Pulmonary Function and Survival 1 Year After Dupilumab Treatment of Acute Moderate to Severe Coronavirus Disease 2019: A Follow-up Study From a Phase 2a Trial

Abstract

Background: We previously conducted a phase 2a randomized placebo-controlled trial of 40 subjects to assess the efficacy and safety of dupilumab use in people hospitalized with coronavirus disease 2019 (COVID-19) (NCT04920916). Based on our preclinical data suggesting that downstream pulmonary dysfunction with COVID-19 induced type 2 inflammation, we contacted patients from our phase 2a study at 1 year for assessment of post-COVID-19 conditions.

Methods: Subjects at 1 year after treatment underwent pulmonary function tests, high-resolution computed tomographic imaging, symptom questionnaires, neurocognitive assessments, and serum immune biomarker analysis, with subject survival also monitored. The primary outcome was the proportion of abnormal diffusion capacity for carbon monoxide (DLCO) or 6-minute walk test (6MWT) at the 1-year visit.

Results: Of those survivors who consented to 1-year visits (n = 16), subjects who had originally received dupilumab were less likely than those who received placebo to have an abnormal DLCO or 6MWT (Fisher exact P = .011; adjusted P = .058). As a secondary endpoint, we saw that 16% of subjects in the dupilumab group died by 1 year compared to 38% in the placebo group, though this was not statistically significant (log-rank P = .12). We did not find significant differences in neurocognitive testing, symptoms, or chest computed tomography between treatment groups but observed a larger reduction in eotaxin levels in those who received dupilumab.

Conclusions: In this observational study, subjects who received dupilumab during acute COVID-19 hospitalization were less likely to have a reduced DLCO or 6MWT, with a nonsignificant trend toward reduced mortality at 1 year compared to placebo.

Keywords: COVID-19; dupilumab, post COVID conditions; pulmonary function; type 2 immunity.

Figure 1.

Bar graph illustrating the percentage of patients who had abnormal pulmonary function testing (defined by abnormal diffusion capacity for carbon monoxide [DLCO] or 6-minute walk test [6MWT]) by treatment group at 1-year follow-up. The orange box depicts the patients randomized to placebo during initial coronavirus disease 2019 (COVID-19) admission and the blue box depicts the subjects randomized to dupilumab during initial COVID-19 admission. Actual number and percentages are displayed in the table at the bottom of the chart. **Two missing values.

Figure 2.

Box plots of percentage predicted pulmonary function measures and 6-minute walk test distance (in meters) between treatment groups. The orange boxes depict the patients randomized to placebo and the blue boxes depict the subjects randomized to dupilumab during initial coronavirus disease 2019 admission. The solid horizontal line within the box is representative of the median value and the open circle within the box is representative of the mean value. A, Diffusing capacity of the lungs for carbon monoxide (DLCO). B, Six-minute walk test (6MWT). C, Forced expiratory volume in 1 second (FEV₁). D, Forced vital capacity (FVC).

Figure 3.

Kaplan-Meier curve depicting 1-year mortality between the 2 treatment groups. Dupilumab group is represented by the blue line. Placebo group is represented by the orange line. Adjusted P value indicative of adjustment for sex in the Cox regression.

Figure 4.

Heatmap showing the standardized differences between baseline and 1-year cytokine, chemokine, or growth factor levels between treatment groups. Each column represents an individual subject labeled by a unique identifier (DP ***). Treatment group is indicated by the top horizontal bar with the blue bar representing the dupilumab group and the orange bar representing the placebo group. Pulmonary function test (PFT) status is represented by the second horizontal bar with subjects who had abnormal PFTs at 1 year (defined as abnormal diffusion capacity for carbon monoxide or 6-minute walk test) represented by the gray bar and those who had normal testing represented by the green bar). Each biomarker is listed on the right of the plot. *Significant differences in 1-year changes between treatment groups. ^Significant differences between 1-year PFT status. P = .065 for IL1Ra.