Identification of prognostic factors for patients with advanced pancreatic cancer by proteomics
4615 Background: Gemcitabine monotherapy is the current standard for patients with advanced pancreatic cancer (PC). Its efficacy, however, varies significantly depending on individuals. This study was aimed at discovering a new diagnostic biomarker that can estimate the outcome of patients after receiving the therapy. Methods: All patients included in this study (304 patients) had metastatic PC and received at least two cycles of gemcitabine monotherapy. We compared the baseline plasma proteome between representative 29 short-term survivors (survived for less than 100 days) and 31 long-term survivors (survived for more than 400 days) using quantitative mass spectrometry (MS). Results: Among a total of 45,277 peptide peaks, we identified 637 peaks whose intensities were significantly different (p<0.001, Welch's t-test). The 2 MS peaks with the highest statistical significance (p=2.57×10-4 and 5.03×10-4) were revealed to be derived from α1-antitrypsin (AT) and α1-antichymotrypsin (ACT), respectively, by tandem MS. The levels of AT and ACT, WBC count, platelet count, alkaline phosphatase, and ECOG performance status were selected using a forward stepwise procedure by Akaike's information criterion, and a scoring system (nomogram) was constructed to estimate the prognosis of individual patients. Among the selected parameters the AT level was found to be the second most significant contributor to the nomogram (p=0.0003; Table ). This survival prediction model was internally validated using a bootstrap approach with 200 resamples. Conclusions: Our survival prediction model including values of AT and ACT seems to have high practical utility and may lead to tailoring the treatment of patients with advanced PC. Modification of therapeutics may need to be taken into consideration for patients with increased AT and ACT. [Table: see text] [Table: see text]