Real‐World Toxicity and Outcomes of KEYNOTE‐522‐Style Neoadjuvant Pembrolizumab for Triple‐Negative Breast Cancer in Australia

Background Following the KEYNOTE‐522 trial, neoadjuvant pembrolizumab with chemotherapy (chemo‐IO) became the standard of care in Australia for patients with triple‐negative breast cancer (TNBC). In KEYNOTE‐522, patients achieved significantly higher pathological complete response (pCR) at the cost of more immune‐related adverse events (irAEs). The objective of this study was to collect real‐world data on the use of neoadjuvant pembrolizumab in patients with TNBC to identify patterns of irAEs and outcomes. Patients and Methods Retrospective data were collected across three tertiary hospitals in Melbourne, Australia, for all patients diagnosed with TNBC who commenced neoadjuvant pembrolizumab combined with chemotherapy from December 2022 until August 2024. Data were collected until October 30, 2025, and included baseline demographics, neoadjuvant treatment and irAEs, breast surgery, and pCR rates, and were analyzed using descriptive statistics and Kaplan‐Meier estimates. Results Fifty‐four female patients received neoadjuvant chemo‐IO (median age = 52 years, median follow‐up = 26 months). Three patients had a history of autoimmune disease. 69% had nodal involvement at diagnosis. 65% of patients experienced at least one irAE, most commonly thyroid dysfunction (28%), adrenal dysfunction (20%), colitis (9%), and arthralgias (9%). 43% of those with irAEs were hospitalized, 9% required intensive care, and 51% have ongoing sequelae. The median time to onset of irAE was 84 days (22–505 days). 50% achieved pCR. There was no difference in pCR between those who experienced an irAE and those who did not ( p = 0.081). Conclusion Real‐world data show a higher proportion of irAEs associated with neoadjuvant pembrolizumab in TNBC than the original KEYNOTE‐522 study, with a notable proportion leading to hospitalization. This information can be used to guide shared decision‐making.