Abstract B050: Validation of body mass index (BMI) as a prognostic factor in patients (pts) treated with immune checkpoint inhibitors (ICI) across multiple cancer types (CT) and the impact of confounding factors (CF)

Abstract

Abstract Introduction: Obesity results in PD1-mediated T-cell dysfunction in preclinical models and improved outcomes from ICI. We wanted to validate this clinical association in a prospective cohort of pts receiving PD1/L1 inhibitors across multiple CT and investigate the potential impact of CF, such as age and concomitant medications (CM) that may interact with obesity. Methods: Clinicopathological data from 310 pts treated with ICI at VHIO phase 1 Unit from Aug'12 to Jul'18 were investigated. Frequent CM included metformin (M), statins (S) and others. Associations between different variables and progression-free survival (PFS) were assessed with univariable and multivariate Cox regression models and survival data were calculated by the Kaplan-Meier method. Tumor types were stratified by proven PD1/L1 inhibitor efficacy versus unknown PD1/L1 sensitivity. Results: Out of 310 pts, median age was 59.8 years (y), 54.2% were male and most frequent tumor type was melanoma (19.4%) and lung (13.2%). In univariate models, pts with BMI>25 kg/m2 (n=147, 47%) had increased PFS (3.5 months [m], CI95% 2.8-5.5) as compared to those with BMI<25 kg/m2 (2.3m, CI95% 1.9-3.7, HR=0.73, p=0.02). Overweight and obese pts were older and more often received S and M than those with normal BMI or underweight. Pts exposed to S (n=50, 16%) had increased PFS (6.3 m, CI95% 3.7-9.5) as compared to those without S use (2.7m, CI95% 2.1-3.5, HR=0.63, p=0.03). Other CM had no impact on pts outcomes. In a multivariable model adjusting for age, tumor type, S and M use, increased BMI had no longer a significant impact on PFS (HR=0.83 [CI95% 0.63-1.11], p=0.21), and only age remained a signficant prognostic fator (10-y increase in age, HR=0.86 [CI95% 0.77-0.95], p=0.005). Conclusion: In univariate models, increased BMI is associated with prolonged PFS in pts receiving ICI. However, after adjustment for CF such as age and CM, the association loses statistical significance. Larger randomized cohorts need to be investigated before conclusive statements on obesity as a determinant of benefit with ICI. Citation Format: Analía Azaro, Alejandra Ivars, Ignacio Matos, Omar Saavedra, Itziar Gardeazabal, Juan Martin-Liberal, Cinta Hierro, María Vieito, Irene Braña, Cristina Viaplana, Guillermo Villacampa, Gemma Sala, María Alsina, Ana Callejo, Helena Verdaguer, Guillem Aegiles, Ana Oaknin, Eva Muñoz, Rodrigo Dienstmann, Elena Garralda. Validation of body mass index (BMI) as a prognostic factor in patients (pts) treated with immune checkpoint inhibitors (ICI) across multiple cancer types (CT) and the impact of confounding factors (CF) [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2019 Oct 26-30; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2019;18(12 Suppl):Abstract nr B050. doi:10.1158/1535-7163.TARG-19-B050