Formal Modeling and Analysis of Pancreatic Cancer Microenvironment

Qinsi Wang, Natasa Miskov-Zivanov, Bing Liu, James R. Faeder, Michael Lotze, Edmund M. Clarke

TL;DR

This paper presents a novel computational model of the pancreatic cancer microenvironment, integrating intracellular signaling within cancer cells and surrounding stromal cells. By extending a rule-based modeling language and employing statistical model checking, the authors validate existing knowledge and uncover potential new therapeutic strategies for this aggressive disease.

What Problem Does This Paper Address?

Pancreatic cancer (PC) remains a deadly disease with poor prognosis, largely due to the complex interplay between cancer cells and their surrounding microenvironment. While much research has focused on targeting cancer cells directly, the tumor microenvironment, particularly pancreatic stellate cells (PSCs), plays a crucial role in promoting tumor progression, invasion, and resistance to therapy. Understanding this intricate multicellular system and its dynamics is essential for developing effective treatments. However, quantitatively modeling such complex biological systems, with their numerous interacting components and feedback loops, presents a significant challenge.

Methods

The authors developed a computational model of the pancreatic cancer microenvironment by integrating intracellular signaling networks of pancreatic cancer cells (PCCs) and pancreatic stellate cells (PSCs), along with their intercellular communication. They extended the BioNetGen rule-based modeling language to handle both discrete (Boolean Networks for intracellular dynamics) and continuous variables, allowing for the specification of multiscale and multicellular interactions. This extended formalism captures the dynamics of signaling cascades and cellular processes. To analyze the model's behavior and predict system properties, they employed stochastic simulation (NFsim) and statistical model checking (StatMC) with Bounded Linear Temporal Logic (BLTL) to formally verify system properties.

Datasets

Not discussed in the paper. The model parameters were informed by existing literature and experimental observations.

Key Insights and Ideas

  • Bridging the Gap in Modeling: The paper successfully extends a rule-based modeling language (BioNetGen) to handle multicellular interactions and multiscale dynamics (molecular and cellular levels), overcoming limitations of existing formalisms.
  • The Crucial Role of PSCs: The model quantitatively confirms the significant role of PSCs in promoting PCC proliferation and survival, aligning with experimental findings.
  • Uncovering Therapeutic Synergies: The model's analysis reveals potential synergistic effects of targeting multiple pathways simultaneously, suggesting novel (poly)pharmacological strategies for PC treatment.
  • Predictive Power of Formal Methods: Statistical model checking proves to be a powerful tool for validating model behavior against known biological phenomena and for predicting the outcomes of various therapeutic interventions.

Key Results

The study's verification results confirmed several key aspects of the PC microenvironment:

  • PSC Promotion of PCCs: The presence of PSCs significantly increases the probability of PCC populations reaching and maintaining a high level (from 0.405 to 0.9961).
  • PCC Activation of PSCs: PCCs are crucial for activating quiescent PSCs, with the probability of PSC activation increasing from 0.1191 to 0.9961 in the presence of PCCs.
  • Apoptosis vs. Autophagy Balance: The model predicts a dynamic shift where apoptosis initially dominates but is eventually overtaken by autophagy as PC progresses, with a probability close to 1.
  • Therapeutic Efficacy:
    • Existing treatments like cetuximab and erlotinib showed limited benefit in simulations.
  • Strategies targeting PSC depletion (nab-paclitaxel) and JAK/STAT inhibition showed promising results (probabilities of 0.7810 and 0.8004, respectively).
  • Combinatorial inhibition of RAS in PCCs and ERK in PSCs was predicted to synergistically improve PC treatment by reducing bFGF concentration.
  • Inhibiting STAT in PCCs and NFkB in PSCs simultaneously was shown to reduce VEGF secretion, a key factor in angiogenesis and metastasis.

Limitations and Future Work

The authors acknowledge that the model's parameters were estimated based on general values from textbooks and wet-lab measurements, and that a large number of rate constants were not experimentally determined. They also note that the model currently focuses on specific cell types and interactions, and that further work is needed to incorporate other components of the microenvironment.

Future directions include:

  • Empirically testing the predicted therapeutic strategies.
  • Extending the model to include cancer-associated macrophages.
  • Further refining parameter estimation through more detailed experimental data.

Summary updated

Abstract

The focus of pancreatic cancer research has been shifted from pancreatic cancer cells towards their microenvironment, involving pancreatic stellate cells that interact with cancer cells and influence tumor progression. To quantitatively understand the pancreatic cancer microenvironment, we construct a computational model for intracellular signaling networks of cancer cells and stellate cells as well as their intercellular communication. We extend the rule-based BioNetGen language to depict intra- and inter-cellular dynamics using discrete and continuous variables respectively. Our framework also enables a statistical model checking procedure for analyzing the system behavior in response to various perturbations. The results demonstrate the predictive power of our model by identifying important system properties that are consistent with existing experimental observations. We also obtain interesting insights into the development of novel therapeutic strategies for pancreatic cancer.
BibTeX
@inproceedings{wang2016formal,
  author = {Qinsi Wang and Natasa Miskov-Zivanov and Bing Liu and James R. Faeder and Michael Lotze and Edmund M. Clarke},
  title = {{Formal Modeling and Analysis of Pancreatic Cancer Microenvironment}},
  booktitle = {CMSB},
  year = {2016},
  doi = {10.1007/978-3-319-45177-0\_18},
}
Figure