This is the warning being issued by medical geneticists who have assessed the outcomes of some of the tens of thousands of DNA tests carried out every month in the US as part of the world’s largest screening programme for cystic fibrosis (see Cystic fibrosis: key facts”, below).

Some of the companies carrying out the tests blame doctors for misinterpreting complex results or requesting the wrong tests. But geneticists say a few companies are at fault too, for failing to stick to clinical guidelines.

And cystic fibrosis is just the first of many diseases for which genetic testing is likely to become routine. If mistakes are being made already, experts warn, what will happen as screening becomes more and more common? Government regulation of genetic testing in many countries is poor, they say.

One in four chance

With cystic fibrosis screening, the aim is to discover early in pregnancy if a woman and her partner carry any of the mutations that cause the disease. If blood or saliva tests reveal that both are carriers, they have a one in four chance of having a child with cystic fibrosis. If a follow-up DNA test on the fetus confirms that it has inherited two defective copies of the CF gene and will thus get the disease, couples face a tough decision whether or not to terminate the pregnancy.

Until recently, antenatal clinics in the US offered CF screening only to couples who already had affected children or family members. Most countries still offer only this limited screening. But two years ago doctors’ groups recommended that cystic fibrosis screening be offered to all pregnant women.

The result has been a huge hike in the number of CF tests carried out by diagnostics labs in the US – and growing concerns about the reliability and interpretation of some of these tests. “The volume of tests has exploded, and front-line service providers are having some difficulty with the hard mutations that are difficult to interpret,” says Michael Watson, executive director of the American College of Medical Genetics (ACMG).

Genetic screening for diseases is far from simple because a mutation in any of the tens of thousands of DNA letters in and around a gene can cause a disease. And while one mutation may be harmless, another may be deadly. As if that were not complex enough, some mutations are dangerous only if other mutations are present too, which is what has caused the problems with cystic fibrosis screening.

900 mutations

Since the CF gene was discovered in 1989, researchers have discovered more than 900 distinct mutations in the gene or nearby DNA that can cause some or all of the symptoms of the disease. Routinely screening for every mutation – most of which are incredibly rare – would be very costly and time-consuming, so most commercial test kits only look for the 25 to 35 mutations responsible for the overwhelming majority of cases.

One common mutation is called 5T. Though present in as many as one in 20 people, it contributes to cystic fibrosis only if there is a second, much rarer mutation called R117H in the same gene.

By itself, the 5T mutation does not cause life-threatening illness, and the consensus among doctors’ groups and ethicists is that its presence in a pregnant woman’s DNA does not justify carrying out risky fetal tests such as amniocentesis or chorionic villi sampling. And no mainstream group supports terminating a pregnancy if these tests, done for whatever reason, reveal 5T without R117H.

But according to Watson, there have been both unnecessary fetal tests and terminations. The extent of the problem is unclear: there are no official government figures for how many women have been tested for CF mutations in the US since population screening began in October 2001, or what the outcome has been in each case.

But at a conference for genetics professionals in March, the ACMG put out an unprecedented alert on an overhead slide warning delegates that “over 20 prenatal tests were performed with 5T alone and with terminations occurring”. Despite repeated requests, the ACMG did not reveal any more details of these cases. But Watson told żěè¶ĚĘÓƵ that it knows of “at least 150 prenatals that have been done that perhaps should not have been done”.

Worrying cases

Other worrying cases have been disclosed by the biggest gene testing company in the US. In an abstract published at the same conference, scientists from Quest Diagnostics of California report having so far received 150 fetal samples for CF testing since October 2001 when population screening began. Of these, 41 were submitted by clinics after one parent tested positive for the 5T mutation. None appeared to carry the accompanying mutation required to put someone with 5T at risk of having a CF child.

Yet the clinics still decided to screen fetal samples obtained by amniocentesis. And while many of the women would have had an amniocentesis for other reasons, in 12 cases the discovery of the 5T mutation was the sole reason for the procedure, which has a one in 200 risk of triggering miscarriage.

