Embed Size (px)
Citation preview
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 1 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
In Depth Human body Medicine
Can we ever fully trustpersonal gene tests?
Kits sold to consumers that predict healthKits sold to consumers that predict healthrisks from our genes have been hit by majorrisks from our genes have been hit by majorconcerns over their reliability. Nic Flemingconcerns over their reliability. Nic Fleminginvestigates why.investigates why.
. � . � . �
Home News Sport Weather Shop Earth Travel
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 2 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
"Neurological conditions, 60% probability," anurse in green scrubs reads in a monotonevoice. A young woman cradles her newbornwhile an anxious new father looks on."Manic depression, 42% probability.Attention hyper-deficit disorder, 89%probability. Heart disorder, 99% probability...life expectancy 30.2 years."
The scene is from the 1997 sci-fi classicGattaca, in which a man refuses to acceptthe rules of a future society condemning himto an underclass only fit for menial workbecause of his less-than-perfect geneticinheritance. Back in early 21st Centuryreality, gene tests don’t yet feature in jobinterviews, thankfully. Yet the idea that thecode of life within us – DNA – can predictour long-term health prospects from themoment of birth is a powerful andpersuasive one.
Since the mid-2000s several companieshave launched tests for the general public,
. . .
By Nic Fleming29 January 2014
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 3 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
offering consumers cheap, simple geneticscreening for a wide range of healthconditions. Spit into a tube or take a samplewith a cheek swab, send it off for testing andfor as little as $99 you could receive inreturn information ranging from ancestrallinks to risks of specific diseases, accordingto the manufacturers.
But all has not been plain sailing for thisnascent personal gene testing industry. In2012, two of the three big players,decodeMe and Navigenics, pulled out of themarket after being acquired by othercompanies. And in November last year, thethird and arguably most high-profilecompany, 23andMe, which says it has morethan half a million customers, stoppedoffering new tests after the US Food andDrug Administration (FDA) said it hadfailed to show its health predictions wereaccurate.
"That evidence is simply not available, so Iwould not be surprised if this marks the endof direct-to-consumer genetic testing formost common diseases," saysepidemiologist Cecile Janssens, of EmoryUniversity, in Atlanta, Georgia.
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 4 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
If this is the case, then why do DNA testssold to consumers appear to be asunreliable as US regulators say? And ifthese tests can't be trusted, where does thatleave the broader hopes that our newfoundability to read our individual genetic codecan help us live longer, healthier lives?
Screen test
Doubts about whether the industry coulddeliver what it promised surfaced early. In2008, for example, I submitted samples ofmy DNA to three companies as part of aninvestigation for the Sunday Timesnewspaper. While deCODEme informedme my risk of developing exfoliationglaucoma, which causes loss of vision, was91% below average, 23andMe said I was3.6 times more likely than average to get it.British company GeneticHealth said I hadfour times the average risk of developingAlzheimer's disease by my late 80s, whiledeCODEme described the risk as 74%above average. I was given conflictingpredictions for other conditions includingage-related macular degeneration, Crohn'sdisease and hypertension.
I wasn’t the only one. In 2009, pioneering
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 5 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
human genome sequencer Craig Venter andseveral colleagues sent samples to23andMe and Navigenics, and also receivedsimilarly contradictory results.
Critics have argued that without guidancepeople are liable to misinterpret their results,potentially leading to them feeling eitherunnecessarily alarmed or falsely reassured.
"What if a woman who learns she does nothave the BRCA1 or 2 mutations that makeher at high risk of breast cancer decides shewill no longer go for her mammograms,even though for the average Americanwoman not having them only cuts the riskfrom around 12% to 11.95%?" says HankGreely, Director of Stanford University'sCenter for Law and the Biosciences.
Others respond that people have the right totheir genetic data, and that access to it canimprove health prospects and even savelives. Last year, Robert Green andcolleagues from Harvard Medical Schoolcompleted a survey of 1,800 customers whohad bought direct-to-consumer genetictests. Although the results haven't yet beenpublished, he says preliminary findingssuggest that around four in 10 claim the
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 6 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
tests motivated them to make changes suchas improving their diet and doing moreexercise, while only 1% reported alteringprescription treatment without consulting adoctor.
"It's certainly an industry that should becarefully examined and potentiallyregulated," says Green, who set out hiscase against the FDA's decision in thejournal Nature earlier this month. "But theemerging evidence suggests thatsometimes people use their results tomotivate lifestyle changes,” he says. “In avery few cases individuals and their relativeshave learnt that they were at increasedbreast and ovarian cancer, and arguably insome of these lives have been saved."
It's an interesting clash of perspectives, butone which may be beside the point now thatthe FDA has slapped a large question markover the basic accuracy of the testpredictions. After being ordered by the FDAto halt its testing service, 23andMe saidcustomers would have access only to itsresults from ancestry and traits like haircolour, instead of disease risks, while itseeks FDA approval for health-relatedresults.
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 7 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
Complex influence
One simple reason genetic tests are limitedin their predictive powers is that manycommon conditions are influenced as much,if not more, by environmental and lifestylefactors than genetic ones. A summary oftwin studies published by Emory’sJanssens in 2010 identified that someconditions direct-to-consumer tests offerpredictions for are highly heritable, such astype 1 diabetes at 88%, schizophrenia at81% and Alzheimer's disease at 79%, butthat most have low-to-moderate heritability,such as prostate cancer at 42%, depression37% and type 2 diabetes 26%. "There is noutility at all in doing a genetic test for adisease with low heritability," says Janssens.
