This is not a great leap forward; this is a serious disappointment, not just for the investigators, but for the entire field. It’s difficult to mount any enthusiasm for a dual HER-2 blockade in the adjuvant setting, at least with the combination shown here. What does this failure tell us? At a simple level, we’ll not be using lapatinib in the adjuvant setting. Not only doesn’t it work, but also the combination is significantly more toxic than trastuzumab alone.

ALTTO requires us to rethink our approach to the development of new drugs for early-stage breast cancer. Two years ago, the FDA released a draft of guidance, outlining the path to accelerated approval for neoadjuvant breast cancer therapies. This guidance suggested that pathologic complete response represented a reasonable surrogate for adjuvant benefit. The logic behind this guidance I thought was quite rational. We know from a large number of studies that pathologic complete response is associated with improved outcomes, both with regard to DFS and OS. Therefore it seemed logical to suggest that significant increases in pathologic complete rates in the context of randomized controlled trials would predict the outcome of adjuvant trials. If this was indeed the case, we might get valuable new agents to patients years before the completion of our large, ponderous cast-of-thousands adjuvant trials.

We need to understand our failures if we are to create new successes. For decades we’ve assumed that activity in the acute metastatic setting will translate to improved cure rates for micro-metastatic disease. This assumption is now under siege under multiple fronts in several disease types. If we cannot trust neoadjuvant and metastatic results to predict the outcomes in the adjuvant setting, how will we move forward?

But move forward we shall, at least in HER-2–positive breast cancer. I hope to someday soon see a happier plenary session on HER-2–positive disease. We have several other ongoing experiments in the adjuvant HER-2 setting. Pertuzumab and T-DM1 are the furthest along, but multiple others are close behind. We’re out of neither ideas nor new agents in the HER-2 setting. I hope, and I believe that at least one of these will prove fruitful, and will move us closer to the day where HER-2-positive breast cancer will no longer be a public health problem. But if ALTTO has taught us anything, it’s that hopes and beliefs are no substitute for the hard work of well-conducted definitive clinical trials.

Many of us felt the combination of lapatinib and trastuzumab would be a hit. Unfortunately, as we heard [at ASCO], that’s not true. The combination is no better with regard to PFS or OS than trastuzumab by itself. Also, there is more toxicity with the combination — principally gastrointestinal and dermatologic toxicities.

This is why we do clinical trials. Breast cancer is not a logical disease, no matter how much we try to make it be so.

The good news is that the patients in this trial actually are doing quite well overall, and that proves once again the very powerful effect that trastuzumab has had on HER-2–positive breast cancer, which used to be a really poor-prognosis cancer.

Although this is not the step forward that we hoped it would be, this means we won’t expose future patients to another drug that has toxicities but doesn’t do them any good. So even though it is a negative study, it is a very important negative study with regard to what we do in the future.

The results of this study really operate on two levels. On the one hand, from a practical point of view, this is the kind of study that if positive would have changed practice tomorrow morning because these drugs are available and the adjuvant setting is one where there is a tremendous drive to reduce recurrence.

But the second level in which this operates is a more technical one, with tremendous potential implications right now. Last year the FDA for the first time ever approved a drug in the preoperative setting on the basis of an increased pathological complete response rate. That drug, pertuzumab (Perjeta, Genentech) has been studied in the adjuvant setting, and the hope is that by getting the drug into the market based on pathological complete response, we would ultimately save lives. That may turn out to be true; we’ll have to wait for the results of the AFFINITY trial. 

But the question we have from the societal point of view is, can we routinely use the preoperative setting and the improvement in pathological complete response as a reliable surrogate for DFS and OS? Then, the answer at least from ALTTO right now is maybe not. This is going to cause a tremendous amount of high-level technical discussion in terms of drug development.