based on final data from a randomized trial.
The IMpassion031 trial showed significant improvement in pathological complete response (pCR) with the addition of atezolizumab to chemotherapy, as well as an acceptable safety profile, said Carlos H. Barrios, MD, of the Latin American Cooperative Oncology Group, Oncoclinicas, in Porto Allegre, Brazil, at the European Society for Medical Oncology (ESMO) Breast Cancer annual congress. Those findings were published in the Lancet in 2020.
Dr. Barrios reported data from a final analysis of the IMpassion031 trial, with data on event-free survival (EFS), disease-free survival (DFS) and overall survival (OS) in the intent-to-treat (ITT) and PD-L1–positive populations.
In the study, patients with early triple-negative breast cancer (eTNBC) and a primary tumor greater than 2 cm were randomized to 840 mg of atezolizumab once every 2 weeks plus a neoadjuvant chemotherapy regimen of nab-paclitaxel 125 mg/m2 once weekly for 12 weeks, followed by doxorubicin 60 mg/m2 plus cyclophosphamide 600 mg/m2 once every 2 weeks for 8 weeks. A total of 333 patients were randomized (165 atezolizumab and 168 placebo). Patients were stratified by stage II versus stage III, and by status of PD-L1, a protein that can predict treatment response (PD-L1 less than 1% vs. 1% or higher).
The primary endpoints (previously reported) were pathological complete response (pCR) in the ITT and PD-L1 populations. After a median follow-up of 39 months, the pCR was 58% in patients treated with atezolizumab versus 41% in those treated with neoadjuvant chemotherapy alone (P = .0044) in the ITT population, Dr. Barrios said. The added benefit from atezolizumab occurred regardless of the status of PD-L1.
Dr. Barrios reported the secondary outcomes of EFS, DFS, and OS in the intent-to-treat and PD-L1–positive populations. “This is a descriptive analysis, with no statistical comparison,” he emphasized.
The 2-year data on EFS, DFS, and OS consistently favored atezolizumab across key clinical subgroups, Dr. Barrios said. In the ITT population, 2-year EFS, DFS, and OS was 85%, 87%, and 95%, respectively, in the atezolizumab group and 80%, 83%, and 90%, respectively, in the placebo group. The results were similar, irrespective of PD-L1 status.
In the PD-L1–positive population, 2-year EFS, DFS, and OS was 89%, 91%, and 95%, respectively, in atezolizumab patients and 80%, 87%, and 91% in placebo patients.
Among patients without pCR at the time of surgery, 14 of 70 patients (20%) in the atezolizumab group and 33 of 99 patients (33%) in the placebo group received additional adjuvant systemic therapy. The most common adjunctive therapy was capecitabine.
As for safety, no new safety signals or treatment-related deaths were observed in the study. Overall, 70% of atezolizumab patients and 62% of placebo patients experienced grade 3 or 4 adverse events (AEs); 59% and 54% of which were treatment related. A total of 1% of patients in each group experienced grade 5 AEs. A total of 25% of atezolizumab patients and 20% of placebo patients experienced AEs leading to treatment discontinuation.
In a further exploratory analysis, pCR was highly predictive of long-term outcomes. Exploratory analysis of circulating tumor DNA (ctDNA) showed clearance in 89% of atezolizumab patients and 86% of placebo patients by the time of surgery.
Looking at the relationship between ctDNA, DFS, and OS, positive ctDNA was associated with a worse prognosis following surgery. As demonstrated in previous studies, pCR patients with negative ctDNA had the best DFS and OS. “In non-pCR patients with positive ctDNA, a numerical trend suggests improved overall survival with atezolizumab,” although the caveat is the very small numbers, Dr. Barrios said.
More research is needed, but in the final analysis, the significant pCR benefit seen with the addition of atezolizumab to chemotherapy for eTNBC translated into numerically improved EFS, DFS and OS, said Dr. Barrios. Additionally, “we should further analyze ctDNA to help select patients for further therapy.”
In a question-and-answer session, Dr. Barrios was asked how the results compared with other studies.
“We should not overinterpret the results,” he said. However, “what the IMpassion031 study shows is consistency; the results are aligned with previous studies addressing the same question of introducing immunotherapy,” in the patient population. Although the numbers in the IMpassion031 study did not reach statistical significance, it is important to recognize that they reflect previous research.
“In my opinion, looking at the whole field, immunotherapy is something we need to consider as part of the treatment of these patients,” said Dr. Barrios. However, more research is needed to better identify which patients do and do not need chemotherapy.