Three years ago, Theo Dingermann, a professor of pharmaceutical biology at Goethe University in Frankfurt, sent a glob of spit to the United States. For $200 a company there sequenced part of his genome using DNA from his cheek cells. The test revealed that if he were to take the cholesterol-lowering drugs statins, they would have little effect. In fact, they might harm him.
"I'm healthy at the moment, but if I took statins, there's a risk I would get muscle pain," he says. Statins work through the blood, coursing through the body until they reach the liver. Here, they pass through a door, known as a carrier, which takes them from the blood into the liver, where they lower cholesterol production. Dingermann's genetic test showed that he has fewer carriers than usual, which means that statins could stay in his bloodstream and start harmfully reducing the cholesterol in his muscles instead.
Realising that he might have to live without these artificial cholesterol-busters, he decided to get healthy. "I lost some kilos and started running. Last year, I jogged 1,400km."
Intolerance to a drug is something most people learn about themselves only after it's prescribed. "There are some tumour drugs, for certain cancers, that 70% of the population won't respond to. For some asthma drugs, it's 40%," says Dingermann, who has been collecting research on the topic for the past four years. His source is a 2001 article published in the journal Trends in Molecular Medicine, which also stated that antidepressants were ineffective for 38% of people, diabetes drugs for 43%, arthritis drugs for 50% and Alzheimer's drugs for as many as 70%.
Ann Daly, a professor of pharmacogenetics at Newcastle University Medical School, says these rates are difficult to verify. Some people, for instance, don't take the medicines they are prescribed. But for a very small minority, the consequences of taking their medications are dire. A paper in the Journal of the American Medical Association in 1998 estimated that adverse drug reactions ranked between the fourth and sixth biggest killers in the United States. "Medical practice hasn't changed all that much since this paper appeared. Adverse drug reactions remain a big killer in Europe, as well as in the US," Daly says.
Part of the problem lies in how medicines are developed. Clinical trials deliberately ignore a patient's background and genetics so they can avoid bias. But this in turn means that, so long as a drug works well for most people, doctors will prescribe it to everyone with that condition. "Such a study doesn't tell you how a particular patient responds to a drug. Some people have a personal biochemistry which means it doesn't work at all," says Dingermann.
The tiny risk of side-effects is something we accept almost every time we swallow a pill, but Dingermann argues it's one we need no longer live with. If everyone had their genomes analysed the way he did and if medical trials included patients' genetic backgrounds, they could steer clear of the drugs that don't suit them, he says. He has been urging governments and health insurance companies to adopt widespread genetic testing. This kind of personalised medicine could save millions in unnecessary medical bills.
Scientists are beginning to back the idea with greater force. A report in January 2012 by the Human Genomics Strategy Group, which advises the UK government, laid out a vision for the NHS in 2020 that includes using patients' genetic profiles to avoid drugs that could harm them. The group's chair, Sir John Bell, a former president of the Academy of Medical Sciences, said that a national DNA database might be needed in the future.
One successful example of genetic testing is the HIV and Aids drug, Abacavir. "Around 1–2% of people develop a severe, allergic-type reaction to it, and for them, it can be life threatening," says Daly. Studies have shown that this reaction is linked to a specific set of genes, so since around 2008, doctors have been advised to screen patients before prescribing Abacavir.
Scientists are developing similar tests for the blood thinner Warfarin, which is already carefully dosed but can cause severe bleeding in up to 5% of patients. There are also "more and more genetic tests on tumours," adds Geoff Tucker, emeritus professor of clinical pharmacology at the University of Sheffield. "In cancer therapy, the application of genetic testing has been very successful."
But he is sceptical that genetic screening will be rolled out across the entire population any time soon. "The reasons that people don't respond to drugs are more than genetic," he says. Tucker suggests a possible alternative might be to pump a variety of indicators, including age, weight, diet and environmental factors, along with genetics, into software that could determine how much a patient's dose of a drug should be.
Research linking genes to drug intolerance "is only halfway there", agrees Daly. "Not to mention, if you wanted to screen everybody, that would be expensive and require a big infrastructure." Keeping this kind of personal information also raises privacy issues. Where they exist, DNA banks are kept under strict controls, but with widespread databases and weaker security, genetic data could be used to discriminate against people based on their genetic traits. Insurers, for instance, might raise premiums for patients with a higher susceptibility to heart disease.
But Daly admits that, as the cost of genetic testing plummets, it may become a bigger issue. Like Theo Dingermann, more people could simply start paying to have their spit checked.