Combination therapies have led to drastic improvements in cancer patient outcomes, and are a routine part of patient care. However, despite promising initial evidence, results from combination therapies have been disappointing. A major challenge is deciding how to optimally administer combination therapies, and currently, most combination therapies are administered based on empirical experience of the individual drugs used as monotherapies. Several preclinical and clinical studies have shown that altering therapy administration schedules can significantly improve survival outcomes, suggesting the current methods of administering combination therapies is suboptimal. However, it is infeasible to systematically test all possible administration schedules experimentally, due to the large search space. Mathematical modeling, however, is perfectly suited to systematically search through the possible dose administration schedules and combination therapies. This project aims to develop mathematical models of two novel combination therapies to treat head and neck cancer and estrogen receptor positive (ER+) breast cancer, respectively, to identify optimal treatment strategies.
The first aim will seek to develop a mathematical model of radiation-Ataxia telangiectasia and Rad3 related (ATR) inhibitor combination therapy for treating head and neck cancer.
The second aim will seek to develop a mathematical model of endocrine therapy-cyclin dependent kinases (CDK) 4/6 inhibitor combination therapy for treating ER+ breast cancer. Schedule optimization has not been performed for either combination therapy. The development of both models follows a similar strategy. First, for a given drug combination, in vivo experiments that measure dynamic treatment response under the entire space of clinically acceptable administration schedules will be used to develop and parameterize the mathematical model. Second, clinical trial data will be used to estimate pharmacokinetic parameters and drug toxicity. Third, the model will be optimized to maximize treatment efficacy, using toxicity limits as constraints. Lastly, the optimal schedules will be validated using in vivo preclinical studies. If the model is not validated, in vivo results will be used to update the model. The new model will then be re-optimized and re-validated. This iterative process will continue until the model is successfully validated.
These aims will lead to novel, preclinically validated schedules that can be evaluated in clinical trials. The environment in which this research will take place is a highly impactful computational biology research group led by Dr. Franziska Michor. The group prioritizes individualized mentorship of each trainee, including weekly one-on-one meetings between Dr. Michor and each trainee, productive collaborations, fostered by twice per week Michor Lab meetings and monthly joint lab meetings with experimental collaborators, and innovative research with clear clinical impact. This highly collaborative project will lead to novel, preclinically validated combination therapy administration schedules that are predicted to outperform current standards of care.

Public Health Relevance

The success rate of combination treatments has thus far been disappointing, in part due to sub-optimal administration schedules. This project aims to develop novel mathematical models of two novel combination therapies to treat head and neck cancer and estrogen receptor positive breast cancer, respectively, using in vitro data to build the models and in vivo data for validation. The models will enable a systematic search through all possible dose administration schedules in order to identify optimal treatment strategies, thereby improving combination treatment efficacy and patient outcomes.

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Predoctoral Individual National Research Service Award (F31)
Project #
5F31CA239565-02
Application #
9979625
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Korczak, Jeannette F
Project Start
2019-08-01
Project End
2021-07-31
Budget Start
2020-08-01
Budget End
2021-07-31
Support Year
2
Fiscal Year
2020
Total Cost
Indirect Cost
Name
Harvard University
Department
Biostatistics & Other Math Sci
Type
Schools of Public Health
DUNS #
149617367
City
Boston
State
MA
Country
United States
Zip Code
02115