![photo of sample blood vials](https://www.mskcc.org/sites/default/files/styles/large/public/node/139672/3x2/istock-668185506_1200x800.jpg)
Scientists are studying how to use blood from patients as a source of tumor DNA to diagnose cancer in its earliest stages.
An ambitious goal among cancer biologists is to develop a blood test that could easily detect cancer in its early stages, before it becomes entrenched in the body. Early detection would offer opportunities to nip cancer in the bud, saving lives.
The results of a study presented at the annual meeting of the American Society of Clinical Oncology (ASCO) suggest that such a test might one day be feasible.
According to Memorial Sloan Kettering medical oncologist Pedram Razavi, the study’s first author, a blood assay that measures circulating tumor DNA (ctDNA) proved useful in identifying, with high sensitivity, genetic mutations that indicated the presence of cancer in the body.
The assay, which is made by a company aptly named Grail, is not the first ctDNA blood test to be evaluated in cancer, but it is the most comprehensive to date.
“This study is the first successful application of a novel high-intensity ctDNA sequencing assay, using an unprecedented combination of broad genomic coverage with ultra-deep sequencing,” Dr. Razavi says. The assay canvasses more of the genome and at a “depth” — the number of times a particular region is sequenced — that outpaces previous assays.
With this new approach, scientists could detect cancer mutations in the blood without already knowing which mutations are in a patient’s tumor, laying the groundwork for an early-stage cancer test that’s as simple as a pinprick.
DNA Hunting
When cells die, they release their contents, including their DNA, into the body. DNA may also be secreted from the cells. Some of the DNA enters the bloodstream. These free-floating bits of cell-free DNA can be extracted and then sequenced to find the order of nucleotide “letters” along the DNA strand. When tumor DNA differs in sequence from normal DNA at a particular region, that’s called a mutation.
Previous studies have shown the value of measuring ctDNA to make diagnoses or treatment decisions on the basis of specific mutations present in tumors. Such uses have been dubbed liquid biopsies. This study was different. Its goal was to evaluate the performance of a high-intensity sequencing assay as an initial step toward the development of blood tests that could detect cancer early, when a doctor wouldn’t necessarily know which specific mutations to look for.
The assay covers 508 genes, a sizeable portion of the genome. It is “almost ten times more of the genome than any other previously reported cell-free DNA assays at these sequencing depths,” Dr. Razavi says. That’s important because even tumors of the same cancer type can carry mutations in different genes. “The more of the genome that you cover, the higher the chances of finding mutations that come from the tumor,” he adds.
The assay is also very sensitive. Because it sequences more deeply, it’s able to detect the presence of tumor-derived DNA at very low levels. “Most of the cell-free DNA fragments in the blood come from normal cells,” Dr. Razavi explains. “Often, only about 1% or less of the cell-free DNA comes from a tumor.” To tackle this needle-in-a-haystack problem, the assay sequences each selected region of the genome 60,000 times on average to capture the rare cancer DNA.
Test-Drive in Patients
To evaluate the performance of their ctDNA assay, the team collected blood from 124 patients with metastatic breast, non-small cell lung, and prostate cancers. They then separated the plasma (the liquid part of the blood) from the blood cells to obtain the cell-free DNA, and then sequenced this DNA at the 508 different gene regions. In addition, as a control, the team sequenced the DNA in patients’ white blood cells. By analyzing the sequencing results of both the cell-free DNA and the DNA from white blood cells, the cancer-derived mutations present in the ctDNA could be identified more accurately. Inherited variations in the DNA were filtered out, as were other forms of non-cancer-related mutations, such as those originating from bone marrow cells. From these samples, the team was able to identify hundreds of tumor-specific mutations that signaled the presence of cancer in the body.
To confirm the test’s reliability, the team then used MSK-IMPACT to compare the mutations identified from the blood assay with those identified in tumor tissues from the same patients. (MSK-IMPACT is a genomic test performed on tumors from all patients with metastatic cancer who are treated at MSK.) Among 864 mutations detected in tumors, including the mutations present at very low levels, 627 (73%) were also found circulating in the blood. In 89% of patients, at least one of the mutations identified in the tumor tissue was also found in the blood.
Importantly, no knowledge about the mutations found in the patients’ tumor tissue was used during the blood ctDNA sequencing part of the study. Rather, the tumor tissue sequences served only to validate the results of the blood test. As Dr. Razavi points out, since the goal of the test is ultimately to detect cancer early, when earlier intervention could improve the cure rate and survival, it’s necessary to have a test that does not rely on knowing about specific mutations in advance.
An Important First Step
The patients in this study had advanced cancer. Whether the assay would work as well in patients with early-stage cancer is an open question, but one that Dr. Razavi says this study was an important first step toward answering.
“The ability of this assay to detect ctDNA without prior knowledge of tumor sequencing results and with an unprecedented breadth of signals are both important findings that support the possible feasibility of applying this approach to detection of cancer at earlier stages,” Dr. Razavi says. “This new assay demonstrates early promise for the future development of blood tests for early cancer detection.”
Achieving that particular prize will require researchers to learn more about how circulating cell-free nucleic acid signals differ between people who have cancer and those who do not. Gathering that data is the goal of the Circulating Cell-Free Genome Atlas study, a prospective, multicenter Grail-sponsored trial. It aims to enroll more than 10,000 individuals, including those without cancer and those with early- to late-stage disease across all of the major tumor types.
The hope is that one day, a person could have a blood test to look for signs of cancer. “And if you do have cancer, where is it coming from? That’s really the ultimate goal,” Dr. Razavi says.