Clinical Evidence for Improved Outcomes with Histamine Antagonists and Aspirin in 22,560 COVID-19 Patients

Introduction

The coronavirus disease 2019 (COVID-19) pandemic has driven great interest in the therapeutic potential of repurposed drugs with well-established gastroenterological benefits, including those that reduce acid production and secretion in the gastric mucosa.¹ Acid-reducing drugs belong to two main classes, based on their mechanisms of action (MOA): (i) Proton-pump inhibitors (PPIs) sterically block H⁺/K⁺-ATPase pumps, impeding the final step of acid release. (ii) Histamine H₂ receptor antagonists (H2RA) competitively bind the H2R, a type of G-protein coupled receptor (GPCR),² and block its natural stimulation by histamine; famotidine (e.g., Pepcid^®) and ranitidine (e.g., Zantac^®) are exemplary H2RAs.

An intricate and dense web of functional linkages exists between histamine and H2RAs, on the one hand, and disparate physiological pathways on the other hand. These pathways include gastrointestinal contexts (acid reduction) as well as the dysregulated inflammatory cascades (cytokine storm) that likely underlie much of the pathophysiology of COVID-19; these linkages have been reviewed recently. ^2,3 Here, we simply note that mounting cellular-scale evidence suggests that the mechanistic basis for a putative role of famotidine in COVID-19 likely involves the roles of H2RA versus, for instance, direct binding of famotidine to the viral protease 3CL^pro (and resultant inhibition)⁴, as was originally suspected from molecular docking studies.

Given the many conceivable mechanistic linkages alluded to above, is famotidine beneficial in treating COVID-19, as gauged by outcomes involving either (i) infection transmissibility, (ii) disease severity indicators (e.g., likelihood of cases reaching the point of ventilation, WHO severity index), or (iii) mortality rates? Unfortunately this question remains unresolved, though not for lack of effort: since a pioneering report⁵ of positive clinical outcomes with famotidine use in COVID-19, the past year has seen over 10 studies interrogate the potential therapeutic benefits of famotidine. Many reports concluded in favor of famotidine use (e.g., refs ^5-9), others found little to no association between famotidine (or PPIs) and measures such as 30-day mortality^10-12 (ranitidine, another H2RA, was scrutinized in one study¹³), and a recent study found a negative association for both PPIs and famotidine.¹⁴ Each of these independent studies were retrospective and observational; most were cohort-based, with some as case-series (e.g., symptom tracking across longitudinal data); most evaluated inpatient cases; and most attempted to account for confounders and other biases (e.g., via propensity-score matching). Given the conflicting reports thus far, particularly the evidence that suggests a beneficial impact of famotidine on mortality and overall disease progression (e.g., mechanical ventilation), we have undertaken a new analysis, reported herein.

We note that all three parallel tracks of findings—those indicating for and against famotidine, as well as neutral (i.e., no association)—rest upon substantially smaller datasets than were drawn upon in our present work. Are any beneficial effects of famotidine detectable on a population-wide, international scale? Is it synergistic to treat with famotidine in conjunction with aspirin, a general-purpose anti-inflammatory? Does famotidine use correlate with any measurable parameters that may serve as biomarkers, perhaps offering mechanistic clues (e.g., serum C-reactive protein [CRP] levels as a proxy for inflammation and the cytokine storm)? This work seeks to address these questions.

Methods

We retrieved data from the COVID-19 Research Network supplied by TriNetX, comprising ≈400M patients from 130 health care organizations in 30 countries. TriNetX provides a global federated health research network providing access to electronic medical records (diagnoses, procedures, medications, laboratory values, genomic information), and the TriNetX platform uses only aggregated counts and statistical summaries of de-identified information; no protected health information (PHI) or personal data is made available on the platform. This work was reviewed by our IRB board (UVA IRB tracking ID #23100), who determined that this project did not meet the criteria for Human Subject Research; therefore, no additional IRB submission/review was deemed necessary (by the IRB) to proceed with this project. We analyzed a cohort of 22,560 COVID-19 patients taking H₁/H₂receptor antagonists, with a special focus on 1,379 severe cases requiring respiratory support (see CONSORT flow diagram, Supplemental Figure 1). We defined death as the primary outcome, and, in order to try to mitigate confounder bias, we performed propensity score matching to achieve stratified and balanced sub-cohorts across age and gender; specifically, we balanced cohorts using a nearest-neighbor greedy matching algorithm with a caliper of 0.25 times the standard deviation. Measures of association, risk ratios (RRs) and odds ratios (ORs), along with their respective 95% CIs, were calculated. Kaplan-Meier survival curves were also computed for each analysis.

• Download figure
• Open in new tab

Supplemental Figure 1:

CONSORT flow diagram for this study.

• Download figure
• Open in new tab

Figure 1.

Kaplan-Meier survival curves are shown for COVID-19 patients with (blue) or without (red) the dual combination treatment of famotidine and aspirin.

