(572ai) Systems Biology of Mucosal Injury In Cancer Patients | AIChE

(572ai) Systems Biology of Mucosal Injury In Cancer Patients

Authors 

Peterson, D. E. - Presenter, University of Connecticut Health Center
Lalla, R. V. - Presenter, University of Connecticut Health Center
Loew, L. M. - Presenter, University of Connecticut Health Center
Srivastava, R. - Presenter, University of Connecticut
White, J. R. - Presenter, University of Connecticut

Background: Mucosal injury (mucositis) caused by high-dose or targeted cancer therapy can result in significant morbidity and financial cost. Understanding the causative complex molecular pathways of this toxicity will enhance design of new therapies. Computational reaction kinetic biological models represent an important strategic advance in this regard. These simulations are cost- and time-efficient and can identify new hypotheses and projected experimental outcomes. The technology also permits integration of complex models based on animal and human data. Methods: A mucositis reaction network structure was developed using COX-2 pathway data derived from literature-based animal and human studies as well as pre- and post-transplant oral mucosal biopsies from three hematopoietic cell transplant patients. Virtual Cell software was used to simulate network dynamics. Results: The nonlinear network mucositis model incorporated feedback loops and accounted for dynamic response of COX-1, COX-2, TNF-α, PGE-2, IL-1β, IL-6, PGI-2, TXA-2 and NF-κB with nonlinear correlation coefficients of -0.73, -4.27, -89.1, -7.85, 0.02, -2.79, 0.63, 0.86 and -0.02 respectively. For 7 of 8 molecular species, the model was able to predict final steady-state value within experimental error with exception of TNF-α. The ninth species, TXA-2, did not achieve steady-state within the timeframe monitored. Conclusions: Using the COX-2 pathway as a prototype, this model successfully incorporated experimental and computational approaches to elucidate mucositis network pathways. The results identify research areas requiring further study such as TNF-α dynamics, thus highlighting benefit of the model. This technology could also facilitate design of new mucositis research protocols including (i) mechanisms of neuropeptide-mediated pain, (ii) identification of tissue-based risk factors and (iii) integration of genetic, cytomorphologic and functional expression of oral and gastrointestinal mucosal injury.