A systematic review of serum autoantibodies as biomarkers for pancreatic cancer detection

Abstract

Pancreatic cancer is a leading cause of cancer-related deaths in the western world. Patients with pancreatic cancer have poor prognosis, partly due to difficulties in detecting it at early stages. While different markers have been associated with pancreatic cancer, many of them show suboptimal sensitivity and specificity. Serum autoantibodies against tumor-associated antigens have recently emerged as early stage biomarkers for different types of cancers. Given the urgent need for early and reliable biomarkers for pancreatic cancer, we undertook a systematic review of the published literature to identify primary articles that evaluated serum autoantibodies in pancreatic cancer detection by searching PubMed and ISI Web of Knowledge. Two reviewers extracted data on study characteristics and results independently. Overall, 31 studies evaluating 124 individual serum autoantibodies in pancreatic cancer detection met the inclusion criteria. In general, single autoantibody markers showed relatively low sensitivities at high specificity. A combination of markers, either multiple serum autoantibodies or serum autoantibodies combined with tumor-associated markers, led to a better diagnostic performance. However, most of the analyzed autoantibodies have only been reported in single studies and therefore need to be independently validated. We conclude that serum autoantibodies might present an option as biomarkers for early detection of pancreatic cancer, but more work is needed to identify and validate autoantibody signatures that are associated with early stage pancreatic cancer.

Keywords: autoantibodies; biomarkers; early detection; pancreatic cancer; systematic review.

Figure 1. Overview of the literature search process (until 27th of April 2015)

Figure 2. Graphical representation of sensitivity versus specificity of all analyzed autoantibodies

Sensitivity is plotted on the y-axis while on the x-axis the false positive rate is presented (100 - Specificity). Autoantibodies evaluated in one study only are labeled in blue. Autoantibodies evaluated in more than one study are labeled in red (p53), green (Annexin A2) and orange (PGK1). Autoantibodies that showed high sensitivity and specificity are labeled directly on the graph. Abbreviations: CLP: Coactosin-like protein; ENOA: Alpha-enolase; PGK1: Phosphoglyceratekinase 1.