Plasma metabolomic profiling to predict durable response and survival with afatinib in EGFR-mutated non-small cell lung cancer: A prospective, multicenter, observational biomarker study

Abstract

While EGFR tyrosine kinase inhibitors (TKIs) like afatinib are standard treatments for EGFR-mutated non-small cell lung cancer (NSCLC), predicting which patients will achieve durable responses remains challenging. Given the emergence of more toxic and costly combination therapies, identifying candidates who can achieve long-term survival with TKI monotherapy is a critical clinical priority. To identify baseline plasma and extracellular vesicle (EV) metabolomic biomarkers that predict progression-free survival (PFS) and overall survival (OS) in patients treated with first-line afatinib. A prospective, multicenter, single-arm observational biomarker discovery study (the EXTRA study). We enrolled 103 patients with advanced or recurrent EGFR-mutated NSCLC. Fasting plasma (n = 101 evaluable) and EV (n = 99 evaluable) samples collected at baseline were subjected to comprehensive metabolomic profiling (630 metabolites) using the MxP® Quant 500 kit (Biocrates Life Sciences AG, Innsbruck, Austria) combined with UHPLC-MS/MS. Durable response was defined as a progression-free survival (PFS) of 18 months or longer. Associations with PFS and OS were evaluated using false discovery rate (FDR)-adjusted univariable Cox models and validated via bootstrap resampling. In this first large-scale concurrent analysis, baseline EV metabolites unexpectedly showed no significant correlation with survival end points. Conversely, baseline plasma metabolites provided robust predictive signatures. Histidine and five lysophosphatidylcholines (LPC16:0, 17:0, 18:0, 18:1, 18:2) were identified as significant positive predictors of PFS. For OS, one phosphatidylcholine (PC ae C30:1) and four amino acids (homoarginine, histidine, threonine, tryptophan) were positive predictors, while one dihydroceramide (Cer [d18:1/24:1]) was a negative predictor. Prognostic scores based on these metabolites successfully stratified patients into three risk groups with significantly distinct survival outcomes (p < 0.001). Plasma-derived, but not EV-derived, metabolomic signatures accurately predict durable response and survival in EGFR-mutated NSCLC treated with afatinib. These novel biomarkers reflect host immunity and tumor microenvironment states, offering a step toward precision medicine by identifying patients likely to benefit from monotherapy over intensive combination regimens. University Hospital Medical Information Network Clinical Trials Registry (UMIN-CTR), https://center6.umin.ac.jp/cgi-open-bin/ctr/ctr_his_list.cgi?recptno=R000028688 (UMIN000024935).