key: cord-0791988-31n93brl
authors: Saadat, Mostafa
title: COVID-19 prevalence and mortality is associated with the allele frequency of CCR5-Δ32
date: 2021-06-03
journal: Croat Med J
DOI: 10.3325/cmj.2021.62.303
sha: 837bd99e3aa4d51c0ecd3407ff4c81495e3371dd
doc_id: 791988
cord_uid: 31n93brl

nan

To the Editor: I read with great interest the article Does the CCR5-Δ32 mutation explain the variable coronavirus-2019 pandemic statistics in Europe? by Starčević Čizmarević et al (1) . The authors found no significant association between COVID-19 prevalence/mortality and the CCR5-Δ32 allele frequency in 39 European countries. As mentioned by the authors, European countries share a relatively similar genetic background. Although the prevalence and mortality of COVID-19 differ across European countries, these epidemiologic parameters vary even more between European and non-European countries. The CCR5-Δ32 frequency in European populations is higher than in Asian, especially East-Asian, populations. To determine whether COVID-19 prevalence/mortality follows the geographical distribution of CCR5-Δ32 worldwide, I extended the analysis performed by Starčević Čizmarević et al to 82 world countries.

COVID-19 prevalence/mortality and the number of performed diagnostic tests (per 10 6 people, as of end of December, 2020) were obtained from the Worldometer web-site (www.worldometers.info/coronavirus/countries). The CCR5-Δ32 frequency was obtained from a previously published article (2). The Human Development Index (HDI) value, reflecting three major dimensions of human development: life expectancy at birth, education, and the gross national income per capita, was used as a potential confounder. Another potential confounder was the number of performed diagnostic tests (Supplementary Table 1 ). While HDI showed normal distribution, COVID-19 prevalence/ mortality, the number of preformed diagnostic tests, and CCR5-Δ32 frequency deviated from the normal distribution and were square root-transformed (SR-transformed).

SR-prevalence (r = 0.516, df = 80, P < 0.001) and SR-mortality (r = 0.456, df = 80, P < 0.001) were significantly associated with the SR-CCR5-Δ32 frequency. To account for the effect of confounding socio-economic factors on COVID-19 prevalence/mortality, multivariable linear regression analysis was used. Table 1 shows the final multivariable models constructed using a backward elimination procedure. In the model, SR-prevalence was significantly positively associated with SR-CCR5-Δ32 frequency (partial r = 0.265, P = 0.017). SR-transformed mortality was positively associated with CCR5-Δ32 frequency (partial r = 0.216), but the difference did not reach significance (P = 0.053). This means that countries with a high frequency of CCR5-Δ32 allele had a higher COVID-19 prevalence and mortality than countries with a low CCR5-Δ32 frequency. The current findings reveal that the frequency of CCR5-Δ32 mutation can partially explain the difference in COVID-19 prevalence/mortality between populations.

Does the CCR5-Δ32 mutation explain the variable coronavirus-2019 pandemic statistics in europe?