NIH pulls the plug on controversial alcohol trial

The National Institutes of Health (NIH) in Bethesda, Maryland, today killed a controversial clinical study that was already on life support: the Moderate Alcohol and Cardiovascular Health (MACH) trial.

A working group of NIH advisers assembled to review the study found that senior officials at the National Institute on Alcohol Abuse and Alcoholism (NIAAA) actively and secretively courted the alcohol industry to fund the $100 million project, and saw to it that a favored principal investigator (PI) won the funding. NIH Director Francis Collins this morning called the conduct “way outside of the acceptable culture of our noble institution” and, following the working group’s recommendation, ordered the study shut down “as quickly as that can be done.”

“Many of the [NIH staff] who have seen the working group report were frankly shocked to see that so many lines were crossed,” Collins said in describing the findings, which were released today. For instance, he said, it was clear that backers of the study had manipulated the research plan to ensure an outcome favorable to the sponsors, which included five companies—Anheuser-Busch InBev, Carlsberg Group, Diageo, Heineken, and Pernod Ricard—and had hidden their activities from other NIH staff, behavior he called “a flashing neon light.”

The 10-year trial aimed to enroll 7800 participants, with the active arm consuming one serving of alcohol each day and the control group none, to measure whether moderate alcohol consumption prevents the development of cardiovascular disease and diabetes, as has been suggested by earlier studies.

But the working group enlisted the opinions of epidemiologists and concluded that the study proponents appeared “to intentionally bias the framing of the scientific premise in the direction of demonstrating a beneficial health effect of moderate alcohol consumption.” The epidemiologists said the study did not plan to enroll enough patients or allow enough follow-up time to assess whether moderate alcohol consumption also increases cancer risks. They concluded that “the trial could show benefits while missing the harms.”

The huge study was to have been funded mostly by the alcohol industry through donations to the Foundation for the National Institutes of Health (FNIH), a nonprofit established by Congress that is meant to provide a firewall between donors and the studies they fund, preventing them from influencing trial designs to promote particular outcomes. The working group found that senior NIAAA officials kept FNIH officials in the dark about their courting of alcohol companies. It also concluded that the officials ensured an investigator who helped them woo the industry—Kenneth Mukamal of Harvard Medical School in Boston—would become the trial’s PI.

Mukamal and three members of NIAAA leadership had “sustained interactions” from at least 2013, prior to and during development of the Funding Opportunity Announcements for both the planning and main grants to fund the MACH trial, said NIH Principal Deputy Director Lawrence Tabak, who led the working group. “These interactions appear to have provided the eventual PI with a competitive advantage not available to any other applicant and effectively this steered funding to this investigator.”

Collins praised The New York Times for revealing the actions of the alcohol institute officials in an article published in March, when, Collins said, he realized “this was a matter that needed really serious consideration.” Last month, Collins suspended enrollment in the trial pending the findings of the working group. At the time the trial was suspended, 105 people were enrolled. “Those are 105 people to whom we owe all kinds of information,” Collins said today.

Before killing the study, Collins invited the opinion of George Koob, the NIAAA director who took the reins of the institute in January 2014. At that point, senior NIAAA officials had begun courting the industry to fund the study, but they hadn’t solicited grant applications for executing the trial or engaged FNIH as a vehicle for industry contributions.

A strained-looking Koob told Collins and his advisory panel today: “I understand and agree with the significant concerns about the MACH study and its ultimate credibility. There were design issues … [and] also significant process irregularities … that undermined the integrity of the research process. We do not see a truly competitive competition [for the award]. The trial is irrevocably damaged and I don’t think we can justify continuing the study.”

NIH is now undertaking a 60-day study to try to locate any other agency-funded projects that may have been compromised by donor influence or improper inside tracks for favored investigators. (In the case of the MACH trial, Mukamal was the only applicant for awards NIH solicited for both planning and execution phases of the study.)

“We will see what we get from the new investigation and come up with an appropriate plan for interventions,” Collins said. “My expectation is that the MACH study is an unusual outlier, but I want to be sure I’m right. And this is the only way to find out.”

For some, NIH’s internal investigation does not go far enough. “The shocking allegations” about the “aggressive campaign” by NIH officials to solicit industry funding require an independent investigation, says Michael Carome, director of the health research group at Public Citizen, a watchdog organization in Washington, D.C. In April, Public Citizen and 13 other groups asked Alex Azar, secretary of the Department of Health and Human Services (HHS), NIH’s parent department, to request an investigation by the HHS inspector general.

*Correction, 18 June, 9:45 a.m.: An earlier version of this story stated that NIAAA officials had not solicited grant applications related to the MACH study before Koob, the current NIAAA director, took the helm in January 2014. In fact, NIAAA solicited applications for planning grants, to allow investigators to begin laying the groundwork for the trial, in July 2013, months before Koob arrived at NIAAA.