Challenging pancreatic cancer-prone pedigrees: a nosologic dilemma

Our objective is to describe 11 pancreatic cancer (PC)-prone families, none of which are consonant with known hereditary cancer syndromes, in an attempt to portray familial aggregations of this disease that might be encountered in a clinical practice setting.

We selected 11 families containing two or more first- and/or second-degree relatives affected with PC from a registry of 200 PC-prone kindreds. Each proband and/or key relative(s) was interviewed and completed a detailed family history questionnaire (after providing informed consent) that allowed us to extend the pedigree as far as possible with retrieval of primary medical and pathology documents, whenever available.

All of the 11 families show PC features that merit clinical attention and raise questions as to whether this familial clustering could be due to "chance" alone, exposure to certain common environmental factors, such as cigarette smoking, and/or polygenic, multifactorial, or Mendelian inherited factors.

It is estimated that about 5% of PC may have a primary hereditary etiology. Because of early death, reduced penetrance, and often profuse phenotypic and genotypic heterogeneity, particularly with respect to variable age of onset and association with diverse patterns of cancer at different anatomic sites, the pedigrees require extension for ultimate diagnosis. Physician knowledge about PC's natural history and syndrome delineation should ultimately foster earlier diagnoses and possibly prevention of this disease. These high-risk patients may provide a source of DNA for formal linkage analysis in the search for culprit cancer-prone susceptibility loci.