Image: Gio.tto/Shutterstock

The UK’s Plan to Sequence 100,000 Human Genomes

Using one new technology as both a clinical and research tool.

|
Jul 17 2015, 5:45pm

Image: Gio.tto/Shutterstock

About seven years ago, Sophie Lewis was taking part in a 10K charity race when she had a heart attack. She was in her 20s and led an active lifestyle.

While her fellow runners were able to perform CPR and save her life, doctors couldn't explain why she collapsed. Now 34, the housing association worker from Cheshire in the UK is no closer to knowing what's wrong with her, though she's had several incidents since.

"They never found out what caused my heart attack," said Lewis. "They fitted me with a defibrillator in case it ever happened again, and they tested me for everything—all the tests came clear."

Lewis is one of the first patients to take part in a UK-wide project that's using a new tool to uncover the causes of rare diseases and cancers using genetic testing: whole genome sequencing. The 100,000 Genomes Project aims to sequence 100,000 genomes from National Health Service patients, the first of which were recruited last year. It's run by Genomics England, a company set up and owned by the Department of Health.

The aim is twofold: to help clinically diagnose people like Lewis who have been left without answers, and to establish a huge dataset for researchers to better understand the causes of disease—and perhaps even develop treatments.

"There's an absolute sea change in terms of the capacity at which you can deliver genetic testing."

Lewis was asked last year if she was interested in taking part. "Obviously I absolutely was, because I just thought it was marvellous there might be some chance of them finding out what condition I had, and test my family for it," she said. Without knowing more about her condition, doctors can't tell if her relatives may be at risk too.

Whole genome sequencing looks at the complete DNA sequence altogether.

From the patient's point of view, it doesn't take much effort—it just means giving a blood sample (two close family members are also asked to contribute). But it's the analysis of the samples that has held sequencing back until recently.

"Both in the research arena and the clinical arena, what you can deliver in genetics and genomics has been entirely predicated on the technology," said Clare Turnbull, who is the clinical lead for the cancer program of the 100,000 Genomes Project. "Both research-wise and clinical-wise, 10 years ago if we wanted to study genes we had to decide which genes we wanted to study and then analyse each gene individually."

Then came the Human Genome Project, which aimed to map all of the human genes.Completed in 2003, it took about ten years, £2 billion, and tens of thousands of scientists.

But since then, the technology has improved. "Next-generation" sequencing allows researchers to produce many sequences in parallel, which makes the process much quicker and cheaper, and makes the huge scope of the 100,000 Genomes Project feasible.

"That's an absolute sea change in terms of the capacity at which you can deliver genetic testing," said Turnbull.

Cost of sequencing the human genome, estimated by the US National Human Genome Research Institute. Image: Ben Moore/Wikimedia

She pointed out that the whole genome sequencing also includes genomic material outside of the "exome," the part of the genome which contains the 20,000-or-so genes we know of but only makes up around 1.5 percent of the overall genome. "The other 98.5 percent is genomic material outside of the genes, but actually we recognise probably a lot of determinants of disease reside in the region outside of the gene," she said.

While patients like Lewis are hoping to get answers about their own condition, and how it could affect them and their family, the full value of the 100,000 Genomes Project is in the scale of data it intends to collect. In sequencing the genomes of so many individuals, patterns may emerge.

"In rare diseases, particularly something that's very rare, you're more likely to understand the cause of it and understand the implications of a particular rare disease if you study groups of patients together," explained Turnbull.

For this reason, the project is not only collecting genomic data, but also phenotype data—information about patients' observed characteristics and traits—and information such as family history.

Multiply that by 100,000 people, and the project becomes as much a data challenge as a health challenge.

"You need substantial IT to store the data and to process it," said Tim Hubbard, head of bioinformatics at Genomics England. He explained that the company is currently in the process of selecting the companies to do this work.

Image: science photo/Shutterstock

Following the dual clinical and research applications of the project, the information has to feed back both to aid the health care of individuals and allow for broader analysis. "On one side, some of these contracted algorithms will run to provide the interpretation back to the health service," he said, "But from the other point of view, it'll provide an environment for doing research over a large number of genomes in an anonymised framework." (Taken to task by the Guardian, the Department of Health admitted it's actually pseudonymous—given the individual nature of the information collected, true anonymity is tough to promise.)

