Preoperative combination radiotherapy (RT) and immune checkpoint inhibitor (ICI) in early-stage breast cancer: Influence of RT and ICI treatment sequencing (NCT04454528 post-hoc analysis).

615 Background: The anti-tumor immune effects of combination radiotherapy (RT) and immune checkpoint inhibitors (ICI) are synergistic. Optimal treatment sequencing, i.e. RT before ICI vs. RT after ICI, remains unclear. Assuming clinical equipoise, we conducted a phase 1b/2 multi-arm adaptive window of opportunity study to evaluate feasibility and treatment response after preoperative (preop) RT before ICI vs. RT after ICI vs. ICI alone. Methods: Participants with early-stage breast cancer planning for upfront surgery were eligible to enroll to receive a single preop intravenous dose of pembrolizumab (pembro, an ICI), 200 mg, either alone (Arm 3) or in combination with RT (single fraction, 7 Gy) given before (Arm 1) or after pembro (Arm 2). Primary outcome measures included 1) feasibility of administration of combination RT + pembro in a 21-day preoperative window, and 2) tumor response as measured by % tumor size change post-treatment, including the proportion achieving > 30% tumor reduction (TΔ30). Secondary endpoints included pathologic response as measured by residual cancer burden (RCB). Study sample size was powered for proportion of participants with TΔ30. Assuming 40% of treated participants would achieve TΔ30 compared to untreated participants, 15 subjects would yield 80% power at α = 0.05. Comparison of proportions was performed using Chi-squared test. Results: Between 1/6/2021 and 1/28/2025, 30 participants enrolled and completed the preop regimen assignment. Median follow-up was 634 days (IQR 256 –1120 days). All participants proceeded to surgery without delay. There were no adverse events (AEs) ≥ grade 3. There was one grade 2 immune-related AE (hypothyroidism). The proportion of participants achieving TΔ30 was 8 of 14 (57.1%), 6 of 11 (54.5%) and 0 of 5 (0%) in Arms 1, 2, and 3 respectively. For TNBC, TΔ30 rate was 6 of 7 (85.7%), 4 of 6 (66.7%) and 0 of 4 (0%) in Arms 1, 2, and 3 respectively. Compared with Arm 3, overall TΔ30 rate was significantly higher for those treated in Arm 1 or 2, p = 0.0307 and p = 0.0433; and for TNBC, p = 0.0088 and 0.0454, respectively. TΔ30 rate was similar in those treated with RT before (Arm 1) or RT after (Arm 2) ICI. The proportion of participants with major pathologic response (MPR) defined as RCB class 0 or 1 (RCB 0/1) treated in Arms 1, 2, and 3 were 7 of 14 (50%), 1 of 10 (10%), and 0 of 4 (0%) overall and 4 of 7 (57%), 0 of 5 (0%), and 0 of 3 (0%) for TNBC, respectively. The proportion of participants with RCB 0/1 in Arm 1 was numerically and statistically higher than Arm 2 overall and for TNBC, p = 0.0448 and p = 0.0479, respectively. Conclusions: RT given before ICI is more effective than RT given after ICI and resulted in MPR in 57% of participants with TNBC. This chemotherapy-free preop regimen may possibly facilitate chemotherapy omission decision in those achieving MPR in future studies. Clinical trial information: NCT04454528 .