Adaptive Cancer Therapy in the Age of Generative Artificial Intelligence

Area of Science:

• Oncology
• Artificial Intelligence
• Computational Biology

Background:

• Therapeutic resistance poses a significant challenge to developing effective cancer treatments.
• Adaptive cancer therapy, utilizing drug combinations with modulated timing and dosage, is a promising strategy to manage cancer's dynamic nature.
• Clinical success of adaptive therapy is hindered by the lack of real-time treatment response prediction capabilities.

Purpose of the Study:

• To explore the potential of generative artificial intelligence (AI) in predicting cancer treatment response.
• To propose an integrated model combining Generative Pre-Trained Transformers (GPTs) with adaptive cancer therapy.
• To address the critical need for real-time disease progression predictions in adaptive treatment protocols.

Main Methods:

• Conceptualizing an integration model of Generative Pre-Trained Transformers (GPTs) within a closed-loop system for adaptive cancer therapy.
• Leveraging clinical, molecular, and radiomics data for training AI prediction models.
• Discussing the challenges associated with implementing AI-driven adaptive treatment strategies.

Main Results:

• Generative AI, particularly GPTs, shows potential for learning complex prediction models of treatment response.
• The proposed model aims to predict disease progression trajectories by integrating GPTs with adaptive treatment regimens.
• Identified key challenges in realizing AI-driven adaptive therapy in clinical oncology.

Conclusions:

• Generative AI offers a promising avenue for overcoming the limitations of current adaptive cancer therapy approaches.
• Real-time prediction of treatment response using AI is crucial for the successful clinical implementation of adaptive therapy.
• Further research and development are needed to integrate AI effectively into oncology for personalized cancer treatment.