Inhaled Sargramostim (Recombinant Human Granulocyte-Macrophage Colony-Stimulating Factor) for COVID-19-Associated Acute Hypoxemia: Results of the Phase 2, Randomized, Open-Label Trial (iLeukPulm)

INTRODUCTION: Granulocyte-macrophage colony-stimulating factor (GM-CSF), a protein produced in the lung, is essential for pulmonary host defense and alveolar integrity. Prior studies suggest potential benefits in several pulmonary conditions, including acute respiratory distress syndrome and viral infections. This trial evaluated the effect of the addition of inhaled sargramostim (yeast-derived, glycosylated recombinant human GM-CSF) to standard of care (SOC) on oxygenation and clinical outcomes in patients with COVID-19-associated acute hypoxemia. MATERIALS AND METHODS: A randomized, controlled, open-label trial of hospitalized adults with COVID-19-associated hypoxemia (oxygen saturation <93% on ≥2 L/min oxygen supplementation and/or PaO(2)/FiO(2) <350) randomized 2:1 to inhaled sargramostim (125 mcg twice daily for 5 days) plus SOC versus SOC alone. Institutional SOC before and during the study was not limited. Primary outcomes were change in the alveolar–arterial oxygen gradient (P(A–a)O(2)) by day 6 and the percentage of patients intubated within 14 days. Safety evaluations included treatment-emergent adverse events. Efficacy analyses were based on the modified intent-to-treat population, the subset of the intent-to-treat population that received ≥1 dose of any study treatment (sargramostim and/or SOC). An analysis of covariance approach was used to analyze changes in oxygenation measures. The intubation rate was analyzed using the chi-squared test. All analyses are considered descriptive. The study was institutional review board approved. RESULTS: In total, 122 patients were treated (sargramostim, n = 78; SOC, n = 44). The sargramostim arm experienced greater improvement in P(A–a)O(2) by day 6 compared to SOC alone (least squares [LS] mean change from baseline [SE]: −102.3 [19.4] versus −30.5 [26.9] mmHg; LS mean difference: −71.7 [SE 33.2, 95% CI −137.7 to −5.8]; P = .033; n = 96). By day 14, 11.5% (9/78) of sargramostim and 15.9% (7/44) of SOC arms required intubation (P = .49). The 28-day mortality was 11.5% (9/78) and 13.6% (6/44) in the sargramostim and SOC arms, respectively (hazard ratio 0.85; P = .76). Treatment-emergent adverse events occurred in 67.9% (53/78) and 70.5% (31/44) on the sargramostim and SOC arms, respectively. CONCLUSIONS: The addition of inhaled sargramostim to SOC improved P(A–a)O(2), a measure of oxygenation, by day 6 in hospitalized patients with COVID-19-associated acute hypoxemia and was well tolerated. Inhaled sargramostim is delivered directly to the lung, minimizing systemic effects, and is simple to administer making it a feasible treatment option in patients in settings where other therapy routes may be difficult. Although proportionally lower rates of intubation and mortality were observed in sargramostim-treated patients, this study was insufficiently powered to demonstrate significant changes in these outcomes. However, the significant improvement in gas exchange with sargramostim shows this inhalational treatment enhances pulmonary efficiency in this severe respiratory illness. These data provide strong support for further evaluation of sargramostim in high-risk patients with COVID-19.