Cyclophosphamide in Patients with Systemic Sclerosis-associated Interstitial Lung Disease A Systematic Review and Meta-Analysis

Background: The American Thoracic Society convened an international, multidisciplinary panel to develop clinical practice guidelines for the treatment of systemic sclerosis-associated interstitial lung disease (SSc-ILD). Objective: To conduct a systematic review and evaluate the literature to determine whether patients with SSc-ILD should be treated with cyclophosphamide. Data Sources: A literature search was conducted across the MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials databases through June 2022 for studies using cyclophosphamide to treat patients with SSc-ILD. Data Extraction: Mortality, disease progression, quality of life, and adverse event data were extracted, and meta-analyses were performed when possible. The Grading of Recommendations, Assessment, Development and Evaluation Working Group method was used to assess the quality of evidence. Synthesis: Five studies were included; two randomized controlled trials compared cyclophosphamide versus placebo, and one randomized controlled trial and two retrospective casecontrol studies compared cyclophosphamide versus mycophenolate. Compared with placebo, there was a 2.83% reduction in the decline at 12 months for forced vital capacity (FVC) % predicted using cyclophosphamide (95% confidence interval [CI], 0.80-4.87; low evidence). There were improvements in breathlessness (Transition Dyspnea Index mean difference [MD], 2.90; 95% CI, 1.94-3.86; minimum clinically important difference, 1; moderate evidence) and disability (Health Assessment Questionnaire-Disability Index MD, 20.16; 95% CI, 20.28 to 20.04; minimum clinically important difference, 20.14; moderate evidence). There were increased risks of leukopenia and constitutional symptoms using cyclophosphamide, but no difference in mortality. When cyclophosphamide was compared with mycophenolate, there were differences in diffusing capacity of the lung for carbon monoxide % predicted favoring mycophenolate at 6 months (MD, 23.67%; 95% CI, 26.3% to 21.1% unadjusted; MD, 24.88%; 95% CI, 27.3% to 22.5% adjusted for alveolar volume; moderate evidence), 12 months (MD, 25.90%; 95% CI, 28.4% to 23.4% adjusted for alveolar volume; moderate evidence), and 18 months (MD, 23.26%; 95% CI, 26.1% to 20.4%; moderate evidence), but not at 24 months. There were no differences in FVC % predicted, mortality, or quality-of-life outcomes, but participants were more likely to prematurely discontinue cyclophosphamide compared with mycophenolate (relative risk, 1.70; 95% CI, 1.10-2.63; highcertainty evidence). Conclusions: A review of the published evidence shows that cyclophosphamide is effective in SSc-ILD compared with placebo, with an increased risk of side effects. However, mycophenolate may be equivocal or better than cyclophosphamide. Clinicians and patients should weigh the potential benefits and risks with respect to individual patient circumstances and preferences.