Generalized Evolutionary Classifier for Evolutionary Guided Precision Medicine

Area of Science:

• Oncology
• Computational Biology
• Translational Medicine

Background:

• Current precision medicine (CPM) faces challenges with treatment resistance due to evolving tumor heterogeneity.
• Dynamic precision medicine (DPM), an evolutionary guided precision medicine (EGPM) approach, addresses this by tracking genetic diversity and adapting treatments.
• DPM aims to proactively manage resistance and improve long-term survival through frequent adaptations and combination therapies.

Purpose of the Study:

• To evaluate the clinical utility of a shortened DPM treatment window, specifically two 6-week adaptations.
• To determine if limited DPM interventions can yield substantial benefits comparable to more extensive treatment sequences.
• To assess the cost-effectiveness and feasibility of frequent monitoring in DPM.

Main Methods:

• Simulated nearly 3 million virtual patients with diverse genetic profiles and treatment responses.
• Compared treatment outcomes for standard CPM, full DPM (40 moves), and a limited DPM protocol (2 moves).
• Analyzed DPM input parameters including initial subclone composition, drug sensitivities, and kinetic properties.

Main Results:

• The initial two DPM adaptations provided comparable average survival benefits to a 40-adaptation sequence across the virtual patient cohort.
• Patients showed no benefit from DPM if the initial two treatment recommendations matched CPM (65% negative predictive value).
• A subset of 20% of patients experienced extraordinary benefits from 2-move DPM, similar to 40-move DPM (hazard ratio 0.03).

Conclusions:

• The first two DPM adaptations capture the majority of the clinical benefit, potentially reducing the need for prolonged, intensive monitoring.
• An evolutionary classifier can identify patients who will benefit most from DPM, particularly when DPM and CPM recommendations diverge early.
• These findings support the clinical feasibility of DPM and other EGPM strategies, moving them closer to patient testing.