RNA profiling of blood platelets noninvasively differentiates colorectal cancer from healthy donors and noncancerous intestinal diseases: a retrospective cohort study

Abstract

BackgroundThe RNA profiles of tumor-educated platelets (TEPs) possess pathological features that could be used for early cancer detection. However, the utility of TEP RNA profiling in detecting early colorectal cancer (CRC) versus noncancerous colorectal diseases has not yet been investigated. This study assesses the diagnostic capacity of TEP RNA profiles in a cohort of patients with CRC and noncancerous diseases.MethodsTranscriptome sequencing for platelets isolated from 132 patients with CRC at early and late stages and 190 controls consisting of healthy donors and patients with ulcerative disease, Crohn’s disease, polyps, and adenomas was performed and analyzed using binary particle swarm optimization coupled with support vector machine to identify genes that contributed to the classification of CRC patients versus controls. The area under the receiver operating curves (AUROCs) and the accuracy of TEP RNA profiles in CRC diagnosis were assessed.ResultsTEP RNA profiling achieved high performance in distinguishing and staging CRC patients from the controls. Using the swarm intelligence algorithm, the 921 most contributive genes that classified CRC patients from the controls were identified. AUROCs of 0.928 for the training set via leave-one-out cross-validation and 0.92 for the validation set were achieved, both of which were significantly higher than the clinically utilized serum biomarkers: carcinoembryonic antigen and cancer antigen 19-9. Notably, an AUROC of 0.915 in an external validation set was achieved. For predicting different CRC stages, an AUROC of 0.984 was achieved in the training set and 1.000 in the internal validation set.ConclusionsRNA profiles of TEPs are of potential diagnostic value for identifying early CRC from noncancerous diseases. Prospective studies are needed to validate its clinical relevance.