Profiling of autoantibodies in sera of pancreatic cancer patients

Yosuke Nagayoshi, Masafumi Nakamura, Kazuhiro Matsuoka, Takao Ohtsuka, Yasuhisa Mori, Hiroshi Kono, Teppei Aso, Noboru Ideno, Shunichi Takahata, Akihide Ryo, Hiroyuki Takeda, Tetsuhide Ito, Yoshinao Oda, Yaeta Endo, Tatsuya Sawasaki, Masao Tanaka

Research output: Contribution to journalArticlepeer-review

23 Citations (Scopus)


Background. Although autoantibodies to cancer antigens are candidates for biomarkers, no comprehensive studies to detect cancer-specific antibodies have been performed. This study identified autoantibodies in the sera of pancreatic cancer (PC) patients using proteomics based on a wheat germ cell-free protein production system. Methods. We constructed a biotinylated protein library of 2,183 genes. Interactions between biotinylated proteins and serum antibodies were detected by AlphaScreen® assay. Relative luminescence signals of each protein in 37 PC patients and 20 healthy controls were measured, and their sensitivity and specificity for PC were calculated. Results. Luminescence signals of nine proteins were significantly higher than those of healthy controls, with calcium and integrin binding 1 (CIB1) protein showing the greatest significance (p = 0.002). Sensitivity, specificity, positive predictive value and negative predictive value of CIB1 autoantibody alone for PC were 76, 70, 82, and 61 %, respectively, and 97, 35, 74, and 88 %, respectively, when the four most significant proteins were combined. Presence of these autoantibodies did not vary significantly with other clinicopathological characteristics. Conclusion. Several autoantibodies, including CIB1, are potential biomarkers for PC.

Original languageEnglish
Pages (from-to)S459-S465
JournalAnnals of Surgical Oncology
Issue numberSUPPL. 3
Publication statusPublished - 2014

All Science Journal Classification (ASJC) codes

  • Surgery
  • Oncology


Dive into the research topics of 'Profiling of autoantibodies in sera of pancreatic cancer patients'. Together they form a unique fingerprint.

Cite this