The Epstein-Barr virus appears to be an important cause of multiple sclerosis

January 14, 2022 • 9:15 am

The Epstein-Barr virus has been associated with a variety of diseases; as Wikipedia notes (my emphasis, and I’ve left in the footnotes so you can consult 11-13, which I’ve put in bold):

The Epstein–Barr virus (EBV), formally called Human gammaherpesvirus 4, is one of the nine known human herpesvirus types in the herpes family, and is one of the most common viruses in humans. EBV is a double-stranded DNA virus.[2]

It is best known as the cause of infectious mononucleosis (“mono” or “glandular fever”). It is also associated with various non-malignant, premalignant, and malignant Epstein–Barr virus-associated lymphoproliferative diseases such as Burkitt lymphoma, hemophagocytic lymphohistiocytosis,[3] and Hodgkin’s lymphoma; non-lymphoid malignancies such as gastric cancer and nasopharyngeal carcinoma; and conditions associated with human immunodeficiency virus such as hairy leukoplakia and central nervous system lymphomas.[4][5] The virus is also associated with the childhood disorders of Alice in Wonderland syndrome[6] and acute cerebellar ataxia[7] and, based on some evidence, higher risks of developing certain autoimmune diseases,[8] especially dermatomyositis, systemic lupus erythematosus, rheumatoid arthritis, Sjögren’s syndrome,[9][10] and multiple sclerosis.[11][12][13][14] About 200,000 cancer cases globally per year are thought to be attributable to EBV.[15][16]

Infection with EBV occurs by the oral transfer of saliva[17] and genital secretions. Most people become infected with EBV and gain adaptive immunity. In the United States, about half of all five-year-old children and about 90% of adults have evidence of previous infection.[18] Infants become susceptible to EBV as soon as maternal antibody protection disappears. Many children become infected with EBV, and these infections usually cause no symptoms or are indistinguishable from the other mild, brief illnesses of childhood. In the United States and other developed countries, many people are not infected with EBV in their childhood years.[19] When infection with EBV occurs during adolescence, it causes infectious mononucleosis 35 to 50% of the time.[20] In 2022, it has been shown that EBV infection increase the risk of developing multiple sclerosis by 32-fold.[21]

EBV infects B cells of the immune system and epithelial cells. Once EBV’s initial lytic infection is brought under control, EBV latency persists in the individual’s B cells for the rest of their life.[17][22]

Here are two EBV virl particles with some proteinaceous spheres (not nuclei!) containing the viruses’ genetic material.  I had mono about twenty years ago, so I’m probably carrying the virus, too.

It has spikes, like Covid-19:


At any rate, note that the association with the virus (henceforth “EBV”) with multiple sclerosis (“MS”) has been suggested before (references 11-13, ref. 14 is this paper). I haven’t read the first three papers, but #14 is just out in Science, and I’ll mention it briefly today. It surely is, given the discussions, the strongest evidence to date for an EBV cause of MS.

The paper was called to my attention by the tweet below from Matthew. And, apparently, this is the strongest suggestion yet that EBV actually causes multiple sclerosis.  If this proves to be the case, and the evidence is pretty strong, then this opens the way to preventing MS, most likely via I suspect shots in the young, because once the diease develops, a shot wouldn’t work. In fact, Moderna is at this moment making an mRNA vaccine against the virus. (Although, coronaviruses like Covid-19 have RNA instead of DNA as their genetic material—EBV has DNA—it doesn’t matter what genetic material the virus uses to replicate, for the mRNA in a vaccine is used by the body to make viral protein that then activates the host’s immune system.)

One note: MS is a disease that appears when your immune system attacks the myelin sheath surrounding the nerves, which disrupts nerve impulses. That in turn can lead to multiple effects, including difficulty in breathing, walking, and seeing. All of us have known people with MS, and you’re probably aware that the disease varies widely in its severity, with of the afflicted dying very quickly and others living a life of nearly normal span. On average, MS takes away five to ten years from your life, and a lot of that life is unpleasant.

Here’s the tweet that alerted Matthew, and then me, to the new results:

The paper below with the nearly dispositive data is free; click screenshot for access or get the pdf here. The reference is at the bottom.

(There’s also a News and Views piece on this article, which you can get for free by clicking the screenshot):

Now the best way to see if the virus causes the disease is to inject virus-free humans with EBV, and see if the injected group gets MS more often than does a control (noninjected) group. But since 90% of adults are infected anyway, and this experiment is highly unethical, one has to find other ways.

