Elevated SARS-CoV-2-Specific Antibody Levels in Patients with Post-COVID Syndrome

With the routine use of effective severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines, the number of life-threatening coronavirus disease 2019 (COVID-19) courses have largely been reduced. However, multiple COVID-19 convalescents, even after asymptomatic to moderate disease, suffer from post-COVID syndrome, with relevant limitations in daily life. The pathophysiologic mechanisms of post-COVID syndrome are still elusive, with dysregulation of the immune system suggested as a central mechanism. Here, we assessed COVID-19 post-infectious symptoms (5–6 months after PCR-confirmed acute infection) together with the humoral immune response against SARS-CoV-2 in non-hospitalized COVID-19 convalescents, early (5–6 weeks) and late (5–6 months) after their first positive SARS-CoV-2 PCR result. Convalescents reporting several post-infectious symptoms (>3) showed higher anti-spike and anti-nucleocapsid antibody levels 5–6 weeks after PCR-confirmed infection with the latter remained increased 5–6 months after positive PCR. Likewise, a higher post-infectious symptom score was associated with increased antibody levels. Of note, convalescents displaying neuro-psychiatric symptoms such as restlessness, palpitations, irritability, and headache, as well as general symptoms such as fatigue/reduced power had higher SARS-CoV-2-specific antibody levels compared with asymptomatic cases. The increased humoral immune response in convalescents with post-COVID syndrome might be useful for the detection of individuals with an increased risk for post-COVID syndrome.


Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), causing coronavirus disease 2019 (COVID- 19), is still spreading around the world, so far reaching more than 600 million documented cases [1]. With approval of effective SARS-CoV-2 vaccines, the number of life-threatening COVID-19 courses has largely been reduced. However, multiple COVID-19 convalescents, even after asymptomatic to moderate disease, suffer from post-COVID syndrome with relevant limitations in daily life [2]. Recently, it was reported that after a mild course of COVID-19 11% of patients still could not fully participate in everyday and work life seven months after disease onset [3]. One of the key challenges for health systems in general and for physicians, dealing with post-COVID patients in particular, is

Assessment of Post-Infectious Symptoms
A questionnaire-based assessment of ten post-infectious symptoms, including grading of severity (no, mild, moderate and severe), was performed at T L . Symptoms were classified into four categories: general symptoms ("fatigue" and "reduced performance"), altered sensory perception ("anosmia/ ageusia" and "hearing loss"), neuro-psychiatric symptoms ("restlessness", "palpitations", "irritability", and "headache"), and pulmonary symptoms ("cough" and "dyspnea"). Patients were included in a category if at least one symptom for that category was reported by the patient. The symptom score was calculated by summing up the self-reported grading (no = 0, mild = 1, moderate = 2, severe = 3) of all of the reported symptoms.
The 96-well Euroline Anti-SARS-CoV-2 ® (Euroimmune, 2606A_A_DE_C03, as constituted on 22 April 2020) was performed on an automated BEP 2000 Advance system (Siemens Healthcare Diagnostics GmbH) according to the manufacturer's instructions. The ELISA assay detected anti-SARS-CoV-2 IgG directed against the S1 domain of the viral spike protein (including the immunologically relevant RBD) and relied on an assay-specific calibrator to report a ratio of specimen absorbance to calibrator absorbance. The final interpretation of positivity was determined by a ratio above a threshold value given by the manufacturer: positive (ratio ≥ 1.1), borderline (ratio 0.8-1.0), or negative (ratio < 0.8). The levels of anti-spike IgG-antibodies are shown in the graphs as a ratio above the threshold value. Quality control was performed following the manufacturer's instructions on each day of testing.
The Elecsys ® anti-SARS-CoV-2 immunoassay is an ECLIA (electro-generated chemiluminescence immunoassay) designed by Roche Diagnostics GmbH and was used according to the manufacturer's instructions (V1.0, as constituted in May 2020). It is intended for the detection of high-affinity antibodies (including IgG and IgA) directed against the nucleocapsid protein of SARS-CoV-2 in human serum. Readout was performed on the Cobas e411 analyzer (Roche Diagnostics). Negative results were defined by a cut-off index of <1.0. The levels of anti-nucleocapsid antibodies are shown in the graphs as an index value. Quality control was performed following the manufacturer's instructions on each day of testing.
Data on antibody levels and post-infectious symptoms were retrieved from previous publications [28,40].

