Serology results after COVID vaccine in multiple sclerosis patients treated with fingolimod

Abstract

Introduction

While it is recommended that patients with multiple sclerosis (MS) be vaccinated against COVID-19, it is unknown what the vaccine response is in MS patients treated with fingolimod, an agent which modulates the humoral response. We aimed to characterize the immune response to the COVID-19 vaccine in MS patients treated with fingolimod and to explore which factors influenced response.

Method

We collected the following data from 59 MS patients treated with fingolimod and vaccinated against COVID-19: age, sex, duration of treatment, number of vaccine doses, date of last vaccination, type of vaccine, lymphocyte count, history of COVID-19, and serology to measure the vaccine response. We used Student's t-test and Chi² test to see whether there was a relationship between these variables and seropositivity. A multivariate logistic regression model was used to identify factors influencing the serology result. A multivariate linear regression model was used to identify factors influencing the antibody titer.

Results

Twenty-eight participants (47%) developed a positive serology. Age (P < 0.001) and the duration of treatment (P = 0.002) were significantly related to seropositivity. Gender (P = 0.73), number of vaccinations (P = 0.78), lymphocyte count (P = 0.46), and the time between the last vaccine dose and blood sampling (P = 0.84) were not significant variables. Multivariate analysis using logistic regression (n = 59) showed that age (P = 0.003, RR = 2.28, 95%CI = 1.28, 4.07) and duration of treatment (P = 0.04, RR = 1.91, 95%CI = 1.04, 3.50) were significantly and independently correlated with COVID serology. Multivariate linear regression analysis of the antibody titer (n = 59) found the duration of treatment to be significant (P = 0.015), but not age (P = 0.53). After removing three outliers, age (P = 0.005, RR = 6.82, 95%CI = 1.66, 27.98) and duration of treatment (P = 0.008, RR = 5.12, 95%CI = 1.24, 21.03) were significantly correlated with the antibody titer.

Conclusion

COVID-19 seropositivity was present in 47% of our sample of 59 MS patients on fingolimod. A strong relationship was found between antibody development, age, and duration of treatment, as well as between antibody titer and age and duration of treatment.

1. Introduction

Since December 2019, the world has experienced a pandemic following the outbreak of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1]. Several types of vaccines are currently available including mRNA vaccines (Moderna and Pfizer), and viral vector vaccines (Johnson&Johnson and Astrazeneca) [2]. These vaccines have already proven their efficacy and safety in the general population [3], [4], [5].

Several studies have already shown that multiple sclerosis (MS) is not a contraindication to vaccination (except for live attenuated vaccines which are contraindicated for patients on immunosuppressive therapies), and that patients vaccinated against COVID-19 do not experience increased disease activity. Consequently, MS patients are advised to be vaccinated early as they are often immunocompromised and therefore constitute a population at risk [6], [7].

MS is a chronic, autoimmune, inflammatory disease of the central nervous system in which the immune system attacks the patient's myelin sheath. Patients are administered immunosuppressive treatments to reduce this autoimmunity and limit relapses.

Fingolimod, an orally administered sphingosine 1 phosphate (S1P) receptor inhibitor, is one of the background treatments for patients with MS. This class of drugs works by blocking the migratory signal of S1P thus preventing the lymphocytes in lymphoid tissues from moving to peripheral tissues where they destroy the myelin sheath. The lymphocytes remain sequestered in the lymph nodes and are not altered [8], [9], [10], [11].

It has already been shown that, unlike other immunosuppressive background treatments, fingolimod does not increase the risk of developing severe COVID (requiring hospitalization) compared with the general population [12], [13], [14], [15], [16], [17]. Furthermore, those patients who do develop severe disease have similar risk factors to those of the general population of patients at risk [18]. This means that patients currently on fingolimod can continue their treatment during the pandemic which is reassuring as there is a risk of rebound after stopping fingolimod [19]. Background immunosuppressive treatment should always be introduced on a case-by-case basis, weighing up the benefits of each treatment on MS and the potential risk of severe COVID that the treatment could cause [20]. In terms of vaccine efficacy, disease-modifying therapies for MS can modulate the vaccine response. S1P modulators, in particular, can produce attenuated vaccine responses [21]. A randomised trial of vaccination in fingolimod-treated patients with MS showed that the immune response was lower in patients on fingolimod than in the control groups (placebo). It is therefore advisable to check the humoral response with serology [22].

The aim of the current study was to describe the qualitative (development of antibodies) and quantitative (antibody titer) immune response of the COVID-19 vaccine in MS patients treated with fingolimod, and to identify factors influencing response.

