Vinayak Muralidhar1, Ryan D Nipp, David P Ryan, Theodore S Hong, Paul L Nguyen, Jennifer Y Wo. 1. 1 Harvard-MIT Division of Health Sciences and Technology, Harvard Medical School, Boston, Massachusetts 2 Division of Hematology and Medical Oncology, Massachusetts General Hospital, Boston, Massachusetts 3 Department of Radiation Oncology, Massachusetts General Hospital, Boston, Massachusetts 4 Department of Radiation Oncology, Dana-Farber Cancer Institute and Brigham and Women's Hospital, Boston, Massachusetts.
Abstract
BACKGROUND: Larger tumor size and lymph node involvement are traditionally associated with increased colon cancer-specific mortality. OBJECTIVE: We sought to determine whether patients with very small tumors associated with lymph node involvement are at paradoxically increased risk of colon cancer-specific mortality in comparison with those who have larger tumors and lymph node involvement. DESIGN: This is a retrospective cohort study using the Surveillance, Epidemiology, and End Results database. SETTING: Geographic areas included in one of the 18 Surveillance, Epidemiology, and End Results registries were used. PATIENTS: We identified 99,594 patients with nonmetastatic colon adenocarcinoma treated with surgery between 1988 and 2001. MAIN OUTCOME MEASURES: The primary predictor variables were regional lymph node involvement and primary tumor size by longest dimension, grouped into the following predetermined strata: <5 mm, 5 to 19 mm, 20 to 39 mm, 40 to 59 mm, ≥ 60 mm. We used competing risks regression to determine differences in the risk of colon cancer-specific mortality between strata after controlling for T stage, tumor grade, age, year of diagnosis, race, and number of dissected lymph nodes. RESULTS: Median follow-up among censored patients was 12.9 years. We found a significant interaction between lymph node involvement and tumor size (p < 0.05). Among those with node-negative disease, colon cancer-specific mortality increased monotonically with tumor size. In contrast, among those with node-positive disease, patients with the smallest tumors (<5 mm) were at increased risk of 10-year colon cancer-specific mortality compared with those with tumors sized 5 to 19 mm, 20 to 39 mm, 40 to 59 mm, and ≥60 mm (53.3% vs. 30.1%, 37.5%, 39.2%, and 39.7%; adjusted hazard ratios, 1.63-2.24; p < 0.05 in all cases). LIMITATIONS: The main limitations are the retrospective design and information available in the study database. CONCLUSION: In the setting of lymph node involvement, very small tumor size may predict for increased colon cancer-specific mortality compared with larger tumors. Smaller tumors associated with lymph node involvement may represent more aggressive malignancies with a distinct biology that merits further investigation.
BACKGROUND: Larger tumor size and lymph node involvement are traditionally associated with increased colon cancer-specific mortality. OBJECTIVE: We sought to determine whether patients with very small tumors associated with lymph node involvement are at paradoxically increased risk of colon cancer-specific mortality in comparison with those who have larger tumors and lymph node involvement. DESIGN: This is a retrospective cohort study using the Surveillance, Epidemiology, and End Results database. SETTING: Geographic areas included in one of the 18 Surveillance, Epidemiology, and End Results registries were used. PATIENTS: We identified 99,594 patients with nonmetastatic colon adenocarcinoma treated with surgery between 1988 and 2001. MAIN OUTCOME MEASURES: The primary predictor variables were regional lymph node involvement and primary tumor size by longest dimension, grouped into the following predetermined strata: <5 mm, 5 to 19 mm, 20 to 39 mm, 40 to 59 mm, ≥ 60 mm. We used competing risks regression to determine differences in the risk of colon cancer-specific mortality between strata after controlling for T stage, tumor grade, age, year of diagnosis, race, and number of dissected lymph nodes. RESULTS: Median follow-up among censored patients was 12.9 years. We found a significant interaction between lymph node involvement and tumor size (p < 0.05). Among those with node-negative disease, colon cancer-specific mortality increased monotonically with tumor size. In contrast, among those with node-positive disease, patients with the smallest tumors (<5 mm) were at increased risk of 10-year colon cancer-specific mortality compared with those with tumors sized 5 to 19 mm, 20 to 39 mm, 40 to 59 mm, and ≥60 mm (53.3% vs. 30.1%, 37.5%, 39.2%, and 39.7%; adjusted hazard ratios, 1.63-2.24; p < 0.05 in all cases). LIMITATIONS: The main limitations are the retrospective design and information available in the study database. CONCLUSION: In the setting of lymph node involvement, very small tumor size may predict for increased colon cancer-specific mortality compared with larger tumors. Smaller tumors associated with lymph node involvement may represent more aggressive malignancies with a distinct biology that merits further investigation.