Pathological complete response to predict long term outcomes in the I-SPY2 TRIAL.

Abstract

597 Background: The I-SPY2 platform trial uses pathologic complete response (pCR) to identify active agents in patients with stage 2/3 breast cancer. Only patients with HR+ HER2- tumors, and low risk 70-gene assay results are excluded. 2118 patients were enrolled from March 2010 through June 2022 with follow-up data from 2005 patients as of January 2024. We have reported the improved event-free (EFS) and distant-recurrence free (DRFS) survival for 950 patients (1) who achieved a pathological complete response (pCR). Here we report the outcomes for the additional 1055 patients not included in our initial report. Methods: Patients have been randomized to one of 25 arms. Residual Cancer Burden (RCB) assessment was performed by site pathologists who were trained to ensure uniform response classification; RCB 0 = pCR. EFS is defined as time to local or distant recurrence or death. DRFS is defined as time to distant recurrence or death. Patients without events were censored at time to last follow-up. Association between pCR and EFS/DRFS was assessed using Cox proportional hazard modeling with significance assessment using the log rank test. EFS and DRFS at 3 years are estimated using the Kaplan Meier method. Results: Median follow up was 3.3 years (0.05-10.2 y). There were 144 EFS and 123 DRFS events. pCR rates varied by subtype, but achieving pCR was associated with an EFS of 96% and DRFS of 96%. For patients without pCR the EFS was 82% and DRFS was 84% at 3 years. The relationship between pCR and EFS was significant and clinically impactful within each subtype. Conclusions: pCR continues to have a strong positive association with EFS and DRFS as in our initial report. It is notable that EFS for non-pCR patients improved to 82% from our initial report of 78%. Initially few patients received additional adjuvant therapy except for standard of care endocrine and HER2-targeted therapy. Increased adoption of post neoadjuvant therapy with the approval of several agents for patients with non-pCR likely contributed to this improvement. I-SPY also reported a shift in residual cancer burden for patients receiving investigational therapies, which also may improve outcome for non-pCR patients. Taken together, these data demonstrate the value of pCR as an early endpoint regardless of tumor type or drugs delivered. These findings emphasize the need for new agents that will allow all patients to achieve pCR and potentially avoid the need for additional therapy after surgery. 1. Yee, et al. JAMA Oncol 2020. 2. Symmans, et al. JAMA Oncol 2021. Clinical trial information: NCT01042379 . [Table: see text]