The promises of targeted drug discovery in the early days of human genome research were coupled with achingly slow progress and costly setbacks, leaving many companies and investors with a painful hangover and a hesitancy to step back in. “The field was generating enormous numbers of false positives, which made many people cynical—including me,” says Alexander Kamb, Amgen's senior vice president for research. Vincent Mooser, former vice president, head of Applied Genetics at GlaxoSmithKline in London, recalls the excitement touched off by the discovery in the early 1990s of a link between variants in the gene encoding apolipoprotein E (ApoE) and Alzheimer's disease. “The idea was that geneticists had nailed down a disease with a common variant that can prompt the discovery of new drugs. But 20 years down the road, people still don't know whether ApoE is the culprit.” Countless genome-wide association studies have scanned the chromosomes of patient and control populations to home in on subtle sequence variations associated with various diseases, but so far these have borne little commercial fruit.
Some believe dramatic improvement of sequencing technology and steady accretion of high-quality genomic data over the past decade could usher in a new era of discovery. Robert Plenge, head of genetics and pharmacogenomics at Merck in Boston, thinks the time is right to focus on programs that examine large populations to identify 'experiments of nature'—genetic perturbations that yield clear physiological changes, which can then be mimicked through chemical or biologic modulators. “Ten years ago we had a map of the genome but not of genetic variation, and even though the HapMap Project gave us that genetic variation map, we didn't have the relationship between those variants and relevant phenotypes,” says Plenge. “Now we're able to start making connections between these genetic variants and relevant clinical phenotypes to drive drug discovery.”
This is a preview of subscription content, access via your institution