At the ninth annual HUGO meeting in Berlin, you might expect to find a series of talks lauding the progress of genomics and the future promise of its applications. But on the subject of pharmacogenomics, attendees received a refreshing dose of humility.
In a symposium that veered from research priorities to the ethics of keeping genomic data private, Klaus Lindpaintner’s comments stood out. Given the lack of large-scale studies, the head of Roche Genetics said, it is not even clear that pharmacogenomics will end up helping patients — although he certainly hopes it will.
“Because of the lack of data, the real hurdle is the lack of conviction that pharmacogenomics actually makes a difference [to patients],” he says. “The best we can hope for is a modest degree of optimism that it will improve healthcare.” He adds: “We as a company are struggling with this ourselves.”
Lindpaintner went on to debunk the most common myths associated with pharmacogenomics, and laid the blame for misconceptions by the general public squarely on the shoulders of the scientists themselves. The ability of pharmacogenomics to “revolutionize” medicine is at the moment mere fantasy, he told attendees, citing the example of an insensitive genetic test that requires a blood test as confirmation. The cost effectiveness of such a genetic test is negligible if a doctor has to perform a blood test anyway, he says.
He went on to say that “individualized” medicine is “one of the more crass oxymorons” currently in use. Pharmacogenomics will never be able to achieve 100 percent accuracy with respect to a single individual, he adds, because the field deals with the probability that a group of people will respond in a certain way. In other words, contrary to another typical misconception, he adds, a pharmacogenomic test is not necessarily absolutely unique.
But the fallacy most important to correct, Lindpaintner says, involves the public’s basic understanding of the nature of a genomic test. Just because a test involves gathering data about a person’s genome does not make it any more insightful than a method for gathering information more conventionally, such as through a doctor’s physical examination, he says.
So what is Roche doing to meet at least some of the expectations for pharmacogenomics? Pharmas are slowly gathering more data on how different groups of people respond to treatments as they go about screening potential new drugs, Lindpaintner says. If, for example, researchers at Roche see a bimodal response to a particular drug, it would be interesting to understand why that happens, and to investigate the potential for capitalizing on the discrepancy, he says. On the other hand, work to prove that medicines already on the market can be more effective for certain subpopulations might be better left to the public sector, he says. “Will pharmacogenomics make an impact? We don’t have the data yet to know,” he says.
— John S. MacNeil