Diagnostic Value of Conventional PET Parameters and Radiomic Features Extracted from 18F-FDG-PET/CT for Histologic Subtype Classification and Characterization of Lung Neuroendocrine Neoplasms

Published on Mar 10, 2021in Biomedicines4.717
· DOI :10.3390/BIOMEDICINES9030281
Philippe Thuillier1
Estimated H-index: 1
,
Virginia Liberini2
Estimated H-index: 2
+ 8 AuthorsFilippo Molinari35
Estimated H-index: 35
Sources
Abstract
Aim: To evaluate if conventional Positron emission tomography (PET) parameters and radiomic features (RFs) extracted by 18F-FDG-PET/CT can differentiate among different histological subtypes of lung neuroendocrine neoplasms (Lu-NENs). Methods: Forty-four naive-treatment patients on whom 18F-FDG-PET/CT was performed for histologically confirmed Lu-NEN (n = 46) were retrospectively included. Manual segmentation was performed by two operators allowing for extraction of four conventional PET parameters (SUVmax, SUVmean, metabolic tumor volume (MTV), and total lesion glycolysis (TLG)) and 41 RFs. Lu-NENs were classified into two groups: lung neuroendocrine tumors (Lu-NETs) vs. lung neuroendocrine carcinomas (Lu-NECs). Lu-NETs were classified according to histological subtypes (typical (TC)/atypical carcinoid (AC)), Ki67-level, and TNM staging. The least absolute shrink age and selection operator (LASSO) method was used to select the most predictive RFs for classification and Pearson correlation analysis was performed between conventional PET parameters and selected RFs. Results: PET parameters, in particular, SUVmax (area under the curve (AUC) = 0.91; cut-off = 5.16) were higher in Lu-NECs vs. Lu-NETs (p 0.05) and TC vs. AC classification. Conventional PET parameters were correlated with N+ status in Lu-NETs. Conclusion: In our study, conventional PET parameters were able to distinguish Lu-NECs from Lu-NETs, but not TC from AC. RFs did not provide additional information.
References45
Newest
#1Yong Han (Capital Medical University)H-Index: 1
#2Yuan Ma (Capital Medical University)H-Index: 7
Last. Xiuhua Guo (Capital Medical University)H-Index: 24
view all 12 authors...
To evaluate the capability of PET/CT images for differentiating the histologic subtypes of non-small cell lung cancer (NSCLC) and to identify the optimal model from radiomics-based machine learning/deep learning algorithms. In this study, 867 patients with adenocarcinoma (ADC) and 552 patients with squamous cell carcinoma (SCC) were retrospectively analysed. A stratified random sample of 283 patients (20%) was used as the testing set (173 ADC and 110 SCC); the remaining data were used as the tra...
15 CitationsSource
#1Lamiaa Zidan (Cairo University)H-Index: 1
#2Amir Iravani (University of Melbourne)H-Index: 15
Last. Rodney J. Hicks (University of Melbourne)H-Index: 90
view all 6 authors...
This study aimed to analyse the molecular imaging (MI) phenotype of typical carcinoid (TC) and atypical carcinoid (AC) by 68Ga-DOTATATE (GaTATE) and 18F-FDG (FDG) PET/CT with the emphasis on its potential theranostic implications for peptide receptor radionuclide therapy (PRRT). Retrospective review of patients with biopsy-proven TC or AC undergoing both GaTATE and FDG PET/CT at presentation. Based on correlative CT or MRI, positive lesions on either scan were defined by uptake above liver paren...
4 CitationsSource
#1Manila Rubino (IEO: European Institute of Oncology)H-Index: 9
#2J.Y. Scoazec (Institut Gustave Roussy)H-Index: 5
Last. Nicola Fazio (IEO: European Institute of Oncology)H-Index: 42
view all 15 authors...
Abstract Highly proliferative lung carcinoids (HPLC) have been recently reported but information about this subset remains scarce. Objectives Clinical and pathological data of 630 patients with lung carcinoids (LC) referred to Gustave Roussy (GR) or European Institute of Oncology (IEO) were retrospectively reviewed to select HPLC and analyze their frequency, behavior and compare their outcome to conventional LC with Ki-67 ≤ 20 % and mitotic count (MC)≤10/2 mm2. Materials and methods Selection cr...
5 CitationsSource
#1Tianyu Tang (ZJU: Zhejiang University)H-Index: 5
#2Xiang Li (ZJU: Zhejiang University)H-Index: 24
Last. Tingbo Liang (ZJU: Zhejiang University)H-Index: 24
view all 18 authors...
