Predicting therapy response in psoriasis

Psoriasis is a disease affecting the skin experienced by approximately 3% of the UK population. Moderate-to-severe disease is treated with systemic therapies however a significant fraction of patients will not achieve adequate response. This results in a trial-and-error approach to treating patients which is both undesirable for patients and costly healthcare systems. Identification of clinically meaningful baseline predictors of response has the potential to improve care. Currently the predominant method of sampling a psoriatic skin lesion is through tissue biopsy, a painful and potentially scarring procedure. This project uses proteins measured in the skin by painless non-invasive tape stripping alongside clinical information and serum samples to discriminatively predict therapy response.
Samples and data were collected from two observational clinical studies and analysed with univariate and multivariate approaches to identify means of predicting response. Analysis of the association between tape strip proteins, samples from other locations in the same individual and other features of disease was also made.
Well known single variable predictors of response were identified (BMI, previous number of biologics) as well as a range of putative new tape strip and serum biomarkers. The protease inhibitor Elafin was identified as a potentially discriminative marker of response where higher levels in the serum were associated with worse response to TNF-α inhibitors, IL-17 inhibitors and methotrexate but better response in IL-23 inhibitors. A range of psoriatic markers measured in skin tapes were also identified as being associated aspects of disease such as PASI and lesion scores.
In conclusion, tape strip measurements were shown to be promising in the development of biomarkers for treatment response, however Elafin in serum was the most promising discriminative biomarker overall.