Oral famotidine versus placebo in non-hospitalised patients with COVID-19: a randomised, double-blind, data-intense, phase 2 clinical trial

Abstract

Objective: We assessed whether famotidine improved inflammation and symptomatic recovery in outpatients with mild to moderate COVID-19.

Design: Randomised, double-blind, placebo-controlled, fully remote, phase 2 clinical trial (NCT04724720) enrolling symptomatic unvaccinated adult outpatients with confirmed COVID-19 between January 2021 and April 2021 from two US centres. Patients self-administered 80 mg famotidine (n=28) or placebo (n=27) orally three times a day for 14 consecutive days. Endpoints were time to (primary) or rate of (secondary) symptom resolution, and resolution of inflammation (exploratory).

Results: Of 55 patients in the intention-to-treat group (median age 35 years (IQR: 20); 35 women (64%); 18 African American (33%); 14 Hispanic (26%)), 52 (95%) completed the trial, submitting 1358 electronic symptom surveys. Time to symptom resolution was not statistically improved (p=0.4). Rate of symptom resolution was improved for patients taking famotidine (p<0.0001). Estimated 50% reduction of overall baseline symptom scores were achieved at 8.2 days (95% CI: 7 to 9.8 days) for famotidine and 11.4 days (95% CI: 10.3 to 12.6 days) for placebo treated patients. Differences were independent of patient sex, race or ethnicity. Five self-limiting adverse events occurred (famotidine, n=2 (40%); placebo, n=3 (60%)). On day 7, fewer patients on famotidine had detectable interferon alpha plasma levels (p=0.04). Plasma immunoglobulin type G levels to SARS-CoV-2 nucleocapsid core protein were similar between both arms.

Conclusions: Famotidine was safe and well tolerated in outpatients with mild to moderate COVID-19. Famotidine led to earlier resolution of symptoms and inflammation without reducing anti-SARS-CoV-2 immunity. Additional randomised trials are required.

Keywords: COVID-19; clinical trials; inflammation; interferon.

Figure 1

Trial overview. (A) The trial schematic and (B) the CONSORT diagram are displayed. IRB, Institutional Review Board; PO, per os, that is, taken by mouth; TDS, three times a day.

Figure 2

Intention-to-treat symptom resolution analyses. (A) The cumulative incidence of total symptom resolution for both study arms as defined in the primary trial endpoint is plotted. The famotidine and placebo arms were compared using stratified log-rank test. (B) The logarithmically transformed patient-level total symptom score (thin lines) and their estimated means based on linear mixed effect model are shown for each study arm. The p value for the interaction term of group and day in study is displayed. (C) The estimated cumulative incidence of symptom resolution for each individual symptom at days 7, 14, 21 and 28 is displayed for each study arm. The results for diarrhoea are not included because neither arm had symptomatic patients at the displayed timepoints. All timepoints with no remaining symptomatic patient are displayed as 100% symptom resolution.

Figure 3

Effect of famotidine on inflammation and immunity. (A) The longitudinal plasma famotidine levels are displayed for patients enrolled in the famotidine arm and for the patient from the placebo arm with detectable plasma famotidine. (B) Participant numbers with detectable plasma interferon α levels in each arm at day 1 and day 7 of the trial are shown. Statistical comparison by χ² test. (C) Enrichment scores for type-I interferon response genes expressed in peripheral blood mononuclear cells (PBMCs) at days 1, 7, 14 and 28 are shown. (D) The correlation of symptom score and enrichment scores for type-I interferon is assessed. Statistical comparison by Spearman rank analysis. (E) The levels (mean and SE of the mean) of class g immunoglobulins reactive to the SARS-CoV-2 core protein are plotted for days 1, 7, 14 and 28 for each study arm. (F) The number of study participants with RT-PCR-detectable viral RNA extracted from nasal swabs on days 1, 7, 14 and 28 are shown for each trial arm. IFN: interferon, IgG: immunoglobulin type G.

Figure 4

Analysis of symptom score and type-I interferon signal correlation at patient category level. (A–F) The correlation of symptom score and enrichment scores for type-I interferon was assessed for the indicated group of patients. Patient counts for each group are displayed. Patients were able to self-assign to more than one race/ethnicity group, and the total patient count for panel C–F is therefore larger than 35, the total number of patients in the experimental medicine section of the trial. Statistical comparison by Spearman rank analysis. IFN: interferon.