These researchers did the next best thing: a retrospective analysis of blood serum left over from AIDS tests on more than 10 million U.S. military personnel.  The criterion for “causality” here is the philosophical one: A causes B if you never get B unless you have A beforehand. (This doesn’t mean, of course that A is the sole cause of B.) As the authors say, “causality implies that some individuals who developed MS after EBV infection would not have developed MS if they had not been affected by EBV.” Note that they say “some individuals”, as there may be other causes of MS. But this is more than an association study, as EBV negative individuals could be tested for infection status during their period of activity duty, and then screened for MS to see whether the disease is associated with earlier infection.

The ten million soldiers were screened over a period of 20 years, and the leftover serum, fortunately, had been stored.  All samples were analyzed for EBV infection and then the MS status of the individuals determined during the period of active duty.

5.3% of individuals whose blood were tested were EBV-negative (as I said, most of us are infected) and in a sample of ten million that’s about half a million people.

Among the personnel examined, 955 MS cases were identified, of which 801 cases had several blood samples available taken at differen times. For each one they looked at three serum samples taken BEFORE onset of the symptoms. Each case was matched with at least one non-MS-afflicted control individual of same age, sex, ethnicity, branch of service, and date of blood collection.

The results were pretty compelling. Under the causation scenario, you’d expect MS to develop almost entirely in the group that were initially EBV negative but then got infected, and only then did they develop MS.

And that’s what they found. To quote the paper:

Only one of the 801 MS cases occurred in an individual who was EBV-negative in the last sample, which was collected at a median of 1 year before MS onset [hazard ratio (HR) for MS comparing EBV-positive versus EBV-negative = 26.5; 95% confidence interval (CI): 3.7 to 191.6; P = 0.001, conditional logistic regression]. At baseline, 35 MS cases and 107 controls were EBV-negative. All but one of these 35 EBV-negative MS cases became infected with EBV during the follow-up, and all seroconverted before the onset of MS (fig. S3). The median time from the first EBV-positive sample to MS onset was 5 years (range: 0 to 10 years), and the median time from estimated EBV seroconversion, defined as the midpoint between the last seronegative sample and the first seropositive sample, to MS onset was 7.5 years (range: 2 to 15 years).

Remember, all of the 801 cases were EBV negative at the first sampling. Then all but one of the individuals who developed MS had gone from EBV negative to EBV positive. (The authors discuss the one outlier case, but you can read that for yourself.) To see if it was really EBV that was associated with the onset of MS, they looked at other viruses as well, and also looked at other disease markers that could show whether MS had already begun (but without physical symptoms) when the patients were still EBV-negative. (They didn’t find that.)

They did other tests as well trying (like good scientists) to try to rule out a causal role of EBV in MS. They ruled out “confounding by unknown factors” because of the strong association between EBV infection and later development of MS. No risk factor could account for the huge increase in MS propensity among those who went from EBV negative to EBV positive.

The other factor was “reverse causation”: perhaps EBV doesn’t cause MS, but the early development of MS, not detected clinically, could make a patient more susceptible to EBV infection. This is ruled out because only the EBV virus was associated with the pathology, while one would expect the “reverse causation” syndrome to make MS patients more susceptible to other viruses.  That wasn’t seen.

I won’t go on except to show this graph, which displays significant differences in the level of antibodies against various human viruses between controls and those who got MS (remember, these are all EBV negative people at the start of the trial.) The blue bars represented antibodies against viral proteins that showed higher levels in controls than in those who got MS, while the orange bars represent the level of antibodies  significantly higher in the blood samples of those those who got MS than the controls. As you see, the level of antibodies against EBV is much, much higher in the pre- and post-MS-onset blood samples than in the control (no MS) samples. In other words, no other virus beside EBV was associated with MS either before or after the symptoms appeared.

The last paragraph of the Science paper suggests MS therapy with monoclonal antibodies against the viral proteins might be better than current therapies, and in fact we’re using monoclonal antibodies now to help patients already infected with Covid-19.

But a better tactic would be not to get the disease in the first place, and the tweet below suggests a vaccine that might do this is in development. And if EBV is associated with all those diseases mentioned above, like cancer and inflammatory bowell disease then a jab when young might stave those off, too!



Bjornevek, K. et al. 2022. Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. Science,10.1126/science.abj8222(2022).

