Pancreatic ductal adenocarcinoma (PDAC) has a 5-year survival of only 3%. Surgery is the only curative option but only 15-20% of patients are suitable and post-operative recovery is long, with significant early mortality and a 5-year survival of around 20%. Thus not all patients benefit from surgery and prognostication and subsequent stratification of patients to surgical or non-surgical management could be better. Initial staging currently relies on imaging; the main pathological prognostic factors are not known until after resection. Initial diagnosis usually involves pancreatic cytology samples. Prognostication is possible in such samples using biomarkers such as miR-21. Prognostication on cytology is not yet clinically routine, perhaps because previous research methodology has involved tissue extracts e.g. RT-PCR and not in situ techniques, as often preferred by pathologists. Our aim was to study the expression of miR-21 in resected PDAC by in situ hybridisation and its relationship to clinico-pathological parameters including prognosis.


We studied PDACs from 69 patients, using tissue microarrays (TMAs). Chromogenic in situ hybridisation (CISH) for miR-21 was performed using an EXIQON kit. Staining was scored based on intensity as: negative or weak (low expression); or moderate or strong (high expression). Kaplan-Meier survival curves were generated for progression free survival (PFS) and overall survival (OS) and correlated with miR-21 scores using univariate and multivariate COX proportion hazard models.


miR-21 expression on CISH was cytoplasmic and mainly in neoplastic PDAC epithelium and stroma; further analysis used epithelial expression data only. miR-21 was highly expressed in 39 (56%) patients; miR-21 was low or absent in 30 (44%) patients. Median PFS was 9 months (3.8-14.2, 95% CI) and 19 months (12.1-26.9, 95% CI) in the high and low expression groups respectively (P<0.002). Median OS was 15 months (12.9-18.1, 95% CI) and 24 months (18.5-28.8, 95% CI) in the high and low expression groups respectively (P<0.001). On univariate analysis with standard clinico-pathological factors, miR-21, tumor grade and lymph node invasion were prognostic for PFS and OS; on multivariate analysis, only high miR-21 was prognostic, being associated with poor PFS (HR 1.8, p<0.01) and OS (HR 2.1, p<0.01).


High epithelial expression of miR-21 assessed by CISH is independently predictive of poorer overall and progression-free survival in patients with PDAC. Prognostication using miR-21 in diagnostic cytology samples is theoretically possible. CISH enables tissue localisation of the biomarker, in this case miR-21. Use of CISH might facilitate adoption of such stratification methods and enable improved allocation of patients with into surgical or non-surgical treatment groups.

Note: This abstract was not presented at the meeting.

Citation Format: Asif Ali, Elisa Giovannetti, Niccola Funel, Roderick Ferrier, Jennifer Morton, Karin A. Oien. miR-21 overexpression assessed by in situ hybridisation is an independent predictor of survival in patients with resected pancreatic ductal adenocarcinoma. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2860. doi:10.1158/1538-7445.AM2014-2860