Low neutralizing antibody titers after asymptomatic or non-severe SARS-CoV-2 infection over a 6-month assessment period.

Dear Editor:

In their recent publication, Hanrath and co-workers describe protection against SARS-CoV-2 re-infection conferred by prior SARS-CoV-2 infection, as assessed by PCR over 6 months in healthcare workers (HCW) with a prior history of COVID-19.[1]

Neutralizing antibodies against SARS-CoV-2 are a key determinant of immunity following infection [2,3]. Data in convalescent patients who experienced moderate to severe disease suggest that this immune response may last for months, and generally correlates with disease severity 4, 5, 6. However, serological responses in patients with asymptomatic or mild SARS-CoV-2 infections are less well defined [7].

We characterized serological responses to SARS-CoV-2 in asymptomatic or non-critical SARS-CoV-2 infections over a 6-month period. Specifically, we monitored the magnitude and kinetics of serological responses in 335 HCW found to be seropositive for SARS-CoV-2 using a Luminex-based technology targeting the trimeric spike protein [8]. This cohort was composed of 198 (59%) seropositive HCW randomly selected throughout our institution[9] without a prior SARS-CoV-2 polymerase chain reaction positive test (PCR+) and 137 (41%) seropositive HCW recruited on the basis of a prior SARS-CoV-2 PCR+ test. SARS-CoV-2-specific antibody response was measured at baseline (end of May 2020), 3- and 6 months thereafter using the luminex-based quantitative anti-spike trimer protein assay to determine quantitative antibody responses, and a cell-free neutralization assay based on the competitive inhibition of trimeric SARS-CoV-2 spike protein binding to the angiotensin-converting enzyme 2 (ACE2) receptor to determine neutralizing capacity of the antibodies developed [8,10]. Importantly, in this neutralizing activity assay a value of 50 percent is predictive of robust neutralization capacity in cell assays [10]. At each time-point, serological assessments were paired with a survey to assess symptoms, risk of exposures or PCR documentation of repeat SARS-CoV-2 infection. Of note, the 6-month period (June to December 2020) coincided with the tail-end of the first COVID-19 wave in Switzerland (at baseline), a very low SARS-CoV-2 infection period (3-month visit) and the beginning of a second COVID-19 wave (6-month visit).

Median range of anti-spike trimer immunoglobulin G (IgG) titers (expressed as mean fluorescence intensity relative to control) decreased from 35 (baseline) to 20 (3-month) and 22 (6-month) (Linear mixed model for the IgG titers across time: p < 0.001). However, median range of neutralizing activity (expressed as inverse of the serum dilution required for 50 percent inhibition) was 40, 44 and 53 at 0-, 3- and 6-months, respectively (Fig. 1 a). Compared with baseline, the mean neutralization activity was 1.34 and 1.62 times higher at 3- and 6-months, respectively (Linear mixed model for the neutralizing activity across time: p < 0.001).

Fig. 1

Evolution of Anti-S IgG titers and neutralization capacity in HCWs across time (A): HCWs assayed for Anti-S protein (red) expressed in mean fluorescence intensity and mean neutralization values (green) expressed in percent of inhibition at 0, 3 and 6 months: 335/334, 302/289 and 294/294 HCWs Evolution of median neutralization activity in HCWs according to disease severity (B). A total of 334 SARS-CoV-2 seropositive HCWs had recorded neutralizing activity at baseline with 289 and 294 presenting for follow-up visits at 3- and 6-months, respectively. Participants were segregated according to disease severity at baseline, as per NIH classification (asymptomatic – blue, mild [myalgia, anosmia, cough, ageusia, fever, sore throat, diarrhea, common cold but without shortness of breath] – purple, moderate to severe [shortness of breath with or without hospitalization] – green). Units of neutralizing activity are expressed as inverse of the serum dilution required for 50 percent inhibition. Distribution according to disease severity at 0/3/6 months: Asymptomatic 105/91/88 HCWs, Mild 139/125/120 HCWs, moderate-severe 90/73/86 HCWs.

Anti-spike IgG titers and neutralizing activity were stratified according to the severity of the clinical manifestations using the NIH COVID-19 classification. Among 334 participants with available neutralizing activity at baseline, 105 (31%) had asymptomatic infection, 139 (42%) had mild infection and 90 (27%) had moderate to severe infection that did not require ICU admission. At baseline, median range of neutralizing activity was significantly lower for the 105 asymptomatic vs. 229 symptomatic participants (median 8 vs. 63, respectively; p < 0.001 Mann-Whitney-U). Thus, the neutralizing activity of sera from asymptomatic participants was very weak.[10] Specifically, the median neutralizing activity of asymptomatic participants was significantly lower compared to those with mild infection (median 8 vs. 59; p < 0.001) and those with moderate-severe infection (median 8 vs. 71; p < 0.001). Albeit not statistically significant, the median neutralizing activity was higher in HCW with moderate-severe infection than in those with mild infections, suggesting an association between levels/titers of neutralizing antibodies and COVID-19 clinical severity (median 59 vs. 71; p = 0.056), adjusting for multiple testing (Benjamini-Hochberg) (Fig. 1b). Neutralizing activity was sustained over time within each severity category, with the same trend being present at 3- and 6-months.

The survey assessed the rate of re-infection in participants post-baseline. No participant reported a SARS-CoV-2 PCR+ test at 3-months. In contrast 9 (3%) participants reported PCR positivity at 6-months. Out of 9 re-infected participants, the initial SARS-CoV-2 infection was asymptomatic in 5, mild in 3 and moderate/severe in one participants respectively. These participants had a significant rise in neutralizing activity at the last time point (median 5, 13 and 180 at baseline, 3-months and 6-months, respectively; p < 0.001), strongly supporting the likelihood of re-infection. Albeit not formally excluded in the absence of a comparison of SARS-CoV-2 sequence data, the possibility of persistent viral shedding is unlikely considering the time lapse and temporal association of symptomatic disease.

While our results support the finding of Hanrath et al. that prior infection with SARS-CoV-2 can protect against re-infection, we observe that re-infection risk varies according to initial infection severity.[1] Consistent with recent data, we find that antibody titers can be influenced by infection severity. 4, 5, 6 However, in this cohort of seropositive HCWs, participants with asymptomatic and mild SARS-CoV-2 infections had relatively lower antibody titers and neutralizing activity and they experienced higher rates of re-infection. We acknowledge that the subset of re-infected participants is small (possibly due to the timing of the survey) and results are only hypothesis generating. Nevertheless our findings strengthen the recommendation to boost by vaccination persons who are seropositive following natural infection with SARS-CoV-2, particularly those who experienced asymptomatic or mild infections in light of the low overall neutralizing activity observed over time in this subgroup.

The study was approved by the local institutional review board (Research Ethics Committee of the Canton de Vaud - CER-VD, Switzerland, Req-2020–00990).

Acknowledgements: Special thanks to our sponsors Philippe Eckert and Oliver Peters. Patricia Berdoz, Anne Jacot, Lyne Bischoff, Justine Gendroz, Sandrine Picon, Laurence Posset and the nursing staff of the Occupational Health of Lausanne University Hospital. Many thanks to the service of logistics for its assistance in shuttling the samples. Thanks to Isabelle Guilleret, Fady Fares and Vassily Soumas of the Clinical Trial Unit for assistance.

All authors have completed the Unified Competing Interest form (available on request from the corresponding author) and declare: no support from any organization for the submitted work [or describe if any]; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years [or describe if any], no other relationships or activities that could appear to have influenced the submitted work [or describe if any].