Abstract PS2-07-01: Cemiplimab in High-risk or Locally Advanced Luminal and Triple Negative Breast Cancer (CemiHALT)

Abstract Introduction: PD-1 inhibitors (PD-1i) are associated with significantly higher pathological complete responses (pCR) in patients (pts) with early-stage triple-negative breast cancer (TNBC) as well as luminal/estrogen receptor-positive (ER+) breast cancer (BC) with higher pCR rates reported in PD-L1-positive pts. However, some PD-L1-negative pts respond as well. Most clinical trials of PD-1i have thus far focused on PD-L1 expression for pt eligibility. The alternative PD-1 ligand, PD-L2, with reported ∼3-fold greater affinity for PD-1, remains understudied. We recently reported that high levels of PD-L2 in pts with treatment-naïve ER+ BC were an independent predictor of shorter progression-free survival. We therefore sought to assess the correlation of PD-L1 and PD-L2 with pCR in TNBC and ER+/HER2-negative BC treated with PD-1i and neoadjuvant chemotherapy (NACT). Methods: CemiHALT (NCT04243616) is a single-arm, open-label Phase II study of PD-1i cemiplimab (Regeneron Pharmaceuticals) and NACT in pts with high-risk TNBC or locally advanced ER+/HER2-negative BC with positive PD-L1 and/or PD-L2 tumor expression. Baseline PD-L1 or PD-L2 expression in ≥1% of cancer cells or stromal immune cells in diagnostic core biopsies was considered positive. Pts were treated with standard taxane and anthracycline-based NACT (weekly paclitaxel x 12 with carboplatin added for TNBC only, followed by adriamycin and cyclophosphamide x 4 cycles). The study was designed to administer two doses of cemiplimab (350mg standard dose) with 1st dose as monotherapy to prime the immune microenvironment and 2nd dose concurrently with NACT start. The trial was later amended to incorporate cemiplimab concurrently throughout NACT for TNBC pts after FDA approval of neoadjuvant pembrolizumab per KEYNOTE-522 in July 2021. Radiographic responses were assessed by RECIST criteria. Wilcoxon paired analyses and Spearman correlation were used to correlate PD-L1 and PD-L2 protein expression with pCR rates and radiographic responses respectively. Fisher’s test and McNemar’s tests were used for comparison of variables and PD-L1/PD-L2 expression between ER+ and TNBC subtypes. All analyses were planned with 80% power and a significance level of 0.05. Results: Thirty-six pts were enrolled between June 2020 and January 2025, with 23 ER+/HER2-negative and 13 TNBC pts (5 with 2 doses, 8 with 8 doses). Median age at diagnosis was 49yrs (range 26-70) with 81% White, 14% African American, and 5.6% Hispanic pts. TNBC pts had high baseline PD-L1 and PD-L2 protein expression (92% and 100% respectively), whereas ER+ pts had a higher PD-L2 expression at 83% vs 61% for PD-L1. Of the evaluable 31 pts, the pCR rate was 23% in the overall cohort with a significantly higher pCR seen in TNBC compared to ER+/HER2-negative pts (50% vs 5.3%, p=0.007) and no differences seen by doses of cemiplimab in TNBC pts. Higher proportion of TNBC pts had complete and partial responses radiographically (25% and 58.3% respectively) while ER+/HER2- pts had more stable disease (41.6%). Baseline PD-L1 and/or PD-L2 expression was not significantly associated with pCR rates or radiographic responses. Treatment-related adverse events were not higher than expected. Conclusion: Addition of cemiplimab to NACT resulted in 23% pCR rate in the overall cohort with a significantly higher pCR (50%) seen in TNBC pts. Baseline PD-L1 and/or PD-L2 expression was not associated with pCR rates or radiographic responses. This study incorporated only 2 doses of cemiplimab for ER+ pts which could explain the low pCR rates in this cohort. Given the discordance in PD-L1 and PD-L2 expression with a higher prevalence of PD-L2 in ER+ pts, further studies exploring PD-L1/PD-L2 as a combined marker for PD-1i use are warranted. To our knowledge, this is the first clinical trial with PD-L2 status as an eligibility criterion for PD-1i treatment. Citation Format: L. N. Chaudhary, H. Abid, J. M. Jorns, A. Szabo, Y. Sun, S. Kamaraju, Y. Cheng, A. Kong, T. W. Yen, C. Patten, C. S. Cortina, E. Weil, I. Chervoneva, C. R. Chitambar, H. Rui. Cemiplimab in High-risk or Locally Advanced Luminal and Triple Negative Breast Cancer (CemiHALT) abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS2-07-01.