Lymph node micrometastasis of poorly differentiated node-negative gastric cancer risks a worse-than-expected survival outcome under standard management algorithm

BackgroundInvestigation of lymph node micrometastasis (mN) of gastric cancer has been focused on either T1 disease or T1-4N0 disease. Yet, it is unclear whether standard management algorithm toward poorly differentiated gastric cancer (PDGC) is more vulnerable to existence of mN, given its inherently biological aggressiveness, as compared with other histological types.Patients and methodsA surgical series (n = 3456) of gastric cancer categorized by histological differentiation was enrolled to analyze survival stratification. Of them, a cohort of T1-T4 N0 PDGC (n = 100) were subjected to cytokeratin immunohistochemistry, a surrogate of mN.ResultsCancer-specific survival by AJCC8 staging system could be nicely differentiated in both well-/moderately differentiated and signet ring cell types, while those between stage IA versus IB (p = 0.105), and stage IB versus IIA (p = 0.141) in PDGC could not. Thirteen (13%) out of 100 node-negative PDGC cases exhibited mN, with 5, 2, 5 and 1 cases occurring in T1, T2, T3, and T4 stage, respectively, without identifiable contributing factors. Prognostic performance of AJCC8 working upon PDGC became more discriminative by incorporating mN, as hazard ratio of stage IIIC referenced to stage IA increased from 43 to 78.ConclusionDefective discriminative survival of PDGC by standard staging algorithm prompted us to survey mN occurring in T1-T4N0 PDGC. The prognostic performance of AJCC8 working upon PDGC was enhanced by incorporating mN. As so, we recommend documentation of mN exclusively on node-negative PDGC that helps unveil stage migration phenomenon and switch to appropriate adjuvant therapy in need.     

Intervention rates and outcomes in medically managed uncomplicated descending thoracic aortic dissections

ObjectiveTo evaluate the long-term incidence and outcome of aortic interventions for medically managed uncomplicated thoracic aortic dissections.MethodsBetween January 2012 and December 2018, 91 patients were discharged home with an uncomplicated, medically treated aortic dissection (involving the descending aorta with or without aortic arch involvement, no ascending involvement). After a median period of 4 (first quartile: 2, third quartile: 11) months, 30 patients (33%) required an aortic intervention. Patient characteristics, radiographic, treatment, and follow-up data were compared for patients with and without aortic interventions. A competing risk regression model was analyzed to identify independent predictors of aortic intervention and to predict the risk for intervention.ResultsPatients who underwent aortic interventions had significantly larger thoracic (P = .041) and abdominal (P = .015) aortic diameters, the dissection was significantly longer (P = .035), there were more communications between both lumina (P = .040), and the first communication was significantly closer to the left subclavian artery (P = .049). A descending thoracic aortic diameter exceeding 45 mm was predictive for an aortic intervention (P = .001; subdistribution hazard ratio: 3.51). The risk for aortic intervention was 27% ± 10% and 36% ± 11% after 1 and 3 years, respectively. Fourteen patients (47%) underwent thoracic endovascular aortic repair, 11 patients (37%) thoracic endovascular aortic repair and left carotid to subclavian bypass, 3 patients (10%) total arch replacement with the frozen elephant trunk technique, and 2 patients (7%) thoracoabdominal aortic replacement. We observed no in-hospital mortality.ConclusionsThe need for secondary aortic interventions in patients with initially medically managed, uncomplicated descending aortic dissections is substantial. The full spectrum of aortic treatment options (endovascular, hybrid, conventional open surgical) is required in these patients.     

