9635 Background: Outcome disparities for African-American (AA) women with breast cancer, including late stage presentation & poorer survival, may in part be due to biologically aggressive tumors. We hypothesized that survival would be worse for AA women with breast tumors which demonstrate abnormal p53 tumor suppressor gene function. Methods: Tumors from 188 consecutive AAs were evaluated for abnormal nuclear accumulation of p53 protein with IHC, using monoclonal antibodies DO-7 & Pab 1801 on paraffin embedded tissue, at the SHCC MBCCOP, in Chicago. Staining was scored on the UK Receptor Group, 0–8 system, with ≥2 as positive, & slides were read by 2 blinded pathologists. Disease-free survival (DFS), distant-disease free survival (DDFS), & overall survival (OS) were computed by the Kaplan-Meier method & correlated with p53 expression. Univariate analyses were performed with the use of a two-sided log-rank test & multivariate analysis was performed with the use of the Cox proportional-hazards model. Results: A total of 72 cases (39.6%) were positive with D07 IHC, 47 cases (27%) were positive with 1801 IHC, & 42 cases (30.2%) were positive for both D07 & 1801 IHC. The mean follow-up was 48 months & young age, high AJCC stage, high tumor grade & negative ER status, were all significantly correlated with p53 expression & poorer survival. p53 expression with 1801 IHC was significantly correlated with DFS, DDFS, & OS. For DFS, 44 events occurred, Hazard Ratio (HR)=2.63, 95% Confidence Interval (95% CI)=1.43–4.85, p=0.002; for DDFS, 34 events occurred, HR=2.68, 95% CI=1.33–5.41, p=0.006; and for OS, 25 events occurred, HR=2.93, 95% CI=1.31–6.55, p=0.009. Multivariate analysis demonstrated that independent predictors for DFS were p53 expression (HR=2.04, 95% CI=1.08–3.85, p=0.027), stage (HR=3.0, 95% CI=1.57–5.73, p=0.0008) & PR status (HR=0.46, 95% CI=0.23–0.95, p=0.036); for DDFS, p53 expression (HR=2.46, 95% CI=1.21–5.0, p=0.012) & stage (HR=3.26, 95% CI=1.58–6.71, p=0.001); and for OS, p53 expression (HR=2.69, 95% CI=1.19–6.10, p=0.017) & stage (HR=2.75, 95% CI=1.22–6.23, p=0.014). Conclusions: Abnormal nuclear accumulation of p53 protein independently predicts poorer survival & agressive biology in AA women with breast cancer. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Aventis, Avon Foundation, NCI

5259 It has been suggested that biologic factors may contribute to the limited efficacy of mammography and poorer breast cancer outcomes in African American (AA) women. To further investigate, we retrospectively evaluated tumor biology of interval cancers diagnosed in 254 consecutive AA women with breast cancer at the John Stroger Hospital of Cook County, Minority-Based Community Clinical Oncology Program, which serves the medically-indigent in Chicago. A total of 51 AA women were diagnosed with interval cancers; 30 had AJCC stage 0-1 and 21 had stage 2-4 cancers. For women with stage 0-1 cancers, interval cancers were significantly more likely to be hormone receptor (ER) negative (33% vs. 13% for non-interval cancers; Chi-square p

5260 We previously demonstrated that lower rates of diagnosis of early-stage breast cancer in minority-underserved women, is in part due to inadequate screening practice, with a need for increased repeat mammography (AACR 2003). In an ongoing study, we retrospectively evaluated the quality of screening mammography reports in 439 consecutive patients with breast cancer, treated at the John Stroger Hospital of Cook County, Minority-Based Community Clinical Oncology Program, which serves the medically-indigent in Chicago. A total of 233 women (53%) had screening mammograms. Significantly more patients with early-stage breast cancer (AJCC stage 0-1) had at least one prior mammogram (74% versus 43% for 2-4; Chi-square p

