Oral azacitidine compared with standard therapy in patients with relapsed or refractory follicular helper T-cell lymphoma (ORACLE): an open-label randomised, phase 3 study.

BACKGROUND: Follicular helper T-cell lymphomas (TFHL) harbour frequent alterations in genes that regulate DNA methylation. Preliminary reports suggest that treatment with 5-azacitidine has clinical activity in patients with relapsed or refractory TFHL. We aimed to compare the oral form of azacitidine with investigator's choice standard therapy (ICT; ie, gemcitabine, bendamustine, or romidepsin) in patients with relapsed or refractory TFHL. METHODS: Patients older than 18 years with relapsed or refractory TFHL (angioimmunoblastic T-cell lymphoma, follicular lymphoma, or nodal T-cell lymphoma with phenotype, ie, positive with two or more markers among CD10, BCL6, CXCL13, PD1, or ICOS) based on the 2017 WHO classification of haematological neoplasms, with an Eastern Cooperative Oncology Group performance status score of 0-3, were recruited in university hospitals from five European countries and from Japan. Patients were randomly assigned 1:1 to treatment with either azacitidine given at a dose of 300 mg once a day (200 mg in Japanese patients) for 14 days in a 28-day cycle or gemcitabine, bendamustine, or romidepsin according to the investigator's choice. Random assignment was stratified by the number of previous lines of therapy and by the presence of previous or concomitant myeloid malignancy. The primary endpoint was investigator-assessed progression-free survival, presented in the intention-to-treat population. This Article is the final analysis of this trial, registered at ClinicalTrials.gov (Europe NCT03593018 and Japan NCT03703375). FINDINGS: 86 patients (median age 69 years [IQR 62-76], 50 patients were male, 36 were female) were enrolled between Nov 9, 2018, to Feb 22, 2021; 42 in the azacitidine group and 44 in the ICT group. With a median follow-up of 27·4 months (IQR 20·2-32·9), the median progression-free survival was 5·6 months (95% CI 2·7 -8·1) in the azacitidine group versus 2·8 months (1·9-4·8) in the ICT group (hazard ratio of 0·63 (95% CI 0·38-1·07); 1-sided p=0·042). Grade 3-4 adverse events were reported in 32 (76%) of 42 patients in the azacitidine group versus 42 (98%) of 43 patients in the ICT group. The most adverse grade 3 or worse adverse events were haematological (28 [67%] of 42 patients vs 40 [93%] of 43 patients), infection (8 [19%] and 14 [33%]), and gastrointestinal (5 [12%] vs 1 [2%] for azacitidine and ICT, respectively). There were two treatment-related deaths in the azacitidine group (one endocarditis and one candidiasis) and three in the ICT group (one heart failure, one COVID-19, and one cause unknown). INTERPRETATION: Although the pre-specified primary outcome of the trial was not met, the favourable safety profile suggests that azacitidine could add to the treatment options in these difficult to treat diseases especially in combination with other drugs. Trials with combination are in preparation in a platform trial. FUNDING: Bristol-Myers Squibb. TRANSLATION: For the French translation of the abstract see Supplementary Materials section.

Whole-Body Positron Emission Tomography with 18F-Fluorodeoxyglucose/Magnetic Resonance Imaging as a Screening Tool for the Detection of Malignant Transformation in Individuals with Neurofibromatosis Type 1.

Malignant peripheral nerve sheath tumors (MPNSTs) are the leading cause of death in patients with neurofibromatosis type 1. They can result from premalignant neurofibromas, including neurofibromas with atypia and atypical neurofibromatous neoplasms of uncertain biologic potential. Some phenotypic characteristics have been described as associated with their development. The aim of this study was to outline our use of whole-body positron emission tomography with 18F-fluorodeoxyglucose/magnetic resonance imaging in adults with neurofibromatosis type 1, especially in the screening of asymptomatic individuals with a higher risk of developing an MPNST, and to study its impact on neurofibroma classification (malignant vs premalignant) and MPNST staging over time. Individuals with neurofibromatosis type 1 who underwent a positron emission tomography with 18F-fluorodeoxyglucose/magnetic resonance imaging between 2017 and 2021 were included, analyzing separately the screened population. Maximum standard uptake value and diffusion-weighted imaging were assessed. Biopsy/surgery confirmed the diagnosis. In all, 345 positron emission tomography with 18F-fluorodeoxyglucose/magnetic resonance imaging were performed in 241 patients, including 149 asymptomatic (62%) but at-risk patients. Eight MPNSTs in 8 screened individuals (5%), 6 neurofibromas with atypia in 4 individuals (3%), and 29 atypical neurofibromatous neoplasms of uncertain biologic potential in 23 individuals (15%) were diagnosed. Over time, the proportion of grade 3 MPNST and the malignant/premalignant ratio in screened individuals significantly decreased (P = .03 and P < .001, respectively). This study emphasizes the diagnostic and screening performances of whole-body positron emission tomography with 18F-fluorodeoxyglucose/magnetic resonance imaging in adults with neurofibromatosis type 1.

