A prognostic nomogram utilized lymph node ratio for signet ring cell gastric cancer patients post-surgery.

BACKGROUND

Signet ring cell (SRC) gastric cancer is known for its aggressive behavior and poor prognosis. To date, no nomogram has been specifically developed for SRC gastric cancer patients post-surgery. Our objective was to create a nomogram to personalize the prediction of both overall survival (OS) and cancer-specific survival (CSS).

METHODS

We analyzed data from 3,481 patients with histologically confirmed SRC gastric cancer, diagnosed between 2004 and 2021, using information from the Surveillance Epidemiology, and End Results (SEER) database. Patients diagnosed between 2004 and 2015 were randomly divided into two groups: one for training and the other for validation. Additionally, patients diagnosed between 2016 and 2021 were selected as the second validation set. Univariate and multivariate Cox regression models were employed to identify key predictors, which were then used to construct a nomogram. Only the variables significantly linked to OS were incorporated into the final model. The nomogram's accuracy and performance were tested using several evaluation tools, including the concordance index (C-index), calibration plots, and receiver operating characteristic (ROC) curves.

RESULTS

Univariate and multivariate analyses identified race, chemotherapy, T and M stages, age, tumor size, primary tumor location, and lymph node ratio (LNR) as independent prognostic factors for OS and CSS. These key variables were used to construct the nomogram. The model's predictive accuracy was reflected by a C-index of 0.748 [95% confidence interval (CI): 0.734-0.763] for OS and 0.763 (95% CI: 0.761-0.764) for CSS. In the first validation cohort, the C-index for OS was 0.746 (95% CI: 0.702-0.791), and for CSS, it was 0.751 (95% CI: 0.706-0.796). In the second validation cohort, the C-index for OS was 0.784 (95% CI: 0.733-0.836), while for CSS, it was 0.818 (95% CI: 0.767-0.870). For CSS in the training set, the AUC values were 0.760, 0.821, and 0.833 at 1, 3, and 5 years, respectively. In the first validation set, the area under the curve (AUC) values were 0.738, 0.800, and 0.824 for the same time points. In the second validation set, the AUC values were 0.808, 0.872, and 0.885 at 1, 3, and 5 years, respectively. For OS predictions, the AUC in the training set was 0.737, 0.812, and 0.825 at 1, 3, and 5 years. In the first validation set, the AUC values were 0.727, 0.789, and 0.813, while in the second validation set, the AUC values were 0.792, 0.851, and 0.839.

CONCLUSIONS

We have successfully developed effective nomograms to evaluate the prognosis of patients with SRC gastric cancer, focusing on both OS and cumulative survival CSS. These nomograms integrate key clinical factors and provide valuable tools for personalized patient prognosis, enhancing clinical decision-making and potentially improving treatment outcomes.