Association of low serum albumin concentration with reduced overall survival for patients with metastatic head and neck cancer receiving anti-programmed death receptor-1 therapy.

Background: Immunotherapy efficacy is modulated by immune competence which lacks well established clinical measures. Nutritional status influences immune competence and has been associated with poor outcomes with antimicrobial and oncologic therapies. The relationship between nutritional status and outcomes with anti-PD-1 checkpoint blockade has not been reported, and we examined this association in patients with metastatic head and neck (H&N) cancer. Methods: 114 patients with metastatic H&N cancer unselected for PD-ligand 1 (PD-L1) status received anti-PD-1 therapy. Low baseline serum albumin concentration was < 4.0 g/dL. Univariate logistic regression analyzed the relationship between albumin and overall response rate (ORR) measured by RECIST v1.1, and the association with body mass index (BMI) was performed with Chi-square analysis. Progression-free survival (PFS) and overall survival (OS) were analyzed using Kaplan-Meier methods and Cox proportional hazard regression. Results: Median follow-up was 8.6 months. Baseline characteristics including known PD-L1+ (n = 27/48 high albumin vs. n = 27/60 low albumin, p = 0.38) were comparable between groups. ORR was 22.3%. Albumin status not associated with ORR (p = 0.41) but was associated with BMI (p = 0.04). There was a trend towards lower median PFS for the low albumin group (2.4 months, 95% confidence interval [CI]: 1.9-3.7) compared to the high albumin group (4.6 months, 95% CI: 2.6-7.5) (p = 0.1). OS was significantly reduced for the low albumin group (median: 5.4 months, 95% CI: 3.6-9.5) compared to the high albumin group (median: 12.5 months, 95% CI: 9.1-17.3) (p-0.03). On multivariate analysis, albumin was the strongest independent predictor of poor OS (p = 0.01). Conclusions: Poor nutritional status as measured by low baseline serum albumin concentration is associated with worse outcomes in patients with metastatic H&N cancer receiving anti-PD-1 therapy and is a potential measure of immune competency. Investigation of clinical measures of immune competency including albumin as a marker of nutritional status in larger patient cohorts is warranted.