BACKGROUND: Lymph node (LN) status is a major determinant of prognosis and treatment of gastric adenocarcinoma. The 1997 American Joint Commission on Cancer/Union Internationale Contre le Cancer guidelines were revised, requiring examination of ≥15 LN for staging.METHODS: We investigated compliance with these guidelines and the correlation with overall survival (OS) by analyzing 10,807 resected gastric cancers in the Surveillance, Epidemiology and End Results (SEER) database, 1988–2002. Kaplan–Meier survival curves were constructed; survival was compared by using Cox proportional hazards.RESULTS: Overall, 29% of cases had ≥15 LN examined. After 1997, the median number of LN assessed increased from 9 to 10 (P < .0001). Factors predictive of adequate LN assessment (ALNA) were higher stage, worse grade, age <74 years, later year of diagnosis, nonwhite race, more extensive surgery, female sex, and SEER region. Differences in the rate of ALNA between regions were noted, ranging from 19.7–53% (P < .0001). Of T1N0 patients, 19% had ALNA. Improved OS was predicted by earlier stage, lower grade, marital status, Asian race, younger age, T‐stage, female sex, SEER region, and ALNA. Median OS was highest in the region with the best ALNA rate and worst in the region with the lowest (33 mos vs. 17 mos, P < .0001). Inadequate LN assessment led to poorer survival at every stage (P < .001).CONCLUSION: The overwhelming majority of patients have an inadequate LN assessment. ALNA was associated with improved OS, with significant variation across regions. Understaging due to inadequate LN assessment may affect eligibility for adjuvant therapy. Education is required to improve LN retrieval. Cancer 2006. © 2006 American Cancer Society.

Few studies have examined the relationship of insurance status with the presentation and treatment of breast cancer. Using a state cancer registry, we compared tumor presentation and surgical treatments at presentation by insurance status (private insurance, Medicare, Medicaid, or uninsured). Student's t-test, Chi-square test, and ANOVA were used for comparison. P-values reflect a comparison to insured patients. From 1996 to 2005, there were 6876 cases of invasive breast cancer with either private (n = 3975), Medicare (n = 2592), Medicaid (n = 193), or no insurance (n = 116). The median age (years) at presentation was 55 for private, 76 for Medicare, 54 for Medicaid and 54 for uninsured. The mean and median tumor size (mm) were 18.5 and 15 for private; 20.9 and 15 for Medicare; 24.2 and 18 for Medicaid; and 29.5 and 17 for uninsured, respectively; (p < 0.001 for all). Fewer women with Medicare and Medicaid presented with node negative breast cancers: private, 73.4% node negative; Medicare, 79.5% (p < 0.001); Medicaid, 60.9% (p < 0.001); and uninsured, 58% (p = 0.005). Significantly more uninsured women had no surgical treatment of their breast cancer: 15.5% versus 4.3% for private (p < 0.001). Among women with non-metastatic T1/T2 tumors, 71.5% with private insurance underwent breast-conserving surgery (BCS), compared with 64.2% of Medicare (p < 0.001), 65% of Medicaid (p = 0.097), and 65.4% of uninsured (p = 0.234). The rate of reconstruction following mastectomy was higher for private insurance (36.6%), compared with Medicare (3.8%, p < 0.0001), Medicaid (26.1%, p = 0.31), and uninsured (5.0%, p = 0.0038). The presentation of breast cancer in women with no insurance and Medicaid is significantly worse than those with private insurance. Of concern are the lower proportions of BCS and reconstruction among patients who are uninsured or have Medicaid. Reduction of disparities in breast cancer presentation and treatment may be possible by increasing enrollment of uninsured, program-eligible women in a state-supported screening and treatment program.