Abstract 36P
Background
Metastatic colorectal cancer (mCRC) poses formidable challenges in therapeutic decision-making due to its complex nature, characterized by diverse genetic and molecular profiles, patient-specific factors, and variable treatment responses. Traditional approaches to treatment selection often lack the nuance needed to address the individualized needs of patients. However, machine learning (ML) presents a promising avenue for transforming personalized treatment strategies in mCRC by leveraging advanced algorithms to analyze and interpret data patterns. The primary objective of this research is to assess the effectiveness of ML algorithms in facilitating clinical decision-making for mCRC treatment, ushering in new dimensions of precision medicine.
Methods
We conducted a study on 29 mCRC patients, collecting comprehensive data including demographics, genetic profiles, treatment responses, and outcomes. ML algorithms, particularly Multiple Kernel Learning, were employed to identify optimal treatment strategies based on individual patient profiles.
Results
The mean progression-free survival (mPFS) was 8.87 months, with shorter durations observed in patients with lung metastases (6.25 months). The CAPOX regimen showed the best results, with an mPFS of 9.25 months. Objective response rates (ORR) were reported in 65.5% of cases, and the clinical benefit rate was 82.8%. These findings suggest that AI algorithms may have a significant impact on both PFS and ORR.
Conclusions
Through the integration of artificial intelligence (AI) and machine learning (ML) algorithms in clinical practice, this research aims to advance the evolution of precision medicine in metastatic colorectal cancer (mCRC). The potential of ML algorithms to facilitate tailored treatment strategies for mCRC holds promise for enhancing the efficacy of the clinical decision-making process, optimizing therapeutic approaches, and maximizing treatment effectiveness. However, further validation of these findings is essential to solidify their role in enhancing precision medicine approaches for mCRC for the most beneficial and personalized care possible.
Clinical trial identification
NCT05396807.
Legal entity responsible for the study
Regional Institute of Oncology Iași, Romania.
Funding
European Project Horizon 2020 SC1-BHC-02-2019 [REVERT, ID:848098].
Disclosure
All authors have declared no conflicts of interest.