Number of positive lymph nodes combined with the logarithmic ratio of positive lymph nodes predicts long-term survival for patients with node-positive parotid gland carcinoma after surgery: a SEER population-based study

Machine learning-based survival prediction nomogram for postoperative parotid mucoepidermoid carcinoma

Parotid mucoepidermoid carcinoma (P-MEC) is a significant histopathological subtype of salivary gland cancer with inherent heterogeneity and complexity. Existing clinical models inadequately offer personalized treatment options for patients. In response, we assessed the efficacy of four machine learning algorithms vis-à-vis traditional analysis in forecasting the overall survival (OS) of P-MEC patients. Using the SEER database, we analyzed data from 882 postoperative P-MEC patients (stages I-IVA). Single-factor Cox regression and four machine learning techniques (random forest, LASSO, XGBoost, best subset regression) were employed for variable selection. The optimal model was derived via stepwise backward regression, Akaike Information Criterion (AIC), and Area Under the Curve (AUC). Bootstrap resampling facilitated internal validation, while prediction accuracy was gauged through C-index, time-dependent ROC curve, and calibration curve. The model's clinical relevance was ascertained using decision curve analysis (DCA). The study found 3-, 5-, and 10-year OS rates of 0.887, 0.841, and 0.753, respectively. XGBoost, BSR, and LASSO stood out in predictive efficacy, identifying seven key prognostic factors including age, pathological grade, T stage, N stage, radiation therapy, chemotherapy, and marital status. A subsequent nomogram revealed a C-index of 0.8499 (3-year), 0.8557 (5-year), and 0.8375 (10-year) and AUC values of 0.8670, 0.8879, and 0.8767, respectively. The model also highlighted the clinical significance of postoperative radiotherapy across varying risk levels. Our prognostic model, grounded in machine learning, surpasses traditional models in prediction and offer superior visualization of variable importance.

Number and ratio of metastatic lymph nodes impacts the prognosis of submandibular gland cancer

This study aimed to assess the impact of the number and ratio of metastatic lymph nodes (LNs) on prognosis in submandibular gland cancer. To this end, patients were selected from the Surveillance, Epidemiology, and End Results database retrospectively. The effect of the number and ratio of metastatic LNs and the American Joint Committee on Cancer (AJCC) N stage on disease-specific survival (DSS) and overall survival (OS) was analyzed. In addition, prognostic models based on LN evaluation methods were developed to predict the OS and DSS. A total of 914 patients were included. Binary recursive partitioning analysis determined the optimal cut-off number of metastatic LNs (0 vs. 1-2. vs. 3+). The presence of 3+ metastatic LNs carried the greatest impact on prognosis, followed by 1-2 positive LNs occurrences. The ratio of metastatic LNs was an independent factor for DSS and OS. The model had a higher likelihood ratio and C-index than those in the Cox model based on the AJCC N stage. Quantitative LN burden and ratio of metastatic LNs provides better survival stratification than the AJCC N stage.

Metastatic lymph node burden impacts overall survival in submandibular gland cancer

Objective

To assess the effect of the number of positive lymph nodes (LNs) on the overall survival (OS) of patients with submandibular gland cancer (SmGC).

Methods

Patients who had undergone neck dissection for SmGC were retrospectively enrolled in this study. The effect of the American Joint Committee on Cancer (AJCC) N stage, the number of positive LNs, LN size, LN ratio, and extranodal extension (ENE) on OS and recurrence-free survival (RFS) was evaluated using Cox analysis. Prognostic models were proposed based on the identified significant variable, and their performance was compared using hazard consistency and discrimination.

Results

In total, 129 patients were included in this study. The number of positive LNs rather than LN ratio, LN size, and ENE was associated with OS. A prognostic model based on the number of positive LNs (0 vs. 1-2 vs. 3+) demonstrated a higher likelihood ratio and Harrell's C index than those according to the 7th/8th edition of the AJCC N stage in predicting OS and RFS.

Conclusions

The effect of LN metastasis on OS and RFS was mainly determined by the number of positive LNs. A validation of this finding is warranted in adenoid cystic carcinomas that were not included in this study.