Abstract 421: Computational Modelling Suggests That Reversing Low HDL May Not Always Lead to Plaque Regression

Abstract

Introduction: Clinical evidence suggests that high numbers of high density lipoprotein (HDL) particles in the bloodstream and high levels of HDL efficacy, reduce the risk of heart attack and stroke. Experimental evidence suggests that increasing HDL particle number too late in the life of the plaque does not lead to regression but only to reduced rates of plaque growth. Hypothesis: We postulate that nonlinear dynamics are intrinsic to how cells respond to HDL in reverse cholesterol transport (RCT) from macrophages and foam cells and that these dynamics have consequences for plaque growth, regression and the plaques’ responses to changes in the plaque environment, such as changes in low density lipoprotein (LDL) influx. Methods: We formulated a computational model for the endothelium and the inside of the artery wall. This model consisted of six partial differential equations and associated boundary conditions. We found what conditions led to model plaques remaining small and what led to model plaques growing unboundedly. We also determined the outcomes if the concentration or action of HDL changes at different times in the life of the model plaque. Results: Model plaques with high concentrations of HDL, highly efficacious HDL and low levels of LDL were small and did not grow. There was a tipping point (or bifurcation point) so that plaques with HDL concentrations below the tipping point grew unboundedly and plaques with HDL concentrations above the tipping point remained small. The position of this tipping point depended on the concentration of modified LDL experienced by the plaque, anti-inflammatory actions of HDL and the phenotypes of macrophages after RCT. Increasing HDL concentration after the model plaque had grown led to plaque regression only if the dynamics of the plaque were such that the model plaque was able to shrink to a small plaque that did not grow. If the reduction in HDL was too late, or not large enough, the model plaque could not regress and continued to grow. Conclusion: The nonlinear nature of the dynamics of plaque growth and regression may result in outcomes to treatment that seem unexpected or inexplicable. Computational modelling provides a means of exploring the consequences of these dynamics.