Epidemiology of multiple sclerosis in Central Europe, update from Hungary

Abstract Objectives Not so long ago, a novel phenotypic classification of multiple sclerosis (MS) and revisions to the McDonald diagnostic criteria were published. Good quality, standardized, and therefore comparable epidemiological data from the Central European region altogether are scarce, and data based on the aforementioned criteria are nonexistent; thus, an update is needed. Materials and Methods Patients residing in Csongrád county with a definitive diagnosis of MS according to the 2017 McDonald criteria were included and evaluated by the 2014 revised phenotypic classification. Results A total of 420 patients were included, of whom 313 were females (female/male ratio 2.925:1). Standardized prevalence was 101.8/100,000, and incidence was 4.44/100,000. Relapsing–remitting disease type was identified in 288 (68.57%) cases, of which 230 patients (79.86%) were treated and of which 202 patients (87.8%) showed no disease activity with their current treatment. Progressive disease type was seen in 132 (31.43%) cases, with 72 patients (54.54%) receiving treatment. More than half of the treated patients (178, 57%) were administered platform therapies, while 134 (43%) received highly active disease modifying therapies. Conclusion The prevalence of MS in Hungary similarly to other countries shows a constant increase in the past decades. The majority of our patients received treatment and had a stable disease while being treated. The distribution of disease courses, phenotypes, and treatment status fell in line with data in the literature based on MS registries with a large number of participants. Ours is the first study to give epidemiological data based on the most recent McDonald criteria and phenotypic classification from the Central European region.


