Dataset Information

Preoperative prediction of lymph node metastasis in nonfunctioning pancreatic neuroendocrine tumors from clinical and MRI features: a multicenter study.

ABSTRACT:

Background

The extent of surgery in nonfunctioning pancreatic neuroendocrine tumors (NF-PNETs) has not well established, partly owing to the dilemma of precise prediction of lymph node metastasis (LNM) preoperatively. This study proposed to develop and validate the value of MRI features for predicting LNM in NF-PNETs.

Methods

A total of 187 patients with NF-PNETs who underwent MR scan and subsequent lymphadenectomy from 4 hospitals were included and divided into training group (n = 66, 1 center) and validation group (n = 121, 3 centers). The clinical characteristics and qualitative MRI features were collected. Multivariate logistic regression model for predicting LNM in NF-PNETs was constructed using the training group and further tested using validation group.

Results

Nodal metastases were reported in 41 patients (21.9%). Multivariate analysis showed that regular shape of primary tumor (odds ratio [OR], 4.722; p = .038) and the short axis of the largest lymph node in the regional area (OR, 1.488; p = .002) were independent predictors for LNM in the training group. The area under the receiver operating characteristic curve in the training group and validation group were 0.890 and 0.849, respectively. Disease-free survival was significantly different between model-defined LNM and non-LNM group.

Conclusions

The novel MRI-based model considering regular shape of primary tumor and short axis of largest lymph node in the regional area can accurately predict lymph node metastases preoperatively in NF-PNETs patients, which might facilitate the surgeons' decision on risk stratification.

SUBMITTER: Zhu HB

PROVIDER: S-EPMC9547759 | biostudies-literature | 2022 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Preoperative prediction of lymph node metastasis in nonfunctioning pancreatic neuroendocrine tumors from clinical and MRI features: a multicenter study.

Zhu Hai-Bin HB Nie Pei P Jiang Liu L Hu Juan J Zhang Xiao-Yan XY Li Xiao-Ting XT Lu Ming M Sun Ying-Shi YS

Insights into imaging 20221008 1

<h4>Background</h4>The extent of surgery in nonfunctioning pancreatic neuroendocrine tumors (NF-PNETs) has not well established, partly owing to the dilemma of precise prediction of lymph node metastasis (LNM) preoperatively. This study proposed to develop and validate the value of MRI features for predicting LNM in NF-PNETs.<h4>Methods</h4>A total of 187 patients with NF-PNETs who underwent MR scan and subsequent lymphadenectomy from 4 hospitals were included and divided into training group (n ...[more]

PMID: 36209332

Similar Datasets

Project description:Lymph node involvement increases the risk of breast cancer recurrence. An accurate non-invasive assessment of nodal involvement is valuable in cancer staging, surgical risk, and cost savings. Radiomics has been proposed to pre-operatively predict sentinel lymph node (SLN) status; however, radiomic models are known to be sensitive to acquisition parameters. The purpose of this study was to develop a prediction model for preoperative prediction of SLN metastasis using deep learning-based (DLB) features and compare its predictive performance to state-of-the-art radiomics. Specifically, this study aimed to compare the generalizability of radiomics vs DLB features in an independent test set with dissimilar resolution. Dynamic contrast-enhancement images from 198 patients (67 positive SLNs) were used in this study. Of these subjects, 163 had an in-plane resolution of 0.7 × 0.7 mm2, which were randomly divided into a training set (approximately 67%) and a validation set (approximately 33%). The remaining 35 subjects with a different in-plane resolution (0.78 × 0.78 mm2) were treated as independent testing set for generalizability. Two methods were employed: (1) conventional radiomics (CR), and (2) DLB features which replaced hand-curated features with pre-trained VGG-16 features. The threshold determined using the training set was applied to the independent validation and testing dataset. Same feature reduction, feature selection, model creation procedures were used for both approaches. In the validation set (same resolution as training), the DLB model outperformed the CR model (accuracy 83% vs 80%). Furthermore, in the independent testing set of the dissimilar resolution, the DLB model performed markedly better than the CR model (accuracy 77% vs 71%). The predictive performance of the DLB model outperformed the CR model for this task. More interestingly, these improvements were seen particularly in the independent testing set of dissimilar resolution. This could indicate that DLB features can ultimately result in a more generalizable model.

Project description:BackgroundRegional lymph node metastasis (LNM) is crucial for planning additional lymphadenectomy, and is directly correlated with poor prognosis in gastroenteropancreatic neuroendocrine tumors (GEP-NETs). However, the patterns of LNM for small (≤20 mm) GEP-NETs remain unclear. This population-based study aimed at evaluating LNM patterns and identifying optimal surgical strategies from the standpoint of lymph node dissemination.MethodsThis retrospective cohort study retrieved data from the Surveillance, Epidemiology, and End Results (SEER) 18 registries database for 17,308 patients diagnosed as having localized well-differentiated GEP-NETs ≤ 20 mm between January 1, 2004, and December 31, 2017. The patterns of LNM were characterized in 6,622 patients who underwent extended resection for adequate lymph node harvest.ResultsOf 6,622 patients with localized small GEP-NETs in the current study, 2,380 (36%) presented with LNM after regional lymphadenectomy. Nodal involvement was observed in approximately 7.4%, 49.1%, 13.6%, 53.7%, 13.8%, 7.8%, and 15.4% of gastric (g-), small intestinal (si-), appendiceal (a-), colonic (c-), rectal (r-), non-functional pancreatic (nfp-), and functional pancreatic (fp-) NETs ≤ 20 mm. Patients with younger age, larger tumor size, and muscularis invasion were more likely to present with LNM. Additional lymphadenectomy conferred a significant survival advantage in NETs (≤10 mm: HR, 0.47; 95% CI, 0.33-0.66; p < 0.001; 11-20 mm: HR, 0.54; 95% CI, 0.34-0.85; p = 0.008) and fp-NETs ≤ 20 mm (HR, 0.08; 95% CI, 0.02-0.36; p = 0.001), as well as g-NETs (HR, 0.39; 95% CI, 0.16-0.96; p = 0.041) and c-NETs of 11-20 mm (HR, 0.07; 95% CI, 0.01-0.48; p = 0.007). Survival benefits of additional lymphadenectomy were not found in a-NETs, r-NETs, and nfp-NETs with a small size.ConclusionsGiven the increased risk for nodal metastasis, primary tumor resection with regional lymphadenectomy is a potential optimal surgical strategy for si-NETs and fp-NETs ≤ 20 mm, as well as g-NETs and c-NETs of 11-20 mm. Local resection is an appropriate and reliable surgical approach for a-NETs, r-NETs, and nfp-NETs ≤ 20 mm.

Dataset Information

Preoperative prediction of lymph node metastasis in nonfunctioning pancreatic neuroendocrine tumors from clinical and MRI features: a multicenter study.

Background

Methods

Results

Conclusions

Publications

Preoperative prediction of lymph node metastasis in nonfunctioning pancreatic neuroendocrine tumors from clinical and MRI features: a multicenter study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets