NEW YORK (GenomeWeb) – Next week, the US Food and Drug Administration will convene with experts in protein and peptide detection using liquid chromatography-mass spectrometry to discuss how to regulate in vitro diagnostic tests based on the method.
So far, the FDA has not had to regulate any aspect of a mass spec-based protein or peptide test, because nobody has submitted one for clearance or approval, according to Julia Lathrop, a scientist at the agency's Center for Devices and Radiological Health.
"We think mass spec has tremendous potential and we'd like to see it used more in the clinic with these peptide-based assays," she told GenomeWeb.
The workshop, which will be held May 2 at the White Oak Campus in Silver Spring, Maryland, will focus on validation considerations for the technology, which the FDA is hoping to elucidate.
With many variables, LC/MS is a powerful but complex method. In advance of the meeting, the FDA has posted a six-page-long discussion paper full of topics and questions for discussion, for example: "Given the variety of proteins measured and the workflows involved, please discuss appropriate calibrators and control materials for quantitative protein/peptide LC-MS/MS devices, and what commonalities can be applied to all protein/peptide LC-MS IVDs. Please discuss when external vs. internal calibration is preferred."
Though the workshop comes as the FDA is gearing up to regulate mass spec-based laboratory developed tests (LDTs) Lathrop stressed that the agency wanted to focus solely on issues related to analytical validity, rather than policy. Still, it's not hard to imagine that the lessons learned won't somehow be applied when the FDA does begin to regulate those tests.
The workshop comes out of what Lathrop described as an increased effort by the FDA to work with the proteomics research community. In June 2014, the FDA held a workshop on proteomics in the clinic. "Mass spec is a natural outgrowth in focusing the output of that meeting into trying to address specific analytical validation considerations around the development of these tests, so we can get more of these protein and peptide-based tests into the clinic and get them to patients," she said.
While the FDA has blessed other kinds of mass spec-based devices for use in clinical settings, like for microbe identification, drug monitoring, and screening, it has yet to clear or approve any in vitro diagnostics.
While a laboratory developed test may be an easier route to take, it is possible test developers don't know what the FDA expects of them. "We are working on developing a regulatory path forward for these devices," Lathrop said. "As we develop guidance for them, we want to know, what are the things we should be looking for?"
There are numerous considerations that must be taken into account: the choice of analyte, instrumentation, sample preparation, and internal controls. Many of them are specific to the LC/MS community. Lathrop indicated that the FDA is interested in hearing if the researchers believe the agency's expectations are unreasonable, or conversely, not rigorous enough.
While the FDA has provided an outline for the workshop, as evidenced by its discussion paper, Lathrop said she's open to addressing other topics that the FDA hasn't deemed urgent.
While raw data and its analysis and processing is one of the slated topics, Lathrop told GenomeWeb that software and informatics could be of major interest to the FDA.
"Informatics is critical," she said. How software packages perform tasks such as peak selection and normalization are of particular interest.
And given that the FDA performs its own analysis for submissions, what it wants to see — and how it wants to receive data from the test developer — is an ongoing internal discussion. "We reevaluate, we re-graph, re-plot. We look at your analyses de novo ourselves," Lathrop said. "We're not going to want to see all 100,000 chromatograms, but are there chromatograms around the cutoffs, the medical decision points, that we want to see?"
While the FDA is interested in receiving data, Lathrop said it doesn't want to put an unbearable burden on sponsors. Demanding a truckload of information would not be a useful expectation, she said. "How we want to manage [data], given the necessity to look deeply at the data, is something we don't quite know, so we are interested in feedback and discussion."
Though the FDA is seeking input, it already has developed guidance for a pre-submission process that test developers can go through.
"Pre-submissions can be super detailed or much more high level," Lathrop said. "We very strongly encourage people who are developing tests to use that process because it gives them a chance to ask us what we're looking for, a chance to ask us what we think of their plan." It's worked well for other types of tests and, best of all, it's free for everybody, whether it's their first time developing an IVD or their fiftieth.
Though the FDA is hoping policy debates don't seep into the discussion, if a workshop held earlier this year on how to regulate NGS panels for cancer treatment is any indication, it's likely to go there.
Lathrop offered this glimpse into the FDA's stance on how it might regulate LC/MS tests submitted for clearance or approval and for which there are predicate devices. Lathrop told GenomeWeb that while LC/MS tests will be judged on their own merits, their risk class would likely be the same as existing tests with the same intended use.
Medical device classification is based on the intended use of the device, the diseases, the patients, and how it's going to be used in diagnosis, she noted. "If other devices with the same intended use are Class 2, or moderate risk, although I wouldn't make any promises, the expectation would be that a mass spec device that had the same intended use would probably be a Class 2 device that also gets cleared," Lathrop said. "If a different device was Class 3 with different technology, the mass spec device with the same intended use would also be Class 3 and subject to FDA approval."
As the FDA settles its expectations for demonstrating analytical validity of LC/MS tests, Lathrop said the agency was noncommittal to the idea of issuing guidance in the wake of the workshop, but did say it was working to develop that guidance.
"We'd like to give some helpful feedback for the regulatory pathway for these devices," she said.