Association of LI-RADS and Histopathologic Features with Survival in Patients with Solitary Resected Hepatocellular Carcinoma.

Background Both Liver Imaging Reporting and Data System (LI-RADS) and histopathologic features provide prognostic information in patients with hepatocellular carcinoma (HCC), but whether LI-RADS is independently associated with survival is uncertain. Purpose To assess the association of LI-RADS categories and features with survival outcomes in patients with solitary resected HCC. Materials and Methods This retrospective study included patients with solitary resected HCC from three institutions examined with preoperative contrast-enhanced CT and/or MRI between January 2008 and December 2019. Three independent readers evaluated the LI-RADS version 2018 categories and features. Histopathologic features including World Health Organization tumor grade, microvascular and macrovascular invasion, satellite nodules, and tumor capsule were recorded. Overall survival and disease-free survival were assessed with Cox regression models. Marginal effects of nontargetoid features on survival were estimated using propensity score matching. Results A total of 360 patients (median age, 64 years [IQR, 56-70 years]; 280 male patients) were included. At CT and MRI, the LI-RADS LR-M category was associated with increased risk of recurrence (CT: hazard ratio [HR] = 1.83 [95% CI: 1.26, 2.66], P = .001; MRI: HR = 2.22 [95% CI: 1.56, 3.16], P < .001) and death (CT: HR = 2.47 [95% CI: 1.72, 3.55], P < .001; MRI: HR = 1.80 [95% CI: 1.32, 2.46], P < .001) independently of histopathologic features. The presence of at least one nontargetoid feature was associated with an increased risk of recurrence (CT: HR = 1.80 [95% CI: 1.36, 2.38], P < .001; MRI: HR = 1.93 [95% CI: 1.81, 2.06], P < .001) and death (CT: HR = 1.51 [95% CI: 1.10, 2.07], P < .010) independently of histopathologic features. In matched samples, recurrence was associated with the presence of at least one nontargetoid feature at CT (HR = 2.06 [95% CI: 1.15, 3.66]; P = .02) or MRI (HR = 1.79 [95% CI: 1.01, 3.20]; P = .048). Conclusion In patients with solitary resected HCC, LR-M category and nontargetoid features were negatively associated with survival independently of histopathologic characteristics. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Kartalis and Grigoriadis in this issue.

Detection and severity quantification of pulmonary embolism with 3D CT data using an automated deep learning-based artificial solution.

PURPOSE: The purpose of this study was to propose a deep learning-based approach to detect pulmonary embolism and quantify its severity using the Qanadli score and the right-to-left ventricle diameter (RV/LV) ratio on three-dimensional (3D) computed tomography pulmonary angiography (CTPA) examinations with limited annotations. MATERIALS AND METHODS: Using a database of 3D CTPA examinations of 1268 patients with image-level annotations, and two other public datasets of CTPA examinations from 91 (CAD-PE) and 35 (FUME-PE) patients with pixel-level annotations, a pipeline consisting of: (i), detecting blood clots; (ii), performing PE-positive versus negative classification; (iii), estimating the Qanadli score; and (iv), predicting RV/LV diameter ratio was followed. The method was evaluated on a test set including 378 patients. The performance of PE classification and severity quantification was quantitatively assessed using an area under the curve (AUC) analysis for PE classification and a coefficient of determination (R²) for the Qanadli score and the RV/LV diameter ratio. RESULTS: Quantitative evaluation led to an overall AUC of 0.870 (95% confidence interval [CI]: 0.850-0.900) for PE classification task on the training set and an AUC of 0.852 (95% CI: 0.810-0.890) on the test set. Regression analysis yielded R² value of 0.717 (95% CI: 0.668-0.760) and of 0.723 (95% CI: 0.668-0.766) for the Qanadli score and the RV/LV diameter ratio estimation, respectively on the test set. CONCLUSION: This study shows the feasibility of utilizing AI-based assistance tools in detecting blood clots and estimating PE severity scores with 3D CTPA examinations. This is achieved by leveraging blood clots and cardiac segmentations. Further studies are needed to assess the effectiveness of these tools in clinical practice.