| INTRODUC TI ON
Multiple sclerosis (MS), although considered as a rare disease, is the most common chronic, autoimmune, demyelinating, and neurodegenerative disease of the central nervous system and is the second most common cause (after traumatic injury) of permanent disability among young adults (Alastair Compston et al., 2005). In recent years, the therapeutic palette has expanded considerably; nowadays, there is at least one drug available for every disease course. Adjacent to the introduction of new disease-modifying therapies, a novel standpoint is gaining ground regarding the therapeutic strategy of MS.
In 2017, the most recent revisions to the McDonald diagnostic criteria were published (Thompson et al., 2018). According to these, the diagnosis of clinically definitive multiple sclerosis can now be set even in patients who previously were classified as clinically isolated syndrome (CIS) patients, based on solely one clinical event supported by magnetic resonance imaging (MRI) and/or cerebrospinal fluid (CSF) examination. Furthermore evidence suggests that the future disease activity can be predicted from the baseline MRI parameters (Davda, Tallantyre, & Robertson, 2019).
In addition to the newly introduced diagnostic criteria, the very first comprehensive therapeutic guideline was also published in 2018 (Montalban et al., 2018a(Montalban et al., , 2018b, which breaks with the previous escalative therapeutic strategy of starting with less potent platform therapies (interferons, glatiramer acetate, dimethyl fumarate, teriflunomide) which, if turn out to be inefficient, can be switched onto more potent disease modifying treatments (DMT). On the contrary, the new guideline recommends the treatment of every patient as soon as possible, with disease modifying treatments matching the patient's disease activity, advocates the use of highly active DMTs (HAMDT-fingolimod, natalizumab, ocrelizumab, cladribine, alemtuzumab, and mitoxantrone) as a first choice for patients with highly active disease. Furthermore, a new phenotype classification was also established (Lublin, 2014). It maintained much of the basic attributes of the originally defined disease courses; however, in contrast to the previous classification system, it lays a much greater emphasis on the activity of the disease and the gradual worsening of symptoms in order to categorize the disease into different phenotypes.
It recognizes two major courses of the disease, one being the relapsing-remitting (R-R and CIS) type, the other being the progressive type (Lublin, 2014). Depending on the clinical and imaging presentation, a patient's disease diagnosed with CIS can be stable and show no activity, or can be clinically and/or radiologically active, in which case the disease is considered relapsing-remitting. Similarly, an R-R disease can be stable or can show clinical and/or radiological activity. Likewise, patients with progressive diseases can be split into two groups whether activity (new, or unequivocally enlarging T2 lesions and/or contrast enhancement can be seen on MRI scans or a clinical relapse) is present (i.e., progressive disease with or without activity).
These two groups can further be divided into categories based on the presence or the lack of continuous worsening of symptoms (i.e., progressive disease type with or without progression; Lublin, 2014).
Previously, when disease subtypes were classified based on clinical courses and on Expanded Disability Status Scale (EDSS) scores (Kurtzke, 1983), no clear cut margin was defined where the relapsing-remitting phase ended and secondary progressive (SP) phase (most commonly still with relapses) began. The EDSS score of R-R patients traditionally ranged from 0 to 5.5, whereas it laid between 3.5 and 10.0 for SP patients. From previous large scale studies exploring the natural history of the disease (Scalfari et al., 2010), and epidemiological studies, it is known that no matter how long it took a patient from disease onset to reach the turning point between relapsing and progressive state; the disease advances roughly with the same speed in all patients thereafter (Leray et al., 2010;Scalfari et al., 2010). Irreversible axonal injury and transition into a progressive disease course occur approximately at an EDSS score of 4 (Lorscheider et al., 2016) and therapeutic window to stabilize a patient's illness and prevent transition is before this (Correale, Gaitan, Ysrraelit, & Fiol, 2017); with drugs currently at our disposal, we can significantly prolong the time between disease onset and irreversible axonal injury. The new phenotypic classification gives pivot to clinicians to assess the disease activity of their patients and to choose the most suitable DMT for them, also when to change, recognize if a given treatment has become inefficient. Recently published data shows that the biggest socioeconomical and financial burden on a society and healthcare giver is put on by the secondary progressive MS population, almost double that of the relapsing-remitting population (Purmonen, Hakkarainen, Tervomaa, & Ruutiainen, 2020). The majority, roughly 2/3rd of the total expense, fell into the categories of direct nonmedical costs and productivity loss, driven mainly by the early retirement of patients with secondary progressive disease (Purmonen et al., 2020). Up to date, comparable epidemiological data are needed according to the novel phenotypic classification system in order to assess the therapeutic, medical, and financial needs of patients with MS on a population basis (Kingwell et al., 2013), since many of the patients previously categorized with an R-R disease now fall into the secondary progressive disease category according to the new classification system and therefore require a fundamentally different therapeutic approach, than before.
The aim of our study was to provide a picture of disease development (comparable age-, and sex-specific crude, and standardized prevalence and incidence of MS) from a previously surveyed demographic region and also to provide information about the disability state (measured by the EDSS score) as well as treatment and disease activity status of our patients almost a decade after the introduction and eventual widespread use of HADMTs based on the novel phenotypic classification system proposed by Lublin et al (Lublin, 2014).

| Study area
All of our patients were residing in Csongrád County, and point prevalence was determined on the prevalence day of 1 January 2019.
Incidence data reflect the incidence of diagnosed patients (not disease onset) in the given time period. Csongrád County is located in the southeastern region of Hungary in the temperate zone, with an area of 4,262.68 km 2 and humid continental climate.
According to the latest census performed by the Hungarian Central Statistical Office on 1 January 2019; 399,012 (accounting for 4.08% of the total population) people lived in Csongrád County, of which 189,420 were males and 209,592 females (www.ksh.hu).

| Data collection
All of the enrolled patient's data (demographic, radiologic, and clini-

