Abstract

While Glaxo insists that LYMErix is safe and effective, the questions continue to mount. Is the vaccine provoking a raging, destructive immune response? Is it turning asymptomatic Lyme into symptomatic forms of infection? Is the disease definition itself arbitrary or wrong? And, finally, why are these questions being asked now, after LYMErix has already been released?

Heading to Bethesda this past January for the Food and Drug Administration's (FDA) meeting on the Lyme disease vaccine LYMErix, I expected a somber and orderly affair. Instead, for the price of gas and tolls from New York, I bought a ringside seat at a raucous, riotous, and bitter free-for-all worthy of Jerry Springer. Before the meeting was through, enraged FDA panel members questioning the manufacturer, GlaxoSmithKline, would alternately yell and laugh at the company's experts, sometimes making such pointed fun of them I thought I was at a roast. When the FDA advisors were done, a lineup of furious, litigious patients - the "LYMErix vaccine victims" - delivered testimony both shocking and heartbreakingly sad. Yet in the end, LYMErix was left standing. Without asking for so much as a label change, the FDA charged GlaxoSmithKline with the task of submitting more data and validating the vaccine's worth. Was there a method to this madness? The answers, I would learn, lay in the politics of drug approvals, a protracted debate about Lyme disease, and the bizarre saga of a molecule called OspA.

The most common vector-borne illness in the United States, Lyme disease, is caused by the spirochete bacterium Borrelia burgdorferi (Bb), and is spread by the Ixodes tick. About 16,000 new cases a year are reported to the Centers for Disease Control and Prevention (CDC), a figure the CDC itself estimates may represent 10 to 20 percent of those meeting its surveillance criteria, although no recent study pinpoints the number exactly. With about 100,000 new cases a year meeting the CDC's strict standard, and an uncertain number of infected individuals the agency says fall outside those parameters, this group presents an important health concern. The reason is that while Lyme disease is usually easily cured if treated early on, late or partial treatment can leave patients extremely ill or even disabled, their arthritic, neuropsychiatric, or gastrointestinal symptoms lasting for months to years.

The problem is compounded by debates over who qualifies for diagnosis at all. The question is so difficult because Bb bacteria are elusive, quickly leaving the bloodstream for tissue of the joints or brain. Tests must, therefore, detect the organism indirectly through measurement of an immune response that varies according to bacterial strain, the infected individual, and stage of the disease. Because there were so many different tests and diagnostic standards in the early 1990s, chaos reigned. All agreed the uncertainty called for new, more accurate testing. And no one wanted to settle the confusion through standardization more than SmithKline Beecham, at the time the parent company for LYMErix. Without an official definition, after all, the company could not tell the difference between adverse event and vaccine failure. Without a definition, it would be impossible to conduct clinical trials or move a product toward approval at the FDA.

To resolve the matter, the company, the CDC, and the FDA got together in the spring of 1994 and agreed upon a case definition, including a two-tier diagnostic test based on measurement of antibody response in the blood. In the first step, an ELISA (enzyme-linked immunoabsorbent assay), scientists looked for antibodies responsive to a mixture of whole-cell Bb spirochetes. Because the mixture included not just proteins specific to the microbe, but others found more widely, the scientists adopted a second, confirmatory test, a Western blot that detected a smaller, more specific set of antigens; these antigens (and the Western blot bands that represented them) were derived from a statistical analysis of patients in a study conducted by the vaccine's chief investigator, Allen Steere, the Yale University rheumatologist who first recognized Lyme disease in Connecticut and now heads the Rheumatology and Immunology Department at the New England Medical Center, Tufts University School of Medicine.

Five months later, in October, the same two-step serological standard was adopted for surveillance and research purposes in Dearborn, Michigan, at the Second National Conference on Lyme Disease Testing, sponsored by the Association of State and Territorial Public Health Laboratory Directors and the CDC. The most divisive part of the two-step diagnostic standard - now called the Dearborn criteria - was elimination from the Western blot of two Bb proteins, outer surface protein A (OspA), from which LYMErix was made, and outer surface protein B (OspB), the intended component of next-generation vaccines. For the vaccine trials, this made sense. In a universe of the vaccinated, testing for OspA antibodies would only serve to blur the line between inoculation and disease. But removal of OspA and OspB for other purposes was viewed with alarm by many practitioners, who knew these proteins were specific to Lyme disease and sometimes the only markers present in those with late-stage disease.

