The idea of a general, quick and simple blood test for a diverse range of cancers just came closer to reality with news of a new study published in Nature Medicine.
Researchers from Stanford University School of Medicine have devised an ultra-sensitive method for finding DNA from cancer tumors in the bloodstream.
Previous research has already shown circulating tumor DNA holds promise as a biomarker for cancer, but existing methods for detecting it are not sufficiently sensitive and do not cover a diverse range of cancers.
Ways to increase the sensitivity and coverage of such tests exist, but these are cumbersome and time-consuming, and need lots of steps to customize for individual patients, so they are not feasible for use in clinics.
The new approach promises to change that. It is highly sensitive and specific and should be broadly applicable to a range of cancers, say the researchers.
Their new test identified around half of patients with stage 1 lung cancer and all patients with stage 2 or higher disease. They also showed the circulation tumor DNA was highly correlated with tumor volume estimated using CT and PET scans.
This suggests an approach based on the new test could monitor tumors at a fraction of the cost of present methods that rely on imaging studies.
In developing the test, they faced two major hurdles, as Maximilian Diehn, co-senior author and assistant professor of radiation oncology, explains:
“First, the technique needs to be very sensitive to detect the very small amounts of tumor DNA present in the blood. Second, to be clinically useful it’s necessary to have a test that works off the shelf for the majority of patients with a given cancer.”
Co-senior author, Ash Alizadeh, assistant professor of medicine, explains why they are interested in developing a general way to detect and measure disease burden in solid cancers, and how they are approaching it:
“Blood cancers like leukemias can be easier to monitor than solid tumors through ease of access to the blood. By developing a general method for monitoring circulating tumor DNA, we’re in effect trying to transform solid tumors into liquid tumors that can be detected and tracked more easily.”
Cancer cells divide and die, even without treatment. When a cancer cell dies, the DNA in its nucleus escapes into the bloodstream. This is present in small concentrations; something like 1 in 1,000 or 10,000 bits of DNA in the blood can be from a dead cancer cell in a person with cancer.
Even in patients with advanced cancer, the vast majority of DNA circulating in their blood is from healthy, normal cells.
So a test that can quickly and non-invasively monitor the tiny concentrations of cancer cell DNA would be really useful to clinicians who need to estimate the size of the tumor, how it changes over time, and monitor a patient’s response to treatment.
The team found a way to do this by boosting existing methods for extracting, processing and analyzing the DNA. They called their approach CAPP-Seq (which is short for Cancer Personalized Profiling by deep Sequencing).
CAPP-Seq is sensitive enough to detect one molecule of tumor DNA among 10,000 DNA molecules from healthy cells in the blood.
In their study, they tested blood from patients with non-small-cell lung cancer (this includes most lung cancers, like adenocarcinomas, squamous cell carcinoma and large cell carcinoma). But they say the approach should also work with solid cancers that occur in other parts of the body.
And while they see the test one day being used to follow the progress of tumors in patients already diagnosed with cancer, the researchers say it also has potential as a cancer screening tool for healthy and at-risk populations.
Although the test is described as a general test for cancer, it by no means just looks for one pattern of DNA. Each cancer is genetically different in different patients, but there are certain sets of DNA mutations that are the same across patients with the same cancer.
So the challenge was to find which DNA sequences were the ones most likely to indicate the presence of a given cancer across a diverse range of patients.
This is why the team decided to take a population-based approach. They looked in national databases that contain DNA sequences of tumors from thousands of patients, and identified the points on the cancer DNA that are different from normal DNA.
From this information, they were able to compile a fingerprint for each cancer type made up of all the DNA mutations recorded – these include insertions or deletions of short pieces of genetic material, plus where sequences of DNA have been shuffled around or even flipped over.
But while no patient will have all these mutations, nearly all of them will have at least one of them. This makes it possible to compile a test that looks for as many of the known mutations for a given cancer as possible. But it only has to find one of them to strike a positive.
The next stage of the study was to examine the genome of the 407 patients with non-small-cell lung cancer recruited for the study.
Prof. Alizadeh explains how, using an approach called bioinformatics, they looked for regions in the genome enriched for cancer-associated mutations:
“We looked for which genes are most commonly altered, and used computational approaches to identify what we call the genetic architecture of the cancer. That allowed us to identify the part of the genome that would be best to identify and track the disease.”
They identified 139 genes that only represent 0.004% of the human genome but are recurrently mutated in non-small-cell lung cancer.
“By sequencing only those regions of the genome that are highly enriched for cancer mutations, we’re able to keep costs down and identify multiple mutations per patient,” Prof. Diehn says.
Other approaches tend to look for single, well-known mutations that occur frequently, but not necessarily in every patient, with a particular cancer. Because it looks for more than one mutation, the CAPP-Seq approach is more sensitive and gives researchers more flexibility in how to track the cancer over time.
Prof. Diehn explains that there are currently no reliable biomarkers for lung cancer, a cancer that claims the most lives. He says they are “very excited” about the study results because “a personalized, clinically useful biomarker could revolutionize how we detect and manage this devastating disease.”
The team is now working on ways to quickly home in on patient-specific mutations and methods to suppress background noise in a sample so they can identify even very tiny amounts of cancer DNA that might be in it.
The researchers say CAPP-Seq may also have potential as a prognostic tool. When they tested one patient thought to have been successfully treated for lung cancer, they found low levels of circulating tumor DNA. The cancer came back in that patient, and they died.
Conversely, scans of another patient who was treated for early stage disease showed a mass that was thought to indicate disease was still present. But CAPP-Seq found no circulating tumor DNA in that patient’s blood, and they remained disease-free for the rest of the study period.
And in a third patient, CAPP-Seq found a mutation that makes non-small-cell lung cancer resistant to the drug that is commonly used to treat it.
Prof. Diehn says this suggests another use for the approach – to monitor how the tumor progresses and look out for the emergence of treatment resistance early on, giving enough time to switch therapy to target the resistant cells.
“It’s also possible we could use CAPP-Seq to identify subsets of early stage patients who could benefit most from additional treatment after surgery or radiation, such as chemotherapy or immunotherapy,” he adds.
Funds from a number of sources helped finance the study, including the Department of Defense and the National Institutes of Health.
Meanwhile, Medical News Today recently learned how another US study led by The Scripps Research Institute (TSRI) found a new biomarker for head and neck cancer and non-small-cell lung cancer. That study focused on CCTα – an antigen that prompts the immune system to make specific antibodies – and concluded it was a better predictor of patient outcomes than expression of ERCC1, which is involved in DNA repair.