Applicability of Bosniak 2019 for renal mass classification on portal venous phase at the era of spectral CT imaging using rapid kV-switching dual-energy CT.

OBJECTIVES: To evaluate the applicability of Bosniak 2019 criteria on a monophasic portal venous phase using rapid kilovoltage-switching DECT (rsDECT). MATERIALS AND METHODS: One hundred twenty-seven renal masses assessed on rsDECT were included, classified according to Bosniak 2019 classification using MRI as the reference standard. Using the portal venous phase, virtual monochromatic images at 40, 50, and 77 keV; virtual unenhanced (VUE) images; and iodine map images were reconstructed. Changes in attenuation values between VUE and 40 keV, 50 keV, and 77 keV measurements were computed and respectively defined as ∆HU40keV, ∆HU50keV, and ∆HU77keV. The values of ∆HU40keV, ∆HU50keV, and ∆HU77keV thresholds providing the optimal diagnostic performance for the detection of internal enhancement were determined using Youden index. RESULTS: Population study included 25 solid renal masses (25/127, 20%) and 102 cystic renal masses (102/127, 80%). To differentiate solid to cystic masses, the specificity of the predefined 20 HU threshold reached 88% (95%CI: 82, 93) using ∆HU77keV and 21% (95%CI: 15, 28) using ∆HU40keV. The estimated optimal threshold of attenuation change was 19 HU on ∆HU77keV, 69 HU on ∆HU50eV, and 111 HU on ∆HU40eV. The rsDECT classification was highly similar to that of MRI for solid renal masses (23/25, 92%) and for Bosniak 1 masses (62/66, 94%). However, 2 hyperattenuating Bosniak 2 renal masses (2/26, 8%) were classified as solid renal masses on rsDECT. CONCLUSION: DECT is a promising tool for Bosniak classification particularly to differentiate solid from Bosniak I-II cyst. However, known enhancement thresholds must be adapted especially to the energy level of virtual monochromatic reconstructions. CLINICAL STATEMENT: DECT is a promising tool for Bosniak classification; however, known enhancement thresholds must be adapted according to the types of reconstructions used and especially to the energy level of virtual monochromatic reconstructions. KEY POINTS: • To differentiate solid to cystic renal masses, predefined 20 HU threshold had a poor specificity using 40 keV virtual monochromatic images. • Most of Bosniak 1 masses according to MRI were also classified as Bosniak 1 on rapid kV-switching dual-energy CT (rsDECT). • Bosniak 2 hyperattenuating renal cysts mimicked solid lesion on rsDECT.

Glomerular Filtration Rate Measured Based on Iomeprol Clearance Assessed at CT Urography in Living Kidney Donor Candidates.