880 Background: African-American (AA) women with breast cancer have increased risk of death compared with white women. Biologic differences may contribute to outcome disparities for AA women. Our prior study demonstrates that AA women treated with preoperative (preop) chemotherapy have reduced pathologic complete response (pCR) rates compared to non-AAs. Methods: To further evaluate differences in tumor biology as a cause for this disparity, we examined tumor histology and markers on samples from women who received preoperative therapy, at the MBCCOP, Stroger Hospital of Cook County, in Chicago. Results: There were 21 AA (0% pCR) and 21 non-AA (19% pCR) women included. The number of women with grade 1, 2 or 3 tumors respectively was 2, 2 & 17 for AAs, and 0, 8 & 13 for non-AAs. Hormone receptor (HR) status was negative for 15 AAs and 11 non-AAs. HER2 overexpression was seen in 6 AAs and 7 non-AAs. Extra-capsular lymph node tumor extension was seen in 5 AAs and 7 non-AAs. Although AAs were more likely to have grade 3 tumors which were HR negative, these differences were not significant. Immunohistochemistry on paired samples for AAs from pre and post chemotherapy specimens, demonstrated mild decreases in HR and estrogen receptor beta immunostaining, and no change in epidermal growth factor receptor or p27 immunoreactivity. However, p53 immunoreactivity with both Pab1801 and D07 antibodies was increased in the post chemotherapy specimens. Conclusions: Higher tumor grade and HR negative status seen in AAs in our study, are more commonly associated with favorable response to preop therapy. However, pCR did not occur in these women. Mutant p53 may be associated with an unfavorable response to preop therapy, and the demonstrated increase in p53 overexpression in AAs in this study may be responsible for the observed lack of pCR. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration This research was funded in part by the NCI (MBCCOP Grant CA95867), Aventis Pharmaceuticals Inc. & the Avon Foundation

876 Background: African-American (AA) women with breast cancer have an increased risk of death, compared with white women. Biologic differences may contribute to outcome disparities for AA women. Response to preoperative (preop) chemotherapy in AA women may provide useful information on tumor biology. Methods: We retrospectively reviewed 442 consecutive minority-underserved women treated for breast cancer at the MBCCOP, Stroger Hospital of Cook County, in Chicago. Results: Preop therapy was given to 63 (23%) women and 42 had records for review. Of these, 50% (21) were AA. Preop clinical stage distribution for AAs showed 1 with 2A, 5 with 2B, 12 with 3A and 3 with 3B disease; for non-AAs the distribution was 1 with 2A, 8 with 2B, 4 with 3A and 7 with 3B disease. Although more AAs had 3B disease and more non-AAs had 2B disease, this was not significant. Preop node positive status was 17 AAs vs. 14 non-AAs. AC only was used almost equally for both groups (6 AA vs. 7 non-AA), and the addition of Taxane was similar (15 AA vs. 13 non-AA). Equal numbers from both groups were non-compliant with preop therapy (2 AA & 2 non-AA), or had disease progression (2 AA & 2 non-AA). Pathologic residual invasive disease was found in 91% AAs and 76% non-AAs. Pathologic microscopic invasive disease was found in 10% AAs and 5% non-AAs. Pathologic complete response (pCR) was not found in AAs, but 19% non-AAs had a pCR. Node positive status post surgery was similar for both groups (N1: 4 AA vs. 3 non-AA; N2: 7 AA vs. 8 non-AA). For non-AAs, 3 of 4 pCRs occurred in patients with preop clinical stage 3B, and 1 in stage 2B; in this group, pathologic down staging was seen in 63% of 2B, 50% of 3A, & 72% of 3B disease. For AA women, pathologic down staging was comparable, with 60% of 2B, 67% of 3A, & 100% of 3B disease, but pCR did not occur. Conclusions: These results indicate that AA women with breast cancer treated with preop therapy have reduced pCR rates compared to non-AAs. It is unlikely that these results are due to differences in stage distribution, type of therapy or compliance issues. Further investigation of potential differences in tumor biology is required. Author Disclosure Employment or Leadership Consultant or Advisory Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration This research was funded in part by the NCI (MBCCOP Grant CA95867), Aventis Pharmaceuticals Inc. & the Avon Foundation