SAN FRANCISCO – Circulating tumor cells may be useful for ascertaining both the presence and extent of pancreatic cancer, according to the findings of a prospective cohort study reported at the annual Gastrointestinal Cancers Symposium sponsored by the American Society of Clinical Oncology.
A "nano-Velcro" assay to find circulating tumor cells (CTCs) was evaluated prospectively in 61 patients with suspicious pancreatic masses, two-thirds of whom were found to have pancreatic adenocarcinoma.
Finding at least one CTC/2 mL of blood had 71% sensitivity and 95% specificity for the presence of pancreatic cancer, and, among the patients with cancer, finding at least two CTCs/2 mL of blood had 69% sensitivity and 96% specificity for the presence of metastases, reported Dr. Jacob S. Ankeny of the department of surgery, University of California, Los Angeles.
Dr. Jacob S. Ankeny
"I think CTCs do have a potential role in aiding in our diagnosis of pancreas cancer and in discriminating which patients are at higher risk of having metastatic disease at the time of their disease presentation," said Dr. Ankeny.
"In our study, circulating tumor cells actually outperformed CA 19-9 [cancer antigen 19-9] for differentiating locoregional vs. metastatic disease, and as a result, I think they show marked promise as a biomarker at the time of disease presentation so that we can perform pretreatment staging and implement more effective initial therapies based on patient stage," he added.
Study limitations included the small sample size and the lack of outcome data, Dr. Ankeny acknowledged. "Additionally, we had minimal inclusion of inflammatory pancreatic conditions and didn’t have any patients with acute pancreatitis. It’s known that some other inflammatory conditions actually lead to false-positive results in some other circulating tumor cell platforms."
A key challenge to using CTCs is their rarity, according to invited discussant Rebecca A. Miksad of Harvard Medical School and the Beth Israel Deaconess Medical Center, both in Boston.
"In the data presented, the difference between cancer and no cancer is one CTC; similarly, the difference between metastatic and locoregional cancer is one CTC. Because CTCs are detected only in whole numbers, these close cutoffs raise questions about the accuracy, reliability, and reproducibility; in other words, will the same result be found in repeated analysis of the same sample and in different samples from the same patient," she commented.
"CTCs in pancreatic cancer are intriguing, although the role remains unclear," Dr. Miksad concluded. "The presented CTC data has a narrow diagnostic predictive window, and we may find with further work that the utility of CTCs as a test to detect the presence of pancreatic cancer may differ from its utility to assist with staging."
The nano-Velcro assay uses antibody-coated silicon nanowires to capture circulating cells with high efficiency, Dr. Ankeny explained. The captured cells are stained immunocytochemically, scanned with fluorescence microscopy, and counted.
In the study, a tumor cell was defined as a cell measuring at least 10 microns that had a DAPI-staining nucleus, was positive for cytokeratin, and was negative for the leukocyte marker CD45.
The investigators tested duplicate 2-mL aliquots from the study patients at the time of their presentation with suspicious pancreatic masses.
The 41 patients determined to have pancreatic adenocarcinoma were roughly equally distributed across tumor stages II, III, and IV. The other 20 patients had more than half a dozen pancreatic conditions, such as intraductal papillary mucinous neoplasms, mucinous cystic neoplasms, and serous cysts.
The assay identified CTCs in 71% of the patients with cancer, reported Dr. Ankeny, who disclosed no conflicts of interest related to the research.
A single patient in the control group was found to have a CTC, meaning a false-positive result. "That was a patient with a nondiagnostic FNA [fine-needle aspiration] of a large cyst containing a large solid component," he explained. "Unfortunately, the patient went elsewhere for surgery and refused surgery, and we don’t have any follow-up pathology. So it’s a theoretical false-positive in our non–adenocarcinoma group."
At an optimal diagnostic cutoff of at least one CTC/2 mL blood, the assay had a sensitivity of 71%, a specificity of 95%, a positive predictive value of 97%, and a negative predictive value of 59% for the presence of pancreatic cancer.
The area under a receiver operating characteristic curve was 0.84 (P less than .001), indicating good discrimination between patients with and without cancer, Dr. Ankeny said.
Among the patients with cancer, at the optimal cutoff of two or more CTCs/2 mL of blood, the assay had a sensitivity of 69%, a specificity of 96%, and a positive predictive value of 92% for the presence of metastases.