by Bob Dozor, MD
The entire human genome has been mapped; well – sort of! We now know every single chemical constituent of DNA and how DNA does its job! Moreover, very soon, we will be able to map every single one of your DNA base pairs for less than $1000. The problem is the “map” is more like a photograph of the earth from space than a Google map. In other words, we can see all the genes but we have only a very rudimentary understanding of how they all work together.
It is also critical to understand that GENES are usually “probabilistic,” and only rarely “deterministic.” GENES are “expressed” in our bodies through complex interactions with each other and the environment, mainly what we eat and breathe. Yes, if you have the Huntington’s chorea (Woody Guthrie) gene, you will get Huntington’s, but most genes don’t determine that you will get a specific disease.
So let’s say you get your genome done, and you find some genes associated -probabilistically- with some serious problem like heart disease or cancer or Alzheimer’s, etc… The point is that having a gene doesn’t mean you are going to get a disease. Our current correlations may not even be relevant to you, because of other interacting genes and environmental factors. Very confusing!!!
One maxim taught to medical students is never to order any test unless the results of the test may affect a treatment decision. At this point very few (but more and more every day) genes are “actionable,” meaning that we know what to do about it. Moreover, going forward, the complexity of such “action-ability” will grow vastly complex. There are 3 -10 million gene variants per individual! As fancy as our current computers and electronic medical records are, they are completely insufficient to deal with the huge amount of data and decision making algorithms. Simply storing your genome requires 10 GB of storage which is 50-fold greater than storing images.
But the vastly greater problem will be manipulating this data, to compare it with known patterns and algorithms to come up with the correct beneficial action. This is almost nothing like the type of medical tests we do today. We don’t yet know what to do with all this data! There will be serious medical misadventures trying to respond to scary gene variants. Harm will occur!
Moreover, there may be grave implications to our healthcare system regarding what insurance companies and the government might do with this data. Current law allows insurance companies to use this data to discriminate. In other words, your GENOME might turn out to be even more prejudicial than the dreaded “pre-existing conditions.”
Without a doubt, GENOMIC medicine will be a source of wonderful and beneficial innovation in health care, but as a profession and as a society we remain pretty far from ready start using this. Yet when the price gets down $1000 undoubtedly many will jump on board, and very likely many will be hurt by premature adoption. Pandora’s box?