"The CDC said the standard was not to be used for diagnosis," said Nick Harris, president of IgeneX, a California reference laboratory that tests for vector-borne diseases, "but they did not seem to realize how difficult they were making that choice for local physicians, who look to CDC definitions for guidance and take test results at face value - positive or negative - without reading between the lines. Without OspA or OspB to serve as markers, many of the sickest patients no longer met any diagnostic standard," Harris says. "By excluding these patients from diagnosis, we excluded them from treatment as well."

One of the most political molecules in the history of science, OspA has, since those early days, become a cause celebre in the embattled world of Lyme disease. Was OspA removed from the Dearborn definition by power players who callously disregarded the sickest of the sick to enable a vaccine to be developed, as many angered physicians believe to this day? Or, as asserted by the CDC and the Dearborn voting panel, was the definition used in both venues because it was, in fact, the most precise?

War Room at Versailles

Whatever the answer, it was under the umbrella of the Dearborn criteria that LYMErix journeyed through the product pipeline and finally received a pass from FDA scientists in the Versailles Room of the Bethesda Holiday Inn in May of 1998. But the stamp of approval was about as ambivalent as members of the committee had ever seen. In fact, despite the go-ahead, concerns were legion. Some panel members wondered whether the OspA vaccine would prevent accurate diagnosis of Lyme disease caught after protection wore off. Others worried that LYMErix might cause relapses in those with previously diagnosed Lyme disease, or worsen symptoms in those with current Lyme disease. The biggest concern was voiced by the chief investigator, Allen Steere. Findings from his lab at Tufts University suggested the possibility that LYMErix could cause a particularly onerous form of treatment-resistant Lyme arthritis in people with a gene called HLA-DR4, present in about 30 percent of the U.S. population and linked to severe rheumatoid arthritis. Published a few months later in the journal Science, Steere's evidence, while circumstantial, showed a striking resemblance between a portion of the OspA molecule and the human protein, LFA-1. In genetically susceptible individuals, Steere's theory went, T cells primed to attack OspA might also recognize and attack human cells lined with the "molecular mimic," LFA-1. The result, Steere suggested, might be autoimmune disease, in which T-cells continued their attack on the mimic even when OspA was gone. [1].

In the end, the committee signed off, reluctantly, declaring a leap into the unknown. The group's sentiment was best expressed by panel member David Karzon, professor at Vanderbilt University Medical Center: "Those who did the trial," he said, "have unearthed some very interesting sinister possibilities that may or may not be real."

Given this turbulent history and the hailstorm of lawsuits that have followed the vaccine's commercial distribution, it's no surprise that this past January, when FDA panelists reconvened in the Versailles Room on the issue of LYMErix, they were prepared to joust.

Plentiful ammo was provided by the sponsor (now, following a merger, called GSK, or GlaxoSmithKline.) To cast doubt on the genetic risk factor, for instance, GSK scientists had inoculated arthritis-prone mice with OspA and found no sign of autoimmune disease. Isn't this study "irrelevant," several panel members asked, since the mice had no analog to the human mimic in question, the protein LFA-1. Glaxo also reviewed a safety study conducted at an HMO. Unfortunately, the vaccine had had such bad press, the study coordinator said, that the uptake had been slow - while 25,000 participants were required for completion, the company had only 2,800 participants to date. How would the sponsor get so many additional participants in the year remaining, asked a panel member, "when it is possible that a hearing like this will make people less comfortable and doctors less comfortable, and there will be a gigantic falloff. Do you have any idea what is going on?" Finally, the company touted a pregnancy registry with "no unexpected findings" and only 4 miscarriages out of 13. "You make it sound as though you find no consequences. I don't consider that . . . no pattern of anything," the Mayo Clinic's Michael O'Fallon fumed. The comments "really disturb me," he said. That was when a tall, well-dressed man swept up to the podium from somewhere in the back and motioned the presenting scientists away. It was time for damage control, and David Wheadon, vice president of regulatory affairs at GSK, took charge. "Certainly spontaneous abortion, within the context of pregnancy, in an overall population, is not something that is unexpected," Wheadon told the panel, "and I think that was, indeed, what was intended to be said."

