John Cunningham Virus Status, Seroconversion Rate, and the Risk of Progressive Multifocal Leukoencephalopathy in Polish John Cunningham Virus-Seronegative Patients with Relapsing-Remitting Multiple Sclerosis

Multiple sclerosis (MS) is a chronic, inflammatory, and degenerative demyelinating disease of the central nervous system (CNS), leading to disability among young adults [1]. Natalizumab (NAT), a monoclonal antibody to α4β1 and α4β7 integrins located on the surface of lymphocytes, prevents lymphocytes from adhering to the endothelium of the blood-brain barrier and decreases their infiltration into the CNS. It is a highly effective treatment in patients with relapsing-remitting multiple sclerosis (RRMS) since it reduces relapse rate, disability progression, and number of new lesions on MRI [2].

The most serious adverse event related with NAT treatment is progressive multifocal leukoencephalopathy (PML) – a potentially fatal infection of the CNS caused by reactivation of the John Cunningham virus (JCV). The major risk factors for PML on NAT therapy are anti-JCV-positive status, prior exposure to immunosuppressive drugs, and NAT treatment duration especially longer than 2 years [3]. The incidence of PML in seropositive patients without prior immunosuppressant treatment was <1/1,000 during the first 24 months of NAT therapy and increased to 6/1,000 after 49–72 months of treatment and with prior immunosuppressant was 1/1,000 during the first 24 months and increased to 13/1,000 after 49–72 months. Incidence of PML among negative for JCV antibodies (JCVAbs) patients during NAT treatment was estimated <1/1,000 patients [4]. But little is known about the true risk of PML among negative for JCVAb patients, given that the JCVAb assay has an estimated analytical false-negative rate of 3% [4]. It is unquestionable that MS patients who are JCVAb seronegative have a lower risk of developing PML, but they can become seropositive during treatment with NAT [5, 6]. In Poland, according to the National Health Fund (NFZ) drug program guidance, only patients with negative JCVAb test have been able to start treatment with NAT. The aim of the study was to assess the true risk of PML, seroconversion rate, and time to seroconversion in all JCVAb-negative RRMS patients treated with NAT in Poland.

It is an observational multicenter prospective study assessing the true risk of PML, JCVAb status, seroconversion rate, and time to seroconversion in Polish patients with RRMS receiving NAT. Demographic and clinical data of all Polish RRMS patients treated with NAT reimbursed by NFZ were prospectively collected in electronic files using the Therapeutic Programme Monitoring System in years 2013–2018. In Poland, NAT is used as a second-line treatment after the first-line disease-modifying therapy’s ineffectiveness or in rapidly evolving severe MS cases. The assessment of JCVAb presence (without collection of JCVAb index value) in serum at the beginning of NAT therapy and repeated every 6 months was done using ELISA (STRATIFY JCV® antibody ELISA) performed at Unilabs, Copenhagen, Denmark. The maximum follow-up time was 4 years. In Poland, according to the NFZ drug program guidance, only patients with negative JCVAb test have started treatment with NAT but could continue treatment in case of seroconversion. Seroconversion was defined as a presence of JCVAb at least once during follow-up after baseline seronegative status. Seroreversion was defined as negative JCVAb test at least once during the observation period after seropositive status. Neurological examination including Expanded Disability Status Scale (EDSS) scoring was performed at baseline and then every 12 months.

The JCVAb seroconversion probability was computed using the Kaplan-Meier estimator accounting for right censoring; it estimated the JCVAb seroconversion probability at any time since prescription starts along with a log-log estimation of the 95% confidence interval (CI). The JCVAb-seroconversion Kaplan-Meier estimator was the response of interest.

The effects of patient age, disease status, and disease duration were investigated using univariate Cox proportional hazards regression models. The candidate explaining variables were more precisely:

• The age at JCV seroconversion or censoring (last available observation while JCVAb seroconversion was not observed).

• The disease duration from first symptoms to JCVAb seroconversion or censoring.

• EDSS at prescription start.

• The last recorded EDSS for the prescription.

The study included 210 JCVAb seronegative patients with RRMS (146 women and 64 men). The median age of the study group was 33 years at baseline, with male patients starting NAT prescription 2.59 years younger than women (p = 0.014, shown in Fig. 1). RRMS patients were treated with NAT for at least 9 months, with a maximum follow-up time of 4 years. Patient characteristics are shown in Tables 1 and 2.

