FDA Finalizes Product Classification Guidance

October 3, 2017

On September 26, 2017, FDA combined and finalized two Draft Guidance documents first issued in 2011 that set forth the Agency’s approach to classifying products as “drugs” or “devices” under the FDC Act § 201. In doing so, FDA trimmed significant portions of both Draft Guidance documents ostensibly “for clarity and ease of reference.” See Classification of Products as Drugs and Devices and Additional Product Classification Issues; Guidance for Industry and Food and Drug Administration Staff; Availability, 82 Fed. Reg. 44,802, 44,803 (Sept. 26, 2017).  However, while the Final Guidance does provide helpful clarification and examples, some of the most significant issues with the Draft Guidance documents persist.

The Draft Guidance documents issued in 2011 (which can be found here and here) were titled “Classification of Products as Drugs and Devices & Additional Product Classification Issues,” and “Interpretation of the Term ‘Chemical Action’ in the Definition of Device under Section 201(h) of the Federal Food, Drug, and Cosmetic Act” (hereinafter “Draft Classification Guidance” and “Draft Chemical Action Guidance,” respectively).  These documents were met in 2011 with numerous objections from the medical device industry, because the Agency’s position significantly diminished the scope of the “device” definition.

Specifically, FDA’s Draft Classification Guidance stated that (1) any therapeutic effect by a product would be considered a “primary intended purpose” pursuant to FDC Act § 201(h), and (2) if a product that depends “even in part” on chemical action within or on the body to achieve a primary intended purpose, it is excluded from the device definition. See Draft Classification Guidance at 5.  The Agency further clarified that no chemical action with therapeutic effect would be permitted in a device, noting that “a product that exhibits chemical action within or on the body of man” could remain within the device definition only if its chemical action contributes to something other than a primary intended purpose (i.e., therapeutic effect). Id. at 4-5.

Moreover, the Draft Classification Guidance acknowledged that the new positions set out in both Draft Guidance documents may result in the FDA Office of Combination Products (OCP) changing its view of products that the Agency had already classified, stating that OCP “may determine that, in light of current scientific understanding, the means by which such a [previously classified product] or constituent part achieves an intended use may warrant a different classification for that product or constituent part in the pending RFD than the Agency previously provided.” Id. at 6.

FDA was challenged in litigation over the Agency’s reliance on the extra-statutory “even in part” standard not found in the FDC Act’s definition of “device,” its seeming prohibition of even de minimus chemical action contributing to a therapeutic effect, and its unexplained departure from prior precedents.  The Agency lost at least two lawsuits over its new interpretation of “device.”

In concession to industry concerns, the Final Guidance’s discussion of “chemical action” makes clear that chemical action occurring within the body, but which does not interact “at the molecular level with bodily components . . . to mediate (including promoting or inhibiting) a bodily response, or with foreign entities (e.g. organisms or chemicals) so as to alter that entity’s interaction with the body,” is not sufficient to render a product a drug for purposes of FDA regulation. Final Guidance at 7.  FDA further clarifies that while “interaction at the molecular level” could include chemical reaction, intermolecular forces (e.g. electrostatic interactions), or both, “[t]he mere exchange of non-chemical energy (e.g., electromagnetic or thermal energy) between a product and the body would not constitute ‘chemical action.’” Id. at 7 n. 12.

For example, FDA notes that a polymethlymethacrylate (PMMA) bone spacer would still be considered a device, even though “the molecules [within the product] interact with each other to create a solid mass to fill a bone void physically,” because that “process does not require an interaction between the PMMA and the bone at the molecular level . . . .” Id. at 9 (emphasis supplied).  And topical surgical adhesives, despite bonding to a cut/incision, do not exhibit chemical action “because that binding does not mediate a bodily response.” Id.

At the same time, FDA continues to take the position that the FDC Act’s “drug” definition subsumes the more restrictive “device” definition (Id. at 5), and that product sponsors bear the burden of proving that their product is correctly classified as a device (or drug, as the case may be) (Id. at 6).  Given the difficulty of proving the absence of biological and chemical effect, this burden shifting gives FDA ample room to make mischief.  Moreover, while the Final Guidance no longer states outright that the Agency considers every therapeutic effect to be a primary intended purpose, the examples it cites of products that exhibit chemical action in or on the body, but that remain “devices,” include only products whose chemical action otherwise meets FDA’s definition, but does not contribute at all to a therapeutic purpose of the device.  There is no indication in the Final Guidance that FDA would accept as a device a product exhibiting even minimal chemical action that contributes to a therapeutic effect.  Finally, while FDA states that it has “reconsidered inclusion of content on the status of prior Agency classification decisions” in the Final Guidance, the Agency makes clear that product sponsors should not rely on classification precedents without confirmation by OCP that those precedents remain valid.  82 Fed. Reg. at 44,803.

In sum, while FDA has taken steps toward providing greater certainty to product sponsors by clarifying its stance on the meaning of “chemical action” in this Final Guidance, the Agency’s positions on the remainder of the statutory “device” definition, and on the sponsor’s burden of proof to establish product classification, continue to heavily favor a “drug” designation for hard-to-classify products. This result is arguably contrary to the statutory text, and – in light of the significantly higher cost associated with seeking premarket approval for a drug product – is very likely to result in fewer beneficial therapeutic products reaching the U.S. market.