Abstract 507: Prediction of colorectal cancer recurrence with plasma miRNAs in the ColoCare Study

Proceedings: AACR 107th Annual Meeting 2016; April 16-20, 2016; New Orleans, LA Background Colorectal cancer is the third most common cancer and third most common cause of cancer-related death in the US. Recurrence of primary tumors occurs in 10-40% of patients depending on tumor stage. Early detection of recurrence can increase the treatment options for those eligible for curative therapy, which may lead to better survival. We determined the levels of multiple different plasma miRNAs and assessed their association with clinical outcomes to identify specific miRNAs that may serve as predictors of CRC recurrence. Methods Colorectal cancer patients (age 18-80, newly diagnosed, stage I-IV) were recruited as part of the ColoCare Study in Seattle, Washington, between 2007 and 2011. Patients who had a blood sample taken >30 and <365 days after surgery and prior to any recurrence were included in the study (n = 83). Candidate miRNAs (miRNA 31, 141, 200b, 203, 16, 17-3p, 29a, 92a, 125b) were quantified using pre-designed TaqMan microRNA assays in triplicate assays. Plasma miRNA expression was normalized and expressed as copy number. Cox proportional hazards regression with robust standard errors was used to estimate hazard ratios of recurrence associated with level of each miRNA, adjusted for age, sex and tumor site and stratified by stage. No recurrences were observed in stage I patients. Statistical analyses were conducted separately for stages II-III and II-IV colorectal cancer. Results During a median of 1,067 days (33-2,254 days) follow-up after surgery, 20 recurrences were observed. In stage II-III patients, the presence of detectable circulating levels of miR-141 (HR = 4.30 [1.10-16.84]) or miR-203 (HR = 8.04 [1.39-46.4]) were associated with risk of recurrence. In stage II-IV patients, detectable miR-203 (HR = 3.52 [1.13-10.96]) was associated with recurrence, and elevated miR-29a levels were associated with greater risk of recurrence (HR: 1.80 per natural log unit, 95% CI: 1.06-3.04). Conclusion The present study identified multiple plasma miRNAs that may help to predict those patients at greatest risk of colorectal cancer recurrence. Our results require validation in order to determine the role for such plasma markers in clinical practice. Citation Format: Brandon Dickinson, Scott V. Adams, Cornelia Ulrich, Kathy Vickers, Michelle Wurscher, William M. Grady, Karen W. Makar. Prediction of colorectal cancer recurrence with plasma miRNAs in the ColoCare Study. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 507.