Decision model for durable clinical benefit from front- or late-line immunotherapy alone or with chemotherapy in non-small cell lung cancer.

BACKGROUND: Predictive biomarkers and models of immune checkpoint inhibitors (ICIs) have been extensively studied in non-small cell lung cancer (NSCLC). However, evidence for many biomarkers remains inconclusive, and the opaqueness of machine learning models hinders practicality. We aimed to provide compelling evidence for biomarkers and develop a transparent decision tree model. METHODS: We consolidated data from 3,288 ICI-treated patients with NSCLC across real-world multicenter, public cohorts and the Choice-01 trial (ClinicalTrials.gov: NCT03856411). Over 50 features were examined for predicting durable clinical benefits (DCBs) from ICIs. Noteworthy biomarkers were identified to establish a decision tree model. Additionally, we explored the tumor microenvironment and peripheral CD8+ programmed death-1 (PD-1)+ T cell receptor (TCR) profiles. FINDINGS: Multivariate logistic regression analysis identified tumor histology, PD-ligand 1 (PD-L1) expression, tumor mutational burden, line, and regimen of ICI treatment as significant factors. Mutation subtypes of EGFR, KRAS, KEAP1, STK11, and disruptive TP53 mutations were associated with DCB. The decision tree (DT10) model, using the ten clinicopathological and genomic markers, showed superior performance in predicting DCB in the training set (area under the curve [AUC] = 0.82) and consistently outperformed other models in test sets. DT10-predicted-DCB patients manifested longer survival, an enriched inflamed tumor immune phenotype (67%), and higher peripheral TCR diversity, whereas the DT10-predicted-NDB (non-durable benefit) group showed an enriched desert immune phenotype (86%) and higher peripheral TCR clonality. CONCLUSIONS: The model effectively predicted DCB after front-/subsequent-line ICI treatment, with or without chemotherapy, for squamous and non-squamous lung cancer, offering clinicians valuable insights into efficacy prediction using cost-effective variables. FUNDING: This study was supported by the National Key R&D Program of China.

Assessment of human leukocyte antigen-based neoantigen presentation to determine pan-cancer response to immunotherapy.

Despite the central role of human leukocyte antigen class I (HLA-I) in tumor neoantigen presentation, quantitative determination of presentation capacity remains elusive. Based on a pooled pan-cancer genomic dataset of 885 patients treated with immune checkpoint inhibitors (ICIs), we developed a score integrating the binding affinity of neoantigens to HLA-I, as well as HLA-I allele divergence, termed the HLA tumor-Antigen Presentation Score (HAPS). Patients with a high HAPS were more likely to experience survival benefit following ICI treatment. Analysis of the tumor microenvironment indicated that the antigen presentation pathway was enriched in patients with a high HAPS. Finally, we built a neural network incorporating factors associated with neoantigen production, presentation, and recognition, which exhibited potential for differentiating cancer patients likely to benefit from ICIs. Our findings highlight the clinical utility of evaluating HLA-I tumor antigen presentation capacity and describe how ICI response may depend on HLA-mediated immunity.

Assessment of Blood Tumor Mutational Burden as a Potential Biomarker for Immunotherapy in Patients With Non-Small Cell Lung Cancer With Use of a Next-Generation Sequencing Cancer Gene Panel.

IMPORTANCE: Tumor mutational burden (TMB), as measured by whole-exome sequencing (WES) or a cancer gene panel (CGP), is associated with immunotherapy responses. However, whether TMB estimated by circulating tumor DNA in blood (bTMB) is associated with clinical outcomes of immunotherapy remains to be explored. OBJECTIVES: To explore the optimal gene panel size and algorithm to design a CGP for TMB estimation, evaluate the panel reliability, and further validate the feasibility of bTMB as a clinical actionable biomarker for immunotherapy. DESIGN, SETTING, AND PARTICIPANTS: In this cohort study, a CGP named NCC-GP150 was designed and virtually validated using The Cancer Genome Atlas database. The correlation between bTMB estimated by NCC-GP150 and tissue TMB (tTMB) measured by WES was evaluated in matched blood and tissue samples from 48 patients with advanced NSCLC. An independent cohort of 50 patients with advanced NSCLC was used to identify the utility of bTMB estimated by NCC-GP150 in distinguishing patients who would benefit from anti-programmed cell death 1 (anti-PD-1) and anti-programmed cell death ligand 1 (anti-PD-L1) therapy. The study was performed from July 19, 2016, to April 20, 2018. MAIN OUTCOMES AND MEASURES: Assessment of the Spearman correlation coefficient between bTMB estimated by NCC-GP150 and tTMB calculated by WES. Evaluation of the association of bTMB level with progression-free survival and response to anti-PD-1 and anti-PD-L1 therapy. RESULTS: This study used 2 independent cohorts of patients with NSCLC (cohort 1: 48 patients; mean [SD] age, 60 [13] years; 15 [31.2%] female; cohort 2: 50 patients; mean [SD] age, 58 [8] years; 15 [30.0%] female). A CGP, including 150 genes, demonstrated stable correlations with WES for TMB estimation (median r2 = 0.91; interquartile range, 0.89-0.92), especially when synonymous mutations were included (median r2 = 0.92; interquartile range, 0.91-0.93), whereas TMB estimated by the NCC-GP150 panel found higher correlations with TMB estimated by WES than most of the randomly sampled 150-gene panels. Blood TMB estimated by NCC-GP150 correlated well with the matched tTMB calculated by WES (Spearman correlation = 0.62). In the anti-PD-1 and anti-PD-L1 treatment cohort, a bTMB of 6 or higher was associated with superior progression-free survival (hazard ratio, 0.39; 95% CI, 0.18-0.84; log-rank P = .01) and objective response rates (bTMB ≥6: 39.3%; 95% CI, 23.9%-56.5%; bTMB <6: 9.1%; 95% CI, 1.6%-25.9%; P = .02). CONCLUSIONS AND RELEVANCE: The findings suggest that established NCC-GP150 with an optimized gene panel size and algorithm is feasible for bTMB estimation, which may serve as a potential biomarker of clinical benefit in patients with NSCLC treated with anti-PD-1 and anti-PD-L1 agents.