Diagnosis of head and neck cancer by AI-based tumor-educated platelet RNA profiling of liquid biopsies

TL;DR

This study uses AI to analyze platelet RNA in blood samples to diagnose head and neck cancer with high accuracy.

Contribution

A novel AI-based method using tumor-educated platelet RNA for early diagnosis of head and neck squamous cell carcinoma is introduced.

Findings

A PSO-SVM model using 245 platelet transcripts achieved an AUC of 0.87 for HNSCC diagnosis.

A LASSO logistic regression model with 1,198 mRNAs reached a median AUC of 0.84, independent of HPV status.

TEP RNA profiling shows promise as a non-invasive diagnostic tool for head and neck cancer.

Abstract

Over 95% of head and neck cancers are squamous cell carcinoma (HNSCC). HNSCC is mostly diagnosed late, causing a poor prognosis despite the application of invasive treatment protocols. Tumor-educated platelets (TEPs) have been shown to hold promise as a molecular tool for early cancer diagnosis. We sequenced platelet mRNA isolated from blood of 101 patients with HNSCC and 101 propensity-score matched noncancer controls. Two independent machine learning classification strategies were employed using a training and validation approach to identify a cancer predictor: a particle swarm optimized support vector machine (PSO-SVM) and a least absolute shrinkage and selection operator (LASSO) logistic regression model. The best performing PSO-SVM predictor consisted of 245 platelet transcripts and reached a maximum area under the curve (AUC) of 0.87. For the LASSO-based prediction model, 1,198 mRNAs were selected, resulting in a median AUC of 0.84, independent of HPV status. Our data show that TEP RNA classification by different AI tools is promising in the diagnosis of HNSCC.

RNA expression profiling of tumor-educated platelets is highly accurate in detecting head and neck cancer, and could aid head and neck cancer diagnostics.

Figures