Oligoclonal Immunoglobulin Gamma Bands and Long-Term Disability Progression in Multiple Sclerosis: A Retrospective Cohort Study

Multiple sclerosis (MS) patients without the typical oligoclonal bands (OCB) distribution of immunoglobulin gamma (IgG) in the cerebrospinal uid (CSF) has a different genetic background and brain MRI features than OCB-positive MS patients. It is less clear if OCB presence indicates differences in clinical aspects of MS. We used Swedish MS register data on clinically denite MS patients with known OCB status. Date of birth, MS onset and date at sustained Expanded Disability Status Scale (EDSS) score milestones 3, 4, 6, date at conversion to secondary progressive (SP) MS, sex, and immunomodulatory treatment (IMTs) duration were collected. Multivariate Cox regression models were used to investigate the association between OCB status and risk of reaching each milestone.


Objective
To determine whether CSF OCB status is associated with long-term disability outcomes.

Methods
We used Swedish MS register data on clinically de nite MS patients with known OCB status. Date of birth, MS onset and date at sustained Expanded Disability Status Scale (EDSS) score milestones 3, 4, 6, date at conversion to secondary progressive (SP) MS, sex, and immunomodulatory treatment (IMTs) duration were collected. Multivariate Cox regression models were used to investigate the association between OCB status and risk of reaching each milestone.

Results
The OCB-positive group reached disability milestones at an earlier time and at younger age. OCB positivity signi cantly increased the risk of reaching EDSS score 3.0 (HR = 1.29, 95%CI: 1.12 to 1.48, P < 0.001) and 4.0 (HR = 1.38, 95%CI: 1.17 to 1.63, P < 0.001) but non-signi cantly for EDSS 6. The OCBpositive group had a 20 % higher risk to convert to SPMS.

Conclusion
Patients with or without CSF OCBs share a risk of long term progression and for conversion to SPMS.
Presence of OCBs in the CSF is associated with higher risk to reach EDSS milestones and a higher risk to convert to SPMS. Our ndings suggest higher disease modifying effect of OCB presence in the early in ammatory stages of MS.

Background
Multiple sclerosis (MS) is an immune-mediated demyelinating and neurodegenerative disorder of the central nervous system (CNS).(1) Cerebrospinal uid (CSF) analysis at the time of diagnosis is used for Results investigation of CSF pathology and for differential diagnosis work up. Routine CSF examination includes measuring the oligoclonal bands (OCB) status, IgG index, albumin ratio and cell counts. OCB in CSF is detected in about 95% of MS patients. (2) It has been shown that HLA-DRB1*04:04 is associated with OCB-negative MS and carriership of HLA-DRB1*15:01 confers risk for OCB-positive MS. (3)(4)(5)(6) We have previously reported that OCB-negative patients MS have less brain white matter atrophy and less regional gray matter loss in speci c areas of the brain including basal ganglia, diencephalon, cerebellum and hippocampus.(7) However, ndings have been inconsistent with regard to clinical and demographic differences such as age at MS onset, sex ratio and disease severity between the two groups. (5,(8)(9)(10) Given the paucity of evidence, we aimed at conducting a registry based study to investigate whether OCB carriership confers higher risk of long-term disability worsening in MS.

Study population
We conducted a retrospective registry based study of patients with MS diagnosis according to the McDonald's diagnostic criteria (11)(12)(13), registered in the Swedish MS registry. (14,15) Patients with available data on OCB status at the time of MS diagnosis, date of birth, date of MS onset, sex, and duration of exposure to rst-and second-line immunomodulatory treatment (IMTs) were included. CSF samples that harbors two or more unique CSF bands in comparison to plasma were considered to be OCB positive.

Statistical analyses
Clinical and demographic data at the time of CSF examination were compared between OCB-positive and OCB-negative patients using parametric and non-parametric tests for normally and non-normally distributed data, respectively. The Kaplan-Meier method was used to estimate and compare the median age at reaching EDSS milestones and SP conversion between OCB-positive and OCB-negative groups. Multivariate Cox regression models were used to investigate the in uence of OCB status on the risk of reaching each clinical milestone. Models were controlled for sex, age at the onset of MS, calendar year of CSF examination, disease phenotype and exposure to IMTs (as time varying covariate).
In total, 7 322 patients were included of whom 828 patients (11.3 %) were OCB-negative. Age at MS onset was 1.9 years earlier in OCB-positive group compared with OCB-negative (p < 0.001). In the OCB-negative group, platform IMTs (interferons, glatiramer acetate, dimethyl fumarate, teri unomide) were started on average at a 2.5 years higher age in the OCB-negative group compared with OCB-positive (p < 0.001). Similarly, more potent IMTs (natalizumab, alemtuzumab, cladribine, ngolimod, rituximab, autologous stem cell transplantation) were started on average at a 3.2 years higher age in the OCB-negative group (p < 0.001). Age at CSF sampling was 2.7 years older in the OCB-negative group (p < 0.001). Table 1 summarized characteristics of the study population. Age at and risk of sustained EDSS score milestones The median age at the sustained EDSS score 3.0 was almost four years younger in OCB-positive patients (57.3 vs. 53.1, Fig. 1A). Similarly, the median ages at the sustained EDSS scores 4.0 and 6.0 were almost three (62.5 vs. 60, Fig. 1B) and two (67.8 vs. 66, Fig. 1C) years younger in OCB-positive compared with OCB-negative, respectively. After controlling for potential confounders, OCB-positive patients had a higher risk of reaching sustained EDSS scores 3.0, 4.0 and 6.0 comparted to OCB-negative patients with HRs of  (Fig. 1D).

