UNSTRUCTURED
Purpose: Currently, selection of patients for sequential vs. concurrent chemotherapy/radiation regimens lacks evidentiary support, and it is based on locally-optimal decisions for each step. We aim to optimize the multi-step treatment of head and neck cancer patients and to predict multiple patient survival and toxicity outcomes, and we develop, apply, and evaluate a first application of deep-Q-learning (DQL) and simulation to this problem.
Patients and methods: The treatment decision DQL digital twin and the patient’s digital twin were created, trained and evaluated on a dataset of 536 oropharyngeal squamous cell carcinoma (OPC) patients with the goal of, respectively, determining the optimal treatment decisions with respect to survival and toxicity metrics, and predicting the outcomes of the optimal treatment on the patient. The models were trained on a subset of 402 patients (split randomly) and evaluated on a separate set of 134 patients. Training and evaluation of the digital twin dyad was completed in August 2020. The dataset includes 3-step sequential treatment decisions and complete relevant history of the patients cohort treated at MD Anderson Cancer Center between 2005 and 2013, with radiomics analysis performed for the segmented primary tumor volumes.
Results: On the validation set, 87.09% mean and 90.85% median accuracy in treatment outcome prediction, matching the clinicians’ outcomes and improving (predicted) survival rate by +3.73% (95% CI: [-0.75%, +8.96%]), and dysphagia rate by +0.75% (CI: [-4.48%, +6.72%]) when following DQL treatment decisions.
Conclusion: Given the prediction accuracy and predicted improvement on medically relevant outcomes yielded by this approach, this digital twin dyad of the patient-physician dynamic treatment problem has the potential of aiding physicians in determining the optimal course of treatment and in assessing its outcomes.