Relationship between nasal SARS-CoV-2 viral load and endemic coronavirus antibody levels
To explore the potential basis for the higher antibody response to SARS-CoV-2 in patients with undetectable or lower SARS-CoV-2 viral load, the levels of antibodies to the S protein of endemic coronaviruses were measured (Figure 3 ). First, we observed that anosmic and normosmic patients had similar levels of nasal and serum IgG and IgA to the S protein of 229E, OC43, HKU1 and NL63 viruses, except for serum levels of IgG to 229E that were slightly lower among anosmic patients(Figure 3a) . We then observed that antibody levels to the four endemic coronaviruses were comparable in SARS-CoV-2 PCR positive and negative patients, whether anosmic or normosmic (Figure 3b) . A moderate negative correlation was observed between SARS-CoV-2 viral load and nasal levels of IgG to OC43 and HKU1 but no other correlations were observed (Figure 3c) . Together these data indicate that overall antibody levels to endemic coronaviruses are not associated with nasal SARS-CoV-2 viral load in our study population. Immunity to endemic coronaviruses can therefore not explain the association between SARS-CoV-2 viral load and SARS-CoV-2-specific antibodies. However, the fact that OC43 and HKU1 have the closest homology with SARS-CoV-2 among endemic coronaviruses suggests that the moderate negative correlation observed between SARS-CoV-2 viral load and serum levels of IgG to these two coronaviruses may indicate some role for cross-reactive immunity in the control of SARS-CoV-2 replication in anosmic patients.
To the best of our knowledge, this is the first study showing an inverse association between nasal SARS-CoV-2 viral load and the levels of antibodies. The specificity of this observation in patients with COVID-19 OD suggests an interaction with other immune components involved in the pathogenesis of the olfactive disorder. The data suggests that boosting local immunity to SARS-CoV-2 may be beneficial to patients with OD and may reduce the duration of symptoms. Systemic vaccination with currently available vaccines provides limited mucosal immunity. The development of mucosal vaccines may offer opportunities for the prevention of local symptoms, in addition to the reduction of SARS-CoV-2 shedding and transmission20 .
The relatively large number of patients and the prospective design are the main strengths of the study. Moreover, patients were recruited and followed-up within relatively narrow time windows, increasing the likelihood of measuring dynamic virological and immunological events at similar times between patients. Some limitations can be mentioned as well. Measurements were performed at a single time point, preventing analysis of the dynamics of the relationship between viral replication and antibody responses. As explained above, the study was conducted from October to November 2020 (second wave of SARS-CoV-2 in Europe) when the dominant strain of SARS-CoV-2 was the Wuhan strain. The interactions between variants of concern and antibodies at the mucosal level could be significantly different, limiting the potential extrapolation of our study results to the current stage of the pandemic. The lack of fully objective methods to assess the olfactory dysfunction is another limitation. The use of Identification Sniffin’Sticks Test (16 items) and not the complete TDI test was made according to the quarantine restrictions of physician-patient contact in the first waves of the pandemic. We are however well aware that this identification test is less effective in the objective analysis of smell disorders.