Could we soon detect cancer in 10 minutes? | Science

AAbout seven years ago, researchers at the American DNA sequencing company Illumina began to notice something odd. A new blood test performed on 125,000 pregnant women looking for genetic abnormalities such as Down syndrome in the fetus gave extremely unexpected signals in 10 cases. Surprisingly, it appeared to them that the abnormal DNA they saw did not come from the fetus, but from an undiagnosed cancer in the mother. Cancers of different types were then confirmed among the ten. "It was not a test developed for cancer screening," says Alex Aravanis, then director of research and development at Illumina. "But it was proof that this could be possible."

In 2016, Illumina created the Silicon Valley-based Grail spin-off company with Aravanis as scientific manager. Backed by more than $ 1.5 billion in funding, including money from Microsoft's co-founder, Bill Gates, and Amazon's founder, Jeff Bezos, Grail is looking for several types of cancer before the symptoms, via a simple blood test. The test examines cell-free plasma for fragments of so-called circulating tumor DNA (cDNA) that are cleared by cancer cells.

Detecting cancer earlier, before the symptoms, means that you can intervene earlier and that people are less likely to die. Although doctors can screen for bad, colon and lung cancer, most forms of the disease can only be detected after the onset of symptoms. And although this is far from the only approach (see box), the beauty of blood is that it is minimally invasive to collect. "A relatively simple blood test to screen for evidence of cancer … could improve or even replace some screening programs over time," says Jacqui Shaw, a professor of translational cancer genetics at the University of Leicester, who is studying SDAC.

CDNA research has become a viable proposition in recent years due to improvements in DNA sequencing technologies that can scan fragments and find those with changes that may indicate cancer. While other blood-based biomarkers are under study, the advantage of cDNA lies in the fact that, because of its direct link with the tumor, it can be very specific for the identification of cancer. For this reason, ctDNA also seems promising as a means of profiling and monitoring advanced cancers, a "liquid biopsy".

Early detection is a more difficult problem. In the beginning, when the tumor is small, there is not as much cDNA to detect. The women identified by Illumina as having cancer were all late, not at an early stage.

To date, a company is proposing a cTADN-based blood test for early detection of cancer: Epigenomics began offering its 2016 colon cancer screening test, based on the detection of a biochemical modification of a tumor. single gene.

But the dream dreamed up by Grail and others is an inexpensive test, perhaps no more than $ 500, that could likely be given every year to those who reach a certain age, with a great chance of detecting many cancers at the same time. times with great accuracy (Grail was not announced a final number but think it will be in the region of 10). This is a test we could all, if it worked, one day pbad. "The big studies are still to be done," says Nitzan Rosenfeld, a researcher studying CTAD at the Cambridge Institute of Cancer Research UK and co-founder of the British Inivata liquid biopsy company, "but considerable progress has been made."

Cancer begins essentially when the DNA of a normal cell is mutated or altered. From this moment, the cell multiplies too often and an abnormal cell mbad or tumor is formed. Part of the cells invariably die and "spill" genetic material into the blood, mixing with larger amounts of DNA fragments from normal cell death.

It has been reported for the first time that DNA fragments carrying carcinogenic mutations floated freely in the blood of cancer patients in the mid-1990s. The results drew the attention of Dennis Lo today. Professor of Medicine and Chemical Pathology at the Chinese University of Hong Kong, who thought that "a baby living in a mother is a bit like a cancer that develops in a patient". Based on this idea, he later discovered fragments of fetal DNA in the maternal blood and was a pioneer in non-invasive prenatal testing (NIPT) (he sold his patents to Illumina and other companies). ). He also began, with others, to apply this knowledge to the way in which tDNA fragments could be used in the monitoring and detection of cancer. In 2017, Grail merged with Lo, Cirina, which also focused on early detection.

Grail's competitors include Guardant Health, a liquid biopsy company founded in 2011 valued at $ 3.5 billion. She is also based in Silicon Valley. He recently expanded his activities to an early detection test for four common cancers: lung cancer, bad cancer, colorectal cancer and ovarian cancer.

There are also multiple academic efforts with designs on marketing. Last year, researchers at the Johns Hopkins University School of Medicine released details of a potential blood test called CancerSEEK, which covers eight cancers. And last December, researchers at the University of Queensland made headlines with a "10-minute test" that they called "methylscape" and which they say could potentially positive or negative response to the presence of cancer in the body, even if it would not help. t identify its location.

Each group or company has developed or is developing a way to detect cDNA. The Grail and Guardant tests are based on cell-free DNA sequencing. This can look for mutations, an increase in the number of chromosomes or genes or unusual biochemical changes called epigenetic modifications, all of which can occur in the DNA of cancer cells. Grail has experimented with these three methods but has not announced the method that his final test will use. The Guardant method takes them all into account. CancerSEEK, meanwhile, looks for a small number of mutations as well as protein markers known to increase certain cancers. Nickolas Papadopoulos, professor of oncology and co-head of the work on the latter method, explains the small panel of changes to reduce costs. Instead of sequencing, methylscapes uses gold nanoparticles to detect epigenetic alterations.

