Cross-sectional analysis of the humoral response after SARS-CoV-2 vaccination in Sardinian Multiple Sclerosis patients, a follow-up study

Study participants

847 MS patients who had received two BNT162b2 vaccine injections, 21 days apart, followed by a booster 6 month later were enrolled between October 2021 and January 2022 from the MS clinical centers of Cagliari and Sassari, Sardinia (Italy). Each injection contained 30μg of BNT162b2 (0.3ml volume per dose). 662 MS patients were analyzed 6 months after the second dose (T2) and 185 MS patients a month after the booster (third) dose (T3). Data from 912 MS patients at 4 weeks after the second dose (T1) were also used (2) (Figure 1).

• Download figure
• Open in new tab

Figure 1. Timeline of MS patients’ enrollment.

Schematic of the vaccination (blue) and sample collection (green) timeline in MS patients.

In summary we analyzed 1,944 serum samples of 1,307 unique MS patients. Of these, only 94 were recruited at each time point (T1, T2 and T3). The overlap between samples was 31.8% between T2 and T1, 22.8% between T2 and T3, and 13.9% between T3 and T1.

All MS patients, diagnosed according to the McDonald criteria, were contacted using different communication methods. They were questioned concerning previous SARS-CoV-2 infection and the manifestation of any adverse events following vaccination. Clinical and demographic information were collected for each patient, including age, sex, disability score, disease subcategory and disease-modifying treatment. All the data collected have been summarized in Table 1.

Detection of SARS-CoV-2 IgG antibodies

Blood samples were collected in vacutainer tubes containing clot activator and gel separator.

Samples were processed within two hours after blood collection to avoid time-dependent artifacts, and subsequently serum was stored at -80°C until use. Quantification of SARS-CoV-2 antibodies direct against the proteins Spike (S) or Nucleocapsid (N) in human serum was performed using the electrochemiluminescence immunoassays Elecsys® Anti-SARS-CoV-S and Elecsys® Anti-SARS-CoV-N (Roche). Anti-S and Anti-N results are expressed as units per ml (U/ml). The anti-N reactive index cutoff (COI) used is > 1.

Statistical analysis

For all categorical variables the absolute number and percentage were reported, and for quantitative variables the median and interquartile range (IQR) were reported. Differences between patient groups - stratified according to therapy and selected for Anti-N antibody negativity - were assessed with a negative binomial generalized linear mixed-effects model, which takes into account the nature of the outcome variable (anti-S, non-negative count data); in addition to therapy, the models also consider the contribution of other variables such as age, smoke, sex, Expanded Disability Status Scale (EDSS), disease duration, and clinical sampling center. Results are presented as Incidence Rate Ratio, calculated as the exponential of the regression coefficient (8). Differences between medians were tested with the Mann-Whitney test. Differences between groups in longitudinal data were tested using the non-parametric Friedman test for repeated measures, followed by a post-hoc pairwise comparison using paired Wilcoxon signed-rank test. All statistical analyses were performed using R v.4.0.3 software with the following CRAN libraries: ggplot2, dplyr, readxl, MASS, kableExtra, rstatix. P values <0.05 were considered statistically significant.

Ethics and data collection

The study was reviewed and approved by the Ethical Review Boards ATS Sardegna - Prot. N° 2492/CE. Patient data and samples were coded anonymously to ensure confidentiality during sample processing and data analysis. The patients/participants provided their written informed consent to participate in this study.

Description of the studied MS cohort

A total of 1,944 sera were analyzed at three different time points following BNT162b2 vaccination. Results from 912 MS patients obtained 4 weeks after the second dose (T1) of vaccine have already been published (2) and have been used here as baseline for the two subsequent time points. 662 MS patients were analyzed 6 months after the second dose (T2) and 185 MS patients 4 weeks after the booster dose (T3) (Figure 1).

The T1 cohort has been previously described (2). The T2 MS cohort included 448 (73.6%) female and 214 (26.4%) male patients of whom 84.9% had relapsing-remitting MS (RR), 1.51% had primary progressive MS (PP), and 13.6% had secondary progressive MS (SP). Finally, the T3 MS cohort included 137 (74.1%) female and 48 (25.9%) male patients of whom 89.7% had RR, 0.54% had PP and 9.7% had SP (Table 1).

