Abstract PS14-08: Safety evaluation from the KEYNOTE-355 study of pembrolizumab plus chemotherapy vs placebo plus chemotherapy in patients with previously untreated, locally recurrent inoperable or metastatic triple-negative breast cancer

Abstract Background: In KEYNOTE-355 (NCT02819518), pembrolizumab (pembro) + chemotherapy (chemo) led to statistically significant and clinically meaningful improvements in PFS and OS (primary endpoints) vs placebo (pbo) + chemo in patients (pts) with previously untreated PD-L1 positive advanced TNBC (combined positive score ≥10). In the overall safety population at final analysis, treatment-related AEs occurred in 96.3% of pts in the pembro + chemo arm and 95.0% of pts in the pbo + chemo arm; immune-mediated AEs occurred in 26.5% and 6.4% of pts, respectively. We report additional safety information, beyond the already reported safety results, on immune-mediated AEs and management from the final analysis of KEYNOTE-355. Methods: Eligible pts were randomized 2:1 to receive pembro 200 mg or pbo Q3W for up to 35 cycles + investigator’s choice of nab-paclitaxel, paclitaxel, or gemcitabine-carboplatin. Safety was assessed in all randomized pts who received ≥1 dose of study treatment. AEs were monitored throughout the study and for 30 d post-treatment (90 d for serious AEs). Results: At final analysis (data cut-off: June 15, 2021), median treatment duration was 6.1 (range, 0.0–48.8) mo in the pembro + chemo arm (n = 562) and 5.3 (range, 0.0–51.5) mo in the pbo + chemo arm (n = 281). Of 157 pts with immune-mediated AEs and infusion reactions in the pembro + chemo arm, 125 had grade 1-2 events and 32 had grade 3-4 events; none were grade 5. The most common immune-mediated AEs with pembro + chemo were hypothyroidism (15.8%) and hyperthyroidism (4.3%); infusion reactions occurred in 3.7% of pts (table). Of 89 pts with hypothyroidism, median time to onset was 105 d (range, 19–707 d) and 70 were treated with thyroid replacement, suggesting an endocrine abnormality and need for continued thyroid replacement. Of 24 pts with hyperthyroidism, median time to onset was 53.5 d (range, 20–209 d) and 2 were treated with corticosteroids. Of 21 pts with infusion reactions, median time to onset was 147 d (range, 1–729 d) and 10 were treated with corticosteroids. Other immune-mediated AEs of interest in the pembro + chemo arm were pneumonitis (2.5%) and adrenal insufficiency (1.4%); most of these events were grade 2–3. Of 14 pts with pneumonitis, median time to onset was 99.5 d (range, 29–288 d) and 11 were treated with corticosteroids; median episode duration was 56 d. All 8 pts with adrenal insufficiency were treated with hormone replacement. There were no reports of hypophysitis. Conclusion: In pts with previously untreated advanced TNBC, pembro + chemo had a manageable safety profile that was generally consistent with the known safety profiles of pembro and the chemo regimens. Most immune-mediated AEs and infusion reactions were grade 1-2, manageable with treatment interruption, corticosteroids, and/or hormone replacement therapy, and did not result in discontinuation of study treatment. Together with the efficacy findings, our results support pembro + chemo as a standard of care regimen for these pts. Table. Citation Format: Hope Rugo, Peter Schmid, Zbigniew Nowecki, David Cescon, Seock-Ah Im, Mastura Yusof, Carlos Gallardo, Hiroji Iwata, Carlos Barrios, Sherene Loi, Xuan Zhou, Xiaoli Zhang, Wilbur Pan, Vassiliki Karantza, Javier Cortés. Safety evaluation from the KEYNOTE-355 study of pembrolizumab plus chemotherapy vs placebo plus chemotherapy in patients with previously untreated, locally recurrent inoperable or metastatic triple-negative breast cancer 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 PS14-08.