More MS news articles for January 2001

Not So Fast: Research on Infectious Links to MS Questioned

http://jama.ama-assn.org/issues/v285n3/ffull/jmn0117-1.html

Brian Vastag

Washington: Amid scattered reports of physicians prescribing antiviral and antibiotic medicines for patients with multiple sclerosis (MS), both unproven against the disease, research into infectious triggers for MS is generating controversy over two pathogenic suspects, one a virus, the other a bacterium.

In 1998, separate teams of researchers announced that they had found higher-than-expected amounts of human herpesvirus 6 (HHV-6) and Chlamydia pneumoniae in the central nervous systems of MS patients, findings that have been roundly debated. And while the two organisms appear in different chapters of the microbiological textbook, their paths through the scientific wringer have been similar.

The advocates present compelling cases and confirmatory data; the skeptics find flaws and fail to replicate findings. Like a boxing match, each round of journal reports and letters tilts the balance, only to be knocked back by the next month's offerings. The instigating scientists, a team at Vanderbilt University seeking C. pneumoniae and a group in Wisconsin chasing HHV-6, stand behind their discoveries, and both are hard on the trail of buttressing evidence from clinical trials. While the dizzying debate leaves observers puzzled, it all sounds very familiar to veteran MS researchers, who have sought an infectious origin of the disease for more than a century. Rabies, rubella, measles, influenza, and "almost the full gamut of viruses associated with neurological diseases" have been implicated in MS, writes Richard T. Johnson, MD, the Johns Hopkins University researcher who helped found the field of neurovirology. Dismissed like stooges in a police station lineup, no suspect remains standing after the withering scrutiny of science. And while trendier pathogens have supplanted the old-time bad guys, the story may well repeat itself.

A Spinal Tap and Questions

Vanderbilt University neurologist Subramianiam Sriram, MD, is eager to talk about his research. Three years ago he published the case study of a 24-year-old man newly diagnosed with MS (Neurology. 1998;50:571-573). After 6 months of worsening symptoms, a spinal fluid sample offered something unexpected: positive cultures for C. pneumoniae. Polymerase chain reaction (PCR) tests confirmed the presence of the bacteria's DNA in cerebrospinal fluid and blood. A cause of respiratory infections, C. pneumoniae had received extra attention for newly reported possible links to heart disease. But until Sriram's finding, no one was seriously pursuing the pathogen in MS.

Following up with a series of patients, a 1999 article from Sriram and colleagues (Ann Neurol. 1999;46:6-14) reported positive cerebrospinal fluid cultures in 64% of patients with MS but in only 11% of control patients. Genetic profiling via PCR picked up even more C. pneumoniae, in 97% of patients with MS vs 18% in controls.

An editorial accompanying the report, from Donald Gilden, MD, at the University of Colorado Health Sciences Center, Denver, said that "if Sriram and colleagues are correct, other laboratories with access to MS material should have no trouble amplifying C. pneumoniae DNA" from brain tissue and spinal fluid.

But there's been nothing but trouble. An onslaught of work from investigators studying Chlamydia has failed to reproduce the Vanderbilt team's results, engendering a debate over methods and qualifications.

Chief among the critics, Margaret Hammerschlag, PhD, a microbiologist at the State University of New York at Brooklyn who has worked with Chlamydia organisms for 25 years, says that the Vanderbilt work "has serious problems." In a recent report, her laboratory and that of Jens Boman, PhD, at Umeå University, Sweden, did not find any C. pneumoniae in blinded samples of brain tissue from 12 MS patients and seven controls.

Sriram responded by saying, "The fact that they can't find it in the brain is not surprising at all" because of difficulties his group had finding C. pneumoniae after injecting it into the brains of mice. "We know we put the bug there," he said. "But we're having difficulty finding out exactly what the conditions are to extract it from brain tissue. I don't know why."

To help settle the dispute, Sriram agreed to participate in a blinded study with three other teams. All received spinal fluid from patients with MS and controls, sent by Michael Kaufman, MD, of the MS Center of the Carolinas Medical Center in Charlotte, NC. The results, presented at the 2000 meeting of the American Neurology Association, further isolate the Vanderbilt team, who found C. pneumoniae DNA in 22 (73%) of 30 MS cases and in 5 (23%) of 22 controls. The other teams, at Johns Hopkins, the Centers for Disease Control and Prevention, and Umeå University, detected no traces of the organism.

