key: cord-0268138-0rxvm8pa authors: Idda, M. L.; Pitzalis, M.; Lodde, V.; Loizedda, A.; Frau, J.; Lobina, M.; Zoledziewska, M.; Virdis, F.; Delogu, G.; Marini, M. G.; Mingoia, M.; Masala, M.; Lorefice, L.; Fronza, M.; Carmagnini, D.; Carta, E.; Pilotto, S.; Castiglia, P.; Chessa, P.; Uzzau, S.; Farina, G.; Solla, P.; Steri, M.; Devoto, M.; Fiorillo, E.; Floris, M.; Zarbo, I. R.; Cocco, E.; Cucca, F. title: Cross-sectional analysis of the humoral response after SARS-CoV-2 vaccination in Sardinian Multiple Sclerosis patients, a follow-up study date: 2022-05-20 journal: nan DOI: 10.1101/2022.05.19.22275317 sha: 8c34ceb3714546e6308c174a21ecd64d7418642a doc_id: 268138 cord_uid: 0rxvm8pa Monitoring immune responses to SARS-CoV-2 vaccination and its clinical efficacy over time in Multiple Sclerosis (MS) patients treated with disease-modifying therapies (DMTs) help to establish the optimal strategies to ensure adequate COVID-19 protection without compromising disease control offered by DMTs. Following our previous observations of the humoral response one month after two doses of BNT162b2 vaccine (T1) in MS patients differently treated, here we present a cross-sectional and longitudinal follow-up analysis six months following vaccination (T2, n=662) and a month following the first booster (T3, n=185). Consistent with results at T1, humoral responses were decreased in MS patients treated with fingolimod and anti-CD20 therapies compared with untreated patients also at the time points considered here (T2 and T3). Interestingly, a strong upregulation one month after the booster was observed in patients under every DMTs analyzed, including those treated with fingolimod and anti-CD20 therapies. And although patients taking these latter therapies had a higher rate of COVID-19 infection five months after the first booster, only mild symptoms that did not require hospitalization were reported for all the DMTs analyzed here. Based on these findings we anticipate that additional vaccine booster shots will likely further improve immune responses and COVID-19 protection in MS patients treated with any DMT. General population data on SARS-CoV-2 vaccination support its effectiveness in preventing 49 infection (1). Still, the magnitude of protection it offers to Multiple Sclerosis (MS) patients receiving certain 50 disease-modifying therapies (DMTs) is not completely clear. Several evidence have already demonstrated that the humoral and cellular response after SARS-CoV-2 53 vaccination were strongly affected by the treatment with certain DMTs used to ameliorate MS symptoms (2-54 4). Notably, azathioprine (AZA), fingolimod (FTY) and anti-CD20 treatments, including ocrelizumab (OCR) 55 and rituximab (RTX), negatively influences the humoral response after SARS-CoV-2 vaccination and likely 56 affect the level of protection against . Additionally, older age, male sex and active smoking 57 were significantly associated with lower antibody titers against SARS-CoV-2 vaccine in MS patients (2,5). Based on these results, it has been suggested that the immune response after SARS-CoV-2 vaccination can 60 be improved through additional booster shots, and by optimally adjusting vaccination timing based on the 61 timing of immune cell repopulation after the last administration of specific immunosuppressive DMTs (6). Specifically, in contrast to the initial international recommendation for timing DMTs (1) of a 3-month 63 waiting time after the last dose of immune-suppressive therapies, we and others have recommended a 6-64 month waiting time (2,7). It is now important to further monitor, over time and after booster doses, the 65 immune responses and clinical efficacy of SARS-CoV-2 vaccination in MS patients. This will provide new 66 insight to define the most effective strategies that will ensure optimal treatment of MS providing at the same 67 time the most effective prevention of COVID-19 and especially of its severe forms. Following our initial observations of humoral response one month after two doses of BNT162b2 vaccine in 70 MS patients treated with different DMTs or untreated (2) here we present a cross-sectional and longitudinal 71 follow-up analysis on the humoral responses to BNT162b2 vaccination 6 months after the second dose, and a 72 month after the third dose (booster). The effect of previous or concomitant detectable SARS-CoV-2 73 infection, as well as age, sex, and active smoking was also considered. Reciprocally, we provide preliminary 74 evidence of the impact of vaccination on the incidence of SARS-CoV-2 infection and its clinical severity 5 75 months after the third dose of vaccine. The findings further help in defining an appropriate immunization 76 strategy in MS patients in relation to DMTs and other factors influencing humoral immunity. 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. 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 100 subsequently serum was stored at -80°C until use. Quantification of SARS-CoV-2 antibodies direct against Patient data and samples were coded anonymously to ensure confidentiality during sample processing and 124 data analysis. The patients/participants provided their written informed consent to participate in this study. 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. (2) and have been used here as baseline for the two subsequent time points. 