SAN FRANCISCO – A new blood-based assay that measures circulating tumor cells (CTCs) shows good performance in detecting colorectal cancer and precancer, investigators reported at the 2018 GI Cancers Symposium.
Although colorectal cancer screening is a grade A recommendation of the U.S. Preventive Services Task Force, poor uptake remains problematic and contributes to more advanced disease at diagnosis, noted lead investigator Wen-Sy Tsai, MD, assistant professor at Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan.
“Currently, one-third of Americans have never been screened for colorectal cancer,” he said, due in part to reluctance to undergo colonoscopy and poor compliance with stool tests. Further complicating matters, the fecal immunochemical test (FIT) has a high false-positive rate.
He and his coinvestigators tested the new assay, called CMx (CTCs in Maximum), among 620 individuals in Taiwan who underwent colonoscopy – some with colorectal cancer, some with precancerous lesions, and some healthy. Results showed the assay’s sensitivity was nearly 87% for cancer and 77% for precancerous lesions. Specificity exceeded 97%.
“The CMx assay is capable of detecting … early-stage cancer with a low false-positive rate. As it is a blood test, higher compliance will lead to better outcomes,” Dr. Tsai proposed. “Because the mechanism of CTC dissemination is similar, the CTC assay can be applied in other cancer types such as prostate, breast, and lung cancer.”
The investigators are also identifying collaborators for testing the new assay among U.S. populations, he noted.
“Screening tests offer some of the greatest potential for getting to zero colorectal cancer deaths,” said invited discussant Douglas A. Corley, MD, PhD, clinical professor and gastroenterologist at the University of California, San Francisco, and a research scientist at Kaiser Permanente, San Francisco.
CTCs are especially attractive for screening because they could be both sensitive and specific, he said. “Unlike something such as fecal immunotesting, which is not testing specifically for cancer, or colonoscopy, which is quite invasive and detects a lot of things that may never progress to cancer, CTCs offer this potential,” said Dr. Corley.
Challenges of interpreting new screening tests include their heavy dependence on the population being tested and the need for replication, according to Dr. Corley. For example, initial results for the septin 9 methylated DNA blood test looked very good, with sensitivity of about 90% (), but after its testing in 14 populations, a meta-analysis showed that pooled sensitivity was just 67% ( ).
“Circulating tumor markers are an incredibly interesting target for screening, particularly because of their potential for being very specific for what you are looking for, and potentially markedly decreasing the subsequent follow-up that would need to be done for invasive tests such as colonoscopy,” Dr. Corley said. “However, this [CTC assay] really requires confirmation in screening populations, especially given some of the information we have from prior tests.”
Dr. Tsai’s team studied 327 patients with colorectal cancer of all stages, 111 patients with precancerous lesions (adenomas, advanced adenomas, carcinoma in situ/stage 0), and 182 healthy controls. All had blood drawn for the CTC assay before undergoing colonoscopy. Results of each test were ascertained with blinding to the results of the other.
CTCs are rarely shed into the circulation from precancerous lesions, with approximate density of only 1 per billion blood cells, Dr. Tsai said. The CMx assay (manufactured by CellMax Life) is able to detect these cells with high sensitivity through use of advanced technologies such as affinity-based microfluidics and a biomimetic surface coating.
The assay is performed with just 2 mL of whole blood. CTCs are defined as intact nucleated cells staining positive for CD20 and negative for CD45; they were combined with patient age in an algorithm, ultimately producing a risk score.
Study results showed that the CTC assay had an accuracy of 87.9% in the entire cohort, reported Dr. Tsai. The false-positive rate was just 3.3%, and the false-negative rate was 15.8%.
Sensitivity was 84.0% overall (76.6% for precancer and 86.9% for cancer), specificity was 97.3% overall (97.3% and 97.3%), and area under the receiver operating characteristic curve was 0.87 overall (0.84 and 0.88).
Dr. Tsai noted that the CTC assay’s sensitivity of nearly 77% for precancer compares favorably with that of a variety of other screening tests, such as the stool guaiac test for fecal occult blood (2%-10%), FIT alone (23.8%), and a stool DNA test combined with FIT (42%), and, in fact, falls within the range reported for colonoscopy (76%-94%).
The symposium was sponsored by the American Gastroenterological Association, the American Society for Clinical Oncology, the American Society for Radiation Oncology, and the Society of Surgical Oncology.
SOURCE: Tsai W et al., ASCO GI