Advances in genomics, proteomics, and metabolomics over the last decade have provided an unprecedented, exquisitely detailed view into the molecular nature of disease and treatment. Biomarkers are among the most practical tools arising from the "omics" revolution.
Biomarkers serve multiple functions throughout a pharmaceutical product's lifecycle. During drug discovery, they help elucidate mechanisms of action and critical disease pathways. Biomarkers can also guide lead optimization efforts by prioritizing candidate compounds according to their biological activity.
When used during drug development, biomarkers can help drug developers tip the risk-benefit scale in their favor, and thus reduce uncertainty, by avoiding side effects that are not countered by therapeutic gain. Several biomarkers are already in use for monitoring the efficacy of drug treatments, for example serum hemoglobin A1c for diabetes drugs and LDL for cholesterol-lowering drugs.
During clinical development, investigators increasingly employ biomarkers to monitor drug efficacy and safety, and to screen patient populations for the likelihood that individuals will respond to treatment or suffer adverse events.
The use of biomarkers to drive clinical development and, eventually, prescribing of new drugs is a relatively new idea. Although slow to catch on, biomarkers have become one of the most vibrant, productive areas of biomedical research. Virtually every major drug and biomedical company is now engaged in biomarker research.
Eventually, the understanding gained through biomarker research will lead to a new treatment paradigm, personalized medicine, of which several examples already exist. Personalized medicine has come to mean many different things, but a good working definition might be "the right drug, at the right time, for the right patient." One defining characteristic of personalized medicine is its reliance on biomarkers to guide which drugs are prescribed and to whom.
The impetus for developing personalized medicines may be safety, efficacy or both. The approval of Nitromed's Bidil, a treatment for heart failure in African Americans, is an example of a targeted medicine. When Bidil was originally developed, it appeared to be ineffective in the general heart failure population, but closer inspection of the data revealed that a significant response occurred in African American patients.
This was not entirely unexpected because for years it was recognized that African Americans do not respond as well as Caucasians to conventional angiotensin converting enzyme (ACE) inhibitor anti-hypertensive agents. Bidil, according to the package insert, corrects that treatment gap. Using race to target medicines is obviously a blunt tool and should be viewed as only a first step in uncovering the molecular basis for differences in treatment response. Once the molecular basis is understood, diagnostics based on biomarkers can be used to select patients for treatment with Bidil.
The next phase of biomarker research, already underway, is to identify biological indicators associated with serious adverse drug events. Tests, say, for liver or kidney toxicity after a drug is prescribed could greatly reduce the incidence of life-threatening side effects while keeping lifesaving drugs on the market. Personalized medicine takes this idea to the next level, by identifying patients in particular response categories before a drug is prescribed.
The arthritis drugs Vioxx and Bextra and the diabetes drug Rezulin are examples of widely effective drugs that were withdrawn due to serious adverse reactions such as stroke, heart attack and liver failure. Another drug, Biogen Idec's Tysabri, a treatment for multiple sclerosis, was withdrawn after three patients experienced a rare brain disease and two died. Had these drugs been coupled with an appropriate biological indicator, each might have remained on the market. Indeed, after further review of available safety data by an FDA expert panel, Tysabri will return but physicians will be required to adhere to a comprehensive safety-monitoring program.
Recently, Roche Diagnostics introduced the AmpliChip CYP450 Test, a sophisticated gene microarray that enables comprehensive analysis of two genes that influence drug metabolism, and hence drug safety and efficacy, the cytochrome P450 2D6 and 2C19 genes. This is the first device intended for broad diagnostic use that allows physicians to personalize treatments according to drug type and dose.
The U.S. healthcare industry still has a long way to go before our understanding of complex disease pathways, and how they are affected by different drugs, leads us to personalized medicine as a way of life.
As we gain more insight into complex disease/treatment pathways, we will be better equipped to personalize drugs, to predict which drugs will work on which types of patients, and which patients should avoid certain treatments.