NCCTG-N9831: Adjuvant Chemotherapy Alone or with Sequential or Concurrent Addition of 52 Weeks of Trastuzumab in HER2-Positive Breast Cancer

DR PEGRAM: I found this presentation intriguing. When I was working in Dennis Slamon’s laboratory at UCLA, we published a series of papers showing that the combination of cytotoxic agents with trastuzumab was superior in cell-based assays and in xenografts than when given in sequence.

Moreover, if you gave the agents in sequence deliberately in an experimental model, both in vitro and in vivo, it was always better to give the trastuzumab first followed by the cytotoxic agent and not the other way around. When we did the original drug development strategy for trastuzumab, we deliberately integrated it in combination with cytotoxics in an effort to exploit any potential synergy between the two. The Perez trial addressed this question clinically in a large adjuvant study that compared the sequence of chemotherapy followed by trastuzumab to AC followed by paclitaxel given in combination with trastuzumab followed by a total of a year of adjuvant trastuzumab.

The study design was nice because it allowed for the first time in the adjuvant setting a head-on comparison of the schedules of trastuzumab — whether it is best to use in sequence or in combination. The European adjuvant trastuzumab trial was all done as a sequence, in which generally chemotherapy was followed by trastuzumab. The common practice outside of North America is to use sequence rather than combination.

Just as they had shown a few years ago in an unplanned interim analysis, these most recent updated data from Dr Perez also showed superiority of the combination over the sequence. The data looked fairly convincing and conclusive until you examined the statistical plan of the protocol. The O’Brian-Fleming p-value that must be achieved in order to call this statistically significant in the preplanned analysis was 0.001. The actual p-value was 0.019. If you’re a purist, you could argue that this did not cross the boundary for statistical significance, even though it looks compelling that the combination is better than the sequence.

This is a little bit of an unsatisfying result. It fits my preconceived notion and my clear bias that the combination is probably better, but there’s this nagging issue that it doesn’t quite hit the mark in terms of the preplanned p-value. It is a nonsignificant trend, but it seems to be a trend of a significant enough magnitude to possibly merit clinical consideration.

Clinicians can digest the data as they see fit. The overarching question is whether or not to use the trastuzumab, and that has been definitively answered. Which chemotherapy or whether you use the sequence or the combination are relatively trivial considerations compared to the magnitude of benefit that we see by simply administering some trastuzumab in any way, shape or form.

Dr Pegram is Full Professor of Medicine and Director for the Translational Research Program at the UM Sylvester Comprehensive Cancer Center’s Braman Family Breast Cancer Research Institute in Miami, Florida.