The LYMErix vaccine victims could not have engineered it better if they had written the Glaxo script themselves - the panel was primed to hear their stories, 20 in all. There was Emily Biegel, who addressed the panel for her husband, John, vaccinated in April and May of 1999. "Some of you may have seen him come in with a walker," she said. An active outdoorsman before vaccination, John has since been through four hospitalizations, atrophy, insulin dependence, compression fractures, tremors, and 25 plasmaphoresis treatments. He is positive for HLA-DR4.

Jenny Marra, a New Jersey hospice nurse positive for HLA-DR4, said she had been living with "severe joint and muscle pain since vaccination in 1999. SmithKline did not include a warning about the potential risk for this information in the product labeling or inform the health care providers of this concern. Had I known this I personally would not have taken the vaccine." Physicians aware of the political controversy, she added, are turning the LYMErix vaccine victims away "with statements like, 'I don't want to get involved.'"

Karen Burke, the mother of two toddlers from the Pocono Mountains of Pennsylvania, said that her husband, the vigorous owner of a construction business, got his second dose of LYMErix in July of 1999. By October, he couldn't roll over in bed. "My standing joke with him is, honey, at least when our kids are big enough to go to Disney World you'll be well enough to sit in a wheel chair, and we'll get to the front of the line. It's not funny, but you have to have some fun in your life," Burke said.

Benjamin Luft, a panel member from the Department of Medicine University Hospital and Medical Center Health Sciences Center at the State University of New York at Stony Brook, described the "twilight zone" of the "disconnect" between the patient testimony and the sponsor's denial of significant adverse events.

Would he be tempted to try this new vaccine. "The answer," said O'Fallon, was emphatically "No!"

"My concern is greater than it was before," said Patricia Ferrieri, University of Minnesota Medical School, Minneapolis, a longtime panel member and the person who chaired the FDA meeting on LYMErix in 1998. "Are we going to be able to resolve these issues expeditiously, or should you put a moratorium on the vaccine until you are able to very critically examine what we have. . . . I've never had to say this before," she told the FDA scientists in the room. But "in all of the years I've been on the committee, I've never heard this type of concern iterated without agency response that has satisfied the dissatisfying. . . . I consider what we're dealing with today to be very, very serious, and I would like to throw back to you the need to reexamine how this fits into your mission and in the public health realm. There are too many ifs here for us to feel secure that the answers will be forthcoming . . . you have to examine where you are and what we owe to the public."

And the Bands Play On

A metaphor for science-gone-wrong among the tick disease crowd at the annual Lyme Disease Foundation Conference this past April in Farmington, Connecticut, LYMErix was, as expected, omnipresent - mentioned in talk after talk, it functioned as data, as anecdote, as the proverbial wrench in the works. In fact, the news coming out at the conference and elsewhere around the country was bizarre. In physician offices, in diagnostic labs, and now in clinical and controlled studies, LYMErix recipients without any known exposure to Bb and no symptoms of Lyme disease were testing Dearborn positive. What's more, those who once had Lyme seemed to be relapsing into the symptoms of the disease.

Paul Fawcett, director of the immunology laboratories at the duPont Hospital for Children in Wilmington, Delaware, and a noted expert on Lyme disease serology, says he'd observed the ability of the OspA vaccine to provoke a wide range of Borrelia-specific bands on Western blots well before the product reached market, as patients involved in clinical trials appeared for routine Lyme disease tests. Fascinated by the phenomenon, he coordinated a study of 20 adult volunteers, all employees of the hospital, who received three vaccine doses each and submitted blood for analysis.

