Next-generation lung-cancer-on-a-chip: Toward personalized therapy, AI, and CRISPR-driven models

Lung-cancer-on-a-chip (LCOC) technologies have advanced rapidly, yet most models evaluate mechanical strain, patient-derived tumors, multi-organ interactions, artificial intelligence (AI) analytics, and clustered regularly interspaced short palindromic repeats (CRISPR) editing in isolation. In this review, we uniquely integrate these emerging components into a unified framework centered on the breathing LCOC. We highlight how embedding patient-derived lung tumor fragments into cyclically stretched microenvironments, then linking them to downstream organ compartments, enables patient-specific mapping of metastatic routes under physiologically relevant mechanics. We further describe how continuous high-resolution imaging from these platforms can feed AI pipelines for automated drug-response prediction and metastatic trajectory simulation, and how on-chip CRISPR editing enables accurate investigation of metastatic drivers within dynamic, strain-modulated microenvironments. By synthesizing these technologies, we outline a next-generation, personalized multi-organ-on-chip architecture capable of predicting individual disease progression without direct patient risk. We also address practical barriers, including tumor fragility under strain, imaging domain shift, and gene-editing delivery challenges, and how to overcome such barriers.