None of these fetal tests led to terminations, says Charles Strom, medical director of genetics at Quest Diagnostics. But the fact that they happened at all, he says, illustrates the confusion over the significance of the 5T mutation. “This is not a unique situation,” Strom told żěè¶ĚĘÓƵ. “I think you can say with some certainty that this is going on throughout the industry.”

Possible explanations, according to Strom, include doctors ordering the wrong tests, genetic counsellors misinterpreting the parental 5T tests, or parents getting the right counselling but becoming so anxious that they request a fetal test anyway.

Strom says that when his company realised there was a problem, it stopped providing doctors with the results of the 5T part of the company’s standard CF screening test. Doctors who want this information from Quest must now request it specifically. And when doctors submit fetal samples solely on the strength of a 5T mutation, Strom says that the company “does the test and has a genetic counsellor call to make sure the patient has been correctly counselled”.

Broken guidelines

But others feel such tests should not be done at all. “These couples are being found with the 5T and no other mutations, and they’re getting all confused and worried,” says Wayne Grody, a medical geneticist at the School of Medicine at the University of California in Los Angeles. “It’s disturbing.”

Grody points out that any population screening for the 5T mutation in pregnant women runs counter to clinical guidelines. Rules drawn up by the ACMG and the American Academy of Obstetricians and Gynecologists state that pregnant women should be screened for the 5T mutation only if they first test positive for the R117H mutation.

“The recommendations as written could not be more explicit on this point,” says Grody. “Why a lab would knowingly violate that, I can’t figure out.”

Yet żěè¶ĚĘÓƵ’s survey of company websites reveals that at least three companies in the US include 5T in their standard CF test. It is not clear whether these companies automatically tell doctors and patients whether the 5T mutation is present or if, like Quest, they withhold the information unless specifically asked for it.

Either way, as far as Grody is concerned, the companies are breaking the guidelines, which state that you should not even test for 5T until R117H is found.

Raised costs

But following the guidelines to the letter could raise companies’ costs. A big part of the $200 or so they charge for a CF test is the cost of the test kit. Waiting until someone tests positive for R117H before screening for the 5T mutation would often mean using up two test kits instead of one.

Another complication is that there is a separate group of patients with a good reason to know whether they carry the 5T mutation: infertile men. Men who inherit two copies of the 5T mutation, or one copy plus a CF mutation, often lack a working vas deferens, the tube that stores and carries sperm.

If the 5T mutation is found in an infertile man, it tells the urologists and IVF clinics treating them that there is a good chance that healthy sperm can be taken directly from the testicle. “It would be unfair to urologists not to offer a test for 5T,” says Strom.

But Strom admits companies currently cannot guarantee that the 5T form of the CF test that is intended for infertile men will only be used by this group. And if doctors and prospective parents discover – by deliberately or accidentally ordering this test during pregnancy – that a male fetus has the 5T mutation and may grow up to be infertile, they may opt to terminate the pregnancy.

“They may decide that this may be a child they don’t want to bring into the world,” says Strom. “It’s not up to us to tell them what reasons they’re allowed to terminate a pregnancy.”

And the link with infertility means that companies that test for 5T as part of CF screening but do not reveal the results risk being sued decades later when a boy grows up and discovers he is infertile. Quest has decided to proceed nevertheless but other companies are not prepared to take the chance.

The US cases also come at a time when countries such as Britain are considering following the US example and setting up their own national screening programmes. As yet, the written guidelines covering CF testing in Britain make no stipulations about when labs should and should not screen women for the 5T mutation.

