The Dynamic Changes of Antibodies against SARS-CoV-2 during the Infection and Recovery of COVID-19

Deciphering the dynamic changes of antibodies against SARS-CoV-2 is essential for understanding the immune response in COVID-19 patients. By comprehensively analyzing the laboratory findings of 1,850 patients, we describe the dynamic changes of the total antibody, spike protein (S)-, receptor-binding domain (RBD)-, and nucleoprotein (N)- specific IgM and IgG levels during SARS-CoV-2 infection and recovery. Our results indicate that the S-, RBD-, and N- specific IgG generation of severe/critical COVID-19 patients is one week later than mild/moderate cases, while the levels of these antibodies are 1.5-fold higher in severe/critical patients during hospitalization (P<0.01). The decrease of these IgG levels indicates the poor outcome of severe/critical patients. The RBD- and S-specific IgG levels are 2-fold higher in virus-free patients (P<0.05). Notably, we found that the patients who got re-infected had a low level of protective antibody on discharge. Therefore, our evidence proves that the dynamic changes of antibodies could provide an important reference for diagnosis, monitoring, and treatment, and shed new light on the precise management of COVID-19.


Introduction
Caused by the infection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), coronavirus disease 2019 (COVID-19) is spreading in more than 210 countries and territories around the World 1-3 . As of April 24, 2020, a total of 2,718,155 confirmed cases were reported, of which 190,636 patients died. Although its mortality is lower than SARS and MERS, the high infection rate leads to its rapid spread 4 .
Approximately 100,000 confirmed cases increased every single day, extremely challenging the public health and medical service around the globe. Therefore, rapid, accurate, and timely diagnosis and monitoring of COVID-19 are urgently needed.
Currently, the diagnosis of COVID-19 is mainly based on the virus RNA load test using aquantitative real-time polymerase chain reaction (RT-PCR) 5 . However, the nucleic acid testing results are subject to many factors, including the specimen location, type, quality, and patients' condition, as well as sample storage. Thus, some COVID-19 patients will be unrecognized if only based on the virus RNA load test 6 . In consideration of the high false-negative rate of virus load detection, on March 3, 2020, the latest release of "Diagnosis and Treatment Protocol for Novel Coronavirus SARS-CoV-2 play an important role in resisting the virus infection by host 8 . It is widely recognized that IgM provides the first line of defense during viral infections, while the production of IgG lags behind IgM, and provides long-term immunity and memory. The coronavirus neutralizing antibodies primarily target the trimeric spike (S) glycoproteins on the viral surface that mediates entry into host cells 9 . The S protein has two functional subunits, the S1 subunit mediates cell attachment, and the S2 subunit mediates fusion of the viral and cellular membrane 10 . Neutralizing antibodies often target the receptor-binding domain (RBD) in the S1 subunit to block the interaction between the virus and host receptor 11 . However, the dynamics of serum antibodies against SARS-CoV-2 during COVID-19 infection and recovery are still unclear, and most of the current opinions are based on the knowledge acquired from previous experience of the SARS-CoV and MERS-CoV infections 12 .
Here, we comprehensively analyzed the laboratory tests of 1,850 hospitalized COVID-19 patients at Wuhan Huoshenshan Hospital, admitted from February 4 to March 30. We describe the dynamics of the SARS-Cov-2-specific antibody levels, including total antibody, as well as the S-, RBD-, and nucleoprotein (N)-specific IgM and IgG levels on admission, during hospitalization and on discharge. Our results would provide an important reference for the understanding of immune response after SARS-Cov-2 infection, and the function of protective antibody during the recovery, and shed new light on the precise management of COVID-19.

Data Collection
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Serum Anti-SARS-CoV-2 Antibodies Assay
Total SARS-CoV-2 IgM or IgG in the serum was measured by chemiluminescence using commercially available kits (Shenzhen YHLO Biotech Co., Ltd.) in 1,850 patients at different time points. The magnetic beads of this kit are coated with recombinant N and S proteins. 416 of these patients were tested for S-, RBD-, and N-specific IgM and IgG levels at different time points by chemiluminescence using commercially available kits (Nanjing RealMind Biotech Co., Ltd.), including 126 mild or moderate patients, and 290 severe or critical patients. Briefly, blood samples were centrifugated at room temperature and the supernatant was removed and incubated separately with SARS-CoV-2 antigens-coated magnetic beads. The antigen-antibody complex captured by the beads slurry was gently precipitated by a magnetic separation rack. The beads were then incubated with acridinium ester-labeled mouse anti-human IgM or IgG antibody and reacted with hydrogen peroxide in excitation buffer. Relative luminescence intensity was recorded in an ACL2800 chemiluminescence system (Nanjing RealMind Biotech Co., Ltd.). Relative All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . luminescence intensity was converted to antibody level and its unit is AU/mL.

