Dataset Information

Deep learning from HE slides predicts the clinical benefit from adjuvant chemotherapy in hormone receptor-positive breast cancer patients.

ABSTRACT: We hypothesized that a deep-learning algorithm using HE images might be capable of predicting the benefits of adjuvant chemotherapy in cancer patients. HE slides were retrospectively collected from 1343 de-identified breast cancer patients at the Samsung Medical Center and used to develop the Lunit SCOPE algorithm. Lunit SCOPE was trained to predict the recurrence using the 21-gene assay (Oncotype DX) and histological parameters. The risk prediction model predicted the Oncotype DX score > 25 and the recurrence survival of the prognosis validation cohort and TCGA cohorts. The most important predictive variable was the mitotic cells in the cancer epithelium. Of the 363 patients who did not receive adjuvant therapy, 104 predicted high risk had a significantly lower survival rate. The top-300 genes highly correlated with the predicted risk were enriched for cell cycle, nuclear division, and cell division. From the Oncotype DX genes, the predicted risk was positively correlated with proliferation-associated genes and negatively correlated with prognostic genes from the estrogen category. An integrative analysis using Lunit SCOPE predicted the risk of cancer recurrence and the early-stage hormone receptor-positive breast cancer patients who would benefit from adjuvant chemotherapy.

SUBMITTER: Cho SY

PROVIDER: S-EPMC8405682 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Contemporary treatment of node-positive (N+) colon cancer consists of adjuvant chemotherapy; however, randomized data supporting this practice were derived from lesions T2 or greater. Minimal data exist regarding the use and need for adjuvant chemotherapy in T1N+ disease.The aim of this study was to determine treatment trends and the effects of adjuvant chemotherapy on T1N+ colon cancers by using the National Cancer Database.This was a retrospective study. Baseline demographics, tumor, and cancer treatment characteristics were compared. Groups were matched on the propensity to receive chemotherapy. Adjusted long-term survival stratified by chemotherapy use was compared by using the Kaplan-Meier method with the log-rank test. Predictors of not receiving chemotherapy were identified by using a multivariable logistic regression model.Data were collected from the National Cancer Database, which collects cancer data from over 1500 cancer centers.We identified patients from 1998 to 2006 with T1N+ disease, excluding those with metastatic disease or previous cancer. Patients were stratified based on whether or not they received chemotherapy.The primary outcome measure of this study was long-term survival.Three thousand one hundred thirty-seven patients had T1N+ disease; 70.6% (n = 2216) received chemotherapy, and utilization significantly increased from 1998 to 2011 (p < 0.001). Unadjusted analysis revealed that patients treated with chemotherapy were statistically younger and healthier, and had shorter postoperative lengths of stay (all p < 0.001). Unadjusted 5-year survival was higher in patients receiving chemotherapy (87.9% vs 63.0% in patients with no chemotherapy; p < 0.001) and this persisted after propensity matching with (83.4% and 63.0% in patients with or without chemotherapy; p < 0.001). Only age (OR, 0.29; p < 0.001) predicted not receiving chemotherapy.Limitations include potential selection bias as well as the inability to compare disease-free survival/recurrence.Adjuvant chemotherapy appears to significantly improve long-term survival in patients receiving chemotherapy in T1N+ disease. Thus, the use of chemotherapy in T1N+ disease is justified and provides a highly significant survival benefit.

Dataset Information

Deep learning from HE slides predicts the clinical benefit from adjuvant chemotherapy in hormone receptor-positive breast cancer patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets