An infectious cause of multiple sclerosis?

January 2002

Multiple sclerosis (MS) is by far the most common demyelinating disease in the United States, a disease known well to infectious disease physicians as a result of personal experiences, or as a result of caring for the often devastating infections these patients acquire during the more advanced stages of their disease. Disease onset is usually early, between 15 and 45 years of age, with a characteristic relapsing — remitting clinical course. Chronic progressive disease is also common, but less frequent.

Several features of the epidemiology of MS suggest that it may be caused or precipitated by an infectious agent, probably a virus. As reviewed by Donald Gilden, MD, (Viruses and multiple sclerosis; JAMA, 2001;286:3127), there are three lines of evidence that support this possibility. The first is the now-famous Faroe Islands studies, in which a careful analysis of all cases of MS that occurred there from 1920-1977 strongly suggested that a point source epidemic occurred there, probably introduced by British troops during World War II (Kurtzke JF, Hyllested K. Ann Neurol. 1979;5:6-21). The disease in the Faroe Islands appeared to be most consistent with a transmissible and probably infectious disease.

The second line of evidence comes from studies of identical twins: if one twin develops MS, there is only a 30% chance that the second twin will develop the disease, thus suggesting that something more than a susceptible genotype is required for the disease to develop. Finally, there is substantially more IgG in the cerebrospinal fluid (CSF) of MS patients than in normal CSF, and there are oligoclonal bands as well. Such bands are found almost solely in patients with infections of the central nervous system, and, where carefully studied, have been determined to represent antibody directed at the specific agent causing the infection.

Coronavirus OC43	Herpes simplex virus
Parainfluenza virus	Respiratory syncytial virus
Some of the many viruses that at one time or another were posited as the cause of MS.

The most recent contribution in attempting to identify an infectious cause of MS appeared in The Journal of the American Medical Association (JAMA) on Dec. 26, 2001. Using banked sera from two large ongoing population cohorts, a group of investigators from Harvard examined the association between definite or probable MS and a number of Epstein-Barr viral (EBV) antibodies. The two population cohorts were the Nurses Health Study (women age 30 to 55 years in 1976) and Nurses Health Study II (women age 25 to 42 in 1989); these two cohorts contained a substantial population of 62,349 women; there were 144 women with definite or probable MS, matched with 288 healthy age-matched controls. Included among the cases were sera from 18 cases of MS with sera drawn prior to the onset of disease.

The results, in a nutshell, were as follows: the 18 cases in whom sera were collected before disease onset had higher serum geometric mean titers (GMTs) of EBV antibodies, but not cytomegalovirus (CMV) antibodies, than did age-matched controls. The strongest significant association was found for Epstein-Barr nuclear antigen 2 (EBNA-2), but significant associations were also found for EBNA-1 and early antigen D (EA-D). For the EBNA-2 antibodies, a fourfold difference in GMTs carried a relative risk of 3.9, with a 95% confidence interval from 1.1 to 13.7. Similar, but generally less dramatic and pronounced, results were found in the 126 cases in which the onset of MS occurred after sera had been collected.

The authors made no inferences about causality, but rather conservatively suggested simply that EBV infection seems to have a role in the etiology of MS, and may act by increasing the risk of the disease.

The search for an infectious agent in multiple sclerosis has been going on for decades, and this is not the first time that Epstein-Barr virus has been incriminated in the etiology of MS. Richard T. Johnson, MD, the neurovirologist at Johns Hopkins University, notes in his excellent textbook, Viral Infections of the Nervous System (Second Edition, Lippincott-Raven Press, Philadelphia, 1985) that an infectious cause was first postulated in 1884. Agents implicated in the past, either as a result of detection of higher antibody levels in CSF or serum of MS patients than in controls, or virus isolation, or both, include: measles, parainfluenza viruses 1-3, influenza A & B, varicella, herpes simplex virus (HSV), EBV, respiratory syncytial virus (RSV), coronaviruses, adenoviruses, human T-cell lymphotrophic virus type 1 (HTLV-I), and human herpesvirus 6. Even that list is incomplete. Each of these agents has had a moment or more in the sun, until eventually it was recognized that key findings could not be confirmed in other laboratories, that there were other explanations for the findings, or that there were other sometimes fatal flaws in the data or the reasoning.

One of the most recent agents implicated in MS was Chlamydia pneumoniae, an agent now incriminated in a broad spectrum of chronic diseases, the best-known of which is coronary artery disease. In 1998, a group of investigators from Vanderbilt University published some quite dramatic findings (Sriram S, Stratton CW, Song-yi Y, et al. Chlamydia pneumoniae Infection of the Central Nervous System in Multiple Sclerosis. Ann Neurol. 1999;46:6-14) including recovery of the organism from the CSF in 64% of 37 patients with definite MS vs. 11% of controls with other neurologic diseases. In addition, polymerase chain reaction assays demonstrated the presence of C. pneumoniae genetic material in the CSF of 97% of MS patients, but in only 18% of controls. Finally, 86% of MS patients had CSF antibodies directed against certain C. pneumoniae antigens that were strikingly higher than those seen in controls with other neurologic diseases.

Quite impressive, one might think, is that these findings seemed to have a great deal of promise. True, but even this has evidently all come to naught. Only a few years later, Dr. Gilden noted in his recent editorial accompanying the report of the Harvard investigators on EBV serologic studies (JAMA. 2001; 286:3127) that multiple studies in various laboratories around the world indicated a lack of any significant association between C. pneumoniae and MS.

Little wonder then that seasoned neurovirologists like Don Gilden are more than a little skeptical about the significance of the findings reported by the Harvard investigators. Of particular concern is the absence of any CSF data, since the IgG in the brain tissue and CSF of MS patients is synthesized locally, reflecting the immune response at the site of infection or disease; serum antibody levels may be only poorly reflective, if at all, of what is going on within the central nervous system. Not to appear too discouraging, however, he argues that the search for a viral cause of MS must go on, even though the cause of MS is not “likely to be found under the EBV lamppost.”