Humoral Immune Response to SARS-CoV-2 in Iceland

Ethical Considerations

The study was approved by the National Bioethics Committee of Iceland. The Health Care sample collection was performed on behalf of Icelandic health authorities in agreement with the Act no. 19/1997 on Health Security and Communicable Diseases. Participants who were part of the other sample collections provided written informed consent.

Antibody Measurements

We measured SARS-CoV-2–specific antibodies in up to 30,576 persons with six established assays, targeting pan-immunoglobulin (pan-Ig: IgM, IgG, and IgA) antibodies against the nucleoprotein (N) (Roche); pan-Ig antibodies against the receptor binding domain (RBD) in the S1 subunit of the spike protein (pan-Ig anti–S1-RBD) (Wantai); IgM and IgG antibodies against N (IgM anti-N and IgG anti-N) (EDI/Eagle); and IgG and IgA against the S1 subunit of the spike protein (IgG anti-S1 and IgA anti-S1) (Euroimmun). Thresholds for positivity were supplied by the assay manufacturers. We used the two pan-Ig antibody assays to evaluate seroprevalence, requiring positive results for both assays for a test result to be considered positive (Fig. S1 in Supplementary Appendix 1). To quantify antibody levels among qPCR-positive persons, we assayed antibodies against SARS-CoV-2 using IgG anti-N, IgM anti-N, IgG anti-S1, and IgA anti-S1.

Sample Collection

We measured antibodies in two groups of qPCR-positive Icelanders and in six groups who had not been qPCR-tested or who had been tested and had received a negative result (Figure 1). We collected samples from a group of hospitalized qPCR-positive persons and invited all qPCR-positive persons who had recovered from infection to donate samples, both shortly after recovery and again approximately 3 months after recovery (a total of 2102 samples from 1237 persons). We used two groups of samples collected before the pandemic (in 2017 and in early 2020) to evaluate assay specificity and to determine when the pandemic reached Iceland. We collected samples from quarantined persons who had not tested qPCR-positive to evaluate infection during quarantine and the effect of exposure type on the probability of infection. We used three groups of samples collected from persons who had neither tested qPCR-positive nor been quarantined to evaluate seroprevalence outside quarantine and the spread of the virus in Iceland (the Health Care, Reykjavik, and Vestmannaeyjar sample groups, totaling 23,452 persons).

Estimation of Infection Rate

The largest of these six groups, the Health Care group, was enriched for older people. To estimate seroprevalence, we weighted this sample by region, sex, and age in the population (see Supplementary Appendix 1). To estimate the number of infected Icelanders, we added together the number of qPCR-positive persons, the number of quarantined persons times the estimated seroprevalence in this group, and the number of persons outside quarantine times the estimated seroprevalence outside quarantine. We estimated the percentage of Icelanders infected by dividing the number of infected persons by the number of Icelanders. We estimated the infection fatality risk by dividing the number of deaths from Covid-19 by the number of infected persons.

Antibody Levels, Age, Sex, and Clinical Characteristics

We tested for associations of age, sex, preexisting conditions (27 phenotypes), and clinical outcome (35 characteristics) with antibody titers (for each of the six assays) in the most recent samples obtained from persons in the Recovered group. We recoded categorical clinical characteristics with their ordinal number in the analysis.

Statistical Analysis

We used a likelihood ratio method to calculate confidence intervals of fractions with the Clopper–Pearson exact method when the estimated fraction was 0 or 1. To test for association between each clinical characteristic and antibody levels, we performed multiple regression analyses with the phenotype as a covariate and quantile normalized antibody levels as a response, adjusting for age, age squared, sex, and time since qPCR diagnosis, excluding the age and sex covariates when testing for association with age and sex, respectively. We quantile-normalized the antibody levels by ranking the levels and transforming them, using the inverse normal transform of the rank divided by one plus the number of observations. Effects estimates were reported in terms of standard deviations of antibody levels. We derived P values and confidence intervals from standard errors estimated by the multiple regression. We used Bonferroni correction to determine significance, with a threshold for significance of P<[0.05÷6(2+27+35)]=0.00013. For effects of the exposure type on the probability of infection among quarantined persons, we used logistic regression to estimate the confidence intervals of odds ratios. We did not adjust confidence intervals for multiple testing.

