Tumor organoids on-a-chip and the role of AI in predictive oncology and personalized cancer medicine.

Abstract

The drug development process in cancer faces significant challenges due to high failure rates in translational studies despite promising<i>in vitro</i>results. Additionally, conventional animal models exhibit inherent limitations and ethical concerns, constraining their relevance to cancer studies. Recognizing the pivotal role of the tumor microenvironment (TME) on cancer development and treatment outcomes, recent advancements in 3D microfluidic devices and tumor-on-a-chip models enabled researchers to explore the TME with enhanced accuracy and reliability, yielding novel insights. Notably, the emergence of physiological tumor models, particularly 3D models such as organoids derived from human tissues, provides a more accurate representation of<i>in vivo</i>tumor features. Moreover, 3D tumor models hold promise for diverse applications, including high-throughput drug testing, disease modeling, and regenerative medicine. Meanwhile, combining artificial intelligence (AI) with patient-derived tumor organoids has become a key strategy in predictive oncology and personalized cancer treatment. Furthermore, incorporating quantitative systems pharmacology and physiologically based pharmacokinetic modeling, and pharmacokinetics/pharmacodynamics analysis with generative AI (Gen-AI) has revolutionized predictive oncology by enabling precise simulations of drug interactions and patient-specific responses, thereby enhancing the predictive accuracy of personalized cancer treatments. These advanced methodologies harness the power of AI algorithms to analyze intricate datasets derived from patient-specific tumor organoids. Moreover, the predictive modeling capabilities of Gen-AI facilitate the development of personalized treatment strategies customized for each patient, thereby revolutionizing oncology practice. This review explores the synergistic impact of tumor-on-a-chip models, organoids derived from patient tumors, and Gen-AI. Together, these technologies mark a significant advancement in precision medicine, offering promising opportunities to improve therapeutic effectiveness and treatment outcomes in cancer care.