Genome Sequencing of Blood Samples May Lead to Detection of Early- Stage Lung Cancer

Get Permission

IF THE INITIAL promise of research presented at the 2018 ASCO Annual Meeting bears fruit, we may one day have a simple blood test to screen for early-stage lung cancer and possibly other cancers. Although it is still very early days for this test, an initial report from the ongoing Circulating Cell-Free Genome Atlas (CCGA) study showed that genome sequencing of blood samples detected about 40% of early-stage lung cancers and about 90% of late-stage lung cancers with excellent specificity.1 

Geoffrey R. Oxnard, MD

Geoffrey R. Oxnard, MD

“Two years ago, it was a pipe dream to conceive of a blood test to identify lung cancer. Today we have shown that it is feasible to find cancer in the blood. It will take some time and further studies to develop a clinically useful test,” said lead author Geoffrey R. Oxnard, MD, of Dana-Farber Cancer Institute, Boston. 

“The U.S. Food and Drug Administration has a clinical path for diagnostic tests, and it requires demonstration of analytic validity. This report shows that the researchers are on that path. The test is not yet developed, but it is exciting preliminary research,” said ASCO Chief Medical Officer, Richard Schilsky, MD, FACP, FASCO, FSCT, at a press conference where these data were presented. 

Richard Schilsky, MD, FACP, FASCO, FSCT

Richard Schilsky, MD, FACP, FASCO, FSCT

Early detection of lung cancer is a high unmet medical need. Although screening high-risk people with low-dose computed tomography identifies early-stage lung cancer and saves lives, several studies have shown an abysmally low uptake of this test in the at-risk population. Reasons for lack of adoption are complex, Dr. Oxnard noted. Low-dose computed tomography has drawbacks including false-positive results and logistical challenges, which might be overcome if a simple blood test for screening were validated and available. “Cell-free DNA–based tests represent an untapped opportunity for cancer detection,” Dr. Oxnard emphasized. 

Study Details 

THE MASSIVE CCGA study has enrolled over 12,000 of 15,000 planned participants at 142 sites across North America: 70% with cancer and 30% without cancer. The substudy on which Dr. Oxnard reported is the first of the data comprising 2,800 samples enrolled in a prespecified case-control study: 1,785 samples for the training set and 1,015 samples for the test set. His report on the training set was based on 561 noncancer samples and 118 lung cancer samples. The report on the test set was based on 362 noncancer samples and 46 lung cancer samples. In general, there were more males and ever-smokers among the lung cancer participants. 

“The sequencing we used is exhaustive,” Dr. Oxnard said. The researchers used three parallel sequencing approaches: targeted sequencing to detect somatic mutations and two different whole-genome sequencing approaches: whole-genome sequencing to detect somatic gene copy number changes; and whole-genome bisulfite sequencing of cell-free DNA to detect abnormal methylation patterns (epigenetic changes). 

“White blood cells are sequenced too, and they can have mutations and copy numbers that pollute the DNA and lead to false-positives. We were able to screen out this ‘interference’ and were left with mutations, copy number, and methylation sequencing. We used this data set to look for cancer in the plasma DNA,” Dr. Oxnard explained. “The findings [about white blood cell mutations] are important and can complicate the diagnosis,” he emphasized. 

“Cell-free DNA–based tests represent an untapped opportunity for cancer detection.”
— Geoffrey R. Oxnard, MD

Tweet this quote

“A major advantage of the the CCGA-targeted sequencing approach is a very low false-positive rate of about 2% and a specificity of 98%,” he continued. 

The authors found that the whole-genome bisulfite sequencing approach correctly identified 41% of early-stage lung cancers and 89% of late-stage lung cancers in the training set, with 98% specificity. The whole-genome sequencing test (copy changes) detected 38% of early-stage lung cancers and 87% of late-stage lung cancers, whereas targeted sequencing detected 51% and 89% of lung cancers, respectively. Results were consistent in the test set, with 50% of early-stage cancers identified by the whole-genome bisulfite sequencing approach. 

“This first interim analysis of the CCGA study showed that comprehensive sequencing of plasma cell–free DNA generates high-quality data across the spectrum of genomic features [copy number, methylation] that allowed detection of noninvasive lung cancer across stages and histologies. These findings support further development of ongoing studies,” Dr. Oxnard said. ■

DISCLOSURE: The study was funded by GRAIL, which is developing the technology. Dr. Oxnard has received honoraria from Bio-Rad, Chugai Pharma, Guardant Health, and Sysmex; has served as a consultant or advisor to AstraZeneca, Boehringer Ingelheim, DropWorks, Genentech/Roche, GRAIL, Ignyta, Inivata; Loxo Oncology, Novartis, and Takeda. Dana-Farber Cancer Institute has a patent pending titled “Noninvasive blood-based monitoring of genomic alterations in cancer,” on which Dr. Oxnard is a coauthor. Dr. Schilsky reported no conflicts of interest. 


1. Oxnard GR, Maddala T, Hubbell E, et al: Genome-wide sequencing for early stage lung cancer detection from plasma cell-free DNA. 2018 ASCO Annual Meeting. Abstract LBA8501. Presented June 4, 2018. 

Related Articles

Expert Point of View: David Graham, MD, FASCO

David Graham, MD, FASCO

David Graham, MD, FASCO

ASCO EXPERT David Graham, MD, FASCO, Medical Director at the Levine Cancer Institute in Charlotte, North Carolina, was enthusiastic about this new technology. “This is an important first step showing there is an easier way to detect lung cancer at earlier stages. If ...




By continuing to browse this site you permit us and our partners to place identification cookies on your browser and agree to our use of cookies to identify you for marketing. Read our Privacy Policy to learn more.