Specificity of SARS-CoV-2 Antibody Assays

Both assays measuring pan-Ig antibodies had low numbers of false positives among samples collected in 2017: there were 0 and 1 false positives for the two assays among 472 samples, results that compared favorably with those obtained with the single IgM anti-N and IgG anti-N assays (Table S3). Because of the low prevalence of SARS-CoV-2 infection in Iceland, we required positive results from both pan-Ig antibody assays for a sample to be considered seropositive (see Supplementary Methods in Supplementary Appendix 1). None of the samples collected in early 2020 group were seropositive, which indicates that the virus had not spread widely in Iceland before February 2020.

SARS-CoV-2 Antibodies among qPCR-Positive Persons

Figure 2. Figure 2. Antibody Prevalence and Titers among qPCR-Positive Cases as a Function of Time since Diagnosis by qPCR.

Shown are the percentages of samples positive for both pan-Ig antibody assays and the antibody titers. Red denotes the count or percentage of samples among persons during their hospitalization (249 samples from 48 persons), and blue denotes the count or percentage of samples among persons after they were declared recovered (1853 samples from 1215 persons). Vertical bars denote 95% confidence intervals. The dashed lines indicated the thresholds for a test to be declared positive. OD denotes optical density, and RBD receptor binding domain.

Table 1. Table 1. Prevalence of SARS-CoV-2 Antibodies by Sample Collection as Measured by Two Pan-Ig Antibody Assays.

Twenty-five days after diagnosis by qPCR, more than 90% of samples from recovered persons tested positive with both pan-Ig antibody assays, and the percentage of persons testing positive remained stable thereafter (Figure 2 and Fig. S2). Hospitalized persons seroconverted more frequently and quickly after qPCR diagnosis than did nonhospitalized persons (Figure 2 and Fig. S3). Of 1215 persons who had recovered (on the basis of results for the most recently obtained sample from persons for whom we had multiple samples), 1107 were seropositive (91.1%; 95% confidence interval [CI], 89.4 to 92.6) (Table 1 and Table S4). Since some diagnoses may have been made on the basis of false positive qPCR results, we determined that 91.1% represents the lower bound of sensitivity of the combined pan-Ig tests for the detection of SARS-CoV-2 antibodies among recovered persons.

Table 2. Table 2. Results of Repeated Pan-Ig Antibody Tests among Recovered qPCR-Diagnosed Persons.

Among the 487 recovered persons with two or more samples, 19 (4%) had different pan-Ig antibody test results at different time points (Table 2 and Fig. S4). It is notable that of the 22 persons with an early sample that tested negative for both pan-Ig antibodies, 19 remained negative at the most recent test date (again, for both antibodies). One person tested positive for both pan-Ig antibodies in the first test and negative for both in the most recent test.

The longitudinal changes in antibody levels among recovered persons were consistent with the cross-sectional results (Fig. S5); antibody levels were higher in the last sample than in the first sample when the antibodies were measured with the two pan-Ig assays, slightly lower than in the first sample when measured with IgG anti-N and IgG anti-S1 assays, and substantially lower than in the first sample when measured with IgM anti-N and IgA anti-S1 assays.

IgG anti-N, IgM anti-N, IgG anti-S1, and IgA anti-S1 antibody levels were correlated among the qPCR-positive persons (Figs. S5 and S6 and Table S5). Antibody levels measured with both pan-Ig antibody assays increased over the first 2 months after qPCR diagnosis and remained at a plateau over the next 2 months of the study. IgM anti-N antibody levels increased rapidly soon after diagnosis and then fell rapidly and were generally not detected after 2 months. IgA anti-S1 antibodies decreased 1 month after diagnosis and remained detectable thereafter. IgG anti-N and anti-S1 antibody levels increased during the first 6 weeks after diagnosis and then decreased slightly.

SARS-CoV-2 Infection in Quarantine

Table 3. Table 3. SARS-CoV-2 Infection among Quarantined Persons According to Exposure Type and Presence of Symptoms.

Of the 1797 qPCR-positive Icelanders, 1088 (61%) were in quarantine when SARS-CoV-2 infection was diagnosed by qPCR. We tested for antibodies among 4222 quarantined persons who had not tested qPCR-positive (they had received a negative result by qPCR or had simply not been tested). Of those 4222 quarantined persons, 97 (2.3%; 95% CI, 1.9 to 2.8) were seropositive (Table 1). Those with household exposure were 5.2 (95% CI, 3.3 to 8.0) times more likely to be seropositive than those with other types of exposure (Table 3); similarly, a positive result by qPCR for those with household exposure was 5.2 (95% CI, 4.5 to 6.1) times more likely than for those with other types of exposure. When these two sets of results (qPCR-positive and seropositive) were combined, we calculated that 26.6% of quarantined persons with household exposure and 5.0% of quarantined persons without household exposure were infected. Those who had symptoms during quarantine were 3.2 (95% CI, 1.7 to 6.2) times more likely to be seropositive and 18.2 times (95% CI, 14.8 to 22.4) more likely to test positive with qPCR than those without symptoms.

