org355: Multiple Sclerosis and EBV

From the New York Times

1. What was the incidence risk of MS? Is this common?

2. What is the exposure? Is it common?

3. What is the study design?

4. Wait a second. Shouldn't we do case-control study?

5. Explain how temporality might be an issue.

6. Do we need to do a randomized trial to establish causation?

Re: Multiple Sclerosis and EBV

by | JOHANN SEBASTIAN ORTIZ - Monday, 17 January 2022, 1:26 PM

I am absolutely thrilled about this, as a paediatric neurologist. However, I think this information should be interepreted with caution.

1) The pooled incidence rate across 75 reporting countries was 2.1 [95% CI: 2.09, 2.12] per 100,000 (Mult Scler. 2020 Dec; 26(14): 1816–1821.) It is not common

2) In England, EBV infects 55% of the population by the age of 12 ( BMC Infect Dis 19, 1007 (2019). https://doi.org/10.1186/s12879-019-4578-y) si inferring from this data, EBV infection is very common

3) Longitudinal cohort?

4) It´s quite difficult to perform a case control study on low prevalent disease. However, if we follow a population from the begginig all of those who are exposed can act either as a case (if they meet criteria to be) or a control who was exposed.

5) Not sure about temporality, but rather on the case definition, it only included men, from a controlled setting (Millitary)

6) Yes, but is highly unlikely due to costs and prevalence of the diesease, it usually starts with mild symptoms that are often underrecognized.

Re: Multiple Sclerosis and EBV

by | FATHIMA MINISHA - Tuesday, 18 January 2022, 12:04 PM

Very interesting read indeed...

1) As per the article, the risk has been mentioned as 0.5%- implying that its not a common disease

2) Exposure is infection with EB virus- very common.. as the virus is associated with infectious mononucleosis which is again quite common and the virus stays in the body for life

3,4) Now these 2 questions are quite tricky...
I think what has happened is they did follow all those ppl- 10 million for over 2 decades in a cohort study manner- but I dont think it was meant specifically for EB virus and MS association...
so I think they went on to do a nested case-control study- they took the patient who developed MS over the follow up period as cases, and controls were those who did no have MS... and looked at if they had EB virus infection before the onset of MS... because for a disease as rare as MS, case control is the way to go... the way the results have been reported "Among the multiple sclerosis patients, 32 out of 33 got infected with Epstein-Barr before they developed M.S"- this is how you would normally report results from a case-control study right?

5) Temporality is a major issue here because they are basing the diagnosis of exposure on antibody testing... and it would be difficult to ascertain if the infection occurred before or after the onset of MS. Unless they were routinely testing these ppl during the follow up period (which is unlikely considering the logistics and costs behind it)...

6) Well, the thing is you can't possibly design an RCT to establish causation here... especially when this virus is popularly known to be linked to other diseases including cancer... So, the answer would be yes we need it but it can't be done.

Fathima

Re: Multiple Sclerosis and EBV

by | LJILJANA LUKIC - Wednesday, 19 January 2022, 10:29 AM

Hi all,

I read this the other day and as well though it was really interesting (ID doctor here :) )

As for your comments, I'd reflect on Fathima's 5) comment:
From what I understood, they went back and tested the blood samples that were kept for years. So if they did test multiple times during the follow-up they could've watched the seroconversion from EBV negative to EBV positive by testing positive IgM antibodies after previous tests showing negative IgM and IgG for EBV.
Do you think this was the case as well?

What is a limitation of the study is that MS is 3 times more common in women than men, but they mostly had men in their study. Considering that we still don't know the exact pathogenesis of the disease, this is quite a relevant issue to address.

Ljiljana

Re: Multiple Sclerosis and EBV

by | FATHIMA MINISHA - Friday, 21 January 2022, 6:08 PM

Hi Lili...
Yup that makes sense.... then temporality could be established right? if there is a clear seroconversion before the diagnosis of MS...
Then again the question remains- the point of diagnosis of MS need not be the time when the pathology sets in... from what I know MS can have a latent silent periods and timing of development of symptoms is different for different people... the disease could have started much before the appearance of symptoms or the diagnosis even... again in such a case temporality would be an issue...

Fathima

Re: Multiple Sclerosis and EBV

by | Sujit Rathod - Wednesday, 19 January 2022, 5:53 PM

Dear Johann, Fathima, Ljiljana - fabulous answers, and I'm excited about the expertise you bring to the topic.

In that case, it would be worth reading over the actual publication in Nature, and thinking about how the NYT journalist made the content access to the wider audience.

It is amazing to read about a study with 10 million people followed up for decades. My own projects tend to have sample sizes in the hundreds, followed up for months. :-)

Re: Multiple Sclerosis and EBV

by | MARK ALAN SWANCUTT - Wednesday, 19 January 2022, 7:20 PM

Hello all,
I will make only a few comments.
As an Infectious Diseases specialist (slowly going into public health) and not a neurologist, people have been trying to link EBV to all sorts of things, including chronic fatigue syndrome. It is difficult to do with a virus that is present (as a life-long infection- herpes virus family) in the majority of adults and in some populations >90%. It may be correct. It is just hard to say.
Temporality is always an issue except that the military personnel were in the military for as long as or more than 10 years. This gives a minimum time frame at least.
It looks like it was a nested case control study from the original article with a control group of 1566 controls.
They had a very large group because they did it in a military population (USA I presume since authors at NIH or NIH centers). A lot of the veterans could go into VA hospitals, which would help with data gathering. It does not look that way from the manuscript in Science. I agree that it wasn't likely just looking for EBV/MS association.
The military explains the huge surplus in men because women were uncommon until recently and only in certain ranks and billets (assignments and places); this is changing or has changed- it does limit the caution a bit due to this noted above.

Re: Multiple Sclerosis and EBV

by | Sujit Rathod - Friday, 21 January 2022, 9:30 AM

Welcome Mark, and many thanks for your input. My question to you then is: do you think this study has advanced our understanding of the cause(s) of MS?

Re: Multiple Sclerosis and EBV

by | MARK ALAN SWANCUTT - Friday, 21 January 2022, 2:07 PM

Yes, I think it is. First,it might be correct. It needs more data as confirmation.

Second, this paper has a hypothesis for those further studies, which is very useful.

This paper alone is not definitive, but it is very interesting.

Re: Multiple Sclerosis and EBV

by | MARK ALAN SWANCUTT - Friday, 21 January 2022, 2:14 PM

Thank you for the welcome. I've read these for awhile.I don't usually reply with responses or opinions. It has always been interesting.

Re: Multiple Sclerosis and EBV

by | JUDITH MARGARET BURCHARDT - Saturday, 22 January 2022, 10:05 PM

I am posting the link to the original article
https://www.science.org/doi/10.1126/science.abj8222

As Fathima thought it is a nested case control study and analysis was limited to the 5% of military recruits who were seronegative for EBV at baseline. Blood samples were taken from the recruits every 2 years to check for HIV, and the leftover serum was stored. The hazard ratio of having seroconverted for EBV was 32x greater in cases than controls. After (but not before) developing EBV higher levels of a biomarker of neurodegeneration were found, giving further weight to the hypothesis that EBV is causing the MS.

Thank you Sujit for drawing our attention to this paper and to everyone else for your helpful contributions

Judith