Early detection of sporadic and hereditary pancreatic cancer using serum protein profiling.

Early detection of pancreatic cancer (PC) leads to lower staged tumors, with an increased chance on microscopic radical resection, which is a pre-requisite for attempting curative treatment. Detection of PC or its precursor lesions with routinely used imaging modalities; endoscopic ultrasound guided fine needle aspiration and/or serum tumor marker testing, lack sensitivity and specificity. This also hampers PC surveillance in individuals with a genetic predisposition for PC. A blood-based test for early detection of PC is an attractive alternative. Serum protein profiling with state-of-the-art mass spectrometry (MS) tools is an established approach for the development of new diagnostic tests based on discovery of clinical biomarkers. Previously, we have shown that the analysis of MS-based serum peptide and protein profiles allows for discrimination between PC patients and control persons. This discriminating PC protein profile was validated and applied to pinpoint individual PC cases in a surveillance setting. With the current research proposal we aim to apply complementary MS-based proteomics strategies for improved diagnostics (A), intensified surveillance programs (B), and testing individuals using dried blood spots (DBS) obtained using a simple finger-stick (c.q.finger-prick) (C). A. Improved diagnostics Previously, we reported peptide and protein profiles from serum samples that were collected following a completely standardized and robotized protocol. A discriminating MS-profile for PC was defined with a sensitivity of 76% and specificity of 91% (AUC of 92%). Classification was validated with a sensitivity of 93% and a specificity of 100% (AUC of 98%). To prepare for implementation, in the current proposal, based on both our and other experiences a set of discriminating peaks will be identified. We will develop a protocol to quantitatively measure this protein panel determined using amongst other bottom-up liquid chromatography(LC)-MS approaches. Moreover, we aim to map the proteoforms of the standard clinical biomarker CA19-9 using top-down MS strategies. This will be evaluated and validated in a large PC cohort series with joined effort of two specialized centers for PC treatment (Leiden, Rotterdam). B. Surveillance programs Recently, we have applied our ultrahigh resolution peptide and protein profile for the analysis of serum samples from 66 new individuals with a CDKN2A germline mutation who participated in a PC surveillance program (5 cases, 61 controls). The results showed that cases could be distinguished from controls with a higher median case-probability score than controls (0.26 vs 0.016; p=0.001). We concluded that these protein signatures are promising candidates for implementation in PC surveillance programs as an additional screening modality. To further validate this approach before implementation a multicentre study is required. Based on and in-line with our experience in the successful TESTBREAST study for hereditary breast cancer we here propose the TESTPAC (Trial Early Serum Test PAncreatic Cancer). C. Development of a DBS test for high-risk candidates (e.g. diabetes, smoking, chronic pancreatitis). Dried blood spots are collected using a finger-stick on a dedicated absorptive card. The advantage of DBS sampling is that it can be performed at the general practise and that samples can be shipped to our laboratory for analysis. The infrastructure for MS-analysis of DBS is available at the Leiden University Medical Center (LUMC), the methods still need to be developed. Moreover this method is applicable to conduct large European studies where shipping of samples is precair.