During the follow-up period, JCVAb status changed from negative to positive in 34 patients (16.2%). Seroconversion was observed within 1 year (±3 months) of NAT therapy among 52.9% RRMS patients (shown in Fig. 2; Table 3). Probability of being seroconversion-free was evaluated using a Kaplan-Meier estimator, involving the right censoring to account for the reduction of the patient panel size with time. After 12 months of treatment, it was 0.93 (0.90–0.97, 95% CI), and after 24 months, it was 0.8 (0.72–0.86, 95% CI) (Table 3). Observation therefore supports a decreasing probability of being seroconversion-free with the duration of the treatment. It means that the probability of seroconversion increased with the time between 0 and 24 months after prescription start. The decrease in the number of patients monitored during the follow-up data beyond 24 months makes data inconclusive.

The JCVAb status changed from positive to negative in 4 patients (1.9%) during the period of observation up to 48 (±3) months. Multivariate Cox proportional hazards regression models showed that patients’ age was not related with the risk of JCVAb seroconversion or censoring (last available observation while JCVAb seroconversion) (p = 0.5). EDSS at treatment start and the last recorded EDSS for the treatment period were not significantly associated with an alteration of JCVAb seroconversion risk (p = 0.22). The disease duration from first symptoms to JCVAb seroconversion or censoring was not related with JCVAb seroconversion (p = 0.94, shown in Table 4). The probability of JCVAb seroconversion increases with treatment duration (Kaplan-Meier estimators, shown in Fig. 2; Table 3). In the NFZ drug program, since its establishment in 2013, no patient treated with NAT has been diagnosed with PML.

In Poland, in a drug program, since 2013, no patient treated with NAT has been diagnosed with PML. Before establishment of NFZ drug program, 4 cases of PML in patients treated with NAT in clinical trials were diagnosed. Our study showed that NAT therapy in JCVAb-seronegative RRMS patients living in Poland in a high-risk area for MS is safe and results in the absence of PML cases [7, 8]. Although seronegative patients have a low risk of PML, repeated measurements of JCVAb are necessary to exclude the possibility of seroconversion during treatment [9, 10] since a case of PML was diagnosed in a patient who had been seronegative for JCVAb 2 weeks before symptoms of this opportunistic infection [10, 11].

Findings from sero-epidemiological studies in which JCVAb has been measured show a worldwide distribution of JCV. More than 50% of the adult population is estimated to have been exposed [3, 12]. A correlation has been reported between the percentage of the seropositive population and decade of life, from 15% in the second decade to 80% in the seventh and eighth decades of life [3]. The median worldwide prevalence of JCVAb among adults with MS or neuromyelitis optica was found to be 58.0% [13]. Previous reports mentioned seropositivity rates between 50.0 and 90.0% [13, 14], but the present review showed that it was between 40.0 and 69.0% [13].

However, patients can convert from seronegative to seropositive, and vice versa, as antibody concentrations fluctuate over time. Furthermore, longitudinal observations of multiple JCV testing showed that most patients present stable JCVAb status and index values over years, although the different worldwide seroconversion rates have been described [15, 16]. Plavina and collaborators stated that 87% of patients with negative JCVAb remained negative during 18 months observational period and only 13% turned from negative to positive serostatus [17]. Seropositive patients with low JCVAb index (<0.9) must be screened for JCVAb determination every 6 months and annually by MRI, during NAT treatment. Also, MS patients who were seronegative before NAT therapy must be screened every 6 months because of possible seroconversion, secondary JCV infection, or initial false-negative assay [18].

According to the results of our study, the JCVAb status changed from negative to positive in 16.2% patients. Seroconversion was observed within 1 year (±3 months) of NAT therapy among 52.9% RRMS patients. The results obtained in our study are similar to previously published data, which showed the worldwide JCVAb seroconversion rates in RRMS patients [19‒21]. In the systematic review and meta-analysis, the incidence of seroconversion was reported between 6 and 41%. The pooled estimate of seroconversion incidence was 19%, which means that in 190 per 1,000 MS cases who had been treated with NAT, anti-JCV antibody status had changed from negative to positive. Subgroup analysis by considering the country of the origin showed that the pooled incidence of seroconversion incidence during the studies was 6% in Asian countries and 21% in European/American countries [6].