Results & Discussion

We statistically analyzed outcomes for treatment with (i) the H1R antagonists loratadine (e.g., Claritin^®) and cetirizine (e.g., Zyrtec^®), (ii) the H2RA famotidine, (iii) the general-purpose medication aspirin, and (iv) a combination of famotidine & aspirin. For cases that reached the point of respiratory support, we found a significantly reduced fatality risk for famotidine treatment (OR 0.73, CI 0.57 to 0.94; Table 1, Supplemental Files 1-4). Hogan et al.⁶ recently suggested that dual-histamine receptor blockade, concurrently targeting both the H₁ and H₂ receptors, can improve COVID-19 clinical outcomes; however, significant improvements were not found in our cohorts, versus famotidine alone (OR 0.75, CI 0.39 to 1.46; Supplemental Files 5-8). Notably, and perhaps unexpectedly, the combination of famotidine and aspirin (344 severe cases before matching) did exhibit a significant synergistic survival benefit (OR 0.55, CI 0.39 to 0.78; Supplemental Files 9-12). The risk ratio for death decreased by 32.5%—an immense benefit, given the more than 2.6 million COVID-19–related deaths thus far.

Can these findings be reconciled with the recent studies (mentioned above) of famotidine and COVID-19? Janowitz et al.⁷ reported a case-series of 10 non-hospitalized COVID-19 patients for whom self-administration of famotidine had uniformly beneficial impact on disease trajectories, based on quantitative symptom-tracking across longitudinal data. Retrospective, single-center^5,6,8 studies also found promising results, in terms of reduced risk of clinical deterioration leading to intubation or death, for famotidine usage in 83 and 84 hospitalized COVID-19 patients, corresponding to 9.5 and 5.1% of the analyzed cohorts of patients, respectively.^5,8 Notably, both Freed-berg et al.⁵ and Mather et al.⁸ found lower levels of serum markers for severe disease (e.g., ferritin, C-reactive protein, or procalcitonin) in the famotidine group, consistent with our findings and with a potential role for this H2RA in attenuating cytokine release. Finally, a new systematic review and analysis (of published reports) suggests that famotidine may be beneficial¹⁵, while two other recent meta-analyses are either neutral or (statistically) inconclusive^16,17 as regards famotidine use in COVID-19.

If indeed famotidine is of therapeutic benefit in (most) COVID-19 cases, what might be its etiological basis? Beyond what is alluded to above, the molecular-scale MOAs that can be envisioned are covered in other recent work.¹⁸ Here, we simply reiterate that famotidine or other H2RAs may aid COVID-19 patients because of the capacity of such compounds to attenuate pro-inflammatory pathways that become dysregulated upon an infection (cytokine storms activate pro-fibrotic path-ways; lung damage eventually results). Thus, a role for famotidine in COVID-19 may stem from cellular mechanisms that are rather indirect and unrelated to this compound’s classic therapeutic role in gastroenterology.

As SARS-CoV-2 infection rates continue surging worldwide, we desperately need more data on potential therapies. The large, international, multi-center retrospective study reported here, sampling over 250,000 COVID-19 cases, hopefully helps clarify the potential benefit of clinically-approved histamine antagonists such as famotidine. We anticipate that at least three prospective, randomized, controlled clinical trials currently underway—NCT04504240, NCT04370262 and NCT-04545008—will illuminate famotidine’s potential therapeutic profile. Given the findings reported here, alongside the cost-effectiveness and mild/manageable side-effects of OTC drugs like famotidine and aspirin, we suggest that further prospective clinical trials—perhaps utilizing the aspirin combination reported here—are advisable.

Document information

Additional notes

The main text includes 1 figure, 1 table, and is accompanied by several items of Supplementary Material.

Funding

This work was supported by TRR295, KFO339, and DFG PR1562/1-1. Portions of this work were also supported by the University of Virginia School of Data Science and by NSF CAREER award MCB-1350957.

Patient consent for publication

Not required; see Methods section for IRB review.

Supplementary Materials

The following supplementary figures, tables, and data files accompany this manuscript:

• Supplemental Figure 1: CONSORT Flow Diagram

• Supplemental Table 1:Lab Values and Standard Deviations for Serum Levels of C-reactive Protein

• Supplemental File 1: Measures of Association Data Graph for Famotidine

• Supplemental File 2: Measures of Association Data Table for Famotidine

• Supplemental File 3: Kaplan-Meier Raw Data Graph for Famotidine

• Supplemental File 4: Kaplan-Meier Raw Data Table for Famotidine

• Supplemental File 5: Measures of Association Data Graph for H1 and H2

• Supplemental File 6: Measures of Association Data Table for H1 and H2

• Supplemental File 7: Kaplan-Meier Raw Data Graph for H1 and H2

• Supplemental File 8: Kaplan-Meier Raw Data Table for H1 and H2

• Supplemental File 9: Measures of Association Data Graph for Famotidine and Asp

• Supplemental File 10: Measures of Association Data Table for Famotidine and Asp

• Supplemental File 11: Kaplan-Meier Raw Data Graph for Famotidine and Asp

• Supplemental File 12: Kaplan-Meier Raw Data Table for Famotidine and Asp