Clinical predictors of treatment response in dry eye disease

Dry eye disease is a multifactorial inflammatory disease characterized by an unstable tear film, most often due to decreased quality of the outer lipid layer produced by modified sebaceous glands in the eyelids. Due to the many different underlying causes of dry eye disease, several treatments and treatment combinations are often used. This is often financially expensive and tiresome for the patient. The present project aims to streamline, optimize and individualize treatment regimens to different dry eye disease subgroups through machine learning and prediction models.

Dry eye disease is a multifactorial disease characterized by inflammation, decreased tear production and/or tear film instability due to dysfunction of the lipid producing glands of the eyelids. As a result, a multimodal treatment approach is often initiated, addressing several pathological clinical signs and modifiable risk factors concomitantly. This is often time consuming, financially expensive and exhausting for the patient. Interpreting signs and symptoms of dry eye disease and making accurate predictions as to which patient will respond which treatment or combination of treatments difficult. With the current project we aim to improve clinical decision making by: 1. Through machine learning and prediction models delineate which group of patients are most likely to benefit from either IPL or cyclosporine in monotherapy, and which patients are most likely to require dual therapy. 2. Construct prediction model protocols to streamline and individualize IPL protocols based on prognostic factors and prediction analysis of treatment outcomes. This is especially applicable in relation to demodex eradication, as demodicosis remains fairly treatment refractory to date and is readily involved in the pathophysiology of MGD. Thus, it is of utmost interest to streamline the current treatment options and, hopefully, contribute to an expanded repertoire of efficacious eradication methods in the future. 3. Contribute to the lack of standardized treatment regimes, recommendations and guidelines concerning IPL treatment. Evaluate the optimum number of treatments and treatment intervals as well as the need for concomitant heat mask and eyelid massage.