A blood test based on cell-free DNA was able to detect cancer as well as the site of origin in patients with a clinical suspicion of cancer, according to results of the Circulating Cell-Free Genome Atlas (CCGA) study presented at the 2020 American Association for Cancer Research (AACR) Virtual Annual Meeting.1
A methylation-based assay was identified as the best test to continue studying, and it had almost 100% specificity and about 50% sensitivity for detecting the presence of 50 different cancers as well as the tumor of origin. Methylation reflects epigenetic changes across the genome to exposure to cancer signals. The CCGA study focused on a large population of 15,254 participants. The test has been granted Breakthrough Therapy status by the U.S. Food and Drug Administration.
“Early detection of cancer with the targeted methylationbased multicancer assay could potentially accelerate the diagnostic workup of patients with a high suspicion of cancer.”— David Thiel, MD
Tweet this quote
“Early detection of cancer with the targeted methylation-based multicancer assay could potentially accelerate the diagnostic workup of patients with a high suspicion of cancer,” said lead author David Thiel, MD, Chair of the Department of Urology, Mayo Clinic, Jacksonville, Florida. “This will be particularly useful in cancers of unknown origin,” said discussant of the trial, David Huntsman, MD, Founder and Chief Medical Officer of Contextual Genomics.
“Many cancers are detected too late, and the large percentage of breast, colorectal, and lung cancers are diagnosed when they are metastatic. The detection of cancer prior to the development of metastatic disease can improve 5-year survival,” Dr. Thiel said.
The estimated 5-year cancer-specific survival in patients diagnosed early is 89%, whereas it drops to 21% in those diagnosed when the disease is more advanced. In lung cancer, for example, 5-year disease-specific survival is 5% if diagnosed at an advanced stage vs 56% for those diagnosed early.
The multicancer assay under development is being studied in many clinical trials. The prospective PATHFINDER trial will assess the utility of the assay in clinical practice. The SUMMITY trial will enroll 50,000 people who do not have a cancer diagnosis at enrollment, but half will be smokers at high risk of developing cancer.
The CCGA trial enrolled 15,254 participants with and without cancer at 142 different sites in North America. In the case-controlled, longitudinal, observational study, the investigators collected blood samples from all participants and tissue samples for participants with cancer. All participants were followed for 5 years for vital status and cancer status.
Eligibility criteria included a confirmed pathologic diagnosis of cancer or a high clinical suspicion of cancer. The assay was designed to test for 50 different types of cancer.
The study has three substudies. Substudy 1 was designed to identify the best performing assay of three comprehensive sequencing strategies (targeted sequencing, whole-genomic sequencing, and whole-genome bisulfite sequencing [a methylation-based assay]). This substudy included 1,785 patients in the training portion and 1,015 in the validation portion.
The methylation-based assay outperformed the other two types, so substudy 2 was conducted to refine the chosen assay. There were 3,113 participants in the training portion and 1,354 in the validation portion. Substudy 3 is currently ongoing and is the final validation study, with a planned enrollment of 5,000.
A subgroup analysis of substudy 2 evaluated the test performance in 303 patients with a high clinical suspicion of cancer. Six weeks after the first blood draw, these participants underwent clinical and/or radiologic assessment, planned biopsy, or surgical resection to establish a definitive diagnosis.
Participants with a confirmed diagnosis of cancer included 164 in the training group and 75 in the validation group. Of those who did not have cancer, 49 were in the training group and 15 in the validation group. Among patients with cancer, 20 different cancers, ranging from stage 1 to 4, were identified, and one patient with stage 3 disease of unknown origin was found in the validation group.
Specificity of the assay in substudy 2 and the subgroup at high clinical suspicion of cancer was 99.8% in the training group and 99.3% in the validation group. The specificity in the subgroup at high clinical suspicion of cancer was 100% in the training group and 100% in the validation group.
“This suggests that the false-positive rate will be very low, and it was not elevated in the subgroup at high clinical suspicion of cancer, compared with the enrolled population,” Dr. Thiel pointed out. “A low false-positive rate would reduce the number of unnecessary diagnostic procedures and workups.”
“Sensitivity of the CCGA assay was lower, around 50%, and this is something we are working to improve,” he noted.
Substudy 2 and the subgroup analysis of those with high clinical suspicion of cancer showed comparable sensitivity. In substudy 2, the sensitivity was 40.2% in the training group and 46.7% in the validation group. In the subgroup including those with high clinical suspicion of cancer, the sensitivity was 50.4% and 59.3%, respectively. “These percentages [for sensitivity] improved when we excluded kidney cancers, which are difficult to pick up in blood,” he explained.
Looking at the test performance in the subgroup of those with high clinical suspicion of cancer, the accuracy of predicting the tissue of origin was 85.5% in the training group and 97.1% in the validation group. The precision in identifying specific cancer types was 100% in both the validation and training groups.
DISCLOSURE: The study was funded by Grail. Dr. Thiel reported no conflicts of interest. Dr. Huntsman is founder and Chief Medical Officer of Contextual Genomics.
1. Thiel DD, Chen X, Kurtzman KN, et al: Prediction of cancer and tissue of origin in individuals with suspicion of cancer using a cell-free DNA multi-cancer early detection test. 2020 AACR Virtual Annual Meeting. Abstract CT021. Presented April 28, 2020.
Discussant of the CCGA study, David Huntsman, MD, Founder and Chief Medical Officer of Contextual Genomics, said: “Screening is the peak [for the number of people who could benefit from early detection], but screening is the most challenging. It requires high sensitivity and specificity that is...