Advocates Call for FDA to Take Tougher Stance on Postmarket Safety Studies

With high dropout rates threatening the viability of 2 postmarket safety studies of silicone breast implants, the US Food and Drug Administration (FDA) in August convened an advisory committee hearing to determine the best way to assess these devices in the absence of results from these studies. The agency required manufacturers to conduct such research as concerns about the safety of these devices emerged.

The meeting was closely watched by consumer advocates and clinical experts, who say it highlights the ongoing difficulties the agency has had in ensuring that device makers follow through on required postmarket studies of medical devices. Even though such research is frequently required by the agency as a condition of approval, companies often fail to follow through. Critics argue that tougher sanctions from the agency or greater collaboration among government agencies to ensure collection of the necessary data may help resolve this problem.

Silicone breast implants have long been shadowed by safety concerns, including worries that breakage of the devices and leakage of silicone into the body could lead to systemic disease. Such concerns led to a 14-year moratorium on the sale of the devices that ended in 2006, when the FDA approved marketing of the products by 2 manufacturers, Allergan and Mentor, on the condition that both companies conduct several postmarket studies of implant safety.

Since then, more data have been collected on implant safety. In a report issued in June, the FDA summarized the state of the evidence, including some preliminary results from required postmarket studies (http://tinyurl.com/3hqbnkn). The report noted frequent local complications and adverse events such as capsular contracture, implant rupture, wrinkling, asymmetry, scarring, pain, and infection. Implant removal rates are also high: 20% to 40% of patients who received them for augmentation purposes had their implants removed within 8 to 10 years, and 40% to 70% of breast reconstruction patients had them removed during that same time frame. The agency also noted a small increased risk of anaplastic large cell lymphoma.

“The longer a woman has breast implants, the more likely she is to experience local complications or adverse outcomes,” the report noted. “Women with breast implants will need to monitor their breasts for local complications for the rest of their lives.”

The agency concluded in the report that there is enough information about the risks and benefits of these devices to allow women to make an informed decision about them. However, efforts to gain a fuller understanding of long-term adverse effects that was expected to emerge from the large, long-term postmarket studies required by the agency have foundered. The trials now appear unlikely to be sufficiently powered to assess rare outcomes.

The agency required each company to follow at least 40 000 women for 10 years to provide data about possible rare adverse events such as autoimmune diseases or events that develop over the life of the implant. But just 3 years into enrollment, both companies are failing to meet targets for 1-year follow-up of patients with implants. The goal was 96.5% follow-up at this point, but so far, Allergan has achieved a follow-up rate of only 60.2% of patients receiving its silicone implants, and Mentor has followed up only 21.8% of patients who received its implants. At these rates, the FDA concluded, the studies will not be sufficiently powered to answer the questions they were intended to address, so the agency turned to its advisory committee for advice.

“It's a disaster in terms of patients ever getting the data they were promised,” said Diane Zuckerman, PhD, president of the National Research Center for Women & Families.

Zuckerman said she was disappointed that the agency didn't ask its advisory committee about the possibility of rescinding approval of the devices because companies have failed to comply with the postmarket study requirements so far.

“These required studies aren't really required,” she said. “[The companies] are required to start them but not required to finish them in a way that provides meaningful data.”

Rita Redberg, MD, editor of the Archives of Internal Medicine and a cardiologist at the University of California–San Francisco, agreed that the FDA must insist that postmarket device safety studies are completed to ensure that adequate safety data are available about a device. She emphasized that in the case of implants, the inability of the companies to carry out such studies is a “patient safety issue.”

“If the FDA requires postmarket studies as a condition of approval and they are not done, approval should be revoked,” Redberg said. However, she noted that despite the failure of many device companies to follow through on required studies, the FDA had yet—as of late September 2011—to revoke the approval of a device for this reason. And without such clear penalties for failure to complete postmarket studies, Redberg said, companies have little incentive to conduct research that may identify safety issues with their products and consequently reduce sales.

At the meeting, the FDA's advisory committee recommended that registries be created of all women who have received silicone implants to collect sufficient data on rare events. It also said that the FDA should work with the companies to improve patient participation in the existing studies.

“The FDA is committed to seeing [the long-term postmarketing silicone breast implant studies] completed and making sure follow-up rates improve,” said Erica Jefferson, a spokeswoman for the FDA in an e-mail response. She explained that the agency will work with the companies and consumer groups on developing incentives for patients and their physicians to participate, on finding ways to streamline the questionnaires, and on exploring alternative ways for patients to report adverse event information, such as through social media or the FDA's MedWatch program.

But some observers say the FDA should also consider other potential models to ensure that postmarket safety data are collected.

Joseph P. Broderick, MD, a neurologist at the University of Cincinnati, argues that the FDA should collaborate more with other federal agencies to ensure that postmarket safety data are collected. The recently published results of the Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (SAMMPRIS) trial demonstrated how effective such a collaboration could be, he wrote in an editorial that accompanied the results (Chimowitz MI et al. N Engl J Med. 2011;365[11]:993-1003 and Broderick JP. N Engl J Med. 2011;365[11]:1054-1055).

The study, a randomized controlled trial in patients with intracranial arterial stenosis, compared 2 interventions intended to reduce stroke risk: aggressive medical management or aggressive medical management plus a stent. The device studied in the SAMMPRIS trial was the Wingspan stent system, which was approved by the FDA under its Humanitarian Device Exemption program in 2005. The Centers for Medicare & Medicaid Services (CMS) chose to provide coverage for use of the device only in patients who were participating in the SAMMPRIS trial. According to Karen Riley, a spokeswoman for the FDA, the CMS does not typically offer blanket coverage for devices approved under this program. The trial's data and safety monitoring board stopped the study in April, after 451 patients were enrolled because the rate of stroke or death within 30 days was 14.7% in those receiving a stent plus medical management vs 5.8% in those who received medical management alone.

Broderick said the trial allowed “us as clinical scientists to learn what is the best role for this device at this time” and may have prevented more widespread use of the device in patients who would not have benefited from it. He acknowledged that there may be barriers to CMS and FDA coordinating efforts, however, because the 2 agencies are governed by different laws.

Redberg agreed that the SAMMPRIS trial is a useful model for ensuring that postmarket data are collected and should be used more often because it allows the CMS to use its funds more effectively and ensures that adequate data are available to guide clinical practice. “It's a win-win,” she said. “CMS is often forced to pay for new devices before there are adequate data.”

For example, she noted that the CMS previously used this strategy successfully to assess the effectiveness of lung-reduction surgery in patients with emphysema. The costly technique had been widely used, but as questions arose about the intervention, the CMS chose to restrict coverage of the procedure to those patients participating in a large multicenter trial. The trial demonstrated that for certain patients the procedure was associated with increased mortality and did not improve lung function (National Emphysema Treatment Trial Research Group. N Engl J Med. 2003;348[21]:2059-2073). Without such a restriction, it can be difficult to conduct needed clinical trials once a device becomes widely used.

Redberg noted that registries will also remain an essential part of collecting postmarket data on devices. In the case of silicone breast implants, it may be the only way to collect sufficient data to assess rare adverse events.

But Broderick said that more strategies need to be used that help meet the needs of the various stakeholders involved. He explained that the US health care system needs to get better value for money and to stop paying for things that aren't helpful, patients need better outcomes, and companies need to profit.

“The resources paying for clinical care can be set aside for trials that are going to answer questions and improve care in the process,” Broderick said.