Major gaps in oversight of human medical research

Introduction

Ten years ago, an 18-year-old student named Jesse Gelsinger died during a medical research trial involving gene therapy at the University of Pennsylvania. The teenager’s death, caused by a violent immune reaction to an injection, triggered a federal inquiry, a review of human protection standards in clinical trials and, ultimately, tougher national standards for patient consent.

Now details have surfaced about a clinical trial at Columbia University Medical Center that took place not long after the Gelsinger case, and which raises questions about the regulatory safety net that was the legacy of the teen’s death.

Federal regulators have cited Columbia for ethical and regulatory mistakes in a clinical trial, completed in 2001, in which more than 200 patients agreed to receive one of four federally approved surgical fluids during open-heart operations. Regulators now have concluded that some patients may have been harmed and have ordered Columbia to track down all participants to inform them of the facts of the study and their surgeries, as the Huffington Post Investigative Fund reported exclusively earlier this month. The hospital has acknowledged flaws in the study but has said there is no evidence those led to the patients’ medical problems.

If the Gelsinger case was a pivotal moment in reform, the Columbia case is a cautionary tale, coming to light months after a sobering national review of clinical trials by independent bioethics research institute, the Hastings Center. In its 2008 review, the Hasting Center concluded the human subjects are no better protected today than they were at the time of Gelsinger’s death.

The federal order that directed Columbia to find its heart patients and tell them more about their drug study illustrates the continuing weakness in safeguards of clinical trials and specifically the procedures for obtaining informed consent—the written acknowledgement by a patient that he or she understands the risks of participation in a clinical trial.

Collaboration between industry and academic institutions to test drugs and medical devices has increased at such a rate in the past decade that oversight panels are hard pressed to fully review all such trials. The gaps mean that a medical product may receive federal approval and be marketed without regulators or doctors being aware of limitations and biases in its testing history.

In the case of Columbia’s study, research that was barred from publication by university officials was nonetheless submitted by the manufacturer to an advisory panel hearing of the U.S. Food and Drug Administration as evidence that its product was safer than a competitor’s.

Vulnerabilities in the oversight system have registered in Washington. Jerry Menikoff, a bioethicist and well-known critic of the research oversight process, was appointed in 2008 by the Bush administration to head the federal Office of Human Research Protections. That office was created in 2001, within the Department of Health and Human Services, in the wake of Gelsinger’s death. [Ed. note: Date of Menikoff’s appointment corrected Oct. 29, 2009.]

Menikoff, who is also a physician, has argued publicly that patients are often not fully informed or protected when they sign consent forms. “There is good reason to believe that in many thousands of cases, research subjects are not being given the information they most need,” Menikoff wrote in “What the Doctor Didn’t Say,” a book he co-authored in 2006.

“Too much of our current system has contained remnants of a disturbing notion that to have the appropriate amount of research, we need to deceive many patients when they are being asked to participate in clinical trials,” Menikoff wrote. All too often, in his view, patients are led to believe they will benefit directly from participating.

An estimated 300,000 studies involving up to 7 million human subjects are conducted each year in the U.S. No government agency or private entity tracks those numbers closely, according to CenterWatch, a Boston-based service that maintains one of the largest databases of clinical trials.

Medical research involving patients is subject to oversight by institutional review boards, or IRBs. There are nearly 4,000 of these boards in the U.S. Most rely on already overscheduled staff at medical centers and research hospitals; some are independent commercial outfits. Each board is responsible for reviewing anywhere from dozens to hundreds of research proposals a year.

Board members are often volunteers with full academic and clinical duties at their hospital or research center. Many describe themselves as hard pressed to keep up with all relevant studies and often rely on the scientific integrity of researchers who propose the clinical trials, according to experts in the medical research field.

“Patient protection is run on an honor system,” said Robin Wilson, a law professor at Washington and Lee University. The federal Office of Human Research Protections assumes, unless a complaint is lodged, that hospital boards are vigilant about oversight in clinical trials. The boards, in turn, largely trust researchers to ensure patient safety.

Greg Koski, a former head of the Office of Human Research Protections, told the Investigative Fund that he believes it is impossible to know if institutional review boards fully understand the proposed studies and are adequately protecting patients.

“Most IRBs operate behind closed doors…in part to protect intellectual property or commercial information,” said Koski, an anesthesiologist at Massachusetts General Hospital. Koski now advocates that researchers be trained and certified before they design clinical trials. He also wants to see uniform requirements for all commercial and federally-funded studies and better data collection on the performance of institutional review boards.

