Common variable immunodeficiency (CVID) affects approximately 10,000 people in the USA and is caused by an expanding array of genetic lesions. It is defined by international consensus criteria that may be summarized as loss of B cell function developing after infancy with or without a well-described constellation of autoimmune or lymphoproliferative complications. The development of a distinctive form of interstitial lung dis- ease, termed granulomatous-lymphocytic interstitial lung disease (GLILD), is particularly problematic in CVID because it is relatively common, associated with significant mortality and morbidity, and has an unmet need for adequate treatments. We found that the biologic drug abatacept showed good efficacy for reversal of severe, often refractory GLILD in a small series of patients with CVID. Abatacept is a recombinant fusion protein incor- porating CTLA-4 (cytotoxic T lymphocyte–associated protein 4) that blocks T cell activation by binding to CD80/CD86, thereby preventing CD28 engagement. Our findings suggest that abatacept would be an effec- tive therapy for GLILD in CVID. However, rigorous clinical trials are needed to prospectively define the risks and benefits of abatacept as a therapy for GLILD. To fill this critical gap, we have designed the ABCVILD trial. In this trial, we plan to treat with either pla- cebo or abatacept in a double-blinded fashion. Following a six-month blinded treatment phase, patients will enter a six-month open-label phase in which they will start or continue abatacept. On the basis of our clinical experience to date, we hypothesize that 6 months of abatacept therapy (compared to placebo) will clinically improve GLILD (as assessed by objective and QOL measures) across CVID genotypes and improve sCD25 and other exploratory biomarkers of T cell activation. We will test this hypothesis by pursuing these aims: Aim 1: Determine the response rate of GLILD after six months of abatacept therapy (versus placebo). Our readouts for the ABCVILD trial include quantitative assessment of chest CT scans, pulmonary function tests, quality of life measures, and radiation-free measure 129Xe magnetic resonance image (MRI). Aim 2: Determine whether genotype and/or lung histology predict GLILD responsiveness to abatacept therapy. As CD4+ T cells are often the most prominent infiltrating cells in GLILD and T follicular helper (Tfh) cells have been shown experimentally to be sensitive to CD28/CTLA4 manipulation, we hypothesize that GLILD is largely a disease of Tfh cells and that responsiveness will correlate with these cells in lesions. Aim 3: Define the utility of sCD25, abatacept pharmacokinetics, or exploratory biomarkers of T cell acti- vation for predicting response to therapy. We will serially assess sCD25, various exploratory biomarkers, and abatacept levels to test the predictive value of each, understand the effects of our adaptive dosing regimen on these parameters, and further test our hypothesize that G...