Untreated MS patients at T2 and at T3 were respectively 73 (11%) and 13 (7%). The remaining 589 (88.9%) at T2 and 172 (92.9%) at T3 were treated with different DMTs. The most common treatments were dimethyl fumarate (DMF) at T2 and FTY at T3. DMTs used less frequently included cladribine (CLA) and RTX, only one patient included in a study with a Bruton’s tyrosine kinase (BTK) inhibitor. No patients treated with methotrexate (MET) were present either at T2 or T3. A detailed description of the specific DMTs received by the MS patients along with their demographic characteristics at T1, T2 and T3 are summarized in Table 1.

Disease-Modifying Therapies Impact Humoral Responses to BNT162b2 Vaccine

To evaluate the effects of different DMTs on humoral responses to BNT162b2 vaccine, we applied a negative binomial generalized linear mixed-effects model in patients negative for anti-N antibodies production for both T2 and T3 time points separately. Only treatments with data available for at least 10 patients were considered in this analysis.

In line with previously reported humoral responses to vaccine one month after the second dose (T1) (2,7), after six months (T2) we observed a significant difference in anti-S antibody levels between patients untreated (UNT) and those treated with FTY (IRR = 0.17, p = 1.82×10⁻²¹), OCR (IRR = 0.20, p = 4.54×10⁻⁴²) and RTX (IRR = 0.33, p = 1.35×10⁻¹⁷). No significant difference was observed for the other DMTs (Table 2 and Figure 2A). Similarly, one month after the booster (T3) we observed a significantly lower level of anti-S antibodies in MS patients treated with FTY (IRR = 0.37, p = 1.52×10⁻¹¹) and OCR (IRR = 0.34, p = 1.20×10⁻¹⁴) compared to untreated patients. Other treatments did not show significant results (Table 3 and Figure 2B).

• Download figure
• Open in new tab

Figure 2. Post-vaccination SARS-CoV-2 anti-S antibody response by disease-modifying therapy (DMT) in MS patients negative for anti-N antibodies.

(A) Antibody response to SARS-CoV-2 vaccination by DMT in MS patients six month after SARS-CoV-2 vaccination (T2). (B) Antibody response one month after booster (T3) by DMT in MS patients. Results are reported as boxplots, showing the median value (in bold) and the quartiles as box limits; whiskers at the top and bottom sides represent the overall maximum value and the overall minimum value, respectively. Data outside boxes and whiskers represent the outliers of the distribution. (*P< 0.05; **P<0.01; ***<0.001).

(UNT=untreated; ALEM=alemtuzumab; IFN=interferon; GA=glatiramer acetate; DMF=dimethyl fumarate; NAT=natalizumab; CLA=cladribine; TER=terifluomide; AZA= azathioprine; FTY=fingolimod; RTX=rituximab; OCR=ocrelizumab)

The nucleocapsid protein antigens analysis, used to discriminate the immune response generated by the vaccination from immune response generated by natural SARS-CoV-2 infection, identified 23 anti-N positive patients at T2 (3.5% of 662 patients) and 9 at T3 (4.9% of 185 patients) (Table 1). Prior natural SARS-CoV-2 infection impact the humoral responses to BNT162b2 vaccine (2). Indeed, in MS patients with evidence of a natural exposure to SARS-CoV-2 virus, postvaccination anti-S antibodies levels were significantly higher than in patients who did not experience SARS-CoV-2 infection at T2 (medians 2,500 vs. 303.4 U/ml, Mann-Whitney test p = 2.36×10⁻¹¹). We observed a comparable trend at T3 (medians 10,965 vs. 9,600 U/ml, Mann-Whitney test p = 0.4816). However, the number of MS patients positive for anti-N proteins was small, especially at T3, and additional data are required to properly evaluate the impact of SARS-CoV-2 infection on the immune response to BNT162b2 vaccine.

The influence of disability status, sex, age, and smoking on anti-S production after BNT162b2 vaccine

We also tested at T2 and T3 the impact of additional factors that had been shown to influence anti-S production after BNT162b2 vaccine at T1 (2). In contrast with findings at T1, at T2 our statistical model did not reveal additional significant effects for sex, while age and the Expanded Disability Status Scale (EDSS) continue to show consistent effects. Specifically, we observed reduced postvaccination levels of anti-SARS-CoV-2 antibodies directed against the S protein in older patients (IRR = 0.98, p = 4.459×10⁻⁰⁷) as well as in patients with reduced EDSS (IRR = 0.95, p = 0.022). At T3 only the reduced EDSS showed a significant effect (IRR = 0.89, p = 0.01) on anti-SARS-CoV-2 antibodies production.

