Blood-based kinase activity profiling to predict response to immune checkpoint inhibitors in patients with advanced stage NSCLC: the prospective IOpener study
Karlijn de Joode, Harry J Groen, Michel van den Heuvel, Jeroen S Kloover, Femke Van Der Meer, Cor H Van der Leest, Ben Van Den Borne, Rik De Wijn, Daan P Hurkmans, Dianne M A Van den Heuvel, Herbert M Pinedo, Ellen Kapiteijn, Reno Debets, Els M E Verdegaal, Sjoerd H Van der Burg, John P Groten, Joachim G J V Aerts, Ron H J MathijssenBackground
The programmed death-ligand 1 (PD-L1) tumor proportion score (TPS) is used as a biomarker to predict benefit from immune checkpoint blockade (ICB) in patients with non-small cell lung cancer (NSCLC). However, additional biomarkers are needed. The potential of kinase activity profiling of peripheral blood mononuclear cells (PBMCs) for predicting response to ICB was previously shown in a discovery study. The goal of this prospective study is to evaluate the predictive value of blood-based kinase profiling for ICB response in patients with NSCLC and to compare the performance and added value to TPS.
Methods
Advanced stage patients with NSCLC, treated with anti-programmed cell death protein 1 ICB (±chemotherapy) were included in this multicenter study (N=210 patients). PBMCs were collected prior to ICB treatment and profiled using a peptide microarray with multiple kinase substrates (PamChip). Classification analysis was performed to discriminate between patients with or without progressive disease by Response Evaluation Criteria in Solid Tumors (RECIST) V.1.1 <24 weeks after treatment start. A predictive model was first established based on TPS and kinase activity profiles. Subsequently, the performance of the model was tested in a second (validation) cohort.
Results
In the validation cohort, a significantly higher progression-free survival (PFS) rate was observed for patients with predicted benefit than for patients without predicted benefit according to the kinome-based prediction model (HR=0.56, p=0.01), which tended better compared with TPS alone (HR=0.66, p=0.07). When the kinome-based model was combined with TPS, the difference in PFS between patients with and without predicted benefit was further increased (HR=0.38, p<0.001). PFS rates were also higher for patients with predicted benefit in subgroups with decreased PD-L1 expression (TPS <1% (HR=0.46, p=0.027) and TPS 1–50% (HR=0.20, p=0.005)).
Conclusions
This study shows the predictive value of kinase activity profiling of PBMCs for ICB response in advanced NSCLC, which was superior compared with TPS. In particular, for patients with NSCLC with TPS <50%, combining TPS with the blood-based kinase test may identify patients with limited benefit from ICB.
Trial registration number
NTR7015/NL6828.