Discussion
In this study, by using the national Swedish MS registry data, that cover 80% of prevalent MS patients we showed that carriage of CSF OCB is associated with a higher risk to conversion to progressive phase of MS and increase risk of reaching EDSS milestones 3 and 4. We observed clinical and demographic differences between OCB-positive and negative patients in agreement with the previously published genetic studies. Disease modifying feature of OCBs maybe most prominent in early stages of MS were in ammatory burden is higher compared with later stages. It is known that in ammatory events like relapses with unfavorable characteristics and number of gadolinium enhancing lesions early in disease are associated with adverse long term outcomes (19) and many studies reported that relapse rate decline with disease duration. (20) This comprehensive study includes the largest number of OCB negative patients, 828 (11.3%) in total of 7322 MS patients allowing us to calculate a more precise estimation of the prevalence of OCB negative patients in Sweden. Our estimates here is two times the prevalence previously reported by our group previously (5.5%)(3) and to be considered as representative for Swedish MS population and comparable to OCB-negative MS prevalence reported in Scandinavia (12%). (5) A study on OCB status and association to speci c genetic risk alleles in a Scandinavian cohort (5) also reported a higher prevalence of male gender and older age of onset in OCB negative group. patients with pattern I lesion type harbor OCBs in 88% of cases and in patients with pattern II and III lesions OCB was present only in 22%. In the latter group two patients were only transiently OCB-positive. (26) Despite that, OCBs are considered to be a diagnostic mainstay feature in MS. (2,27) The presence of OCBs is a clear indication of the in ammatory process in the CNS (28) which is believed even though sometimes debated to be followed by neurodegeneration (29) that is re ected clinically in disability progression. However their speci city and role in the in ammatory process is still under investigation. (28) In a group of CIS and early MS patients, CSF cell counts were observed to be high in OCB positive patients and correlated with percentage of intrathecal IgG production, in addition the latter correlated strongly with percentage of plasma cells in the CSF.(30) Moreover, the levels of serum neuro lament light chain, a well-established marker for neurodegeneration,(31) is higher in OCB positive patients and correlates with CSF IgG levels,(32) indicating higher extension neuro-axonal damage and loss as well as chronic degeneration in OCB positive patients. This is in line with the increased brain atrophy in OCB positive patients but still only con rms the end but not the means or the how.
OCB could be suggested as a potential clinical covariate in randomized clinical trials (RCT). Our current study endpoints were not designed to answer this question, however. In RCT, short term disability progress outcomes are used, such as "con rmed disability progression at 3 months" and "con rmed disability progression" at 6 months. (33,34) International RCT with high number of included RRMS Page 7/12 patients ( n > 4000) and low EDSS, long time follow up period (3 years) could be characteristics of studies were number of OCB-negative patients would be substantial and time su cient to evident the OCB effect on disability progression.
Limitation of this study include high number of patient without recorded OCB status. Hence, Kaplan-Meyer estimates might somewhat be affected by exclusion of old patients that reached milestones but were lacking OCB data. Another limitation of this study is lack of data with regard to factors known to have disease modifying effects such as comorbidities, vitamin D levels, smoking (35)(36)(37)(38)

Consent for publication. Not applicable
Availability of supporting data. The data that support the ndings of this study are available from Neuroregister.se but restrictions apply to the availability of these data, which were used under license of Neuroregister Research Board and Ethical permission, issued by Stockholm´s Regional Ethical Committee for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of Swedish Neuroregister Research Board, provided that the requesting person holds ethical permission to analyze the data.
Competing interests. SKB and AM report no con icts of interest. JH received honoraria for serving on advisory boards for Biogen, Sano -Genzyme and Novartis and speaker's fees from Biogen, Merck-Serono, Bayer-Schering, Teva and Sano -Genzyme. He has served as P.I. for projects sponsored by, or received unrestricted research support from, Biogen, Merck-Serono, TEVA, Novartis, Sano -Genzyme and Bayer-Schering. VDK has received nancial support from Stockholm County Council (grant ALF 20160457); Biogen (recipient of grant and scholarship, PI for project sponsored by); Novartis (Scienti c Advisory board member, recipient of scholarship and lecture honoraria); Merc (Scienti c Advisory Board member, recipient of lecture honoraria).
Funding. This study was supported by Biogen. Biogen reviewed and provided feedback on the article content to the authors. The authors had full editorial control of article.
Authors' contributions. Conceived and designed the analysis: JH, AM, VDK. Collected the data AM, VDK, JH; Contributed data or analysis tools VDK, JH, AM, SKB. Performed the analysis: AM; Wrote the paper: VDK, AM, SKB, JH. All authors read and approved the nal manuscript.