Non-invasive pre-natal testing (NIPT) might be the new gold standard for Downs Syndrome screening, but as I found out while pregnant, when it comes to other genetic conditions, this £450 test can deliver more harm than good
I was looking for certainty at a time when there is none. At 41, after enduring five difficult years in which my husband battled and survived two kinds of cancer, I was pregnant after just one round of IVF with sperm that we had frozen prior to his chemotherapy.
We were thrilled but cautious. Because of my advanced age we knew our baby was at a higher risk of Downs Syndrome and after the previous five years, we knew we did not have not have the emotional reserves to raise a child with intellectual disability.
During the IVF process I had heard about the Harmony Test, one of the Non-Invasive Prenatal Tests (NIPT) offered by pharmaceutical companies like Roche through private laboratories and clinics, which are considered to be a more accurate and safer way to screen than the older methods of ultrasound and amniocentesis, the latter of which can cause miscarriage.
This test is due to be brought onto the NHS in England for Downs Syndrome towards the end of 2018.
The test was invented for Downs, but has since been extended to screen for other genetic abnormalities, including abnormalities in the sex chromosomes, and at eleven weeks pregnant, I paid £450 privately at Glasgow Centre for Reproductive Medicine (GCRM), an IVF clinic for the Harmony NIPT.
During the appointment I was asked to tick a box if I wanted to know the gender.
I hesitated. I was not in a hurry to know the gender, it felt like peeking under the hood.
“If you don’t tick the box, then they can’t test for abnormalities in the sex chromosomes,” said the nurse.
She had just told me in a breezy voice that my baby’s nuchal fold and nasal bones looked normal on the accompanied ultrasound and so with a sigh and a shrug, I ticked the box.
As I left, she told me that if the test results were fine, I would receive an email, and if there was a problem, I would get a phone call.
Ten days later the phone rang.
Professor Alan Cameron is both a consultant obstetrician and maternal fetal medicine specialist at the private IVF clincic and head of fetal medicine at the NHS Queen Elizabeth University Hospital in Glasgow. As the IVF clinic receptionist told me proudly over the phone, “He’s the guy who brought the Harmony Test to Scotland.”
In a kind voice, as I sat shivering in a towel, Professor Cameron told me that the test had come back as high risk for Turner Syndrome, a sex chromosomal disorder that only affects girls. Girls with Turner Syndrome are infertile, with many only diagnosed in adolescence when they fail to start puberty. He added that this genetic disorder is not hereditary and has nothing to do with maternal age
At no point in the phone call did he say the test might be wrong. What he did say was that the Harmony Test was just a screening test, not a diagnosis, and for this result to be confirmed, we would have to have either chorionic villus sampling (CVS), where a biopsy is done of the placenta by inserting a needle through the mother’s stomach, or an amniocentesis, where fluid is removed from the amniotic sac, also via needle, both of which have a risk of miscarriage.
What he did ask me though, was why I chose to have the test.
The words: “because I could,” flashed through my brain. Any pregnant woman is able to buy this test, at her own discretion, through a clinic. An obstetrician does not have to recommend her for it. If she has £450 spare, she can buy it, in the belief that she is doing the best thing for herself and her baby. Forewarned is forearmed, right?
As I put down the phone, shattered, I thought this was a fait accompli. I believed I had paid for the best new pre-natal technology on the market.
And as I read through the list of Turner syndrome symptoms sent to me, by the Professor, via email, my spine slumped, what little first-trimester energy I had disappeared through my fingers, and I sobbed and sobbed as I read what sounded like a busy week on a hospital ward for ten patients.
Girls with Turner Syndrome are shorter than their peers and require growth hormone to achieve adult height. They can have a hole in their heart. They can have misshapen kidneys which increase the frequency of UTI infections. They can have thick necks and lower ears and can be born with swollen feet. They are susceptible to ear infections, coeliac disease, underactive thyroid. They do not have intellectual disability, but often struggle with spatial reasoning and mathematics. They are often bullied.
Life is hard enough, and bringing a child into the world that was staring down the lens of a life lived close to a hospital seemed cruel. I began to consider that termination was the kindest thing for our much-wanted baby. The mother’s instinct to protect saw destruction as the greatest kindness.
That afternoon, through unstoppable tears, I told my best friend over the phone. Her response stopped me in my tracks.
“It could be wrong,” she said.
“How can it?” I asked. “These tests are considered to the best new technology out there.”
“Don’t be so sure. Look into it before you make any more decisions,” she said.
