Two factors — body mass index (BMI) and the Psoriasis Epidemiology Screening Tool (PEST) — especially predicted development of psoriatic arthritis (PsA) during a 2-year period in patients with psoriasis, according to findings published in Journal of the American Academy of Dermatology.
Researchers conducted a prospective, multicenter, cohort study using data from CorEvitas Psoriasis Registry between April 2015 and November 1, 2020. They analyzed registry data from 1489 patients with psoriasis without previous diagnosis of PsA to determine risk factors that predicted onset of PsA within a 24-month follow-up period. The researchers divided the 1489 patients into 2 data sets: 1042 were included the training data set and 447 were included in the testing data set.
Researchers developed and tested 9 different predictive models — each with unique combinations of patient-related variables — to assess each model’s sensitivity in predicting the onset of PsA within a 2-year period. Predictive variables included patient sex, race, age, ethnicity, insurance provider, educational level, work status, smoking status, alcohol use, comorbidities, BMI, PEST, psoriasis subtype, duration of psoriatic disease, body surface area affected, and several outcome measures on severity of patient-reported symptoms, quality of life, and global assessments.
Dermatologists diagnosed 119 (11.4%) of the 1042 patients in the training data set with PsA after 2 years. The researchers observed a similar PsA occurrence in the testing data set with 41 (9.2%) of the 447 patients receiving a PsA diagnosis after 2 years.
Among the 9 predictive models, the optimal model, which best predicted onset of PsA within 24 months, analyzed 6 variables, including BMI, PEST, the modified Rheumatic Disease Comorbidity Index, work status, alcohol use, and patient-reported fatigue (area under the curve [AUC]=68.9%; 82.9% sensitivity; 48.8% specificity). When analyzing combinations of these 6 variables, the model that included only BMI and PEST resulted in similar predictive capability (AUC=68.8%; 92.7% sensitivity; 36.5% specificity).
While these models demonstrated high sensitivity, AUC scores below 70% and specificities below 50% indicated weak discrimination and high-false positivity rates, respectively.
Limitations of the study include possible misclassification bias for PsA by dermatologists, the lack of consideration of change in therapies between baseline and 24-month follow-ups, and the inherent limitations accompanying model prediction and performance.
Study authors conclude, “[A]mong patients with [psoriasis] treated with systemic therapy, PEST and BMI were important factors in predicting development of PsA over 2 years.” They add, “This study is a first step in the development of a model that could potentially be used to predict future PsA and be easily adapted for routine use in the dermatology clinic.”
Disclosures: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.
Ogdie A, Harrison RW, McLean RR, et al. Prospective cohort study of psoriatic arthritis risk in patients with psoriasis in a real-world psoriasis registry. J Am Acad Dermatol. Published online August 17, 2022:S0190-9622(22)02543-9. doi:10.1016/j.jaad.2022.07.060
This article originally appeared on Rheumatology Advisor