Slow start, but gene knowl­edge comes good

Prom­ises of be­spoke medicines and magic bul­lets’’ for mankind’s big­gest killers may be un­ful­filled, but the true value of ge­net­ics is now be­ing re­alised, re­ports Mark Henderson

The Weekend Australian - Travel - - Health -

IT WAS quite a vi­sion of our ge­netic fu­ture: shortly be­fore the first drafts of the hu­man genome were un­veiled in 2001, Francis Collins spelt out in bold terms what the project he had pi­o­neered would ul­ti­mately mean for medicine. By 2010, sci­en­tists would un­der­stand how genes con­trib­ute to at least a dozen com­mon ill­nesses such as di­a­betes and heart dis­ease, the di­rec­tor of the US Na­tional Hu­man Genome Re­search In­sti­tute said.

A new era of be­spoke medicine lay ahead, in which drugs tai­lored to in­di­vid­u­als’ ge­netic pro­files would re­place the tra­di­tional ‘‘ one­size-fits-all’’ approach. In­sights from hu­man­ity’s ge­netic code were poised to trans­form health care.

Just a year ago, Collins’s timetable looked like fan­tasy. Science had had a ver­sion of the genome for five years, and parts of it for longer, yet had added sur­pris­ingly lit­tle to knowl­edge of the ge­netic roots of com­mon dis­eases. The promised revo­lu­tion had yet to be­gin.

It is now un­der way. The an­nounce­ment in May that four com­mon genes have been re­li­ably linked to breast can­cer is only the latest and most spec­tac­u­lar of a se­ries of re­mark­able re­sults that have led cau­tious sci­en­tists to speak openly of a tip­ping point in their un­der­stand­ing of how DNA af­fects our health.

Scores of genes that cause or in­flu­ence dis­ease are now known to science, but un­til re­cently it has been dif­fi­cult to iden­tify any that con­trib­ute to the great­est causes of ill-health — di­a­betes and heart dis­ease, men­tal ill­ness and most can­cers.

The suc­cesses of the first phase of genome re­search were lim­ited largely to ge­netic mu­ta­tions that have dev­as­tat­ing ef­fects — but for very few peo­ple, such as those that cause Hunt­ing­ton’s dis­ease and cys­tic fi­bro­sis.

Four out of five women with mu­tated BRCA1 or BRCA2 genes will de­velop breast can­cer. Only one wo­man in 500, how­ever, car­ries th­ese de­fects. Their dis­cov­ery has trans­formed screen­ing for a small mi­nor­ity, but it means noth­ing for more than 95 per cent of the 44,000 Bri­tish women, and over 13,000 Aus­tralian women whose breast can­cer is di­ag­nosed each year.

Sci­en­tists have long known, from fam­ily and twin stud­ies, that other ge­netic fac­tors af­fect women’s risk, but hith­erto it has been im­pos­si­ble to pin­point what they are. The same has been true of other dis­eases.

All that is start­ing to change. Af­ter years of tread­ing wa­ter, ge­neti­cists have started to pub­lish a cascade of data about very com­mon ge­netic vari­ants that can sub­tly in­flu­ence peo­ple’s risk of de­vel­op­ing com­mon dis­eases.

The new breast can­cer genes, dis­cov­ered by a team as­sem­bled by Can­cer Re­search UK, have much less of an ef­fect than the BRCA genes: the most dam­ag­ing ge­netic profile raises the life­time risk to about 17 per cent.

But they are much, much more com­mon: one vari­ant is car­ried by one in six women, and the rarest by one in 16.

That dis­cov­ery fol­lowed the an­nounce­ment in April of the first com­mon gene that has been shown re­li­ably to con­trib­ute to obe­sity. Sim­i­lar stud­ies have iden­ti­fied com­mon ge­netic vari­ants that pre­dis­pose to heart dis­ease and type 2 di­a­betes. In June a Well­come Trust con­sor­tium pub­lished de­tails of many more that are linked to seven ma­jor con­di­tions, in­clud­ing high blood pres­sure, bipo­lar dis­or­der and rheumatoid arthri­tis.

Where ge­net­ics was once ca­pa­ble of pin­ning down only rare mu­ta­tions with a cat­a­strophic im­pact, it is now trac­ing vari­ants with smaller ef­fects that are much more wide­spread. You might call it the democrati­sa­tion of the genome.

‘‘ Democrati­sa­tion is a good word for it,’’ said Mark Wal­port, di­rec­tor of the Well­come Trust, the Bri­tish bio­med­i­cal char­ity that has funded much of the key re­search. ‘‘ What we are find­ing now is rel­e­vant to ev­ery­one, not just to those rare fam­i­lies who hap­pen to be ex­cep­tion­ally un­for­tu­nate.’’

Pro­fes­sor Bruce Ponder of the Univer­sity of Cam­bridge, whose team iden­ti­fied the new breast can­cer genes, said: ‘‘ The genes we have seen so far are strong genes for spec­tac­u­lar stuff, but this is the real ge­net­ics that un­der­lies vari­a­tion in the pop­u­la­tion at large. Th­ese are the genes that af­fect how peo­ple re­act to their en­vi­ron­ments, and that in­flu­ence, how­ever slightly, their chances of de­vel­op­ing com­mon dis­eases.’’