And so I did. And I was stunned.
I began on the Internet. On baby forums I found woman after woman in the United States whose babies were considered high risk for Turners Syndrome, and for whom the test had been wrong. One woman said her doctor told her the test was so unreliable, you may as well flip a coin.
I read about something called a Positive Predictive Value (PPV), which refers to the proportion of positive and negative results that are indeed true positives and true negatives – basically how precise the test is – and I found a Bayesian mathematical model that that estimated for a woman with a maternal age of 41, the NIPT for Turner Syndrome had a 59% chance of being wrong, and a 41% chance of being correct.
Two days later, I called the Professor back to ask if my Internet scouring was pure fancy, or did it hold some truth?
Was it true that the PPV for this test could around or below 40%?
“Yes, but we don’t know,” he said.
Could this result could be false positive?
“Yes,” he said.
Why didn’t you tell me this when we spoke on the phone? Was it because you didn’t want to give me false hope?
“Yes,” he said.
What I wouldn’t have given for a flicker of hope over these last two days when it felt like I was drowning, when I could barely stand up for longer than half an hour, when I lay in my husband’s arms and cried and cried, and began to pray for a miscarriage so that this would all be over.
Three days later, the Professor performed a second ultrasound. Many babies with Turners have large growths on their neck – cystic hygromas – and heart defects. Most die in-utero, with an estimated only 1% of girls with Turners syndrome making it to birth. My baby had a perfect neck, a strong heartbeat – although a proper examination of the heart could only be done at 20 weeks – and measured correctly for her gestational age. But we were not out of the woods. It’s also true that some girls with Turners show no symptoms in-utero. If wanted to know for sure, my only choice was an invasive test with the accompanied risk of miscarriage.
I was handed over to the midwife, who did her best to give us all the information we needed to make our decision, deploying that NHS non-directional style that does not offer any personal advice, opinion or steer.
It seemed an irony that a laboratory was prepared to charge me £450 for a test that they had insufficient validation for, and a midwife with years of experience was not permitted to share that experience with me as I made one of the hardest decisions of my life.
I stared deep into my soul and couldn’t find the answer, so I called my Yorkshire auntie. Her fury was quick and palpable.
“You can’t terminate a baby because she might be short, flat-chested and can’t do maths. And besides, we are from strong Yorkshire stock. You know that. Don’t let them stick a needle into you. They have already treated you like a science experiment. Don’t let them do it again.”
She was right. I had paid £450 in a bid for certainty, and had inadvertenly bought myself a seat on a rollercoaster of doubt and fear. It was time for me to take back that power.
And so I put away the tears and began a proper journalistic investigation.
I began with the email the Professor had forwarded from the lab.
The first line read: “There is currently not enough data to make sensitivity and specificity claims for the individual sex chromosome conditions.”
What the hell did that actually mean?
I typed “sensitivity” and “specificity” into Google and found myself in a world of medical statistics. Slowly I deciphered that it meant that neither the pharmaceutical company that devised the test, nor the labs and clinics that sell it know how good it is in demonstrating whether someone really has the condition they are testing for. That simply do not know how likely the test is to bring back false positives.
I called TDL Genetics, the laboratory who had tested my blood and asked them why they do not know this.
“The test is primarily designed for Trisomy 13, 18 and 21, there is just a limited number of patients who have had the sex chromosome conditions to be able to give accurate results,” said clinical scientist, Elaine Holgado
“And how often do you follow-up to see if babies with your high risk score, do indeed have Turners?” I asked.
“Because a lot of our patients are private, it’s not something that we always get the follow-up information because a lot of the follow-up testing happens in the NHS and there is no link between us, the private clinic and the hospitals,” she said.
So they don’t.
I asked why a friend, who had the Harmony Test the previous year, was not offered to test for sex chromosome abnormalities.
“Because of the test’s limitations, some clinics don’t offer it at all, or only when there is clinical indications on the ultrasound. We do educate the clinicians to discuss with the patients, and most of the time everything is fine and it’s not an issue, so maybe they become too blasé with what they are offering until something like this happens. In your case, the pre-test counselling seems to have fallen short somewhat.”
At my next ultrasound appointment, I asked Professor Cameron why I wasn’t given proper counselling at GCRM before the test?
“There aren’t enough genetic counsellors in the UK. If you bought this this test in the US, you would be given four hours of genetic counselling beforehand, but that we don’t have the people in the UK.”
But we wouldn’t put planes in the sky if we didn’t have sufficient pilots, so why are we testing for genetic conditions, if we don’t have the staff to protect vulnerable people and keep us safe?
I asked him why the National Institute of Clinical and Health Excellence (NICE) wasn’t watching this, protecting patients from this situation, and he said it currently fell beyond their remit. So whose remit was it? The next day I called every regulator I could think of. Every press officer shook their head down the phone. By the end of the day I was in tears again. Could it be that nobody in the UK was regulating this? That nobody was making sure that pregnant women weren’t being treated like science experiments?
A few days later, my suspicion was confirmed by Catherine Joynson, the assistant director of the Nuffield Council of Bioethics.
In early 2017, the council had released a detailed report on the ethical worries around the brave new world of NIPT.
“NIPT is not regulated. It has fallen between the cracks. I’m trying to get someone to notice it. I’ve written to the Care Quality Commission who’ve sort of ignored us, the health minister who has now lost his job in the reshuffle. The trouble is, with Brexit going on, these kinds of issues are just not a priority.”
What’s more, in their study, Nuffield couldn’t find any good scientific studies on testing for sex chromosome aneuploidies.
“We think it’s important that people have reproductive autonomy but anecdotally we heard that these tests are being offered without upfront information. We heard from many women who’ve you’ve ended up in a situation where they have some results that are really scary and they are not sure what to do with them,” says Johnson.
Which was how I felt. Turner syndrome, as I was to discover, was a broad church. While an internet search will list all sorts of conditions that make your heart grow cold, practising geneticists and endocrinologists will tell you that Turner Syndrome can actually be a very mild condition, with many families not even knowing they have it until the girls hit puberty and fail to get their period.
My vexation was further inflamed by a 2015 meta-analysis study published in Ultrasound, Obstetrics and Gynecology Journal. The authors examined all the current studies on screening for sex chromosome aneuploidies by cfDNA testing – effectively the Harmony test – and concluded that: “It may be inappropriate to offer pregnant women screening for sex chromosome aneuploidies by cfDNA testing just because it is feasible. There are several reasons for this: first, the phenotype of these aneuploidies is generally mild; second, the test has a high failure rate and relatively low DR [detection rate] and high FPR [false positive rate]; third, fetal mosaicism accounts for up to 50% of these aneuploidies; and fourth, the test may uncover a previously unknown maternal aneuploidy; up to 90% of women with 47,XXX are not aware that they have a third X chromosome.”
With further digging, I was gobsmacked to find that Ariona, the Harmony Test’s inventor (before selling it to Roche) had said as much in their own 2007 study, and even went as far as to state: “The inclusion of general sex chromosome aneuploidy (SCA) assessment has not been studied extensively and may not fit into classic or well-recognized criteria for prenatal screening. As such, NIPT with evaluation of SCA may be best utilized in cases where a clinical indication such as an ultrasound finding or family history is present.”
So if the Harmony Test has not been tested in the general population, if it was invented to be used in a population where clinical indications are already present, then why is it being sold to any pregnant woman who can afford £450, at her discretion? And why is there no regulator making sure that vulnerable pregnant women are not at the mercy of a brave new world of genetic testing that we are psychologically ill prepared for?
My story has a positive outcome. After presenting my research to Marco Gaudoin, head of the Glasgow Centre for Reproductive medicine, Dr Gaudoin did his own digging, and agreed that the level of false positives were unacceptable for this test, and has since decided that at his clinic “SCA screening is off the table.”
Professor Cameron will still offer it if any pregnant woman wants SCA screening as part of the NIPT, but they will need to see the professor first so he can counsel them accordingly.
Furthermore, Health Improvements Scotland, who regulates the private healthcare sector in Scotland, agreed that there was a gap in regulations and they would set out to rectify this.
As for my baby. Fintry Annie Bell was born on 19 June 2018. Although on the small size at birth (but there are short genes in our family), she continues to have no symptoms of Turners. On the day after she was born, excited paediatricians gathered around us, pushing us to have the genetic test to find out whether she did in fact have the condition. We declined. They pushed. We declined again. Our priority in the days after her birth, and for the rest of her life, was to feed her, nurture her, love her. It’s well documented that anxiety is bad for breastfeeding, and so I declined anything that would put further stress between me and her. She is not a science experiment. We have agreed that – perhaps – when she is one year old we will do a test to see if her X chromosomes are different to those of other girls. In the meantime though, it doesn’t matter. She is healthy, beautiful, full of smiles, and hitting all her textbook milestones. And to think that I considered terminating my pregnancy.