As it turned out, the elaborate banding patterns showed up in all but one subject in Fawcett's experimental group. In fact, the banding was so robust that 30 days after the second dose of vaccine, the only two commercial Western blots then approved by the FDA were "rendered virtually useless for diagnostic purposes." On one of the FDA-approved tests, for instance, he found that OspA vaccinees tested with "antigens covering the whole length of the strip, so that they were positive for Lyme disease by CDC criteria. These people were so reactive," adds Fawcett, "that they often showed 15 to 20 bands," far more than the minimum requirement of five. The other FDA-approved Western blot, he notes, "showed several bands below the OspA region and one dark gray smear of reactivity at OspA and above."

What's more, Fawcett found that the odd patterns were sometimes accompanied by adverse events. Two of the 20 in his study - one physician and one therapist - developed severe arthritic pain, and the strange symptom, not generally seen in Lyme disease itself, of swelling hands. "There's absolutely no question these are the result of the vaccine," Fawcett states. His feeling was only strengthened in yet another study, this one of children participating in LYMErix clinical trials. Here, he found the vaccine could literally retrigger or worsen symptoms of the disease. In one instance, a 16-year old presenting with severe, recurrent arthritis had been infected at around the time of his third LYMErix dose. "This was a vaccine failure," says Fawcett, treatable with antibiotics, "but LYMErix apparently worsened the course of the disease." In another instance, a six-year-old vaccinee with previous, neurological Lyme disease but no evidence of current infection experienced a full-blown return of symptoms as his banding pattern bloomed.

What could be going on? Describing himself as a "fan of data," Fawcett reviewed his findings and concluded the only explanation was a "hyperactivation" of the immune system after exposure to the vaccine. "This test group was clean," he says of his adult trial, "with absolutely no serological evidence of prior exposure to B. burgdorferi at baseline. Part of what we see could be cross-reactivity, with OspA stimulating antibodies that match, even if imperfectly, the Borrelia burgdorferi bands on the Western blot. But that can't be all of it." The rest, Fawcett theorizes, "may be a generalized, nonspecific, broad-spectrum activation of the immune system." It is this phenomenon, he notes, that would account for adverse events.

A study from Sam Donta, professor of infectious disease at Boston University School of Medicine, suggests that LYMErix can retrigger old, presumably "cured" cases of Lyme. Donta says he was alerted to the possibility after the vaccine hit the market and he began to see, within his own practice, LYMErix recipients who appeared to have the symptoms of chronic Lyme disease, most often reported after the third shot. Donta found that these patients tended to test positive for Lyme bacteria proteins other than Osp-A on Western blots. Moreover, treating them with antibiotics, he found most got well, just as he would expect in bona fide cases of the disease. In a formal study of 50 such patients, 25 within his own practice, Donta has found the observations hold.

Why does he believe these adverse events represent reactivation of previous Lyme disease instead of the autoimmune reaction suggested by Steere? "Because in cases where patients have had Lyme before, the flare-ups induced by the vaccine caused the same types of symptoms in the same location of the body, revealing a disease fingerprint specific to each patient, and generally observed in those who relapse. Either they coincidentally got Lyme disease during the series of vaccinations, or they had the disease already," Donta adds. The latter seems more likely, he says, "because patients have responded to antibiotics after suffering from their vaccine reaction for months or years."

Providing a third perspective is rheumatologist Philip J. Molloy, medical director of Imugen, the Massachusetts diagnostic laboratory identified by many in the mainstream as the sine qua non for diagnosing vector-borne disease. Molloy's investigation, published last year in Clinical Infectious Diseases, concludes that the problem is not the vaccine, but the Western blot itself [2]. Like everyone else, Molloy found that vaccination led to a complex pattern of multiple bands, including CDC diagnostic bands, on Western blots, making it difficult to determine which bands came from the vaccine and which ones from infection. "It was possible to tell whether or not they had been vaccinated," says David Persing, vice president of research for Corixa, who did the study with Molloy, "but not whether they had Lyme."

