IVD Roundtable at the FDA, Jan. 13, 2010

At the FDA, the OBRR Director Jay Epstein welcomed us and introduced the first speaker, Sayah Nedjar, who gave us a CBER update. Nedjar is a reviewer in Biologics and ombudsman for biologic combination products, and he discussed CBER record reviewing IVDs, IVD products and equipment associated with transfusion, cell therapy, etc. The average total 510(k) review time last year was 121 days, with average FDA time 80 days. He stated that CBER welcomes meetings, especially pre-IDE. For formal meetings they prefer participants to employ their SPOP 8101, and for informal meetings SOP 8104. CBER and OBRR are working more closely with CMS and are involved in laboratory surveillance. They have assisted in updates to MEDRA nomenclature and ICD 9 and 10, and are working with the EU (ISPC) on a project to improve health care product information.

The next speaker was Dr. Alberto Gutierrez. Now Director of OIVD, Dr. Gutierrez stated that CDRH would soon have a new Director. (Post-roundtable note: Jeffry Shuren, MD, is now confirmed as Director, CDRH. At the time of the meeting he was still Acting Director.) The Center will also have deputies for Science and Policy. Dr. Gutierrez stated that OIVD’s interests were primarily the following:

  • Safety issues
    • Public health issues created by use of glucose monitors: he stated that the FDA wants increased accuracy in these meters
    • Personalized medicine: OIVD has charged Liz Mansfield with developing policy for these tests and the Commissioner has spoken and will speak more about this soon.
  • Transparency
    • The Center is now messaging on Twitter, and the FDA is on YouTube
    • The Center is planning an open meeting in February on Science and Regulatory decisions (Post-roundtable note: took place on Feb. 9)
  • Review of 510(k)’s
    • CDRH has a contract with an outside vendor to review the 510(k) process and make recommendations, including answering the following questions:

      • Are 510(k)’s still a viable way to get products to market
      • Is it failing to ensure safe and effective products?
      • Do we need major changes?

In other news, Radiology is joining OIVD; Don St. Pierre will be the Director. For guidances, the de Novo guidance is complete, but is being held up by Counsel’s office. The Emergency Use Authorization Guidance will be available within twelve months.

Dr. Gutierrez then asked for questions from the floor.

When asked about IVDMIA, Dr. Gutierrez responded with a long silence and then said that the new administration was very absorbed with health care reform and that once a new health care bill is signed, he expects them to move on to other things. FDA believes that the administration is favorable to publication of the final guidance and expects publication sometime this year (this was of course before the health care bill slipped back into chaos).

When asked about waiver applications, Dr. Gutierrez said OIVD only got six and approved one. In many cases the guidance was not followed. The 510(k) and waiver pilot examined two applications. However, it did not work out well, and did hold up clearance of the 510(k).

About the future of 510(k), he couldn’t say. They are looking at the whole 510(k) program. For IVDs, “replacement reagents” is a better approach than Specials.

Dr. Gutierrez made a point of stating that OIVD has always cleared products based on data, which is a complaint many have about other CDRH 510(k) review groups (the complaint being that the other groups don’t require data).

Asked about enforcement against LDTs, Dr. Gutierrez commented that the administration understands the problem, and he expects movement in this area to begin “soon”. Dr. Mansfield added if an LDT presents a public health risk, the FDA will move on it. When asked how the FDA would move, Dr. Gutierrez stated, “We are working the issue.”

In response to the question of whether the de Novo process will be part of the 510(k) review, Dr. Gutierrez said yes, he thinks the de Novo is a good idea and a growing trend.

Dr. Tezak spoke about the FDA’s concern that components of tests offered as laboratory diagnostics are actually labeled RUO. It is easy to translate a research test to actual clinical diagnostic use, but the mix and match system for molecular methods has caused the FDA some concern. The issues are that the products are not manufactured under GMP, and that there may be unacceptable variation blot-to-lot. Reagents may be substituted, when in fact for some applications the reagents are not interchangeable. There is still no IVD sequencer manufactured under GMP. Tezak referenced the respiratory viral panel guidance and its treatment of ancillary reagents (author’s note: I think she meant “Guidance for Industry and FDA Staff—Class II Special Controls Guidance Document: testing for Detection and Differentiation of Influenza A Virus Subtypes Using Multiplex Assays”). She stated that the entity offering the test as a diagnostic is responsible for assuring that the ancillary reagents to work consistently with their test (author’s note: ancillary reagents are reagents that are required but not provided by the manufacturer).

Asked by the audience when the RUO guidance will come out, Dr. Tezak responded there should be a draft in 2010 (maybe).

Next Dr. Mansfield spoke on personalized medicine. There is a group actually working on a co-development (drug/diagnostic) draft, and one working on statistical applications suitable for multiplex diagnostics. They are working with the NCI for 510(k) guidance for proteomic assays employing multiple markers. She called our attention to a mock pre-submission published in Clin Chem (Regnier et al. 56 (2): 165. 2010). (Author’s note: I have a copy—it is not very informative, but I can provide it to anyone who asks.)

The most interesting of the afternoon sessions was the explanation of the migration studies guidance draft, especially the session by Marina Kondratocvich, who is a statistician at OIVD. This draft has offered a number of statistical approaches to evaluating the similarity of two methods for the same indication when a manufacturer wants to support a change in the methods (for example, from use of instrument A to instrument B). The guidance suggests far more rigorous criteria than are usually done when a company changes a method, and additionally suggests far more rigorous criteria than are usually employed to show similarity in performance. The applicability of the methods was questioned by some of the audience, but in general most of the complaints were reserved for behind the scenes. You should review and comment on this draft guidance, as it will mean big changes for companies if/when it is really finalized.

AMDM should be publishing the overheads from this meeting soon on their website.