Deep Learning Model for Predicting Immunotherapy Response in Advanced Non-Small Cell Lung Cancer

Abstract

Importance: Only a small fraction of patients with advanced non-small cell lung cancer (NSCLC) respond to immune checkpoint inhibitor (ICI) treatment. For optimal personalized NSCLC care, it is imperative to identify patients who are most likely to benefit from immunotherapy.

Objective: To develop a supervised deep learning-based ICI response prediction method; evaluate its performance alongside other known predictive biomarkers; and assess its association with clinical outcomes in patients with advanced NSCLC.

Design, setting, and participants: This multicenter cohort study developed and independently validated a deep learning-based response stratification model for predicting ICI treatment outcome in patients with advanced NSCLC from whole slide hematoxylin and eosin-stained images. Images for model development and validation were obtained from 1 participating center in the US and 3 in the European Union (EU) from August 2014 to December 2022. Data analyses were performed from September 2022 to May 2024.

Exposure: Monotherapy with ICIs.

Main outcomes and measures: Model performance measured by clinical end points and objective response rate (ORR) differentiation power vs other predictive biomarkers, ie, programmed death-ligand 1 (PD-L1), tumor mutational burden (TMB), and tumor-infiltrating lymphocytes (TILs).

Results: A total of 295 581 image tiles from 958 patients (mean [SD] age, 66.0 [10.6] years; 456 [48%] females and 502 [52%] males) treated with ICI for NSCLC were included in the analysis. The US-based development cohort consisted of 614 patients with median (IQR) follow-up time of 54.5 (38.2-68.1) months, and the EU-based validation cohort, 344 patients with 43.3 (27.4-53.9) months of follow-up. The ORR to ICI was 26% in the developmental cohort and 28% in the validation cohort. The deep learning model's area under the receiver operating characteristic curve (AUC) for ORR was 0.75 (95% CI, 0.64-0.85) in the internal test set and 0.66 (95% CI, 0.60-0.72) in the validation cohort. In a multivariable analysis, the deep learning model's score was an independent predictor of ICI response in the validation cohort for both progression-free (hazard ratio, 0.56; 95% CI, 0.42-0.76; P < .001) and overall survival (hazard ratio, 0.53; 95% CI, 0.39-0.73; P < .001). The tuned deep learning model achieved a higher AUC than TMB, TILs, and PD-L1 in the internal set; in the validation cohort, it was superior to TILs and comparable with PD-L1 (AUC, 0.67; 95% CI, 0.60-0.74), with a 10-percentage point improvement in specificity. In the validation cohort, combining the deep learning model with PD-L1 scores achieved an AUC of 0.70 (95% CI, 0.63-0.76), outperforming either marker alone, with a response rate of 51% compared to 41% for PD-L1 (≥50%) alone.

Conclusions and relevance: The findings of this cohort study demonstrate a strong and independent deep learning-based feature associated with ICI response in patients with NSCLC across various cohorts. Clinical use of this deep learning model could refine treatment precision and better identify patients who are likely to benefit from ICI for treatment of advanced NSCLC.

Figure 1.. Deep Learning Model and Immune Checkpoint Inhibitors (ICIs) by Clinical Outcome

A and B, Progression-free survival and overall survival in response to ICI, stratified by the Deep-IO model scores in the validation cohort. C, Distribution of Deep-IO probability scores across ORR subgroups in the validation cohort using the Mann-Whitney U-test. CR indicates complete response; HR, hazard ratio; mOS, median OS in months; mPFS, median PFS in months; ORR, objective response rate; PD, progressive disease; PR, partial response; and SD, stable disease.

Figure 2.. Multivariable Analysis in the Validation Cohort

Cox proportional hazard models of significant independent predictive factors associated with progression-free survival and overall survival. Variables with P < .25 from univariate analyses were included in the Cox regression analysis. Number of events and C-index are indicated for each model. C-index indicates concordance index; ECOG, Eastern Cooperative Oncology Group performance status; ICI, immune checkpoint inhibitors; LUAD, lung adenocarcinoma; and LUSC, lung squamous cell carcinoma.

Figure 3.. Performance of Immune Checkpoint Inhibitors (ICIs) Biomarkers vs the Deep-IO Model

Analysis of the performance power of individual biomarkers (Deep-IO, PD-L1, TMB, TILs) and the combined Deep-IO and PD-L1 in differentiating ICI objective response rate binary groups in the A, test and B, validation cohorts. C, Proportion of ICI responses in PD-L1 and Deep-IO subgroups within the validation cohort, with Deep-IO subgroups classified into tertiles as low (lower tertile), medium (middle tertile), and high (upper tertile). PD-L1 subgroups are categorized as low (<1%), medium (1%-49%), and high (≥50%). D, Combination of Deep-IO scores and PD-L1 expression subgroups in relation to the ICI response rate within the validation cohort. The color intensity of the squares represents the response rate, with darker colors indicating a higher response rate and lighter colors indicating a lower response rate (shown as percentages). AUC indicates area under the receiver operating characteristic curve; HR, hazard ratio; ORR, objective response rate; PD-L1, programmed death-ligand 1; TILs, tumor-infiltrating lymphocytes; TMB, tumor mutational burden; and TPS, tumor proportion score.