Comparison of patient self-reported data to physician-reported data in eosinophilic granulomatosis with polyangiitis (Preprint)

Abstract

Patient-based registries can help advance research in rare diseases such as eosinophilic granulomatosis with polyangiitis (EGPA), a complex, multi-organ form of antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis.

To compare patient-reported vs. physician-reported manifestations, treatments, and outcomes for patients with EGPA.

Comparative analysis of patients ≥18 years with EGPA in Canada or the United States from two separate cohorts: i) The Vasculitis Patient-Powered Research Network (VPPRN), a self-enrolled, secure portal with patient-entered data updated quarterly (2014-2019), vs. ii) The Vasculitis Clinical Research Consortium (VCRC) observational studies, a physician-entered database (2003-2019) of patients who fulfilled the 1990 American College of Rheumatology classification criteria for EGPA. Studied parameters included demographics, clinical manifestations, ANCA status, treatments, and relapses.

Data from 195 patients with a validated diagnosis of EGPA in the VPPRN and 354 patients enrolled in the VCRC were analyzed. Compared to the VCRC cohort, the patients in the VPPRN cohort were more commonly female (69.2% vs. 59.0% in the VCRC cohort; P =.02), younger at diagnosis (47.3 vs. 50.0 years; P =.03), reported similar frequencies of asthma (96.2% vs 92.9% in VCRC; P =.13), cardiac manifestations (28.8% vs 21.2%; P =.06), but less frequent lung manifestations other than asthma, and more frequent disease manifestations in all other organ systems. ANCA positivity was 48.9% in the VPPRN patients vs. 38.9% (P=.05) in the VCRC cohort. Relapsing disease after study enrollment was reported in 32.3% patients in the VPPRN compared 35.7% of patients in the VCRC. Most therapies (glucocorticoids, cyclophosphamide, mepolizumab) were used at similar frequencies in both groups, except for rituximab with VPPRN patients reporting more use than VCRC cohort (24.1% vs. 10.5%; P =<.001).

Overall, patients and physicians report manifestations of EGPA at similar frequencies. However, observed differences between patient/ physician reports imply possible selection biases occurred. These results support the use of patient-reported data in EGPA, but also the need for careful consideration of disease-specific definitions for the study of EGPA and how patient- and physician-reported data are collected.

ClinicalTrials.gov: (1) VCRC Longitudinal Study (LS) NCT00315380 https://clinicaltrials.gov/ct2/show/NCT00315380 and (2) One-Time DNA (OT) study NCT01241305 https://clinicaltrials.gov/ct2/show/NCT01241305