ObjectiveComplex phenotypes captured on histological slides represent the biological processes at play in individual cancers, but the link to underlying molecular classification has not been clarified or systematised. In colorectal cancer (CRC), histological grading is a poor predictor of disease progression, and consensus molecular subtypes (CMSs) cannot be distinguished without gene expression profiling. We hypothesise that image analysis is a cost-effective tool to associate complex features of tissue organisation with molecular and outcome data and to resolve unclassifiable or heterogeneous cases. In this study, we present an image-based approach to predict CRC CMS from standard H&E sections using deep learning.DesignTraining and evaluation of a neural network were performed using a total of n=1206 tissue sections with comprehensive multi-omic data from three independent datasets (training on FOCUS trial, n=278 patients; test on rectal cancer biopsies, GRAMPIAN cohort, n=144 patients; and The Cancer Genome Atlas (TCGA), n=430 patients). Ground truth CMS calls were ascertained by matching random forest and single sample predictions from CMS classifier.ResultsImage-based CMS (imCMS) accurately classified slides in unseen datasets from TCGA (n=431 slides, AUC)=0.84) and rectal cancer biopsies (n=265 slides, AUC=0.85). imCMS spatially resolved intratumoural heterogeneity and provided secondary calls correlating with bioinformatic prediction from molecular data. imCMS classified samples previously unclassifiable by RNA expression profiling, reproduced the expected correlations with genomic and epigenetic alterations and showed similar prognostic associations as transcriptomic CMS.ConclusionThis study shows that a prediction of RNA expression classifiers can be made from H&E images, opening the door to simple, cheap and reliable biological stratification within routine workflows.
This study confirmed that the majority of patients with MCRC prefer oral to i.v. therapy, although the OPdG regimen appears to be the most popular i.v. option. Capecitabine clearly represents an effective, well-tolerated oral alternative to i.v. 5-FU/LV.
Background:People with colorectal cancer have impaired quality of life (QoL). We investigated what factors were most highly associated with it.Methods:Four hundred and ninety-six people with colorectal cancer completed questionnaires about QoL, functioning, symptoms, co-morbidity, cognitions and personal and social factors. Disease, treatment and co-morbidity data were abstracted from case notes. Multiple linear regression identified modifiable and unmodifiable factors independently predictive of global quality of life (EORTC-QLQ-C30).Results:Of unmodifiable factors, female sex (P<0.001), more self-reported co-morbidities (P=0.006) and metastases at diagnosis (P=0.036) significantly predicted poorer QoL, but explained little of the variability in the model (R2=0.064). Adding modifiable factors, poorer role (P<0.001) and social functioning (P=0.003), fatigue (P=0.001), dyspnoea (P=0.001), anorexia (P<0.001), depression (P<0.001) and worse perceived consequences (P=0.013) improved the model fit considerably (R2=0.574). Omitting functioning subscales resulted in recent diagnosis (P=0.002), lower perceived personal control (P=0.020) and travel difficulties (P<0.001) becoming significant predictors.Conclusion:Most factors affecting QoL are modifiable, especially symptoms (fatigue, anorexia, dyspnoea) and depression. Beliefs about illness are also important. Unmodifiable factors, including metastatic (or unstaged) disease at diagnosis, have less impact. There appears to be potential for interventions to improve QoL in patients with colorectal cancer.
Many patients would be willing to have GPs share their cancer follow-up with the caveat that they had received extra training and were appropriately supported by secondary care specialists. Patients attending shared care clinics appreciated a local service and longer appointment times. GPs stress the importance of maintaining their own clinical skills and reliable clinical and administrative support from secondary care.
Levels of anxiety caseness were similar to those of non-clinical samples, but depression caseness was higher, particularly in those who had received neo-adjuvant radiotherapy. Most factors associated with possible or probable depression may be modified with appropriate intervention.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.