A current challenge in cancer research is the development of therapeutic strategies aimed at reducing the toxicity of treatments, since Adverse Events (AEs) typically cause substantial problems and long-term damages to the patients. A possible solution to this issue lies in the personalization of therapy dosages according to demographic factors and in the employment of optimized data-driven drug administration protocols. Control theory can be exploited to this end, as its application in pharmacology allows to define optimized dosages and schedules, aimed at minimizing AEs and maximizing the therapy efficacy. However, an effective application of control theory approaches to this issue is constrained by our ability in inferring the parameters of the mathematical models from currently available data.We here present a closed-loop optimization framework of patient-specific pharmacokinetics (PK) and pharmacodynamics (PD) models, combined with a mathematical model of a liquid tumor, which aims at overcoming such limitations. The most relevant feature of our framework is the ability to learn the value of patient-specific parameters via a Bayesian update, by exploiting a feedback signal obtained monitoring the tumor burden dynamics of the patient. Our framework employs CasADi, an open-source tool for nonlinear optimization, and guarantees a good and robust numerical estimation of the optimized schedule and a parsimonious use of computational time.As a case study, we present the application of our framework to Tyrosine Kinase Inhibitor administration in Chronic Myeloid Leukemia (CML), in which we show that our optimized protocols result in a faster decay of CSCs and in a reduction of the overall toxicity.

Angaroni, F., Pennati, M., Patruno, L., Maspero, D., Antoniotti, M., & Graudenzi, A. (2020). A closed-loop optimization framework for personalized cancer therapy design. In 2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB) (pp.1-9) [10.1109/CIBCB48159.2020.9277647].

A closed-loop optimization framework for personalized cancer therapy design

Angaroni, Fabrizio
Primo
;
Patruno, Lucrezia;Maspero, Davide;Antoniotti, Marco
Co-ultimo
;
Graudenzi, Alex
Co-ultimo
2020

Abstract

A current challenge in cancer research is the development of therapeutic strategies aimed at reducing the toxicity of treatments, since Adverse Events (AEs) typically cause substantial problems and long-term damages to the patients. A possible solution to this issue lies in the personalization of therapy dosages according to demographic factors and in the employment of optimized data-driven drug administration protocols. Control theory can be exploited to this end, as its application in pharmacology allows to define optimized dosages and schedules, aimed at minimizing AEs and maximizing the therapy efficacy. However, an effective application of control theory approaches to this issue is constrained by our ability in inferring the parameters of the mathematical models from currently available data.We here present a closed-loop optimization framework of patient-specific pharmacokinetics (PK) and pharmacodynamics (PD) models, combined with a mathematical model of a liquid tumor, which aims at overcoming such limitations. The most relevant feature of our framework is the ability to learn the value of patient-specific parameters via a Bayesian update, by exploiting a feedback signal obtained monitoring the tumor burden dynamics of the patient. Our framework employs CasADi, an open-source tool for nonlinear optimization, and guarantees a good and robust numerical estimation of the optimized schedule and a parsimonious use of computational time.As a case study, we present the application of our framework to Tyrosine Kinase Inhibitor administration in Chronic Myeloid Leukemia (CML), in which we show that our optimized protocols result in a faster decay of CSCs and in a reduction of the overall toxicity.
No
paper
Scientifica
Mathematical model,Medical treatment,Tumors,Cancer,Optimization,Drugs,Bayes methods
English
IEEE Symposium on Computational Intelligence and Bioinformatics and Computational Biology (CIBCB) 27-29 Oct. 2020
978-1-7281-9468-4
Angaroni, F., Pennati, M., Patruno, L., Maspero, D., Antoniotti, M., & Graudenzi, A. (2020). A closed-loop optimization framework for personalized cancer therapy design. In 2020 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB) (pp.1-9) [10.1109/CIBCB48159.2020.9277647].
Angaroni, F; Pennati, M; Patruno, L; Maspero, D; Antoniotti, M; Graudenzi, A
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/10281/298078
Citazioni
  • Scopus 2
  • ???jsp.display-item.citation.isi??? 0
Social impact