Robotic-assisted central pancreatectomy: A minimally invasive approach for benign and low-grade lesions

BackgroundCentrally located pancreatic lesions are often treated with extended pancreaticoduodenectomy or distal pancreatectomy resulting in loss of healthy parenchyma and a high risk of diabetes and exocrine insufficiency. Robotic central pancreatectomy (RCP) is a parenchyma sparring alternative that has been shown safe and feasible [[1], [2]].MethodsIn this article, we describe our operative technique and the perioperative outcomes of a series of RCP for low-grade or benign pancreatic tumors.ResultsSix patients (5 female and 1 man) with a median age of 51.5 (44–68) years underwent a RCP for 2 serous cystadenomas, 2 mucinous cystic tumors, 1 neuroendocrine tumor, and 1 autoimmune pancreatitis. There were no conversions, intraoperative complications, or perioperative transfusions. Median operative time and was 240 (230–291) minutes and median blood loss was 100 (100–400) ml. The median hospital stay was 8 (5–27) days. There were no mortalities, reoperations, or readmissions. One patient developed a grade B pancreatic fistula which was successfully managed conservatively. All resections had free margins and the median tumor size was 2.5 (1.5–3.5) cm. After a mean follow-up of 46 months, no patients presented new-onset diabetes or exocrine insufficiency.ConclusionsRCP represents the least invasive option for both the patient and the pancreatic parenchyma. With a standardized technique, RCP results in low postoperative morbidity and excellent long-term pancreatic function. Although our results are excellent, POPF still represents the main complication of central pancreatectomy with an incidence ranging from 0 to 80% depending on multiple factors such as the surgeon, technique, and pancreatic texture.     

Development and validation of artificial intelligence models for preoperative prediction of inferior mesenteric artery lymph nodes metastasis in left colon and rectal cancer

BackgroundDissection of lymph nodes at the roots of the inferior mesenteric artery (IMAN) should be offered only to selected patients at a major risk of developing IMAN involvement. The aim of this study is to present the first artificial intelligence (AI) models to predict IMAN metastasis risk in the left colon and rectal cancer patients.MethodsA total of 2891 patients with descending colon including splenic flexure, sigmoid colon and rectal cancer undergoing major primary surgery and IMAN dissection were included as a study cohort, which was then split into a training set (67%) and a testing set (33%). Feature selection was conducted using the least absolute shrinkage and selection operator (LASSO) regression model. Seven AI algorithms, namely Support Vector Machine (SVM), Logistic Regression (LR), Extreme Gradient Boosting (XGB), Light Gradient Boosting (LGB), Decision Tree Classifier (DTC), Random Forest (RF) classifier, and Multilayer Perceptron (MLP), as well as traditional multivariate LR model were employed to construct predictive models. The optimal hyperparameters were determined with 5 fold cross-validation. The predictive performance of models and the expert surgeon was assessed and compared in the testing set independently.ResultsThe IMAN involvement incidence was 4.6%. The optimal set of features selected by LASSO included 10 characteristics: neoadjuvant treatment, age, synchronous liver metastasis, synchronous lung metastasis, signet ring adenocarcinoma, neural invasion, lymphovascular invasion, CA199, endoscopic obstruction, T stage evaluated by MRI. The most accurate model derived from MLP showed excellent prediction power with area under the receiver operating characteristic curve (AUROC) of 0.873 and produced 81.0% recognition sensitivity and 82.5% specificity in the testing set independently. In contrast, the judgment of IMAN metastasis by expert surgeon yield rather imprecise and unreliable results with a significantly lower AUROC of 0.509. Additionally, the proposed MLP had the highest net benefits and the largest reduction of unnecessary IMAN dissection without the cost of additional involved IMAN missed.ConclusionMLP model was able to maintain its prediction accuracy in the testing set better than other models and expert surgeons. Our MLP model could be used to help identify IMA nodal metastasis and to select candidates for individual IMAN dissection.     

