Critical number of lymph node involvement in esophageal and gastric cancer and its impact on long-term survival-A single-center 8-year study.

BACKGROUND

Nodal disease in esophageal and gastric cancer is associated with poor survival.

OBJECTIVES

To determine the critical level of lymph node involvement where survival becomes significantly compromised.

METHODS

Survival analyses using multivariable Cox regression and receiver operator characteristics (ROC) were undertaken to determine what number of positive lymph nodes were most sensitive and specific in predicting survival.

RESULTS

A total of 317 patients underwent esophagectomy (n = 190, 59.9%) and gastrectomy (n = 127, 40.1%) for adenocarcinoma. At multivariable analyses, four nodes positivity (irrespective of T-category) was associated with nearly a fivefold increased risk of mortality when compared to node-negative patients (hazard ratio [HR], 4.9; interquartile range 2.0-11.5; P < .001). A positive ratio of up to 50.0% was not associated with worse survival than having four nodes positive (HR, 4.6; 95% confidence interval, 2.6-8.1; P < .001). ROC analysis demonstrated four lymph nodes positive to have a sensitivity of 80.5%, a specificity of 60.1%, and an accuracy of 77.8 (P < .001).

CONCLUSION

The absolute number of nodes positive for cancer is more important than the proportion of positive nodes in predicting survival in esophageal/gastric cancer. Four positive lymph nodes are associated with a fivefold increase in mortality. Beyond this, increasing numbers of positive lymph nodes make no appreciable difference to survival.