Hammerschlag said the mass of evidence points to contamination or a lack of experience with delicate screening tests on the part of Sriram and his colleagues. "All of his work is being published in neurology journals, and they just kind of accept the methods. None of it would get published in microbiology journals," she said.

No Bias

But Sriram countered that if contamination were the issue, C. pneumoniae would appear in a higher proportion of his laboratory's control samples. He points to the multilaboratory study as vindication of his group's ability to amplify small amounts of DNA from C. pneumoniae. "All of us were blinded, there was no bias," he said, adding that his "simplistic explanation" for the discrepancy is "technical differences," specifically in sample preparation.

Kaufman said that the Vanderbilt team does use a different PCR method than the other labs. In fact, no standard test for C. pneumoniae exists, a situation Hammerschlag calls "infuriating." But she hopes that a forthcoming standard, developed by a panel of Chlamydia experts and set for publication later this year, will provide the right tools to settle the matter.

In the meantime, heavyweights in MS research are lining up against C. pneumoniae. Hopkins' Johnson, editor of the Annals of Neurology, called the association between the bacterium and MS "weak." Stephen Reingold, PhD, vice president for research at the National Multiple Sclerosis Society (NMSS), said, "The bulk of the papers are against C. pneumoniae right now. And Sriram is kind of sitting there with a fortress mentality, which is not helpful." Reingold added that the NMSS, in the interest of finding an answer, funds research at Vanderbilt as well as at facilities that have not found C. pneumoniae. "This will all work itself out in the scientific process," he said.

The scientific process has already worked out a great deal about MS. Because the disease is directly caused by the immune system attacking the nerves and brain tissue, a ream of circumstantial evidence points to an infectious connection. A famous incident in the Faroe Islands, where the rate of MS skyrocketed after British troops disembarked during World War II, is one of some 300 epidemiological and family linkage studies that suggest an infectious agent at work. Genetic predisposition plays a role, as confirmed by twin studies, and geneticists speculate that a few dozen genes are involved. But they note that environmental factors must weigh just as heavily.

Over his long career, Johnson has seen enough to say, "There's an environmental exposure that occurs early in life, but it's unknown." Several studies show that miscellaneous viral respiratory infections can exacerbate symptoms, and the immune systems of people with MS tend to overreact to all infections. But exactly how a virus or bacterium may trigger the disease is a mystery.

Molecular mimicry, one theory gaining momentum, holds that the immune system develops its appetite for the nervous system after fighting off an infection. Under this hypothesis, the part of the infectious agent recognized by immune cells is nearly the same shape as key proteins coating cells of the nervous system. Immune cells targeted to that shape proliferate to fight the infection. After the organism is defeated, perhaps even years later, the immune system attacks the nerve cells in a case of mistaken identity.

At a recent meeting of the NMSS in Washington, DC, Roland Martin, PhD, a researcher at the National Institute of Neurologic Disorders and Stroke (NINDS) and a leading proponent of molecular mimicry, said, "There is strong evidence that viruses can trigger the disease." He added, "But I'm convinced that in different patients it could be different viruses; there is no single virus responsible."

The Latest Virus

If true, Martin's assertion would not bode well for the work of Donald Carrigan, PhD, and Konstance Knox, PhD, who 4 years ago created the Institute for Viral Pathogenesis (IVP) in Milwaukee. Carrigan and Knox are working to prove that HHV-6 is not only associated with MS, but causes it. "I think that HHV-6 is necessary, but not sufficient, for multiple sclerosis," said Carrigan. According to this thinking, eliminating the virus would halt the disease.

The model Carrigan and Knox prefer holds that MS begins as an acute viral encephalitis caused directly by HHV-6. After a time, perhaps years, the body's immune system learns, through molecular mimicry, to attack the myelin sheath that coats nerves and other parts of the nervous system. Carrigan said that the hunt for a mechanism is turning up proteins from HHV-6 that resemble proteins on the myelin sheath.

Much of their work focuses on detecting HHV-6 in the blood and brains of patients with MS. Most recently, they found signs of "active infection" in the blood of 22 (54%) of 41 MS patients but none in 61 controls. They also reported finding HHV-6 in the nervous systems of 8 (73%) of 11 MS patients and in 2 (7%) of 28 controls (Clin Infect Dis. 2000;31:894-903). Other groups, most notably that of Hugh McFarland, PhD, at NINDS, have published work that at least partially supports IVP's findings (J Infect Dis. 2000;185:1321-1325).

