Predicting response to immunotherapy in gastric cancer via multi-dimensional analyses of the tumour immune microenvironment

Abstract

A single biomarker is not adequate to identify patients with gastric cancer (GC) who have the potential to benefit from anti-PD-1/PD-L1 therapy, presumably owing to the complexity of the tumour microenvironment. The predictive value of tumour-infiltrating immune cells (TIICs) has not been definitively established with regard to their density and spatial organisation. Here, multiplex immunohistochemistry is used to quantify in situ biomarkers at sub-cellular resolution in 80 patients with GC. To predict the response to immunotherapy, we establish a multi-dimensional TIIC signature by considering the density of CD4⁺FoxP3^-PD-L1⁺, CD8⁺PD-1^-LAG3^-, and CD68⁺STING⁺ cells and the spatial organisation of CD8⁺PD-1⁺LAG3^- T cells. The TIIC signature enables prediction of the response of patients with GC to anti-PD-1/PD-L1 immunotherapy and patient survival. Our findings demonstrate that a multi-dimensional TIIC signature may be relevant for the selection of patients who could benefit the most from anti-PD-1/PD-L1 immunotherapy.

Fig. 1. Identification and characterisation of tumour-infiltrating immune cells in gastric cancer tissues.

a Schematic representation of the experimental design and analytical methods used in this study. b Selection of the regions of interest (ROIs) in representative images of haematoxylin and eosin (H&E)-stained formalin-fixed paraffin-embedded tissues. TC, tumour core; IM, invasion margin; N, normal tissue. Scale bar: 3 mm. c–f Representative composite and single-stained images of the multiplex immunohistochemistry panels used. Scale bar: 200 µm. g Overview of the automated image analysis pipeline.

Fig. 2. Automated image analysis highlights the ordered immune composition in gastric cancer.

a Constitution of the main tumour-infiltrating immune cell (TIIC) populations. Kruskal–Wallis test with the Dunn’s multiple comparison test. b Density of TIICs across the regions of interest (n = 80). TC, tumour core; IM, invasion margin; N, normal tissue. Immunofluorescence staining images refer to the co-expression of the corresponding markers and DAPI (nuclei). Scale bar: 20 µm. Box and whiskers represent mean ± 10–90 percentile. Kruskal–Wallis test with Dunn’s multiple comparison test. c TIIC density grouped by subtypes. d Overall survival of 80 patients based on the density of TIICs. The individual TIICs were divided into high (>two-thirds of the patients; blue line) or low density (≤two-thirds of patients; red line). Log-rank (Mantel–Cox) test was used. A two-sided P  <  0.05 was considered statistically significant.

Fig. 3. Spatial analysis of gastric cancer shows a hierarchy of organisation of tumour and immune cells.

a Illustration of the distance analysis involving immune and tumour cells. Red dots: tumour cells; green dots: immune cells. The white translucent circle represents the radius. Effective score = number of paired immune cells and tumour cells/number of immune cells. Scale bar: 100 µm. b The distribution of the effective score of tumour-infiltrating immune cell (TIIC) populations in the tumour core in 10-, 20- and 30 µm increments (n = 80). Error bars represent mean ± SEM. c Effective score of TIICs in patients grouped by gastric cancer subtypes. EBV, Epstein–Barr virus status; MMR, DNA mismatch repair; CPS, combined positive score. d Overall survival of the 80 patients based on the effective densities (0–10 µm and 0–20 µm) of TIICs. The individual immune infiltrate values were divided into high (> two-thirds of the patients in the cohort; blue line) or low density (≤ two-thirds of patients in the cohort; red line). Statistical relevance was defined using the log-rank (Mantel–Cox) test. A two-sided P  <  0.05 was considered statistically significant.

Fig. 4. The TIIC signature predicts the response to anti-PD-1/PD-L1-based immunotherapy.

a Definition of the tumour-infiltrating immune cell (TIIC) signature. Red arrows highlight specific immune cells. b Average area under the curve (AUC) of TIIC signature and combined TIIC signature (TIIC+ Epstein–Barr virus status + mismatch repair status + PD-L1 combined positive score) in the four machine learning models in the validation cohort. c Representative receiver operating characteristic (ROC) curves for the performance of the identified TIIC signature and combined TIIC signature in gastric cancer patients subjected to immunotherapy in the validation cohort. ETC extra tree classifier, GBC gradient boosting classifier, ABC AdaBoost classifier, MLP multi-layer perceptron.

Fig. 5. Feature importance of the TIIC signature and predictive value of the TIIC signature in immune-related survival.

a, b The feature importance of each marker in the prediction model. c, d Kaplan–Meier curves of the (c) immune-related progression-free survival (irPFS) and (d) immune-related overall survival (irOS) of anti-PD-1/PD-L1-treated patients stratified by the tumour-infiltrating immune cell (TIIC) signature in the validation cohort. Log-rank (Mantel–Cox) test was used for analysis.