Optimal Chemotherapy for Leukemia: A Model-Based Strategy for Individualized Treatment

Abstract

Acute Lymphoblastic Leukemia, commonly known as ALL, is a predominant form of cancer during childhood. With the advent of modern healthcare support, the 5-year survival rate has been impressive in the recent past. However, long-term ALL survivors embattle several treatment-related medical and socio-economic complications due to excessive and inordinate chemotherapy doses received during treatment. In this work, we present a model-based approach to personalize 6-Mercaptopurine (6-MP) treatment for childhood ALL with a provision for incorporating the pharmacogenomic variations among patients. Semi-mechanistic mathematical models were developed and validated for i) 6-MP metabolism, ii) red blood cell mean corpuscular volume (MCV) dynamics, a surrogate marker for treatment efficacy, and iii) leukopenia, a major side-effect. With the constraint of getting limited data from clinics, a global sensitivity analysis based model reduction technique was employed to reduce the parameter space arising from semi-mechanistic models. The reduced, sensitive parameters were used to individualize the average patient model to a specific patient so as to minimize the model uncertainty. Models fit the data well and mimic diverse behavior observed among patients with minimum parameters. The model was validated with real patient data obtained from literature and Riley Hospital for Children in Indianapolis. Patient models were used to optimize the dose for an individual patient through nonlinear model predictive control. The implementation of our approach in clinical practice is realizable with routinely measured complete blood counts (CBC) and a few additional metabolite measurements. The proposed approach promises to achieve model-based individualized treatment to a specific patient, as opposed to a standard-dose-for-all, and to prescribe an optimal dose for a desired outcome with minimum side-effects.

Synthesis and cytotoxic activity evaluation of some novel 1-(3-(aryl-4,5-dihydroisoxazol-5-yl)methyl)-4-trihalomethyl-1 H -pyrimidin-2-ones in human cancer cells

“…Several structural changes in both the nitrogenated base and the furanose ring M A N U S C R I P T A C C E P T E D ACCEPTED MANUSCRIPT 3 As a result of these studies, a series of DNA bases and nucleoside analogues have been developed and have been used in various types of cancer treatment. Among these, one can highlight 5-fluorouracil, 6-mercaptopurine and cytarabine ( Figure 1) [8][9][10].…”

Model-Based Simulation of Maintenance Therapy of Childhood Acute Lymphoblastic Leukemia

“…The model contained the BSA as a covariate in the clearance and thus provides individualized PK profiles. The PK model of Jayachandran et al (2014) was validated on concentration data of 8 patients (adults) from Hindorf et al (2006). However, the simulated 6-TGN concentrations coincided with data from pediatric patients reported by Hawwa et al (2008); hence, it was a priori unclear which would give better results.…”

“…During the model development, we replaced the 6MP PK model of Jayachandran et al (2014) with the PK model described by Hawwa et al (2008) to obtain a better response to 6MP dosage. The model contained the BSA as a covariate in the clearance and thus provides individualized PK profiles.…”

The Union between Structural and Practical Identifiability Makes Strength in Reducing Oncological Model Complexity: A Case Study

“…The practical parameter identification framework considered is based on (simulated) noisy measurements of the dynamic system (see (1)) output taken over a finite horizon at discrete time points { 푗 , = 1, . .…”

“…Many mathematical models of tumor growth at different levels from gene expression to the macroscopic tumor development have been formulated, [1][2][3][4][5][6][7][8][9][10][11]. Recently, mathematical models of tumor-immune interactions have also been considered to evaluate the efficacy of immunotherapy in the context of tumor challenge.…”