24 thoughts on “The Epstein-Barr virus appears to be an important cause of multiple sclerosis

  1. I just read that and is interesting, but not the entire story. My ex-wife is in final stages of MS and her younger sister died of complications of MS. Thirty years ago, I kept up with the literature on MS focusing on epidemiology: 1. seafoods and olive oil protect, 2, sunlight (vitamin D) protects, 3, diverse gut microflora protects. Since nearly everybody has been infected with Epstein-Barr, the focus for prevention of MS should focus on the above factors I mention.

      1. Yes, about 3 to 1. Also, as you note, there is a familial/genetic role in the disease; and an association with specific HLA-DR genes; and a differential geographical distribution; a differential racial distribution; even an association with heavy metal exposures. Like many diseases, MS is a very complex interaction of the environment and genetic susceptibility.

    1. Yes, these may be complicating factors, but if you CANNOT GET MS without Epstein-Barr, then the overwhelming need is to prevent infection with the virus. Seafood and olive oil do not, as far as I know, prevent one from getting MS. Not having the virus, on the other hand, does.

      And let me add my sympathies for your wife and her sister. That disease sucks.

  2. All i can think of to say is ‘Wow!’. To a non-scientist this looks like one of those deciphering-the-Rosetta-stone breakthroughs. This should be knocking pretty much every other story off the front pages!

  3. Can confirm anecdotally. I had mono at age 9 (unknown due to EBV or some other virus), but also at age 14 (my guess is that was EBV). I never fully recovered from either, but especially after the latter infection, fatigue was especially bad. Other symptoms started at age 18, and finally the MS diagnosis at age 26.

  4. Great to see science at work making our existence more tolerable. I hope they can eventually find a cure as well as a vaccine, and that it leads to treatments for numerous other probably related diseases such as rheumatoid arthritis in all its forms.

  5. The study was possible because the serum samples had been retained. I guess those samples could have been destroyed before the study took place for various reasons e.g. storage cost and patient data-privacy. Hopefully this result will be taken into account by labs holding similar human biological samples and considering how long they should keep them and if/when they should get rid of them. Cost and privacy are both important issues of course but hopefully can be managed in ways that permit the research potential of the material to be properly exploited for societal benefit.

  6. As a virologist, I am beyond thrilled. But I think discovering this elusive link may be mostly a win for the computational biologists who were able to put it all together. And let’s hear it, too, for today’s massive computer power – whew, this is just a glorious win for many areas of science!! And such potential…happy day.

  7. Amazing work. People who store sera always wonder if anything useful will come out of the effort, all those freezers chugging away running up the electricity bill. These “slam-bang” effects settle the question of causality. Now, what to do about it?

    Only 5.3% of U.S. recruits reaching military age had evaded EBV infection to that time and all cases (but one) of MS occurred in that group, after they eventually did become infected (for those who had enough sera to confirm timing.). This implies that infection in adulthood is much more dangerous than infection in early childhood as only one case of MS occurred in that latter, much larger group. (The articles are both paywalled for me,). So if you had a vaccine that prevented EB virus infection, would you give the vaccine to all infants/children or only to those who had got through adolescence without infection? This latter policy would require a blood test in all vaccine candidates, with exclusion of those testing positive. (Not all “mono” is EBV. Cytomegalovirus produces an identical illness. And much infection produces no symptoms.)

    Generally speaking, vaccine policy frowns on testing before vaccination because teenagers rebel at blood tests and it requires two visits to get the vaccine into those who will benefit. But if the vaccine is very expensive (like early hepatitis B vaccine) or is dangerous in people already seropositive (like dengue), this may be necessary. On the other hand, to prevent MS throughout a long life, vaccine immunity would have to remain vigorous for many years. It might be better to delay vaccination until just before the peak years of onset of both adult EBV and, later, MS. All conjectural but relevant to policy.

    Finally, MS is more common in light-skinned women (particularly redheads) who live in high latitudes, an opposite demographic to American service people. Preventing EBV infection in this group would have even greater impact on population health. I can see places like Canada, Scandinavia, and the British Isles adopting universal childhood vaccination (assuming science makes it possible.). Exciting.

    Edit: there may be good reasons related to certain cancers in children that having them immunized against EBV would be an additional benefit.

  8. This is indeed a thrilling breakthrough.

    Add: diet. There is some science finding the possibility of relief for sufferers with a low carb diet. Evidence of remyelination.

    I don’t know of anything about MS prevention and low carb.

    Anecdotal: I have a relative inflicted with MS, and she confirms that her experience with low-carb gives her significant abatement. The closer she sticks to below 30 Gr. carb/day, or even less, the better she feels. She finds it hard to stick to Keto/low-carb because of life habit of sugar-like foods. She is BMI 40.

