The good, the bands, and the ugly: do oligoclonal bands mean more brain shrinkage?


Oligoclonal bands (OCBs) are now back in the diagnostic criteria for MS – they can be used to support a diagnosis. If someone has a clinically-isolated syndrome – an episode of symptoms or signs due to demyelination without any evidence of previous events – the presence of OCBs makes it much more likely that they will go on to develop clinically-definite MS

Additionally, the presence of OCBs is associated with a higher chance of disability progression in defnite MS. Over 95% of people with clinically-definite MS have evidence of bands in their CSF. So, there’s quite good evidence that these bands are telling us something meaningful about the disease process and are not just epiphenomena (a ‘by-product’).

OCBs are a strange phenomenon. When you take a sample of any fluid, you can separate out the proteins in it according to their size and charge. This is called electrophoresis. If you use this technique to look at the proteins in cerebrospinal fluid, you can detect a few distinct ‘bands’ of protein, which represent antibodies either being produced inside the CNS, or leaking into the CNS from the blood. These bands are detectable in the majority of pwMS (over 95% of people with definite MS), but can also occur in some infections of the CNS and other inflammatory diseases. In MS, these bands are typically present in CSF but not in blood, indicating that the antibodies are being produced inside the CNS.

This is what they look like:
nicked from the MS trust website

What is so strange and remarkable about these bands is that, although we know they are antibodies, we do not know…
3. Whether they are ‘pathogenic’ (involved in causing / perpetuating the disease)

Another marker of future disease progression that has generated lots of interest is brain atrophy (shrinkage). Atrophy is measured by MRI. It is clinically-important because it is thought to reflect neurodegeneration, which we know occurs from early in the disease and predicts long-term disability progression.

A new study looked at the relationship between OCBs and brain atrophy. PwMS were included if they had a new (<12 month) diagnosis of MS and had not had steroids in the preceding 30 days. 145 people were included, comprising:
– 20 pwMS without OCBs
– 25 pwMS with OCBs
– 15 healthy controls
The OCB-negative group were slightly older, slightly more male and had a slightly lower EDSS at baseline.

They compared brain volumes – as measured by MRI – between the three groups. The brain volumes included white matter volume (the connections between nerve cells), grey matter volume (the cell bodies of nerve cells), and total brain volume. 

Normalised white matter volume was significantly reduced in the group of pwMS who had OCBs, compared to both healthy controls and pwMS without OCBs. This effect was still significant after adjusting for differences in age, EDSS, and disease duration. There was no difference in whole brain volumes or grey matter volumes between the groups.

This is partly consistent with previous similar work. Other groups have reported an association between OCBs and increased grey matter atrophy and whole brain atrophy. It is therefore slightly surprising that these authors don’t find the same thing, and only see an effect in the white matter.

This study is cross-sectional, and so we are seeing a snapshot in time. To show causation it would be useful to see that people with OCBs, when followed over time, develop more atrophy than people without OCBs.

The authors also don’t control for previous or current disease-modifying therapy use. As we know that effective therapies can wipe out OCBs and slow brain atrophy, this is a major issue for me. One which limits interpretation of the findings considerably.

Another reason to be wary of overinterpreting these findings is that ‘brain volume’ on MRI is not the same thing as atrophy. Loss of brain volume can be a combination of loss of neurons, glia (supporting cells), and resolution of inflammation. The phenomenon in which the brain appears to shrink in the initial phase following DMT initiation is called ‘pseudoatrophy’, and is thought to reflect a reduction in inflammatory swelling rather than a genuine loss of brain tissue. As far as I can see the authors don’t correct their volume measures for lesion count. 

So in short this is interesting stuff but I’m not wholly convinced, on the basis of this, that there is such a stark difference in brain atrophy between pwMS with and without OCBs. It will be important to follow these people over time to see whether the difference in white matter volume is robust and sustained. To really test the hypothesis that OCB+ MS and OCB- MS are different diseases, it would be great to see the results of the big phase III trials analysed according to OCB status. 




Oligoclonal bands of IgG (OB) are proposed as an early prognostic factor of the disease. Growing attention is directed towards brain volume evaluation as a possible marker of the severity of MS. Previous studies found that MS patients lacking OB have less brain atrophy.


to evaluate a possible relationship between OB and cerebral volume in a cohort of early MS patients.


Inclusion criteria were: diagnosis of relapsing-remitting MS; CSF analysis and MRI acquired simultaneously and within 12 months from clinical onset. A total of 15 healthy controls underwent MRI.


In 20 MS patients, CSF analysis did not show OB synthesis (OB negative group). A control group of 25 MS patients in whom OB was detected was also randomly recruited (OB positive group). T test showed a significant difference in NWV between the OB positive and OB negative groups (P value = 0.01), and between the OB positive group and the healthy controls (P value = 0.001). No differences were detected between OB negative group and healthy controls.
Multivariable linear regression showed a relationship between NWV and OB synthesis (P value = 0.02) controlling for age, gender, and EDSS.


