SAN FRANCISCO — Pharmacogenomics has already entered the practice of medicine, with several recent developments that went largely unnoticed, Richard M. Weinshilboum, M.D., said at the annual meeting of the American College of Physicians.
New guidelines on pharmacogenomics from the Food and Drug Administration that were first drafted in 2003 were finalized and published in March, noted Dr. Weinshilboum, professor of medicine and pharmacology at the Mayo Clinic, Rochester, Minn.
The guidelines state that when a company with an investigational new drug has pharmacogenomic information on the drug, it should be submitted with the rest of the approval data. Currently, the submission of these data is considered voluntary.
On the one hand, industry is embracing pharmacogenomics because it can provide essential information in the initial studies of an investigational drug about whether it is worth the effort to continue pursuing expensive development. But on the other hand, industry is being “dragged” unwillingly into pharmacogenomics by the FDA because information on patients who may not respond to a drug or who may have side effects could spell the end of blockbuster drugs, Dr. Weinshilboum said.
The guidelines identify two pharmacogenomic biomarkers that the FDA will consider valid biomarkers. They are cytochrome P450 2D6 and thiopurine S-methyltransferase (TPMT). Information on other markers that identify possible genetic variation of response to a new drug should be submitted also, but the information needs to include background data on the biomarker.
TPMT is an enzyme known to be involved specifically in the metabolism of mercaptopurine, which is commonly used for childhood acute lymphoblastic leukemia, and the immunosuppressant azathioprine (Imuran). It has been shown that individuals have wide variation in TPMT activity. There are three important mutations in the TPMT gene, and individuals with low TPMT activity build up high drug levels leading to myelosuppression that is sometimes severe.
Cytochrome P450 2D6 is one of the isoforms of the cytochrome P450 enzyme family. It has been shown that individuals can have more than one copy of the gene for cytochrome P450 2D6, and that persons with more copies are high metabolizers of the affected drugs. At least 41 different drugs are known to be metabolized through cytochrome P450 2D6; an important one is codeine, which cytochrome P450 2D6 transforms to morphine.
Dentists already know about cytochrome P450 2D6, because just this year Roche Diagnostics received approval for its AmpliChip CYP450 test, a microarray test for clinical use. Patients are coming in armed with information from the Internet, demanding to have their cytochrome P450 2D6 status checked to see if they are among those who do not respond to codeine, he said.
Pharmacogenomics also has recently been an issue in the approval of the new “smart” cancer drug gefitinib (Iressa), which is used for non-small cell lung cancer. The drug was given accelerated approval in 2003, based on preliminary data on a response in some end-stage patients. But a new, placebo-controlled trial of the drug was recently halted because gefitinib failed to show any survival benefit.
The study may have failed to show a benefit because the investigators did not obtain genetic profiles of the patients. It now appears that gefitinib can produce a marked response, but only in those patients with a particular mutation of the epidermal growth factor receptor gene—about 10% of all patients, he said.
AstraZeneca, manufacturer of Iressa, has recently announced that it will begin cancer biomarker studies of the drug.