  9. If we’re the last generation to have this disease that would be a wonderful thing, but still, a cure for those of us who already have it would be nice, too.

  10. It may also be complicated by the age which the person becomes infected. Most people are infected very early in life with little symptoms so the immune response is not extreme. Those who seroconvert later as adults also have a mild response. It is only the young adults who tend to develop glandular fever. Lots of this is misdiagnosed and unless you were tested for EBV IgM then it not really proof positive based on symptoms alone.
    What is really interesting is the level of the immune response. If they could get data on those who has mono and ms the puzzle would become more complicated

    EBV also causes many diseases you mentioned but often in association with other diseases like npc is pretty well linked with plasmodium vivax (the milder malaria). It is a fascinating virus

  11. This is fascinating information and I will be discussing this paper w/ my neurologist during my next visit. I was diagnosed with MS over 20 years ago, when I was 42. My symptoms are mostly cognitive, however, I do have a week right side and a slight limp.

    I’m certain I had mononucleosis when I was in high school. I remember that because my Mother was upset that I had the “kissing disease”. Boy, was I in trouble. LOL

    Also, I am being treated with Ocrevus (an infusion every 6 months that costs $67,000 a pop). I just looked the drug up and it turns out IT IS a monoclonal antibody treatment.

    Now, if they could just come up with a treatment to undo some of the damage to my nervous system. Unfortunately, my neurologist doesn’t think there is “anything in the pipes” during my lifetime that will help me w/ my symptoms.

    Such is life.

    1. Hope you don’t mind my commenting on your health situation. I’m not asking you for any further information and I’m not giving any advice of course. I will apologize to you if you say I shouldn’t have commented.

      If you had Epstein-Barr virus infection during adolescence as the cause of your mono-sounding illness, then you would have been among the seropositive recruits had you volunteered subsequently for the U.S. military. Only one in that entire cohort of 9,500,000 got MS. So one of:

      1) rare events happen, rarely;

      2) your high school illness was not due to EBV infection but some other virus, perhaps not even acquired from kissing. (The “Monospot” test which identifies IgM antibodies against EBV is not always done since it doesn’t change treatment.) If you are in fact seropositive for EBV now, (as nearly everyone our age is) you could have been infected later in life as the initially seronegative recruits were;

      3) there are other causes of MS that might operate more strongly in an unselected civilian and 50:50 female population that were not identified in the U.S. military population. Prospective recruits are not representative of the general population and become even more filtered after health and fitness screening, before they undergo blood testing upon enlistment. I read recently that only 3% of the current U.S. population of military age would meet recruitment standards.

      Apparent exceptions like your situation are very important in testing a rule and I thank you for telling us about it.
      Wish you well.

  12. I’ve never had mono but I was treated for Hodgkin’s(nodular sclerosing) about twenty years ago. I was diagnosed a couple of years after I got out of Peace Corps, and my doctor told me the disease had a viral cause and I likely picked it up in Africa. When I was there I would get sick for a couple of days every other month it seemed. Just a couple days of fatigue, and sometimes a full body rash. I stopped going to see the doctor when this happened because I’d usually be symptom free by the time I got to the capital. So. I guess one or more of those viruses I picked up was EB?

  13. From the text that you quoted above, it appears to me that the 801 number comes from the number of MS patients who had more than one sample available for testing.

    Only 35 of the MS cases were EBV negative on their first draw.

    At baseline, 35 MS cases and 107 controls were EBV-negative.

    Granted, I didn’t read this paper at work today, and I’ll have to log onto my university’s VPN to access the paper.

    I’m curious how many of the ~5% of people who were EBV negative at first draw also became positive during the course of the draws.

  14. The mRNA vaccines may be one of the positives after the pandemic becomes endemic (where other vaccines may be the better long term protection, of course).

    I got the news through an Ars Technica article in my feed, and there a commenter mentioned the Moderna trial – I found it while checking up on references.

    Since I have had two lasting and extremely painful infection consequences – shingles and AMSAN – I’ll add emphasis to the bleeding obvious advice: If there is a vaccine against it, and it is recommended as useful enough, get it.

  15. Are the EBV associated diseases exclusionary – that is, if one takes hold (MS), does that mean another (Burkitt’s Lymphoma) won’t?

  16. I have had MS for 33 years. I don’t remember getting mono. This is good news for future generations and I hope no one else will have to deal with it.

Leave a Reply