The differential effects of statins on the risk of developing pancreatic cancer: a case-control study in two centres in the United Kingdom.

INTRODUCTION: There are plausible biological mechanisms for how statins may prevent pancreatic cancer, although the evidence from epidemiological studies in the general population is conflicting. This study aims to clarify whether statins exert their effects in specific sub-groups, namely, gender, smoking status and diabetes. METHODS: A matched case-control study was conducted in patients diagnosed with pancreatic cancer, and a group of dermatology patients of similar ages and gender, diagnosed with basal cell carcinoma. Participants' medical records were reviewed for information on statin use prior to diagnosis. Odds ratios and 95 % CIs for the development of pancreatic cancer were estimated using conditional logistic regression. Subgroup analysis was performed in men, women, smokers and those with type 2 diabetes. RESULTS: Two hundred fifty-two cases (median age 71 years, range 48-73 years, 51 % women) and 504 controls were identified, of which 23 % of cases were regular statin users versus 21 % of controls. In the general study population there was no association between pancreatic cancer and regular statin use (OR 0.82, 95 % CI 0.53-1.23, p = 0.33). However, in male smokers, regular statin use was associated with significantly reduced odds of pancreatic cancer compared to male smokers not prescribed a statin (OR 0.11, 95 % CI 0.01-0.96, p = 0.05). In patients with type 2 diabetes statins use was not associated with reduced odds (OR 0.92, 95 % CI 0.35-2.45, p = 0.80), with no gender effects. CONCLUSIONS: In male smokers, statins may reduce the odds of pancreatic cancer. Statin use should be measured in aetiological studies of pancreatic cancer but analysed in specific sub-groups. Future work should investigate statins as chemopreventative agents in this high risk sub-group.

Allocation of epiblast cells to germ layer derivatives during mouse gastrulation as studied with a retroviral vector.

The embryonic ectoderm, or epiblast, is the source of the three primary germ layers that form during gastrulation in the mouse embryo. Previous studies have investigated the fate of epiblast cells in early gastrulation stages using clonal analysis of cell lineage and in late gastrulation stages using transplantation of labeled grafts. In this study, we studied the fate of late gastrulation stage epiblast using a clonal analysis based on a retroviral vector encoding the Escherichia coli lacZ gene. We found that by reducing the volume of viral suspension injected into each embryo, it was possible to achieve single infectious events. Our analysis of 20 embryos singly infected at the late streak stage and 21 at the head fold stage revealed clonal descendants in only a single germ layer in each embryo. These results indicate that allocation of epiblast progenitors to a single germ layer fate has occurred by late gastrulation in mouse embryos.

Relationship between the debrisoquine hydroxylase polymorphism and cancer susceptibility.

There have been a series of reports on the association of a genetic polymorphism at the cytochrome P450 CYP2D6 gene locus with cancer susceptibility. Many of these reports have remained contradictory either because of small numbers of patients studied or because of the limitations and controversy surrounding the pharmacokinetic assay used to identify affected individuals (poor metabolizers; PMs). We have recently developed a DNA-based assay that will allow the unequivocal identification of poor metabolizers and have applied this to the study of 1635 patients with different forms of cancer. Out of 361 lung cancer patients studied no statistically significant change in the proportion of PMs relative to controls was found. However, a significant increase in the proportion of poor metabolizers or heterozygotes was seen in leukaemia, bladder cancer and melanoma patients. This could be explained by a role for CYP2D6 in carcinogen detoxification or by linkage to another cancer-causing gene.