FDA TOO QUICK IN APPROVALS

The quality of data used by the US Food and Drug Administration (FDA) for accelerated drug approvals and approvals of high-risk medical device modifications is relatively weak and raises questions about the rigor of the regulatory process, according to two studies published online August 15 in JAMA.

In one of the studies, Sarah Y. Zheng, MD, from the department of psychiatry at the University of California at San Francisco, and colleagues analyzed 83 studies that supported approvals for 78 modifications of high-risk devices between 2006 and 2015. High-risk devices are defined as those that support or sustain human life, such as coronary stents and hip prostheses.

"[F]ewer than half were randomized, blinded, or controlled, and most primary outcomes were based on surrogate end points," they write. Specifically, just 45% of the studies were randomized and only 30% were blinded.

Moreover, 71 (91%) approvals were supported by a single study.

Researchers found that the median number of patients per study was 185 and median follow-up was 6 months.

Of the primary end points, 121 of 150 (81%) were surrogates, defined as "a laboratory measurement or a physical sign used as a substitute for a clinically meaningful end point that measures directly how a patient feels, functions or survives."

While such end points allow for smaller and less costly trials, "surrogate measures can lead to uncertainty about clinical outcomes," the researchers note.

Remarkably, the device modifications the researchers studied were the ones expected to significantly change design or performance or indication for the device, and require the most rigorous approval process. Thus, even though the data uncovered in the current study are less than ideal, other levels of approval for device modifications would require even less rigorous evidence.

The researchers explain that modifying high-risk devices can have a direct effect on millions of people.

They use the example of the gastric Lap-Band, whose indications were expanded when the FDA approved the modification based on a single-group study of 160 people without controls.

Lap-Band's expanded indication meant that 19 million more people were able to have the band placed.

However, the authors report, "18.5% of Medicare beneficiaries who have received this device have undergone reoperation by 4.5 years, with an average of 3.8 procedures per patient."

Data for Accelerated Approvals

In the second study, Huseyin Naci, PhD, MHS, from LSE Health, department of health policy, London School of Economics and Political Science in London, United Kingdom, and colleagues find similar problems in the data used to support accelerated drug approvals.

Drugs to treat serious or life-threatening conditions can get fast-tracked based on surrogate end points that "are only reasonably likely to predict clinical benefit." The results then need to be confirmed in trials to see if they really lead to improvements.

The researchers found that between 2009 and 2014, 22 drugs received 24 indications via accelerated approval based on 30 studies. Among the studies, just 12 (40%) were randomized and 6 (20%) were double-blind. Moreover, 14 indications were initially approved after single-arm trials that lacked comparator groups, and the mean number of patients in all of the preapproval trials was just 132.

In addition, postmarketing studies were slow to come. "Among 22 drugs with 24 indications granted accelerated approval by the FDA in 2009-2013, efficacy was often confirmed in postapproval trials a minimum of 3 years after approval," the authors write.

The researchers also found that in many cases design of the pre- and postmarketing studies was similar and many of the postmarketing trials also used surrogate end points instead of clinical end points.

"Sweeping Overhaul Needed"

Robert Califf, MD, former FDA commissioner and currently a cardiologist with Duke Health and Duke University School of Medicine in Durham,

North Carolina, writes in an accompanying editorial, that he agrees there are systemic problems that need to be addressed.

"A sweeping overhaul of the overall system is needed," he writes. "Fortunately, such a transformation is under way. Regardless of the type of medical product, substantial progress in balancing safety with access to effective therapies will come from systemic changes in the ecosystem rather than incremental modifications made by imposing more severe demands on individual products."

However, there are reasons behind the lack of blinding and control groups, Dr Califf explained.

When patients have life-threatening diseases without effective therapies, patients and their physicians are understandably hesitant to use a control group or placebo, he says. Similarly, blinding in trials for medical devices might be either impractical or expose patients to risk with no possible benefit.

In a related podcast, Dr Califf says that when there is a life-threatening disease with no effective treatment, "patient groups have been very clear that they are willing to take a high degree of risk to have earlier access."

He adds, "The goal is to keep the degree of regulation in sync with some benefits that might be anticipated."

As far as the small number of randomized, controlled clinical trials, he says that harnessing the power of electronic medical records to conduct such trials in a nontraditional, inexpensive way will help enable more randomized trials and help with continual surveillance of efficacy.

He said that despite the issues raised by the new studies, "[I]t is important to remember that decisions about postmarket requirements and monitoring of these studies are overseen by full-time FDA employees with no financial conflicts. These civil servants are charged with considering the complex issues involved when a presumably effective therapy is on the market but additional evidence is needed."

Physician Finds Editorial Disconcerting

It's that assertion in the editorial that most troubles Vinay K. Prasad, MD,  MPH, a hematologist-oncologist and assistant professor of medicine at Oregon Health and Sciences University in Portland.

"Both of these studies continue to highlight deficiencies in the US Food and Drug administration's permissive approval and poor postmarketing enforcement," he told Medscape Medical News. 

"Dr Califf tries to reassure us that the FDA's employees have no financial conflicts, and, thus, the implication seems they are already doing the best job in balancing speed and safety. However, the fact remains these articles suggest we can and should be doing a better job in that balance," he continues.

Dr Prasad points to a study he coauthored that showed that 60% of FDA reviewers later go to work for pharma. [S]o there may be a bias for reviewers to act in a way that is favorable to industry in order to secure future employment if they were ever to leave."

Dr Prasad said that in walking the line between making products available quickly and ensuring safety and efficacy, the FDA has tipped toward making things available.

"Dr Califf points out that this move is acclaimed by patients and their families, but this is wrong," Dr Prasad says. "First, the patient voice in the drug approval space is dominated by patient advocacy groups that are funded by the industry. Whether these groups speak for all patients is uncertain and unlikely. Second, the role of a federal agency like the FDA is to protect people from choices that don't offer net benefit. If it were just about choices, why have the FDA at all? It is about having good choices, and these articles highlight ways in which that falls short."

Dr Naci is supported by the Higher Education Funding Council of England. A coauthor reports receiving unrelated grants from the FDA Office of Generic Drugs and Division of Health Communication and support from the Engelberg