In any case, attention to privacy is key; given the sensitivity of data that's so intrinsically personal, safeguards have to be in place to protect the patients involved. After all, the information collected by the project will include details such as an individual's medical history—the kind of thing you'd usually only share with your doctor, in confidence.

The privacy process starts as soon as patients join the project, with a consent form that asks for permission for their data to be used by researchers in the future as well as immediately analysed for mutations. When they give a sample, it's shipped to a biorepository where the DNA is extracted and sent to a sequencing operation, and the results of the sequencing are transferred electronically to a secure environment in a data centre. "And that's where it'll stay," said Hubbard.

As part of the security around the project, researchers will not be able to move the data elsewhere—they have to go to it. "The users are all authenticated and there'll be a protocol for ensuring researchers are vouched for by their organisations and the grouping they're working in," Hubbard added. Part of the agreement with researchers also states that they shouldn't try to re-identify anyone; the sequencing data is separated from explicit identifiers.

For her part, Lewis said she was more than happy for her data to be shared, though she admitted she wasn't really sure what the worst that could happen was. "My more initial priorities are my family," she said.

"Many patients are totally shocked when we ask them for research consent that you're not already using their data and samples for research, because surely it would make sense to do so."

While the project has only just started, and clearly still has a lot of nuts and bolts to figure out in its implementation, its legacy may extend beyond the realm of genomic research. For Turnbull, one of the biggest breakthroughs of the project is bringing the objectives of clinical care and advancing research together: the same data that helps one patient in the short term could help others in the long term.

"Philosophically, as a clinician in the NHS but also a researcher, I think the enormous value of this program in one regard is about collecting a single set of high quality samples, a single set of high quality clinical data, an overarching consent, and using your samples and data to make a diagnosis and give information to that patient, but also provide samples and data for the research community to further advance research into the disease," she said.

Usually, a researcher would have to independently go through a lot of work (and funding) before approaching patients to get the kind of data that may have already been collected in the clinic. With its overarching consent process, the 100,000 Genomes Project makes that data available to researchers from the get-go.

She said that when she started out as a medical student, she kind of presumed that was how the unified health system of the NHS would work anyway. "Many patients are totally shocked when we ask them for research consent that you're not already using their data and samples for research, because surely it would make sense to do so," she added.

There are now 11 "genomic medicine centres" across the UK that will recruit patients to the 100,000 Genomes Project. While it sounds like a lot—and it is massive compared to the scale of genomic research done previously in the UK—Turnbull pointed out that the 100,000 figure quickly dwindles when you take into account that this is split 50/50 between participants with cancer and those with rare diseases, and each patient will also bring a couple genomes from family members.

But with the technology and systems implemented, the future could see whole genome sequencing used with much more regularly. "We anticipate at the end of this program that it will segue into some structure of NHS centres routinely commissioning whole genomes for diagnosis," Turnbull said.

Meanwhile, Sophie Lewis is still waiting for answers. Since she got her defibrillator, she said she's been shocked dozens of times and had a second heart attack last summer while training to climb the Three Peaks. Luckily, especially as she was alone and in the middle of nowhere, the defibrillator kicked in.

The main problem, she said, is that her doctors never quite know what's safe for her to do. "The doctors are unsure what to tell me about my lifestyle and what I can or can't do," she said. "That advice seems to change a lot, because they don't know what's wrong with me. Quite often they'll say, 'Oh you're safe to do this,' and then I'll get a shock from my defibrillator because I wasn't safe to do it."

She's backed out of a skydive and zip wire she was planning to do with work as doctors gave her conflicting advice, and she didn't want to have shocks in front of her colleagues.

It's been about a year since Lewis gave her samples to the 100,000 Genomes Projects, and she hasn't heard anything yet. But as more people take part, perhaps with similar conditions, maybe researchers will gain more insight.

"The ideal outcome would be that not only can they identify what condition I have, but also they'd be able to use that information to tell me for definite what I can and can't do, and I can test my family and see if they have that condition as well," she said. "I'm just keeping my fingers crossed now."

Modern Medicine is a series on Motherboard about how health care and medical technology can move forward so rapidly while still being stuck in the past. Follow along here.