No abstract

Based on several case-control studies, it has been estimated that familial aggregation and genetic susceptibility play a role in up to 10% of patients with pancreatic cancer, although conclusive epidemiologic data are still lacking. Therefore, we evaluated the prevalence of familial pancreatic cancer and differences to its sporadic form in a prospective multicenter trial. A total of 479 consecutive patients with newly diagnosed, histologically confirmed adenocarcinoma of the pancreas were prospectively evaluated regarding medical and family history, treatment and pathology of the tumour. A family history for pancreatic cancer was confirmed whenever possible by reviewing the tumour specimens and medical reports. Statistical analysis was performed by calculating odds ratios, regression analysis with a logit-model and the Kaplan-Meier method. Twenty-three of 479 (prevalence 4.8%, 95% CI 3.1-7.1) patients reported at least 1 first-degree relative with pancreatic cancer. The familial aggregation could be confirmed by histology in 5 of 23 patients (1.1%, 95% CI 0.3-2.4), by medical records in 9 of 23 patients (1.9%, 95% CI 0.9 -3.5) and by standardized interviews of first-degree relatives in 17 of 23 patients (3.5%, 95% CI 2.1-5.6), respectively. There were no statistical significant differences between familial and sporadic pancreatic cancer cases regarding sex ratio, age of onset, presence of diabetes mellitus and pancreatitis, tumour histology and stage, prognosis after palliative or curative treatment as well as associated tumours in index patients and families, respectively. The prevalence of familial pancreatic cancer in Germany is at most 3.5% (range 1.1-3.5%) depending on the mode of confirmation of the pancreatic carcinoma in relatives. This prevalence is lower than so far postulated in the literature. There were no significant clinical differences between the familial and sporadic form of pancreatic cancer.

Peptide presentation by HLA class I and II antigens regulates specific antigen recognition by T cells. The present study aimed to investigate T cell infiltration and its relation to HLA antigen expression in pancreatic neuroendocrine tumors. Fresh tissue samples were collected from five insulinomas and six other neuroendocrine tumors (one gastrinoma, one glucagonoma, two carcinoid, and two neuroendocrine carcinomas). Normal pancreatic and splenic tissue samples were used as controls. Investigation of infiltrating lymphocyte populations, as well as staining of HLA class I and II antigens, were performed by standard immunohistochemistry. The majority of investigated tumors demonstrated an intratumoral infiltration by CD3+, CD4+ and CD8+ T cells that was significantly higher than in normal pancreatic islets. Only a minority of tumor-infiltrating T cells showed the CD45RO+ phenotype. The expression of HLA class I antigen was altered in 10 of 11 tumors. A loss of beta-2microglobulin represented the most frequent type of alteration to HLA class I expression, although the total loss of HLA class I was found in only one case of neuroendocrine carcinoma. HLA class II molecules were expressed by endothelial and lymphoid cells and not by tumor cells. In conclusion most neuroendocrine pancreatic tumors induce a T cell mediated immune response resulting in an intratumoral infiltration with CD3+, CD4+ and CD8+ T cells. Loss of beta-2microglobulin is a frequent alteration in these tumors, which may influence the normal function of the HLA class I antigen complex. In contrast to malignant tumors of the exocrine pancreas, expression of HLA class II was absent in neuendocrine pancreatic tumor cells.

Our established effector cells possess some regulatory features induced by unspecific TCR-activation that could be prevented by coculture with DCs. CIK cells have desirable properties for immunotherapeutical approaches, especially after coculture with DCs, which could be used additionally for induction of a specific immune response.

Our results suggest that Ad-p53 may synergistically enhance 5-FU-chemosensitivity most strikingly in pancreatic cancer cells lacking p53 function. These findings illustrate that the anticancer efficacy of this combination treatment is dependent on the p53 gene status of the target tumor cells.