IVDMIA 101

PLEASE NOTE: This page describes a guidance that no longer reflects FDA's current thinking. It is here for historical reference only.

FDA intends to extend the enforcement of medical device regulations to some LDTs. FDA has published draft guidance document entitled Draft Guidance for Industry Clinical laboratories, and FDA Staff: In Vitro Diagnostic Multivariate Index Assays (IVD MIA).

The ﬁrst draft of this document was issued in September 2006 and after a comment period and holding public comment meetings a second draft of the guidance was issued on July 26, 2007. Although there have been legal challenges made that assert FDA does not have authority to regulate LDTs, FDA has already cleared three assays as a IVD MIAs (MammoPrint, XDx’s AlloMap assay and the Pathwork Tissue of Origin assay).

In the IVD MIA guidance FDA makes it clear that they are not interested in regulating all LDTs. But rather, is taking a closer look at tests that are “based on observed correlations between multivariate data and clinical outcome, such that the clinical validity of the claims is not transparent to patients, laboratorians, and clinicians who order these tests.”

The current working definition of an IVDMIA according to FDA is an assay that:

Combines the values of multiple variables using an interpretation function to yield a single, patient-specific result (e.g., a “classification,” “score,” “index,” etc.), that is intended for use in the diagnosis of disease or other conditions, or in the cure, mitigations, treatment or prevention of disease, and
Provides a result whose derivation is non-transparent and cannot be independently derived or verified by the end user.

An example of an assay that is likely to be considered an IVD MIA is the Genomic Health Oncotype DX assay. This assay is ordered by physicians to predict the likelihood of breast cancer recurrence in women with newly diagnosed, early stage invasive breast cancer. A physician is provided with a ‘Recurrence Score’ between 0 and 100 based on a gene panel that GHI does not fully reveal. GHI has been able to offer the Oncotype DX assay for years without going through an FDA review process.

The advantages of avoiding FDA review are obvious, faster time to market and less regulation on business activities and promotion. If the new IVD MIA guidance document is finalized, companies that market IVDMIAs will need to file an application with FDA within 12 months of the final guidance and would be subject not only to FDA review but also will need to comply with the FDA’s Quality System Regulation (QSR). This would be a much larger hurdle for laboratories to get over and will burden them with additional expensive, high-impact activities such as design control.

In the July 2007 version of the IVD MIA Guidance FDA offered some examples of assays that are outside of the IVD MIA definition. These include:

Multivariable tests that provide enough information to the physician to allow independent interpretation (triple screen testing, hCG and estriol)
Genotype determinations such as CFTR testing.
Chromosomal copy number determinations
Common clinical calculations such as creatinine clearance and cholesterol ratios
Clinical decision support tools
Common public risk calculations such as Framingham scores because information is publicly available to the physician to independently assess.

The current status of the IVD MIA guidance is officially unknown but recent conversations with FDA and public FDA statements indicate that significant efforts are continuing and the Agency hopes to finalize the document in 2009.

Will All Assays Based on Multiplex Technology be Considered IVDMIAs?

Because multiplex technologies are specifically designed to test for multiple parameters simultaneously there has been some speculation when the IVD MIA guidance document was first released that any assay run on multiplex would be subject to the IVD MIA. FDA has made it clear via the second draft of the document that they are not determining whether an assay is an IVD MIA because it is on an array or other multiplex system, but rather how is the final result presented to the physician.

There is no mention of technology choices in the IVD MIA guidance document. Just because an assay is performed on a particular type of system does not mean that it is subject to the IVD MIA guidance document. Just as performing multivariate analysis on the results of more traditional ELISA assays and combining the information does not exempt a laboratory from the IVD MIA.

The important factors in considering whether an assay is subject to the new IVD MIA rules is whether the specific assay is multivariate and whether presentation of the results is transparent. If the assay result is a ‘black-box’ to the physician than FDA wants to see the data that supports the end result.