This step change has been brought about by new tech­nol­ogy and a change of approach. Tra­di­tional gene hunts have re­lied on a tech­nique called link­age anal­y­sis, in which sci­en­tists study fam­i­lies in which sev­eral mem­bers have de­vel­oped an in­her­ited dis­or­der. DNA sam­ples from in­di­vid­u­als with the dis­ease are com­pared with sam­ples from those who are un­af­fected. Around 100 sub­jects are usu­ally enough, and no more than 300 DNA mark­ers gen­er­ally need to be scanned.

While this is ef­fi­cient, it works only for genes with a large im­pact, rais­ing risk by at least 200 per cent. Most ge­netic vari­ants do not work like that, and to find them sci­en­tists have turned to a new method called whole genome as­so­ci­a­tion.

Such stud­ies use un­re­lated sub­jects, and need to be much larger. The Well­come Trust Case Con­trol Con­sor­tium, which found the FTO gene and pub­lished more re­sults in June, com­pared 2000 peo­ple with par­tic­u­lar dis­eases with 3000 un­af­fected con­trols. As the sub­jects are not re­lated, sci­en­tists must also look at hun­dreds of thou­sands of DNA mark­ers to find any ef­fects.

This approach can find com­mon vari­ants, which raise risks by as lit­tle as 10 to 20 per cent. The draw­back is the vast amount of data that must be pro­cessed — it has be­come pos­si­ble only with the ad­vent of mi­croar­rays or ‘‘ gene chips’’ that can scan hun­dreds of thou­sands of ge­netic mark­ers at once.

It is this that has en­abled the spec­tac­u­lar re­cent ad­vances. ‘‘ It’s been known in prin­ci­ple that whole genome as­so­ci­a­tion stud­ies should be more pow­er­ful, but only in the last two years has it be­come af­ford­able to do them with the large sam­ple sizes we need,’’ said Peter Donnelly, of the Univer­sity of Ox­ford, who chairs the Case Con­trol Con­sor­tium.

‘‘ Since the genome was pub­lished, progress has been slower than peo­ple had hoped, but the pace has zoomed now. Each week and month, ex­cit­ing new find­ings for im­por­tant dis­eases are be­ing pub­lished. That pace will con­tinue. What we knew just a year ago is enor­mously dif­fer­ent from what we know now.’’

This en­hanced knowl­edge is chang­ing the way sci­en­tists think about genes and dis­ease. It is now recog­nised that there are very few con­di­tions that are caused by sin­gle genes with big ef­fects — the in­flu­ence usu­ally comes from dozens of genes, each of which has only a mi­nor im­pact on its own.

‘‘ We are mostly find­ing that for any given dis­ease there are zero, or at best one or two, genes with large ef­fects,’’ said Mark McCarthy of the Univer­sity of Ox­ford, who led the team that found FTO’s ef­fect on obe­sity.

‘‘ Then there is a sprin­kling of genes, per­haps five to 10, with mod­est ef­fects of 10-20 per cent, and there may be many hun­dreds with even smaller ef­fects.’’

For many con­di­tions, in­deed, th­ese smaller ef­fects may be all there are.

Robert Plomin of the In­sti­tute of Psy­chi­a­try in Lon­don, who spe­cialises in be­havioural disor­ders such as autism, doubts whether whole genome as­so­ci­a­tion will even find many mod­er­ate ef­fects in his field.

This means that the genome will rarely pro­vide medicine with ‘‘ magic bul­lets’’ that tackle dis­ease by cor­rect­ing faulty genes.

First, most of the vari­ants that con­trib­ute to a high risk of breast can­cer or di­a­betes are com­mon, and are thus likely to have other im­por­tant func­tions. We would use ge­netic en­gi­neer­ing or gene ther­apy to al­ter them at our peril. Then there is the prob­lem of num­bers. So many dif­fer­ent genes are play­ing a mi­nor role that it is im­plau­si­ble to think we might cor­rect them all.

The real value of this knowl­edge will thus be in un­der­stand­ing the ba­sic bi­ol­ogy of com­mon dis­eases, so sci­en­tists can de­sign bet­ter drugs to tar­get the pro­cesses that our genes set in train. It should also al­low the de­vel­op­ment of tests for gene com­bi­na­tions that cre­ate a sig­nif­i­cant risk.

That would then al­low pa­tients and their doc­tors to ma­nip­u­late a fac­tor that al­most in­vari­ably in­ter­acts with genes to af­fect dis­ease, and which is much eas­ier to con­trol: the en­vi­ron­ment. Women with a high breast can­cer risk could be re­ferred for more fre­quent screen­ing. Those in dan­ger of heart dis­ease might change their diet and ex­er­cise pat­terns. In some cases, drugs will be ap­pro­pri­ate, to change the bio­chem­i­cal en­vi­ron­ment in which genes op­er­ate in the body.

‘‘ What we are go­ing to get out of this in the end is not ge­netic en­gi­neer­ing, but en­vi­ron­men­tal en­gi­neer­ing,’’ Plomin said. ‘‘ A lot is go­ing to be about chang­ing be­hav­iour, about ed­u­ca­tion. It’s not al­ways go­ing to be: you’ve got this ge­netic prob­lem and here’s a pill. Most ge­netic ef­fects are go­ing to be too small for that.’’ The Times

Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.