Inhaled granulocyte-macrophage colony-stimulating factor (molgramostim) as host-directed therapy for pneumonia-related acute respiratory distress syndrome: a multicentre, randomised, placebo-controlled phase 2a trial (GI-HOPE).

BACKGROUND: Granulocyte-macrophage colony-stimulating factor (GM-CSF) enhances pulmonary host defence and promotes re-establishment of alveolar barrier function. We investigated the safety, feasibility and efficacy of nebulised recombinant human (rh)GM-CSF (molgramostim) in pneumonia-related acute respiratory distress syndrome (ARDS). METHODS: In this multicentre, randomised, double-blind, parallel-group, placebo-controlled, investigator-initiated phase 2a trial, patients were randomised to receive nebulised low-dose (150 μg) or high-dose (450 μg) rhGM-CSF or placebo for 3 days. Bronchoalveolar lavage (BAL) was performed before the first dose and after dosing. The primary outcome parameter was the composite GI-HOPE score, representing expression changes of CD80, CD86, CD206 and human leukocyte antigen (HLA)-DR on alveolar macrophages after dosing compared to baseline. Secondary outcomes included oxygenation, Sequential Organ Failure Assessment (SOFA) scores and clinical end-points at day 28. RESULTS: 46 participants were randomised, 43 completed treatment (n=15 placebo, n=16 low dose and n=12 high dose), and BAL was performed on 38 participants before and after treatment. Although the composite biological GI-HOPE score did not reach significance (p>0.05), high-dose treatment caused upregulation of HLA-DR and CD206 on alveolar macrophages, indicating deposition in the alveolar compartment and beneficial macrophage activation (HLA-DR: p=0.0406 versus low dose, p=0.1841 versus placebo; CD206: p=0.009 versus low dose, p=0.0179 versus placebo). High-dose rhGM-CSF treatment revealed a favourable profile of oxygenation at the end of analysis compared to time-points before inhalation (p=0.095 and p=0.0175) or compared to the low-dose and placebo groups (p=0.0063 and p=0.0198; visit 13), and was associated with decreased SOFA scores in the high-dose group over time (p=0.0013 and p=0.00023 compared to low dose and placebo). Importantly, inhaled rhGM-CSF did not increase alveolar or systemic inflammation. CONCLUSION: Inhalation of 450 μg rhGM-CSF was safe, and promoted a favourable profile regarding alveolar macrophage activation, oxygenation and SOFA scores over time; however, it did not lead to differences in the GI-HOPE score.