Review 1: "SARS-CoV-2 Antibodies Cross-React and Enhance Dengue Infection"

RR:C19 Evidence Scale rating by reviewer:

• Misleading. Serious flaws and errors in the methods and data render the study conclusions misinformative. The results and conclusions of the ideal study are at least as likely to conclude the opposite of its results and conclusions than agree. Decision-makers should not consider this evidence in any decision.

***************************************

Review:

This manuscript by Jakhar and colleagues reports a study that suggests the presence of cross reactivity between anti-SARS-CoV-2 spike (S) antibodies with dengue virus-2 (DENV-2) E protein and that the resultant cross reactivity enhances DENV-2 infection. The authors used a combination of experimental animal infection and convalescent sera from Covid-19 subjects to show the presence of cross-reactive antibodies and enhancement of DENV-2 infection in myeloid-derived cell lines, in vitro. They suggest that SARS-CoV-2 infection could elevate the risk of severe dengue and thus explain the rising trend of dengue incidence in various Asian countries. 

The main problem with antibody-dependent enhancement of DENV infection (ADE) is that the in vitro findings do not correlate with clinical observations. This issue has been reviewed recently (Ooi and Kalimuddin, Sci Transl Med 2023). In vitro studies must thus be grounded by unequivocal epidemiological and clinical evidence of increased rates of severe dengue following convalescence from SARS-CoV-2 infection. If not, any findings could be mere in vitro phenomenon. In such a context, this manuscript falls short on several key areas. Major concerns are as follows:

1. Lines 59-66: The numbers are not only misleading, they are also inaccurate. Comparing year to year changes in dengue cases should always be avoided as dengue incidence follows a cyclical trend. Peak incidence occurs every 3-7 years, depending on the force of infection and thus varies from country to country. They are also inaccurate as dengue incidence declined (rather than increased as described in these lines) in 2021 compared to 2020 both in Singapore (Ho et al, PLoS Negl Trop Dis 2023) and in Malaysia (Iderus et al, Front Public Health 2023). The authors have thus provided no convincing epidemiological trends to suggest that convalescence from SARS-CoV-2 infection was associated with increased risk of dengue, let alone severe dengue.

2. Another major limitation of this study is the lack of specificity of the observed DENV-2 infection enhancement. If indeed the goal of this study was to show that anti-S antibodies from SARS-CoV-2 infection enhanced dengue, then a useful negative control is the inclusion of antibodies and sera produced from seasonal human CoV infection. The use of rabies anti-sera or even just isotype IgG control is insufficient to show specificity needed to support the conclusion.

3. The serum samples used in this study lack history on prior DENV infection. The cross reactivity and infection enhancement observed could simply be explained by prior DENV infection. This is particularly likely since these subjects were recruited in India, which has experienced regular epidemics of dengue.

4. Lines 133-137: The authors tried to use the rates of K562 infection by DENV-2 when reacted with convalescent sera to suggest rising trend of ADE. The description of the results and the statistical tests are both misleading. The comparison should not be with DENV-2 only control. Instead, to suggest a trend in rising prevalence of enhancing antibodies, the authors should have compared pre- to post-covid samples. In fact, there is no statistically significant difference between the rates of K562 infection with samples obtained from either the delta wave or omicron wave with the pre-delta wave.