We also tested persons in two regions of Iceland affected by cluster outbreaks. In a SARS-CoV-2 cluster in Vestfirdir, 1.4% of residents were qPCR-positive and 10% of residents were quarantined. We found that none of the 326 persons outside quarantine who had not been tested by qPCR (or who tested negative) were seropositive. In a cluster in Vestmannaeyjar, 2.3% of residents were qPCR-positive and 13% of residents were quarantined. Of the 447 quarantined persons who had not received a qPCR-positive result, 4 were seropositive (0.9%; 95% CI, 0.3 to 2.1). Of the 663 outside quarantine in Vestmannaeyjar, 3 were seropositive (0.5%; 95% CI, 0.1 to 0.2%).

SARS-CoV-2 Seroprevalence in Iceland

None of the serum samples collected from 470 healthy Icelanders between February 18 and March 9, 2020, tested positive for both pan-Ig antibodies, although four were positive for the pan-Ig anti-N assay (0.9%), a finding that suggests that the virus had not spread widely in Iceland before March 9.

Of the 18,609 persons tested for SARS-CoV-2 antibodies through contact with the Icelandic health care system for reasons other than Covid-19, 39 were positive for both pan-Ig antibody assays (estimated seroprevalence by weighting the sample on the basis of residence, sex, and 10-year age category, 0.3%; 95% CI, 0.2 to 0.4). There were regional differences in the percentages of qPCR-positive persons across Iceland that were roughly proportional to the percentage of people quarantined (Table S6). However, after exclusion of the qPCR-positive and quarantined persons, the percentage of persons who tested positive for SARS-CoV-2 antibodies did not correlate with the percentage of those who tested positive by qPCR. The estimated seroprevalence in the random sample collection from Reykjavik (0.4%; 95% CI, 0.3 to 0.6) was similar to that in the Health Care group (0.3%; 95% CI, 0.2 to 0.4) (Table S6).

We calculate that 0.5% of the residents of Iceland have tested positive with qPCR. The 2.3% with SARS-CoV-2 seroconversion among persons in quarantine extrapolates to 0.1% of Icelandic residents. On the basis of this finding and the seroprevalence from the Health Care group, we estimate that 0.9% (95% CI, 0.8 to 0.9) of the population of Iceland has been infected by SARS-CoV-2. Approximately 56% of all SARS-CoV-2 infections were therefore diagnosed by qPCR, 14% occurred in quarantine without having been diagnosed with qPCR, and the remaining 30% of infections occurred outside quarantine and were not detected by qPCR.

Deaths from Covid-19 in Iceland

In Iceland, 10 deaths have been attributed to Covid-19, which corresponds to 3 deaths per 100,000 nationwide. Among the qPCR-positive cases, 0.6% (95% CI, 0.3 to 1.0) were fatal. Using the 0.9% prevalence of SARS-CoV-2 infection in Iceland as the denominator, however, we calculate an infection fatality risk of 0.3% (95% CI, 0.2 to 0.6). Stratified by age, the infection fatality risk was substantially lower in those 70 years old or younger (0.1%; 95% CI, 0.0 to 0.3) than in those over 70 years of age (4.4%; 95% CI, 1.9 to 8.4) (Table S7).

Age, Sex, Clinical Characteristics, and Antibody Levels

Table 4. Table 4. Association of Existing Conditions and Covid-19 Severity with SARS-CoV-2 Antibody Levels among Recovered Persons.

SARS-CoV-2 antibody levels were higher in older people and in those who were hospitalized (Table 4, and Table S8 [described in Supplementary Appendix 1 and available in Supplementary Appendix 2]). Pan-Ig anti–S1-RBD and IgA anti-S1 levels were lower in female persons. Of the preexisting conditions, and after adjustment for multiple testing, we found that body-mass index, smoking status, and use of antiinflammatory medication were associated with SARS-CoV-2 antibody levels. Body-mass index correlated positively with antibody levels; smokers and users of antiinflammatory medication had lower antibody levels. With respect to clinical characteristics, antibody levels were most strongly associated with hospitalization and clinical severity, followed by clinical symptoms such as fever, maximum temperature reading, cough, and loss of appetite. Severity of these individual symptoms, with the exception of loss of energy, was associated with higher antibody levels.

