Dataset Information

Biology-guided deep learning predicts prognosis and cancer immunotherapy response.

ABSTRACT: Substantial progress has been made in using deep learning for cancer detection and diagnosis in medical images. Yet, there is limited success on prediction of treatment response and outcomes, which has important implications for personalized treatment strategies. A significant hurdle for clinical translation of current data-driven deep learning models is lack of interpretability, often attributable to a disconnect from the underlying pathobiology. Here, we present a biology-guided deep learning approach that enables simultaneous prediction of the tumor immune and stromal microenvironment status as well as treatment outcomes from medical images. We validate the model for predicting prognosis of gastric cancer and the benefit from adjuvant chemotherapy in a multi-center international study. Further, the model predicts response to immune checkpoint inhibitors and complements clinically approved biomarkers. Importantly, our model identifies a subset of mismatch repair-deficient tumors that are non-responsive to immunotherapy and may inform the selection of patients for combination treatments.

SUBMITTER: Jiang Y

PROVIDER: S-EPMC10447467 | biostudies-literature | 2023 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Biology-guided deep learning predicts prognosis and cancer immunotherapy response.

Jiang Yuming Y Zhang Zhicheng Z Wang Wei W Huang Weicai W Chen Chuanli C Xi Sujuan S Ahmad M Usman MU Ren Yulan Y Sang Shengtian S Xie Jingjing J Wang Jen-Yeu JY Xiong Wenjun W Li Tuanjie T Han Zhen Z Yuan Qingyu Q Xu Yikai Y Xing Lei L Poultsides George A GA Li Guoxin G Li Ruijiang R

Nature communications 20230823 1

Substantial progress has been made in using deep learning for cancer detection and diagnosis in medical images. Yet, there is limited success on prediction of treatment response and outcomes, which has important implications for personalized treatment strategies. A significant hurdle for clinical translation of current data-driven deep learning models is lack of interpretability, often attributable to a disconnect from the underlying pathobiology. Here, we present a biology-guided deep learning ...[more]

PMID: 37612313

Similar Datasets

Project description:ObjectivesThis study aimed to investigate the differentiation state and clinical significance of colorectal cancer cells, as well as to predict the immune response and prognosis of patients based on differentiation-related genes of colorectal cancer.IntroductionColorectal cancer cells exhibit different differentiation states under the influence of the tumor microenvironment, which determines the cell fates.MethodsWe combined single-cell sequencing (scRNA-seq) data from The Cancer Genome Atlas source with extensive transcriptome data from the Gene Expression Omnibus database. We obtained colorectal cancer differentiation-related genes using cell trajectory analysis and developed a colorectal cancer differentiation-related gene based molecular typing and prognostic model to predict the immune response and prognosis of patients with colorectal cancer.ResultsWe identified 5 distinct cell differentiation subsets and 620 colorectal cancer differentiation-related genes. Colorectal cancer differentiation-related genes were significantly associated with metabolism, angiogenesis, and immunity. We separated patients into 3 subtypes based on colorectal cancer differentiation-related gene expression in the tumor and found differences among the different subtypes in immune infiltration status, immune checkpoint gene expression, clinicopathological features, and overall survival. Immunotherapeutic interventions involving a highly expressed immune checkpoint blockade may be selectively effective in the corresponding cancer subtypes. We built a risk score prediction model (5-year AUC: .729) consisting of the 4 most important predictors of survival (TIMP1, MMP1, LGALS4, and ITLN1). Finally, we generated and validated a nomogram consisting of the risk score and clinicopathological variables.ConclusionThis study highlights the significance of genes involved in cell differentiation for clinical prognosis and immunotherapy in patients and provides prospective therapeutic targets for colorectal cancer.

Project description:Programmed death-ligand 1 (PD-L1) IHC is the most commonly used biomarker for immunotherapy response. However, quantification of PD-L1 status in pathology slides is challenging. Neither manual quantification nor a computer-based mimicking of manual readouts is perfectly reproducible, and the predictive performance of both approaches regarding immunotherapy response is limited. In this study, we developed a deep learning (DL) method to predict PD-L1 status directly from raw IHC image data, without explicit intermediary steps such as cell detection or pigment quantification. We trained the weakly supervised model on PD-L1-stained slides from the non-small cell lung cancer (NSCLC)-Memorial Sloan Kettering (MSK) cohort (N = 233) and validated it on the pan-cancer-Vall d'Hebron Institute of Oncology (VHIO) cohort (N = 108). We also investigated the performance of the model to predict response to immune checkpoint inhibitors (ICI) in terms of progression-free survival. In the pan-cancer-VHIO cohort, the performance was compared with tumor proportion score (TPS) and combined positive score (CPS). The DL model showed good performance in predicting PD-L1 expression (TPS ≥ 1%) in both NSCLC-MSK and pan-cancer-VHIO cohort (AUC 0.88 ± 0.06 and 0.80 ± 0.03, respectively). The predicted PD-L1 status showed an improved association with response to ICIs [HR: 1.5 (95% confidence interval: 1-2.3), P = 0.049] compared with TPS [HR: 1.4 (0.96-2.2), P = 0.082] and CPS [HR: 1.2 (0.79-1.9), P = 0.386]. Notably, our explainability analysis showed that the model does not just look at the amount of brown pigment in the IHC slides, but also considers morphologic factors such as lymphocyte conglomerates. Overall, end-to-end weakly supervised DL shows potential for improving patient stratification for cancer immunotherapy by analyzing PD-L1 IHC, holistically integrating morphology and PD-L1 staining intensity.SignificanceThe weakly supervised DL model to predict PD-L1 status from raw IHC data, integrating tumor staining intensity and morphology, enables enhanced patient stratification in cancer immunotherapy compared with traditional pathologist assessment.

Dataset Information

Biology-guided deep learning predicts prognosis and cancer immunotherapy response.

Publications

Biology-guided deep learning predicts prognosis and cancer immunotherapy response.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets