VARIATIONS IN A MINOR KEY: SOME THOUGHTS ON PRENATAL TESTING IN AN ERA OF WHOLE GENOME SEQUENCING

James Watson is many things – geneticist, Nobel laureate, agent provocateur – but in the realm of psychiatry he is first and foremost the parent of a son with schizophrenia.  So when he spoke in 2007 at the World Congress of Psychiatric Genetics, it was as a family member, albeit a family member with an unusually good grasp of the science.  And it was as a family member that he exhorted the scientists in the audience to keep up the good work, so that “someday we could identify those individuals destined to suffer from mental illness in utero, and weed them out.”

How often do you hear an audible gasp in the midst of a plenary talk?  The dismay and the indignation were palpable.  Researchers throughout the day interrupted their talks on GWAS to express in the strongest possible language that the goal of their work was to understand the pathophysiology of the disease and perhaps to aid in diagnosis – not to provide pre-symptomatic risk  assessment and not – no, never – not to be used prenatally.

“But if this is what families want,” I asked one speaker later that day.  “How do you propose to restrict testing, once the means to test is available?”

“They can’t,” he replied.  “They must not.”

Ah.  Of course.  They must not – I will pass that along.

Five years later, it is not GWAS but whole exome sequencing and whole genome sequencing providing all the buzz at conferences.  Solving the diagnostic odyssey!  Revolutionizing cancer treatment!  Ushering in an era of personalized medicine!  It’s very exciting.  Prenatal testing is rarely mentioned, and then only in passing – while prognosticators sing happy songs of a not-so-far-off day when every baby will be sequenced at birth.

Sequenced at birth?  Will it even be necessary?  Maybe Mom and Dad have baby’s DNA already, on a hard drive or a memory stick or downloaded onto their cell phones along with the ultrasound pics.

This is not the genome sequencing story you are seeing in the papers or the blogs.  It’s not what researchers are excited about.  The ones we hear are all about science journalists getting their DNA decoded and setting off on odysseys of self-discovery that involve hours of consultation with clinical and academic superstars who donate their time. We hear about kids with strange constellations of symptoms finding answers after years of disappointment.  Those are heartwarming tales: anecdotal and difficult to imagine at scale, but hopeful and exciting nonetheless.  But there is another theme playing, in a minor key, and I hear it faintly, hidden beneath the violins and the trumpets.

I hear it, an unspoken question, when we debate the utility of genomic information.  What does to mean to say that information is actionable? (Prevention? Treatment? Cure?  Prenatally, there is only Yes or No.)  Can patients handle uncertainty?  (And what will we lose, when pregnancies are terminated just to be on the safe side?)  Doesn’t everyone have the right to know what is in their own DNA? (The information is available – why not use it?  What could possibly go wrong?)

Whatever tests are available postnatally will find their way into prenatal use.  The gateway technologies – PGD, cell-free fetal DNA testing – are in place. And there is no use saying, “they can’t, they won’t, they shouldn’t” because they can and they will – and sometimes they should.  There will be good uses too: success stories and disasters averted.  A blanket “no” is not an option, and granting anyone authority to pick and choose which uses are worthwhile vests altogether too much power in the hands of any one person, or profession, or bureaucratic entity.

The same tests can be done before or after birth, but the experience is entirely different.  Uncertainty after birth is an opportunity.  The least useful information is that which will absolutely come true, no matter what you do.  Uncertainty before birth is a crisis.  Anyone who has ever discussed a variant of uncertain significance with a pregnant mother can tell you that.  But what are the chances there will be developmental delay?  Are you certain that the heart will be affected? How sure are you that this means anything?  Not nearly sure enough.  Please understand that.

In general, notions of genetic determinism increase the likelihood that genomic testing will have negative consequences.  Fatalistic attitudes about the power of genes could lead people to overestimate the meaning of elevated risks and underestimate the meaning of reduced risks.  Anxiety, stress, missed mammograms – you have heard this before.  Shrug.  People are grown ups.  They will figure this out.  Information is power.

But we are in a whole new universe trying to reconcile underpowered and often misunderstood predictive testing in the context of prenatal use.  So please, in telling tales of all the wonderful things that genome sequencing will do, save space for a mention of what it cannot do.  Make sure they understand that there are great wide cracks in our crystal ball.  Do not oversell the value of genotype in the absence of phenotype.  Remember that in the end neither researchers nor physicians nor genetic counselors will dictate how this new technology will be used.  Others will make that call, and we will be in the choir, singing songs of praise laced with sorrow.

Follow me on Twitter:

https://twitter.com/laurahercher