We estimate that during the first wave of the SARS-CoV-2 pandemic, the incidence of infection in Iceland was 0.9% (95% CI, 0.8 to 0.9) and the infection fatality risk was 0.3% (95% CI, 0.2 to 0.6). Our estimate of the infection fatality risk is lower than but consistent with estimates described by others.^6-8 We estimate that of the infected persons, 56% had cases previously diagnosed by qPCR, 14% had been in quarantine (but either had not been qPCR-tested or had tested negative), and 30% neither were known to be qPCR-positive nor had been placed in quarantine. We therefore conclude that, despite extensive screening by qPCR, a substantial fraction of infections were not detected, which indicates that many infected persons did not have substantial symptoms.

The case fatality risk is straightforward to estimate but may differ across countries and over time. An accurate calculation of infection fatality risk requires an accurate estimate of the number of infections, both diagnosed and undiagnosed. In Iceland, the high percentage of infections identified through qPCR (56%) as compared with that of other countries (for example, approximately 9% in Spain⁴) renders a commensurately accurate estimate of the total number of infections.

Each of the pan-Ig SARS-CoV-2 antibody assays that we used has high specificity (99.8%, according to the manufacturers’ literature), which raises the question of whether using a single pan-Ig assay would have sufficed. One sample obtained in 2017 was positive on only one pan-Ig antibody assay, a finding that supports the use of two separate assays to determine seroprevalence, if the infection rate is below 1%, as in Iceland.

By April 30, a total of 20,766 Icelanders had been placed in quarantine. Of the 1797 Icelanders who tested positive by qPCR, 1088 (61%) were in quarantine when tested. Despite substantial qPCR testing of persons in quarantine, 2.3% of persons in quarantine who did not receive qPCR-positive result (i.e., a diagnosis of infection) developed SARS-CoV-2 antibodies. Household exposure was more likely to lead to infection than other types of exposure, which suggests that people who share a household with an infected person should not have contact during quarantine and that contacts of household members should be quarantined. Seroprevalence in the two regional hot spots (Vestfirdir and Vestmannaeyjar) was absent or low outside quarantine, which indicates that most infections were detected by qPCR screening and that quarantine, social distancing, contact tracing, and limits on public gatherings were effective in limiting spread.

Over 90% of qPCR-positive persons tested positive with both pan-Ig SARS-CoV-2 antibody assays and remained seropositive 120 days after diagnosis, with no decrease of antibody levels as detected by the two pan-Ig assays. We observed some diminution of antibody titer with some of the single-Ig assays. Previous smaller studies reported reduction of IgG antibodies against the N protein and a peptide representing the S protein within 21 to 28 days⁵ and against trimeric S protein within 56 days¹⁴ after a positive test by qPCR. These discrepancies may be explained partly by differences in the specificity and sensitivity of the assays used as well as differences in the design and performance of the semiquantitative assays used, including the antigen targeted and the analytic sensitivity and range, as well as differences in the study populations. For example, because of widespread qPCR testing and screening, it is likely that the Icelandic qPCR-positive persons were healthy, as compared with the participants in other studies. Repeated SARS-CoV-2 exposure is unlikely to affect the persistence of antibody levels in Iceland, given the low prevalence of infection. Comparative studies using validated quantitative SARS-CoV-2 antibody assays are needed; those described in the published literature are based on small sample sizes.^9-12

Of the 22 recovered persons who had a negative result (using the combined pan-Ig antibody tests) for an early sample and who had another sample tested at least a month later, 19 (86%) received a second negative result. Thus, either some persons infected by SARS-CoV-2 produce no antibodies or undetectable levels of antibodies reactive to the S1 and N proteins, even 3 months after infection, or some qPCR delivered false positive results.

Among recovered persons, antibody levels are higher in older persons and in those more severely affected by SARS-CoV-2 infection. Women, who tend to become less sick than men, had lower antibody levels in two spike protein antibody assays. SARS-CoV-2 antibody levels were lower in smokers. Smoking increases the probability of severe Covid-19 illness among young adults,¹⁵ and smoking has been reported to increase the expression of ACE2,¹⁶ the receptor for cellular entry of the SARS-CoV-2 virus.

The humoral immune response is critical for the clearance of cytopathic viruses and is generally important for the prevention of viral reinfection.¹⁷ A relationship between a humoral immune response to SARS-CoV-2 infection and protection against reinfection by this virus has been shown in rhesus macaques¹⁸ but has yet to be established in humans. Regardless of the relationship or lack thereof between seropositivity against SARS-CoV-2 and protection against reinfection, the low SARS-CoV-2 antibody seroprevalence in Iceland indicates that the Icelandic population is vulnerable to a second wave of infection.