PERSPECTIVES ON THE CLINICAL APPLICATION OF LEWIS ANTIGENS IN PANCREATIC CANCER: TOWARDS A PERSONALIZED TUMOR MARKER STRATEGY

Background. Carbohydrate antigen 19.9 (CA19.9) is used as a tumor marker in gastrointestinal cancers, assuming that it is produced by cancer cells. Recent data has suggested that, in pancreatic ductal adenocarcinoma (PDAC), elevation of CA19.9 could depend on its reabsorption in the bloodstream. No specific association exists between cancer and CA19.9 with respect to the other two type 1 chain Lewis antigens, namely Lewis a (Lea) or Lewis b (Leb), because each one can be reabsorbed in the blood, according to their specific and individual expression. This thesis aims to verify if Lea and Leb can support CA19.9 in the management of pancreatic cancer, and to assess the relationship between the amounts and pattern of Lewis antigens detected in the pancreatic tissue (both normal and neoplastic) and circulating in the blood of patients. Methods. Serum and fragments of surgical resection of PDAC patients are being collected. ELISA was performed on sera of 118 patients and 52 healthy controls, using anti-CA19.9, anti-Lea and anti-Leb antibodies. Real Time PCR (qPCR) was performed on cDNA of 14 tissue specimens (normal and cancer counterpart) to assess the level of each glycosyltransferase involved in the synthesis of Lewis antigens. PCR on genomic DNA evaluated the presence of FUT2/FUT3 null alleles on 18 tissue’s patients. Immunofluorescence (IF) was performed on 18 normal and cancer counterpart tissue sections using anti-CA19.9, anti-Lea and anti-Leb antibodies. Results. In 14 out of 118 patients (12%), ELISA on sera showed an increase in CA19.9 and/or Lea with respect to healthy controls. In all the patients, no significant statistical difference in glycosyltransferases expression was assessed by qPCR between normal and cancer tissue. 11 out of 18 patients (61%) patients were found to be FUT2-/+ and/or FUT3-/+ but this does not always correlate with a reduction of Lewis antigen expression. IF did not show a correlation between serum and tissue levels of Lewis antigens. Conclusions. These results show that Lea could be used as a management marker in 12% of PDAC patients that underwent surgical resection. It could be an alternative marker in patients that are CA19.9 negative, and a complementary marker in those CA19.9 borderline positive.