How will we handle the rapidly approaching flood of genomic information on individual patients and consumers?
This commentary in the Genomics Law Report’s ongoing series What ELSI is New? is contributed by Hank Greely, Stanford Law School.
The cost of advanced genome analysis is falling rapidly. The fabled “thousand dollar genome” is less than five years away. It will be as cheap to order a whole genome as to order any one genetic test. As the price falls, many people will buy “their genomes,” or, at least, information on billions of base pairs, millions of SNPs, and unknown numbers of copy number variations and translocations. What will we do with that information?
Some may think we already are facing this problem with the “consumer genomics” companies like Navigenics and 23andMe, but those SNP-chip-based companies have the “advantage” that they produce weak information, linked to only small variations in disease risk. Fully detailed genomic information will unearth something frightening in all humans’ genomes, for themselves or for their (existing or possible) children and other family members.
Patients need to understand the true implications of this genomic information, but how will we accomplish that? Some primary care physicians may be able to make sense of information about a few famous disease-related genomic variations – perhaps well-known mutations in BRCA 1 or 2, the expanded CAG repeats of Huntington’s disease, or the most common cystic fibrosis-associated mutations of CTFR1. Most won’t know even that; none have a clue about the thousands of rarer genetic disease associations, let alone the hundreds of published pharmacogenomic associations. Even clinical geneticists and genetic counselors will not know all the important variations, and in any event, those professionals are far too rare to handle any significant part of the demand. Even if we had the capacity to provide counseling, our current regulatory scheme does not require that genome consumer get any professional explanation – good, bad, or indifferent – to genome consumers. The age of cheap full genomes is almost upon us – and we are not close to ready for it.
Comments
One Response to “How will we handle the rapidly approaching flood of genomic information on individual patients and consumers?”Trackbacks
Check out what others are saying about this post...-
[...] [...]
Great points; I appreciate that you recognized that there is a resource limitation not only in the number of trained genetics professionals, but in the ability of any single person to keep abreast of the hundreds of genetic associations being published on a weekly basis, much less the algorithms that will increasingly be required to integrate and interpret them. Discussions must take into account this resource limitation when considering the restriction of access to personal genetic information: should professional consultation be required for an individual to obtain their genetic information for health-related reasons, even if lack of availability might equate to a severe limitation on access? How about for ancestry (though even ancestry-related genetic information can have health consequences)? Of course, you know where 23andMe stands on these questions. :)
I would also like to point out that genetic information does not need to be frightening to be useful. You mentioned pharmacogenomic associations–genetic associations that can affect the efficacy, safety, or toxicity of a drug and thus can have effects on the treatment of disease (and some of which are available on genome-wide SNP arrays). Francis Collins said in his recent NEJM interview (http://content.nejm.org/cgi/content/full/361/14/1321) that pharmacogenomic information is most useful when a person already has access to the information before it is needed (i.e. at the time a drug is prescribed). Perhaps it may take the advocacy of individuals who have received their pharmacogenomic data and would like to use it to convince health care professionals of the utility of this information (provided that it is accurate and valid, of course).
There’s also the question of reimbursement for tests purely for screening/predictive purposes. A representative from Palmetto Government Benefits Administrators made very clear at DXcon09west that Medicare would not reimburse tests that are not being used in the active treatment of disease, such as pharmacogenomic tests–does that mean that you’d have to pay out of pocket to have your pharmacogenomic data in advance of being prescribed a drug? Will this be a further barrier to adoption?