Blood-based tumour mutation index act as prognostic predictor for immunotherapy and chemotherapy in non-small cell lung cancer patients

Abstract

Abstract Background Circulating tumour DNA (ctDNA)-based sequencing might provide a simple test for the stratified model of non-small cell lung cancer (NSCLC). Here, we aimed to assess the ctDNA sequencing-based tumour mutation index (TMI) model for screening responders (from non-responders) among NSCLC patients who received monotherapy with docetaxel or atezolizumab. Methods We performed a retrospective analysis of the POPLAR (NCT01903993) and OAK (NCT02008227) trials. We identified three biomarkers, blood tumour mutation burden (bTMB), sensitive blood tumour mutation burden (sbTMB) and unfavourable mutation score (UMS), of the ctDNA profiles. After integrating the advantages and disadvantages of the three independent biomarkers, we developed the TMI model and identified NSCLC patients who may benefit from monotherapy with docetaxel or atezolizumab in terms of overall survival (OS). Results The TMI model as a stratified biomarker for docetaxel responders provided a median OS duration of 5.55 months longer than non-responders in the OAK cohort, with a significantly decreased hazard ratio (HR). Moreover, atezolizumab responders had a 10.21-month OS advantage over atezolizumab non-responders in the OAK cohort via TMI stratification, and the HR was also decreased significantly. The TMI demonstrated effectiveness for stratifying responders in the POPLAR cohort. Importantly, we found that the TMI model could screen additional responders upon combining the cohorts from the POPLAR and OAK trials after adjustment. Conclusion In the present study, we provide a novel TMI model for screening responders (from non-responders) among NSCLC patients who received the 2nd-line monotherapy with docetaxel or atezolizumab. We believe that the biomarker TMI will potentially be effective for the clinical treatment of NSCLC in the future.