Background: Minimally invasive surgery is the standard approach in colorectal cancer (CRC), yet the clinical value of robotic-assisted surgery (RAS) compared with laparoscopy remains under debate. This study aimed to compare surgical quality using textbook outcome (TO) and textbook oncological outcome (TOO) in patients undergoing minimally invasive resection for locally advanced CRC. Methods: A retrospective analysis of patients with locally advanced CRC (cT2-4N0-2M0) treated in a high-volume centre was performed. Patients undergoing laparoscopic or robotic surgery were included. Propensity score matching (PSM) was applied to balance baseline characteristics. TO was defined as an optimal perioperative course without complications, conversion, reintervention, prolonged length of stay, or mortality. TOO extended TO by including oncological parameters such as R0 resection and adequate lymph node yield. Results: A total of 123 patients were included (80 laparoscopic, 43 robotic), with 80 patients analyzed after PSM (40 per group). RAS was associated with significantly higher rates of intracorporeal (97% vs. 18.9%) and mechanical anastomoses (96.9% vs. 48.6%). No differences were observed in postoperative complications, reintervention rates, length of stay, or mortality. Although the comprehensive complication index was lower in the robotic group, this did not translate into improved TO or TOO rates. After matching, TO was achieved in 72.5% of RAS and 85.0% of laparoscopic cases (p = 0.1745), while TOO rates were also comparable between groups. No independent predictors of TO or TOO were identified in multivariable analysis. Conclusions: Robotic surgery for locally advanced CRC provides comparable perioperative safety and oncological quality to laparoscopy. The implementation of RAS in a high-volume centre does not compromise short-term outcomes, supporting its safe integration into clinical practice.

Background: Composite immunonutritional biomarkers reflecting systemic inflammation, nutritional status, and immune competence have emerged as promising tools for perioperative risk stratification in colorectal cancer (CRC). However, their prognostic value during the perioperative period following minimally invasive CRC surgery remains unresolved. This study aimed to evaluate the association between perioperative inflammatory biomarkers, including the neutrophil-to-lymphocyte ratio (NLR), hemoglobin–albumin–lymphocyte–platelet (HALP) score, and Host Index (H-index), and short-term postoperative outcomes after minimally invasive colorectal cancer surgery. Methods: This retrospective single-center study included 105 patients who underwent minimally invasive curative-intent surgery for CRC between 2023 and 2025. Biomarkers were assessed preoperatively and during the first four postoperative days. Severe postoperative morbidity was defined as a Comprehensive Complication Index (CCI) ≥ 20.9. Results: Among all parameters, NLR measured on postoperative day 3 demonstrated the highest apparent discriminatory ability for severe complications (AUC = 0.87, 95% CI: 0.74–0.95; sensitivity 91.7%; specificity 73.7%). Preoperative HALP was the most informative preoperative marker (AUC = 0.78, 95% CI: 0.66–0.87; sensitivity 76.5%; specificity 87.7%). Conclusions: Perioperative inflammatory and immunonutritional biomarkers are associated with postoperative morbidity following CRC surgery. Preoperative HALP and postoperative day 3 NLR demonstrated the strongest apparent discrimination for CCI ≥ 20.9; however, these findings require confirmation in larger, independent prospective cohorts.

Introduction: By leveraging dedicated datasets and predictive modeling, machine-learning (ML) algorithms can estimate the probability of both short- and long-term outcomes after surgery. The aim of this study was to evaluate the ability of ML-based models to predict postoperative complications in patients with gastric cancer (GC) undergoing multimodal therapy. In particular, we aimed to develop a free, publicly accessible online calculator based on preoperative variables. Materials and Methods: Patients with histologically confirmed locally advanced (cT2-4N0-3M0) GC who underwent multimodal treatment with curative intent between 2013 and 2023 were included in the study. ML models evaluation pipeline was used with Stratified 5-Fold Cross-Validation. Results: A total of 368 patients were included in the final analytic cohort. Among five algorithm classes under 5-fold cross-validation, Compute Area Under the Receiver Operating Characteristic Curve (ROC AUC) was 0.9719, 0.9652, 0.9796, 0.8339 and 0.7581 for XGBoost, Catboost, Random Forest, SVM and Logistic Regression, respectively. Macro F1 was 0.8714, 0.5094, 0.8820, 0.8714 and 0.4579 for XGBoost, SVM, Random Forest, CatBoost and Logistic Regression, respectively. Overall Accuracy was 0.8897, 0.5980, 0.8885, 0.8750 and 0.5466 for XGBoost, SVM, Random Forest, CatBoost and Logistic Regression models, respectively. Conclusions: In this Central and Eastern European cohort of patients with locally advanced GC, ML models using non-linear decision rules-particularly Random Forest and XGBoost- substantially outperformed conventional linear approaches in predicting the severity of postoperative complications. Prospective external validation is needed to clarify the model’s clinical utility and its potential role in perioperative decision support.