2. Materials and methods

2.1. Participants and study design

We included all patients with MS who consulted at the Neurology Department of Nancy University Hospital between 15/08/2021 and 09/12/2021 and who met the following criteria: patients vaccinated against COVID-19; patients treated by fingolimod.

For each patient, we collected the following data: age, sex, duration of fingolimod treatment, number of vaccine doses, date of last vaccination, type of vaccine, lymphocyte count, history of COVID-19, and the results and date of the serology showing the patient's vaccine response.

Antibodies were detected by DiaSorin's liaison SARS-COV2 trimericS IgG control set in a dry tube with gel separator. Lymphocytes were measured by complete blood count (CBC). Patients with BAU/mL values over 33.80 (threshold) were considered seropositive. The LIAISON® SARS-CoV-2 TrimericS IgG assay uses chemiluminescence immunoassay technology (CLIA) for the quantitative determination of specific IgG antibodies to the SARS-CoV-2 trimeric Spike protein in human serum and plasma samples. The main components of the test are magnetic particles (solid phase) coated with recombinant SARS-CoV-2 trimeric Spike protein and a conjugate reagent containing a mouse monoclonal anti-human IgG antibody linked to an isoluminol derivative (isoluminol-antibody conjugate). The amount of isoluminol-antibody conjugate is translated into a light signal and is measured by a photomultiplier as relative light units (RLU), which indicates the presence of anti-SARS-CoV-2 antibodies in standards, samples and controls.

2.2. Statistical analysis

Quantitative variables (age, duration of treatment, lymphocyte count, time between last vaccination and blood sampling, antibody titer) are described as means and standard deviations after checking the normality of the variables. Categorical variables (gender, number of doses) are described as numbers and proportions (Table 1 ).

Table 1.

Descriptive analyses (n = 59).

	Min	Q1	Median	Mean	Q3	Max	Standard deviation
Age (years)	19	34	44	43.29	54	65	11.84
Length of treatment (month)	6	20	44	50.63	82	126	33.13
Lymphocyte count (G/L)	0.2	0.38	0.479	0.51	0.61	1.419	0.19
Time between serology and last dose (month)	0	1	3	2.73	4	8	2.07
Antibodies titer (BAU/mL)	4.81	5.68	30.10	214.44	218	2080	472.58

A Student's t-test was used after checking the normality of the variables to compare age, duration of treatment, lymphocyte count, and time between the last vaccination and blood sampling between the group that developed antibodies and the group that did not. A Chi² test was used to check whether there was a relationship between seropositivity and age, gender, or the number of doses received. A multivariate logistic regression model was used to identify factors influencing the serology result. The variable to be explained was antibody development and the explanatory variables were age and duration of treatment. A multivariate linear regression model was used to identify factors influencing the antibody titer. The variable to be explained was the antibody titer, the explanatory variables were the duration of treatment and age. We defined age (40 years) and treatment duration (4 years), thresholds, according to literature data, that correspond to the median. We only kept variables with a P  < 0.10 in the multivariate model.

Statistical analyses were performed using Excel and Rstudio software.

2.3. Ethics

Patients were identified through the “Registre Lorrain de la Sclérose en Plaques” (RelSEP). All the patients gave their informed consent. Data collection was approved by the French National Commission for Data Protection and Liberties (CNIL No. 913001-2014.01.06).

3. Results

Of the 368 patients with MS who consulted at our Neurology Department during the study period, 59 patients met the inclusion criteria (43 women (73%) and 16 men (27%)) with an average age of 43 years (SD = 12). Thirty-three people (56%) had received two doses of vaccine, 23 (39%) three doses, and three (5%) one dose, two of whom had active COVID-19 infection at the time of sampling. Eight participants (13.6%) had a history of COVID infection. None of these eight patients required hospitalisation. Overall, 28 participants (47%) were seropositive.

Age was significantly the main explanatory factor for seropositivity (P  < 0.001) (Table 2 ). We modelled the proportion of COVID seropositivity in four age groups: under 30, between 30 and 39, between 40 and 50 and over 50. Seropositivity was found in eight of the nine patients under 30; eight of the 13 patients between 30 and 39; eight of the 17 patients between 40 and 50; and four of the 20 patients over 50 (Fig. 1 ).

Table 2.

Characteristics of patients with a positive and a negative serology.

	Positive serology	Negative serology	P-value
N	n = 28	n = 31	/
Sex (W/M)	21/7	22/9	P = 0.73
Age, years mean (SD)	37.07 (10.83)	48.90 (9.72)	P < 0.001
Duration of treatment, month mean (SD)	37.04 (27.56)	62.90 (32.94)	P = 0.002
Lymphocyte count, G/L mean (SD°)	0.486 (0.148)	0.524 (0.223)	P = 0.46
Time between last dose and blood sampling, month mean (SD)	3 (2.01)	2.68 (2.12)	P = 0.84

Fig. 1.

