Derek S. Park

Supplementary Information File

Supplementary Information File

Supplementary Information File

Supplementary Information File

Standard-of-care treatment regimens have long been designed for maximal cell killing, yet these strategies often fail when applied to metastatic cancers due to the emergence of drug resistance. Adaptive treatment strategies have been developed as an alternative approach, dynamically adjusting treatment to suppress the growth of treatment-resistant populations and thereby delay, or even prevent, tumor progression. Promising clinical results in prostate cancer indicate the potential to optimize adaptive treatment protocols. Here, we applied deep reinforcement learning (DRL) to guide adaptive drug scheduling and demonstrated that these treatment schedules can outperform the current adaptive protocols in a mathematical model calibrated to prostate cancer dynamics, more than doubling the time to progression. The DRL strategies were robust to patient variability, including both tumor dynamics and clinical monitoring schedules. The DRL framework could produce interpretable, adaptive strategies based on a single tumor burden threshold, replicating and informing optimal treatment strategies. The DRL framework had no knowledge of the underlying mathematical tumor model, demonstrating the capability of DRL to help develop treatment strategies in novel or complex settings. Finally, a proposed five-step pathway, which combined mechanistic modeling with the DRL framework and integrated conventional tools to improve interpretability compared with traditional “black-box” DRL models, could allow translation of this approach to the clinic. Overall, the proposed framework generated personalized treatment schedules that consistently outperformed clinical standard-of-care protocols.Significance:Generation of interpretable and personalized adaptive treatment schedules using a deep reinforcement framework that interacts with a virtual patient model overcomes the limitations of standardized strategies caused by heterogeneous treatment responses.

Standard-of-care treatment regimens have long been designed for maximal cell killing, yet these strategies often fail when applied to metastatic cancers due to the emergence of drug resistance. Adaptive treatment strategies have been developed as an alternative approach, dynamically adjusting treatment to suppress the growth of treatment-resistant populations and thereby delay, or even prevent, tumor progression. Promising clinical results in prostate cancer indicate the potential to optimize adaptive treatment protocols. Here, we applied deep reinforcement learning (DRL) to guide adaptive drug scheduling and demonstrated that these treatment schedules can outperform the current adaptive protocols in a mathematical model calibrated to prostate cancer dynamics, more than doubling the time to progression. The DRL strategies were robust to patient variability, including both tumor dynamics and clinical monitoring schedules. The DRL framework could produce interpretable, adaptive strategies based on a single tumor burden threshold, replicating and informing optimal treatment strategies. The DRL framework had no knowledge of the underlying mathematical tumor model, demonstrating the capability of DRL to help develop treatment strategies in novel or complex settings. Finally, a proposed five-step pathway, which combined mechanistic modeling with the DRL framework and integrated conventional tools to improve interpretability compared with traditional “black-box” DRL models, could allow translation of this approach to the clinic. Overall, the proposed framework generated personalized treatment schedules that consistently outperformed clinical standard-of-care protocols.Significance:Generation of interpretable and personalized adaptive treatment schedules using a deep reinforcement framework that interacts with a virtual patient model overcomes the limitations of standardized strategies caused by heterogeneous treatment responses.

Standard-of-care treatment regimens have long been designed for maximal cell killing, yet these strategies often fail when applied to metastatic cancers due to the emergence of drug resistance. Adaptive treatment strategies have been developed as an alternative approach, dynamically adjusting treatment to suppress the growth of treatment-resistant populations and thereby delay, or even prevent, tumor progression. Promising clinical results in prostate cancer indicate the potential to optimize adaptive treatment protocols. Here, we applied deep reinforcement learning (DRL) to guide adaptive drug scheduling and demonstrated that these treatment schedules can outperform the current adaptive protocols in a mathematical model calibrated to prostate cancer dynamics, more than doubling the time to progression. The DRL strategies were robust to patient variability, including both tumor dynamics and clinical monitoring schedules. The DRL framework could produce interpretable, adaptive strategies based on a single tumor burden threshold, replicating and informing optimal treatment strategies. The DRL framework had no knowledge of the underlying mathematical tumor model, demonstrating the capability of DRL to help develop treatment strategies in novel or complex settings. Finally, a proposed five-step pathway, which combined mechanistic modeling with the DRL framework and integrated conventional tools to improve interpretability compared with traditional “black-box” DRL models, could allow translation of this approach to the clinic. Overall, the proposed framework generated personalized treatment schedules that consistently outperformed clinical standard-of-care protocols. Significance: Generation of interpretable and personalized adaptive treatment schedules using a deep reinforcement framework that interacts with a virtual patient model overcomes the limitations of standardized strategies caused by heterogeneous treatment responses.

The immune system is a robust and often untapped accomplice of many standard cancer therapies. A majority of tumors exist in a state of immune tolerance where the patient's immune system has become insensitive to the cancer cells. Because of its lymphodepleting effects, chemotherapy has the potential to break this tolerance. To investigate this, we created a mathematical modeling framework of tumor-immune dynamics. Our results suggest that optimal chemotherapy scheduling must balance two opposing objectives: maximizing tumor reduction while preserving patient immune function. Successful treatment requires therapy to operate in a “Goldilocks Window” where patient immune health is not overly compromised. By keeping therapy “just right,” we show that the synergistic effects of immune activation and chemotherapy can maximize tumor reduction and control.Significance:To maximize the synergy between chemotherapy and antitumor immune response, lymphodepleting therapy must be balanced in a “Goldilocks Window” of optimal dosing.

Expanded model parameter analysis for memory cell growth rates. Greater memory cell expansion leads to a larger Goldilocks treatment window and a greater possible reduction in tumor sizes.

Expanded parameter analysis for Treg death rates. Increasing death rates leads to a larger Goldilocks Window and greater maximum tumor reduction.