Pathologic complete response (pCR) to neoadjuvant treatment with or without atezolizumab in triple-negative, early high-risk and locally advanced breast cancer: NeoTRIP Michelangelo randomized study

•Atezolizumab with neoadjuvant carboplatin/nab-paclitaxel led to non-significantly higher pCR rate in PD-L1+ TNBC. •In other trials neoadjuvant ICIs and chemotherapy significantly improved pCR rate, but EFS was independent of pCR. •Our analysis of NeoTRIP supports that pCR may not be an appropriate surrogate endpoint for the role of ICIs in early TNBC. •Lack of pCR improvement may be misleading as to the impact of atezolizumab on long-term efficacy in high risk TNBC. Background High-risk triple-negative breast cancers (TNBCs) are characterized by poor prognosis, rapid progression to metastatic stage and onset of resistance to chemotherapy, thus representing an area in need of new therapeutic approaches. Programmed death-ligand 1 (PD-L1) expression is an adaptive mechanism of tumour resistance to tumour-infiltrating lymphocytes, which in turn are needed for response to chemotherapy. Overall, available data support the concept that blockade of PD-L1/programmed cell death protein 1 checkpoint may improve efficacy of classical chemotherapy. Patients and methods Two hundred and eighty patients with TNBC were enrolled in this multicentre study (NCT002620280) and randomized to neoadjuvant carboplatin area under the curve 2 and nab-paclitaxel 125 mg/m2 intravenously (i.v.) on days 1 and 8, without (n = 142) or with (n = 138) atezolizumab 1200 mg i.v. on day 1. Both regimens were given q3 weeks for eight cycles before surgery followed by four cycles of an adjuvant anthracycline regimen. The primary aim of the study was to compare event-free survival (EFS), and an important secondary aim was the rate of pathological complete response (pCR defined as the absence of invasive cells in breast and lymph nodes). The primary population for all efficacy endpoints is the intention-to-treat (ITT) population. Results The ITT analysis revealed that pCR rate after treatment with atezolizumab (48.6%) did not reach statistical significance compared to no atezolizumab [44.4%: odds ratio (OR) 1.18; 95% confidence interval 0.74-1.89; P = 0.48]. Treatment-related adverse events were similar with either regimen except for a significantly higher overall incidence of serious adverse events and liver transaminase abnormalities with atezolizumab. Conclusions The addition of atezolizumab to nab-paclitaxel and carboplatin did not significantly increase the rate of pCR in women with TNBC. In multivariate analysis, the presence of PD-L1 expression was the most significant factor influencing the rate of pCR (OR 2.08). Continuing follow-up for the EFS is ongoing, and molecular studies are under way. High-risk triple-negative breast cancers (TNBCs) are characterized by poor prognosis, rapid progression to metastatic stage and onset of resistance to chemotherapy, thus representing an area in need of new therapeutic approaches. Programmed death-ligand 1 (PD-L1) expression is an adaptive mechanism of tumour resistance to tumour-infiltrating lymphocytes, which in turn are needed for response to chemotherapy. Overall, available data support the concept that blockade of PD-L1/programmed cell death protein 1 checkpoint may improve efficacy of classical chemotherapy. Two hundred and eighty patients with TNBC were enrolled in this multicentre study (NCT002620280) and randomized to neoadjuvant carboplatin area under the curve 2 and nab-paclitaxel 125 mg/m2 intravenously (i.v.) on days 1 and 8, without (n = 142) or with (n = 138) atezolizumab 1200 mg i.v. on day 1. Both regimens were given q3 weeks for eight cycles before surgery followed by four cycles of an adjuvant anthracycline regimen. The primary aim of the study was to compare event-free survival (EFS), and an important secondary aim was the rate of pathological complete response (pCR defined as the absence of invasive cells in breast and lymph nodes). The primary population for all efficacy endpoints is the intention-to-treat (ITT) population. The ITT analysis revealed that pCR rate after treatment with atezolizumab (48.6%) did not reach statistical significance compared to no atezolizumab [44.4%: odds ratio (OR) 1.18; 95% confidence interval 0.74-1.89; P = 0.48]. Treatment-related adverse events were similar with either regimen except for a significantly higher overall incidence of serious adverse events and liver transaminase abnormalities with atezolizumab. The addition of atezolizumab to nab-paclitaxel and carboplatin did not significantly increase the rate of pCR in women with TNBC. In multivariate analysis, the presence of PD-L1 expression was the most significant factor influencing the rate of pCR (OR 2.08). Continuing follow-up for the EFS is ongoing, and molecular studies are under way.