In the study published by Jan Kolcava et al. [19], there was no significant correlation between JCVAb index and NAT treatment duration, but there was a positive correlation between the baseline JCVAb index and patient’s age. In contrary to these results, in our study, patients’ age, disease duration, and baseline EDSS score seem not to be a risk factor for seroconversion in Polish patients with RRMS. In our study, there was an increase in the seroconversion rate with NAT treatment duration. The association between NAT exposure and increased seropositivity has been previously reported, with seroconversion rates calculated to be between 8 and 10.3% per year [10, 21‒23]. The increase in the seroconversion rate with NAT treatment duration has also been reported in another study, in which the annualized seroconversion rates were 12% during the first 2 years, 18% between the second and fourth years, and 16% between the fourth and seventh [10, 24]. These results may be explained by a phenomenon of an asymptomatic reactivation and enhanced replication of JCV under NAT therapy. NAT leads to a reduction in CD4, CD8, and Th17 cells and significantly lowers CSF immunoglobulin M and G levels, contributing to a decline in immunity to the JCV within prolonged treatment duration [25]. Therefore, the duration of treatment with NAT may be a confounding factor on the prevalence of JCVAb in patients with MS [13]. This leads to further decreased specificity over time and periods of unknown increased risk depending on how often serology is assessed [26].

According to the results of our study, the JCVAb status changed from positive to negative in 4 patients (1.9%) during the period of observation up to 48 (±3) months and was in line with other studies [19]. In the systematic review and meta-analysis, the incidence of seroreversion was reported between 1 and 11% [6]. In the same study, the pooled estimate of seroreversion incidence rate was 5%, meaning that in 50 per 1,000 MS cases who were treated with NAT, anti-JCV antibody status changed from positive to negative.

We would like to emphasize that there is a high need for optimization of the benefit-risk factors in case of treatment with NAT. The management of patients with RRMS is challenging, and NAT still remains one of the most effective therapeutic options in RRMS. The results of our study confirm the importance of the JCVAb assessment as a tool for PML risk stratification. NAT therapy conducted according to the rules of PML risk stratification seems to be safe. In agreement with the Czech study [19], which showed no PML cases during NAT treatment, also in Poland, since 2013, no patient has been diagnosed with PML. There is a need to the development of new biomarkers that may better predict an individual’s risk of PML and further improve upon NAT safety.

NAT therapy in JCVAb-seronegative RRMS patients is safe and results in the absence of PML cases. In Poland, the JCVAb seroconversion rate is similar to that observed in other European countries.

The authors would like to thank the neurologists from all MS centers in Poland who participated in data collection.

The study was approved by the Regional Medical Ethics Committee at the Medical University of Bialystok (R-I-002/141/2018), and written consent to use the anonymized data was obtained from the President of the NFZ.

The authors declare that they have no conflict of interest.

The authors received no specific funding for this work.

Conceptualization: Katarzyna Kapica-Topczewska, Joanna Tarasiuk, Agata Czarnowska, and Alina Kułakowska. Data curation: Katarzyna Kapica-Topczewska, Joanna Tarasiuk, Francois Collin, Agata Czarnowska, Monika Chorąży, Anna Mirończuk, Jan Kochanowicz, and Alina Kułakowska. Formal analysis: Katarzyna Kapica-Topczewska, Joanna Tarasiuk, Francois Collin, Agata Czarnowska, and Alina Kułakowska Investigation: Katarzyna Kapica-Topczew­ska, Joanna Tarasiuk, Monika Chorąży, Anna Mirończuk, Agata Czarnowska, Jan Kochanowicz, and Alina Kułakowska. Methodology: Katarzyna Kapica-Topczewska, Joanna Tarasiuk, Francois Collin, Agata Czarnowska, and Alina Kułakowska. Resources: Katarzyna Kapica-Topczewska, Joanna Tarasiuk, Agata Czarnowska, Monika Chorąży, Anna Mirończuk, Jan Kochanowicz, and Alina Kułakowska. Software: Francois Collin. Supervision: Katarzyna Kapica-Topczewska, Jan Kochanowicz, and Alina Kułakowska.