erable potential, said Evans. Genomic screening will not replace the current metabolic-based screening in the near term, because it remains closer to the phenotype of interest and has much greater specificity. For example, elevated phenylalanine has much more clinical utility than a variant of uncertain significance in the phenylalanine hydroxylase gene. But genomic screening could help resolve ambiguous biochemical results and detect a subset of treatable disorders that do not have good metabolic markers, such as storage diseases, deafness, and neonatal diabetes.
Finally, genomic tests can inform a variety of reproductive decisions, which is an area that Evans believes will “take off tremendously.” Preconceptual carrier screening (see Chapter 4) is currently recommended for a few disorders, but these have been chosen essentially based on cost and mutation prevalence. Screening is conducted for cystic fibrosis or Tay-Sachs disease because it is affordable and because reliable testing is available, not because Tay-Sachs is any worse than, for example, Batten disease, said Evans. “That is not what couples really want to know. They want to know if [their] child is likely to have a really bad, untreatable disease.” Genomic sequencing can help address these concerns by potentially being used to screen for all serious diseases.
Preconceptual carrier screening for serious diseases could have “a potentially profound and very welcome impact on family planning,” said Evans. Some people will treat such information as highly actionable. Others will regard it as morally problematic. The formulation of policy in this area will be difficult, Evans warned.
Effectively harnessing genomic screening faces significant challenges. Because of the large number of bases in the complete haplotype genome, even an accuracy of 99.99 percent will produce 300,000 errors per patient, said Evans, though accuracy will gradually improve.
In addition, each person has about 4 million genetic variants, and our current understanding makes their interpretation difficult. Should information about all of them be gathered or stored? As genome sequencing becomes more accurate and cheaper, it may be more practical to do sequencing when the information is needed, Evans said.
Another significant challenge, said Evans, is that the genome is an unpredictable—and not necessarily friendly—place. For some people, whole genome sequencing will uncover things they were not looking for and might not want to know. Some people will discover that they are at high risk for untreatable and horrific conditions, such as fatal familial insomnia, Huntington’s disease, or early-onset Alzheimer’s disease. The potential for returning information when there is no medical action that can be taken