To resolve the problem, the researchers have patented an OspA-less strain of Bb, which they now use to make vaccine-specific ELISAs and Western blots. "We know the bands that show up are due to infection," says Molloy, "when they show up on Western blot strips made without OspA." While Molloy says the phenomenon "has yet to be fully explained," the theory he favors is degradation of OspA into smaller fragments and buildup of OspA into larger particles, resulting in Western blot tests with a diversity of bands that seem to confirm the disease. One interesting implication of the finding, he adds, is that the banding pattern chosen at Dearborn may "represent an immune response to OspA, which is being 'counted' several times, while other bands presumed to be present are not really there at all." In fact, says Molloy, "Dearborn is irrelevant, an artifact of the Western blot strip" misinterpreted by experts for years. Many of the bands the CDC considers diagnostic for Lyme disease, he adds, may often appear in reaction to antibodies for OspA - the very molecule removed as statistically insignificant in 1994. "We're working on finding more appropriate banding patterns for our own tests," he adds.

Defending the accuracy of the serological criteria adopted at Dearborn, Ned Hayes, chief of epidemiology at the Bacterial Zoonoses Branch of the Division of Vector-borne Infectious Diseases at the CDC, says that "the two-stage testing has close to 100 percent sensitivity in patients with Lyme arthritis and appears to be somewhat less sensitive in patients with late neurologic disease." Moreover, adds Hayes, "we believe that experienced laboratories are able to discriminate between vaccination and infection in most cases through careful interpretation of the blot pattern."

"If the CDC is correct and Dearborn is relevant, then the vaccine may be triggering an immune response to Borrelia burgdorferi in people we never recognized as infected with the spirochete, although they were. It could mean there is a lot more Lyme out there than we ever realized," says Harris of IgeneX. "But, if Imugen is right, we need to go back to the drawing board and determine a new definition for the disease."

Fawcett takes a middle ground, disputing the notion that latent infection emerges, but contending the hyperactive banding patterns represent an immune reaction of serious concern. "This indicates not only that the vaccine is immunologically reactive," he says, "but also that the disease pathology is far more complex than we have understood. Forgive me for being a scientist," Fawcett adds, "but these OspA bands are the most interesting thing to come along in Lyme disease in years."

As for GlaxoSmithKline, communications director Carmel Hogan says the company cannot add to the debate about the banding patterns right now. "We would need more time," she states, "for our scientists to study the papers and reports in depth."

In June 2001, five months after the meeting in Bethesda, LYMErix is now in data-gathering mode - one aimed at determining whether it causes irreversible autoimmune disease or other adverse events. Yet the concerns expressed by the panel members have not been provided to patients, who must be inoculated with LYMErix to enable the review. According to Robert Ball, director of the FDA's Vaccine Adverse Event Reporting System (VAERS), "the stories that people tell you are terrible and your heart goes out, but you have to look for patterns in the data to determine whether a problem is really there." Especially important to the FDA, Ball explains, are "serious events that result in life-threatening illness, hospitalization, or disability." And here, he says, no clear pattern emerged. "Only a small minority, 85 people or eight percent, of the reported adverse events following LYMErix administration were classified as serious, according to this definition," states Ball. "Of this group, we have not identified any clear patterns in the reports. The neurological events were diverse and no single condition predominated. Events involving stroke were reported, but these events are relatively common in the older age group in which these events occurred. Only hypersensitivity reactions, which are uncommonly reported, can be plausibly linked to the vaccine because of their specific timing and clinical features."

As to the risk of arthritis, the main concern expressed by the theoretical work, or the retriggering of Lyme itself, Ball says that, based on his study of the VAERS reports, no disturbing patterns could be found. "If there is an effect, it is pretty small," Ball adds, "and the VAERS system may not be sensitive enough to pick it out." To dig deeper, the FDA is planning internal studies of HLA types and cellular reactivity to OspA. On labeling issues, including warnings about the possible risk of the arthritis gene or prior Lyme, the FDA can only say it is "working closely with the sponsor," but not whether changes will be made.

