Evidence for treatment with estradiol for women with SARS-CoV-2 infection

The consolidated standard of reporting trial (CONSORT) flow diagram, shown in Fig. 1, illustrates the data extraction process from the TriNetX Real-World database. The full cohort size of n = 68,466 patients consists of 37,086 women and 29,609 men who are older than 15 years (database accessed on July 16, 2020) and who are SARS-CoV-2 infected. After excluding (i) boys and girls up to 14 years of age, (ii) all men, and (iii) those without gender information, a sub-cohort of 37,086 women was used for further analysis of sex hormone status.

Regarding sex and gender, analyzing the data in terms of 5-year age strata, the incidence rates of women and men with known SARS-CoV-2 infection can be seen to significantly differ by sex (Fig. 2) (t test, p < 0.01). The highest SARS-CoV-2 infection numbers among women, compared to age-matched men, is observed in women with what would be expected to be high levels of serum estradiol—i.e., between 15 and 49 years of age. The incidence of COVID-19 in women in this pre-menopausal age group is about + 15% higher versus age-matched men (n = 21,229 women vs. n = 15,918 men). Starting with 60 and up to 85 years of age, the data show a relative decrease in frequency among women (n = 7334 women vs. n = 7171 men). In older women and men (age strata 65 to 80), the statistical susceptibility to viral infection converges (see the closely-matched red and blue pairs of bars in Fig. 2, in the range of 65–85 years old). Near the end of life, defined as 85+ years old, the incidence of SARS-CoV-2 infection is higher in women (n = 5865) than in men (n = 5406). At the opposite end of the age spectrum, all pre-adolescents (girls and boys) have similar risks of infection.

This study focuses on the incidence and outcome of COVID-19 infections by considering an age- and sex-disaggregated data analysis. We identified a sex-specific distribution of COVID-19 incidence rates, with the highest frequencies being among premenopausal women in the 20–55-year age range. We also found a higher fatality rate of men compared to age-matched women, beginning at 50 years of age, and that estradiol hormone use reduced fatality rates for women in this 50+ age range.

The patterns of incidence rates of women and men with SARS-CoV-2 infection, measured using 5-year age strata, differs from the trends in the number of deaths following the same age strata (Fig. 3). This age-dependent, sex- and gender-based discrepancy might hint at different factors as contributing to the differential risks of infection and COVID-19 fatality among women versus men; note that these other factors may be either intrinsic (e.g., physiological/mechanistic) or extrinsic (e.g., smoking, lifestyle patterns or other circumstances) in nature. Gebhard et al. recently published the first sex-disaggregated data of confirmed COVID-19 cases and deaths, provided by Global Health 50/50 data tracker. Data from China and Europe were evaluated. Regardless of country-specific demographics, case fatality in men was higher than in women for all age groups and was more evident in middle age [17]. Similarly, Klein et al. reported in the USA a significant male-female difference in COVID-19 cases, hospitalizations, and deaths [18].

The occurrence of most cases of SARS-CoV-2 infection within the reproductive ages—i.e., before age ~ 50 (see Fig. 2 and the dip towards the right-hand side)—hints at an association with relatively higher estrogen levels. General features of the datasets used in our large international COVID-19 cohort, comprising n = 68,466 cases, are consistent with the public data for the incidence of coronavirus infection and the fatality rate for COVID-19 disease in Germany, as presented by the Robert-Koch Institute (RKI) website [19]. In our study, the incidence of coronavirus infection in young women is about + 15% higher than age-matched men; however, the fatality rate is about + 50% higher in men than women.

The data in our present study indicate that pre-menopausal women are disproportionately more infected with coronavirus than men in the same age brackets, but they do not become as seriously ill, as evidenced by lower fatality rates; we believe this to be an interesting observation for sex and gender medicine experts.

Sex- and gender-specific epidemiological differences, with pathophysiological bases, have been published in the recent past, pointing to potential mechanistic roles for the sex hormones. However, the X-chromosomal location of the ACE2- and the AT2-receptor genes—and the downregulation of the pleiotropic cytokine interleukin 6 (IL-6) by estrogens and its upregulation by androgens—should also be considered, along with other biochemical and physiological factors to which sex-correlated discrepancies might be attributed [20]. A central molecular component here is human ACE2, which is known to interact strongly with the SARS-CoV-2 viral spike protein [11]; together with the “transmembrane protease, serine 2” (TMPRSS2) enzyme, ACE2 plays an essential role in viral entry of SARS-CoV-2 into host cells [6].

