1 Department of Pharmaceutical and Biomedical Sciences (FARMABIOMED), University of Salerno, Fisciano (SA), Italy2 BIOUNIVERSA, University of Salerno, Fisciano (SA), Italy3 IRCCS “Casa Sollievo della Sofferenza”, San Giovanni Rotondo, Dipartimento di Scienze Chirurgiche, San Giovanni Rotondo (FG), Italy4 Dipartimento di Chirurgia Generale e Trapianti d’Organo, La Sapienza Università di Roma e Umberto I Policlinico di Roma, Rome, Italy5 Department of Biochemistry and Biophysics, Second University of Naples, Naples, Italy6 Department of Oncology, Fatebenefratelli Hospital, Benevento, Italy7 Department of Experimental and Clinical Sciences, University G. D’Annunzio and Fondazione G. D’Annunzio, Ce.S.I., Chieti, Italy8 ARC-NET Centre for Applied Research on Cancer and Department of Pathology and Diagnostics, University and Hospital Trust of Verona, Verona, Italy9 Pancreas Unit, Department of Digestive Diseases and Internal Medicine, Sant’Orsola- Malpighi Hospital, Bologna, Italy10 These authors contributed equally to this work
Checking for direct PDF access through Ovid
To the Editor: Pancreatic ductal adenocarcinoma (PDAC) is the only one of the five most lethal malignancies (lung, colorectal, breast, pancreatic, and prostate cancer) for which both the incidence rate and death rate have increased in recent years, and with a projected increase in the number of deaths. The lack of early symptoms results in late-stage detection and a near to 100% mortality rate, despite improvements in therapies (1,2,3). Therefore, there is an urgent need for novel screening strategies to recognize early cancers or precursor lesions in the general population, and in individuals that are considered at highest risk on the basis of clinical or genetic characteristics. Owing to the vast genetic heterogeneity of this cancer, no single biomarker exists that is strongly correlated with its diagnosis. Efforts for detecting pancreatic adenocarcinoma at early stages are, therefore, focused on the identification of a large panel of appropriate sensitive and specific serum biomarkers, including some already characterized candidates and clearly novel ones (1,4).Recently, we reported that the intracellular anti-apoptotic protein BAG3 (5) was expressed in all 346 PDAC tumor samples examined, and not in the surrounding non-neoplastic tissue or in normal pancreas (6). Furthermore, in a cohort of 66 patients who underwent R0 resection, the level of BAG3 expression inversely correlated with patient survival (6). These results prompted us to investigate whether BAG3 protein was also present in patients’ sera and could potentially be used as a biomarker for PDAC.In a preliminary analysis by western blotting, we detected BAG3 protein in sera from four PDAC patients, but not in those from four healthy donors (Figure 1a). Furthermore, we found that BAG3 was complexed with anti-BAG3 antibodies: indeed, by dissociating such complexes and isolating antibodies, we demonstrated that these recognized BAG3 in western blotting (Figure 1b). We, therefore, decided to verify the presence and measure the levels of BAG3 and BAG3/antibody immune complexes (IC) in serum samples of a higher number of patients, by enzyme-linked immunosorbent assay. We analyzed 52 pretreatment serum samples from PDAC patients (stage:T3; average age+s.e.:64.0+1.3; males:29, females:23), 44 samples from healthy subjects (average age+s.e.:58.7+1.6; males:31, females:13), 17 sera samples from patients having proven chronic pancreatitis (average age+s.e.:51.7+4.2; males:14, females:3), 28 sera from patients having main duct and/or branch duct intraductal papillary mucinous neoplasm (IPMN) (average age+s.e.:66+4.2; males:13, females:15) and 19 sera from resected IPMN patients with histologically demonstrated carcinoma in situ (average age+s.e.:72+2.1; males:8, females:11). Levels of BAG3 (Supplementary Figure 1) and BAG3 IC (Figure 1c) were significantly higher in PDAC patients’ than in healthy donors’ or pancreatitis patients’ samples. Furthermore, also the levels detected in patients affected by IPMN or IPMN having carcinoma in situ were significantly (P<0.05) higher than those measured in healthy donors’ or pancreatitis patients’ samples (Figure 1c). The test for BAG3 IC (Figure 1c) discriminated better in the different groups than that for BAG3 protein (Supplementary Figure 1). It is not clear at this stage why this is the case; it is possible that at the time of the diagnosis most of the protein is bound to antibodies reducing the discriminating potential of the protein detection test. The test discriminated between neoplastic lesions (PDAC, Ca in situ and IPMN) and healthy subjects or pancreatitis patients, with a sensitivity of 71% and a specificity of 71% at a cutoff value of 0.078 AU (P<0.