Stephen Sheller, the Philadelphia attorney representing some 250 LYMErix vaccinees in personal injury suits against GlaxoSmithKline, says Ball is manipulating data "by focusing almost exclusively on 'serious' events that result in hospitalization, permanent disability, or death, while discounting the far more prevalent 'severe' event. I've been contacted by hundreds of individuals whose lives have been drastically affected by a chronic inflammatory process which is neither life-threatening nor requires hospitalization, and whose permanence has not yet been determined," Sheller explains. "These people don't meet the requirements of the FDA standard, but they do usually fit the definition of the 'Grade 3 Severe' adverse event, defined as a problem that prevents normal everyday activities, and approved by the FDA for the human clinical trials of LYMErix. In a young child, for instance, this kind of reaction would prevent attendance at school or day care, and would cause parents to seek medical advice. I believe the FDA and GSK have information demonstrating the true rate of Grade 3 Severe adverse reactions in both adult and/or pediatric clinical trials to be in excess of 20 percent." Spearheading a class action suit seeking labeling changes for LYMErix, Sheller says "this information must be fully and clearly disclosed to the medical community and to consumers at once."

His fight with GlaxoSmithKline appears headed to court. "Based on all our scientific evidence we believe the lawsuits to be without merit and will defend against them," states GlaxoSmithKline spokesperson Hogan. "All the evidence to date from clinical trials and post marketing data establishes LYMErix to be safe and effective. Over 15,000 people took part in the clinical trials, and GSK has distributed over 1.3 million doses representing some 400,000 vaccinees," Hogan states. (Sheller argues that no more than 100,000 have completed all three doses.)

"There is no scientific evidence that individuals with HLA DR4 genotype are at increased risk of developing adverse events from the product. The company was aware of this theoretical issue and, indeed, this matter was examined in study participants in clinical trials," Hogan goes on. As to Sheller's assertion that 20 percent of those in the pediatric trial suffered Grade 3 adverse reaction, she clarifies: "Approximately 20 percent of the children who participated in the pediatric clinical studies did report adverse events that were characterized as 'severe,' a term that was defined in the study protocols as 'preventing normal daily activity.' But the vast majority of these reports involved localized injection site events such as soreness, pain, or swelling that prevented the children from throwing a ball or playing with others for a limited period of time following vaccination." Hogan also notes that "in a study in which 4,087 healthy children between the ages of 4 and 18 were vaccinated, arthritis was no more frequent in those who received the vaccine than in those who received the placebo."

For those investigating LYMErix outside the FDA-pharma circle, the questions just mount. Is the vaccine sparking a devastating autoimmune reaction that places a third of its recipients at risk for something much worse, much less treatable, than Lyme disease? Is it igniting asymptomatic Lyme disease (an entity researchers now say may be as pervasive as the symptomatic kind), revealing infection with the Bb spirochete to be persistent or far more common than generally thought? Do the bizarre Western blot patterns reflect a raging, expansive immune response to the OspA molecule, as Paul Fawcett believes, suggesting that Lyme disease pathology is still misunderstood? Or are the bands merely artifacts, testament to gross misinterpretation of lab results by the field's leading lights going back years? Are the answers even knowable in the face of what could be the ultimate nightmare scenario - vaccine trials performed according to a disease definition that is incomplete, arbitrary, or wrong? Perhaps most pertinent, why are we asking these questions now, after the product's release?

Some insight comes from Alan Barbour, professor of infectious disease at the University of California at Irvine, and one of two inventors on the OspA patent that was filed in 1988 by Symbicom AB, a small Swedish biotech company (now part of AstraZeneca). Though Barbour himself did not work on the vaccine, he recalls the race to market between two companies, SmithKline Beecham and Pasteur Merieux Connaught , now Aventis Pasteur (which eventually dropped out). "That race would be an intriguing topic for an article in the New Yorker magazine," he states.

A result of that competition, says one scientist close to the action, was pressure to complete clinical trials so the vaccine could move through the approval process. "And in retrospect," he says, "the approval itself was rushed, mainly because it was not known how often booster immunizations would be needed and what the consequences of getting or not getting the booster would be."

