Integrating spatial lymph node patterns and multimodal clinicopathological features to predict post-neoadjuvant recurrence in gastric adenocarcinoma: a machine learning nomogram

Comprehensive Summary

Neoadjuvant therapy (NAT) has been found to significantly improve survival rates and pathology responses for gastric cancer. However, NAT makes it difficult to predict post-therapeutic recurrence risk for traditional TNM. The researchers developed a novel lymph node tumor regression grade (LN-TRG) classification system using potential biomarkers for therapeutic efficacy in metastatic lymph nodes to best tailor treatment plans to specific tumors. 112 gastric patients receiving NAT were analyzes according to LN-TRG, clinical (complete blood counts), and other pathological features (metastatic lymph node location, residual tumor ratio, and lymph node dissection count) to build the predictive model. Fazhi et al. found 12 significant prognostic factors for recurrence-free survival and overall survival, 6 of which were key predictors for recurrence-free survival. These factors were used to develop a highly accurate prediction model that was able to distinguish high and low risk groups, identifying a negative correlation between recurrence risk and lymph node dissection count. The researchers also identified higher residual tumor ratios and hazard ratios in specific anatomical lymph node groups, suprapyloric and infrapyloric, as well as an increasing number of positive lymph node groups. This model may prove significant in the clinical setting through usage as a precise risk stratification tool for post-NAT gastric cancer and proactive dissections occurring.

Outcomes and Implications

This model allows for anatomical specificity and lymph node therapeutic response to be incorporated into diagnostic and risk stratification tools for gastric cancer post-neoadjuvant therapy. With the ability to inform high risk patients of adjuvant therapy intensification, it can minimize excessive treatments for these vulnerable populations. Moreover, the model allows for proactive dissection of high-risk nodal stations for patients. This will allow for conventional staging systems to be altered to address the existing clinical inadequacy through incorporation of these new biomarkers and radiomic features.