Cystic fibrosis: key facts

• About 30,000 Americans, 3000 Canadians and 20,000 Europeans have cystic fibrosis

• The disease is caused by defects in a gene on chromosome 7 that codes for the cystic fibrosis transmembrane regulator protein, or CFTR

• CFTR’s job is to move chloride ions and water through the membranes of cells lining the passageways of the body. If it is defective, mucus accumulates in the intestines and lungs, causing malnutrition, breathing difficulties and eventually permanent lung damage

• The most common mutation in the CFTR gene accounts for up to 80 per cent of all cases. People with this classic form of the disease now live to age 30 or more. But the CFTR gene is very large, and at least 900 mutations can cause the disease. Symptoms vary greatly depending on which mutations someone inherits

• A child only gets CF if they inherit a defective gene from both parents. So the first step in testing is to find out if both parents have a mutated copy: around one in 29 white people carry a single defective gene

• Only if both parents are carriers is the fetus tested to see if it has inherited two defective genes. But not all CF mutations can be detected and testing, by amniocentesis or chorionic villi sampling, is risky for the mother and fetus

• Some parents decide to terminate the pregnancy if tests show the baby will have cystic fibrosis. But in many cases tests cannot determine how serious the disease will be

However, the scientists may get little chance of a breather. Opponents of Britain’s farm-scale trials have chosen the lull before the first findings are published in a few month’s time to mount the most detailed attack on the science yet.

The trials were set up to address fears that the broad-spectrum herbicides used with many GM varieties would harm farmland wildlife. But a 47-page dossier of arguments and allegations, seen in advance by żěè¶ĚĘÓƵ, claims the multimillion-pound experiment cannot succeed in delivering a definitive answer.

The trials are likely to play a key role in determining whether – and on what basis – the British government sanctions commercial growing later in 2003. For now, the findings remain strictly under wraps. But based on the trials’ methods and first-year pilot observations, which are already in print, campaigners claim the prospects for a clear verdict are bleak.

“That the trials look set to produce uncertain results is not a reflection on the scientists involved,” says Pete Riley, whose team at Friends of the Earth compiled the report. “Rather it highlights the inherent problems of embarking on politically motivated science.”

Field furore

Opposition to the trials is not new. From the start activists periodically ripped up trial crops while others claimed farmers were biasing the outcome by treating GM fields with less herbicide than would be used commercially – a charge the trial scientists rejected. Now, with the endgame in sight, campaigners are keen to shift the focus onto what even some neutral experts see as the experiment’s potential Achilles’ heel: its statistical power.

The trials involved farmers growing both conventional and GM varieties of sugar beet, maize or oilseed rape (canola) in neighbouring fields. There were up to 25 sites per crop per season, which researchers would regularly visit to count weeds, beetles and other biodiversity “indicators”. The goal was to discover if the GM fields held significantly less, or more, wildlife than those with conventional crops. But what counts as “less” or “more”?

For weeds and insects, the scientists designed the trials to be sensitive enough to have an 80 per cent chance of detecting 1.5-fold differences between conventional and GM fields. However, the report claims this sensitivity target is unlikely to be met for every species because of “noise” in the data.

Based on the trials’ own pilot observations, the report claims that levels of some key indicator organisms, including beetles and broad-leaved weeds, are likely to vary from field to field by far more than the 50 per cent margin that the trial allows for. If so, that could make detecting a 50 per cent difference between GM and non-GM fields impossible even if the difference is there.

Pure speculation

The report also takes issue with the 1.5-fold target difference itself, arguing that it is set too high. Previous research on the impact of herbicides on grey partridges found that much smaller differences in weed numbers – as slight as 13 per cent – were ecologically significant.

Les Firbank, who has coordinated the trials from the Institute of Terrestrial Ecology in Cumbria in north-west England, rejects the criticisms. “They’re speculating on whether the experiment has been capable of delivering the stated power. That won’t be answered until the data are published,” he says.

Firbank says that the sensitivity target was never intended to be met for every species. “The interpretation comes not from looking at each species in isolation but from combining results from different species and looking for patterns.”

Peter Green, president of Britain’s Royal Statistical Society, says that while many of the report’s points about statistical power are valid, such problems are not unique to these trials and there are well-established ways of handling them. “The danger with an issue that is so highly charged politically,” he says, “is that some people will seek black and white answers when they are not attainable.”

When Mark Anderson encounters his first Kanuri tribesman on the southern fringes of the Sahara it makes a compelling television moment. All his life, the 23-year-old black Londoner has wondered about his African roots. Now his quest seems over. He is face to face with the people and landscape from which, generations ago, his ancestors were wrenched and sold into slavery.