But Knox and Carrigan have been the loudest proponents. They created a media buzz at the 1998 American Neurological Association meeting after presenting a study that found HHV-6 in the brain lesions of MS patients, and they continue to be regularly quoted in the lay press.

This outspokenness has some researchers on edge. "They are overinterpreting their data," said Reingold. "This is all association. It's interesting to know, but cause and effect is what matters."

Establishing cause and effect will be difficult. For one, HHV-6 is ubiquitous; that is, almost everyone is exposed to it as a child. (The virus sometimes causes roseola infantum, a condition with rashes and high fevers.) The virus then turns off, becoming latent, leaving behind clues like antibodies and DNA. So finding HHV-6, by either blood studies or PCR, does not mean a person harbors an active infection. On top of that, people with MS tend to overreact to all pathogens, rendering the finding of higher-than-normal antibody levels in the blood dubious.

To get around the problem, Carrigan says the test he and Knox developed detects active infections in the blood and central nervous system. Several groups have published work that does not support IVP's theoriesby finding DNA from HHV-6 in controls at the same rate as in MS patients, for example. Carrigan says those experiments fail to distinguish between active and latent infection.

But there are other critics. Johnson, for one, said that no one has proven HHV-6 is more than an innocent bystander, perhaps ferried to the damaged areas of the brain by immune system cells during or after the disease process.

Toward Clinical Trials

With a host of complicated immune reactions muddying the picture, many researchers say that proving any infectious connection with MS will be difficult, if not impossible. At a 1999 meeting on the origins of MS, held in Brighton, England, discussion turned to Koch's postulates, time-honored rules for establishing an infectious etiology. Much of the talk focused on whether the postulates apply to MS. "With respect to HHV-6 and C. pneumoniae, there was a general sense that we would probably never be able to develop a causation," said Reingold.

Nevertheless, both IVP and Vanderbilt are plowing ahead with clinical trials. The Institute for Viral Pathogenesis had planned a trial with ganciclovir (Cytovene, Roche), a drug originally developed to treat cytomegalovirus, a close cousin to HHV-6. But the clinicians recruited for the study balked at ganciclovir's means of delivery, twice a day through a central line. Carrigan said the manufacturer had pledged enough drug for the study but that he and his colleagues are anticipating an oral version, currently wending its way through the FDA approval process.

While waiting, Carrigan says he gets several calls a week from neurologists wondering if they should prescribe antiviral drugs for their patients with MS. "If the patients really want to try an antiviral," Carrigan says, he suggests acyclovir (Zovirax, Glaxo Wellcome). One of Carrigan's collaborators, Joseph Brewer, MD, at St Luke's Hospital in Kansas City, said he has treated 10 patients with acyclovir, despite data questioning its effectiveness against HHV-6, a problem that Carrigan admits.

The practice is drawing fire. In an open letter to patients, John Fleming, MD, director of the MS clinic at the University of Wisconsin, wrote, "I think that it is premature to be taking antiviral drugs; at current doses, these may not go to the brain effectively, and long-term use may cause significant side effects." David Irani, MD, a Johns Hopkins researcher who develops viral models of MS, said he regularly encounters patients taking antivirals. "It's not being done in a rigorous manner, it's sort of off the cuff. And that makes it hard to know; it doesn't allow you to understand why or what's going on."

Reingold sees another problem. "Some people would like to think that a specific virus is responsible for the disease. But the more general view is that there is an immune reactivity problem. It's how a person's genetically predisposed immune system responded to infection a long time ago that triggers the disease. And if that's the case, then treating an adult with an antiviral medication won't do much good."

Meanwhile, the Vanderbilt team has begun a phase 2 trial of 40 patients, using the antibiotics rifampin and erythromycin. Supported by the NMSS, the study will use magnetic resonance imaging scans to image the size and location of brain lesions, tracking how they respond to the drugs. But because MS tends to progress and regress spontaneously, the brain lesions waxing and waning on their own, sorting out the drug's effects from background noise may be tricky. In addition, critics like Hammerschlag say that the two drugs under study may not kill C. pneumoniae or effectively cross the blood-brain barrier.

But until scientists' understanding of autoimmunity deepens, clinical research trials of drug treatments may be the best approach. Johnson summarizes the situation this way: "With such a complex web of cause and effect, you almost get down to the point where you say, if I give patients antibiotics and they get well, that's pretty good evidence. That's a throwback, a primitive way we used to do things." For the moment, though, it may be the only way.

© 2001 American Medical Association.