Integrative biomarker and drug target discovery in osteosarcoma: traditional experimental approaches and AI-enabled insights

Osteosarcoma is the most common primary malignant bone tumor and remains a major clinical challenge due to frequent metastasis, chemoresistance, and pronounced molecular heterogeneity. Despite substantial advances in understanding disease biology, clinically actionable biomarkers and therapeutic targets that can reliably support precision treatment decisions remain limited. Traditional experimental approaches have yielded important mechanistic insights into osteosarcoma pathogenesis, but their hypothesis-driven nature and limited scalability constrain the ability to capture complex regulatory interactions. Recent progress in high-throughput sequencing and multi-omics profiling, together with advances in artificial intelligence (AI), has enabled more systematic interrogation of high-dimensional molecular landscapes. By integrating heterogeneous datasets, AI-based analytical frameworks can identify composite biomarker patterns, regulatory hubs, and candidate druggable vulnerabilities that better reflect tumor complexity and treatment heterogeneity. In parallel, computational strategies for drug sensitivity prediction and drug repurposing are emerging as complementary tools for accelerating therapeutic hypothesis generation and prioritizing candidate interventions in osteosarcoma. In this mini-review, we summarize recent progress in biomarker discovery and therapeutic target identification, with an emphasis on how traditional experimental evidence and AI-driven analyses function as complementary components within an integrated discovery-to-validation framework. We discuss key challenges in translational validation and highlight future directions for integrating data-driven discovery with pharmacological and clinical research to advance precision therapy for osteosarcoma.