BACKGROUND: In pancreatic cancer, methods to predict early recurrence (ER) and identify patients at increased risk of relapse are urgently required. PURPOSE: To develop a radiomic nomogram based on MR radiomics to stratify patients preoperatively and potentially improve clinical practice. STUDY TYPE: Retrospective. POPULATION: We enrolled 303 patients from two medical centers. Patients with a disease-free survival
23 CitationsSource
#1Giulia Polverari (UNITO: University of Turin)H-Index: 6
#2Francesco Ceci (UNITO: University of Turin)H-Index: 24
Last. Silvia Novello (UNITO: University of Turin)H-Index: 2
view all 14 authors...
Objectives: (1.1) to evaluate the association between baseline 18F-FDG PET/CT semi-quantitative parameters of the primary lesion with progression free survival (PFS), overall survival (OS) and response to immunotherapy, in advanced non-small cell lung carcinoma (NSCLC) patients eligible for immunotherapy; (1.2) to evaluate the application of radiomics analysis of the primary lesion to identify features predictive of response to immunotherapy; (1.3) to evaluate if tumor burden assessed by 18F-FDG...
23 CitationsSource
#1Alex Zwanenburg (DKFZ: German Cancer Research Center)H-Index: 18
#2Martin Vallières (McGill University)H-Index: 18
Last. Steffen LöckH-Index: 24
view all 63 authors...
The Image Biomarker Standardization Initiative validated consensus-based reference values for 169 radiomics features, thus enabling calibration and verification of radiomics software.
534 CitationsSource
#1Xiaofeng LiH-Index: 80
#2Guotao YinH-Index: 1
Last. Wengui XuH-Index: 8
view all 9 authors...
ABSTRACT Radiomics has become an area of interest for tumor characterization in 18F-Fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) imaging. The aim of the present study was to demonstrate how imaging phenotypes was connected to somatic mutations through an integrated analysis of 115 non-small cell lung cancer (NSCLC) patients with somatic mutation testings and engineered computed PET/CT image analytics. A total of 38 radiomic features quantifying tumor morph...
26 CitationsSource
#1V. Grøndahl (UCPH: University of Copenhagen)H-Index: 1
#2Tina Binderup (UCPH: University of Copenhagen)H-Index: 24
Last. Ulrich Knigge (UCPH: University of Copenhagen)H-Index: 54
view all 8 authors...
Abstract Background Bronchopulmonary neuroendocrine tumours are divided into typical carcinoid (TC), atypical carcinoid (AC), large cell neuroendocrine carcinoma (LCNEC), and small cell lung cancer (SCLC). Aim To thoroughly describe a cohort of 252 patients with TC, AC and LCNEC (SCLC excluded). Material and methods Collection of data from 252 patients referred to and treated at Rigshospitalet 2008-2016. Data was collected from electronic patient files and our prospective NET database. Statistic...
12 CitationsSource
#2David BourhisH-Index: 8
Last. Ronan AbgralH-Index: 20
view all 10 authors...
Aim Characterizing tumor heterogeneity with textural indices extracted from 18F-fluorodeoxyglucose positron emission tomography (FDG PET/CT) is of growing interest in oncology. Several series showed promising results to predict survival in patients with head and neck squamous cell carcinoma (HNSCC), analyzing various tumor segmentation methods and textural indices. This preliminary study aimed at assessing the inter-observer and inter-segmentation method variability of textural indices in HNSCC ...
9 CitationsSource
#1Yuanyuan Jiang (Peking Union Medical College Hospital)H-Index: 4
#2Guozhu Hou (Peking Union Medical College Hospital)H-Index: 4
Last. Wuying Cheng (Peking Union Medical College Hospital)H-Index: 2
view all 3 authors...
AbstractBackground:Pulmonary carcinoids (PC) are histologically classified into typical carcinoid (TC) and atypical carcinoid (AC). The diagnosis of pulmonary carcinoid and possibly the differentiation between TC and AC could make a significant effect on the treatment planning as well as prognosis.[
17 CitationsSource
Cited By1
Newest
#1Annette Erle (University Hospital Bonn)H-Index: 2
#2S Moazemi (University Hospital Bonn)H-Index: 1
Last. Ralph A. Bundschuh (University Hospital Bonn)H-Index: 27
view all 6 authors...
The importance of machine learning (ML) in the clinical environment increases constantly. Differentiation of pathological from physiological tracer-uptake in positron emission tomography/computed tomography (PET/CT) images is considered time-consuming and attention intensive, hence crucial for diagnosis and treatment planning. This study aimed at comparing and validating supervised ML algorithms to classify pathological uptake in prostate cancer (PC) patients based on prostate-specific membrane ...
Source