Dataset Information

Noncontrast Radiomics Approach for Predicting Grades of Nonfunctional Pancreatic Neuroendocrine Tumors.

ABSTRACT:

Background

Endoscopic ultrasound-guided fine-needle aspiration is associated with the accurate determination of tumor grade. However, because it is an invasive procedure there is a need to explore alternative noninvasive procedures.

Purpose

To develop and validate a noncontrast radiomics model for the preoperative prediction of nonfunctional pancreatic neuroendocrine tumor (NF-pNET) grade (G).

Study type

Retrospective, single-center study.

Subjects

Patients with pathologically confirmed PNETs (139) were included.

Field strength/sequence

3T/breath-hold single-shot fast-spin echo T₂ -weighted sequence and unenhanced and dynamic contrast-enhanced T₁ -weighted fat-suppressed sequences.

Assessment

Tumor features on contrast MR images were evaluated by three board-certified abdominal radiologists.

Statistical tests

Multivariable logistic regression analysis was used to develop the clinical model. The least absolute shrinkage and selection operator method and linear discriminative analysis (LDA) were used to select the features and to construct a radiomics model. The performance of the models was assessed using the training cohort (97 patients) and the validation cohort (42 patients), and decision curve analysis (DCA) was applied for clinical use.

Results

The clinical model included 14 imaging features, and the corresponding area under the curve (AUC) was 0.769 (95% confidence interval [CI], 0.675-0.863) in the training cohort and 0.729 (95% CI, 0.568-0.890) in the validation cohort. The LDA included 14 selected radiomics features that showed good discrimination-in the training cohort (AUC, 0.851; 95% CI, 0.758-0.916) and the validation cohort (AUC, 0.736; 95% CI, 0.518-0.874). In the decision curves, if the threshold probability was 0.17-0.84, using the radiomics score to distinguish NF-pNET G1 and G2/3, offered more benefit than did the use of a treat-all-patients or treat-none scheme.

Data conclusion

The developed radiomics model using noncontrast MRI could help differentiate G1 and G2/3 tumors, to make the clinical decision, and screen pNETs grade.

Level of evidence

4 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:1124-1136.

SUBMITTER: Bian Y

PROVIDER: S-EPMC7895298 | biostudies-literature | 2020 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Noncontrast Radiomics Approach for Predicting Grades of Nonfunctional Pancreatic Neuroendocrine Tumors.

Bian Yun Y Zhao Zengrui Z Jiang Hui H Fang Xu X Li Jing J Cao Kai K Ma Chao C Guo Shiwei S Wang Li L Jin Gang G Lu Jianping J Xu Jun J

Journal of magnetic resonance imaging : JMRI 20200428 4

<h4>Background</h4>Endoscopic ultrasound-guided fine-needle aspiration is associated with the accurate determination of tumor grade. However, because it is an invasive procedure there is a need to explore alternative noninvasive procedures.<h4>Purpose</h4>To develop and validate a noncontrast radiomics model for the preoperative prediction of nonfunctional pancreatic neuroendocrine tumor (NF-pNET) grade (G).<h4>Study type</h4>Retrospective, single-center study.<h4>Subjects</h4>Patients with path ...[more]

PMID: 32343872

Similar Datasets

Project description:ImportanceThe number of patients with small nonfunctional pancreatic neuroendocrine tumors (NF-PanNETs) is increasing. However, the role of surgery for small NF-PanNETs remains unclear.ObjectiveTo evaluate the association between surgical resection for NF-PanNETs measuring 2 cm or smaller and survival.Design, setting, and participantsThis cohort study used data from the National Cancer Database and included patients with NF-pancreatic neuroendocrine neoplasms who were diagnosed between January 1, 2004, and December 31, 2017. Patients with small NF-PanNETs were divided into 2 groups: group 1a (tumor size, ≤1 cm) and group 1b (tumor size, 1.1-2.0 cm). Patients without information on tumor size, overall survival, and surgical resection were excluded. Data analysis was performed in June 2022.ExposuresPatients with vs without surgical resection.Main outcomes and measuresThe primary outcome was overall survival of patients in group 1a or group 1b who underwent surgical resection compared with those who did not, which was evaluated using Kaplan-Meier estimates and multivariable Cox proportional hazards regression models. Interactions between preoperative factors and surgical resection were analyzed with a multivariable Cox proportional hazards regression model.ResultsOf the 10 504 patients with localized NF-PanNETs identified, 4641 were analyzed. These patients had a mean (SD) age of 60.5 (12.7) years and included 2338 males (50.4%). The median (IQR) follow-up time was 47.1 (28.2-71.6) months. In total, 1278 patients were in group 1a and 3363 patients were in group 1b. The surgical resection rates were 82.0% in group 1a and 87.0% in group 1b. After adjustment for preoperative factors, surgical resection was associated with longer survival for patients in group 1b (hazard ratio [HR], 0.58; 95% CI, 0.42-0.80; P < .001) but not for patients in group 1a (HR, 0.68; 95% CI, 0.41-1.11; P = .12). In group 1b, interaction analysis found that age of 64 years or younger, absence of comorbidities, treatment at academic institutions, and distal pancreatic tumors were factors associated with increased survival after surgical resection.Conclusions and relevanceFindings of this study support an association between surgical resection and increased survival in select patients with NF-PanNETs measuring 1.1 to 2.0 cm who were younger than 65 years, had no comorbidities, received treatment at academic institutions, and had tumors of the distal pancreas. Future investigations of surgical resection for small NF-PanNETs that include the Ki-67 index are warranted to validate these findings.

