Decoding small cell lung cancer vulnerabilities: the AI pathway to next-gen therapeutics

Small cell lung cancer (SCLC) is aggressive, treatment-sensitive, and prone to relapse. This Review evaluates how generative AI, graph neural networks, molecular language and diffusion models, structure-based design, and multimodal learning can exploit SCLC dependencies for small-molecule discovery. It integrates biomarker-guided patient stratification, resistance, combination therapy, medicinal chemistry, pharmacology, safety, manufacturing, and clinical translation, emphasizing benchmarked, uncertainty-aware, prospectively validated design–build–test–learn workflows.