Government watchdogs say the problem may be conflict of interest, an issue recently investigated by the General Accounting Office, an arm of Congress. In a two-part report released this month (posted online at www.gao.gov/new.items/d01755.pdf and www.gao.gov/new.items/d01787r.pdf), the GAO found no profound conflict, stating that "federal agencies generally meet requirements for disclosure and review of financial interests related to Lyme disease." Yet patient advocacy groups hold that, while not illegal, the potential conflicts of interest on the part of decision makers are of concern. According to "Conflicts of Interest in Lyme Disease," a report from the Lyme Disease Association to which this reporter contributed, the Dearborn panel setting the disease definition had particular potential for bias. Indeed, the nine voting consultants hired by CDC included a scientist holding the patent for OspA; the inventor of the canine Lyme vaccine, Lymevac; the CDC scientist named as inventor of the "P37/FlaA protein antigen," with potential for use in next generation vaccines and diagnostic tests; and Allen Steere, who was both an author of the study used to generate the case definition and lead investigator for clinical trials of the vaccine.

As to the FDA panel that approved LYMErix in 1998, the report highlights a State University of New York at Stony Brook scientist given voting rights by the FDA. According to the official transcript, this researcher disclosed a consulting relationship with the pharmaceutical manufacturer and received a waiver. However, the transcript does not mention that the scientist and his colleague, also a researcher at Stony Brook and a voting member of the panel, were principals of a company with a product line directly dependent on the availability of the OspA vaccine.

Moreover, the LDA adds, the government and corporate entities with vested interest in LYMErix and associated Lyme disease products are vast. U.S. government agencies, including the CDC, the National Institutes of Health, and the Department of Defense own partial rights to revenue from more than a third of the 56 U.S. patents identified as especially significant for Lyme disease vaccines and tests. What's more, GSK may not be the only company with revenue rights to LYMErix. Also poised to derive benefit based on possible interest in the patent are multinational life science giants Aventis and AstraZeneca.

Attorney Stephen Sheller, meanwhile, compares the LYMErix situation to the handling of the diet drugs fen-phen and Redux. "These drugs caused heart valve problems in hundreds of thousands of people before industry or the FDA chose to act in September of 1997 to protect consumer safety." In fact, Sheller notes, a recent editorial in The Lancet documented private FDA communications that apparently "subverted official procedures" and "suppressed scientific debate and open review" within the agency in the case of another GlaxoSmithKline drug, Lotronex. Hogan of GlaxoSmithKline counters that The Lancet article is misleading. "As with all our medicines and vaccines, we have and will continue to work closely with FDA, in line with all regulatory requirements and obligations," she states. Sheller, however, believes that "the conduct criticized in The Lancet might be repeating itself with the LYMErix vaccine."

But if the powers that be have been spinning Lyme disease for profit, they have made a bumbling mistake: The hub of the so-called strategy, the OspA protein, now wreaks havoc, careening through Western blots like a zany free radical and bringing out more bands than Woodstock. Whether these bands signal a true immune response or just decades of misinterpretation, they demand that we cipher their meaning. In doing so, we'll be forced to rethink the Dearborn criteria, the meaning of Lyme disease, and the clinical trials that propelled the vaccine through approval at FDA.

In the end, the problems of LYMErix may be rooted in something far less organized than insidious - the hubris of medical science, which has sold its soul to industry for the funding it needs to survive. To be true to itself, science must acknowledge the gray areas, but to fit the needs of business, it must deal in black and white. The Nobel Prize-winning geneticist Barbara McClintock put it best. To do real science, she said, scientists must have "a dialog with nature." But to send a product down the FDA pipeline, it is the outcome, not the dialog, that counts. Experimental design and disease definition created in the shadows of the drug approval process must, by McClintock's standard of science, be forever flawed.

Pamela Weintraub is a former staff writer at Discover, former editor-in-chief of Omni Internet, and the author of 15 books on health and science.