Background There is little evidence base to guide clinicians in the management of patients with rheumatoid arthritis (RA) related interstitial lung disease (ILD). In the subgroup who have rapidly deteriorating respiratory function, pulsed cyclophosphamide has been advocated but no assessment of its efficacy has been published. The British Rheumatoid InterstitiaL Lung (BRILL) network has collected data on the treatment and outcomes of 260 patients with RA-ILD, which allow assessment of this therapeutic strategy.
Objectives To assess the effect of pulsed IV cyclophosphamide on the outcome of patients with RA-ILD
Methods We identified 260 patients across 16 UK centres who met the 2010 EULAR criteria for RA and had confirmation of ILD on high resolution computed tomography (HRCT) of the lungs. We analysed demography, all drug therapy, duration of disease, subtype and extent of ILD on HRCT, smoking history, serology and baseline pulmonary function. For each patient who had received pulsed intravenous cyclophosphamide, we identified two case controls with RA-ILD who had never received this drug and who were matched for age, gender, disease subtype and extent. We compared duration of both RA and of ILD between the two groups, as well as vital capacity (VC) and gas transfer (TLco). Mortality and mean survival times in each group were compared.
Results We identified 21 patients who had received pulsed cyclophosphamide for progressive RA-ILD. Ten were male and the group median age was 71 (58-86) years. Most patients were smokers (13/21) and all were seropositive. Median duration of ILD and RA were 2 and 11 years respectively in both the cyclophosphamide treated group and in the case controls. In both groups, 14 patients had UIP (with 7 NSIP), and 14 patients had extensive disease (with 7 limited). Median baseline % predicted VC (range) was lower in the cyclophosphamide patients at 71% (40-117%) than in case controls 84% (55-134%) [p=0.015], as was TLco at 45% (25-109%) vs 57% (27-82%) [p=0.04]. Mortality was identical between the groups at 24%, with mean survival better in those treated with cyclophosphamide than in case controls (72 vs 43 months)[p=0.13].
Conclusions This data from the BRILL network shows that the use of pulsed cyclophosphamide for patients with progressive RA-ILD is associated with mortality and survival at least as good as in RA-ILD patients with lesser impairment of baseline pulmonary function. This retrospective analysis supports the early use of such therapy in patients with deteriorating RA-ILD. A prospective study similar to that conducted in scleroderma lung disease would now seem appropriate to confirm these results and define an optimal treatment regime.
Disclosure of Interest None declared