The results have been published or presented for all these methods, demonstrating that cancer-related signals can be seen. They are based on small studies of 1,000 or fewer people with cancer at different stages. The key to testing is to achieve a high probability of detection (good sensitivity) and a low false-positive rate to avoid unnecessary anxiety and unnecessary follow-up. (In a test designed for a general population, where cancer is not widespread, the latter is particularly important because even a low false-positive rate will result in a significant number of incorrect diagnoses.)

The best detection rates of Grail, based on a prototype of detection of epigenetic changes, ranged from 80% to 47% for nine cancers (respectively ovary, liver, lymphoma, multiple myeloma, pancreatic, colorectal, esophageal, cervicofacial and pulmonary). Breast cancer ranged from 56% to 11% depending on the type. The false-positive rate was set at 2%, although according to Aravanis, other studies suggest that it could reach less than 1%. "From a single test, we detected a very large fraction of the cancers most affected by early stage mortality, with very high specificity," he explains.

Among Grail's competitors, CancerSEEK's sensitivity ranged from 98% for ovarian cancer to 33% for bad cancer, with a false-positive rate of less than 1%. Guardant showed that he could detect lung cancer in 71% of cases and colorectal cancer in 67% of cases with a false positive rate of 2%. Methylscape had a sensitivity of 90% but the rate of false positives was higher, between 10 and 15%.

Yet while companies are optimistic about what they see, not everyone is convinced. Eleftherios Diamandis, professor of clinical biochemistry at the University of Toronto, has made a name for criticizing major revolutionary blood test projects. Long before the media question the validity of the blood test technology of the young company Theranos – the company and its CEO were accused last year of "mbadive fraud" by the Securities and Exchange Commission of the United States Unis – Diamandis had doubts in the scientific literature.

Last year, his partner Clare Fiala and himself published a series of newspaper articles in which they questioned the true utility of cTDNA in early diagnosis of cancer. Diamandis calculations, based on experimental literature data, give an idea of ​​the tumor size that a cDNA-based blood test might be able to detect. His work indicates that with the technology available to badyze cDNA, tumors should have a diameter of about 1 cm or more to be detected. "It's a pretty big tumor," he says. It is simply unlikely that there is a single tumor fragment of this size in the 10 ml of blood constituting a standard sample. Although more blood can help a little, it would make things a lot less enjoyable.

According to Diamandis, the tests seem to be going well, but that's because they are applied to people who have already been diagnosed with cancer. "If you're talking to a real population of asymptomatic individuals, their success rate will be, as I predicted, much less," he says. "For the [multi-cancer] blood test – to this day, I am relatively pessimistic. "

For their part, groups and companies recognize that large studies of people without diagnosed cancer are needed. CancerSEEK is recruiting 10,000 cancer-free women to get their blood test; He will then follow them for five years to see if they develop cancer. And Grail has two large and long studies on those who are asymptomatic, one in 100,000 and one in 50,000, in preparation. "These are large clinical validity studies that complement the evidence we generate," says Aravanis.

They also point out that they discover many of the most clinically aggressive and potentially life-threatening solid tumors. "Even if you find 10%, it's 10% more than zero," notes Papadopoulos.

Aravanis wonders if the discovery of a large number of cancers at a very early, slow-growing, and potentially inconsequential stage could be really helpful. "Whether or not you detect every cancer at the earliest stage or pre-cancer is an interesting scientific question, but I do not know if it's really the important clinical issue," says Aravanis. "[Overdiagnosis] is something that some existing screening programs are struggling to ".

Perhaps the solution, say other people, such as Rosenfeld and Shaw, is to take a more limited approach: attacking high-risk populations or targeting cancers for which the diagnosis is often late. "[Then] I think we could find places where we can improve the current clinical situation, "says Rosenfeld. In addition, Diamandis does not rule out success in liquids other than blood, such as urine and cervical fluids for bladder and ovarian cancer, where cDNA concentrations could be much higher because they are in contact with, or near, the tumor.

The question in millions of dollars is: when can we expect a blood test once a year to detect multiple cancers? The typical answer from the experts is that it will take at least five years before there is enough data to show if it can work and then it will be necessary to contact regulators and health economists. Patience, it seems, is a virtue in the fight against cancer.

Detection of breathing and cancer

Although blood is relatively easy to collect to look for a cancer signal, what if you can detect it on your breath – no needle needed? The British company Owlstone Medical is among those who want to know if this is possible. He developed a breathalyzer based on the idea that there may be chemicals – volatile organic compounds – in the breath, indicators of early stage cancer.

He is currently conducting a partnership trial with the NHS to determine if there are differences between breath-detectable chemicals between people with lung cancer and those who are n & # 39; 39, do not suffer. Lung cancer, says Max Allsworth, scientific director of the company, is a good starting point because the air we breathe pbades directly into the lungs and beyond the tumors.

The company has also recently launched a separate smaller study to determine if six other types of cancer can also be detected earlier (in this case, the chemicals would end up less directly in the breath, via the blood, which exchange volatile chemicals). with air in the lungs). Depending on the results of the test, it may be a more generic cancer test – not telling you exactly where the cancer is, but a common signal has been found. In any case, says Allsworth, if chemical substances in the breath suggest cancer, they will be present very early, before you probably have circulating tumor DNA (ctDNA).