In the last nine years, the federal office has cited more than 40 medical centers for lapses ranging from misinforming patients about potential risks to failing to report serious complications.

Some of the cases involved studies that were funded in part by pharmaceutical, biotech and medical device industries. Such collaborations are mutually beneficial. Companies seek out researchers at well-known academic centers who have access to patients willing to participate in trials. Universities find that industry-driven clinical trials enhance their reputations and financial bottom lines.

As collaborations have increased, the clinical trial divisions at many university hospitals have been reorganized in ways that resemble the structures of commercial drug companies, said Jennifer Washburn, author of “University Inc.: The Corporate Corruption of Higher Education.” “They now function far more like for-profit research organizations whose purpose is to serve their clients,” Washburn said in an interview.

Washburn points out that members of review boards “may be reluctant to interfere or impede the work of their colleagues…especially when those colleagues are powerful faculty who bring in substantial revenue into the medical school.”

The weaknesses in the oversight system can have consequences that reach beyond the immediate research. If studies are carried out with flaws in design or procedures, the results can affect many more patients than just those who participated in the trial.

In the case of the Columbia trial, Abbott Laboratories contributed $150,000 to the research; a blood expander that Abbott manufactures was one of four used on open-heart patients at the hospital from 1999 to 2001.

Patients signed consent forms but federal regulators have now determined that the patients were unaware of the “true nature” of the study.

Testing the safety of the fluids at high doses was one purpose of the study, Columbia investigators later concluded. The hospital’s internal reviews found that the consent form failed to inform patients that hetastarch — a substance used in two of the blood expanders in the study, including the one sold by Abbott – could prevent blood from clotting properly, especially when used at higher doses. About half of the 215 people in the study were given hetastarch, and some received more than three times the upper limit recommended by the manufacturers, according to documents later filed in court.

After conducting an internal investigation – one of three the hospital would launch about the drug trial – Columbia in 2002 communicated its concerns to federal regulators and barred from publication any scientific findings from the clinical trial.

The findings never were published. However, records show, Abbott Laboratories in 2002 provided data from the Columbia study, along with other unpublished studies, to an FDA advisory board deciding what kind of safety labels to put on the Abbott product, Hextend, and that of its main competitor, Hespan, manufactured by B. Braun Medical Inc. Abbott was lobbying the FDA to have its product exempted from a requirement that it contain a warning about bleeding complications during cardiac surgery.

According to FDA records, Abbott used the unpublished studies to raise questions about “excessive bleeding during cardiac surgery” by its competition.

Gary Haynes, chair of the department of anesthesia and critical care at St. Louis University School of Medicine, was among those who provided expert testimony to the FDA panel about commercial blood expanders. Haynes, interviewed for this story, recalls that the unpublished study from Columbia was central to Abbott’s argument that its product was safer for heart surgery than its competitor’s.

The FDA decided in Abbott’s favor, requiring its competitor to include the additional warning about bleeding risks. Abbott’s product was sold without the warning about heart surgery.

An Abbott spokesman told the Investigative Fund that the company has no record of how it obtained the data from the Columbia study. The FDA did not respond to questions about the matter.

After the FDA’s decision, Abbott paid for full-page advertisements in at least one medical journal, promoting its product for use “even at higher volumes.” Hospitals around the country subsequently included the Abbott’s blood expander on their list of preferred drugs. The U.S. military, engaged in two wars for much of this decade, names Abbott’s Hextend in its Army guide for medical personnel.

Yet the claims that the Abbott product is safer than its competitors have been called into question by subsequent reviews in the medical literature.

In 2003, a published review of 40 studies of intravenous fluids in the journal Anesthesia & Analgesia found that virtually all the studies of blood expanders were poorly conducted or reported. The author concluded that the use of blood expanders by anesthesiologists and other physicians was being guided by “dogma and personal beliefs” rather than sound science.

In 2005, another study published in the Journal of Cardiothoracic and Vascular Anesthesia found that both Abbott’s Hextend, and its competition, Hespan, had the same anti-clotting effect.

A third study, cited in the journal Chest in March 2009, had to be halted early because of serious bleeding among heart surgery patients. Those patients were given just one-fifth the average dose of the Abbott drug that patients received in the Columbia study.

Jeanne Lenzer is an independent medical investigative journalist and a frequent contributor to the BMJ (formerly the British Medical Journal). She can be reached at jeanne.lenzer@gmail.com. Shannon Brownlee is a senior research fellow at the New America Foundation and the author of Overtreated: Why Too Much Medicine is Making Us Sicker and Poorer. She can be reached at shannon.brownlee@comcast.net.