Lack of trial diversity: Is it really an unavoidable reality?

Diversity in clinical trials is a common refrain at advisory committee meetings these days, with panelists often recommending postmarket trials or subgroup analyses for minorities or patients with specific co-morbidities.

But that could be too little too late for some patients, Anna Mazzucco, of the National Research Center for Women and Families, reminded the Anti-infective Drugs Advisory Committee (AIDAC) at a March 31, 2014, meeting on two antibiotics intended to treat acute bacterial skin and skin structure infections caused by gram-positive bacteria, including methicillin-resistant Staphylococcus aureus (MRSA).

Research shows that sponsors take about 10 years to complete and analyze a postmarket study – if they complete it. That’s a long time to wait for better data when dealing with a product that could have serious risks for specific populations, Mazzucco said during the adcom public hearing. Meanwhile, the drug or device is still on the market, being used in patient groups in which its benefit-risk profile isn’t fully known.

Instead of relying on postmarket studies to fill in the knowledge gaps, Mazzucco urged the FDA to require sponsors to test their therapies in a more diverse population to begin with.

The March meeting, in which AIDAC considered the data for Cubist Pharmaceuticals Inc.’s Sivextro (tedizolid phosphate) and Durata Therapeutics Inc.’s Dalvance (dalbavancin), provides two case studies of how adcoms and the FDA weigh trial diversity. While panelists do raise diversity issues at committee meetings, in both of these cases, concerns about the lack of diversity were first raised during the public comment period or by the patient representative.

For instance, Mazzucco pointed out that the phase III trials for both antibiotics enrolled few patients older than 65. Yet elderly patients, who generally are hospitalized more frequently than younger adults, are more susceptible to MRSA. In general, patients older than 65 had poorer outcomes in the trials, Mazzucco noted. (See BioWorld Today, March 31, 2014.)

One of Cubist’s phase III trials for Sivextro was open to patients 18 or older, but the mean age was 43. Hoping to get a pediatric indication for the drug, Cubist opened the second phase III study to patients who were 12 or older, but only two subjects in the 12-18 age range enrolled. They were both 17.

Together, the two Sivextro trials enrolled a total of 1,324 patients; 69 of those patients, or about 5 percent, were 65 or older. The numbers were even smaller when the oldest patients were singled out. Only about 1.5 percent, or 18 subjects, were 75 or older. None of the AIDAC members expressed concerns about the lack of older subjects in the trial, but several of them recommended the FDA not approve Sivextro for people younger than 18 until it is tested in a broader pediatric population.

Mazzucco also pointed out the lack of ethnic diversity in the phase III trials for Sivextro. “While studies have shown that Hispanic and African-American patients are more likely to contract MRSA, 85 percent of the patients in all the phase III trials were white. With such a low percentage, any safety or efficacy problems specific to minority patients, potentially because of different metabolism of the drug, would be invisible” in the much larger sample of Caucasian patients, she said.

In the two pivotal trials combined, 77 subjects – or about 6 percent – were African-American and only about 1 percent represented other racial groups, Mazzucco claimed. About 80 percent of the patients in the first trial were enrolled in North America, while 47 percent in the second trial were enrolled at North American sites. Although the majority of the subjects in the trials were white males, gender was not mentioned as a cause of concern.

REFLECTIVE OF PATIENT POPULATION

Later that day, the committee’s patient representative, Cynthia Cauhan, brought up diversity again when AIDAC discussed Durata’s Dalvance. “Does the population that was studied reflect the population that is affected?” she asked the FDA reviewers.

Dmitri Iarikov, an FDA medical officer, said that in clinical practice, elderly patients would be more likely to develop MRSA-related acute skin infections than patients younger than 65. Although the Dalvance trials were open to adults between the ages of 18-85, the majority of the subjects were younger, with a mean age of about 49. The FDA briefing document showed that 16 percent of the participants were 65 or older, but it didn’t break the age group down further.

Noting that 60 percent of the subjects were male and about 92 percent were Caucasian, Iarikov acknowledged that the percentage of minorities in the Dalvance trials “probably” didn’t reflect the U.S. population, because about 60 percent of the patients were enrolled in Eastern Europe. As a result, the trial demographics had certain limitations when compared with the U.S. population, he said, “but I would not characterize them as great limitations.”

Iarikov’s FDA colleague Ed Cox disagreed. “It seems like we could do a better job here with greater diversity as far as the racial and ethnic groups that are involved in the clinical trials. I think that’s something to work on,” he said. While he stressed the need to select study sites that would provide a greater degree of racial and ethnic diversity, he was silent on age diversity.

Responding to Cox, Mads Rasmussen, a corporate project vice president at Novo Nordisk Inc. and the industry rep for the adcom, said, “I’d just like to remind the panel that most sponsors have to develop drugs for global use in order to justify the cost of development. And there is no reasonable expectation to make sure that the global representation is equal to the American population.”

Instead, he said, the focus should be on making sure the population recruited in the U.S. is reflective of the U.S. population at risk.

Some countries avoid this problem by requiring trials to be conducted in the domestic market. China, for example, requires all clinical trials used to support approval to be conducted in China. And sponsors of new drugs for the Japanese market must conduct clinical trials in the Japanese population, although biosimilars can be tested in other Asian populations.

In the U.S., trials are supposed to be representative of the intended population. While the FDA requires sponsors to analyze clinical data by age, gender and ethnicity, it doesn’t actually require diversity. Thus, if a product were approved only for the population it was adequately tested in, most drugs and devices approved in the U.S. would be indicated only for white men younger than 65.

By not insisting that sponsors evaluate drugs and devices in women, ethnic groups or the elderly as part of the approval process, the FDA is sending the message that a product’s efficacy and safety in those groups doesn’t matter, several speakers said at a hearing held earlier this year to discuss ways to improve diversity in trials. The hearing was a first step in the FDA’s development of a plan – mandated by the FDA Safety and Innovation Act – to improve the quality and availability of data by demographic subgroups. (SeeBioWorld Today, April 2, 2014.)

The speakers recommended a number of ways to make trials more reflective of the real-world patient population. At the top of the list was enforcing current requirements for making trial populations representative. The FDA also was encouraged to ban ageism in trials, eliminate unnecessary exclusion criteria, require sponsors to develop a plan for inclusion, strengthen regulations on subgroup analysis, require trial data to be broken out by ethnicity rather than lumping all the minority data together, and include trial data based on age, gender and ethnicity on product labeling.

Source