Spectrum of mammographically detected breast cancers

Mammographic screening of women at both ends of the age spectrum presents a number of challenges. The purpose of this study was to characterize experi ence with mammographic detection of breast cancer. The two goals were 1) to establish the cancer detection rate of screening mammography and 2) to com pare the tumor size of cancers found by mammography, physical examination, or both modalities. From January 1994 through June 1997, data on 609 consec utive female primary breast cancer patients were collected concurrent with definitive surgical therapy. The method of detection was determined by the surgeon, after reviewing mammogram and physical examination. Screening ultr asound was not used. For the 184 patients under 50 years of age, 53 (29%) c ancers were detected by mammography only and 48 (26%) by physical examinati on only. Women under 50 years of age had fewer cancers detected by mammogra phy only (P < 0.001) and more cancers detected by physical examination only (P = 0.0014) than those over 50. With increasing age, the proportion of wo men with ductal carcinoma in situ decreased (P = 0.004), and the proportion with T-1c or T-2 tumors increased (P = 0.006). We conclude that I) when ex amining women under 50 years of age, the surgeon must be clearly focused on the double-edged sword of screening mammography in this age group, and 2) community cancer programs should encourage annual screening of women over 4 0 years of age but focus on those over 70, without an arbitrary upper age l imit.