"It's my body and my genes, and I have the right to know what's in them," a woman once told me. She was at risk of breast cancer and had been tested for that gene as part of research study but wanted to know what else was in her DNA. The researcher was planning to conduct other tests but thought it would be too complicated to tell her all of the results and thus decided that he would not do so. Understandably, she was upset. She knew she could get such additional testing done online through direct-to-consumer (DTC) testing companies but was unsure if she should do so -- how accurate the online tests were.
Since ancient times, Oracles, soothsayers, and fortune tellers have been offering prophesies about the future. Palm readers now occupy store fronts and many newspapers run horoscopes. This desire for foreknowledge leads many people to read horoscopes. And to seek answers in their genes.
This week, however, 23andMe, the Google-backed DTC genetic test company, stunned many observers by agreeing to stop sales of its $99 genetic test kit online, saying it will now release only ancestry information and raw data without interpretation. This decision follows a week after the FDA told the company to cease selling the kit, after the company had failed over several years to provide evidence to support its claims about the test's benefits. Many critics of DTC testing companies applaud the agency and the fact that a major problem has now been resolved. Others chastise the FDA, arguing that each of us the right to our own genetic information. Some contend that the FDA is charged with overseeing food and drugs and thus should not even get involved in testing. But the agency has merely sought that the company provide evidence of its claims of medical benefits. As such, the FDA has acted correctly.
Yet the company's planned response to the FDA remains insufficient, and many questions arise about what the company and the FDA each will or should do next. Many people are wondering whether DTC testing has a future, and if so, what.
This case, however, highlights several broader issues that need attention -- how to understand and communicate about genetics, what rights patients should hold, and what responsibilities doctors, labs or agencies have concerning genetics. While older types of tests (offered by 23andMe) have provided too little information, new tests threaten to overload us with data. Two sets of problems thus loom with which we must grapple and that are easy to get confused.
Genetics is a rapidly evolving target, posing many dilemmas for us all -- as patients, physicians and society as a whole. Ten years ago, for the first time, scientists mapped the human genome -- the chains of 3 billion molecules (or "letters") that are the unique blueprint for making each of us. The cost was about $1 billion and has since fallen to about $3,000. Some predict it will soon be $1,000 or less.
Genes have been found that are associated with around 2,000 diseases, but only 23 of these ailments are serious, begin in adulthood, are treatable, and have highly predictive genetic tests. The most common of these diseases is breast cancer: the presence of certain genes confers about a 50 percent chance of getting the disease. But only about 10 percent of breast cancer is hereditary. Hence, around 50 percent of women with the mutation will never get the disease, and many women without the mutation may still get this kind of cancer. Most disorders for which predictive genes have been found are extremely rare. For the vast majority of cases of common disease, genes have not been identified. Instead, common diseases probably result from highly complex interactions of multiple genes and various environmental factors -- from pollutants to stress -- combinations of both nature and nurture.
23andMe has examined not the whole genome, but only one out of every 3,000 letters. The results are even thus more limited. But the company continued to hype the benefits - though for the vast majority of customers, these have been nil. The company has now admitted, however, that its business model focuses on building a database of 1 million samples, which could then yield insights into disease, and also valuable patents, and profits for the company.
The company's new plan to offer raw data, however, poses other major concerns of how the company will market these tests. These data are essentially uninterpretable by the vast majority of people. DTC companies themselves have been found to disagree in how they interpret the same raw information. Some company supporters say the physicians can do the interpretation. But research I recently published in the Journal of Genetic Counseling found that 74 percent of internists feel their knowledge of genetic is poor, and 77 percent felt they needed more training on how to interpret results. Patients are even more bewildered. Still, the company may now claim to consumers that it is providing health information -- even if it is incomprehensible to most physicians and patients.
Moreover, many customers of 23andMe do not realize that the company's main goal has been to amass a large profitable data base, rather than provide useful health information to consumers. It will be crucial that the company make this purpose far clearer to its consumers.
Thousands of patients are also now getting their whole genomes mapped. DTC companies are already starting to sell whole genome testing, and many observers say that 23andMe should and eventually will do so. Whole genome mapping provides 3,000 times more data than 23andme's test, much of which is more helpful. But we still don't know how to interpret most of it -- what its specific medical interpretations are for most people.
In addition, as whole genomes are tested, many patients will be found to have genes for various disorders -- including the 23 treatable ones above. This past spring, the American College of Medical Genetics announced that all labs testing whole genomes should give results concerning those 23 diseases to the patients' doctors, presumably to be returned to the patient -- regardless of the patients' ages or desires for the test. Hence, the parents of infants will be told that their child has the breast cancer gene, and has a 50 percent chance of developing the disease perhaps 40 years from now. In the future, this information will also undoubtedly be linked to our electronic medical records.
Yet that's too much information to give without asking patients (or their parents) whether and how they want it. Instead, we need to find out more about what these genes mean -- how often they appear in healthy individuals who never develop the diseases in question, and how these doctors and patients will react to all this information. Discrimination could potentially also ensue. Here, we need to avoid being too paternalistic, and respect patients' autonomy more.
Genetic technology is fast outpacing our grasp of it. It is crucial that we begin to prepare as soon as we can -- to understand the benefits, costs and potential dangers involved in these various approaches. We should avoid tests that provide little useful information and oversell what they provide; and prepare for newer tests that give us more useful information than we may want. We need to improve education about genetics for all of us -- whether as doctors, patients, or ordinary citizens. Labs, doctors and researchers need to act responsibly, consider how most patients will respond, and not to rush to quick responses. The FDA should continue to monitor this sector carefully. DTC testing has a future, but needs to be done right. The future of testing our genetic futures is not written in our genes -- it is up to each of us.