DESCRIPTION (provided by applicant): This study, submitted in response to PAR-06-294, proposes to investigate whether breast cancer incidence decreased, following dramatic declines in hormone therapy (HT) use after publication of the Women's Health Initiative (WHI) results in mid-2002. The WHI found, contrary to expectations, that HT did not reduce, and may in fact have elevated, risk of cardiovascular disease, while also confirming that long-term use of the combined estrogen plus progestin HT was associated with increased risk of breast and ovarian cancer. Suggestive evidence, albeit based on only 2 studies, indicates US breast cancer incidence declined by 7-11% between mid-2002 and 2003, especially among older women with estrogen-receptor positive (ER+) tumors. If cessation of HT does reduce risk of breast cancer, it follows that the greatest post-2002 US declines in breast cancer incidence should be most evident among (a) women most likely to use HT (i.e., white affluent women with access to health care), and (b) types of breast cancer most strongly linked to HT use (estrogen receptor positive (ER+), lobular histologic type, and age at diagnosis >= 50 years old). To test these hypotheses, we seek to add breast cancer data for 2003-2005 to an existing geocoded data set we previously generated for our study on "Socioeconomic Trends in Breast Cancer Incidence." This study includes data on all incident cases of primary invasive breast from two US cancer registries: (1) the original 5- county catchment area of Northern California Cancer Center (NCCC, 1973-2002), and (2) the Massachusetts Cancer Registry (MCR) (1988-2002). Focusing on years 1995-2005 (n ~ 53,335 cases), we will geocode the new 2003-2005 data to census tract (CT) level, link them to 2000 CT-level area-based socioeconomic data, and test 3 related sets of hypotheses: (1) the first significant decline in breast cancer incidence occurred after mid-2002; (2) since mid-2002, breast cancer incidence rates have dropped most among white non-Hispanic women living in more affluent CTs, compared to (a) white non-Hispanic women living in less affluent CTs, and (b) women of color living in both more and less affluent CTs; and (3) post-2002 declines in breast cancer incidence were greatest among: (a) ER+ compared to ER- tumors; (b) lobular and ductal-lobular tumors compared to all other histologic types; and (c) women age 50 and older compared to women under age 50. Results are likely to enhance etiologic understanding of the epidemiology of breast cancer and ways its incidence can be reduced. To date, only 2 studies have examined whether US breast cancer incidence decreased following the dramatic declines in hormone therapy (HT) use after publication of the Women's Health Initiative results in mid-2002, which found, contrary to expectation, that HT use did not decrease, and may in fact have elevated, risk of cardiovascular disease, while also confirming HT use was associated with increased risk of both breast and ovarian cancer. We propose to build on these 2 studies by using data from two US population-based cancer registries for the years 1995-2005 to examine whether breast cancer incidence has declined since mid-2002, both overall and especially among: (1) women most likely to use HT (i.e., white affluent women with access to health care), and (2) types of breast cancer most strongly linked to HT use (estrogen receptor positive (ER+), lobular histologic type, and age at diagnosis >= 50 years old). The knowledge gained will lead to new insights about the causes of breast cancer and ways its incidence can be reduced.