C. M. Court1,4, J. S. Ankeny1,4, S. Sho1, S. Hou1, P. Winograd1, Q. Li1, M. Song1, T. R. Donahue1,4, O. J. Hines1, H. A. Reber1, Z. A. Wainberg3, H. R. Tseng4, J. S. Tomlinson1,4 2University Of California – Los Angeles,Molecular And Medical Pharmacology,Los Angeles, CA, USA 3University Of California – Los Angeles,Hematology/Oncology,Los Angeles, CA, USA 4VA Greater Los Angeles,Surgery,Los Angeles, CA, USA 1University Of California – Los Angeles,Surgery,Los Angeles, CA, USA
Introduction: Current preoperative evaluation and staging of patients with pancreatic ductal adenocarcinoma (PDAC) is hampered by the limited sensitivity of cross-sectional imaging for micrometastatic disease. Furthermore, the majority of patients who undergo resection ultimately succumb to metastatic disease, suggesting that current staging systems are likely routinely understaging patients. Our goal was to investigate the utility of circulating tumor cells (CTCs) as a preoperative predictor of metastatic disease and postoperative outcomes.
Methods: A total of 49 patients were taken to the operating room for attempted pancreatic resection and enrolled in the study. 15 (30.6%) had undergone neoadjuvant therapy but were deemed to be surgical candidates based on post-treatment imaging. Four milliliters of venous blood (VB) was evaluated for the presence and number of CTCs using the microfluidic NanoVelcro chip. CTCs were defined by immunocytochemical staining (CK+ or CEA +, CD45-, DAPI+). CTC number was then correlated with both surgical outcomes and standard clinicopathologic findings.
Results: Of the 49 PDAC patients taken to the operating room, 11 (22.4%) were found to have metastatic disease intraoperatively, 6 with liver metastases and 5 with peritoneal metastases. CTC presence and number per 4 mL VB were found to correlate with stage and distinguished patients with resectable disease from those with metastatic disease. CTCs were present in all (100%) of the 11 patients with metastatic disease versus only 63% of those with resectable PDAC and at significantly higher levels (Average CTCs/4mL VB- 11.5 vs. 1.7, p<0.01). Using a cutoff of ≥3 CTCs/4mL VB, CTCs were able to distinguish patients with metastatic disease from those with resectable cancer with a sensitivity of 90.9%, specificity of 84.2%, PPV of 62.5%, NPV of 97.0%, and AUROC of 0.898 (95% CI = 0.796 – 0.999, p < 0.001). Of note, CA19-9 levels did not statistically differ between the groups (p=0.88). Furthermore, higher CTC counts correlated with worse recurrence-free and overall survival for both the entire cohort as well as the group of patients who successfully underwent an operation.
Conclusion: In this small prospective study, preoperative CTC enumeration demonstrated a correlation with the presence micrometastatic disease not detected by preoperative cross sectional imaging. Larger studies with longer follow-up are needed to firmly establish CTCs as a predictive biomarker in PDAC.
