Spatial analysis of tumor‐infiltrating lymphocytes in histological sections using deep learning techniques predicts survival in colorectal carcinoma

Abstract

This study aimed to explore the prognostic impact of spatial distribution of tumor‐infiltrating lymphocytes (TILs) quantified by deep learning (DL) approaches based on digitalized whole‐slide images stained with hematoxylin and eosin in patients with colorectal cancer (CRC). The prognostic impact of spatial distributions of TILs in patients with CRC was explored in the Yonsei cohort (n = 180) and validated in The Cancer Genome Atlas (TCGA) cohort (n = 268). Two experienced pathologists manually measured TILs at the most invasive margin (IM) as 0–3 by the Klintrup–Mäkinen (KM) grading method and this was compared to DL approaches. Inter‐rater agreement for TILs was measured using Cohen's kappa coefficient. On multivariate analysis of spatial TIL features derived by DL approaches and clinicopathological variables including tumor stage, microsatellite instability, and KRAS mutation, TIL densities within 200 μm of the IM (f_im200) remained the most significant prognostic factor for progression‐free survival (PFS) (hazard ratio [HR] 0.004 [95% confidence interval, CI, 0.0001–0.15], p = 0.0028) in the Yonsei cohort. On multivariate analysis using the TCGA dataset, f_im200 retained prognostic significance for PFS (HR 0.031 [95% CI 0.001–0.645], p = 0.024). Inter‐rater agreement of manual KM grading was insignificant in the Yonsei (κ = 0.109) and the TCGA (κ = 0.121) cohorts. The survival analysis based on KM grading showed statistically significant different PFS in the TCGA cohort, but not the Yonsei cohort. Automatic quantification of TILs at the IM based on DL approaches shows prognostic utility to predict PFS, and could provide robust and reproducible TIL density measurement in patients with CRC.

Integrative Spatial Analysis of H&E and IHC Images Identifies Prognostic Immune Subtypes Correlated with Progression-Free Survival in Human Papillomavirus (HPV)-Related Oropharyngeal Squamous Cell Carcinoma

“…Briefly, we first applied a DL-based algorithm to H&E images to identify the tumor, stroma, and TIL regions within the tumor. The accuracy of the segmentation process for identifying tumors, TILs, and stroma regions based on H&E images has been previously published [14].…”

“…Consequently, within each 0.5 mm 2 grid, there were approximately 17ൈ17 tiles (Figure 2A). The tiles were classified into three categories: TIL, tumors, and stroma using an in-house Resnet18-based pre-trained deep learning model [14].…”

“…Advancements in digital pathology and deep learning (DL) algorithms have paved the way for automated TIL quantification and analysis from pathology slides [12]. DL-based whole-slide hematoxylin and eosin (H&E) image analyses offer a cost-effective and readily applicable approach to quantifying TILs within tumors [12, 14]. H&E-based evaluation of TILs provides information on the presence of specific cell types within the TME but is limited in the evaluation of the functional roles and characteristics of these cells.…”

A pathomic approach for tumor-infiltrating lymphocytes classification on breast cancer digital pathology images

“…Sun et al [ 21 ] developed a deep learning-based tool for an automatic til score assessment and achieved a F1-score of 0.856 for nuclei (also including TILs) classification. Similar approaches were also developed for detecting TILs in other cancer types [ 11 , [49] , [50] , [51] ].…”

Cell spatial analysis in Crohn's disease: unveiling local cell arrangement pattern with graph-based signatures

“…4 They also developed a method to detect prognostic tumor infiltrating lymphocytes (TIL) density in colorectal carcinoma patients . 5 Saltz et al generated TIL maps from H&E images, correlating them with survival in diverse tumor types . 6 However, applying spatial analysis to inflammatory bowel disease (IBD) remains under investigation.…”