Early pembrolizumab clearance as prognostic biomarker for non-response in patients with advanced non-small cell lung cancer.

Immune checkpoint inhibitors have improved survival rates in patients with advanced-stage non-small cell lung cancer; however, the majority obtain no long-term benefits. We investigated pembrolizumab clearance as an early prognostic biomarker and evaluated its accuracy using a limited sampling strategy. Pembrolizumab clearance was calculated using non-linear mixed effects modeling, and cut points were determined using maximally selected rank statistics. The prognostic value for survival was estimated using univariate Cox regression analysis. Sensitivity, specificity, and positive and negative predictive values were calculated to evaluate the performance in identifying response (defined as disease control at 6 months). The accuracy of a limited sampling strategy was evaluated using MPE and NRSME. Among 303 patients included, 65% experienced disease progression, and 60% died. Patients with pembrolizumab clearance above 0.232 L/day at the first dose were more likely to have disease progression (HR = 1.98, 95% CI [1.21, 3.26], p = .007) or poor survival (HR = 2.04, 95% CI [1.16, 3.59], p = .014). A diminished decrease in clearance (<15.8%) at 6 weeks was also significantly associated with progression (HR = 1.46, 95% CI [1.12, 1.92], p = .006) and poor survival (HR = 1.82, 95% CI [1.35, 2.45], p = .000). Pembrolizumab clearance showed high sensitivity (0.96, 95% CI [0.92, 0.99]), but moderate positive predictive value (0.48). Limited sampling matched comprehensive sampling accuracy (MPE = +4.5% vs. +3.2%, NRSME = 16.8% vs. 14.2%). Early pembrolizumab clearance is a feasible prognostic biomarker for survival, with opportunities to enhance its positive predictive value before clinical implementation.