Even if a disease is highly heritable, thatdoes not necessarily mean DNA tests canmake an accurate prediction of risks. Formany conditions, scientists have found thatlarge numbers of genes may have smalleffects, and the way these interact with eachother and non-genetic causes such assmoking, diet and physical activity levels isfar too complicated for the simple modelsused by the tests to untangle. "In mostcommon diseases such as type 2 diabetes
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 8 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
or heart disease, there are many riskfactors, both genetic and non-genetic, thatplay a role," says Janssens. "These caninteract in unique and complex ways indifferent people, and it's difficult to capturethis complexity in a model."
This view is supported by research carriedout by John Burn, Professor of ClinicalGenetics at Newcastle University, wholooked into the effects of six genes on therisks of babies born with spina bifida. "Theeffects of these variants didn't just addtogether in a simple way," says Burn. "Therewere some situations where a variant was adisadvantage unless it was in the presenceof another variant when it became anadvantage. Companies that have beenoffering genetic tests for common diseasesimply they understand these relationshipsbut in fact they grossly underestimate theirimmense complexity."
On top of these problems many cases of thedisease-associated markers (or SNPs asthey are known) discovered so far onlyaccount for a small part of the knownheritability of those conditions, and there islittle consensus about why this might be."The case of the missing heritability is the
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 9 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
biggest issue right now in human genetics,"says Greely.
In research published last summer,Janssens and colleagues found variations inthe predictions produced by 23andMe,deCODEme and Navigenics were due inpart to them using widely different numbersof SNPs for the same conditions, andsometimes using drastically differentaverage disease risks. For coeliac disease,for example, deCODEme had an averagepopulation risk of 1% – eight times higherthan that used by 23andMe and 16 timeshigher than Navigenics. Variations in thedisease predictions provided for individualstherefore might be more to do with whichcompany they decide to do their tests withthan their genes.
The health risk reports 23andMe providesfor its users do acknowledge the role ofenvironmental factors in diseasesusceptibility and also state the informationprovided "is intended for research andeducational purposes only, and is not fordiagnostic use". However in its Novemberletter to the company, the FDA made itclear it is now among those who see thisdisclaimer as disingenuous when sitting
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 10 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
alongside statements such as the healthreports being a "first step in prevention"enabling users to "take steps towardmitigating serious diseases".
Risk assessment
The question is whether more advancedgenetic tests will provide a clearer picture.Direct-to-consumer genetic tests, and othertypes of DNA analysis, use "SNP chips"which can identify up to one million geneticvariants. New techniques mean that readingall six billion bases of the human genome –known as whole genome sequencing(WGS) – is getting cheaper and quicker.The current cost is around $5,000. Somepredict that once this drops to around$1,000, it won't be long before everyone hastheir genomes sequenced. In DecemberBritish Health Secretary Jeremy Hunt calledfor the country to become the first in theworld to routinely sequence everyone'sgenomes at birth.
"I think we're moving from a SNP worldwhere the associations are weak andquestionable, to a whole genome worldwhere there will be more strong andclinically valid associations," says Greely.
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 11 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
"For the common diseases for which theknown genetic contribution to heritability islimited, whole genome sequencing isn'tgoing to make any difference, but it could dofor the thousands of rare monogenicdiseases as well as in other non-rareconditions for which there are strongassociations like Alzheimer's. That maymean the medical implications are notenormous for most of us, he says, “but if itgets cheap enough I suspect that goodhealthcare systems will get whole genomesequences on their patients in another 10years or so."
Sequencing everyone's genomes couldimprove existing disease screeningprogrammes such as those for breast andbowel cancer. Those identified as being at agreater genetic risk could be tested morefrequently and at an earlier age, while thoseat lower risk could be tested less often,making it possible to pick up illness soonerand save money. "We might, for example,give all women a cancer predisposition testat the age of 30," says Burn.
Another benefit of universal DNA analysis islikely to be significant reductions in childrenborn with rare "recessive" disorders, which
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 12 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
only occur when a child inheritsmalfunctioning genetic variants from boththeir mother and father. These illnessesoften come as a complete surprise to thefamilies concerned because those whocarry just one copy of the geneticabnormality are "carriers" but do not showany sign of ill health. About one in 25Caucasians of northern European descentcarries the gene variant for cystic fibrosisand around one in 12 African-Americans isa carrier of sickle cell anemia, for instance.In a world in which we all know our carrierstatus, couples who know their unbornoffspring are at risk could undergo IVF andselect embryos without these conditions.
With more and more people seeking to findinformation about their reproductive risksbefore or during pregnancy, many aredeciding not to wait for the advent ofwidespread genome sequencing. Themarket for "carrier screening" has beengrowing rapidly in recent years, especially inthe US. San Francisco-based companyCounsyl, for example, sells tests that lookfor some 400 mutations and that causearound 100 genetic disorders for $599.
Non-invasive prenatal DNA tests involving
7/19/18, 1:41 PMBBC - Future - Can we ever fully trust personal gene tests?
Page 13 of 17http://www.bbc.com/future/story/20140129-personal-gene-tests-what-next
obtaining foetal DNA from pregnantwomen's blood are also gaining inpopularity. These are safer and do not carrythe risk of miscarriage that come withtraditional screening methods such asamniocentesis. They are currently beingused for detecting Down’s syndrome and forearly sex detection of X-linked diseases likeDuchenne Muscular Dystrophy andhaemophilia, however they have thepotential to be used for many moreconditions in future.
Both science fiction such as the film Gattacaand current problems faced by 23andMe etal provide important warnings about thedangers of overly simplistic approaches togenetics. These are valuable lessons,however they shouldn't detract from thedramatic impact this is already having on thequality and length of many human lives, andits potential to power improvements for evenmore of us in the future.
If you would like to comment on this, oranything else you have seen on Future,head over to our Facebook or Google+page, or message us on Twitter.