Due to the impact of smoking on antibodies production in healthy and unhealthy cohorts of smokers compared to non-smokers (2,9), we examined the effects of active cigarette smoking on humoral response to SARS-CoV-2 vaccine in a subset of MS patients negative for anti-N antibodies production for whom smoking status was available (T2=510, T3=157). Our analyses showed that active cigarettes smoking reduced anti-S antibodies production (median = 128.0 U/ml) compared to non-smokers (median = 349.6 U/ml) in response to BNT162b2 vaccine (Mann-Whitney test p = 4.9×10⁻⁴) at T2. No differences were observed at T3 (Mann-Whitney test p = 0.3). The discrepancy between the three time points analyzed could be due to the effect of vaccination over time or the sample size, significantly smaller at T3 than T2.

Cross sectional and longitudinal study on anti-S and anti-N SARS-CoV-2 antibody levels

Overall, evaluating the median of anti-S antibody level at each timepoint (T1-T2-T3) (Figure 1), we found that 6 months after the second dose of BNT162b2 vaccination (T2), the levels of anti-S decrease 3-fold compared to T1 (median T1= 962.2 U/ml and T2 = 323.7 U/ml). Furthermore, in line with previous reports (10), a strong upregulation in serum anti-S antibody levels was observed a month after the BNT162b2 booster (median T3 = 9,758 U/ml). The anti-S antibody median at T3 is ∼30-fold higher than T2 and ∼10-fold higher than T1. After stratification for each DMTs analyzed in this study, we observed that antibody levels significantly decrease when comparing T1 with T2 for the following treatments: interferon (IFN), DMF, natalizumab (NAT) and teriflunomide (TER) (Figure 3). Furthermore, the booster significantly increased anti-S production for DMF, NAT, TER, FTY and OCR. Interestingly, the increase in anti-S antibodies due to the booster was also observed for FTY and OCR even if with a lower impact (Figure 3). The upregulation of SARS-CoV-2 anti-S antibodies at 4 weeks after the booster (T3) was confirmed in an additional longitudinal analysis of a subgroup of 94 MS patients with data at every time points (Median T1= 888 U/ml, T2= 562 U/ml and T3 = 13,346 U/ml) (Table 4).

• Download figure
• Open in new tab

Figure 3. Cross-sectional humoral response over time (T1, T2 and T3) in MS patients treated with different DMTs or untreated.

Post-vaccination anti-S antibody response by disease-modifying therapy (DMT) in MS patients. Results are represented as histogram showing the median value. Significance was tested using Mann-Whitney test. (*P< 0.05; **P<0.01; ***<0.001).

(UNT=untreated; ALEM=alemtuzumab; IFN=interferon; GA=glatiramer acetate; DMF=dimethyl fumarate; NAT=natalizumab; CLA=cladribine; TER=terifluomide; AZA= azathioprine; FTY=fingolimod; RTX=rituximab; OCR=ocrelizumab)

Longitudinal analysis of anti-N antibodies also revealed a progressive decline in anti-N proteins from T1 to T3. Indeed, all 13 MS patients positive for SARS-CoV-2 infection at T1 (as determined by anti-N positivity), and with data for at least another time point, were anti-N negative ∼7-9 month later demonstrating a consistent loss of circulating humoral response against N antigens over time (data not shown). It is important to take this into account when testing samples without information on COVID-19 disease status. Indeed, if exposure occurred well before the time of antibody testing, it may be impossible to detect individuals with natural humoral immunity against the virus.

Impact of different DMTs on COVID-19 risk and severity

To evaluate the risk of COVID-19 in patients receiving anti-CD20 (OCR and RTX) and FTY therapies, which are associated with reduced humoral responses to vaccine, we assessed the number of MS patients who became positive for COVID-19 five months after the booster. The presence of symptomatic infection and its severity was evaluated. A subset of 140 MS patients was monitored and among them 52 patients (37%) were treated with anti-CD20 or FTY while 88 (62.9%) received other therapies or were untreated. Overall, 28 MS patient (20%) of the 140 examined became positive for COVID-19 during the five months following the booster. Among them, COVID-19 infection was significantly more common (risk ratio = 1.95, 95% C.I =1.01 - 3.77, p=0.046) in patients under anti-CD20 and FTY therapies as compared to patients receiving other therapies or untreated. The higher incidence of disease in the group treated with anti-CD20 and FTY compared with the other therapies correlates with the lower antibody titers observed in those DMT categories. Importantly, regardless of the DMT received, all patients reported mild symptoms that did not require hospitalization.