The Fallacies of Genomic testing

When physicians (and investors) look at the potential for genomic testing, the emphasis should be on the word potential.  It is an understatement to say that we are just not there yet, and the abundance of data to be analyzed from sequencing 5B bases pairs of DNA (and roughly 20000 genes) will lead to confusion and possibly bad decisions.  So of course, this being America, we are already seeing the “marketing” of gene testing by various cancer centers.  I also note that my local hospital is reinventing itself as a “Genomics Center”. It cannot be an accident that vulnerable and frightened patients are first singled out for tests that are not covered by insurance.

And of course as Mencken said: Noone ever went broke underestimating the intelligence of the American public.

Firstly, it must be understood that DNA sequence does not in and of itself describe all the ails a human being.

Second, we do not know what most of our genes encode for, or how they interrelate, or become modified by the environment.

Third, we are only just beginning the process of what to do if a SNP, or an isolated bad sequence of DNA is detected.

Fourth, there are very specific diseases for which some patients may benefit from gene testing.

As for the use of DNA sequence to treat all cancers, this too is early; BRCA is a good example of a gene that is only of interest in a small percentage of all breast cancers. Other biomarkers for cancer that are protein based (like PSA) are such poor indicators that many have questioned whether they do more harm than good. Only when there is a certain drug for a certain tumor does it currently make sense, and this does not require sequencing the entire genome.

As regards arthritis, I have been involved in a project that showed about 100 proteins aberrant in that disease.  This finding is likely to recur in other diseases,  where it is the sum total of all the “mistakes’ in DNA that eventually result in a real problem. The classic genetic diseases, like sickle cell anemia, will look simple in comparison.

Which is NOT to say that gene testing is not of interest; it just should not be marketed to all patients as important for treatment.  Not yet.