Proportion by age group of COVID seropositivity. Global P-value = 0.004.

Gender (P  = 0.73), number of vaccinations (P  = 0.78), lymphocyte count (P  = 0.46), and the time between the last vaccine dose and blood sampling (P  = 0.84) were not significant variables (Table 2). We added a univariate analysis to see if there was an association between any of these variables and the serology results, for example, gender, but we did not find a significant relationship (P  = 0.728).

The duration of fingolimod treatment was significantly related to COVID seropositivity (P  = 0.002) (Table 2). We modelled the proportion of COVID seropositivity in four treatment duration groups: less than 22 months, between 22 and 44 months, between 45 and 80 months, more than 80 months. Each group contained, respectively: 15 patients (of whom 11 were seropositive), 15 patients (of whom eight were seropositive), 14 patients (of whom six were seropositive), and 15 patients (of whom three were seropositive) (Fig. 2 ).

Fig. 2.

Proportion of COVID seropositivity for a given treatment du ration. Global P-value = 0.03.

In the multivariate logistic regression analysis (n  = 59), age (P  = 0.003) and duration of treatment (P  = 0.04) emerged as being significantly and independently correlated with COVID seropositivity (Table 3 ). Thus, the higher the age and the longer the duration of treatment, the greater the risk of having a negative serology.

Table 3.

Multivariate analysis (logistic regression): development of antibodies according to age and duration of treatment.

	RR (95% CI)	P-value
Age > 40 years	2.28 (1.28, 4.07)	0.003
Duration of Fingolimod treatment > 48 months	1.91 (1.04, 3.50)	0.04

The risk of being seronegative was 2.28 times greater after 40 years of age and 1.91 times greater after 48 months of treatment.

Multivariate linear regression analysis (n  = 59) showed a positive significance of the duration of treatment (P = 0.015) but not age (P = 0.53). We repeated this analysis by excluding three patients aged 46, 50 and 32 years (two of whom had had COVID) with an antibody titer greater than 2080 BAU/mL, which could bias the result. In the multivariate linear regression analysis with the remaining 56 patients, age (P = 0.005) and duration of treatment (P = 0.008) emerged as being significantly correlated with the antibody titer (Table 4, Table 5 ).

Table 4.

Descriptive analyses (n = 56).

	Min	Q1	Median	Mean	Q3	Max	Standard deviation
Age (years)	19	34.5	43	43.32	54	65	12.02
Duration of treatment (months)	6	25.5	44.5	52.14	82	126	33.04
Antibodies titer (BAU/mL)	4.81	5.41	23.35	114,53	110	838	197.16

Table 5.

Multivariate analysis (linear regression): antibody titer according to age and duration of treatment.

	RR (95% CI)	P-value
Age > 40 years	6.82 (1.66, 27.98)	0.005
Duration of Fingolimod treatment > 48 months	5.12 (1.24, 21.03)	0.008

The risk of having an antibody level < 114 BAU/mL was 6.82 times greater after 40 years and 5.12 times greater after 48 months of treatment.

Concerning COVID history, we added a univariate analysis to see if there was a possible association between covid history and serology results and found no significant result (P  = 0.112). However, 2 of the 3 patients had a very high antibody titer with a history of covid, so the antibody titer is significantly increased in patients with a history of COVID (P = 0.002) but the small number of patients does not allow us to retain this result.

4. Discussion

In this study investigating the immune response in 59 MS patients treated with fingolimod and vaccinated against COVID-19, nearly 50% were seropositive. We also showed a strong relationship between antibody development and age and duration of treatment as well as between antibody titer and age and duration of treatment.

Around 2.8 million people have MS and most are treated by a background immunotherapy such as fingolimod [23]. While the anti-COVID-19 vaccination is recommended for people with MS [24], more data are required to quantify the immune response in MS patients and to better understand which factors influence the response. This study provides additional data about MS patients treated with fingolimod and could contribute to supporting the current recommendations to vaccinate all MS patients.

Overall, 28 participants (47%) in our sample were seropositive, which is lower than the rate in the general population which is close to 100% for several types of vaccine [3], [4], [5]. Our results are in line with those of the randomised study by Kappos et al., which found that patients on fingolimod developed a diminished immune response compared with the placebo group [22]. Other studies have also found a diminished immune response in patients on fingolimod compared with patients treated with different disease-modifying therapies [25]. Finally, one study also showed an attenuated antibody response under immunosuppressive therapy (including fingolimod) [26] and a more recent paper shows that the vaccine response is attenuated in patients treated with fingolimod compared to other patients without background treatment [27]. We found that the development of antibodies was correlated to age: the proportion of people with a positive COVID serology decreases as age increases. This relationship between age and the development of antibodies was confirmed by logistic regression.

