D2.385 - Granulomatous and Lymphocytic Inerstitial Lung Disease (GLILD) in Common Variable Immunodeficiencies (CVIDs): preliminary results of a multicenter retrospective observational study

GLILD is one of the most severe non-infectious complications of CVIDs. Despite its clinical relevance, prognostic stratification is poorly defined, and no evidence-based management guidelines are currently available.

This study aimed to identify clinical and immunological parameters associated with poor prognosis and to evaluate the efficacy of different treatment strategies, with particular focus on rituximab monotherapy (RTXm).

Patients with a clinical–radiological and/or histological diagnosis of GLILD were retrospectively enrolled from six Italian centers. Poor prognosis was defined by the need for immunosuppressive therapy for GLILD, development of lymphoma, or death. We subsequently compared the characteristics of the poor-prognosis group with those of the remaining patients. The efficacy of rituximab monotherapy was assessed by HRCT and, when available, spirometry and PET-CT.

86 patients were included, of whom 46 met criteria for poor prognosis: 38 required immunosuppressive therapy due to GLILD progression, 7 developed lymphoma, and 12 died. Notably, patients with poor prognosis had a significantly longer delay in CVID diagnosis (9 vs. 5 years, p = 0.032). From an immunological perspective, these patients exhibited a lower CD4+/CD8+ ratio in bronchoalveolar lavage fluid (1.10 vs. 2.20, p = 0.002). No significant differences were observed in peripheral blood lymphocyte subpopulations. Regarding treatment, 24 patients received at least one cycle of rituximab monotherapy with a follow-up longer than 6 months. 19 showed radiological improvement on HRCT, with documented reduction in ground-glass opacities, pulmonary nodules, and lymphadenopathies. Among them, 14 also underwent PET-CT evaluation, demonstrating reduced hypermetabolic activity in lymph nodes and/or lung tissue. Spirometry data were available for 21 patients and showed a significant improvement in DLCO (74% vs 84.6%, p = 0.002). 9 patients achieved a clinically meaningful improvement (>10% increase in DLCO and/or >5% increase in FVC). After a median follow-up of 22 months (IQR 10–32), 9 patients (41% of patients treated with RTXm) experienced clinical–radiological relapse, with a median time to relapse of 25 months (IQR 19–31).

This study identifies potential prognostic markers in GLILD and suggests that rituximab monotherapy may be an effective induction treatment in patients with a severe phenotype. However, maintenance therapy may be necessary to ensure sustained long-term disease control.