Foundational advancement toward development and deployment of digital twins for neoadjuvant therapy of triple negative breast cancer
Please login to view abstract download link
We aim to establish practical cancer digital twins to enable optimization of triple-negative breast cancer (TNBC) response to neoadjuvant chemotherapy (NAC). This study employed a cohort of 105 TNBC patients in the ARTEMIS trial, who received four cycles of Adriamycin/Cytoxan (A/C) every 2-3 weeks followed by 12 cycles of weekly Taxol (T). Longitudinal multiparametric MRI were acquired before, during, and after A/C to evaluate tumor morphology, perfusion, and cellularity. A biology-based mathematical model was calibrated to the MRI data collected, creating personalized digital twins. The digital twins accurately predicted patient-specific response to NAC, with an AUC of 0.89 identifying pathological complete response (pCR) status. We then used each patient’s digital twin to theoretically optimize outcome by identifying their optimal A/C-T schedule from 128 options. The patient-specifically optimized treatment yielded a significant improvement in pCR rate of 20.95–24.76%. Retrospective validation was conducted by virtually treating the twins with AC-T schedules from historical trials and obtaining identical observations on outcomes: bi-weekly A/C-T outperforms tri-weekly A/C-T, and weekly/bi-weekly T outperforms tri-weekly T. This proof-of-principle study demonstrates that our digital twins can provide a practical methodology to support the design of prospective trials that address the unmet need of patient-specific treatment tailoring for TNBC. More recently, we designed an interactive interface for clinicians to conveniently review digital twin outputs, including summarized metrics of predicted patient-specific responses to three different clinically feasible treatment options (i.e., default, escalation, de-escalation) and the visualization of associated tumor dynamics, so using them to support decision making. Efforts on structured interviews to collect clinician feedback on interface design and output feasibility, as well as clinically compliable implementation are ongoing. This multi-year project has been making fundamental progress to enable clinically practical implementation and validation of digital twins for personalizing breast cancer care.
