At-home genetic testing might revolutionize medicine. And not in the way you think.

Technological advancement often outpaces society’s ability to understand how to use new advances. Direct-to-consumer genetic testing (DTC-GT) is one such technology whose true power far exceeds our collective mental bandwidth to comprehend.

Home testing kits have come into mainstream culture over a short period of time. In 2010, the direct to consumer testing market was valued at $15 million. It grew almost nine times to $130 million in just five years and is projected to reach $350 million by 2022. FDA commissioner Scott Gottlieb’s November statement to streamline the approval process for genetic testing companies ensures the personal genetic testing market will grow substantially in the near future.

These companies market their products as tools for consumers to understand their risks for future health problems (assuming there are ways to prevent those health problems) and to learn about their ancestral histories. However, there are so many factors that dramatically limit the health value of DTC-GT.

My own 23andMe results are an excellent example of how worthless genetic testing can be when performed without appropriate context. I spat into a tube several months ago and waited six weeks without apprehension to get my somewhat underwhelming results. Besides proving what I already knew, that I am 100 percent South Asian and incorrectly labeling me lactose intolerant, my report stated I am at “slightly increased risk” for age-related macular degeneration. I have two copies of a Y402H mutation in the CFH gene. The medical literature shows a clear association between having two copies of this gene mutation (called “homozygous”) and increased risk of age-related macular degeneration (AMD). Age-related macular degeneration is one of the most common causes of blindness in older adults in the developed world.

Despite all of this being true, a simple overview of statistics in medicine shows why this information is of little value. All tests are defined by something called a positive likelihood ratio — which, in general terms, means how strongly a positive test result means the disease being tested for is actually there. For example, if you designed a test for AMD with a likelihood ratio of 100 (which is very high), a positive test result strongly points to someone truly having AMD. The higher the likelihood ratio is above one, the more believable the test result. For most tests in actual practice, the likelihood ratio is not this high, closer to five or ten. Tests with likelihood ratios between one and two are less useful. They are like when your flaky friend says yes to dinner plans, but you can tell from their shaky voice and body language that they will cancel last minute.

According to my 23andMe report, my two CFH gene mutations have a likelihood ratio of 1.64 for developing AMD in the future.

How useful a test is also depends on how likely we think someone has the disease we are testing for before we have gotten the test results — a concept called pre-test probability. In many cases, the biggest determinant of pre-test probability is how common a disease is. While AMD is a common cause of blindness in older adults, it is still a relatively rare disease, occurring in about 1.5 percent of people over age 40. That means that even if a test tripled my risk of getting AMD, my risk would still be relatively low at 4.5 percent. Consider an alternative hypothetical scenario in which 15 percent of all people developed AMD. Now tripling my risk means I have a 45 percent chance of developing AMD, a far more significant and medically important result. Using a likelihood ratio of 1.64, my risk for macular degeneration goes up from 1.5 percent to 2.44 percent, a largely meaningless increase.

A test becomes even more useless when there is nothing to do with the result. The National Eye Institute’s Facts About Age-Related Macular Degeneration article states that avoiding smoking, exercising regularly, maintaining normal blood pressure and cholesterol levels and eating a diet rich in green, leafy vegetables and fish can reduce the risk of AMD. These are all things that I should do for a dozen other reasons like preventing heart attacks and strokes that are more important than preventing AMD. If those reasons didn’t convince me, why should this? There are certain vitamin combinations that have some evidence for preventing progression of AMD from the early stage to advanced stages, but no conclusive evidence that any vitamin or supplement prevents the disease from happening in the first place. In the end, these results caused me 30 minutes of anxiety but added nothing to my health or well-being.

What is most surprising is how infrequently people make lifestyle changes based on their personal genetic testing results. A study of more than 1,000 people who used either the 23andMe or Pathway Genomics personal genetic testing kits showed that the people at higher risk for colorectal, prostate or breast cancer based on their genetic test results were no more likely to get cancer screening, change their diet or exercise compared to people with normal or lower risk for cancer. They did find that men with higher prostate cancer risk were three times more likely to start taking vitamins or herbal supplements compared to men of normal or low risk, despite the lack of quality evidence proving vitamins or supplements prevent prostate cancer.

The true value of personal genetic testing is its ability to collect the genetic data from millions of individuals into one large database for research purposes. Too often, fragmentation in the scientific community or high costs impede aggregation of data and high-quality research. Through partnerships with academic universities, pharmaceutical companies and charitable foundations like the Michael J. Fox Foundation, scientists in numerous disciplines have access to the data they need to generate tomorrow’s advances. There are real concerns regarding privacy of genetic data and possible negative ramifications if insurance companies could access this data. Still, the potential for meaningful medical discoveries is there.

For now, personal genetic testing represents a technological novelty that tells us if we have cheek dimples or have skin flushing when we drink alcohol. The genetic health risk assessments remain of dubious value for most people. Whether the collective genetic database leads to an important medical discovery that will save some of us in the future remains to be determined.

Suneel Kamath is an oncology-hematology fellow.

Image credit:

Leave a Comment

✓ Join over 130,000 subscribers 
✓ Get KevinMD's 5 most popular stories
✓ Never miss an article
Subscribe. It's free.