Statistical Analysis
Continuous and categorical variables were presented asmedian (IQR) and n (%), respectively. We used the two-sided Wilcoxon rank-sum test or Fisher's exact test to compare the difference between groups where appropriate.

Temporal profiles of antibodies against SARS-CoV-2 of mild/moderate and severe/critical COVID-19 patients
To explore the temporal dynamics of immune response after SARS-Cov-2 infection, we analyzed the antibody levels at different time points after symptoms onset, and the timing and level were compared between mild/moderate and severe/critical COVID-19 patients.
The level of total IgM was extremely low in the first week (median=5.34 AU/ML), and gradually increased until the 5th week (median=43.98 AU/ML), followed by a continuous decrease to the initial level. The level of total IgG was more detectable than IgM at the first week, and continuously increased to seventh week (median=154.54 AU/ML), and slightly decreased from the 8th week, but still kept considerably high level until the end of our observation (12th week, median=95.94 AU/ML) (Supplementary Figure S1A). We further quantified the total antibody levels of confirmed patients after disease onset, and found that the positive rates of the total IgG were 62.5% in the first week. Total IgG could be detected in 94.7% of patients after 5 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . https://doi.org/10.1101/2020.05.18.20105155 doi: medRxiv preprint weeks of disease onset, and 97.6% of patients retained considerable abundance even after 12 weeks. While IgM was rarely detected (32.5%) during the early stage (SupplementaryFigure S1B). 97.4% of the confirmed patients have positive IgM or IgG at the first 1 week after symptom onset, indicating that the combination of IgM and IgG is necessary for auxiliary diagnosis.
As shown in Figure S1C, the level of S-, RBD-, and N-specific IgM reached a high level in the second week after symptom onset in the mild/moderate COVID-19 patients, while it took 3 weeks for the severe/critical COVID-19 patients to get the comparable antibody level. Furthermore, in the mild/moderate COVID-19 patients, the level of S-, RBD-, and N-specific IgG generated at the first 2 weeks after symptom onset, then sharply increased in the third week, and reached their peaks at the fifth or sixth week.
However, in the severe/critical COVID-19 patients, the level of S-, RBD-, and N-specific IgG generated in the third week after symptom onset and reached their peaks in the seventh week. These IgG levels kept considerably high level until the end of our observation (12thweek). These results indicated the immune response of the severe/critical COVID-19 patients was later than the mild/moderate ones for approximately one week. We found that the S-, RBD-, and N-specific IgG levels were higher in the mild/moderate COVID-19 patients at the early stage of infection (first 2~3 weeks after onset) than the severe/critical COVID-19 patients, suggesting the lack of sufficient protective antibodies in patients who had more severe symptoms at the disease onset. But at the middle and late stages of infection (7~10 weeks after onset), the IgG levels were significantly higher in the severe/critical COVID-19 patients than All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . the mild/moderate ones (P<0.05). We compared the total antibody, as well as the S-, RBD-, and N-specific antibody levels between the cured mild/moderate and severe/critical COVID-19 patients at the time of admission, hospitalization, and discharge. Results showed that total IgG, S-, RBD-, and N-specific IgG levels of the severe/critical COVID-19 patients were lower than that of the mild/moderate patients on admission, but these levels sharply increased during hospitalization and on discharge ( Figure 1, Table 1). The RBD-specific IgG level was approximately 1.5-fold higher in the severe/critical COVID-19 patients than in the mild/moderate ones (P=0.006) during hospitalization, and this ratio is up to 1.8 on discharge (P=0.001).
The S-specific IgG levels were also significantly higher in the severe/critical COVID-19 patients during hospitalization and on discharge (P<0.01, Table 1). The significant increase levels of protective antibody against SARS-Cov-2 implied that there are more activated immune responses during the recovery of the severe/critical COVID-19 patients.
Taken together, this evidence indicated that compared with the mild/moderate ones, the severe/critical COVID-19 patients had a late response at the beginning but stronger defense at the middle and late stage of infection. These results emphasized the crucial role of timely medical intervention for the severe/critical COVID-19 patients.

The IgG levels are significantly higher in older COVID-19 patients during hospitalization
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We further evaluated the differences in antibody levels among patients of different age groups. As shown in Table 2, the patients were divided into three groups based on their age, young (Age<40 years), middle-aged (40-65 years), and older patients (Age>65 years). On admission, though not statistically significant, the RBD-specific antibody level in patients who were older than 65 years was relatively lower than in young and middle-aged patients, while the N-specific antibody levels of older patients were higher, probably because N protein had high immunogenic activity and was abundantly expressed during the early stage of infection, while the level of protective RBD-specific antibody was lower in older patients owing to their weak immune defense system. Besides, we observed that the S-, RBD-, and N-specific IgG levels were gradually elevated along with the age increase during hospitalization and on discharge (P<0.05). For example, the median level of RBD-specific antibody in young, middle-aged and older patients was 7.7 AU/ML, 22.4 AU/ML, and 30.7 AU/ML, respectively (young vs. middleage: P<0.0001, middle-age vs. older: P=0.003) during hospitalization. Because the older COVID-19 patients were usually more severe 1 , the more activated immune defense was observed in these patients.

The dynamic changes of the IgG levels are related to the clinical outcome of the severe/critical COVID-19 patients
As the S-, RBD-, and N-specific IgG levels significantly increased in the severe/critical COVID-19 patients during hospitalization, we attempted to evaluate the function of these antibodies in the recovery of the severe/critical COVID-19 patients. We All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . compared the dynamics of these IgG levels between cured and dead severe/critical COVID-19 patients (Figure 2A). Results showed that the S-, RBD-, and N-specific IgG levels gradually increased from admission to discharge of survivors. However, the Sand RBD-specific IgG levels sharply decreased in the fifth week after admission in non-survivors, although these IgG reached a higher level than the survivor before the fifth week. The N-specific IgG level was much higher in the non-survivors than To evaluate whether the antibody levels are correlated with the neutrophil and lymphocyte percentage, we analyzed the blood routine examination and antibody test data of the same cured severe/critical COVID-19 patients on the same day ( Figure 2B). All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We observed that the percentage of lymphocytes was significantly lower in patients with a low IgG level than in patients with high IgG levels (Figure 2, S-specific IgG: P=0.0002, RBD-specific IgG: P=0.007, N-specific IgG: P=2.1e-05). Nevertheless, a higher percentage of neutrophils was observed in patients with low IgG levels (S-specific IgG: P=1.6e-05, RBD-specific IgG: P=0.001, N-specific IgG: P=7.5e-06).
These results further revealed that the S-, RBD-, and N-specific IgG could prevent severe/critical COVID-19 patients from progress.

S-and RBD-specific IgG levels are significantly higher in virus-free patients
To illustrate the relationship between SARS-COV-2 virus load and antibody levels, we analyzed the test results of patients who tested both virus load and antibody level on the same day and compared the S-, RBD-, and N-specific antibody levels between virus-positive and virus-negative status. We observed that the S-specific and RBDspecific antibody levels were significantly higher in the virus-negative status than in thevirus-positive status ( Table 3)

The dynamic changes of antibody levels after convalescent plasma transfusion (CPT) therapy
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . https://doi.org/10.1101/2020.05.18.20105155 doi: medRxiv preprint Because the plasma of patients who have recovered from infections contains theneutralizing antibodies against the virus, human CPT is an option for the prevention and treatment of COVID-19 disease and has been applied in China 14,15 . We analyzed the total antibody level in patients who received CPT at different time points in our cohort. The total IgM and IgG level of 5 patients at 3 days before CPT, and 1-3, 4-7, 7-14 days after CPT were showed in Figure 3. The results showed that although some fluctuations were observed in 2 patients, the total IgM levels decreased within 14 days after CPT. While the total IgG levels decreased during the first 7 days and increased eventually in the second week. These observations demonstrate that CPT therapy may provide the long-term antibody against the SARS-Cov-2 virus for patients and help the COVID-19 patients recover. However, the molecular mechanism of the decrease of IgG level in the first week after CPT needs to be further illuminated.

The low level of protective antibody on discharge is associated with a high risk of re-infection
According to the latest version of "Diagnosis and Treatment Protocol for Novel Coronavirus Pneumonia of China", patients who meet the following criteria can be discharged: 1) body temperature is back to normal for more than three days; 2) respiratory symptoms improve obviously; 3) pulmonary imaging shows obvious absorption of inflammation; 4) nucleic acid tests negative twice consecutively on respiratory tract samples such as sputum and nasopharyngeal swabs (sampling interval being at least 24 hours). However, we are wondering whether the discharged All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . https://doi.org/10.1101/2020.05.18.20105155 doi: medRxiv preprint patients have an efficient protective antibody to prevent them from re-infection. We observed that some patients tested free of virus load, but harboured low levels of protective antibody (Figure 4). In our cohort, patient #288 was discharged on March 5, 2020, with a relatively low level of S-(15.1 AU/ML), RBD-(19.2 AU/ML), and N-specific (15.9 AU/ML) antibodies, which was much lower than average level. This patient got re-infected on March 15 and was back to the hospital for treatment. His second time of discharge was on April 2, still with a low level of protective antibody.
Another 3 patients who also got re-infected, their antibody levels were not tested at their first-time discharge. On their second time of discharge, the test data showed that they still had a low level of protective antibody, implying the probability that these patients were at high risk of re-infection. Therefore, our results indicated that a fair number of virus-free patients were discharged with a low level of protective antibody.
These patients may need close monitoring after discharge.

Discussion
Although several case series about the SARS-Cov-2 antibody responses were previously reported 16-18 , the general pattern of the dynamics of serum anti-SARS-CoV-2 antibody during the infection and treatment of COVID-19 was still unclear due to insufficient sample size of test data. By analyzing the laboratory data of patients from Wuhan Huoshenshan Hospital, which is one of the biggest designated hospitals for COVID-19 in Wuhan, we profiled the temporal dynamic changes of the antibody level from disease onset to 12 weeks, including the S-, RBD-, and N-specific antibodies. We found that compared with the mild/moderate ones, the S-, RBD-, and All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
These IgG levels are significantly higher in the patients who are more severe or older during hospitalization, indicating that the severe and older patients have more activated immune defense during recovery. The increase of these IgG levels indicates a better prognosis of the severe/critical COVID-19 patients. In addition, we found that the RBD-specific IgG level was much higher in virus-negative COVID-19 patients, indicating an important role of the neutralizing antibodies in virus clearance.
Importantly, the patients who got discharged with low levels of protective antibody are found to be at high risk of re-infection, indicating the prognostic role of antibody level for discharged COVID-19 patients. Overall, our results suggested that these IgG, especially S-specific or RBD-specific IgG played an important role in virus clearance and recovery of COVID-19 patients.
Furthermore, according to the previous report about SARS in 2003, IgM could be detected in patients' blood after 3-6 days of disease onset, while IgG could be detected after 8 days of disease onset 17 . Our observations showed that the total anti-SARS-Cov-2 IgG level was already at a relatively high level in the first week after disease onset, probably because some COVID-19 patients are asymptomatic at the beginning of disease onset 19,20 . The clinicians record the first day of the patients' symptoms, such as fatigue, fever, cough or diarrhea as the disease onset time point.
However, the record date may be later than the actual date of infection due to the asymptomatic situation of COVID-19, so that the observed IgG level was high during the first week after record onset. All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . https://doi.org/10.1101/2020.05.18.20105155 doi: medRxiv preprint Moreover, some re-infected patients were reported lately 21,22 , attracting attention to the patients' condition on discharge. The current discharge criteria are mainly based on virus clearance and radiologic findings. However, our study shows that those re-infected patients were discharged with a low protective antibody level, emphasizing that antibody tests are necessary when discharging patients. In our opinion, two reasons may be involved in the re-infection of COVID-19 patients.
Firstly, studies have shown that SARS-CoV-2 mainly infects the lower respiratory tract 23 , but the collection of the bronchoalveolar lavage requires skilled operators and specific devices, exposing the clinicians to a high risk of infection. So, the nasopharyngeal swab samples are usually used to assess the virus load, which easily leads to the false negative of discharged patients. Secondly, the immune defense system of some discharged patients is relatively weak. Lacking sufficient protective antibody makes these patients infected again. Thus, it is important to test the S-specific or RBD-specific IgG level before discharging patients and keep close monitoring of the patients with a low level of protective antibody.
In conclusion, our results suggest that the protective antibodies against (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. . https://doi.org/10.1101/2020.05.18.20105155 doi: medRxiv preprint   (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.  (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020.

Days Days
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 21, 2020.