Individualised computational modelling of immune mediated disease onset, flare and clearance in psoriasis

Abstract

Despite increased understanding about psoriasis pathophysiology, currently there is a lack of predictive computational models. We developed a personalisable ordinary differential equations model of human epidermis that features two stable steady states: healthy skin and psoriasis. In line with experimental data, an immune stimulus initiated transition from healthy skin to psoriasis and apoptosis induced by UVB phototherapy returned the epidermis back to the healthy state. The flexibility of our model permitted the development of a patient-specific "UVB sensitivity" parameter that enabled accurate simulation of individual patients' clinical response trajectory. In a prospective clinical study of 94 patients, serial individual UVB doses and clinical response (Psoriasis Area Severity Index) values collected over the first three weeks of UVB therapy informed estimation of the "UVB sensitivity" parameter and the prediction of patient outcome at the end of phototherapy. Notably, our model was able to distinguish disease flares and offers the potential for clinical application in early assessment of response to UVB therapy outcome, and for individualised optimisation of phototherapy regimes to improve clinical outcome.

Computational Modelling of Immune Interaction and Epidermal Homeostasis in Psoriasis

“…A previous computational study [10] showed that the cell population ratio remained the same in both healthy and psoriasis. In addition, the overall number of epidermal cells is known to increase by up to two to five times as compared to the healthy epidermis [25,33].…”

“…The growth model for both healthy and psoriatic epidermis are based on various nutrients and immune cytokines that have been found in literature and in clinical studies [7,10,11]. Cell division of proliferative stem and TA cells are based on having two interconvertible modes as mentioned in [12] where proliferative cells change their division type based on an expanding or balanced state.…”

“…The Monod-based kinetics allows stem and TA cells to produce GF and differentiated cells to secrete calcium into the environment. The concept of GF production by stem and TA cells is based on a previously developed ODE model [10], while calcium is stored in the proliferative cells as they grow, divide and terminally differentiate and gets secreted when a differentiated cell reaches the stratum granulosum layer.…”

“…Psoriasis is modelled by activating an immune cytokine stimulus such as IL-22 and TNFα into the model, which causes an onset between 7 days to 14 weeks and a time lag between the time the patient gets an immunological response (such as from a streptococcal sore throat) to the appearance of the disease itself [7,21,22]. In this model, the immune cytokine stimuli is induced for a total of 7 days following the previously developed ODE model [10].…”

“…Some proposals for future studies include modelling NB-UVB phototherapy treatments, exploring different doses and frequencies of NB-UVB treatments and using single or combination mechanisms of action for clearance, similar to the 2D model previously developed by [7]. In addition, the model outputs can also be fed into a machine learning algorithm previously developed [10] to make predictions of how each simulation can be clustered to emulate real life patients. It can also be modified to investigate specific immune cytokine stimuli and T cell species or a combination of them to better understand how each cytokine affects the skin to cause different variations of psoriasis.…”