Cyclophosphamide added to glucocorticoids in acute exacerbation of idiopathic pulmonary fibrosis (EXAFIP): a randomised, double-blind, placebo-controlled, phase 3 trial

Summary Background The use of cyclophosphamide in patients with acute exacerbation of idiopathic pulmonary fibrosis (IPF) is unknown. Our study was designed to evaluate the efficacy and safety of four cyclophosphamide pulses in addition to high-dose methylprednisolone in this population. Methods In this double-blind, placebo-controlled trial done in 35 departments across 31 hospitals in France, adult patients (≥18 years) with acute exacerbation of IPF and those with suspected acute exacerbation of IPF were randomly assigned in a 1:1 ratio using a web-based system to receive either intravenous pulses of cyclophosphamide (600 mg/m2) plus uromitexan as haemorrhagic cystitis prophylaxis (200 mg/m2) at the time of cyclophosphamide administration and then again, 4 h later, or placebo at days 0, 15, 30, and 60. Random assignment was stratified according to the severity of IPF and was block-balanced with variable block sizes of four or six patients. Patients receiving mechanical ventilation, with active infection, with active cancer, or who were registered on the lung transplant waiting list were excluded. All patients received standardised high-dose glucocorticoids. The investigators, patients, and the sponsor were masked to the treatment assignments. The primary endpoint was 3-month all-cause mortality, analysed by a χ2 test adhering to an intention-to-treat principle. The trial is now complete and registered with ClinicalTrials.gov , NCT02460588 . Findings Between Jan 22, 2016, and July 19, 2018, 183 patients were assessed for eligibility, of whom 120 patients were randomly assigned and 119 patients (62 [52%] with severe IPF) received at least one dose of cyclophosphamide (n=60) or placebo (n=59), all of whom were included in the intention-to-treat analysis. The 3-month all-cause mortality was 45% (27/60) in patients given cyclophosphamide compared with 31% (18/59) in the placebo group (difference 14·5% [95% CI −3·1 to 31·6]; p=0·10). Similar results were found after adjustment by IPF severity (odds ratio [OR] 1·89 [95% CI 0·89–4·04]). The risk of death at 3 months, independent of the treatment received, was higher with severe than non-severe IPF (OR 2·62 [1·12–6·12]) and was lower with the use of antifibrotic therapy (OR 0·33 [0·13–0·82]). Adverse events were similar between groups by 6 months (25 [42%] in the cyclophosphamide group vs 30 [51%] in the placebo group) and their proportion, including infections, did not differ. Overall infection was the main adverse event and occurred in 20 (33%) of 60 patients in the cyclophosphamide group versus 21 (36%) of 59 patients in the placebo group. Interpretation In patients with acute exacerbation of IPF, adding intravenous cyclophosphamide pulses to glucocorticoids increased 3-month mortality. These findings provide evidence against the use of intravenous cyclophosphamide in such patients. Funding Programme Hospitalier de Recherche Clinique of the French Ministry of Health (PHRC 2014–502), Roche Pharmaceuticals.