FDA Knew Problems At Vaccine Plant Were Pervasive...

FDA Knew Problems at Vaccine Plant Were Pervasive, Documents Show BY MIKE MCGRAW Knight Ridder Newspapers KANSAS CITY, Mo. - (KRT) - American regulators knew five years ago that contamination at the now-infamous vaccine plant in Liverpool, England, was both pervasive and persistent, documents show. In fact, evidence was so overwhelming that regulators should have shut down the plant or sought alternative flu vaccine sources long ago, say some vaccine and pharmaceutical experts who have reviewed the documents. Instead, the United States has lost 48 million doses of flu vaccine - half its supply - after British regulators found more contamination and suspended the plant's license last month. Documents obtained by The Kansas City Star show the Food and Drug Administration five years ago noted that the plant, then owned by the British firm Medeva, had failed to correct longstanding problems. But the agency continued to allow the plant to supply the U.S. market despite clear departures from what the agency calls "Good Manufacturing Practices." Indeed, Medeva plant managers failed to identify the source of serious and potentially deadly contamination in partially processed vaccines. They attempted to get around the problem by taking such steps as mixing samples of highly contaminated vaccines with cleaner lots in order to reduce bacteria levels. FDA officials say later inspections show the plant made improvements, but four experts who reviewed the documents at the newspaper's request questioned why the FDA did not take stronger action in 1999. "The license should have been pulled in `99 regardless of vaccine access issues," said Sarah Sellers, a pharmacist, former FDA adviser and consultant on pharmaceutical policy and regulation. "The problem was not new then, and it is not new today." The findings in the 1999 FDA inspection report angered one congressman who is helping to lead an investigation into this year's vaccine shortage. "To have years of warnings like this and do nothing is government malpractice," said Rep. Henry Waxman, a California Democrat and ranking minority member of the House Committee on Government Reform. "It is clear from this report that it is no unexpected accident that we have a flu vaccine shortage. The problems at the plant were systemic and known years earlier." Although the 1999 report showed widespread contamination in the manufacturing process, including unfinished vaccines, tests indicated that later sterilization made the completed vaccine usable. In fact, Lester M. Crawford, acting FDA commissioner, said Friday in a written response that this year was the first time inspectors found contamination in the finished vaccine. Although the 1999 inspection found a problem in the final vaccine, it was a sterility failure confined to one lot manufactured the previous year, he said. "The most significant issues identified in 1999 were the lack of validation for its manufacturing processes, including establishing proper limits for bioburden (including bacteria) and issues related to assuring sterility in the manufacturing process," Crawford said. Inspections in 2001 and 2003, which the FDA has not yet released, found that the plant had made improvements, he said. Crawford did not respond to suggestions by The Kansas City Star's experts that the FDA should have taken stronger action after the 1999 inspection. He also did not say whether the earlier contamination problems were solved. His statement did say that the FDA required additional corrective actions in 2001 and 2003. Officials of a Belgian firm that now owns Medeva did not return calls requesting comment. At the time, however, Medeva officials agreed to correct many problems the FDA identified. In some cases, they objected to the FDA's findings. Sellers and others said that, based on their experience, they think it is likely that the 1999 and 2004 contamination problems are related. But they cannot be sure. The source of this year's contamination is not known. British and U.S. officials have yet to release further details or more recent inspection documents. Waxman, however, said one thing was clear. "The tragic reality is that tens of thousands of vulnerable seniors will not be protected from the flu this season," he said. "If the government had taken strong action, the current crisis could have been avoided." The plant, which is more than 25 years old, has had a history of problems under a succession of owners. Just a few months after the FDA's 1999 inspection, polio vaccines made at the plant were recalled because of concerns they were contaminated with mad cow disease. In 2001, hepatitis A vaccines being filled into syringes at the plant were also ruined by contamination problems. The next year, routine testing of tuberculosis vaccine made there showed low potency levels. During those years, Medeva merged into Celltech Pharmaceuticals Ltd., another British firm. In 2000, the plant was sold to PowderJect Ltd., another U.K. pharmaceutical company. Chiron Corp., a U.S. biotech firm, purchased PowderJect in mid-2003. The plant has changed hands so many times that the name of one former owner can still be seen under the light green paint on the brick-and-metal building. Most owners said they had invested in plant improvements. Throughout all the changes, the vaccine, Fluvirin, has remained the same. About the time of the Chiron purchase, the FDA returned and again found contamination in the flu vaccine process, but no significant problem with the vaccine itself. The FDA relied on Chiron to fix the problems, but the company reported in August that some vaccine was contaminated. FDA inspectors returned in October after British regulators suspended the plant's license for three months, citing contamination in the flu-vaccine-making process. Crawford said the FDA in each inspection in recent years had reviewed the company's plans to respond to its findings. "If fully implemented, the company's plans appeared adequate to correct deficiencies," he said. In a class-action lawsuit filed last month, Chiron stockholders allege that the company knew the plant had serious contamination problems since at least 2000 and failed to address them or warn its stockholders. "Rather than coming clean about the serious deficiencies at the Liverpool plant," the lawsuit said, "defendants concealed these problems for as long as they could." A Chiron representative said: "We are taking a prospective rather than retrospective view right now with our public communications. Right now the company is focused on our remediation plan and working with (U.S. and British authorities) to