毒性
医学
癌症研究
肿瘤科
细胞周期检查点
内科学
细胞周期
癌症
作者
Megan Othus,Sandip Pravin Patel,Young Kwang Chae,Eliana Dietrich,Howard Streicher,Elad Sharon,Razelle Kurzrock
摘要
Abstract Background Associations between immune-related adverse events (irAEs) from checkpoint inhibitor therapy and outcomes have been previously evaluated, with most prior research finding a positive association between toxicity and survival. This prior research has generally reported on more common tumor types. We use a unique data resource of a federally-funded basket trial ((NCT02834013) for patients with rare cancers (N = 684) to evaluate associations between irAEs and overall survival and progression-free survival. Methods Patients were treated with nivolumab and ipilimumab; the trial was opened at > 1000 sites. Landmark Cox regression models were used to assess first cycle irAE associations with progression-free and overall survival. Results We found that grade 1-2 treatment-related irAEs in the first cycle of therapy were associated with longer overall survival (OS) (multivariable hazard ratio, 95% confidence interval, p-value: 0.61, 0.49-0.75, p < .001) compared to no treatment-related irAE, while grade 3-4 irAEs were associated with shorter OS (HR = 1.41, 95% CI = 1.04-1.90, p = .025). Similar, but weaker, associations were observed with progression-free survival (PFS) and grade 1-2 treatment-related irAEs: HR = 0.83, 95% CI = 0.67-1.01, p = .067 and grade 3-4: HR = 1.35, 95% CI = 1.02-1.78, p = .037 compared to no treatment-related irAEs. Grade 1-2 dermatologic toxicity was associated with improved OS compared to other grade 1-2 toxicities (HR = 0.67, 95% CI = 0.52-0.85, p = .002). There was no significant OS difference between patients with Grade 1-2 fatigue, gastrointestinal, metabolic, hepatic, endocrine, and thyroid toxicities vs other Grade 1-2 toxicities. Conclusions In this large cohort of patients with rare tumors receiving checkpoint inhibitor therapy, grade of irAE in the first cycle was predictive for survival.
科研通智能强力驱动
Strongly Powered by AbleSci AI