Background Estimating glomerular filtration rate (GFR) from serum creatinine can be inaccurate, and current procedures for measuring GFR are time-consuming and cumbersome. Purpose To develop a method for measuring GFR based on iomeprol clearance assessed at CT urography in kidney donor candidates and compare this with iohexol clearance (reference standard for measuring GFR). Materials and Methods This cross-sectional retrospective study included data from kidney donor candidates who underwent both iohexol clearance and CT urography between July 2016 and October 2022. CT-measured GFR was calculated as the iomeprol excretion rate in the urinary system between arterial and excretory phases (Hounsfield units times milliliters per minute) divided by a surrogate for serum iomeprol concentration in the aorta at the midpoint (in Hounsfield units). Performance of CT-measured GFR was assessed with use of mean bias (mean difference between CT-measured GFR and iohexol clearance), precision (the distance between quartile 1 and quartile 3 of the bias [quartile 3 minus quartile 1], with a small value indicating high precision), and accuracy (percentage of CT-measured GFR values falling within 10%, 20%, and 30% of iohexol clearance values). Intraobserver agreement was assessed for 30 randomly selected individuals with the Lin concordance correlation coefficient. Results A total of 75 kidney donor candidates were included (mean age, 51 years ± 13 [SD]; 45 female). The CT-measured GFR was unbiased (1.1 mL/min/1.73 m2 [95% CI: -1.9, 4.1]) and highly precise (16.2 mL/min/1.73 m2 [quartiles 1 to 3, -6.6 to 9.6]). The accuracy of CT-measured GFR within 10%, 20%, and 30% was 61.3% (95% CI: 50.3, 72.4), 88.0% (95% CI: 80.7, 95.4), and 100%, respectively. Concordance between CT-based GFR measurements taken 2 months apart was almost perfect (correlation coefficient, 0.99 [95% CI: 0.98, 0.99]). Conclusion In living kidney donors, GFR measured based on iomeprol clearance assessed at CT urography showed good agreement with GFR measured based on iohexol clearance. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Davenport in this issue.

Chest Computed Tomography Characteristics of Critically Ill COVID-19 Patients with Auto-antibodies Against Type I Interferons.

PURPOSE: Patients with auto-antibodies neutralizing type I interferons (anti-IFN auto-Abs) are at risk of severe forms of coronavirus disease 19 (COVID-19). The chest computed tomography (CT) scan characteristics of critically ill COVID-19 patients harboring these auto-Abs have never been reported. METHODS: Bicentric ancillary study of the ANTICOV study (observational prospective cohort of severe COVID-19 patients admitted to the intensive care unit (ICU) for hypoxemic acute respiratory failure between March 2020 and May 2021) on chest CT scan characteristics (severity score, parenchymal, pleural, vascular patterns). Anti-IFN auto-Abs were detected using a luciferase neutralization reporting assay. Imaging data were collected through independent blinded reading of two thoracic radiologists of chest CT studies performed at ICU admission (± 72 h). The primary outcome measure was the evaluation of severity by the total severity score (TSS) and the CT severity score (CTSS) according to the presence or absence of anti-IFN auto-Abs. RESULTS: Two hundred thirty-one critically ill COVID-19 patients were included in the study (mean age 59.5 ± 12.7 years; males 74.6%). Day 90 mortality was 29.5% (n = 72/244). There was a trend towards more severe radiological lesions in patients with anti-IFN auto-Abs than in others, not reaching statistical significance (median CTSS 27.5 (21.0-34.8) versus 24.0 (19.0-30.0), p = 0.052; median TSS 14.5 (10.2-17.0) versus 12.0 (9.0-15.0), p = 0.070). The extra-parenchymal evaluation found no difference in the proportion of patients with pleural effusion, mediastinal lymphadenopathy, or thymal abnormalities in the two populations. The prevalence of pulmonary embolism was not significantly different between groups (8.7% versus 5.3%, p = 0.623, n = 175). CONCLUSION: There was no significant difference in disease severity as evaluated by chest CT in severe COVID-19 patients admitted to the ICU for hypoxemic acute respiratory failure with or without anti-IFN auto-Abs.

Automated CT LI-RADS v2018 scoring of liver observations using machine learning: A multivendor, multicentre retrospective study.

