Deep learning quantified mucus-tumor ratio predicting survival of patients with colorectal cancer using whole-slide images

Abstract

Background In colorectal cancer (CRC), mucinous adenocarcinoma differs from other adenocarcinomas in gene-phenotype, morphology, and prognosis. However, mucinous components are present in a large amount of adenocarcinoma, and the prognostic value of mucus proportion has not been investigated. Artificial intelligence provides a way to quantify mucus proportion on whole-slide images (WSIs) accurately. We aimed to quantify mucus proportion by deep learning and further investigate its prognostic value in two CRC patients cohorts. Methods Deep learning was used to segment WSIs stained with hematoxylin and eosin. Mucus-tumor ratio (MTR) was defined as the proportion of mucinous component in the tumor area. A training cohort (N = 419) and a validation cohort (N = 315) were used to evaluate the prognostic value of MTR. Survival analysis was performed by the Cox proportional hazard model. Result Patients were stratified to mucus-low and mucus-high groups by 24.1% as the threshold. In the training cohort, patients with mucus-high had unfavorable outcomes (hazard ratio for high vs. low 1.88, 95% confidence interval 1.18-2.99, P = 0.008), with 5-year overall survival rates of 54.8% and 73.7% in mucus-high and mucus-low groups, respectively. The results were confirmed in the validation cohort (2.09, 1.21-3.60, 0.008; 79.8% vs. 62.8%). The prognostic value of MTR was maintained in multivariate analysis for both cohorts. Conclusion The deep learning quantified MTR was an independent prognostic factor in CRC. With the advantages of advanced efficiency and high consistency, our method is suitable for clinical application and promotes precision medicine development.

Development and validation of a scoring system incorporating tumor growth pattern and perineural invasion for risk stratification in colorectal cancer

“…Second, the evaluation processes of TGP and PNI are time-consuming. With the development of artificial intelligence in CRC prognostic prediction, 31 an automated method to evaluate TGP and PNI is necessary to be developed in our future study. Additionally, the prognostic value of the tumor-invasion score needs prospective and widespread validations in the future.…”

Multi-gene models for survival prognosis of colorectal cancer patients

“…Using biomimetic analysis, signi cant changes in lncRNA were found in patients with mismatch repair CRC, and LncRNA AC123023.1 was found to promote CRC invasion and migration. Silencing of LncRNA AC123023.1 in cellular assays caused a signi cant decrease in the expression of JAG2, a ligand of the Notch pathway, and thus promoted CRC invasion and migration [31][32][33] . These ndings provide new ideas for studying the pathogenesis and prognosis of CRC.…”

Artificial Intelligence Applications in the Treatment of Colorectal Cancer: A Narrative Review

“…In the validation cohort, high TSR based on the tested CNN model was correlated with increased OS ( P < 0.004). In another study, 93 the CNN-quantified mucus proportion was also validated to be correlated with the prognosis of colorectal mucinous adenocarcinoma patients ( P < 0.008). Another innovative study 94 focused on classifying consensus molecular subtypes (CMSs) based on histopathology images of colorectal tumors.…”