Project description:ObjectivesTo explore the potential of radiomics features to predict the histologic grade of nonfunctioning pancreatic neuroendocrine tumor (NF-PNET) patients using non-contrast sequence based on MRI.MethodsTwo hundred twenty-eight patients with NF-PNETs undergoing MRI at 5 centers were retrospectively analyzed. Data from center 1 (n = 115) constituted the training cohort, and data from centers 2-5 (n = 113) constituted the testing cohort. Radiomics features were extracted from T2-weighted images and the apparent diffusion coefficient. The least absolute shrinkage and selection operator was applied to select the most important features and to develop radiomics signatures. The area under receiver operating characteristic curve (AUC) was performed to assess models.ResultsTumor boundary, enhancement homogeneity, and vascular invasion were used to construct the radiological model to stratify NF-PNET patients into grade 1 and 2/3 groups, which yielded AUC of 0.884 and 0.684 in the training and testing groups. A radiomics model including 4 features was constructed, with an AUC of 0.941 and 0.871 in the training and testing cohorts. The fusion model combining the radiomics signature and radiological characteristics showed good performance in the training set (AUC = 0.956) and in the testing set (AUC = 0.864), respectively.ConclusionThe developed model that integrates radiomics features with radiological characteristics could be used as a non-invasive, dependable, and accurate tool for the preoperative prediction of grade in NF-PNETs.Clinical relevance statementOur study revealed that the fusion model based on a non-contrast MR sequence can be used to predict the histologic grade before operation. The radiomics model may be a new and effective biological marker in NF-PNETs.Key pointsThe diagnostic performance of the radiomics model and fusion model was better than that of the model based on clinical information and radiological features in predicting grade 1 and 2/3 of nonfunctioning pancreatic neuroendocrine tumors (NF-PNETs). Good performance of the model in the four external testing cohorts indicated that the radiomics model and fusion model for predicting the grades of NF-PNETs were robust and reliable, indicating the two models could be used in the clinical setting and facilitate the surgeons' decision on risk stratification. The radiomics features were selected from non-contrast T2-weighted images (T2WI) and diffusion-weighted imaging (DWI) sequence, which means that the administration of contrast agent was not needed in grading the NF-PNETs.

Project description:BackgroundThe prognosis of patients with pancreatic neuroendocrine tumors (PanNET), the second most common type of pancreatic cancer, varies significantly, and up to 15% of patients develop metastasis. Although certain morphological characteristics of PanNETs have been associated with patient outcome, there are no available morphology-based prognostic markers. Given that current clinical histopathology markers are unable to identify high-risk PanNET patients, the development of accurate prognostic biomarkers is needed. Here, we describe a novel machine learning, multiclassification pipeline to predict the risk of metastasis using morphological information from whole tissue slides.MethodsDigital images from surgically resected tissues from 89 PanNET patients were used. Pathologist-annotated regions were extracted to train a convolutional neural network (CNN) to identify tiles consisting of PanNET, stroma, normal pancreas parenchyma, and fat. Computationally annotated cancer or stroma tiles and patient metastasis status were used to train CNN to calculate a region based metastatic risk score. Aggregation of the metastatic probability scores across the slide was performed to predict the risk of metastasis.ResultsThe ability of CNN to discriminate different tissues was high (per-tile accuracy >95%; whole slide cancer regions Jaccard index = 79%). Cancer and stromal tiles with high evaluated probability provided F1 scores of 0.82 and 0.69, respectively, when we compared tissues from patients who developed metastasis and those who did not. The final model identified low-risk (n = 76) and high-risk (n = 13) patients, as well as predicted metastasis-free survival (hazard ratio: 4.71) after adjusting for common clinicopathological variables, especially in grade I/II patients.ConclusionUsing slides from surgically resected PanNETs, our novel, multiclassification, deep learning pipeline was able to predict the risk of metastasis in PanNET patients. Our results suggest the presence of prognostic morphological patterns in PanNET tissues, and that these patterns may help guide clinical decision making.

