Abstract
BACKGROUND: The association between small-bowel adenocarcinoma (SBA) tumor size and prognosis is unclear, and we used the Surveillance, Epidemiology, and End Results (SEER) database to assess the prognostic value of SBA tumor size. METHODS: Patients with postoperative SBA were selected from the SEER database, and overall survival (OS) and cancer-specific survival (CSS) were used as outcome variables. Tumor size was used as a categorical and continuous variable, respectively, to adjust for confounders and analyze the association between SBA tumor size and prognosis using Cox proportional hazard regression, and the results were visualized using restricted cubic splines (RCS). Spearman correlation coefficient was used to evaluate the statistical correlation between tumor size and tumor invasion depth (T-stage). Kaplan-Meier survival curves were used to estimate OS at different T stages. RESULTS: When the tumor size was analyzed as a quantitative variable, the adjusted covariate model showed that the HR was 1.008 (P = 0.04) for OS and 1.021 (P = 0.03) for CSS. And regardless of OS or CSS, when the tumor size < 3-4 cm, there was a close linear relationship between tumor size and HR. What's more, in the SEER database, the 5-year survival rates of T1, T2, T3 and T4 patients were 81.8 %, 81.1 %, 66.0 % and 50.9 % (P < 0.001) according to AJCC T-stage. However, in the modified T-stage (mT), these rates were 82.8 %, 70.6 %, 60.7 % and 39.8 % (P < 0.001). When patients within each of the AJCC T stages were stratified by mT stages, significant survival heterogeneity was observed within each of the AJCC T1 to T4 stages(P < 0.001). CONCLUSION: When tumor size is used in a quantitative way, tumor size is an independent predictor of poor outcome in patients with SBA. Furthermore, we established a modified T-stage based on tumor size and depth of invasion.