[Paper Review] Investigating the development of chemotherapeutic drug resistance in cancer: A multiscale computational study

Gibin Powathil, Mark A. J. Chaplain+1 more|arXiv (Cornell University)|Jul 3, 2014

Mathematical Biology Tumor Growth58 references14 citations

TL;DR

This multiscale computational study uses a hybrid individual-based model to investigate how intratumoral heterogeneity and slow-cycling cancer cells drive cell-cycle-mediated chemotherapeutic resistance. The model demonstrates that phase-specific drugs spare non-targeted cell-cycle phases, enriching resistant slow-cycling subpopulations, and shows that sequential multi-dose regimens targeting all phases significantly improve cell kill and reduce resistance emergence.

ABSTRACT

Chemotherapy is one of the most important therapeutic options used to treat human cancers, either alone or in combination with radiation therapy and surgery. Recent studies have indicated that intra-tumoural heterogeneity has a significant role in driving resistance to chemotherapy in many human malignancies. Multiple factors including the internal cell-cycle dynamics and the external microenvironement contribute to the intra-tumoural heterogeneity. In this paper we present a hybrid, multiscale, individual-based mathematical model, incorporating internal cell-cycle dynamics and changes in oxygen concentration, to study the effects of delivery of several different chemotherapeutic drugs on the heterogeneous subpopulations of cancer cells with varying cell-cycle dynamics. The computational simulation results from the multiscale model are in good agreement with available experimental data and support the hypothesis that slow-cycling sub-populations of tumour cells within a growing tumour mass can induce drug resistance to chemotherapy and thus the use of conventional chemotherapy may actually result in the emergence of dominant, therapy-resistant, slow-cycling subpopulations of tumour cells. Our results indicate that the appearance of this chemotherapeutic resistance is mainly due to the inability of the administered drug to target all cancer cells irrespective of the stage in the cell-cycle they are in i.e. most chemotherapeutic drugs target cells in a particular phase/phases of the cell-cycle, and hence always spare some cancer cells that are not in the targeted cell-cycle phase/phases. The results also suggest that this cell-cycle-mediated drug resistance may be overcome by using multiple doses of cell-cycle, phase-specific chemotherapy that targets cells in all phases and its appropriate sequencing and scheduling.

Motivation & Objective

• To investigate the role of intratumoral heterogeneity and slow-cycling subpopulations in mediating chemotherapeutic resistance.
• To understand how cell-cycle-specific drugs fail to eliminate all cancer cells due to phase-dependent targeting.
• To evaluate the impact of drug scheduling and sequencing on tumor cell kill and resistance enrichment.
• To develop a predictive multiscale model that integrates intracellular cycle dynamics with extracellular microenvironmental factors such as oxygen levels.
• To provide a computational framework for designing patient-specific, adaptive chemotherapy protocols that overcome intrinsic resistance mechanisms.

Proposed method

• A hybrid, multiscale, individual-based mathematical model was developed using the Compucell3D framework to simulate tumor growth and drug response.
• The model incorporates detailed internal cell-cycle dynamics with four phases (G1, S, G2, M), including regulation by cyclin-dependent kinases and cell-cycle inhibitors.
• Intracellular mutations were introduced to generate a slow-cycling subpopulation, mimicking quiescent or G0-like states with prolonged cycle times.
• Oxygen concentration gradients were simulated to reflect the tumor microenvironment's influence on cell-cycle progression and drug efficacy.
• Drug effects were modeled as phase-specific cytotoxicity: G1-phase drugs target cells in G1, and S-G2-M drugs target cells in S, G2, and M phases.
• Multiple dosing regimens were simulated to assess the impact of sequencing and scheduling on overall cell kill and resistance development.

Experimental results

Research questions

• RQ1 How does intratumoral heterogeneity, particularly slow-cycling subpopulations, contribute to chemotherapeutic resistance?
• RQ2 To what extent do phase-specific chemotherapeutic drugs fail to eliminate all cancer cells due to cell-cycle phase variation?
• RQ3 What is the optimal sequencing and scheduling of multiple phase-specific drugs to maximize tumor cell kill and minimize resistance enrichment?
• RQ4 How do oxygen gradients and microenvironmental factors interact with cell-cycle dynamics to influence drug response?
• RQ5 Can a multiscale computational model accurately reproduce experimental observations of resistance emergence in heterogeneous tumors?

Key findings

• Slow-cycling subpopulations, particularly enriched in G2/M phase, exhibit intrinsic resistance to phase-specific chemotherapeutic drugs due to their non-proliferative state.
• Conventional chemotherapy regimens that target only specific cell-cycle phases fail to eliminate slow-cycling cells, leading to their enrichment post-treatment.
• Simulation results show that two doses of G1-phase-specific drugs achieve the highest cell-kill rate, while two doses of S-G2-M drugs result in the lowest kill rate.
• The combination of G1 and S-G2-M phase-specific drugs yields the most effective cell kill, confirming that targeting all phases is superior to single-phase targeting.
• The model’s predictions are in qualitative agreement with experimental data showing increased G2/M phase prevalence in slow-cycling subpopulations.
• The results support the need for patient-specific, adaptive therapy protocols that account for tumor heterogeneity and cell-cycle dynamics to improve clinical outcomes.