LAUSR

COVID-19 caused by SARS-CoV-2 infection has caused substantial morbidity and mortality worldwide and paralyzed the international economy. Understanding the magnitude and duration of the antibody response to SARS-CoV-2 is important to achieve a balance between curbing the pandemic and minimizing adverse effects on society. Although the antibody response to SARS-CoV2 within 9 months has been extensively studied, little is known about the magnitude and kinetics of antibody responses for over 9 months. Moreover, with limited observations over 9 months (n < 100), several studies have produced inconsistent conclusions about antibody dynamics, suggesting different rates of antiviral antibody positivity at the last follow-up. These studies have been limited by a lack of measurement of neutralizing antibodies (NAbs), of inclusion of mild or asymptomatic cases, and of further exploration of potential predisposing factors for antibody dynamics. Considering the individual heterogeneity (such as disease severity) and time-dependent nature of the immune response, in-depth characterization of SARS-CoV-2 antibody kinetics across disease severity groups over a long period is urgently needed. Therefore, we repeatedly tested IgM, IgG, viral spike protein receptor-binding dom (anti-RBD) IgG, and NAb titers in COVID-19 patients during a follow-up period of up to 10 months and explored potential predisposing factors of antibody titers during follow-up. A total of 215 COVID-19 patients consisting of both mild and severe cases were recruited and regularly followed up. During a median follow-up of 275 (range, 241–303) days, 803 serum samples were collected for longitudinal SARS-CoV-2 serological tests. Enzyme-linked immunosorbent assays were performed to measure IgM, IgG, and anti-RBD IgG. NAbs were evaluated using a SARS-CoV-2 pseudotyped virus neutralization assay. The halfmaximal inhibitory concentration (NT50) for plasma was calculated as the NAb titer. Samples from the same patient were tested together and in double-blind randomization. Details of the study design and antibody test are provided in the Supplementary Methods. Of the 215 patients, the median (interquartile range, IQR) age was 61 (51–67) years, and 105 (48.8%) individuals were male (Table S1). Their symptoms were classified as mild (142 participants, 66.0%) or severe (73 participants, 34.0%) according to Chinese national treatment guidelines. A total of 112 (52.1%) patients had a long duration of viral clearance (≥34 days), and 38 patients (17.7%) had redetectable positive RNA tests. Hypertension (78, 36.3%), diabetes (30, 14.0%), and cardiovascular disease (17, 7.9%) were the most common comorbidities. The median (IQR) duration between symptom onset and the first antibody test was 37 (23–100) days. Most participants (116, 54.0%) underwent four antibody tests. Details of the definition of severity, duration of viral clearance, redetectable positive RNA test, and comorbidities are provided in the Supplementary Methods. The positivity rates for IgM, IgG, anti-RBD IgG, and NAb fell to 20.4% (39/191), 97.9% (187/191), 97.4% (186/191), and 95.8% (183/ 191), respectively, during 9–10 months post symptom onset (Fig. 1A). As shown in Fig. 1B and Fig. S1, a rapid decline in IgM titers was observed from 1 to 6 months post symptom onset, with median (IQR) optical density (OD) values at 1–2, 3–4, and 5–6 months post symptom onset of 0.65 (0.15–1.24), 0.18 (0.05–0.51), and 0.06 (0.02–0.21), respectively. Afterwards, IgM titers remained relatively stable for 6–10 months. In addition, progressive declines in IgG and anti-RBD IgG were observed during 1–10 months after symptom onset, with median OD (IQR) values from 2.61 (2.20–3.00) and 3.11 (2.69–3.51) at 1–2 months to 2.08 (1.74–2.64) and 2.29 (1.95–2.67) at 9–10 months post symptom onset. Similarly, a significant decrease in NAbs was identified during the whole observation period. The median (IQR) NT50 was 1096 (430–3222) during 1–2 and decreased to 249 (105–501) during months 9–10. To explore potential predisposing factors of antibody dynamics during each postinfection phase, we examined distributions of NAb by age, sex, disease severity, duration of viral clearance, redetectable positive RNA test, and hypertension, as depicted in Fig. 1C. NAb titers during 1–2 months post symptom onset were significantly higher in elderly participants, severe cases, and patients without repositive RNA tests or with hypertension. Compared to the group with a long viral clearance duration, the group with a short viral clearance duration had a higher NAb titer during 1–2 months and a lower titer during 5–10 months post symptom onset. Longitudinal assessment of IgM, IgG, and anti-RBD IgG titers by different