he controversy surrounding a study of whether moderate drinking might prevent cardiovascular disease isn’t over: If one interpretation of federal regulations is correct, the study may be in violation of Food and Drug Administration requirements meant to protect the health of research volunteers.
STAT has learned that the study’s leaders failed to seek a form of regulatory approval intended to protect study participants and ensure they understand the possible health risks of the research. By not seeking approval from the FDA, said public health researcher Dr. Michael Siegel of Boston University, the study “is in violation of federal law.”
But the hospital whose physician designed the trial said he asked the FDA whether the investigational new drug (IND) application was required and was told no. Other experts said the requirement for an IND in this case is not crystal clear. Of seven additional researchers or regulatory consultants interviewed by STAT, four agreed with Siegel that FDA approval via an IND was required and the others weren’t sure.
The idea that a study involving a food or drink might be required to file a document covering a new drug seems bizarre. But in a 2013 regulatory guidance, the FDA said that a food is considered a drug if it is “intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease.” A human study “intended to evaluate the effect of a food on a disease would require an IND.”
The alcohol study is intended “to determine whether drinking one alcoholic beverage per day will decrease the chance of developing heart disease,” according to the study’s website.
The $100 million study has already been derailed by other issues. Enrollment in the “Moderate Alcohol and Cardiovascular Health” study was suspended in May over concerns that federal officials solicited funding for it from alcohol beverage makers, and last week Anheuser-Busch InBev announced that it had withdrawn its funding.
A spokesperson for Beth Israel Deaconess Medical Center in Boston, where lead investigator Dr. Kenneth Mukamal is an internist, said that prior to the study’s launch, Mukamal contacted the FDA about the need for an IND. The FDA, said the spokesperson, concluded that none was required because the study raised sufficiently minimal concerns.
The FDA did not directly answer questions from STAT about whether the study, called MACH15, requires an IND. But a spokesperson said that, for regulatory purposes, alcohol is considered a food. Like other foods, alcoholic beverages can “meet the definition of a drug” if they are being studied for their effects on a disease.
In a statement, Beth Israel Deaconess said it has “strong policies in place to ensure the scientific and ethical integrity of any research study involving our investigators.” MACH15, it said, was evaluated by its internal review board, and after additional review “we have not found any reason to believe that it does not adhere to our institutional requirements.”
The significance of failing to file an IND goes beyond a possible regulatory misstep. Its purpose is to let the FDA review a proposed study’s safety and risk information and assess whether the study is justified or places participants at unreasonable risk.
“An IND would have laid out to the FDA what risks study participants would have been told about, allowing the agency to require revisions if it found the warnings fell short,” said BU’s Siegel. MACH15 does not mention that alcohol increases the risk of cancer, especially breast cancer among women. That risk “is being increased by being enrolled in the study and assigned to the alcohol group,” Siegel said. “That could be defined as unnecessary risk, especially since the study could include just males.”
One reason the IND requirement is somewhat murky is because the FDA’s explanation of when a food is a drug, and therefore needs an IND, includes whether it’s being studied for use in “diagnosis, cure, mitigation, treatment, or prevention” of disease — but not reducing the risk of one. Experts disagreed on whether prevention and risk reduction were essentially synonymous.
“It seems plausible that alcohol would be considered a drug” in this context, said Dr. Michael Carome of Public Citizen, “but it’s a legitimate question” whether MACH15 needs an IND.
When nutrition researcher Connie Weaver of Purdue University wanted to study whether blueberries are good for bone health, she said, it took her a year to persuade the FDA that she didn’t need an IND. “But the basis for exemptions is unclear,” she said.
Others argued, however, that the requirement was clear. “If you wanted to study the effects of milk on bone density, milk would be considered a drug and you’d probably need an IND,” said Kevin Maki, an expert on clinical trial regulation at MB Clinical Research and Consulting. “Since we’re talking about preventing heart attacks,” which MACH15 intended to study, “that’s clearly a drug-related claim” and so would trigger the IND requirement.
The need for an IND is driven by whether the study aims to study endpoints such as the prevention and mitigation of disease, said Claire Kruger, president of Spherix Consulting, which focuses on regulatory issues. “It doesn’t matter if you’re studying pizza or Advil: If that’s the question you’re asking, it immediately means you have to do the study under an IND. If you’re asking can alcohol prevent heart attacks, you have a drug study” as FDA defines “drug.”
Kruger added, “it surprises me” that MACH15 did not file an IND.
Joshua Miller, chairman of the department of nutritional sciences at Rutgers University, said that although the legal status of the IND regulation for foods is somewhat murky, “I believe that an IND would be required because [alcohol] would be considered a food and the outcome is therapeutic.”
“Any kind of study where you’re trying to claim a clinical benefit, you have to file an IND,” said Akhilesh Nagaich of consulting firm FDA Insight and a former FDA scientist. “My understanding is they should have.”