| Diagnosis and follow-up examinations
The diagnosis was established by the given patient's neurologist.
Until 2001, the criteria of Poser were applied (Poser et al., 1983), and in subsequent years, always the latest available criteria of McDonald were used. All patients with diagnoses of MS according to Poser (Poser et al., 1983), and previous McDonald criteria ( McDonald et al., 2001;Polman et al., 2005Polman et al., , 2011, were reviewed according to the latest McDonald criteria (Thompson et al., 2018).
At the time of diagnosis, in addition to physical examination, an MRI scan of the brain and, if necessary, the spinal cord was conducted (using a 1.5 T MRI scanner for both the brain and spinal cord before 2017, and only for the spinal cord after 2017, when a 3 T MRI scanner was implemented for brain imaging, always in adherence to the latest MAGNIMS guidelines; Filippi et al., 2016). A lumbar puncture was made to acquire CSF to be analyzed in our clinic's accredited laboratory (according to ISO 9002 standards) by the means of laser nephelometry for the quantitative determination of proteins, isoelectric focusing, and IgG immunoblotting for the detection of oligoclonal bands (OGP). Also, in necessary, visual-evoked potential, somatosensory-evoked potential, and brainstem-evoked response audiometry tests were conducted.
After the diagnosis was established, all CIS, R-R patients (according to the classical phenotype classification), and all other patients still receiving any kind of disease modifying therapy were routinely examined every 3 months. Every other patient (with progressive disease types, and patients without a DMT) was re-evaluated at least annually. In the event of a relapse, an unscheduled, out-of-turn appointment was always provided for the patients. All patients diagnosed with CIS are advised to attend a routine check-up annually indefinitely, unless another disease responsible for their symptoms is diagnosed or disease activity presents, from which point the regular visit schedule is recommended (i.e., at least every 3 months). During every control check-up, the patient's current neurological status, EDSS score, and the clinical form of the disease were laid down and revised in the patient records. After the baseline MRI examination, control MRI scans were conducted on a regular basis; always in adherence with the SMPC of a DMT, a given patient is receiving, but at least yearly. In the event of a relapse before the initiation of intravenous corticosteroids, a contrast enhanced MRI scan was always conducted. A control brain MRI scan was done to all the patients presenting with CIS disease type both 3 and 6 months after the appearance of the first symptoms, and yearly thereafter. Every patient's medical chart participating in our study has been reviewed by four neurology specialists independently and was excluded from it, if at least one specialist felt uncertain about the diagnosis.
According to the aforementioned principles and in line with data from recent large scale studies, all patients who previously fell into the relapsing-remitting disease type were re-evaluated for secondary progressive disease course using the objective, 3-strata criteria proposed by Lorscheider et al. for defining secondary progressive MS (Lorscheider et al., 2016).

| RE SULTS
Our database registered 420 patients with MS on the prevalence day, 107 males and 313 females (the female-male ratio was 2.92:1).
In the 6 years since our last epidemiological data collection from 2013, 108 new MS cases were diagnosed. In that period, 30 patients had died and 28 had moved away from the geographical area. The crude prevalence of MS for the whole cohort was 105.3/100,000, 56.5/100,000 for men and 149.3/100,000 for women (Table 1).
The standardized prevalence was 101.8/100,000. Age-, and sexadjusted, standardized prevalence was 53.9/100,000 for men and 144.8/100,000 for women. The 2013 European standard population was used for the standardization (Table 1).
Disease activity could be established in 31 patients and 23 persons' (17.42%) disease showed progression adjacent to activity (activeprogressive arm, A-P), while progression was not seen despite disease activity in eight patients (6.06%, active-not progressive arm,

10.05
Note: The disease characteristics of our whole cohort as well as stratified by disease phenotype and disease activity. Disease duration, age at diagnosis, and patient age are given in years.
Abbreviations R-R, relapsing-remitting; SD, standard deviation; SP, secondary progressive disease.   (Table 5). Note: Columns one and two show how many patients in each subgroup have ever been treated during some point in their disease, and the potency of the used DMT for patients who are treated on the prevalence day. Columns three and four show how many patients in each subgroup have started their treatment with low or high potency drugs. Most of the still treated R-R patients started their treatment with platform drugs, and escalation to a HADMT was necessary with 48 of them. On the contrary, a higher fraction of patients with a progressive disease started their treatment with a HADMT. Furthermore, a bigger ratio of PP and SP patients switched onto a HADMT after being started on platform drugs, than did patients in the R-R disease type group.