Project description:IntroductionThe role of parenchyma-sparing resections (PSR) and lymph node dissection in small (<3 cm) nonfunctional pancreatic neuroendocrine tumors (PNET) is unlikely to be studied in a prospective randomized clinical trial. By combining data from 4 high-volume pancreatic centers we compared postoperative and long-term outcomes of patients who underwent PSR with patients who underwent oncologic resections.MethodsRetrospective review of prospectively collected clinicopathologic data of patients who underwent pancreatectomy between 2000 and 2021 was collected from 4 high-volume institutions. PSR and lymph node-sparing resections (enucleation and central pancreatectomy) were compared to those who underwent oncologic resections with lymphadenectomy (pancreaticoduodenectomy, distal pancreatectomy). Statistical testing was performed using χ 2 test and t test, survival estimates with Kaplan-Meier method and multivariate analysis using Cox proportional hazard model.ResultsOf 810 patients with small sporadic nonfunctional PNETs, 121 (14.9%) had enucleations, 100 (12.3%) had central pancreatectomies, and 589 (72.7%) patients underwent oncologic resections. The median age was 59 years and 48.2% were female with a median tumor size of 2.5 cm. After case-control matching for tumor size, 221 patients were selected in each group. Patients with PSR were more likely to undergo minimally invasive operations (32.6% vs 13.6%, P <0.001), had less intraoperative blood loss (358 vs 511 ml, P <0.001) and had shorter operative times (180 vs 330 minutes, P <0.001) than patients undergoing oncologic resections. While the mean number of lymph nodes harvested was lower for PSR (n=1.4 vs n=9.9, P <0.001), the mean number of positive lymph nodes was equivalent to oncologic resections (n=1.1 vs n=0.9, P =0.808). Although the rate of all postoperative complications was similar for PSR and oncologic resections (38.5% vs 48.2%, P =0.090), it was higher for central pancreatectomies (38.5% vs 56.6%, P =0.003). Long-term median disease-free survival (190.5 vs 195.2 months, P =0.506) and overall survival (197.9 vs 192.6 months, P =0.372) were comparable. Of the 810 patients 136 (16.7%) had no lymph nodes resected. These patients experienced less blood loss, shorter operations ( P <0.001), and lower postoperative complication rates as compared to patients who had lymphadenectomies (39.7% vs 56.9%, P =0.008). Median disease-free survival (197.1 vs 191.9 months, P =0.837) and overall survival (200 vs 195.1 months, P =0.827) were similar for patients with no lymph nodes resected and patients with negative lymph nodes (N0) after lymphadenectomy.ConclusionIn small <3 cm nonfunctional PNETs, PSRs and lymph node-sparing resections are associated with lower blood loss, shorter operative times, and lower complication rates when compared to oncologic resections, and have similar long-term oncologic outcomes.

Project description:BackgroundEverolimus, an oral inhibitor of mammalian target of rapamycin (mTOR), has shown antitumor activity in patients with advanced pancreatic neuroendocrine tumors, in two phase 2 studies. We evaluated the agent in a prospective, randomized, phase 3 study.MethodsWe randomly assigned 410 patients who had advanced, low-grade or intermediate-grade pancreatic neuroendocrine tumors with radiologic progression within the previous 12 months to receive everolimus, at a dose of 10 mg once daily (207 patients), or placebo (203 patients), both in conjunction with best supportive care. The primary end point was progression-free survival in an intention-to-treat analysis. In the case of patients in whom radiologic progression occurred during the study, the treatment assignments could be revealed, and patients who had been randomly assigned to placebo were offered open-label everolimus.ResultsThe median progression-free survival was 11.0 months with everolimus as compared with 4.6 months with placebo (hazard ratio for disease progression or death from any cause with everolimus, 0.35; 95% confidence interval [CI], 0.27 to 0.45; P<0.001), representing a 65% reduction in the estimated risk of progression or death. Estimates of the proportion of patients who were alive and progression-free at 18 months were 34% (95% CI, 26 to 43) with everolimus as compared with 9% (95% CI, 4 to 16) with placebo. Drug-related adverse events were mostly grade 1 or 2 and included stomatitis (in 64% of patients in the everolimus group vs. 17% in the placebo group), rash (49% vs. 10%), diarrhea (34% vs. 10%), fatigue (31% vs. 14%), and infections (23% vs. 6%), which were primarily upper respiratory. Grade 3 or 4 events that were more frequent with everolimus than with placebo included anemia (6% vs. 0%) and hyperglycemia (5% vs. 2%). The median exposure to everolimus was longer than exposure to placebo by a factor of 2.3 (38 weeks vs. 16 weeks).ConclusionsEverolimus, as compared with placebo, significantly prolonged progression-free survival among patients with progressive advanced pancreatic neuroendocrine tumors and was associated with a low rate of severe adverse events. (Funded by Novartis Oncology; RADIANT-3 ClinicalTrials.gov number, NCT00510068.).

Dataset Information

Noncontrast Radiomics Approach for Predicting Grades of Nonfunctional Pancreatic Neuroendocrine Tumors.

Background

Purpose

Study type

Subjects

Field strength/sequence

Assessment

Statistical tests

Results

Data conclusion

Level of evidence

Publications

Noncontrast Radiomics Approach for Predicting Grades of Nonfunctional Pancreatic Neuroendocrine Tumors.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets