Latest estimates by the agency are that the document will be published next week, although it has already been subject to numerous delays, according to its Center for Drug Evaluation and Research.
The FDA is hoping that official guidance on pharmacogenomics, which looks for associations between genetic variations and responses to drugs, will encourage pharmaceutical companies to include this type of data in their drug submissions, broadening the agency's knowledge in the area.
From a clinical perspective, the potential of pharmacogenomics is expected to be significant, both for the patient and the healthcare system providing treament. In time, it could be possible to carry but genetic profiling of patients, using microarrays, to drive the best treatment selection. For example, some patients are fast metabolisers of particular drugs, rendering the treatment ineffective, and identifying them could allow alternative, effective drug treatment to be delivered more quickly.
This drug metabolism area is probably the most advanced, braod-based application of pharmacogenomics to date, but more focused applications are already in widespread use.
For example, patients with breast cancer can now ben screened before receiving treastment with Roche and Genentech's drug Herceptin (trastuzumab), which targets a specific form of the disease in which the tumours overexpress a protein known as HER2. Gene amplification or protein expression testing identifies those patients with tumours that will respond to the treatment. Imilar testing startegies are also in place for Novartis' new lymphoma drug Glivec/Gleevec (imatinib) and Erbitux (cetuximab), a coloreactal cancer treatment eveloped by ImClone Systems, Bristol-Myers Squibb and Merck KGaA.
In a presentation a year ago, the FDA's director of the Office of Clinical Pharmacology and Biopharmaceutics, Lawrence Lesko, said that he expected screening of patients for genetic variations - single nucleotide polymorphisms (SNPs) - that could guide treatment decisions to be in routine use by 2007-2010.
At the time, he presented a model showing that the rate of potentially life-threatening hypersensitivity reactions to an HIV drug, GlaxoSmithline's Ziagen (abacavir), could be cut from 4 per cent to 0.25 per cent.
The drug industry wants firm guidelines on the type of data that will be accepted by the FDA, as well as assurances that these data can be shared with the agency without leading to unnecessary interruptions in the clinical development of drugs.
On the whole. drug companies are fully behind the concept, despite initial concerns in the 1990s that the more targeted therapy achievable with pharmacogenomics could reduce the number of patients receiving drugs, and eat into revenues. But with Herceptin, Glivec and Erbitux all posting blockbuster sales, these concerns have receded.