Abstract PO5-15-12: Spatial immune correlates of response to eribulin and pembrolizumab in metastatic triple negative breast cancer (mTNBC) on the ENHANCE1 trial

Abstract Background: PD-L1 is the only approved biomarker for pembrolizumab in metastatic breast cancer for response to combination chemo-immunotherapy. As it is not predictive of response in all cases, additional biomarkers are needed. Previously we have demonstrated that stromal tumor infiltrating lymphocytes (sTILs) are associated with response to therapy in patients treated with front-line combination chemoimmunotherapy on the ENHANCE-1 study with eribulin and pembrolizumab. However, sTILs was not predictive of response in patients treated with prior lines of chemotherapy in the metastatic setting. We used multiplexed immunofluorescence on whole slide images to characterize associations between specific immune populations and outcomes in patients treated prospectively on the ENHANCE-1 study. Methods: ENHANCE-1 was a single arm phase Ib/II trial which evaluated the efficacy and safety of eribulin and pembrolizumab in 167 patients (pts) with mTNBC who had received 0-2 prior lines of therapy (66 pts in the first line setting [stratum 1] and 101 pts with 1-2 prior lines of therapy [stratum 2]). 138 patient samples were evaluable for our study: 58 samples from stratum 1 and 80 samples from stratum 2. Objective response rate (ORR) was defined as percentage of pts with either complete response (CR) or partial response (PR) by RECIST 1.1. The ORR was 25.8% in stratum 1 and 22.5% in stratum 2 for the 138 patients analyzed. We utilized whole tissue sections and quantitative multiplexed immunofluorescence (qmIF) stained with: CD4, FoxP3, CD8, CD56 and pancytokeratin to characterize T-cells, NK cells and tumor cells. Halo (Indica labs) was used to segment and threshold cells for positive markers as well as to identify tumor and stromal tissue areas. Results: We found that the ratio of CD8+ to CD4+FoxP3+ T-cells were associated with response, irrespective of stratum (p=0.005). As biopsies were not required immediately prior to enrollment on ENHANCE-1, we also evaluated samples based on prior number of treatments since biopsy to enrollment. We found that only 58/138 patient samples evaluated were obtained without intervening treatment: 35 out of 58 from stratum 1 and 12 out of 80 from stratum 2. When samples were evaluated by stratum and the timing of tissue acquisition we found the ratio of CD8+ to CD4+FoxP3+ T-cells was more significantly associated with response (< 0.001) in patients whose sample were collected prior to enrollment, regardless of patient stratum. The ratio was not predictive in samples taken prior to treatments before enrollment. In addition, we found stromal and intratumoral CD8+ T-cell density is associated with response (p=0.02 and p=0.03 respectively) across both stratum. We did not find any significant changes assessed by qmIF in patient samples which had intervening treatment between sample collection and enrollment onto ENHANCE-1 trial. Conclusion: In this population of patients with mTNBC treated prospectively with eribulin and pembrolizumab, we find that a higher ratio of CD8+ to CD4+FoxP3+ T-cells and stromal and intratumoral CD8+ T-cell density is associated with response in patients with biopsy samples obtained prior to enrollment without intervening treatments after sample collection. The tumor microenvironment on patient samples obtained before one or more lines of therapy prior to enrollment on study are not predictive of response to treatment. Further characterization of the TME via quantitative immunofluorescence is ongoing. This study was funded by Eisai Citation Format: Matthew Kearney, Hua Guo, Rami Vanguri, Qi Wang, Eileen Connolly. Spatial immune correlates of response to eribulin and pembrolizumab in metastatic triple negative breast cancer (mTNBC) on the ENHANCE1 trial [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 PO5-15-12.