Pre-menopausal women are more protected from cardiovascular, lung, and kidney diseases than their age-matched male peers. Recently, it was shown that the estrogen-mediated upregulation of the Mas-receptor contributes to the prevention of acute lung injury and affects endothelial barrier stabilization [21]. A relative protection of females over males has also been observed in other studies with experimental animal models of acute lung injury [22]; intriguingly, this protection was lost in ovariectomized mice, but restored upon estrogen replacement [23]. The 17β-estradiol molecule attenuates lung vascular permeability and edema [24], and estrogens reduce the pulmonary vasoconstriction during hypoxia by increasing levels of prostacyclin and nitric oxide (NO) [24].

The ACE2 and AT2R genes map to the X chromosome. In genetic terms, this sex-linkage would allow women to be heterozygous and differently assorted compared to men, who are hemizygous [25]. The second X chromosome is not inactivated in approximately 15% of genes and another 15% of genes vary between individuals or tissues in whether they are subject to, or escape from, inactivation [26]. This may account for some of the differences that are seen between men and women (sexual dimorphism) and could be a reason for higher expression levels of ACE2- and ATIIR proteins in women (i.e., a gene dosage effect).

Another pertinent mechanistic pathway, potentially triggered or modulated by the types and levels of sex hormones, could be immunological—namely, sex-specific differences in the immune response to SARS-CoV-2 infection. A recent study has linked higher mortality among men to a “cytokine storm,” which in turn closely relates to the severity of symptoms such as pulmonary edema, fibrosis, and other deleterious downstream effects associated with acute lung injury [27]. An individual’s immune response to viral infections can vary with fluctuations in sex hormone concentrations, including steroid hormones such as estrogens, progesterone, and testosterone. Each immune cell type is influenced by all three of these major sex hormones, albeit with different and partly opposing effects. From a biomedical perspective of sex, interleukin-6 (IL-6) is particularly interesting. IL-6 is a cytokine with both anti- and pro-inflammatory effects. It can be produced by almost all stromal and immune system cells (e.g., monocytes, lymphocytes, macrophages, endothelial cells, mast cells, dendritic cells), and it is believed to play a central role in precipitating the cytokine storms that, in turn, yield such severe symptoms from a SARS-CoV-2 infection [28]. Notably, the endogenous female sex hormone 17β-estradiol blocks the IL-6 cytokine pathway, while the male sex hormones (the androgens) increase IL-6 production [20]. We again employed the COVID-19 cohort to evaluate the impact of exogenous estradiol sex hormonal use, either for contraceptive purposes or for postmenopausal symptoms, on COVID-19 fatality.

Among post-menopausal women, we observed a significant difference in the rates of death between women with regular estradiol use (user group) and those without estradiol sex hormone intake (non-user group). To the best of our knowledge, this important finding—that the fatality risk for women > 50 years receiving estradiol therapy (user group) is reduced by more than 50% (OR 0.33, 95% CI [0.18, 0.62] and hazard ratio 0.29, 95% CI [0.11,0.76]) compared to the non-users group—is described here for the first time. Furthermore, our data analysis included distinct sub-groups of young women (15–49 years of age) with and without oral contraceptives. The dominant form of estrogen used for “the pill” is ethinylestradiol. The magnitude of the protective effect of the usage of oral contraception, compared to the non-users sub-population, is smaller among younger versus older women; this may be the case because endogenous estradiol levels are typically already higher in younger women than for post-menopausal women, thus drowning out any differences between user/not-user groups. Moreover, the level of exogenous hormone intake for purposes of contraception is generally less than that used for post-menopausal estradiol hormone therapy.

Note that some current trials are testing the effect of sex hormones (estrogen and testosterone) on COVID-19 outcomes. A brief, 7-day course of estradiol, delivered via a transdermal patch, could be a safe approach to reduce symptom severity in adult men and in older women, when administered prior to intubation (ClinicalTrials.gov; Identifier: NCT04359329). The time-course for estradiol treatment may need to be evaluated for its positive effects on lung protection, and whether it could be an effective therapeutic approach not only for women but also for men with COVID-19. This type of question cannot be answered solely by the data presented here, and instead addressing this question is a key topic for further research.

Based on the main findings of our present study, we believe there are no concerns for continuing the use of sex hormones that contain estradiol prior to SARS-CoV-2 infection. Even though it can be seen in the data that the risk of infection is higher in pre-menopausal women with higher endogenous estradiol levels, compared to either men of the same age strata or to post-menopausal women, the clinical course of COVID-19 disease, and the ultimate mortality rate, is lower in women with higher estradiol levels. Higher survival probabilities are particularly evident in post-menopausal women who are infected with SARS-CoV-2 and who regularly use exogenous estradiol (e.g., for postmenopausal complaints).

Pre-menopausal women are at a relatively high risk for SARS-CoV-2 infection, but the survival probability in this below-50 age range is significantly higher in women than in men. A chief finding of this study is the strong positive effect of regular estradiol hormone therapy on the survival rates of post-menopausal women.