ABSTRACT
BACKGROUND: PD-L1 tumor proportion score (TPS) and tumor mutational burden (TMB) are key predictive biomarkers for immune checkpoint inhibitors (ICI) efficacy in non-small cell lung cancer (NSCLC). Data on their variation across multiple samples are limited. METHODS: Patients with NSCLC and multiple PD-L1 TPS and/or TMB assessments were included. Clinicopathologic and genomic data were analyzed according to PD-L1 and TMB variation. RESULTS: In total, 402 PD-L1 sample pairs and 413 TMB sample pairs were included. Concordance between pairs was moderate for PD-L1 (ρ=0.53, P<0.0001) and high for TMB (ρ=0.80, P<0.0001). Shorter time between biopsies correlated with higher concordance in PD-L1, but not in TMB. Major increases (ΔTPS≥+50%) and decreases (ΔTPS≤-50%) in PD-L1 were observed in 9.7% and 8.0% of cases, respectively. PD-L1, but not TMB, decreased with intervening ICI (P=0.02). Acquired copy number loss of CD274, PDCD1LG2, and JAK2 were associated with major decrease in PD-L1 (q<0.05). Among patients with multiple PD-L1 assessments before ICI, cases where all samples had a PD-L1 ≥1%, compared to cases with at least one sample with PD-L1 <1% and another with PD-L1 ≥1%, achieved improved objective response rate and progression-free survival (PFS). Among patients with at least one PD-L1 <1% and one ≥1% before ICI, cases where the most proximal sample was PD-L1 ≥1% had longer median PFS compared to cases where the most proximal PD-L1 was <1%. Among patients with multiple TMB assessments before ICI, patients with a TMB ≥10 mut/Mb based on the most recent assessment, as compared to those with a TMB <10 mut/Mb, achieved improved PFS and OS to ICI; instead, no differences were observed when patients were categorized using the oldest TMB assessment. CONCLUSION: Despite intrapatient concordance in PD-L1 and TMB, variation in these biomarkers can influence ICI outcomes, warranting consideration for reassessment prior to ICI initiation when feasible.