To advance understanding of environmental, behavioral and genetic causes of brain tumors in adults, DCEG investigators initiated a multicenter case-control study of malignant and benign tumors in adults in 1994. This four-year study was conducted at participating hospitals in Boston, Phoenix, and Pittsburgh. Eligible cases were individuals greater than or equal to 18 years newly diagnosed with an intracranial glioma, meningioma or acoustic neuroma and treated at one of the participating hospitals. The controls were patients admitted to neurological, neurosurgical or general surgical services at the same three hospitals for any of a variety of non-neoplastic conditions. By the end of the study, 811 brain tumor cases had been accrued.

Information about a broad range of possible environmental, lifestyle, and genetic risk factors was obtained from both cases and controls through a computer-assisted personal interview (CAPI). The family history component obtained history and age at diagnosis of cancer or benign brain tumors and selected other diseases, for all living and deceased first degree relatives. A supplemental self-administered questionnaire covered diet, vitamin supplements, alcohol consumption, and household use of electrical appliances. Blood samples were obtained as a source of DNA. Currently, data analysis is in the early stages.

To increase our ability to examine both genetic and environmental components of brain tumor risk, we decided to add a family studies component to the case-control study, focusing on families of glioma cases. Initial contact with each family is made through the cases or, if a case is deceased, through the next of kin. Cases or next of kin are asked to complete a Family Health Questionnaire that updates the family medical history and provides contacting information for all adult first degree relatives and more distant relatives with cancer. Then, we contact all first degree relatives greater than or equal to age 18 years, and the next of kin of deceased eligible relatives and invite them to complete a modified risk factor interview conducted over the telephone. This interview obtains information about each relative's personal and family history of cancer and other diseases, and history of risk factor exposures, including all the major categories covered in the case-control study. Study participants who complete the interview are then asked to provide buccal cells as a source of DNA for future genotyping.

The glioma cases and their relatives will serve as a unique resource for both epidemiologic and genetic analyses. Selected relatives can serve as controls for association thereby eliminating concerns and population stratification. The study design also permits assessment of specific genetic hypotheses that cannot be evaluated in a traditional case-control study. Data from all first degree relatives of the glioma cases will be used in association studies and segregation analysis. In addition, we can screen DNA from members of multiplex families (families with 2 or more relatives with a primary CNS tumor) for mutations in candidate genes known to be associated with glioma, and contribute data from selected multiplex families to collaborative linkage studies to search for new genes conferring susceptibility to brain and possibly related tumors.