Sniffing Out Lung Cancer: Biomimetic Breath Analysis via a Deep Eutectic Solvent-Driven Colorimetric Sensor Array

Lung cancer remains the leading cause of cancer-related mortality worldwide, primarily due to the lack of accessible, noninvasive early diagnostic tools. Here, we present a novel paper-based colorimetric sensor array that leverages deep eutectic solvents (DES) and pH indicators to detect lung-cancer-specific volatile organic compounds (VOCs) in exhaled breath. The hydrophobic DES enhances sensor stability and selectivity by mitigating moisture interference, while amplifying VOC interactions through hydrogen bonding and proton exchange. Breath samples from 91 participants (46 lung cancer patients, 28 noncancer respiratory disease patients, and 17 healthy controls) were analyzed using a portable, sterilized collection system. The sensor array achieved 93% accuracy in discriminating cancer patients from noncancerous individuals via linear discriminant analysis (LDA), with 98% sensitivity and specificity at the optimal cutoff. Notably, the system also distinguished lung cancer from other respiratory diseases (87% accuracy), addressing a critical clinical challenge. Hierarchical cluster analysis (HCA) and ROC curves further validated the robustness of the platform. This cost-effective, rapid, and noninvasive approach demonstrates significant potential for large-scale lung cancer screening, offering a transformative alternative to conventional diagnostic methods.