically looked at is all subgroups in a trial plus or minus an agent. But at least in the HER2 space, I think we have fairly good data that a complete pathologic response leads to close to 90 or greater long-term disease-free survival if the disease is eradicated after receiving neoadjuvant therapy.

Triple-negative, I think the correlation between complete pathologic response and long-term outcome is not as clear, at least in my mind, in part because the trials that have been looked at are AC versus nothing, ACT versus AC. So I think there’s a confounder there and it makes it difficult to offer patients a precise long-term outcome based on a complete pathologic response.

ER-positive, I don’t think we have very good or precise data as to what a complete pathologic response means in terms of long-term survival, because if you think about it, a complete pathologic response in ER-positive breast cancer typically means that that tumor is not particularly an indolent, luminal A, ER-positive, and so those patients might actually have a survival benefit with long-term endocrine therapy. We just don’t know those numbers.

DR MAMOUNAS: I agree. I think for particularly the indolent ER-positive, HER2-negative patient, low grade, Grade I and II, there is really not much correlation between pathologic response and outcome. What I mean by that, though, is that they do well, whether they have a pathologic response or they don’t. The Grade IIIs, the correlation is better, and that was in the FDA overview analysis that indeed, there is a correlation between PCR and outcome. And, obviously, in the triple-negative and HER2-positives, the correlation is much more robust. So these are the patients — that’s another point why, for example, we can show this incremental benefit, because in all these neoadjuvant trials, we included the ER-positives, HER2-negatives and Grade I and II. So those are essentially there to dilute the effect. But if you take those out, the subsets that benefit from chemotherapy show this correlation much more robustly.

DR BLACKWELL: Yes. I think it would be great to have better numbers to be able to provide a patient. And I guess — I think — we know there’s a correlation. But the question is, when the patient comes back to me and has a complete pathologic response after having, let’s say, ACT for triple-negative breast cancer, I’m always a little uncomfortable quoting them a long-term average disease-free survival after they’ve had a complete pathologic response, because I don’t think we have very precise numbers in that population. I can tell them they’re going to do better than if they hadn’t responded. But in terms of, “This is the percent chance that your cancer will come back,” those numbers are very hard to get at.

Local disease control and neoadjuvant treatment

Neoadjuvant treatment of HER2-positive tumors

Neoadjuvant treatment of triple-negative tumors

Neoadjuvant treatment of ER-positive, HER2-negative tumors

Other issues in neoadjuvant treatment