What determines complications and prognosis among patients subject to multivisceral resections for locally advanced gastric cancer?

BACKGROUND: Locally advanced gastric cancer (GC) extending to the surrounding tissues may require a multivisceral resection (MVR) to provide the best chance of cure. However, little is known about how the extent of organ resection affects the risks and benefits of surgery. METHODS: An electronic database of patients treated between 1996 and 2020 in an academic surgical centre was reviewed. MVRs were defined as partial or total gastrectomy combined with splenectomy, distal pancreatectomy, or partial colectomy. RESULTS: Suspected intraoperative tumour invasion of perigastric organs (cT4b) was found in 298 of 1476 patients with non-metastatic GC, and 218 were subject to MVRs, including the spleen (n = 126), pancreas (n = 51), and colon (n = 41). MVRs were associated with higher proportions of surgical and general complications, but not mortality. A nomogram was developed to predict the risk of major postoperative morbidity (Clavien-Dindo's grade ≥ 3a), and the highest odds ratio for major morbidity identified by logistic regression modelling was found for distal pancreatectomy (2.53, 95% CI 1.23-5.19, P = 0.012) and colectomy (2.29, 95% CI 1.04-5.09, P = 0.035). Margin-positive resections were identified by the Cox proportional hazards model as the most important risk factor for patients' survival (hazard ratio 1.47, 95% CI 1.10-1.97). The extent of organ resection did not affect prognosis, but a MVR was the only factor reducing the risk of margin positivity (OR 0.44, 95% CI 0.21-0.87). CONCLUSIONS: The risk of multivisceral resections is associated with the organ being removed, but only MVRs increase the odds of complete tumour clearance for locally advanced gastric cancer.

Factors predicting adequate lymph node yield in patients undergoing pancreatoduodenectomy for malignancy.

BACKGROUND: Most pancreatoduodenectomy resections do not meet the minimum of 12 lymph nodes recommended by the American Joint Committee on Cancer for accurate staging of periampullary malignancies. The purpose of this study was to investigate factors affecting the likelihood of adequate nodal yield in pancreatoduodenectomy specimens subject to routine pathological assessment. METHODS: Six hundred sixty-two patients subject to pancreatoduodenectomy between 1990 and 2013 for pancreatic, ampullary, and common bile duct cancers were reviewed. Predictors of yielding at least 12 lymph nodes were evaluated with a logistic regression model, and a survival analysis was carried out to verify the prognostic implications of nodal counts. RESULTS: The median number of evaluated nodes was 17 (interquartile range 11 to 25), and less than 12 lymph nodes were reported in surgical specimens of 179 (27 %) patients. Tumor diameter ≥20 mm (odds ratio [OR] 2.547, 95 % confidence interval [CI] 1.225 to 5.329, P = 0.013), lymph node metastases (OR 2.642, 95 % CI 1.378 to 5.061, P = 0.004), and radical lymphadenectomy (OR 5.566, 95 % CI 2.041 to 15.148, P = 0.01) were significant predictors of retrieving 12 or more lymph nodes. Lymph node counts did not influence the overall prognosis of the patients. However, a subgroup analysis carried out for individual cancer sites demonstrated that removing at least 12 lymph nodes is associated with better prognosis for pancreatic cancer. CONCLUSIONS: Few variables affect adequate nodal yield in pancreatoduodenectomy specimens subject to routine pathological assessment. Considering the ambiguities related to the only modifiable factor identified, appropriate pathology training should be considered to increase nodal yield rather than more aggressive lymphatic dissection.

Nomogram-Based Prognostic Evaluation of Gastric Cancer Patients with Low Counts of Examined Lymph Nodes Outperforms the Predictive Ability of the 7th and 8th Editions of the American Joint Committee on Cancer Staging System.

BACKGROUND: The American Joint Committee on Cancer (AJCC) staging system has limited accuracy in predicting survival of gastric cancer patients with inadequate counts of evaluated lymph nodes (LNs). We therefore aimed to develop a prognostic nomogram suitable for clinical applications in such cases. METHODS: A total of 1511 noncardia gastric cancer patients treated between 1990 and 2010 in the academic surgical center were reviewed to compare the 7th and 8th editions of the AJCC staging system. A nomogram was developed for the prediction of 5-year survival in patients with less than 16 LNs evaluated (n = 546). External validation was performed using datasets derived from the Polish Gastric Cancer Study Group (n = 668) and the SEER database (n = 11,225). RESULTS: The 8th edition of AJCC staging showed better overall discriminatory power compared to the previous version, but no improvement was found for patients with < 16 evaluated LNs. The developed nomogram had better concordance index (0.695) than the former (0.682) or latest (0.680) staging editions, including patients subject to neoadjuvant treatment, and calibration curves showed excellent agreement between the nomogram-predicted and actual survival. High discriminatory power was also demonstrated for both validation cohorts. Subsequently, the nomogram showed the best accuracy for the prediction of 5-year survival through the time-dependent ROC curve analysis in the training and validation cohorts. CONCLUSIONS: A clinically relevant nomogram was built for the prediction of 5-year survival in patients with inadequate numbers of LNs evaluated in surgical specimens. The predictive accuracy of the nomogram was validated in two Western populations.