get back on track for the 2005-2006 season." The Kansas City Star obtained a copy of the FDA's 1999 inspection of the plant last week. The report was part of a lawsuit filed on behalf of a woman who allegedly became paralyzed as a result of a 1998 flu shot manufactured at the Liverpool plant. Chiron settled the case earlier this year, but the details are sealed, said the attorney in the case. The report shows that FDA officials visited the plant for over a week during a biennial inspection in 1999 and found 17 problem areas. For example, the plant: Mixed samples of highly contaminated vaccines with cleaner ones in an effort to improve test results. Failed for several years to verify to the FDA's satisfaction that its production process met rigorous standards. Raised its limits for allowable contaminants, such as bacteria and toxins. Failed to control air flow and other possible contamination sources. Could not produce critical documents about problems in the manufacturing process. Inspectors also said the plant, unlike newer facilities, lacked alarms that would alert workers to bacteria entering through the heating and cooling system. They noted that at least 21 batches of the vaccine distributed during the 1999-2000 flu season had to be filtered more than once to make them sterile. Before the vaccine went through its final filtering, testing found such bacteria as E. coli, which produces a powerful toxin and can cause severe illness, and Chryseomonas luteloa, which can cause meningitis. Testing showed that at some points in the process, the vaccine had bacterial counts hundreds of thousands of times higher than plant tolerances. Although testing in 1999 did not identify all of the specific bacteria found, there was no mention that tests detected Serratia, the bacteria that entered this year's batch of flu vaccine during the final step of the manufacturing process. The report noted that FDA inspectors had found additional problems in 1997, and that Medeva officials were in the midst of their ongoing investigation into increased contamination inside the plant. The FDA inspection quoted from a draft of a management report that said monitoring had begun to show an increase in contamination since about 1998. "To date the evidence suggests that the source of the microorganisms lies within the process equipment or materials rather than chance contamination from the environment or operations," the Medeva report said. The FDA issued a warning letter to Medeva the following October, citing "significant deviations from the acceptable standards and requirements of the Food, Drug and Cosmetic Act." Problems included failure to maintain and sanitize equipment "to prevent malfunction or contamination." License suspension was one possible result, the agency said. The warning letter was one of three the FDA issued among 20 vaccine manufacturers it inspected that year. The company responded on Nov. 15, accepting some of the findings and arguing in other cases that the FDA had misinterpreted some of the information its inspectors gathered. Company officials noted they were still looking into increased microbial levels but there were no "conclusive findings." The FDA responded in March 2000, saying it would verify the company's corrective actions during the next inspection in 2001. In the meantime, the director of the FDA's case management division wrote, "The FDA continues to have concerns." Specifically, she said, the company's allowable tolerance of contaminants was unacceptable. After reviewing the inspection report, several experts said the problems were significant. The experts comprised a physician who specializes in virology, two pharmaceutical experts and a former industry executive who now does consulting work. "If I were the FDA, I would have said, `Let's go get a contract with a new plant,' " said Ashim Mitra, chairman of pharmaceutical sciences at the University of Missouri-Kansas City. The process of mixing cleaner vaccines with more contaminated samples alone "is a reprehensible practice," Mitra said. While the finished vaccines were most likely safe, Mitra said, the continuing contamination problems rendered the plant "a time bomb" in which contamination could eventually reach the final product. "Oh, dear," said William Schaffner, a Vanderbilt University infectious disease expert, when told of the mixing operations and other practices. "If you are a plant manager, that is one way of meeting a standard, technically, but you are evading the intent." Schaffner said he was not an expert on flu vaccine manufacturing, but as a doctor, "I would not expect a physician or someone interested in vaccines to use a vaccine where a manufacturer gets this kind of report card." Sellers said the state of affairs inspectors found at the plant was frightening. She said she was surprised the FDA's warning letter did not determine the company's vaccines were "adulterated" based on a strict interpretation of federal food and drug laws. Such a finding would require the FDA to pull products from the market. The fact that the plant's manufacturing process had not been validated (certified by the company as effective) since the early 1990s was a "stunning" finding by the FDA inspection team, said Sellers, showing possible lapses in earlier FDA inspections. Sellers thinks that the FDA needs more resources for overseas inspections, and that it missed an opportunity to avoid today's health crisis. "If the license was pulled and the problems dealt with in `99, we may not be facing a national public health crisis," Sellers said. The British license suspension this year should not have surprised U.S. officials, one expert said. "It should have come as no surprise that there was a potential for this episode," said David Webster, a former Aventis executive and now a biotech consultant. "You have all sorts of evidence, including these documents from 1999, the earlier recalls and manufacturing problems. It's not like this plant had batted a thousand for 10 years and then got into a slump for a month." In fact, Schaffner, a physician and chairman of the Department of Preventative Medicine at Vanderbilt, said he was surprised no one at the FDA seemed to go into attack mode when the most recent contamination problems became evident two months ago - especially given what they now appear to have known five years ago. "Nobody at the FDA said `Holy Toledo, we've had bad contamination problems there in the past and these new problems imperil half the supply of flu vaccine in the U.S.,'" he said. "The FDA seemed to handle this in its routine, bureaucratic fashion. "The light never went on; they dropped the ball."