The best magazine
Methods
Patients
The Breast Cancer Family History Clinic (FHC) in South Manchester, established in 1987, collects information (demographic, pedigree, screening and disease symptomatology) on individuals and families with a family history of breast and ovarian cancer. Families within the north-west region of England who have a family history of breast or ovarian cancer are referred to the FHC by their general practitioners. Women who attend the FHC have a detailed family tree elicited including all first, second and if possible third degree relatives. The genetic status of all family members is recorded (BRCA1, BRCA2 and negative results) if testing has occurred. Details of all tested and untested women and relatives are entered onto a database.
Data on women in this prospective study were verified against medical records, NHS summary care records and cancer registrations made by the North West Cancer Intelligence Service (NWCIS). All cases of ovarian cancers were confirmed by means of hospital/pathology records, cancer registrations (from 1960) or death certification. The date the patient first attended the FHC was considered the ascertainment date. Follow-up was censored at the latest date from 31 December 2010; ovarian cancer diagnosis; oophorectomy; or date of death (obtained from death certification either directly or via NWCIS). Women were excluded if they had ovarian cancer or oophorectomy prior to first referral to the FHC. All data on cancer incidence was collected prospectively.
Mutation Testing
Families with an available living family member with breast or ovarian cancer were eligible for NHS testing if there was at least a 20% likelihood of a mutation as per National Institute for Health and Clinical Excellence (NICE) guidelines. Testing was carried out by sequencing and multiple ligation dependent probe amplification (MLPA), which we have shown has high sensitivity in detecting mutations. Testing was also available through a research study (Familial Breast Cancer Study (FBCS)) which had much looser criteria with only two affected relatives with breast cancer usually being required. This study used a combination of conformation sensitive gel electrophoresis in 86 fragments and MLPA. Through the FBCS study, 698/1140 (61%) families were tested. These samples were also tested for the 1100DelC mutation in CHEK2.
Analysis
Women were grouped by genetic status and by ovarian cancer type (invasive epithelial or borderline). Person-years at risk were calculated from the woman's ascertainment date to date censored. The expected numbers of ovarian cancer cases over the follow-up period, for each family genetic status (BRCA1 positive, BRCA2 positive, BRCA negative, BRCA untested), were calculated based on person-years at risk using population level data for the north-west of England from NWCIS. Women testing negative for a family BRCA1/2 mutation were excluded from this calculation as there were few in number and there were no ovarian cancer events. The relative risk (RR) of developing ovarian cancer was calculated as the number of observed cases divided by the expected number of cases in the general population. Subgroups defined by their personal or family genetic status and ovarian cancer type (invasive epithelial or borderline) were analysed separately. Exact confidence limits for the relative risk were calculated using the Poisson distribution. Statistical results are presented as the main effect with a 95% CI unless otherwise stated. All calculations were performed with Stata V.12 (http://www.stata.com).
Source: ...