RPS 801 - Advances in liver imaging

Author Block: N. Abou Zeid, C. Nelles, Z. Gurbanova, N. Große Hokamp, T. Persigehl, S. Lennartz; Cologne/DE
Purpose: To investigate the diagnostic utility of spectral reconstructions for determining cystic nature of small, hypoattenuating liver lesions.
Methods or Background: Patients with portal venous phase dual-layer dual-energy CT (dlDECT) who were diagnosed with hypoattenuating liver lesions smaller than one centimeter that were verified as cysts in corresponding MRI examinations were retrospectively included. ROI-based measurements were conducted by two raters in conventional images (CI), virtual unenhanced images (VUE) and iodine images. CT-based determination of cystic nature of the lesions was conducted using a HU attenuation threshold of less than 20 HU or an iodine concentration threshold of less than 0.5 mg/dl, the latter of which has been reported as the scanner-specific lower limit of iodine detection. Accuracy for determining cystic nature was compared between CI, VUE and iodine images.
Results or Findings: 77 patients with 287 small liver cysts were included. Mean attenuation in CI for small liver cysts was 20.8 ± 24.4 HU, and 11.1 ± 16.8 HU in VUE images. Mean iodine concentration was 0.46 ± 0.57 mg/dl. Using the 20 HU threshold in CI resulted in an accuracy of 60.3 % (173/287), whereas the corresponding accuracy using the same threshold in VUE images was 76.7 % (220/287). Accuracy solely based on the iodine threshold was 57.5 % (165/287). Combining the VUE and iodine threshold resulted in an accuracy of 92.7 % (266/287) for determining cystic nature of the lesions in dlDECT.
Conclusion: Combining quantitative VUE and iodine measurements using established thresholds facilitated correctly diagnosing 92.7% of small hypoattenuating liver lesions as cysts, compared to an accuracy of 60.3 % when using HU measurements in conventional images. This approach may help reducing correlative imaging and thereby accelerating staging of cancer patients.
Limitations: Retrospective, mono-centric study
Funding for this study: None to report
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: IRB waiver due to retrospective nature of the study.

Author Block: T. Russo, A. Belardo, A. Della Corte, D. Santangelo, F. Calabrese, M. M. Vincenzi, M. Mori, C. Fiorino, F. De Cobelli; Milan/IT
Purpose: To investigate the potentials of MRI radiomics to predict recurrence (R), hepatic recurrence (IR), and overall survival (OS) in a cohort of patients with colorectal liver metastases (CLM) who underwent microwave ablation (MWA) performed alone or in combination with surgical resection.
Methods or Background: 121 CLM patients with pre-operative Gadoxetic acid-MRI treated at our Institute between October 2015 and December 2022 were analyzed. One observer manually segmented the largest CLM on T2 scans. The abdominal aorta at the level of second lumbar vertebrae was used for the z-score normalization of the lesion. Cox multivariate analysis was run to establish a few-features radiomic model (RAD-T2), to predict recurrences and death. A bootstrap-based methodology for robust feature selection, including redundancy filtering, was optimized to select the best combination of two, three, four features. Correction of the models for optimism was then performed by internal bootstrap-based validation.
Results or Findings: For R, IR and OS the median follow-ups were respectively 12, 13 and 23 months; the number of events were 80, 68 and 34. After corrections for optimism, the resulting best RAD-T2 models were based on the combination of 2-3 features; they were able to predict R with C-index=0.65 (p=0.0002), IR (C-index=0.64, p=0.0029) and OS (C-index=0.71, p=0.0046). As an example, based on the best cut-off value of the RAD-T2 index, OS at 2 year was 58% and 88% when the cohort was stratified accordingly.
Conclusion: T2-MRI-based radiomic evaluation of CLMs is feasible and potentially useful for outcome prediction.
Limitations: The limited number of patients and the retrospective nature of the study.
Funding for this study: The limited number of patients and the retrospective nature of the study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: All procedures were carried out in accordance with the Declaration of Helsinki (1964) and its later amendments.