Our preliminary results suggest that OB positive patients show more atrophy of white matter since early phases of the disease, supporting the role of CSF analysis as a prognostic factor in MS.

About the author


  • "20 pwMS without OCBs"

    Accordig to Professor Gavin Giovannoni those people dont have ms

    " which represent antibodies either being produced inside the CNS, or leaking into the CNS from the blood."

    If you have a closed BBB is this possible?

    Nice post

    • All I have said that is when we pin down the cause of MS we will find that patients who are 'truly OCB-ve' will have another disease.

      Please note that at present you can fulfill the diagnostic criteria for having MS being OCB-ve. Similarly you don't necessarily need a CSF analysis to make the diagnosis of MS.

    • Please note that people with MS who are OCB-ve do better (progress more slowly) than people who are OCB+ve. This includes both relapse-onset and PP MS.

    • Thanks Luis and ProfG – I'd also add that the blood-brain barrier probably is 'leaky' throughout the disease. This is a bit controversial but certainly the old dogma that it is 'closed' in advanced disease has been questioned recently.

    • Well I personally know 02 pwMS that claim to be negative OCBs and in them MS seems to be a little more active, had at least 03 relapses in 02 years.
      They are currently using Natalizumab.
      It may be that CSF electrophoresis for OCB gives a false negative result, does not it?!
      I think even the MD has published about a study that pointed out that in fact if they "searched for right" the OCBs are there.

      Now as far as I know maybe only the Natalizumabe seems to "erase" the OCBs, it remains to be seen if Cladribine, Alemtuzumab and Ocrelizumab will be able to eliminate OCBs as well …

    • I had the same question, what percentage of pwMS are OCB- after all? Is it around 30% or is it around 5% (based on the older post with the study about the right process of electroaphoresis that Cinara mentioned)? Thanks.

  • Do the bands change over time, for example if a person had a LP for diagnosis of MS and then a second LP several years later? Going from OCB-ve to OCB+ve or the other way round?

    • I came across the following paper about neurological disorders, which states:

      " Three of 24 MS patients did not show OCB at the first CSF analysis, but in the second. A loss of OCB was often associated with remissions from diseases, e.g., during effective treatment. In patients with neurological diseases, both initially positive and negative OCB results may change over time, which often parallels the clinical condition. Such variability must be taken into account for the interpretation of OCB results".

      J Neurol. 2014 Mar;261(3):554-60. doi: 10.1007/s00415-013-7234-2. Epub 2014 Jan 22.
      The presence of oligoclonal IgG bands in human CSF during the course of neurological diseases.

      Haertle M1, Kallweit U, Weller M, Linnebank M.

    • Do you, at Barts, monitor the OCB levels annually of your patients on highly effective DMTs and switch them if the increase?

  • Antibodies
    More than fifty years ago, it was observed that MS patients had an increased
    level of antibodies in CSF.44 Later on it became clear that this increased
    production of antibodies were oligoclonal in distribution, i e only a limited
    number of plasma cell clones are contributing to the increased levels of
    antibodies.45 A further development was the development of the IgG-index,
    which is an estimate of intrathecal IgG production,46 and today the
    demonstration of intrathecal IgG production is part of the clinical routine in
    establishing a diagnosis of MS. However, the specificity of these antibodies
    has not been established. Most of the oligoclonal antibodies present in the
    CSF are not directed to the major myelin components,47 and some
    controversy exists as to the importance of those that do exist.36 Additionally,
    intrathecal antibody production can be seen in a variety of conditions.48 At
    present it is unclear whether these antibodies are harmful, protective, neither
    or both. It has been demonstrated that patients with RRMS and SPMS have
    antibodies directed towards oligodendrocyte precursor cell lines, but only the
    SPMS patients had antibodies directed towards a neuronal cell line.49 This
    supports the idea that the concept of epitope spreading is important in MS.

    Burman, J. 2014. Curing Multiple Sclerosis. How to do it and how to prove it. Digital
    Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1005.
    74 pp. Uppsala: Acta Universitatis Upsaliensis. ISBN 978-91-554-8964-9.

  • I've had some thoughts about OCBs and their potential significance, particularly with respect to disability worsening and neurodegeneration linked to microglia. Should probably put my thoughts down in a paper.

  • Do you, at Barts, monitor the OCB levels annually of your patients on highly effective DMTs and switch them if the increase?

  • What are the implications of cis with 1 area of inflammation on the spinal cord, coupled with no brain lesions on MRI but a positive lumbar puncture for Oligoclonal bands only in the csf ?

    Is the MRI or LP the better indicator of risk of progression from CIS to MS ?

  • At this video Dr. Freedman explains in a very simple way the importance of CSF fluid, the potential use of the blood neurolifaments and more

  • Thanks. The video is interesting but doesn’t address my jest ion read Oligoclonal bands v mri as most reliable indicator of likely progression from cis to ms.

    Thanks for taking the time to comment, the video was an infesting watch

By DrBenJ



Recent Posts

Recent Comments