Nomogram for predicting lymph node metastasis rate of submucosal gastric cancer by analyzing clinicopathological
Zhixue Zheng1, Yinan Zhang1, Lianhai Zhang1
1Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education/Beijing), 1 Department of Gastrointestinal Surgery, 2 Department of Pathology, Peking University Cancer Hospital & Institute, Beijing 100142, China.
Background:
To combine clinicopathological characteristics associated with lymph node metastasis for submucosal gastric cancer into a nomogram.
Methods:
We retrospectively analyzed 262 patients with submucosal gastric cancer who underwent D2 gastrectomy between 1996 and 2012. The relationship between lymph node metastasis and clinicopathological features was statistically analyzed. With multivariate logistic regression analysis, we made a nomogram to predict the possibility of lymph node metastasis. Receiver operating characteristic (ROC) analysis was also performed to assess the predictive value of the model. Discrimination and calibration were performed using internal validation.
Results:
A total number of 48 (18.3%) patients with submucosal gastric cancer have pathologically lymph node metastasis. For submucosal gastric carcinoma, lymph node metastasis was associated with age, tumor location, macroscopic type, size, differentiation, histology, the existence of ulcer and lymphovascular invasion in univariate analysis (all P<0.05). The multivariate logistic regression analysis identified that age ≤50 years old, macroscopic type III or mixed, undifferentiated type, and presence of lymphovascular invasion were independent risk factors of lymph node metastasis in submucosal gastric cancer (all P<0.05). We constructed a predicting nomogram with all these factors for lymph node metastasis in submucosal gastric cancer with good discrimination [area under the curve (AUC) =0.844]. Internal validation demonstrated a good discrimination power that the actual probability corresponds closely with the predicted probability.
Conclusions:
We developed a nomogram to predict the rate of lymph node metastasis for submucosal gastric cancer. With good discrimination and internal validation, the nomogram improved individualized predictions for assisting clinicians to make appropriated treatment decision for submucosal gastric cancer patients.