Background & Aims: Assessment of computed tomography (CT)/magnetic resonance imaging Liver Imaging Reporting and Data System (LI-RADS) v2018 major features leads to substantial inter-reader variability and potential decrease in hepatocellular carcinoma diagnostic accuracy. We assessed the performance and added-value of a machine learning (ML)-based algorithm in assessing CT LI-RADS major features and categorisation of liver observations compared with qualitative assessment performed by a panel of radiologists. Methods: High-risk patients as per LI-RADS v2018 with pathologically proven liver lesions who underwent multiphase contrast-enhanced CT at diagnosis between January 2015 and March 2019 in seven centres in five countries were retrospectively included and randomly divided into a training set (n = 84 lesions) and a test set (n = 345 lesions). An ML algorithm was trained to classify non-rim arterial phase hyperenhancement, washout, and enhancing capsule as present, absent, or of uncertain presence. LI-RADS major features and categories were compared with qualitative assessment of two independent readers. The performance of a sequential use of the ML algorithm and independent readers were also evaluated in a triage and an add-on scenario in LR-3/4 lesions. The combined evaluation of three other senior readers was used as reference standard. Results: A total of 318 patients bearing 429 lesions were included. Sensitivity and specificity for LR-5 in the test set were 0.67 (95% CI, 0.62-0.72) and 0.91 (95% CI, 0.87-0.96) respectively, with 242 (70.1%) lesions accurately categorised. Using the ML algorithm in a triage scenario improved the overall performance for LR-5. (0.86 and 0.93 sensitivity, 0.82 and 0.76 specificity, 78% and 82.3% accuracy for the two independent readers). Conclusions: Quantitative assessment of CT LI-RADS v2018 major features is feasible and diagnoses LR-5 observations with high performance especially in combination with the radiologist's visual analysis in patients at high-risk for HCC. Impact and implications: Assessment of CT/MRI LI-RADS v2018 major features leads to substantial inter-reader variability and potential decrease in hepatocellular carcinoma diagnostic accuracy. Rather than replacing radiologists, our results highlight the potential benefit from the radiologist-artificial intelligence interaction in improving focal liver lesions characterisation by using the developed algorithm as a triage tool to the radiologist's visual analysis. Such an AI-enriched diagnostic pathway may help standardise and improve the quality of analysis of liver lesions in patients at high risk for HCC, especially in non-expert centres in liver imaging. It may also impact the clinical decision-making and guide the clinician in identifying the lesions to be biopsied, for instance in patients with multiple liver focal lesions.

Development and validation of a new equation based on plasma creatinine and muscle mass assessed by CT scan to estimate glomerular filtration rate: a cross-sectional study.

Background: Inter-individual variations of non-glomerular filtration rate (GFR) determinants of serum creatinine, such as muscle mass, account for the imperfect performance of estimated GFR (eGFR) equations. We aimed to develop an equation based on creatinine and total lumbar muscle cross-sectional area measured by unenhanced computed tomography scan at the third lumbar vertebra. Methods: The muscle mass-based eGFR (MMB-eGFR) equation was developed in 118 kidney donor candidates (iohexol clearance) using linear regression. Validation cohorts included 114 healthy subjects from another center (51Cr-EDTA clearance, validation population 1), 55 patients with chronic diseases (iohexol, validation population 2), and 60 patients with highly discordant creatinine and cystatin C-based eGFR, thus presumed to have atypical non-GFR determinants of creatinine (51Cr-EDTA, validation population 3). Mean bias was the mean difference between eGFR and measured GFR, precision the standard deviation (SD) of the bias, and accuracy the percentage of eGFR values falling within 20% and 30% of measured GFR. Results: In validation population 1, performance of MMB-eGFR was not different from those of CKD-EPICr2009 and CKD-EPICr2021. In validation population 2, MMB-eGFR was unbiased and displayed better precision than CKD-EPICr2009, CKD-EPICr2021 and EKFC (SD of the biases: 13.1 vs 16.5, 16.8 and 15.9 mL/min/1.73 m2). In validation population 3, MMB-eGFR had better precision and accuracy {accuracy within 30%: 75.0% [95% confidence interval (CI) 64.0-86.0] vs 51.5% (95% CI 39.0-64.3) for CKD-EPICr2009, 43.3% (95% CI 31.0-55.9) for CKD-EPICr2021, and 53.3% (95% CI 40.7-66.0) for EKFC}. Difference in bias between Black and white subjects was -2.1 mL/min/1.73 m2 (95% CI -7.2 to 3.0), vs -8.4 mL/min/1.73 m2 (95% CI -13.2 to -3.6) for CKD-EPICr2021. Conclusion: MMB-eGFR displayed better performances than equations based on demographics, and could be applied to subjects of various ethnic backgrounds.

Chest computed tomography characteristics of critically ill COVID-19 patients with auto-antibodies against type I interferons.

