The need for effective pancreatic cancer detection and management: a biomarker-based strategy

Abstract

Pancreatic cancer (Pa) is generally a very aggressive disease, with few effective approaches available for early diagnosis or therapy. These factors, combined with the aggressiveness and chemoresistance of Pa, results in a bleak outcome post-diagnosis. Cancer-related biomarkers have established capabilities for diagnosis, prognosis and screening and can be exploited to aid in earlier less-invasive diagnosis and optimization of targeted therapies. Pa has only one US FDA-approved biomarker, CA19-9, which has significant limitations. Hence, it is vital that novel biomarkers are identified and validated to diagnose, treat, control and monitor Pa. This review focuses on existing and potential Pa-associated markers and discusses how they may be applied in cohort for improved management of Pa.

Keywords:

• biomarkers
• CA19-9
• diagnostics
• pancreatic cancer
• therapeutics
• tumor

Financial & competing interests disclosure

AS Crawley is supported by Irish Research Council Postgraduate Government of Ireland, grant code GOIPG/2014/165. RJ O’Kennedy is supported by the Science Foundation Ireland, grant code 10/CE/B1821. RJ O’Kennedy has carried out research for a number of companies including Randox Teo and Biosurfit. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Key issues

• Pancreatic cancer (Pa) is associated with a very poor survival rate. Issues and limitations affect both disease detection and treatment. Pa is rarely caught prior to stage 4 and the general treatment, gemcitabine, is primarily for palliative purposes only. Of the 20% of patients with resectable Pa, the majority will suffer disease relapse. Diagnosis is difficult and the methods are not sufficient.

• The biological mechanisms of pancreatic cancer are incredibly intrinsic. It is known to develop from the accumulation of a wide variety of genetic mutations including, but not limited to, the genes KRAS, p16, TP53 and SMAD4. The cancer often originates from small pre-malignant lesions, such as pancreatic intraepithelial neoplasia, intrapancreatic mucinous neoplasia lesions, which with the accumulation of mutations can eventually lead to fully invasive Pa. While the exact cause is unknown, it is thought it is linked to many risk factors, the most prominent of which is smoking. Moreover, up to 10% of cases can have a genetic predisposition. The untreatable reputation of this cancer derives from its chemoresistance from cancer stem cells and the formation of the desmoplastic stroma (a hallmark of Pa), which causes a decrease in the penetration abilities of the therapeutic agents into the tumor microenvironment. Elements such as these require targeting for a more successful therapy.

• Biomarkers have previously proven applications. Markers from tissue/serum may have the potential to improve how effectively Pa is diagnosed and treated. The current Pa diagnostic marker, CA19-9, has limitations and other markers need to be validated and included with CA19-9 to increase the chances of early diagnosis. Determining the markers present and monitoring their levels can provide information on the cancer origin, the ideal course of treatment, may assist in the choice of treatments, establish the level of aggressiveness of the cancer, assist monitoring for recurrence and predict survival. Thus, markers and their expression profiles must be thoroughly investigated and validated for clinical utility in Pa.

• Research has indicated that certain markers may have potential clinical ability. Pa has a wide range of diagnostic, prognostic and predictive biomarkers. Diagnostic markers include, KRAS, miRNA, CEA, OPN and TIMP-1. SMAD4, AACT and MUCINs can be useful prognostic markers, while MSLN, EGFR and integrins are potentially beneficial predictive markers. Each marker provides individual data that can be utilized for the improved diagnosis and treatment of Pa. While none of these markers thus far can be used alone, much work has shown promise for using combinations of these markers.

• Recent developments in pancreatic biomarker studies have expanded our knowledge of the applications of these biomarkers. In particular, next-generation sequencing is a very beneficial technology that can be used to determine genetic mutations specific to an individual. The applications of this technology can detect genetic mutations known to be cancer-causing (risk evaluation), identify early mutations benefiting early diagnosis and potentially indicate optimum targets for improved therapies.

• This review provides a hypothetical approach, termed the ‘multi-marker-multi-panel’ approach, for potentially improving the diagnosis and treatment of Pa. This approach exploits the idea of using various biomarkers in combinations that will provide a myriad of beneficial data from a patient’s blood and tumor.