Software and Statistical Analysis
Data are displayed as mean with standard deviation (SD), box plots with median and 25th and 75th quartile, min/max whiskers, and individual data points. Groups were tested using the Mann-Whitney U test or Kruskal-Wallis test and were corrected for multiple comparison if applicable. Statistical analyses were conducted using JMP Pro (SAS Institute, v.16, Cary, NC, USA) software. p-values < 0.05 were considered statistically significant. Missing data were denoted in tables and in the descriptive analysis. Graphs were plotted using GraphPad Prism v.9.1.2 (San Diego, CA, USA).

Clinical Characteristics of COVID-19 Convalescents and Prevalence of Post-Infectious Symptoms
For this study, we analyzed post-infectious symptoms after SARS-CoV-2 infection at T L (5-6 months after PCR-confirmed COVID-19) with SARS-CoV-2-specific antibody responses at T E (5-6 weeks after positive PCR) and T L of 51 convalescent individuals with an asymptomatic to moderate COVID-19 course. The mean age of convalescents at the time of infection was 43.4 (SD ± 13.7) years, with an equal gender distribution (female to male ratio of 1.04:1). No convalescent in this study had severe symptoms according to WHO criteria and no one was hospitalized due to COVID-19 infection [39]. Most convalescents (76%) reported at least one post-infectious symptom at T L (median 2, range 0 to 9 symptoms, Table 1).

Anti-SARS-CoV-2 Antibody Levels and Number of Post-Infectious Symptoms
Anti-SARS-CoV-2 anti-nucleocapsid and anti-spike antibody levels were assessed at T E and T L. 84% and 75% of convalescents were positive for anti-spike antibodies at T E and T L , respectively, whereas positivity for anti-nucleocapsid antibodies remained stable with a seroconversion rate of 88% at both timepoints (Table 1).
To assess SARS-CoV-2-directed humoral immune responses in convalescents with different post-infectious symptoms, convalescents were grouped according to the number of reported symptoms into two groups (≤2 symptoms (n = 30) versus ≥3 symptoms (n = 21) corresponding to post-COVID syndrome). Convalescents with post-COVID syndrome displayed significantly increased anti-nucleocapsid antibody levels at T E and T L , as well as significantly increased anti-spike-antibody levels at T E , as well as a trend for increased anti-spike levels at T L (p-values: 0.007, 0.045, 0.024, and 0.062, respectively; Figure 1A,B). TE and TL. 84% and 75% of convalescents were positive for anti-spike antibodies at TE and TL, respectively, whereas positivity for anti-nucleocapsid antibodies remained stable with a seroconversion rate of 88% at both timepoints (Table 1).
To assess SARS-CoV-2-directed humoral immune responses in convalescents with different post-infectious symptoms, convalescents were grouped according to the number of reported symptoms into two groups (≤2 symptoms (n = 30) versus ≥3 symptoms (n = 21) corresponding to post-COVID syndrome). Convalescents with post-COVID syndrome displayed significantly increased anti-nucleocapsid antibody levels at TE and TL, as well as significantly increased anti-spike-antibody levels at TE, as well as a trend for increased anti-spike levels at TL (p-values: 0.007, 0.045, 0.024, and 0.062, respectively; Figure 1A,B). Anti-nucleocapsid antibody (Ab) (left) and anti-spike Ab levels (right) were assessed in convalescent donors (n = 51) at T E ("early" timepoint 5-6 weeks after positive PCR, (A,C)) and T L ("late" timepoint 5-6 months positive PCR, (B,D)). Convalescents were grouped according to the number of postinfectious symptoms at T L . Levels of anti-spike Ab are shown as a ratio above the threshold value. Levels of anti-nucleocapsid Ab are shown as an index value. Data are presented as box plots with 25th and 75th percentiles and min/max whiskers. p-values were calculated using the Mann-Whitney U test (A,B) and Kruskal-Wallis test (C,D). p, p-value; Ab, antibody.
In the next step, we divided the group of convalescents without post-COVID-syndrome into asymptomatic individuals and individuals reporting one to two symptoms to further investigate the differences in SARS-CoV-2-specific humoral immune response especially between these two populations. Therefore, three groups were evaluated (asymptomatic, 1-2 symptoms and ≥3 symptoms corresponding to post-COVID syndrome). In line with the previous findings, convalescents reporting post-COVID syndrome had significantly increased anti-nucleocapsid antibody levels at T E in comparison with both asymptomatic individuals and convalescents with only one to two symptoms (p-values: 0.035 and 0.019, respectively; Figure 1C), with a similar trend at T L (p-values: 0.161 and 0.059, respectively; Figure 1D). Anti-spike antibody levels were similarly increased in convalescents reporting many symptoms (≥3) at T E , being only significant in between strongly affected and mildly affect individuals (p-value: 0.032; Figure 1C). At T L the same trend for higher anti-spike antibody levels in individuals with post-COVID syndrome was observed, failing to reach significance in comparison to asymptomatic or mildly affected convalescents (p-values: 0.270 and 0.057, respectively; Figure 1D). Of note, there were no significant differences between asymptomatic individuals and convalescents with one to two symptoms for antinucleocapsid and anti-spike antibody levels at both timepoints T E and T L (p-values: 0.849, 0.597, 0.751, and 0.446; Figure 1C,D).
Taken together, our data showed elevated SARS-CoV-2-specific humoral immune response in individuals with post-COVID syndrome (≥3 symptoms) in comparison with both asymptomatic and mildly affected convalescents, whereas the antibody levels did not differ significantly between the last two mentioned.

