Background: Tumor-educated platelets (TEPs) represent a promising biosource for non-invasive multi-cancer early detection (MCED). While machine learning (ML) has been applied to TEP data, the integration of explainability to reveal gene-level contributions and regulatory associations remains underutilized. This study aims to develop an interpretable ML framework for cancer detection using platelet RNA-sequencing data, combining predictive performance with biological insight. Methods: This study analyzed 2018 TEP RNA samples from 18 tumor types using seven machine learning classifiers. SHAP (Shapley Additive Explanations) was applied for model interpretability, including global feature ranking, local explanation, and gene-level dependence patterns. A weighted SHAP consensus was built by combining model-specific contributions scaled by Area Under the Receiver Operating Characteristic Curve (AUC). Regulatory insights were supported through network analysis using GeneMANIA. Results: Neural models, including shallow Neural Network (NN) and Deep Neural Network (DNN) achieved the best performance (AUC ~0.93), with Extreme Gradient Boosting (XGB) and Support Vector Machine (SVM) also performing well. Early-stage cancers were predicted with high accuracy. SHAP analysis revealed consistent top features (e.g., SLC38A2, DHCR7, IFITM3), while dependence plots uncovered conditional gene interactions involving USF3 (KIAA2018), ARL2, and DSTN. Multi-hop pathway tracing identified NFYC as a shared transcriptional hub across multiple modulators. Conclusions: The integration of interpretable ML with platelet RNA data revealed robust biomarkers and context-dependent regulatory patterns relevant to early cancer detection. The proposed framework supports the potential of TEPs as a non-invasive, information-rich medium for early cancer screening.

Cancer is a prominent global cause of mortality, primarily due to delayed detection leading to limited treatment options. Current screening methods are mostly invasive and involve complex lengthy processes with high costs. Moreover, each screening typically focuses on a single type of cancer. This imposes a growing need for innovative, precise, and minimally invasive methods for early cancer detection. With the current advances in assay technologies and data science, multi-cancer early detection (MCED) tests are gaining increased interest in the research community as they offer potential for earlier diagnosis and improved patient outcomes. Different approaches are followed for MCED, and multiple machine learning methods are considered. In this paper, we systematically explore various MCED studies and their applied machine learning (ML) models for different types of biomarker data. We discuss the strengths and limitations of different study designs and compare their performance. Future directions are proposed, emphasizing the importance of integrating multi-omics data, enhancing model transparency, and fostering collaborative efforts to develop robust, cost effective and clinically applicable MCED tools.