TA B L E 4 Present and past treatment status of our patients by disease type and disease activity
Abbreviations: A-NP, active-not progressive (patient showing disease without progression); A-NT, active-not treated (patients showing disease activity without receiving treatment); A-P, active-progressive (patient showing disease and progression at the same time); A-T, active-treated (patients showing disease activity despite receiving treatment); CIS, clinically isolated syndrome; HADMT, highly active disease modifying treatment-fingolimod, natalizumab, ocrelizumab, cladribine, alemtuzumab, and mitoxantrone; NA-NP, not active-not progressive (patient not showing disease activity nor progression); NA-NT, not active-not treated (patient not showing disease activity without receiving treatment); NA-P, not active-progressive (patient not showing disease activity while showing progression); NA-T, not active-treated (patient not showing disease activity while receiving treatment); Platform, platform therapies-interferons, glatiramer acetate, dimethyl fumarate, and teriflunomide; PP, primary progressive disease; R-R, relapsing-remitting; SP, secondary progressive disease.
According to the phenotypic classification (Lublin, 2014)  their current DMT, also their mean EDSS (1.39) was low compared to their relatively long mean disease duration (11.55 years). Most of the patients with a relapsing-remitting disease course were using platform therapies (62.6%), roughly one third of them were treated with a HADMT (37.4%).
On the prevalence day, 62.1% of the patients with a progressive disease were treated, more people received a HADMT (58.5%), than platform therapies (41.5%). Compared to patients with an R-R disease course, an even bigger ratio of patients (87.4%) started their treatment with platform drugs, as previously no therapy was approved for primary progressive disease, also when most patients, who now have a SP disease began their treatment at a time when only platform drugs were available for the R-R disease type and converted before the introduction of HADMTs. The currently seen high number of patients on a HADMT among patients with a progressive disease is owed to the recent approval of ocrelizumab for the treatment of primary progressive disease in Hungary. Regarding the whole cohort, an escalative therapeutic approach was used in the history of most patients, as at the time when we began to treat most of our patients HADMTs were either not available yet at all or were not reimbursed by the healthcare provider as a first choice of treatment; therefore, only a lateral change or escalation was possible, induction with a highly potent drug was not.
In the near future, an even higher ratio of patients is expected to be using HADMTs (especially among patients with progressive disease), as more patients are going to be diagnosed with a secondary progressive disease type, than nowadays, not only due to the introduction of the new diagnostic criteria for secondary progressive MS, but because several highly active DMTs are already in the pipeline awaiting EMA approval for the treatment of SPMS.
A limitation of our study is the relatively small sample size, however taking into consideration that only our clinic maintains an MS registry in the country and that the examined area comprises of 4% of the country's total population, and all our diagnostic tools and treatment options are available countrywide, our findings can be considered as representative for the whole of the MS population in Hungary. Strong points include the good quality data extracted from our registry, and our current study appears to be one of the first epidemiological studies that have used the most recent McDonald criteria as well as the newly proposed phenotypic classification of disease type, and one that has also evaluated the treatment status of the patients. Other strong point is that because of a previous epidemiological study in this area (Bencsik et al., 2001;Zsiros et al., 2014) it provides a picture of disease development. Thus, it gives valuable information to the healthcare provider not only about the size of the population that needs to be treated, but of the actual status of their disease and efficacy of the used therapy, and possible future therapeutic and financing needs. Furthermore, fresh epidemiological data from the Central European region based on recent diagnostic and classification criteria were lacking, which demand our study addresses.

| CON CLUS ION
The new phenotype-based classification system, new therapeutic guidelines and the most recent revisions to the McDonald diagnostic criteria invoke the need for a fundamentally different therapeutic approach than used before. In contrast to the previous escalative practice, a personalized treatment strategy is urged. To achieve this, the constant re-evaluation of the patient's disease course along with disease activity is needed. With the timely start of an adequate DMT and rapid changes in treatment, when necessary, long-term stability and significant slowing of disease progression can be achieved not only in patients with R-R, but with progressive disease as well.

ACK N OWLED G M ENTS
This paper was supported by the Foundation for the Neurology of

CO N FLI C T O F I NTE R E S T
The authors report no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.