Anti-SARS-CoV-2 Antibody Levels and Severity of Post-Infectious Symptoms
To assess humoral immune response and the severity of post-infectious symptoms convalescents were grouped according to the symptom score into two groups (score ≤ 3 (n = 32) versus score > 3 (n = 19) corresponding to post-COVID syndrome). Convalescents with post-COVID-syndrome showed significantly higher anti-nucleocapsid antibody levels at T E as well as significantly increased anti-spike antibody levels at T E and T L (p-values: 0.019, 0.031 and 0.033, respectively; Figure 2A,B).
To further investigate differences in SARS-CoV-2-specific humoral immune response, especially between asymptomatic individuals and individuals with a score of one to three, we divided the group of convalescents without post-COVID-syndrome into these two populations. Therefore, three groups were evaluated (asymptomatic with a score of 0, low with a score of 1-3, and high with a score >3 corresponding to post-COVID syndrome). As observed before, convalescents with a high symptom score had significantly increased anti-nucleocapsid antibody levels at T E in comparison with both asymptomatic individuals and convalescents with a low symptom score (p-values: 0.049 and 0.045, respectively; Figure 2C), with a similar trend also at T L (p-values: 0.162 and 0.103; Figure 2D). Anti-spike antibody levels were increased likewise in convalescents with a high symptom score in comparison to both other groups at T E and T L , being significant only when compared with mildly affected convalescents at both time points (p-values: 0.128, 0.040, 0.167, and 0.036; Figure 2C,D). Of note, there were no significant differences between asymptomatic individuals and convalescents with a low symptom score for anti-nucleocapsid or anti-spike antibody titers at any time point (p-values: 0.923, 0.712, 0.892, and 0.381; Figure 2C,D).
In summary, we observed increased SARS-CoV-2-specific antibody levels in individuals with post-COVID syndrome (symptom score > 3) in comparison to both asymptomatic and mildly affected convalescents, whereas no differences could be found in between the last two mentioned.

Anti-SARS-CoV-2 Antibody Levels and Single Post-Infectious Symptoms
In the next step, a potential association of SARS-CoV-2-directed humoral immune response with specific single post-infectious symptoms was assessed. Significant differences were observed for the single post-infectious symptoms "headache" and "irritability". Convalescents reporting "headache" exhibited higher anti-nucleocapsid antibody levels at T E , with a similar trend at T L (p-values: 0.023 and 0.316, respectively; Figure 3A). No relevant differences were shown for anti-spike antibody levels at T E and T L (p-values: 0.272 and 0.858, respectively; Figure 3B).
In terms of the symptom "irritability", convalescents had significantly higher antinucleocapsid antibody levels at T L and trended to elevated anti-nucleocapsid antibody levels at T E (p-values; 0.026 and 0.065, respectively; Figure 3C). In addition, anti-spike antibody levels were significantly increased at T E with a similar, but not significant, trend To further investigate differences in SARS-CoV-2-specific humoral immune response, especially between asymptomatic individuals and individuals with a score of one to three, we divided the group of convalescents without post-COVID-syndrome into these two populations. Therefore, three groups were evaluated (asymptomatic with a score of 0, low with a score of 1-3, and high with a score >3 corresponding to post-COVID syndrome). As observed before, convalescents with a high symptom score had Figure 2. Severity of post-infectious symptoms and antibody response after SARS-CoV-2 infection. Anti-nucleocapsid antibody (Ab) levels (left) and anti-spike Ab levels (right) were assessed in COVID-19 convalescent donors (n = 51) at T E ("early" timepoint 5-6 weeks after positive PCR, (A,C)) and T L ("late" timepoint 5-6 months after positive PCR, (B,D)). Convalescents were grouped according to post-infectious symptom severity at T L , as defined by the symptom score. The existence and severity (no, mild, moderate, and severe) of 10 individual symptoms were assessed using a questionnaire. The symptom score was determined by adding up the gradings of all single symptoms according to self-reported grading (no = 0, mild = 1, moderate = 2, severe = 3). The levels of anti-spike Ab are shown as a ratio above the threshold value. Levels of anti-nucleocapsid Ab are shown as an index value. Data are presented as box plots showing the median with 25th and 75th percentiles and min/max whiskers. p-values were calculated using the Mann-Whitney U test (A,B) and Kruskal-Wallis test (C,D). p, p-value; Ab, antibody.

