Genetic validation of the use of tocilizumab, statins and dexamethasone in COVID-19

  1. CM Schooling
  2. SL Au Yeung
  3. MK Kwok
  4. JV Zhao

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background

New means of treating COVID-19 are urgently needed. Genetic validation of drugs can foreshadow trial results, and help prioritize investigations. We assessed whether common drugs, suggested as possible treatments for COVID-19 (tocilizumab, anakinra and statins) with established genetic proxies, are effective in COVID-19. We also included dexamethasone as a positive control exposure because the RECOVERY trial suggested benefit in severe COVID-19.

Methods

We assessed, using Mendelian randomization, whether genetic proxies of tocilizumab, anakinra, statins and dexamethasone use affected risk of very severe (cases=536, non-cases=329391) or hospitalized (cases=3199, non-cases=897488) COVID-19 using a recent genome-wide association study.

Results

Using rs2228145 ( IL6R ) to proxy effects of tocilizumab use, no association with very severe COVID-19 was found, but possibly an inverse association with hospitalized COVID-19 (odds ratio (OR) 0.83 per standardized effect of higher soluble interleukin-6r, 95% confidence interval 0.67 to 1.02). Using rs12916 ( HMGCR ) to proxy effects of statins use, an inverse association with very severe COVID-19 was found (OR 0.30 per standardized effect, 95% CI 0.10 to 0.89). Using rs6743376 and rs1542176 to proxy effects of anakinra use, no associations with COVID-19 were found. Dexamethasone, instrumented by cortisol, was possibly inversely associated with very severe COVID-19 (OR 0.20 per standardized effect 95% CI 0.04 to 1.04).

Conclusion

Our study provides some genetic validation for the use of both tocilizumab and statins in COVID-19, but not anakinra, whilst being consistent with the findings from the RECOVERY trial about dexamethasone. Investigation of the underlying mechanisms might facilitate re-purposing and development of effective treatments.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.07.09.20149450: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Institutional Review Board StatementConsent: This study of publicly available genetic summary statistics, obtained with consent, does not require ethical approval.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variablenot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    We used the MendelianRandomization R package to obtain univariable estimates, the MR-Base ld_clump R package to obtain independent genetic variants in the multivariable MR and the MVMR R package to obtain multivariable MR estimates, F-statistic and Q-statistic.
    MendelianRandomization
    suggested: None

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Despite providing information that may be relevant to ameliorating the COVID-19 pandemic, this study has limitations. First, valid genetic instruments for MR should fulfil three assumptions, i.e., relate strongly to the effects of drug use, not be associated with potential confounders of the effects of drug use and satisfy the exclusion restriction assumption, i.e., only affect COVID-19 via the drug in question. The genetic variants used to predict effects of tocilizumab, anakinra and statins were originally selected as functionally relevant to the corresponding phenotype, i.e., sIL-6r, IL-1Ra and LDL-cholesterol, in genes that reflect each drug’s specific mechanism of action (IL6R, IL1F10, and HMGCR, respectively).6-8 28 As such, the genetic effect of the drug, encompassing the mechanism of action may extend beyond the target phenotype. Statins, for example, are well-known to have a wide range of effects beyond lipid modulation.8 Both effects via the target phenotype as well as “off-target” effects may contribute to a drug’s success in a new indication. MR studies can be confounded by population stratification. However, we used genetic associations from GWAS mainly comprising people of European ancestry (https://www.covid19hg.org/results/). We also rely on the integrity of the design of the underlying genetic studies to be free from bias, particularly that the non-cases come from the same underlying population as the cases, which is achieved by largely using existing cohort ...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2020.07.09.20149450v1 on medRxiv