Purpose: patients with auto-antibodies neutralizing type I interferons (anti-IFN auto-Abs) are at risk of severe forms of coronavirus disease 19 (COVID-19). The chest computed tomography (CT) scan characteristics of critically ill COVID-19 patients harboring these auto-Abs have never been reported. Methods: Bicentric ancillary study of the ANTICOV study (observational prospective cohort of severe COVID-19 patients admitted to the intensive care unit (ICU) for hypoxemic acute respiratory failure) on chest CT scan characteristics (severity score, parenchymal, pleural, vascular patterns). Anti-IFN auto-Abs were detected using a luciferase neutralization reporting assay. Imaging data were collected through independent blinded reading of two thoracic radiologists of chest CT studies performed at ICU admission (±72h). The primary outcome measure was the evaluation of severity by the total severity score (TSS) and the CT severity score (CTSS) according to the presence or absence of anti-IFN auto-Abs. Results: 231 critically ill COVID-19 patients were included in the study (mean age 59.5±12.7 years; males 74.6%). Day 90 mortality was 29.5% (n=72/244). There was a trend towards more severe radiological lesions in patients with auto-IFN anti-Abs than in others, not reaching statistical significance (median CTSS 27.5 (21.0-34.8] versus 24.0 (19.0-30.0), p=0.052; median TSS 14.5 (10.2-17.0) versus 12.0 (9.0-15.0), p=0.070). The extra-parenchymal evaluation found no difference in the proportion of patients with pleural effusion, mediastinal lymphadenopathy or thymal abnormalities in the two populations. The prevalence of pulmonary embolism was not significantly different between groups (8.7% versus 5.3%, p=0.623, n=175). Conclusion: There was no significant difference in disease severity as evaluated by chest CT in severe COVID-19 patients admitted to the ICU for hypoxemic acute respiratory failure with or without anti-IFN auto-Abs.

Pretreatment CT Texture Parameters as Predictive Biomarkers of Progression-Free Survival in Follicular Lymphoma Treated with Immunochemotherapy and Rituximab Maintenance.

The purpose of this study was to determine whether texture analysis features present on pretreatment unenhanced computed tomography (CT) images, derived from 18F-fluorodeoxyglucose positron emission/computed tomography (18-FDG PET/CT), can predict progression-free survival (PFS), progression-free survival at 24 months (PFS 24), time to next treatment (TTNT), and overall survival in patients with high-tumor-burden follicular lymphoma treated with immunochemotherapy and rituximab maintenance. Seventy-two patients with follicular lymphoma were retrospectively included. Texture analysis was performed on unenhanced CT images extracted from 18-FDG PET/CT examinations that were obtained within one month before treatment. Skewness at a fine texture scale (SSF = 2) was an independent predictor of PFS (hazard ratio = 3.72 (95% CI: 1.15, 12.11), p = 0.028), PFS 24 (hazard ratio = 13.38; 95% CI: 1.29, 138.13; p = 0.029), and TTNT (hazard ratio = 5.11; 95% CI: 1.18, 22.13; p = 0.029). Skewness values above -0.015 at SSF = 2 were significantly associated with lower PFS, PFS 24, and TTNT. Kurtosis without filtration was an independent predictor of PFS (SSF = 0; HR = 1.22 (95% CI: 1.04, 1.44), p = 0.013), and TTNT (SSF = 0; hazard ratio = 1.23; 95% CI: 1.04, 1.46; p = 0.013). This study shows that pretreatment unenhanced CT texture analysis-derived tumor skewness and kurtosis may be used as predictive biomarkers of PFS and TTNT in patients with high-tumor-burden follicular lymphoma treated with immunochemotherapy and rituximab maintenance.

Synthetic MR image generation of macrotrabecular-massive hepatocellular carcinoma using generative adversarial networks.

PURPOSE: The purpose of this study was to develop a method for generating synthetic MR images of macrotrabecular-massive hepatocellular carcinoma (MTM-HCC). MATERIALS AND METHODS: A set of abdominal MR images including fat-saturated T1-weighted images obtained during the arterial and portal venous phases of enhancement and T2-weighted images of 91 patients with MTM-HCC, and another set of MR abdominal images from 67 other patients were used. Synthetic images were obtained using a 3-step pipeline that consisted in: (i), generating a synthetic MTM-HCC tumor on a neutral background; (ii), randomly selecting a background among the 67 patients and a position inside the liver; and (iii), merging the generated tumor in the background at the specified location. Synthetic images were qualitatively evaluated by three radiologists and quantitatively assessed using a mix of 1-nearest neighbor classifier metric and Fréchet inception distance. RESULTS: A set of 1000 triplets of synthetic MTM-HCC images with consistent contrasts were successfully generated. Evaluation of selected synthetic images by three radiologists showed that the method gave realistic, consistent and diversified images. Qualitative and quantitative evaluation led to an overall score of 0.64. CONCLUSION: This study shows the feasibility of generating realistic synthetic MR images with very few training data, by leveraging the wide availability of liver backgrounds. Further studies are needed to assess the added value of those synthetic images for automatic diagnosis of MTM-HCC.