And it’s all thanks to a modest sliver of DNA called HVS-1. A genetic test, arranged by the makers of the flagship BBC programme featuring the reunion, reveals that Anderson’s version of HVS-1 resembles the corresponding sequences from some Kanuri people stored in a laboratory database in Washington DC. Culturally the tribesman is from another world; genetically he is kin, a living symbol of this one Londoner’s African origins.

Or is he? With ancestral pride on the increase, more and more people, white and black, are being encouraged, like Mark, to turn to genetics to fill in the blanks in written records. Several firms already offer “DNA genealogy” tests similar to those used in the programme. But behind the scenes geneticists are divided over just how meaningful and reliable the tests are.

“The science has been over-interpreted,” says Bert Ely, an expert on African-American DNA at the University of South Carolina, Columbia. “Just because you find a match doesn’t mean the person came from that place.” Even geneticists who carried out tests for the TV programme point out the limitations. “The danger is that people think they’re being told where they came from,” says Mark Jobling of the University of Leicester in Britain.

Sailing down the generations

So what’s the problem? At face value the science seems straightforward. Ancestral tests focus on two types of genetic material: the mitochondrial DNA, of which HVS-1 forms part, that we inherit solely from our mothers; and/or the Y-chromosome that boys get from their fathers.

Most of our chromosomes become jumbled up as they are passed on, but these two types sail down the generations undisrupted, providing stable havens where distinctive DNA sequences and markers can settle and take root – first within families, then, as those families expand, within clans and ethnic groups.

This is why cataloguing the various types of mtDNA and Y chromosomes found across the world has proved so useful for reconstructing the major migrations and dispersals of our deep past. But painting broad-brush pictures of population shifts is one thing. DNA genealogy uses the same tools to trace the ancestral origins of individuals – and that is more controversial.

First, not everyone gets a precise answer. Using today’s tools and databases, it turns out that only about 1 in 10 people of African-Caribbean origin have mtDNA sequences that are sufficiently distinctive to be linked to particular areas of Africa. The overwhelming majority can expect to receive answers no more precise than “west Africa”, which given the history of the slave trade and the fact that Africa has hundreds of ethnic groups is perhaps not telling them very much.

Probabilistic estimate

Why the imprecision? Today’s ancestral tests, say critics, look at comparatively little DNA. Another factor is that very few of the mutations or markers they detect are exclusive to any one geographical region or ethnic group. Moreover, even those sets of mutations and markers that are associated with particular ethnic groups or regions represented in DNA databases do so on a statistical basis: any information they can provide about personal ancestry is a probabilistic estimate rather than a clear-cut verdict.

A further worry is the patchiness of today’s databases. By no means all ethnic groups have so far been sampled for DNA, and vast areas such as central Africa are poorly represented. A DNA signature that looks typical of a particular west African group today might look a little less so once more samples have been collected. By then the owner of the signature may already have invested time, emotional energy and perhaps even money bonding with the region and its people.

Even a reliable DNA match traces the ethnic or regional origins of just one ancestor out of many, as the programme itself acknowledges. Going back just 10 generations and ignoring cousin marriages, each of us has about 1000 ancestors; 25 generations ago the figure is more than 30 million. A mtDNA test tells you about just one (the mother of your mother of your mother of your…), while a Y-chromosome test picks out one more (the father of your father of your father…). The bulk of your genetic and cultural heritage is simply not on the radar.

In contrast to medical genetics, genealogy testing is a largely unregulated commercial frontier. That might not seem to matter much, since it is easy to think of these tests as harmless diversions for people curious about their ancestors. The problem is that for many people, not least the descendants of slaves, the question of origins is a deep and sensitive issue.

In the 1970s, the phenomenally successful television series Roots reawakened the curiosity of many African Americans about their family history before slavery. Let’s hope that “Roots II, the genetic remake” does not backfire or disappoint.

Motherland – A Genetic Journey will be shown in Britain on BBC2 on Friday 14 February at 2100.