As Industry Grumbles, CDRH Touts its Review Improvements

While U.S. industry leaders continued to complain about slow FDA device reviews and have been swayed to move a growing share of product development overseas, there’s been a little-acknowledged internal review revolution going on inside CDRH.

That was the message delivered by Center deputy director for science William Maisel to a spring Food and Drug Law Institute update session in Washington, just as lawmakers on Capitol Hill were moving on new user-fee legislation to grant industry additional relief from agency caprices.

Ongoing improvements to CDRH’s premarket review program are paying off, Maisel told the session, describing actions under the ongoing 510(k) revamp that are helping bring greater predictability about data requirements for certain product areas, reducing unnecessary data requests, and implementing new policies allowing more balanced benefit-risk determinations.

“Some of the preliminary data points are starting to point in the right direction,” he said. For example, the 510(k) backlog, which grew from 2005 to 2010, began to drop in 2011 and, in 2012, it continues to go down further.

Although improvements are being seen, Maisel did acknowledge that there is still a lot of work to be done, doubtless raising some skeptical eyebrows among attendees more interested in quick fixes than in small incremental gains over time.

The Center’s Innovation Pathway offers much promise, he said, and “it’s our test bed and our place where we can innovate, try new things and develop new tools. One tool under development is a decision support tool for FDA reviewers to use and evaluate first in human clinical trials coming through the Innovation Pathway. It is a semi-quantitative way to assess decisions and really focus in on the most important issues.”

Maisel said that the decision to allow a first-in-human trial to go forward “is one of the most complicated and challenging ones we make because it is typically based on relatively little information, but we also see the great promise and potential. The tool, which is based on our early feasibility draft guidance will help our reviewers explicitly consider the benefits and risks.”
CDRH is also looking at novel IT tools within the Innovation Pathway, he said, to help promote collaboration during the device evaluation phase. “We have set up a secure online conference center where our device team and industry can work together in a shared space, share documents, exchange communication through that shared space, and have a more collaborative back and forth rather than us throwing something over the wall to industry and industry throwing it back over to us.”

Another area in focus is benefit and risk determinations, Maisel said, and a final guidance issued this year includes a worksheet for reviewers on principal factors to consider when making these decisions.

“We have planned training modules for our own staff,” he said. “We’ll be adding sections within the labeling that summarize the benefit-risk determination, we’ll be making our worksheets available that show what our thinking was, and we’ll be auditing how our review staff and stakeholder like it.”

While the benefit-risk guidance applies to PMAs and de novo devices, CDRH has begun thinking about benefit-risk for postmarket surveillance, Maisel said. “This has been a long standing challenge. In 2012, we plan to produce a vision and comprehension plan that will provide what we suggest as the direction we will be heading. It will be an opportunity to start the dialog and get feedback from stakeholders on what we need or not need to do and other areas where we can move forward.”

What about the Inspection Process?

In reaction to Maisel’s presentation, panel commentator and QualityHub president Timothy Wells said he found it interesting that with all the reinventing occurring at CDRH lately, there is no talk about the inspection process.

“I’m worried,” he said, “that FDA allows a CAPA plus 1, which is a two-day FDA inspection. I’m worried that they do foreign inspections in four days and my clients in the U.S. can have FDA there for 14 weeks. That’s not fair. I know companies that are putting plants overseas to take advantage of the four-day inspection.”

Wells said FDA has had internal discussions about accepting notified bodies’ audits as agency inspections, “and that would be a mistake in my opinion. The notified bodies are the ISO auditors who get paid by the companies and they are very superficial, in my opinion.

“I know FDA is strapped for resources, Wells continued. “I’m just wondering about the resources that they are putting into pre-approval, and how many resources are they putting into the field inspection process? I know we are pulling them in a lot of directions. You may get all these product approvals, but if you don't have good manufacturing practices then that is one of the reasons we have recalls and design issues... And so it is the responsibility of the companies to follow QS and GMPs, but I think it is important also to have good FDA inspections and ask the right questions.”

Additionally, Wells went on, FDA should look into the tools and the training of the investigators. “Some inspections are ‘knock it out of the ballpark’ homeruns and the other half are just leisurely visits,” he told FDLI. “I’m not sure what happened to investigator certification... because having qualified and skilled FDA investigators will benefit everyone and level the playing field.”

Regarding inspections, Maisel responded that “it is simply not possible for us to go out and inspect every manufacturing plant and, with an increasing global economy, we’ve had to strategize about how we are going to use the limited resources we have. We use a risk-based approach to select inspection sites and try to focus on areas where we are most likely to find issues or problems. We have also started pursuing the concept of a single audit program where a single inspection could satisfy multiple regulatory agencies. There would have to be certain circumstances under which those inspections are performed, and it would probably not be available for someone who has a long history of many deficiencies.”

A Business Case for Quality

Maisel also pointed out that, last year, CDRH announced its business case for quality, which is a recognition that slapping someone on the wrist may not be the best approach for getting a quality device. “So we have been meeting with industry stakeholders to try to come up with ways to promote quality systems and practices, which may be separate from compliance actions, because what we really want is quality devices.”

Asked to comment on industry appeals, Maisel told FDLI that there needs to be an avenue to get supervisory input. “We are trying to put more concrete timelines on appeals and we will be coming forward with a guidance document on this,” he said. “The big challenge for companies, I think, is the unpredictability, and the appeals that seem to get lost in a vast morass and a company not having any idea when it will come out on the other side... We end up with somewhere between 30-40% of appeals that are turned over or partially turned over.”

FDLI panel commentator and Hyman, Phelps & McNamara director Jeff Shapiro said that, from industry’s perspective, the appeals process has not been working well and timelines is an issue. The supervisor review process has a problem because there is an inherent conflict-of-interest by having a supervisor reviewing the work of their employees.

“I’ve suggested that there should be an Office of Appeals where it would not be the direct supervisors,” Shapiro said. Another appeals issue is the lack of transparency, he added, and nobody knows how many are pending and how long they take. He explained: “When reforms are introduced, we won’t know if there are improvements unless metrics are developed.”