No Boost in pCR With Added Atezolizumab in HER2-Positive Early Breast Cancer

Adding atezolizumab (Tecentriq) to standard of care treatment for patients with high-risk, HER2-positive early breast cancer did not improve pathological complete response (pCR), according to the primary analysis of the phase III IMpassion050 trial.

In the intent-to-treat (ITT) population, pCR was achieved by 62.4% in the atezolizumab arm compared with 62.7% in the placebo arm, and by 64.2% and 72.5%, respectively, in the PD-L1-positive population, reported Jens Huober, MD, of Cantonal Hospital, Breast Center St. Gallen in Switzerland, during the European Society for Medical Oncology virtual meeting.

“One thing that should be considered are the statistical assumptions of this trial,” said Evandro de Azambuja, MD, PhD, of the Jules Bordet Institute in Brussels, who provided a critical analysis of the study during the plenary. “Which was that the pCR would be increased from 60% to 80% in the ITT population, and in the PD-L1-positive population from 70% to 90%. These are already quite high pathological complete response rates. Maybe [the researchers] were too ambitious in these endpoints.”

IMpassion050 was designed to test the efficacy and safety of atezolizumab compared with placebo when given in combination with standard of care, neoadjuvant doxorubicin plus cyclophosphamide, followed by paclitaxel-trastuzumab (Herceptin)-pertuzumab (Perjeta), in patients with HER2-positive early breast cancer.

The trial included 454 patients with HER2-positive early breast cancer, who were node positive and had a primary tumor size >2 cm. They were randomized 1:1 to the atezolizumab and placebo arms, and received 6 months of neoadjuvant therapy, followed by surgery. Patients then followed their allocated treatment to 52 weeks in total.

The primary outcome of the trial was the percentage of patients with pCR in the ITT population and the PD-L1-positive population.

The safety profile was consistent with other combination studies involving atezolizumab. Here, there were higher rates of grade 3 and 4 adverse events (AEs) in the neoadjuvant phase with atezolizumab (51.8% vs 43.6%), as well as higher rates of serious AEs (19.5% vs 13.3%). There were four grade 5 AEs in the neoadjuvant phase, and one in the adjuvant phase — all with atezolizumab, and all of which were confounded by comorbidities and concurrent events, such as COVID-19, the researchers said.

Huober suggested that the findings regarding pCR should be interpreted with caution until long-term results are available for event-free survival (EFS).

In this analysis, EFS results were considered to be immature at a median follow-up of 15.7 months in the atezolizumab arm and 15.9 months in the placebo arm. However, both Huober and de Azambuja noted that in the GeparNUEVO study, the addition of durvalumab (Imfinzi) to standard neoadjuvant therapy for triple-negative breast cancer only modestly increased pCR.

Yet, “patients in the durvalumab arm had very good invasive disease-free survival, distant disease-free survival, and overall survival,” de Azambuja pointed out.

“So pCR may not be the appropriate endpoint to catch the long-term benefit of the response to immunotherapy,” he added, suggesting that trials should be powered to detect the difference in survival endpoints like EFS and overall survival, “because this may allow a better measurement of immunotherapy’s antitumor effects over time.”

The failure of atezolizumab to show a benefit in pCR also “might reflect the need to better select patients who benefit most from the addition of immunotherapy,” de Azambuja said. “And for that we really need to have predictive biomarkers.”

The study’s results don’t mean that immunotherapy has no future in the HER2-positive setting, commented Nadia Harbeck, MD, PhD, of the University of Munich, who moderated the virtual plenary.

“I think there are so many unknown questions right now that we need answers for,” she said. “We have to find out which is the right setting, the right combination partner, and the right patient selection.”