MRI features associated with HCC histologic subtypes: a western American and European bicenter study.

OBJECTIVES: To evaluate if preoperative MRI can predict the most frequent HCC subtypes in North American and European patients treated with surgical resection. METHODS: A total of 119 HCCs in 97 patients were included in the North American group and 191 HCCs in 176 patients were included in the European group. Lesion subtyping was based on morphologic features and immuno-histopathological analysis. Two radiologists reviewed preoperative MRI and evaluated the presence of imaging features including LI-RADS major and ancillary features to identify clinical, biologic, and imaging features associated with the main HCC subtypes. RESULTS: Sixty-four percent of HCCs were conventional. The most frequent subtypes were macrotrabecular-massive (MTM-15%) and steatohepatitic (13%). Necrosis (OR = 3.32; 95% CI: 1.39, 7.89; p = .0064) and observation size (OR = 1.011; 95% CI: 1.0022, 1.019; p = .014) were independent predictors of MTM-HCC. Fat in mass (OR = 15.07; 95% CI: 6.57, 34.57; p < .0001), tumor size (OR = 0.97; 95% CI: 0.96, 0.99; p = .0037), and absence of chronic HCV infection (OR = 0.24; 95% CI: 0.084, 0.67; p = .0068) were independent predictors of steatohepatitic HCC. Independent predictors of conventional HCCs were viral C hepatitis (OR = 3.20; 95% CI: 1.62, 6.34; p = .0008), absence of fat (OR = 0.25; 95% CI: 0.12, 0.52; p = .0002), absence of tumor in vein (OR = 0.34; 95% CI: 0.13, 0.84; p = .020), and higher tumor-to-liver ADC ratio (OR = 1.96; 95% CI: 1.14, 3.35; p = .014) CONCLUSION: MRI is useful in predicting the most frequent HCC subtypes even in cohorts with different distributions of liver disease etiologies and tumor subtypes which might have future treatment and management implications. KEY POINTS: • Representation of both liver disease etiologies and HCC subtypes differed between the North American and European cohorts of patients. • Retrospective two-center study showed that liver MRI is useful in predicting the most frequent HCC subtypes.

Generative adversarial networks (GAN)-based data augmentation of rare liver cancers: The SFR 2021 Artificial Intelligence Data Challenge.

PURPOSE: The 2021 edition of the Artificial Intelligence Data Challenge was organized by the French Society of Radiology together with the Centre National d'Études Spatiales and CentraleSupélec with the aim to implement generative adversarial networks (GANs) techniques to provide 1000 magnetic resonance imaging (MRI) cases of macrotrabecular-massive (MTM) hepatocellular carcinoma (HCC), a rare and aggressive subtype of HCC, generated from a limited number of real cases from multiple French centers. MATERIALS AND METHODS: A dedicated platform was used by the seven inclusion centers to securely upload their anonymized MRI examinations including all three cross-sectional images (one late arterial and one portal-venous phase T1-weighted images and one fat-saturated T2-weighted image) in compliance with general data protection regulation. The quality of the database was checked by experts and manual delineation of the lesions was performed by the expert radiologists involved in each center. Multidisciplinary teams competed between October 11th, 2021 and February 13th, 2022. RESULTS: A total of 91 MTM-HCC datasets of three images each were collected from seven French academic centers. Six teams with a total of 28 individuals participated in this challenge. Each participating team was asked to generate one thousand 3-image cases. The qualitative evaluation was performed by three radiologists using the Likert scale on ten randomly selected cases generated by each participant. A quantitative evaluation was also performed using two metrics, the Frechet inception distance and a leave-one-out accuracy of a 1-Nearest Neighbor algorithm. CONCLUSION: This data challenge demonstrates the ability of GANs techniques to generate a large number of images from a small sample of imaging examinations of a rare malignant tumor.

