Pulmonary function trajectories in non-UIP ANCA-positive idiopathic interstitial pneumonia under real-world treatment

Antineutrophil cytoplasmic antibody (ANCA)–positive idiopathic interstitial pneumonia (ANCA-IIP) is an interstitial lung disease (ILD) phenotype with ANCA seropositivity but no systemic evidence of vasculitis, and its disease position and treatment strategy remain uncertain. Patients with non-usual interstitial pneumonia (non-UIP) ANCA-IIP are frequently treated with glucocorticoids and immunosuppressive agents in clinical practice. However, previous studies have mainly focused on clinical characteristics and prognosis, while longitudinal changes in pulmonary function under real-world treatment conditions remain insufficiently described. This study aimed to address the following research question: among patients with ANCA-positive interstitial lung disease (ANCA-ILD), how do pulmonary function trajectories in patients with non-UIP ANCA-IIP differ from those in patients with microscopic polyangiitis-associated interstitial lung disease (MPA-ILD)? This single-center retrospective cohort study screened 1,024 patients with ILD and available ANCA testing between January 2012 and January 2022. Patients with non-UIP ANCA-IIP and MPA-ILD were included according to final multidisciplinary diagnosis and high-resolution computed tomography (HRCT) pattern. Locally weighted scatterplot smoothing with bootstrap 95% CIs was used to visualize 36-month trajectories of forced expiratory volume in 1 s, forced vital capacity (FVC), total lung capacity, diffusing capacity of the lung for carbon monoxide (DL CO) , and prednisone-equivalent dose. Segmented linear mixed-effects models with a 12-month knot compared pulmonary function slopes during 0 to 12 months and 12 to 36 months. Seventy-five patients were included: 39 with non-UIP ANCA-IIP and 36 with MPA-ILD. Compared with MPA-ILD, non-UIP ANCA-IIP showed lower inflammatory activity and no systemic involvement. Baseline DL CO was lower in MPA-ILD (53.50% vs. 65.90%; p = 0.034). During 0 to 12 months, FVC increased in non-UIP ANCA-IIP by 8.12% per 12 months, with a greater improvement than in MPA-ILD (slope difference, 8.38%/12 months; p = 0.038). During 12 to 36 months, the FVC slope was more negative in non-UIP ANCA-IIP than in MPA-ILD (slope difference, − 5.94%/12 months; p = 0.046). All-cause mortality was similar. Non-UIP ANCA-IIP showed early FVC improvement followed by later decline under real-world treatment, supporting HRCT–based phenotyping and continued pulmonary function monitoring.