C-C3-02: Ethnic Differences in Interval Diagnosed Breast Cancers in South Carolina

Abstract

Background/Aims: The incidence of breast cancer (BrCA) among African-American (AA) women is lower than European-American (EA) women, yet their mortality rate is twice as high. Numerous hypotheses have been proposed to explain this disparity ranging from the biological to factors related to access to care. The relatively under-studied area of interval-diagnosed BrCAs, tumors which arise between regular mammograms and represent one of the most aggressive types of tumors, among AA women may represent such a biological factor. The goal of this investigation was to describe and compare detection patterns of BrCAs and their related histopathology among AA and EA women in South Carolina. Methods: A longitudinal file on 65,766 women was created with multiple clinic visits representing each mammography or diagnostic service performed and the final findings of these procedures. Women with a malignancy diagnosis were classified as cases provided they had an abnormal screening and were subsequently found to have histopathologically-confirmed malignancies. Interval cancers were defined as those that occurred between regular screenings of the American Cancer Society recommended 1–2 years. After electronic identification of the interval cancers, the actual mammography films from each interval cancer case were combined with a random sampling of negative and positive films and underwent a “blinded” review by a radiologist. Chi-square and t-tests were used to assess for differences between AA and EA women and between screen-detected and interval-detected malignancies. Polytomous regression was also employed to describe and compare patterns of BrCA detection among our cohort of women. Results: Among the study population, AA women were more likely to be younger than EA women. AA women were also more likely to have poorly differentiated tumors and to have interval-detected malignancies. Variables found to be significantly associated with interval cancers included age and marital status. Furthermore, AA women were found to be at a 1.8 times higher risk of interval-detected cancers than EA women. Conclusions: Taken as a whole, these results indicate that the characterization of interval-detected tumors among AA women may prove to be a promising avenue to help explain ethnic disparities in BrCA and offer an opportunity for medical intervention such as increased screening guidelines.