Ian E. Krop, MD, PhD: At the San Antonio Breast Cancer Symposium in 2019 we saw the results of the ATEMPT trial. This was a trial that built on a previous study, called APT, which was designed to try to reduce the amount of treatment given to patients with early-stage HER2-positive breast cancer. We know that because HER2 is a more aggressive subtype of breast cancer, even small cancers that are HER2 positive have a substantial risk of recurrence. Therefore, some adjuvant therapy is needed for these patients. But the large adjuvant trials of trastuzumab-based therapy really didn’t include for the most part small HER2-positive, node-negative cancer. There really was not a standard of care for how to manage such patients.
The APT trial looked at the combination of trastuzumab and paclitaxel, with the paclitaxel given for 12 weeks and the trastuzumab given for a year. That showed a very low recurrence rate after that regimen for patients with largely stage I, HER2-positive cancers. APT did include patients who had cancers up to 3 cm that were node negative, but the majority of the patients on the trial were stage I.
Based on those results, where the recurrence rate was very small, it was clear that you could tell a patient with a relatively low-risk HER2-positive cancer that giving the TH [paclitaxel, trastuzumab] regimen would be associated with a very low risk of recurrence. Given that it’s substantially better tolerated than multiagent chemotherapy with trastuzumab, it is a very reasonable standard of care, and that’s become widely used in the United States and globally.
The ATEMPT trial was an effort to further build on the results of the APT, with the idea that perhaps you could completely get rid of conventional chemotherapy in this setting by substituting T-DM1 [trastuzumab emtansine], which essentially is targeted chemotherapy, along with HER2-directed therapy. The design of the ATEMPT trial was slightly complicated. It randomized patients with stage I HER2-positive breast cancer to either the standard regimen of paclitaxel and trastuzumab that was used in APT or, essentially, 1 year of T-DM1 as a single agent.
To be clear, even though this was a randomized trial, 1 of the primary objectives was to determine whether those patients who received T-DM1 had a very favorable outcome—essentially, the same type of analysis that was done in APT. The randomization in ATEMPT was 3 to 1. About 375 patients were treated with T-DM1. As I said, 1 of the objectives was to just show that in that group patients did well. In fact, in the results that were presented by Sara Tolaney of the Dana-Farber Cancer Institute, that was the case.
The patients who received T-DM1 had a very favorable outcome with a disease-free survival at 3 years of about 98%, with only 2 distant recurrence cases within the entire population of almost 400 patients. Clearly, a year of T-DM1, which does not involve standard chemotherapy, is associated with a very favorable outcome. But the other objective of the trial was to compare the toxicity between a year of T-DM1 and a year of the TH regimen.
Somewhat to our surprise, the clinically relevant toxicities were very similar between the 2 arms. We had anticipated that T-DM1 would have fewer clinically relevant toxicities. But when using that particular composite end point of clinically relevant toxicities, it included patients who stopped T-DM1 early. About 18% of patients in the trial did stop T-DM1 early, and that was for a number of reasons. The majority of those patients switched to trastuzumab and, overall, did very well, as was seen in the primary end point of the study. But those did count as toxicity events. When you include those patients, the toxicity, at least by that 1 end point, was very similar.
We can’t show that the T-DM1 regimen was less toxic by that end point compared with paclitaxel-trastuzumab. That being said, there are definitely toxicities between TH and T-DM1. With TH, patients lost their hair and they had more neuropathy and more neutropenia. With T-DM1, in addition to early discontinuation, there were mild elevations in liver function tests and mild—generally transient—thrombocytopenia. Those are not things patients typically are bothered by. In fact, overall, quality of life was substantially better with T-DM1 than with TH. Putting all these data together, I do think that for some patients, particularly those who are likely to be particularly bothered by the toxicities of standard paclitaxel-trastuzumab, the T-DM1 could represent a reasonable alternative, as long as they appreciate that the potential toxicities of T-DM1 as well.
I should point out that the results of this trial are a relatively early look. This was looking at a roughly 3-year end point, and we certainly want to wait for more mature data overall before strong recommendations are made. But I think we were pleased by the relatively highly favorable outcome seen with the TDM-1 arm. The toxicity profiles, as I mentioned, overall weren’t substantially different in the quantitative sense, but the qualitative toxicities were different between the 2 arms. And quality of life seemed better with T-DM1.
Transcript Edited for Clarity