Focal Benign Liver Lesions and Their Diagnostic Pitfalls.

Focal hepatic lesions are frequently discovered incidentally on cross-sectional imaging or abdominal ultrasound, and in the general population, a vast majority of those incidental findings are benign entities. However, the formal diagnosis of benign liver lesions is not always straightforward and may require advanced imaging modalities, such as MRI with hepatobiliary contrast agent or contrast-enhanced ultrasound (CEUS). This review presents the typical features of the main benign liver lesions, including focal nodular hyperplasia (FNH), hepatocellular adenoma (HCA), hepatic cysts, hemangioma, angiomyolipoma, and pseudotumors, such as inflammatory pseudotumors or hepatic granulomas. However, beyond the specific and classical MRI features, some lesions may present atypical patterns. Moreover, arterial phase hyperenhancement, often present in benign liver lesions, can be seen in malignant lesions such as hepatocellular carcinoma. Hence, accurate analysis of clinical and biological contexts is mandatory to optimize our diagnostic performance. The objective of this investigation was, therefore, to review the specific presentations of benign liver tumors and to illustrate their diagnostic pitfalls.

Macrotrabecular-Massive Hepatocellular Carcinoma: Light and Shadow in Current Knowledge.

The subject of this narrative review is macrotrabecular-massive hepatocellular carcinoma (MTM-HCC). Despite their rarity, these tumours are of general interest because of their epidemiological and clinical features and for representing a distinct model of the interaction between the angiogenetic system and neoplastic cells. The MTM-HCC subtype is associated with various adverse biological and pathological parameters (the Alfa-foetoprotein (AFP) serum level, tumour size, vascular invasion, and satellite nodules) and is a key determinant of patient prognosis, with a strong and independent predictive value for early and overall tumour recurrence. Gene expression profiling has demonstrated that angiogenesis activation is a hallmark feature of MTM-HCC, with overexpression of both angiopoietin 2 (ANGPT2) and vascular endothelial growth factor A (VEGFA).

Liver resection for single large hepatocellular carcinoma: a prognostic factors study.

INTRODUCTION AND OBJECTIVES: Liver resection is the only curative therapeutic option for large hepatocellular carcinoma (> 5 cm), but survival is worse than in smaller tumours, mostly due to the high recurrence rate. There is currently no proper tool for stratifying relapse risk. Herein, we investigated prognostic factors before hepatectomy in patients with a single large hepatocellular carcinoma (HCC). MATERIAL AND METHODS: We retrospectively identified 119 patients who underwent liver resection for a single large HCC in 2 tertiary academic French centres and collected pre- and post-operative clinical, biological and radiological features. The primary outcome was overall survival at five years. Secondary outcomes were recurrence-free survival at five years and prognostic factors for recurrence. RESULTS: A total of 84% of the patients were male, and the median age was 66 years old (IQR 58-74). Thirty-nine (33%) had Child-Pugh A cirrhosis, and the mean Model for End-Stage Liver Disease (MELD) score was 6 (6-6). The aetiology of liver disease was predominantly alcohol-related (48%), followed by nonalcoholic steatohepatitis (22%), hepatitis B (18%) and hepatitis C (10%). The mean tumour size was 70 mm (55-110). The median overall survival was 72.5 months (IC 95%: 56.2-88.7), and the five-year overall survival was 55.1 ± 5.5%. The median recurrence-free survival was 26.6 months (95% CI: 16.0-37.1), and the five-year recurrence-free survival rate was 37.8 ± 5%. In multivariate analysis, preoperative prognostic factors for recurrence were baseline alpha-fetoprotein (AFP) > 7 ng/mL (p<0.001), portal veinous invasion (p=0.003) and cirrhosis (p=0.020). Using these factors, we created a simple recurrence-risk scoring system that classified three groups with distinct disease-free survival medians (p<0.001): no risk factors (65 months), 1 risk factor (36 months), and ≥2 risk factors (8.9 months). CONCLUSION: Liver resection is the only curative option for large HCC, and we confirmed that survival could be acceptable in experienced centres. Recurrence is the primary issue of surgery, and we proposed a simple preoperative score to help identify patients with the most worrisome prognosis and possible candidates for combined therapy.