Linear regression did not find a positive significance of age, which was inconsistent with our results in terms of antibody development. We noticed that three of our patients–aged 46, 50 and 32 years (two of whom had had COVID) had an antibody titer greater than 2080 BAU/mL. Given that no other patient exceeded 1000 BAU/mL, we hypothesized that these outliers could distort the results. Once they were removed from analysis, age was found to be significantly correlated with antibody titer: the higher the age, the higher the risk of having a lower antibody titer.

The fact that our results showed an influence of age on vaccine response is in line with studies showing that vaccine responses in the elderly are weaker for several types of vaccine (excluding COVID) [28]. Although a few studies have shown that the elderly can develop robust vaccine responses against COVID [29], [30], many show that the vaccine response in this population is attenuated both qualitatively (antibody development) and quantitatively (antibody titer). A study of the COVID vaccine response in 45,965 adults from the general population in United Kingdom showed that seropositivity varied with age at the first vaccination, and that in patients with and without evidence of previous infection, younger subjects were more likely to be seropositive. The antibody titer followed the pattern of binary representation (older people are found to have lower antibody titer) [31]. Other studies on vaccine response tend to show similar results [32], [33], [34], [35], [36].

We demonstrate a significant association with the duration of treatment and the development of antibodies: the shorter the duration of treatment, the higher the probability of having a positive serology. Similarly, linear regression showed a positive significance of the duration of treatment with respect to antibody titer: the longer the duration of treatment, the greater the risk of having a lower antibody titer. The same was true for the patient sample once the outliers had been excluded.

Gender, the number of vaccinations, lymphocyte count, and time between last vaccine dose and blood sampling were not significant variables. According to the literature, gender may play a role in vaccine response [37], [38], [39], [40]. In addition, some studies show that the number of circulation lymphocytes could predict the vaccine response, which we do not find in our study [41], [42].

Our results are somewhat discrepant with previously published data on COVID seropositivity in MS patients treated by fingolimod [43]. For example, in the study conducted by Achiron et al., only one patient (out of 26) treated with fingolimod developed a positive serology to COVID, whereas in our sample almost half of the patients developed a humoral response. Furthermore, age did not seem to affect the vaccine response in their study whereas in our sample, age was significantly related to the development of antibodies. However, our results are similar to those published by Guerrieri et al. [44], which showed a positive COVID serological response in 10 of their 16 patients. They found no relationship between serological response and duration of treatment, time between vaccination and last dose of treatment, and white blood cell count. This is broadly consistent with the results of our study except for duration of treatment which was found to be significant in our sample.

The first limitation of our study is the relatively small sample size. Another limitation is that we did not measure cellular immunity: it is feasible that even if a patient did not develop humoral immunity, they might have cellular immunity and thus be protected. Tallantyre et al. studied the COVID-19 vaccine response in MS patients and, while they showed that seroconversion under fingolimod was lower, they also found that one of the six patients who had a negative humoral response had developed a cellular response [45]. An investigation of humoral, and T-cell-specific, immune responses to COVID vaccination in MS patients on different therapies by Tortorella et al. showed firstly a lower antibody response rate and antibody titer in patients on fingolimod compared with other disease-modifying therapies, and also that the lowest frequency of T-cell response was observed in patients treated with fingolimod [46]. The T-cell response in MS patients treated with fingolimod has been measured for other viruses, such as the varicella zoster virus, and the antiviral T-cell response was also lower [47]. Finally, one study recommends measuring cellular immunity to check vaccine efficacy in patients with treatments that have an action on B cells [48].

5. Conclusion

Our results show a COVID seropositivity of 47% in our sample of 59 MS patients on fingolimod. We also showed a strong relationship between antibody development and age and duration of treatment, as well as between antibody titer and age and duration of treatment. Therefore, we cannot afford not to recommend vaccination in MS patients treated with fingolimod, especially as research has shown that the third booster dose (mRNA vaccine in the study) revives the humoral response independently of any clinical variables [49]. Further studies on this topic are needed to support our results, especially regarding the influence of the duration of treatment on antibody development and titer (scarce data at present), although one study supports our point and also shows a relationship between treatment duration and vaccine response [50], ideally with a larger sample size and a measurement of cellular immunity.

Disclosure of interest

The authors declare that they have no competing interest.