Oncologic benefit of adjuvant chemotherapy for locally advanced rectal cancer after neoadjuvant chemoradiotherapy and curative surgery with selective lateral pelvic lymph node dissection: An international retrospective cohort study

IntroductionIntensive local treatment comprising total mesorectal excision (TME) with selective lateral pelvic lymph node dissection (LPND) after neoadjuvant chemoradiotherapy (CRT) for locally advanced rectal cancer (LARC) has received attention among clinicians treating rectal cancer. It remains unclear whether adjuvant chemotherapy (ACT) after intensive local treatment is beneficial for these patients. We evaluated the oncologic benefit of ACT for patients with LARC who received intensive local treatment.Materials and methodsThis international multicentre retrospective cohort study included 737 patients treated in Japan and Korea between 2010 and 2017. The effectiveness of ACT on recurrence-free survival (RFS) was evaluated using univariable and multivariable Cox proportional hazards models, with subgroup analyses to identify subpopulations potentially benefiting from ACT.ResultsThe median follow-up was 49 months; the 5-year RFS and local recurrence rates for the entire cohort were 72.1% and 4.9%, respectively; 514 patients (69.7%) received adjuvant chemotherapy, without an oncologic benefit (hazard ratio, 1.14; 95% confidence interval [CI]: 0.79–1.68) demonstrated in the multivariable Cox regression analysis. In subgroup analyses, the distributions of the 95% CI in patients aged ≥70 years and those with ypStage 0 tended to place a disproportionate emphasis that favoured the non-ACT treatment strategy.ConclusionDespite achieving good local control with intensive local treatment strategy, the effectiveness of ACT for the LARC patients with CRT followed by TME with selective LPND was not proved. Elderly patients and those with ypStage0 may not receive benefit from ACT after CRT and TME ± LPND.     

Biopsia selectiva de ganglio centinela en cáncer de mama, ¿está justificada la espera al resultado intraoperatorio?

IntroductionIn our institution, the study of selective sentinel node biopsy (SLNB) is performed intraoperatively. The main objective of our study is to know the proportion of patients who benefits from the waiting of the results of SLNB.MethodsA retrospective analysis of patients operated on our center between January 1 st, 2018 and June 30, 2019 was carried out. We included women diagnosed with T1–T2 tumors, treated by lumpectomy and SLNB studied using OSNA method.ResultsOur study included 149 women. There were not statistically significant differences in terms of demographic data between the group treated with axillary lymph node dissection (ALND) and exclusively SLNB group. After analysis of SLN intraoperatively, there were performed 18 axillary lymphadenectomies. Only in six of these 18 cases, three or more sentinel nodes were founded. The location of the tumor, the presence of lymphovascular permeation and the total tumor load (TTL) showed statistically significant differences between groups. Only the TTL was established as the independent factor of the need for ALND.ConclusionsObtaining a deferred result of the SLNB allowed reducing the time of anesthesia and occupation of the operating room, since in a high percentage of cases an additional procedure is not performed.     

基于TCGA数据库挖掘甲状腺乳头状癌淋巴结转移相关的突变基因及其机制探讨北大核心

目的:挖掘并筛选与甲状腺乳头状癌(papillary thyroid carcinoma,PTC)淋巴结转移相关的突变基因及其潜在的机制。方法:从TCGA数据库获得377例PTC患者的测序数据及完整的临床资料,并分为淋巴结转移组(LM,n=212)与无淋巴结转移组(NLM,n=165)。利用R语言(v3.6.2)对转移组特有的突变基因进行富集分析。采用String在线软件绘制蛋白互作网络,Cytoscape软件筛选网络中的核心基因。在UALCAN网站上验证基因表达量与淋巴结转移之间的关系。利用荧光实时定量PCR测定候选基因在细胞系中mRNA的表达量。结果:一共筛选出1197个仅在LM组发生突变的基因,它们主要富集在黏着连接、细胞黏附分子(cell adhesion molecules,CAMs)通路。CAMs通路的核心基因ITGB1与VCAN的表达量与淋巴结转移相关。与正常甲状腺细胞系相比,VCAN基因在PTC细胞系TPC-1和B-CPAP中mRNA的表达量较高,尤其是B-CPAP细胞系。结论:CAMs通路可能是PTC淋巴结转移相关机制之一,其中VCAN基因可能是潜在的分子标志物。