662 MS patients were analyzed 6 133 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) Untreated MS patients at T2 and at T3 were respectively 73 (11%) and 13 (7%). The remaining 589 (88.9%) 141 at T2 and 172 (92.9%) at T3 were treated with different DMTs. The most common treatments were dimethyl 142 fumarate (DMF) at T2 and FTY at T3. DMTs used less frequently included cladribine (CLA) and RTX, only 143 one patient included in a study with a Bruton's tyrosine kinase (BTK) inhibitor. No patients treated with 144 methotrexate (MET) were present either at T2 or T3. A detailed description of the specific DMTs received 145 by the MS patients along with their demographic characteristics at T1, T2 and T3 are summarized in Table 146 1. To evaluate the effects of different DMTs on humoral responses to BNT162b2 vaccine, we applied a 150 negative binomial generalized linear mixed-effects model in patients negative for anti-N antibodies 151 production for both T2 and T3 time points separately. Only treatments with data available for at least 10 152 patients were considered in this analysis. In line with previously reported humoral responses to vaccine one month after the second dose (T1) (2,7), 154 after six months (T2) we observed a significant difference in anti-S antibody levels between patients 155 untreated (UNT) and those treated with FTY (IRR = 0.17, p = 1.82x10 -21 ), OCR (IRR = 0.20, p = 4.54x10 -42 ) 156 and RTX (IRR = 0.33, p = 1.35x10 -17 ). No significant difference was observed for the other DMTs ( Table 2 157 and Figure 2A) . Similarly, one month after the booster (T3) we observed a significantly lower level of anti-158 S antibodies in MS patients treated with FTY (IRR = 0.37, p = 1.52x10 -11 ) and OCR (IRR = 0.34, p = 159 1.20x10 -14 ) compared to untreated patients. Other treatments did not show significant results (Table 3 and 160 Figure 2B ). The nucleocapsid protein antigens analysis, used to discriminate the immune response generated by the 163 vaccination from immune response generated by natural SARS-CoV-2 infection, identified 23 anti-N 164 positive patients at T2 (3.5% of 662 patients) and 9 at T3 (4.9% of 185 patients) ( Table 1) . Prior natural 165 SARS-CoV-2 infection impact the humoral responses to BNT162b2 vaccine (2). Indeed, in MS patients with 166 evidence of a natural exposure to SARS-CoV-2 virus, postvaccination anti-S antibodies levels were 167 significantly higher than in patients who did not experience SARS-CoV-2 infection at T2 (medians 2,500 vs. 168 303.4 U/ml, Mann-Whitney test p = 2.36x10 -11 ). We observed a comparable trend at T3 (medians 10,965 vs. The influence of disability status, sex, age, and smoking on anti-S production after BNT162b2 vaccine 174 We also tested at T2 and T3 the impact of additional factors that had been shown to influence anti-S 175 production after BNT162b2 vaccine at T1 (2). In contrast with findings at T1, at T2 our statistical model did 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 20, 2022. ; https://doi.org/10.1101/2022.05.19.22275317 doi: medRxiv preprint 5 Due to the impact of smoking on antibodies production in healthy and unhealthy cohorts of smokers 183 compared to non-smokers (2,9), we examined the effects of active cigarette smoking on humoral response to 184 SARS-CoV-2 vaccine in a subset of MS patients negative for anti-N antibodies production for whom 185 smoking status was available (T2=510, T3=157). Our analyses showed that active cigarettes smoking 186 reduced anti-S antibodies production (median = 128.0 U/ml) compared to non-smokers (median = 349.6 187 U/ml) in response to BNT162b2 vaccine (Mann-Whitney test p = 4.9X10 -4 ) at T2. No differences were 188 observed at T3 (Mann-Whitney test p = 0.3). The discrepancy between the three time points analyzed could 189 be due to the effect of vaccination over time or the sample size, significantly smaller at T3 than T2. Overall, evaluating the median of anti-S antibody level at each timepoint (T1-T2-T3) (Figure 1) , we found 193 that 6 months after the second dose of BNT162b2 vaccination (T2), the levels of anti-S decrease 3-fold 194 compared to T1 (median T1= 962.2 U/ml and T2 = 323.7 U/ml). Furthermore, in line with previous reports 195 (10), a strong upregulation in serum anti-S antibody levels was observed a month after the BNT162b2 196 booster (median T3 = 9,758 U/ml). The anti-S antibody median at T3 is ∼30-fold higher than T2 and ∼10-197 fold higher than T1. After stratification for each DMTs analyzed in this study, we observed that antibody 198 levels significantly decrease when comparing T1 with T2 for the following treatments: interferon (IFN), 199 DMF, natalizumab (NAT) and teriflunomide (TER) (Figure 3) . Furthermore, the booster significantly 200 increased anti-S production for DMF, NAT, TER, FTY and OCR. Interestingly, the increase in anti-S 201 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) 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. Regarding the effects of the first booster dose of vaccine, at T3 we observed a strong upregulation of the 241 humoral response in both treated and untreated patients, although the humoral responses observed in patients 242 treated with anti-CD20 (OCR) and FTY remained significantly lower than those observed in untreated 243 patients (Table 3 and 4) . Yet, the first booster dose significantly increases anti-S levels even in most patients 244 taking these immunosuppressive therapies. For example, in anti-N negative patients under OCR and FTY the 245 median anti-S antibody levels were, respectively, 40.5 and 28.6-fold higher at T3 than at T1. Likewise, prior natural SARS-CoV-2 infection, documented through the presence of anti-N antibodies, also 248 strongly potentiated the humoral responses to BNT162b2 vaccine by inducing a strong increase in levels of 249 anti-S antibodies at both T2 and T3 regardless of the DMTs analyzed and including, albeit to a lesser extent, 250 patients treated with anti-CD20 or FTY. Anti-N antibody analysis also provided additional data on the rate of decline in anti-N seropositivity which is 253 essential for identifying patients with asymptomatic SARS-CoV-2 infection. We found that anti-N antibody 254 levels declined at a sharper rate over time until they were no longer detectable, at least with the assay used in 255 this study, 7 to 9 months after infection. Thus, it is possible that natural asymptomatic SARS-CoV-2 256 infections during the early phase of the pandemic were not detected in some of the MS patients considered 257 here, and that these early infections still had a residual but not ascertainable effect on anti-S antibodies 258 produced after vaccination. This is in agreement with a cross-sectional study by Alfego and colleagues that 259 showed a decay rate of anti-N responses of 68.2% after only 9.7 months following infection, whereas anti-S 260 seropositivity maintained a positivity rate of 87.8% 10 months later (11). In addition to DMTs and prior SARS-CoV-2 infection, in agreement with previous result at T1 (2), we 263 confirmed that both older age and reduced EDSS also influence antibody response to vaccine. Anti-S 264 antibodies levels, however, were significantly lower in older patients at T2 but not a T3. The discrepancy 265 between T1, T2 and T3 is probably due to the smaller sample size we analyzed at T3. Similar findings have 266 been reported in a general population (12) where older age was associated with lower seroconversion rates. By contrast, Cohen and colleague have reported associations between older age and higher immune 268 responses to natural infection, including IgG neutralizing antibody and memory B cell levels (13). We also evaluated the relationship between antibodies titers in response to BNT162b2 vaccine and active (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 20, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 7 be thus expanded to larger case series and further followed over time, including consideration of the immune 285 and clinical impact of additional vaccine booster shots. In this regard, the results presented here refer to the evaluation of the effects of the initial anti-Sars-Cov2 288 vaccination followed by a single booster dose of the vaccine completed several months ago and should 289 therefore be placed in a more current context. Initial reports suggest that an additional booster -already 290 approved by many regulatory agencies for the elderly and immunocompromised individuals -improves 291 protection without affecting safety (15) We thank all the patients and volunteers who generously participated in this study. We would also like to 308 thank all the medical doctors, nurses, and students who have collaborated to the realization of this study. 309 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. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 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 20, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 Impact of vaccination on new SARS-CoV-2 infections in the United 313 Effect of Different Disease-Modifying Therapies on Humoral Response to 316 BNT162b2 Vaccine in Sardinian Multiple Sclerosis Patients COVID-19 vaccination in patients with multiple sclerosis: What we have 320 learnt by Cellular and humoral immune responses following SARS-CoV-2 mRNA 323 vaccination in patients with multiple sclerosis on anti-CD20 therapy Determinants of early antibody responses to COVID-19 mRNA vaccines in a cohort 327 of exposed and naïve healthcare workers Vaccinations in multiple sclerosis patients receiving 330 disease-modifying drugs Effect of SARS-CoV-2 mRNA vaccination in MS patients treated with 334 disease modifying therapies Negative binomial regression. Cambridge Central obesity, smoking habit, and hypertension are associated with lower antibody 338 titres in response to COVID 19 mRNA vaccine COVID-19 vaccination in patients with multiple sclerosis: Safety and humoral 9 2 infection with persisting antibody responses and memory B and T cells Association studies of up to 1.2 million individuals yield 356 new insights into the genetic etiology of tobacco and alcohol use Effectiveness of a second BNT162b2 booster vaccine against hospitalization and death from COVID-360 19 in adults aged over 60 years Abbreviations: UNT=untreated; ALEM=alemtuzumab; IFN=interferon; GA=glatiramer acetate; DMF=dimethyl 421 fumarate; NAT=natalizumab; CLA=cladribine; TER=teriflunomide; AZA=azathioprine; FTY=fingolimod; 422 RTX=rituximab; OCR=ocrelizumab 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 20, 2022. ; https://doi.org/10.1101/2022.05.19.22275317 doi: medRxiv preprint 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. (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. 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 20, 2022. ; https://doi.org/10.1101 /2022 doi: medRxiv preprint 511 512 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 20, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022 17 524 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 20, 2022. ; https://doi.org/10.1101 https://doi.org/10. /2022