Interstitial lung disease (ILD) is a serious complication that represents the second leading cause of death in patients with rheumatoid arthritis (RA). Treatment of RA-ILD remains controversial. The absence of randomized clinical trials and specific ACR or EULAR therapeutic guidelines makes it difficult to establish solid therapeutic recommendations on this issue. In this scenario, real-world data is especially valuable.

To review the literature evidence on the efficacy and safety of abatacept (ABA) for the treatment of rheumatoid arthritis (RA) with associated interstitial lung disease (ILD), given its clinical relevance and the lack of consensus on its therapeutic management.

PUBMED and EMBASE were searched from the date of approval of ABA to the end of 2020 using a combination of RA, ILD and ABA terms following PRISMA guidelines. Identified studies were evaluated by two independent investigators.

Nine original studies (1 case series and 8 observational studies) were selected for inclusion in the systematic review. No randomized trial or meta-analysis were identified. The mean age of patients ranged from 61.2 to 75 years and the mean RA duration varied from 7.4 to 18 years. Subcutaneous ABA (74.5%-91%) predominated in combination with conventional synthetic DMARDs (csDMARDs) (58%-75%), and it was used as first-line biologic agent in 22.8%-64.9% of the patients. The mean course of ILD ranged from 1 to 6.7 years, being usual and nonspecific interstitial pneumonia the most frequent patterns. Improvement or stabilization of ILD imaging (76.6%-92.7%) and FVC or DLCO (>85%) was described after a mean follow-up of 17.4-47.8 months, regardless of the pattern of lung involvement, being more remarkable in patients with shorter evolution of ILD. ABA led to significantly lower ILD worsening rates than TNF inhibitors (TNFi) and was associated with a 90% reduction in the relative risk of deterioration of ILD at 24 months of follow-up compared to TNFi and csDMARDs. Combination with methotrexate may have a corticoid-sparing effect. No unexpected adverse events were identified.

Current evidence suggests that ABA may be a plausible alternative to treat RA patients with ILD. It would be highly desirable to develop prospective randomized controlled studies to confirm these findings.