Prediction accuracy of biomarkers for response to immune checkpoint inhibitors in advanced non-small cell lung cancer: A systematic review and meta-analysis.

person Wei Zhuang State Key Laboratory of Molecular Oncology, Department of Medical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China info_outline Wei Zhuang, Jie Zhao, Boyang Sun, Hua Bai, Zhijie Wang, Jia Zhong, Rui Wan, Lihui Liu, Jianchun Duan, Jie Wang

Authors person Wei Zhuang State Key Laboratory of Molecular Oncology, Department of Medical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China info_outline Wei Zhuang, Jie Zhao, Boyang Sun, Hua Bai, Zhijie Wang, Jia Zhong, Rui Wan, Lihui Liu, Jianchun Duan, Jie Wang Organizations State Key Laboratory of Molecular Oncology, Department of Medical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China, Department of Medical Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China Abstract Disclosures Research Funding National key R&D program of China NSFC general program, NSFC special program, CAMS Innovation Fund for Medical Sciences, CAMS Key Laboratory of Translational Research on Lung Cancer, Ministry of Education Innovation Team development project, Aiyou foundation Background: To compare the predictive accuracy of PD-L1 IHC, tissue or blood tumor mutation burden (tTMB, bTMB), gene expression profiling (GEP), driver gene mutation, and combined marker for immunotherapy response of advanced non-small cell lung cancer (NSCLC). Methods: Systematic searches of PubMed, Cochrane CENTRAL, and Embase were conducted from inception to July 6, 2023. Clinical trials involved with predictive biomarkers exploration for immune checkpoint inhibitors (ICIs) in advanced NSCLC were eligible. The response according to the biomarkers were extracted. The primary outcome was the area under the curve (AUC) of the summary receiver operating characteristic (SROC). Sensitivity, specificity, likelihood ratios and predictive values were evaluated. The subgroup analysis of different PD-L1 cut-off, TMB test and driver gene mutation were performed. Results: 36 articles of 59 biomarker tests were included. The AUC of combined marker (0.75; 95%CI, 0.71-0.78) was higher than PD-L1 (0.64; 95%CI, 0.60-0.68), tTMB (0.64; 95%CI, 0.60-0.68), bTMB (0.68; 95%CI, 0.64-0.72), GEP (0.67; 95%CI, 0.63-0.71), and driver gene mutation (0.51; 95%CI, 0.47-0.55). Combined marker also had higher specificity, positive likelihood ratio and positive predictive value than single biomarkers (Table). In subgroup analysis, PD-L1 with cut-off 50% had higher AUC (0.67; 95%CI, 0.63-0.71) than cut-off 1% (0.62; 95%CI, 0.58-0.66); EGFR mutation had higher AUC (0.69; 95%CI, 0.65-0.73) than KRAS (0.53; 95%CI, 0.49-0.58); tTMB measured by Whole Exosome Sequencing (WES) had approximate AUC to that measure by targeted gene panel. The predictive accuracy of the biomarkers was further compared for patients receiving ICI-based combination therapy. The AUC of combined marker (0.70; 95%CI, 0.66-0.74) remained the highest for combination therapy. Conclusions: Combined marker (PD-L1+tTMB/bTMB/GEP) has superior predictive accuracy than single biomarkers for immunotherapy response of NSCLC. bTMB is a promising liquid biopsy biomarker. The prescription of ICI for patients with EGFR mutation should be cautious. Further investigation is warranted to precisely identify biomarkers in various clinical settings. Biomarkers Specificity (95% CI) Positive Likelihood Ratio (95% CI) Median Positive Predictive Value (Q1, Q3) PD-L1 IHC 0.70 (0.63 to 0.76) 1.70 (1.40 to 1.90) 0.50 (0.37, 0.68) tTMB 0.64 (0.59 to 0.69) 1.50 (1.30 to 1.80) 0.48 (0.42, 0.72) bTMB 0.72 (0.64 to 0.78) 1.30 (0.70 to 2.50) 0.49 (0.24, 0.70) GEP 0.66 (0.59 to 0.73) 1.70 (1.30 to 2.20) 0.52 (0.39, 0.75) Driver gene 0.42 (0.27 to 0.59) 1.10 (0.90 to 1.20) 0.20 (0.12, 0.40) Combined marker 0.86 (0.79 to 0.92) 2.50 (1.60 to 3.90) 0.68 (0.42, 0.79)