Abstract 2579: Genomewide transcriptomic profiling identifies a novel miRNA signature for predicting lymph node metastasis in patients with pancreatic ductal adenocarcinoma

Abstract

Abstract Purpose: Pancreatic ductal adenocarcinoma (PDAC) is an extremely lethal cancer, and is projected to become the second leading cause of cancer-related deaths in the US by 2030. The lymph node (LN) status is an important predictor of survival outcomes in PDAC patients undergoing curative resection. The National Comprehensive Cancer Network guidelines recommend neoadjuvant therapy in patients with high-risk features such as large regional LNs, elevated CA-19-9 levels, large primary tumors, excessive weight loss, and extreme pain. Therefore, a pre-treatment diagnosis of LN metastasis (LNM) is critical in developing a more personalized treatment strategy in PDAC patients. Herein, for the first time, we performed a genomewide analysis to identify and develop a miRNA signature for the detection of LNM in PDAC patients. Methods: We analyzed a total of 518 PDAC patients, which included 269 patients from three genomewide miRNA-expression datasets (TCGA, GSE24279 and GSE35688) and 249 from two, independent clinical patient cohorts. During the biomarker discovery phase, we used rigorous computational and bioinformatic approaches to identify candidate miRNA biomarkers, which later were subjected to in-silico validations. The robustness and performance of this miRNA-signature was subsequently interrogated in two independent clinical cohorts (training cohort: n=150, validation cohort: n=99) using qRT-PCR assays. Furthermore, the clinical potential value of the miRNA signature was evaluated and compared to several clinicopathological factors using logistic regression analysis. Results: Through a comprehensive miRNA expression profiling analysis of LNM-positive and negative patients, we identified a panel of 7 miRNAs that significantly predicted LNM [the area under curve (AUC)=0.76, 0.75 and 0.92, respectively]. Using logistic regression analysis, we then optimized and trained a 6-miRNA risk-prediction model in the training cohort of patients, which robustly distinguished PDAC patients with LNM (AUC: 0.84, 95%CI: 0.76-0.89). The performance of this trained model was subsequently confirmed in an independent, large clinical validation cohort (AUC: 0.73, 95%CI: 0.64-0.82). Interestingly, multivariate analysis which incorporated several pre-operative clinical factors, revealed that our 6-miRNA signature was as an independent predictor of LNM in both patient cohorts (training cohort: OR=34.97; P<0.01, validation cohort: OR=10.11; P<0.01, respectively). Conclusion: We identified and developed a novel 6-miRNA signature that is highly robust in predicting presence of LNM in PDAC patients; highlighting the clinical impact of these findings in more appropriate patient selection and development of improved individualized treatment strategies for PDAC patients. Citation Format: Satoshi Nishiwada, Kensuke Yamamura, Raju Kandimalla, Takahiro Akahori, Kota Nakamura, Hideo Baba, Masayuki Sho, Ajay Goel. Genomewide transcriptomic profiling identifies a novel miRNA signature for predicting lymph node metastasis in patients with pancreatic ductal adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2579.