Pancreatic ductal adenocarcinoma (PDAC) is often diagnosed at advanced disease stages and invasive procedures are needed to confirm diagnosis. Innovative liquid biopsy-based approaches are warranted to diagnose early stages and prevent unnecessary treatment for advanced disease or misdiagnosed benign diseases. CA19.9 is the only clinically implemented biomarker, but its sensitivity is inadequate and it lacks specificity. Spliced messenger RNA (mRNA) of tumor-educated platelets (TEPs) emerged recently as new potent pan-cancer biomarker (Best et al, Cancer Cell 2015&2017). The aim of this study was to evaluate PDAC-specific TEP profiles as diagnostic and staging biomarker. Samples were collected at four international academic centers. Firstly, TEPs were isolated from blood of 135 PDAC patients including early disease (stage I-IIb), advanced disease (stage III) and metastatic patients (stage IV). In addition, 250 age-matched healthy controls (HC) and 72 patients with a benign disease in the pancreaticobiliary tract were included to create a heterogeneous control group that reflects the real patient population at the outpatient clinic. After total RNA was isolated, SMARTer amplification was performed and whole transcriptome mRNA was sequenced on the Illumina platform. Intron-spanning platelet mRNA reads were selected for further differential expression analysis. PDAC patients had a significant different mRNA TEP profile compared to HC (p-value <0.0001). Unsupervised clustering with our gene panel robustly separated HCs from of PDAC. In order to train our swarm-enhanced classification algorithm, all samples were allocated to either the training, evaluation or validation cohorts to determine and evaluate the optimal algorithm and gene panel. In our preliminary first validation cohort, our algorithm predicted PDAC from our control group with an accuracy over 90%. Moreover, we classified early stage disease correctly as PDAC, proving the diagnostic power of our method. This study underlines the value of profiling spliced mRNA of TEPs for cancer diagnostics. Here we showed the capability of TEPs to diagnose PDAC. By the swarm-enhanced classification algorithm, our method showed robust differences in PDAC and non-cancer spliced mRNA profiles, even at early stage disease. Future analyses including other hepatopancreaticobiliary cancers are needed to establish the discriminative power of our algorithm. Evaluation of premalignant lesions and treated patients is ongoing to prove its monitoring value, and large-scale multicenter validation is warranted to facilitate implementation in the clinic.

Citation Format: Tessa Ya Sung Le Large, Laura L. Meijer, Sjors G. In ‘t Veld, Nik Sol, Myron G. Best, Giulia Mantini, Ewoud van der Lelij, Francois Rustenburg, Heleen Verschueren, Niccola Funel, Ingrid Garajova, Maarten F. Bijlsma, Marc G. Besselink, Elisa Giovannetti, Tom Wurdinger, Geert Kazemier. Spliced messenger RNA of tumor-educated platelets yields a new diagnostic prospective for pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2596.