673combination of genes or proteins is a better indicator of disease or disease risk than
a single gene or protein. FDA considers regulation of such tests because the algo-
rithms used are usually proprietary, making it diffi cult for physicians to interpret
the test results. Therefore, the agency needs to look at the data on which these tests
are developed. The FDA would decide case by case what to do about the tests
already on the market. Some might have to come off the market until the developer
can provide enough data for approval. The FDA approach will meet the need for an
oversight of genetic tests, which have proliferated and are becoming increasingly
complex. Government agencies have been criticized for not doing more to clamp
down on questionable genetic tests that are being sold directly to consumers.
Three components are needed to ensure the safety and quality of genetic tests: (1)
the laboratories that conduct the tests must have quality control and personnel stan-
dards in place to prevent mistakes; (2) the tests themselves must be valid and reli-
able – i.e., detect genes that are actually related to disease or disease risk accurately
over time; and (3) health care providers must understand when to order the tests,
how to in interpret them, and what to do with the results. Once these mechanisms are
in place, uses and outcomes also must be evaluated over time in order to pinpoint
any problems that may require attention, particularly as new tests enter wider use.
However, the requirement could also discourage the development of diagnostics
by raising the costs of introducing them. Requiring clinical trials and FDA approval
would discourage development of tests, which do not usually command the same
profi ts as drugs. The requirement could discourage gradual improvements of tests
because each change in a test might require a new regulatory submission. The draft
policy has raised speculation that the FDA will eventually move to regulate addi-
tional laboratory tests beyond the complex ones.
In 2007, the FDA classifi ed gene expression-based breast cancer prognostic tests
as Class II devices and released a “special controls” guidance for companies devel-
oping such tests. The document is designed was a prototype guidance to provide a
general framework for how the FDA’s Offi ce of In Vitro Diagnostics approaches
IVD MIAs. The FDA cleared the fi rst such IVDMIA device – Agendia’s MammaPrint
test – in 2007. In a Federal Register notice in 2007, FDA explained that it had origi-
nally classifi ed MammaPrint as a Class III device, which would have required full
premarket approval, but Agendia fi led a petition requesting that the device be reclas-
sifi ed into Class II, which only requires 510(k) premarket notifi cation. The FDA
determined that MammaPrint, as well as future genomic breast cancer prognostics
tests, can be classifi ed as class II devices with the establishment of special controls,
which are outlined in the guidance document as follows: “Any fi rm submitting a
510(k) premarket notifi cation for a gene expression profi ling test system for breast
cancer prognosis will need to address the issues covered in this special controls
guidance,” the agency said in the document. The recommendations in the guidance
document apply to RNA expression assays used for cancer prognosis, including
RT-PCR and gene expression microarrays, in which an algorithm is applied to such
measurements to yield a result that can be used by physicians as a prognostic
marker, in combination with clinicopathological factors, to assess the risk of cancer
recurrence. The process for reviewing such tests is “contingent on the intended use
of the device” therefore, design of studies and data sets required will be infl uenced
FDA and Personalized Medicine