Anti-SARS-CoV-2 Antibody Levels and Classification of Symptoms
For further analysis, post-infectious symptoms were classified according to general symptoms, altered sensory perception, neuro-psychiatric, and pulmonary symptoms. The appearance of general symptoms ("fatigue" and "reduced performance") was observed in convalescents with significantly higher anti-spike antibody levels at T L , with the same trend at T E (p-values: 0.047 and 0.149, respectively; Figure 4A,B).    Individuals that reported altered sensory perceptions ("anosmia/ageusia" and "hearing loss") had significantly higher anti-nucleocapsid antibody levels at T E , with the same tendency at T L (p-values: 0.008 and 0.102, respectively; Figure 4C,D). Convalescents with a neuro-psychiatric symptom ("restlessness", "palpitations", "irritability", and "headache") showed significantly higher anti-nucleocapsid antibody levels at T E with a similar trend at T L (p-values: 0.016 and 0.150, respectively; Figure 4E,F). A tendency towards higher anti-spike antibody levels could also be noticed at T E in this classification (p-values: 0.118; Figure 4E). Interestingly, for pulmonary symptoms ("cough" and "dyspnea") no significant differences in anti-nucleocapsid and anti-spike antibody levels were observed at both time points (p-values: 0.974, 0.574, 0.795 and 0.574, respectively; Figure 4G,H).

