Abstract PO1-03-07: Revisiting the potential of pathological complete response as a surrogate for long-term survival outcomes in triple-negative breast cancer at the trial-level

Abstract Background: Pathological complete response (pCR) has been used as a primary endpoint in neoadjuvant clinical trials for non-metastatic breast cancer and as a surrogate endpoint for long-term survival outcomes. The U.S. Food and Drug Administration (FDA) accepted pCR as a surrogate endpoint for the approval of new drugs in the neoadjuvant setting in high-risk early-stage breast cancer. However, though the Collaborative Trials in Neoadjuvant Breast Cancer (CTNeoBC) pooled analysis in 2014 showed that at the patient level achieving pCR was associated with longer overall survival (OS) and event-free survival (EFS), at the trial-level the association between high pCR rate and better OS and EFS outcomes was weak (coefficient of difference [R2]; 0.03 and 0.24, respectively) among all subtypes of breast cancer. A recent study reported that this association at the trial-level was similarly weak among Human Epidermal Growth Factor Receptor-2 (HER-2) positive breast cancer (R2; 0.02 and 0.23, respectively). However, the potential surrogacy of pCR rate for OS and EFS in triple-negative breast cancer (TNBC) in the neoadjuvant setting has not been well studied in a comprehensive manner. Moreover, because of significant advances in systemic treatment strategies in TNBC over the past decade including a combination of immune checkpoint inhibitors (ICIs) with conventional cytotoxic agents, the trial-level surrogacy needs to be reassessed in recently published neoadjuvant clinical trials for TNBC. Our primary objective is to assess the association between pCR rate and survival outcomes in neoadjuvant clinical trials for TNBC by collecting data from articles published after the CTNeoBC article in 2014. Methods: We performed a systematic literature search based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines in the PubMed, Cochrane Library, and Web of Science databases. Articles published between January 1, 2014, and March 31, 2023, were searched. We included only randomized control trials of neoadjuvant treatments for clinical stage I-III TNBC that compared two or more regimens followed by curative intent breast surgery. The selected trials had data for odds ratio (OR) for pCR and hazard ratios (HRs) either for OS or for EFS or disease-free survival (DFS). We excluded trials including estrogen receptor positive, progesterone receptor-positive, or HER-2 positive diseases, and those without data enough to obtain ORs for pCR, HRs for OS, EFS, or DFS. Unpublished data as a full article was also excluded. If EFS could not be obtained from the articles, DFS was used in place of EFS. Two independent investigators (TY and TF) searched and extracted data. Any disagreement was resolved by discussion with the third author (ZZ). A linear regression model on a logarithmic scale weighted by the sample size of each trial was performed and the R2 was used to assess the trial-level association between OR for pCR and HRs for OS or EFS. Results: We identified eight trials with a total of 2,926 patients. Three of the eight trials used ICIs in the intervention arm. One trial did not provide OS data and was not included in the OS analysis. In four trials, EFS was not available and DFS was used for EFS analysis. The R2 was 0.191 (95% confidence interval [CI], 0.165 to 0.218) for OS and 0.193 (95% CI, 0.167 to 0.219) for EFS. Conclusions: Similarly to the previous reports, our findings suggest that the surrogacy of the pCR rate for OS and EFS in recent neoadjuvant clinical trials for TNBC is weak. At this moment, there is no strong evidence to support using pCR as a surrogate for long-term survival outcomes in neoadjuvant clinical trials for TNBC. Citation Format: Toru Yoshino, Zao Zhang, Ryota Sato, Stanley Lipkowitz, Takeo Fujii. Revisiting the potential of pathological complete response as a surrogate for long-term survival outcomes in triple-negative breast cancer at the trial-level [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-03-07.