Regulatory Editor

The FDA has finalized a July 2015 draft guidance governing 510(k) applications for magnetic resonance imaging systems, and in the process added three pages to the document. The agency heard from industry that a proposed requirement to provide descriptions of all imaging protocols on MR systems would be impractical, and wrote the final guidance so as to require only the manufacturer-provided imaging protocols in 510(k) submissions.

The FDA said in the draft guidance that the agency would incorporate several existing standards into the framework for 510(k)s for MRI systems, although the draft made note of the fact that no standards are as yet available for MR spectroscopy performance. The draft further indicated that the FDA was intent on applying the document's provisions to combination modality systems such as PET/MR as well. (See Medical Device Daily, July 15, 2015.)

The draft guidance listed 11 standardized test methods by Arlington, Va.-based National Electrical Manufacturers Association (NEMA), but the final drops NEMA's MS-11 standard for determination of gradient-induced electric fields without offering a specific replacement. Another standard found in the draft that was excised before the final guidance was drafted is the UL 94 test for flammability of plastics, although both ISO 10993-1 for biological compatibility testing and NEMA PS 3.1-3.8 for digital imaging and communication both survived the cut into the final.

The Medical Imaging & Technology Alliance, a division of NEMA, said in Oct. 9, 2015, comments to the docket that the association was "concerned" that the draft's mandate for detailed descriptions of "all imaging protocols ... and related testing" was unduly burdensome, in part because modern MR systems may include "hundreds or thousands of pre-programmed imaging protocols."

MITA's director of MR imaging, Andrew Northrup, said that these protocols are "routinely customized on site to optimize for actual clinical usage," and suggested the agency consider a requirement that is limited to systemic controls such as pulse sequence types. The FDA seems to have heeded this advice as the final guidance describes the term "imaging protocol" as "a workflow tool consisting of multiple pulse sequences prescribed in a defined order." Among the items of interest for the agency are descriptions of target anatomy, coil preference/restriction, and "a list of the pulse sequences included in each protocol."

At least partly at MITA's request, the agency expanded the draft's discussion of device modifications, which the association said called for more clarity. Whether a modification would require a new 510(k) filing is discussed briefly in the draft, but the final breaks out what had been a subsection of requirements for performance data into a stand-alone section of the final guidance.

Among the modifications that would constitute a sufficiently significant change to require a new filing is any change to the main static magnetic field, which the agency noted could alter the system's resonant frequency. Changes to the gradient system and any one of several coils could likewise trigger a need for a new regulatory filing, although this section also said that changes to a protocol might not require a new 510(k) so long as there is no change to intended use that includes claims regarding a specific disease state.

The agency added that the addition of a new protocol to a particular system might not require a new filing if that protocol is already in use in another system made by that same manufacturer, assuming the two systems possess similar technological characteristics. The section on software is largely unchanged with the exception of an additional discussion of cybersecurity and one of validation of off-the-shelf software.


FDA also finalized the draft guidance for investigational device exemptions (IDEs) for devices intended to treat neurological diseases, needing only eight months to put the finishing touches on the March draft guidance. The agency sent a clear signal about the importance of the document given the unmet needs of patients with Alzheimer's and Parkinson's diseases, but noted that such devices should "target either the cause or progression" of the disease rather than the related symptoms.

The agency indicated it is amenable to the use of surrogate and intermediate clinical endpoints in the design of a study, providing definitions of both types of endpoints. However, the FDA said assessment of such measures is no mean feat, and noted that biomarkers "may not be accompanied by clinically meaningful [and] observable changes." The guidance said these considerations are particularly pertinent when a patient is asked to pass on approved treatments in early disease stages, particularly when pharmaceutical therapies are both available and better understood.

The guidance said that it may prove difficult to distinguish between treatments that affect symptoms versus disease progression, recommending that a study be designed so as to provide quantitative information on the impact of device treatment on biomarkers and assessments of clinical outcomes.

The FDA noted also that an IDE study could be extended in order to develop an understanding of disease progression when other therapeutic alternatives are available. Other sections of the draft take up matters such as informed consent and labeling.

No Comments