Harnessing Machine Learning and AI-Driven Analytics to Identify Novel Drug Targets and Predict Chemotherapy Efficacy in NSCLC

Shaojia Qin ¹ Biyu Deng ¹ Dan Mo ¹ Zhengyou Zhang ² Xuan Wei ² Zhougui Ling ^2*

• ¹ Department of Pulmonary and Critical Care Medicine, Laibin People's Hospital, Laibin, China
• ² Department of Pulmonary and Critical Care Medicine, the Fourth Affiliated Hospital of Guangxi Medical University,, Liuzhou, China

Non-small cell lung cancer (NSCLC) constitutes the majority of lung cancer cases and displays marked heterogeneity in both clinical presentation and molecular profiles, leading to variable responses to chemotherapy. Emerging evidence suggests that mitochondria-derived RNAs (mtRNAs) may serve as novel biomarkers, although their role in predicting chemotherapy outcomes has yet to be fully explored. In this study, we combined traditional clinical factor modeling and artificial intelligence (AI) analyticsspecifically a BiomedGPT-based prediction module-to evaluate and enhance chemotherapy response stratification in NSCLC. Peripheral blood mononuclear cells were obtained from patients for mtRNA ratio analysis (mt_tRNA-Tyr-GTA_5_end to mt_tRNA-Phe-GAA), while thoracic CT images were processed to refine the AI-driven BiomedGPT variable. Individual clinical factors (Sex, Age, History_of_smoking, Pathological_type, Stage) offered limited predictive power when used in isolation.However, integrating these clinical variables into a random forest model improved sensitivity in the training set yet showed diminished generalizability in the validation cohort. Integrating the BiomedGPT score and mtRNA ratio significantly enhanced predictive performance across both training and validation datasets.Finally, an allinclusive model (clinical + AI + mtRNA) produced a risk score capable of discriminating patients into high-and low-risk groups for progression-free survival and overall survival with statistically significant differences. These findings highlight the potential of combining mtRNA biomarkers with advanced AI methods to refine therapeutic decisionmaking in NSCLC and underscore the importance of integrative approaches in precision oncology.