Davis, Todd

PURPOSE Triple-negative breast cancer (TNBC) is an aggressive subtype of breast cancer. Early-stage, high-risk patients with TNBC are treated with chemotherapy and pembrolizumab per the KEYNOTE-522 regimen. This report examines real-world clinical effectiveness and toxicity of this regimen. METHODS This 19-site retrospective analysis included all patients who received at least one dose of the KEYNOTE-522 regimen. Safety, efficacy, and health care encounters were collected. RESULTS 675 patients from 19 sites in the United States who received chemotherapy and pembrolizumab for early-stage TNBC were included. 547 (81.0%) patients underwent surgery and 250 (54.7%) patients had a pathologic complete response (pCR). No baseline characteristics were predictive of pCR status. 224 patients (33.2%) had an adverse drug event resulting in ≥1 dose reduction of chemotherapy and were significantly more likely to have residual disease. Immune-related adverse events (irAEs) were observed in 388 (57.5%) patients and grade 3+ irAEs were observed in 176 (26.1%) patients. There was no difference in pCR status between provider's choice chemotherapy schedules. There was no difference of pCR rate between Black and White patients (55.6% v 51.6%, P = .489). White patients were significantly more likely to have any grade and grade 3+ irAEs. Older patients (age ≥65 years) had similar pCR rates compared with younger patients; however, older patients were significantly more likely to have grade 3+ irAEs. CONCLUSIONS In this multi-institutional real-world cohort, we observed a lower pCR rate, higher rates of toxicity, frequent emergency health care utilization, and the reporting of rare irAEs at higher frequency than the KEYNOTE-522 regimen. These findings have important implications for clinical management of patients with early-stage TNBC and warrant further validation with real-world data sets.

Abstract Title: Real-world analysis of adverse events in patients with triple-negative breast cancer receiving therapy per KEYNOTE-522 Background: KEYNOTE-522 was a randomized, double-blind, placebo-controlled phase 3 trial which resulted in the FDA approval of pembrolizumab with neoadjuvant chemotherapy for patients (pts) with newly diagnosed, high-risk, early-stage triple-negative breast cancer (TNBC). Given the improvement in pathological complete response (pCR) and event-free survival rates, this regimen has emerged as standard-of-care (SOC) therapy. To date, real world outcome analyses are limited. Methods: In this retrospective, multicenter study, we examined pts with early-stage TNBC who received planned treatment per KEYNOTE-522 as SOC. 16 sites were included in the analysis. IRB approval was obtained from each participating site. Number and length of treatment delays, treatment related toxicities (both chemotherapy and immune-related) of all grades, and pCR rate were collected from the electronic medical record of each participating site, and deidentified data were shared for central analysis. Results: 577 pts were included in this analysis. The median age of the cohort was 52 [range 27-77]. 457 pts had T1-2 disease and 139 had T3-4 disease. 316 pts had N0 disease and 261 had N1-3 disease. 506 pts had baseline ECOG 0, 62 had ECOG 1, and 9 had ECOG 2-3. Of the 482 patients who had surgery at the time of this analysis, 219 patients had pCR (54.5%) and 192 pts were still receiving treatment. 217 patients (37%) had an adverse drug event (ADE) causing dose reductions and 228 patients (39.5%) had to discontinue treatment early. There were 360 unplanned care visits, with 91 patients having 2+ visits. 66 patients had hospitalizations due to the regimen with 102 total hospitalizations. 179 (31%) of patients had grade 3+ immune related adverse events (irAEs), including many that are rare and potentially serious. 317 (54%) had an immune related adverse effect (see Table 1). There was no significant difference in the frequency of grade 3+ irAE based on age or body mass index. We observed no difference in residual disease between patients who discontinued treatment early and those who did not. However, if patients had an ADR causing a dose reduction, pts were significantly more likely to have residual disease (P=0.039) Conclusions: In a multicenter real-world analysis, the KEYNOTE-522 regimen was associated with a high frequency of dose reductions, early treatment discontinuations, ER visits, and hospitalizations. This may have accounted for a lower pCR rate than observed in the landmark clinical trial. Most grade 3+ irAEs occurred at greater frequency in our real-world analysis. Providers should carefully monitor for toxicity to ensure optimal patient outcomes. Table 1 Citation Format: Mara Hofherr, Katherine Clifton, Spenser January, Emily Owen, Farah Raheem, Lida A. Mina, Suganya Arunachalam Karikalan, Lauren Lyons, Meredith Watson Rose, Katherine Madden, Jerline Hsin, Aimee Keegan, Wai Yu, Shawna Kraft, Allison Schepers, Emily Armgardt, Douglas Mazewski, Alison Svoboda, Kayla Harwood, Jodi Taraba, Yonatan Resnick, Shelly Hummert, Lisa Grate, Sidney Keisner, Jacob Hobbs, Todd Davis, Kristin Bastian, Dawn Minikel, Traci White, William Adler, Avneek Singh Sandu, Fouad Boulbol, Kelsey Finch, Andrew Davis. Real-World Analysis of Adverse Events in Patients with Triple Negative Breast Cancer Receiving Therapy per KEYNOTE-522 [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 PS14-06.