Mammographic surveillance in women younger than 50 years who have a family history of breast cancer: tumour characteristics and projected effect on mortality in the prospective, single-arm, FH01 study.

BACKGROUND: Evidence supports a reduction in mortality from breast cancer with mammographic screening in the general population of women aged 40-49 years, but the effect of family history is not clear. We aimed to establish whether screening affects the disease stage and projected mortality of women younger than 50 years who have a clinically significant family history of breast cancer.
METHODS: In the single-arm FH01 study, women at intermediate familial risk who were younger than 50 years were enrolled from 76 centres in the UK, and received yearly mammography. Women with BRCA mutations were not explicitly excluded, but would be rare in this group. To compare the FH01 cohort with women not receiving screening, two external comparison groups were used: the control group of the UK Age Trial (106,971 women aged 40-42 years at recruitment, from the general population [ie, average risk], followed up for 10 years), and a Dutch study of women with a family history of breast cancer (cancer cases aged 25-77 years, diagnosed 1980-2004). Study endpoints were size, node status, and histological grade of invasive tumours, and estimated mortality calculated from the Nottingham prognostic index (NPI) score, and adjusted for differences in underlying risk between the FH01 cohort and the control group of the UK Age Trial. This study is registered with the National Research Register, number N0484114809.
FINDINGS: 6710 women were enrolled between Jan 16, 2003, and Feb 28, 2007, and received yearly mammography for a mean of 4 years (SD 2) up until Nov 30, 2009; surveillance and reporting of cancers is still underway. 136 women were diagnosed with breast cancer: 105 (77%) at screening, 28 (21%) symptomatically in the interval between screening events, and three (2%) symptomatically after failing to attend their latest mammogram. Invasive tumours in the FH01 study were significantly smaller (p=0·0094), less likely to be node positive (p=0·0083), and of more favourable grade (p=0·0072) than were those in the control group of the UK Age Trial, and were significantly less likely to be node positive than were tumours in the Dutch study (p=0·012). Mean NPI score was significantly lower in the FH01 cohort than in the control group of the UK Age Trial (p=0·00079) or the Dutch study (p<0·0001). After adjustment for underlying risk, predicted 10-year mortality was significantly lower in the FH01 cohort (1·10%) than in the control group of the UK Age Trial (1·38%), with relative risk of 0·80 (95% CI 0·66-0·96; p=0·022).
INTERPRETATION: Yearly mammography in women with a medium familial risk of breast cancer is likely to be effective in prevention of deaths from breast cancer.