Sex differences in burden of adverse events in patients receiving immunotherapy.

Abstract

2646 Background: The association between sex and the occurrence of immune-related adverse events (irAEs) in cancer patients who receive immune checkpoint inhibitors (ICIs) is an area of growing interest. Despite the collection of irAEs data in registrational studies, the impact of sex differentiation is rarely published, and comparative analyses using meta-analytic approaches of retrospective data has produced contrasting results. G-DEFINER is a multicentre project evaluating the irAEs inequalities between female (F) and male (M) cancer patients treated with ICIs. The project explored the sex and gender differences spanning from clinical to personal and lifestyle traits (including diet and quality of life). Gene-expression profiling for immune assessment, SNPs, cytokines (blood samples) and microbiota analyses (faecal samples) are also being performed to identify the features associated with irAEs development. Methods: irAE burden (B) was estimated as the sum of the average and worst grade of irAEs occurring at each time. The cumulative sum of time-specific Bs was used as irAE burden index (BI). Median BI (mBI) was modelled using quantile regression models. A univariable analysis (UA), 1st G≥2 irAE crude incidence curves (CCI) were estimated in a competing risks setting (competing event: deaths without irAE), and event-free survival (EFS) curves were estimated with the Kaplan-Meier method and compared with the log-rank test. Multivariable analysis (MA) of CCI and EFS was performed using Fine&Gray and Cox models, respectively. Results: At closing date (31st January 2023) the study has enrolled 256 patients. Our subgroup analysis includes 204 patients (86 F,118 M) with median follow-up of 13 months; IQR: 6.1-20 months). Patients’ age, previous oncologic treatments, cancer type (lung, melanoma, head&neck, colorectal MSI high, urogenital), PS ECOG, ICI type, concomitant chemotherapy (about 20%), and disease setting were well balanced in the two sex groups. F temporarily interrupted ICI administration more frequently than M (44 vs 29%, p=0.034) mainly due to irAEs (27 vs 11%); 83% F vs 71% M (p=0.069) contributed with 204 vs 240 any grade irAEs (significant differences: 36 vs 21% endocrine, p=0.016; 20% vs 9% hepatic, p=0.040). F more frequently developed G1 (64 vs 53%), G2 (49 vs 40%), G3 (24 vs 17%), G4 (6% vs 2%) irAEs. mBI was significantly higher in F both at UA (p=0.024) and MA (p=0.048). First G≥2 irAE6-(12) month estimates in F vs M were 55% vs 45% (65% vs 53%) (p=0.073 UA and 0.140 MA). EFS between group difference was not significant both at UA (p=0.206) and MA (p=0.729). Conclusions: F experience a higher irAE burden and require close monitoring in order to avoid ICI interruption. The results also underline the need of using a measure of irAE cumulative load to gain insights into between groups differences. Clinical trial information: NCT04435964 .