By: Jeffrey N. Gibbs

One category of risks—regulatory risks—is well-known and can readily be characterized. The U.S. Securities and Exchange Commission filings for IVD companies routinely spell out their regulatory risks. When risks are known and understood, they can often be mitigated.

 

Uncertainty is different. As described by Frank Knight in Risk, Uncertainty and Profit, “a measurable uncertainty, or risk proper … is so far different from an unmeasurable one that it is not in effect an uncertainty at all.” By definition, risk can be more easily controlled and managed than uncertainty.

 

IVD companies are now experiencing increasing regulatory uncertainty, particularly with product reviews. These are not isolated events or limited to applications by inexperienced regulatory personnel. Companies are reporting that 510(k) submissions are being delayed or rejected for unexpected reasons. Pre-IDEs are eliciting unforeseen questions and concerns.

 

For example, companies that have submitted multiple applications for a product type are finding, without prior warning, that the data previously deemed adequate no longer suffice. One IVD manager recently found that a 510(k) submission was held to higher data requirements than the 510(k) for the predicate device—a product he had developed just a few years earlier for another company—without OIVD publicly disclosing any review criteria changes. Protocols that had not been challenged during pre-IDE meetings are now considered deficient during product review. The role of research-use-only products in 510(k)s was altered without public notice, surprising multiple applicants. 

 

According to some OIVD reviewers, the latest criteria for 510(k)s is found not in guidelines, or even in draft guidelines, but in review memoranda of other companies’ 510(k) submissions. Developing a product, generating the data, and submitting an application require a long lead time. Review memoranda are informal, ad hoc, and application specific. While they may be instructive, citing these documents to communicate requirements introduces uncertainty.

 

FDA unquestionably needs the ability to modify data requirements. When problems emerge, and as technology and clinical practice change, data requirements can and should evolve. Policy modifications will also be needed. Yet many data changes confronting companies are divorced from specific external events. Absent new circumstances necessitating changed review criteria, establishing new requirements introduces significant uncertainty. To many companies, they appear almost to introduce an element of randomness. These concerns are further exacerbated by the prospect of broader changes by CDRH in the 510(k) process.

 

Increased regulatory uncertainty is not benign. It can delay or prevent useful products from being commercialized. It increases the costs of introducing new products. It inhibits investment. Investors can tolerate foreseeable risks, including costly but predictable clinical requirements, more easily than they can tolerate uncertainties. Uncertainty can inhibit innovation. As one senior IVD executive whose company encountered an unexpected stumbling block told me, this uncertainty “creates an environment which deters industry from creating innovation.” She added that the company is unable to introduce new products, even though they are better than their current products.

 

Other effects may be more subtle. Spending the time, money, and resources to develop a new product is a calculated gamble. If regulatory uncertainty becomes too great, companies may conclude that they can’t meaningfully assess the odds of success, and that may affect the resources allocated to introducing new assays in the United States.

 

What is the solution? No single step would eliminate this problem. Better, more open communications between OIVD and industry is a key ingredient. Industry needs to listen better when OIVD expresses new concerns or questions. Industry should be sensitive to new developments that could affect data requirements about particular products, such as journal articles and presentations at medical meetings. 

 

FDA’s transparency initiatives could help, as could better adherence to Good Guidance Practices. So could the recognition by reviewers that regulatory predictability is itself benecicial. When long-standing data requirements are changed, OIVD should provide notice and, preferably, a rationale. This would help companies understand the changes and provide an appropriate response—and even, perhaps, anticipate future shifts.