A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours
Predicting grade 1 (G1) and 2 (G2) primary pancreatic neuroendocrine tumour (panNET) is crucial to foresee panNET clinical behaviour. Fifty-one patients with G1-G2 primary panNET demonstrated by pre-surgical [68Ga]Ga-DOTANOC PET/CT and diagnostic conventional imaging were grouped according to the tu...
Main Authors: | , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-05-01
|
Series: | Diagnostics |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-4418/11/5/870 |
_version_ | 1797534491629060096 |
---|---|
author | Alessandro Bevilacqua Diletta Calabrò Silvia Malavasi Claudio Ricci Riccardo Casadei Davide Campana Serena Baiocco Stefano Fanti Valentina Ambrosini |
author_facet | Alessandro Bevilacqua Diletta Calabrò Silvia Malavasi Claudio Ricci Riccardo Casadei Davide Campana Serena Baiocco Stefano Fanti Valentina Ambrosini |
author_sort | Alessandro Bevilacqua |
collection | DOAJ |
description | Predicting grade 1 (G1) and 2 (G2) primary pancreatic neuroendocrine tumour (panNET) is crucial to foresee panNET clinical behaviour. Fifty-one patients with G1-G2 primary panNET demonstrated by pre-surgical [68Ga]Ga-DOTANOC PET/CT and diagnostic conventional imaging were grouped according to the tumour grade assessment method: histology on the whole excised primary lesion (HS) or biopsy (BS). First-order and second-order radiomic features (RFs) were computed from SUV maps for the whole tumour volume on HS. The RFs showing the lowest <i>p</i>-values and the highest area under the curve (AUC) were selected. Three radiomic models were assessed: A (trained on HS, validated on BS), B (trained on BS, validated on HS), and C (using the cross-validation on the whole dataset). The second-order normalized homogeneity and entropy was the most effective RFs couple predicting G2 and G1. The best performance was achieved by model A (test AUC = 0.90, sensitivity = 0.88, specificity = 0.89), followed by model C (median test AUC = 0.87, sensitivity = 0.83, specificity = 0.82). Model B performed worse. Using HS to train a radiomic model leads to the best prediction, although a “hybrid” (HS+BS) population performs better than biopsy-only. The non-invasive prediction of panNET grading may be especially useful in lesions not amenable to biopsy while [68Ga]Ga-DOTANOC heterogeneity might recommend FDG PET/CT. |
first_indexed | 2024-03-10T11:30:17Z |
format | Article |
id | doaj.art-fb51eb4e0bca40159f253ba1333106fe |
institution | Directory Open Access Journal |
issn | 2075-4418 |
language | English |
last_indexed | 2024-03-10T11:30:17Z |
publishDate | 2021-05-01 |
publisher | MDPI AG |
record_format | Article |
series | Diagnostics |
spelling | doaj.art-fb51eb4e0bca40159f253ba1333106fe2023-11-21T19:20:27ZengMDPI AGDiagnostics2075-44182021-05-0111587010.3390/diagnostics11050870A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine TumoursAlessandro Bevilacqua0Diletta Calabrò1Silvia Malavasi2Claudio Ricci3Riccardo Casadei4Davide Campana5Serena Baiocco6Stefano Fanti7Valentina Ambrosini8Advanced Research Center for Electronic Systems (ARCES), University of Bologna, I-40125 Bologna, ItalyDepartment of Nuclear Medicine, DIMES, Alma Mater Studiorum, University of Bologna, I-40126 Bologna, ItalyAdvanced Research Center for Electronic Systems (ARCES), University of Bologna, I-40125 Bologna, ItalyIRCCS Azienda Ospedaliero Universitaria di Bologna, I-40138 Bologna, ItalyIRCCS Azienda Ospedaliero Universitaria di Bologna, I-40138 Bologna, ItalyIRCCS Azienda Ospedaliero Universitaria di Bologna, I-40138 Bologna, ItalyAdvanced Research Center for Electronic Systems (ARCES), University of Bologna, I-40125 Bologna, ItalyDepartment of Nuclear Medicine, DIMES, Alma Mater Studiorum, University of Bologna, I-40126 Bologna, ItalyDepartment of Nuclear Medicine, DIMES, Alma Mater Studiorum, University of Bologna, I-40126 Bologna, ItalyPredicting grade 1 (G1) and 2 (G2) primary pancreatic neuroendocrine tumour (panNET) is crucial to foresee panNET clinical behaviour. Fifty-one patients with G1-G2 primary panNET demonstrated by pre-surgical [68Ga]Ga-DOTANOC PET/CT and diagnostic conventional imaging were grouped according to the tumour grade assessment method: histology on the whole excised primary lesion (HS) or biopsy (BS). First-order and second-order radiomic features (RFs) were computed from SUV maps for the whole tumour volume on HS. The RFs showing the lowest <i>p</i>-values and the highest area under the curve (AUC) were selected. Three radiomic models were assessed: A (trained on HS, validated on BS), B (trained on BS, validated on HS), and C (using the cross-validation on the whole dataset). The second-order normalized homogeneity and entropy was the most effective RFs couple predicting G2 and G1. The best performance was achieved by model A (test AUC = 0.90, sensitivity = 0.88, specificity = 0.89), followed by model C (median test AUC = 0.87, sensitivity = 0.83, specificity = 0.82). Model B performed worse. Using HS to train a radiomic model leads to the best prediction, although a “hybrid” (HS+BS) population performs better than biopsy-only. The non-invasive prediction of panNET grading may be especially useful in lesions not amenable to biopsy while [68Ga]Ga-DOTANOC heterogeneity might recommend FDG PET/CT.https://www.mdpi.com/2075-4418/11/5/870[68Ga]Ga-DOTANOCpancreatic neuroendocrine tumourmachine learningbiomarkerstandardized uptake value |
spellingShingle | Alessandro Bevilacqua Diletta Calabrò Silvia Malavasi Claudio Ricci Riccardo Casadei Davide Campana Serena Baiocco Stefano Fanti Valentina Ambrosini A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours Diagnostics [68Ga]Ga-DOTANOC pancreatic neuroendocrine tumour machine learning biomarker standardized uptake value |
title | A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours |
title_full | A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours |
title_fullStr | A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours |
title_full_unstemmed | A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours |
title_short | A [68Ga]Ga-DOTANOC PET/CT Radiomic Model for Non-Invasive Prediction of Tumour Grade in Pancreatic Neuroendocrine Tumours |
title_sort | 68ga ga dotanoc pet ct radiomic model for non invasive prediction of tumour grade in pancreatic neuroendocrine tumours |
topic | [68Ga]Ga-DOTANOC pancreatic neuroendocrine tumour machine learning biomarker standardized uptake value |
url | https://www.mdpi.com/2075-4418/11/5/870 |
work_keys_str_mv | AT alessandrobevilacqua a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT dilettacalabro a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT silviamalavasi a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT claudioricci a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT riccardocasadei a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT davidecampana a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT serenabaiocco a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT stefanofanti a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT valentinaambrosini a68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT alessandrobevilacqua 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT dilettacalabro 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT silviamalavasi 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT claudioricci 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT riccardocasadei 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT davidecampana 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT serenabaiocco 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT stefanofanti 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours AT valentinaambrosini 68gagadotanocpetctradiomicmodelfornoninvasivepredictionoftumourgradeinpancreaticneuroendocrinetumours |