Discussion
The clinical course of post-COVID syndrome can be very variable and it is frequently reported even in non-hospitalized convalescents with an asymptomatic to moderate course of COVID-19 [14][15][16]. In this work, we observed post-COVID syndrome in about 40% of all individuals 5-6 months after PCR-confirmed acute infection, as well as increased anti-SARS-CoV-2-specific antibody levels (5-6 weeks and 5-6 months after positive PCR) in post-COVID syndrome patients. In addition, convalescents reporting general symptoms, altered sensory perceptions, or neuropsychiatric symptoms showed higher anti-SARS-CoV-2-specific antibody levels.
In this non-hospitalized group of convalescents, 76% of individuals reported at least one post-infectious symptom 5-6 months after PCR-confirmed COVID-19, which was higher compared with another study reporting at least one post-infectious symptom in about 40% of convalescents six months after positive PCR [33]. Of note, in hospitalized convalescents similar rates (73%) of post infectious symptoms were observed [14]. Differences in the prevalence of post-COVID syndrome were most likely due to differing assessment with regards to time points and methods, divergent definitions of post-COVID syndrome, and the baseline characteristics of the convalescents [9,10,14,33,37,41]. Currently, the definition of post-COVID syndrome relies on the persistence of symptoms, which cannot be explained by an alternative cause 3 months after COVID-19 and a general definition of symptoms for the diagnosis is missing [42]. In this study, most convalescents reporting symptoms 5-6 months after PCR confirmed acute infection attributed symptoms to be post-infectious after COVID-19. Of note, this study was conducted early in the SARS-CoV-2 pandemic and was influenced by a lack of awareness of post-COVID syndrome and in addition, the self-reported assessment of symptoms by questionnaire has to be taken with caution [43,44]. Therefore, all of the reported (post-infectious) symptoms were evaluated and convalescents with more than two symptoms or a symptom score > 3 were considered as presenting with post-COVID syndrome. This rate is in line with a prospective and longitudinal study reporting that 35% of non-hospitalized COVID-19 patients suffer from post-COVID syndrome 7 months after infection [3]. The most frequently documented symptoms in our cohort were "fatigue", "anosmia and ageusia", "headache", and "reduced performance", as previously reported [14].
Post-COVID syndrome significantly limits the activity of daily life and can thus lead to tremendous socioeconomical consequences [2,4]. Therefore, a key challenge for physicians is the identification of COVID-19 patients being at high risk of developing post-COVID syndrome [4]. So far, the cause of post-COVID syndrome is not fully understood and is even discussed contradictory, which further complicates the search for reliable predictive markers [6,[17][18][19][20][21][22][23][24][25]. Some risk factors for post-COVID syndrome have been identified, such as age, body mass index, gender, and severity of symptoms during acute COVID-19 [45][46][47][48]. In addition, our and other studies have reported on differential SARS-CoV-2-specific humoral immune responses in convalescents with post-COVID syndrome and the role of the SARS-CoV-2-specific humoral immune response in post-COVID syndrome is still a matter of debate as data are incomplete and even partially contradictory, which may be attributable to the different methodological approaches [3,[32][33][34][35][36][37][38]. In our cohort, 84% and 75% of convalescents were positive for anti-spike antibodies at T E and T L , respectively, whereas positivity for anti-nucleocapsid antibodies remained stable with a seroconversion rate of 88% at both timepoints. In this work, we used the Elecsys ® anti-SARS-CoV-2 anti-N immunoassay (Roche Diagnostics) for the detection of anti-nucleocapsid antibodies. This assay is known to be more sensitive during the complete course of a SARS-CoV-2 infection compared to many other assays, which is consistent with our results [49,50]. In addition, the competitive Elecsys ® immunoassay showed longer seropositivity rates, while antibody titers over time decreased over time [40,49,50]. This is most likely explained by the influence of antibody avidity and the relative abundance of the different immunoglobulin classes in the competitive assay [49][50][51]. Recently, increased SARS-CoV-2-specific antibody levels (anti-spike and anti-nucleocapsid) were observed in patients with post-COVID syndrome 6 months after infection [33]. Likewise, we could show that anti-nucleocapsid antibody levels 5-6 months after PCR-confirmed acute infection were associated with post-COVID syndrome, significantly for the quantity and tendentially for the severity of the experienced post-infectious symptoms. In addition, we observed significantly increased anti-nucleocapsid antibody levels in post-COVID syndrome patients at an earlier time point 5-6 weeks after positive PCR, which might help to identify convalescents with an increased risk for post-COVID syndrome, as anti-nucleocapsid antibodies are not induced by most of the EMA-approved vaccines. Moreover, faster symptom resolution of acute COVID-19 is associated with higher anti-nucleocapsid antibody levels during the first week after diagnosis, which additionally underlines their importance [37].
In this work, we observed significantly increased anti-spike-antibody levels early and late after acute infection in convalescents with post-COVID syndrome, while the overall anti-spike-antibodies in all convalescents decreased over time. This is consistent with reports by others showing that anti-spike antibody levels wane over time and convalescents with more and more severe post-infectious symptoms have higher anti-spike antibody levels after infection [26][27][28][29][30][31][32]. Our results resemble the observations made by Durstenfeld et al. reporting on higher anti-receptor-binding-domain antibody levels in convalescents with post-infectious cardio-pulmonary symptoms 7 months after acute infection [35].
Besides the number and severity of post-infectious symptoms, we also assessed the character of post-infectious symptoms with SARS-CoV-2-specific humoral immune response. Convalescents reporting post-infectious neuropsychiatric symptoms had significantly higher anti-nucleocapsid antibody levels 5-6 weeks after PCR-confirmed COVID-19. This is of special importance, as post-infectious neuropsychiatric symptoms have been observed in many convalescents suffering from post-COVID syndrome [41,52]. In addition, we showed in convalescents with altered sensory perception significantly higher anti-nucleocapsid antibody levels, including both IgG and IgA, 5-6 weeks after positive PCR. Interestingly, others reported on lower anti-spike IgA antibody levels in convalescents with persisting post-infectious up to 3 months after acute infection anosmia or ageusia, highlighting the importance of IgA antibodies on mucosal tissue [38].
With increasing reports of post-COVID syndrome and post-infectious fatigue in SARS-CoV-2 convalescents, COVID-19, similar to many other viral infections (e.g., Epstein-Barr virus, human herpes virus-6, or the human parvovirus B19), has become increasingly discussed as a trigger for Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) [53,54]. In this study, we observed higher anti-spike antibody levels early after COVID-19 infection in convalescents reporting post-infectious "fatigue" and/or a "reduced performance". Of note, a post-infectious dysregulated immune response and autoimmunity, including the development of autoantibodies, have been intensively discussed as a key feature of the ME/CFS pathophysiology [53,55,56].
Together, this study provides further insights into the association of SARS-CoV-2specific humoral immune response with post-COVID syndrome: increased SARS-CoV-2specific antibody levels were observed early after infection in convalescents with more and more severe post-infectious symptoms approximately 6 months after infection, in particular in individuals suffering from neuro-psychiatric symptoms. Based on this observation, it is tempting to hypothesize that assessment of the humoral immune response, including anti-nucleocapsid antibody levels, can help to identify convalescents who are at increased risk for post-COVID syndrome.