Author Block: A. Del Gaudio¹, K. Kalisz¹, D. Kruse¹, F. Ria¹, D. De Santis², L. Lofino¹, D. Marin¹; ¹Durham, NC/US, ²Rome/IT
Purpose: To intraindividually compare image quality and lesion conspicuity of abdominal MRI using gadopiclenol at 0.05 mmol/kg and gadobenate dimeglumine (Gd-BOPTA) at 0.1 mmol/kg.
Methods or Background: From September 2023 to March 2024, consecutive patients who had undergone two clinically indicated abdominal MRIs within 12 months using gadopiclenol and Gd-BOPTA on the same scanner were retrospectively enrolled. One independent radiologist manually measured the signal intensity of abdominal organs, arterial and venous vessels, and abdominal lesions (liver, pancreas, and kidneys) on unenhanced, late arterial, venous, and equilibrium phases. SNR, CNR, and magnitude of contrast enhancement (ΔE) were calculated for all organs and vessels on each contrast-enhanced phase. Percentage enhancement (%E) was calculated for all lesions on contrast-enhanced phases. Subjective image quality was assessed using a 5-point Likert scale, including: organs' and vessels' enhancement, liver-to-vessels contrast, and overall image quality. Lesion characteristics were also evaluated. Statistical analysis employed paired t- and Wilcoxon tests.
Results or Findings: One hundred subjects (64 years ± 14; 55 men) and 21 abdominal lesions were included. Compared to Gd-BOPTA, gadopiclenol yielded significantly higher CNR and SNR for pancreas, porta, and kidney in the late arterial phase (p ≤ .040). No significant differences in CNR and SNR were observed between gadopiclenol and Gd-BOPTA across all organs in the portal venous and equilibrium phases. Gadopiclenol showed significantly higher pancreatic ΔE in all contras-enhanced phases (p ≤ .049) compared to Gd-BOPTA.
The %E of abdominal lesions was comparable between gadopiclenol and Gd-BOPTA for all contrast-enhanced phases (p ≥ .100). No significant differences were observed in readers’ perception of image quality and lesions’ characteristics.
Conclusion: Gadopiclenol at 0.05 mmol/kg yields similar image quality and improved pancreatic enhancement compared to Gd-BOPTA at 0.1 mmol/kg.
Limitations: Retrospective study design
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Written informed consent was waived and Institutional Review Board approval was obtained

Author Block: J. Lohrke, T. Brumby, S. Herbert, T. Frenzel, G. Jost, M. Berger, H. Pietsch; Berlin/DE
Purpose: The established liver-specific gadolinium-based magnetic resonance imaging (MRI) contrast agents, gadoxetate disodium and gadobenate dimeglumine are based on linear, DTPA like ligands. In the present study a new early preclinical macrocyclic liver-specific candidate will be presented.
Methods or Background: The MRI efficiency (r1-relaxivity) of the candidate BAY 3393081 was determined at 37°C 1.41 T in human plasma. The kinetic inertness of the complex stability was investigated using an established zinc transmetallation assay. The in vitro liver cell uptake was assessed in rat hepatocytes and human transfected organic anion transporter protein 1B1 or 1B3 embryonic kidney cells. Pharmacokinetic parameters were evaluated in rodent (rats) and non-rodent (dogs) species by analyzing the gadolinium (Gd) concentrations in plasma over time. The in vivo liver elimination was examined in bile-duct-cannulated rats and the bile was analyzed using inductively coupled plasma mass spectroscopy. Contrast-enhanced liver MRI was performed in mice, pigs and a VX2 tumor model in rabbits.
Results or Findings: The relaxivity of BAY3393081 was determined with 8.7±0.3 L/(mmols) in human plasma at 1.41 T. The kinetic inertness of the complex stability was comparable to marketed macrocyclic GBCAs. The in vitro cell uptake results revealed that BAY3393081 is specifically taken up by rat and human OATPs. BAY 3393081 showed a high plasma clearance in rat and dog. In rats ~80% of the injected dose were eliminated in an unchanged form via the bile. Strong signal enhancement of liver parenchyma was demonstrated in rats, rabbits and pigs.
Conclusion: The preclinical candidate BAY3393081 showed the high kinetic inertness of macrocyclic GBCAs and a strong liver parenchyma enhancement in the MRI.
Limitations: Limited transferability of preclinical data to human liver elimination due to significant interspecies variability in hepatobiliary transporter expression.
Funding for this study: Funding for this preclinical study was provided by Bayer AG.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: All animal experiments were approved by LaGeSo.

Author Block: D. Catucci, S. U. Von Däniken, V. Obmann, A. Berzigotti, L. Ebner, J. T. Heverhagen, A. Christe, P. Vermathen, A. T. Huber; Bern/CH
Purpose: This study aimed to analyse the performance of intravoxel incoherent motion (IVIM) imaging parameters (tissue-diffusivity D, perfusion-fraction PF and pseudo-diffusion-coefficient D*) and liver stiffness (LS) measured by magnetic resonance elastography (MRE) to screen for chronic liver disease (CLD) and clinically significant portal hypertension (CSPH) on liver MRI examinations.
Methods or Background: This prospective study included 103 patients without CLD (noCLD-group, n=103) and 82 patients with biopsy-proven CLD who underwent liver MRI examinations including MRE and IVIM imaging between 03/2016 and 11/2023. Patients with CLD were subdivided based on their liver fibrosis degree: early CLD (F0-F1; eCLD-group, n=21), intermediate CLD (F2; iCLD-group, n=19), advanced CLD (F3-F4; aCLD-group, n=20) and aCLD with CSPH according to the BAVENO VII consensus (aCLDPH-group; n=22). IVIM imaging parameters (D, PF and D*) of the liver as well as LS were measured in all patients. Statistical analysis included the Kruskal-Wallis test for group comparison and receiver operating characteristic (ROC) curve analysis with multiple logistic regression analysis for group differentiation.
Results or Findings: D, PF, D* and LS differed significantly between all groups (p<0.001). The overall best parameter for detecting CLD and CSPH was LS with a cut-off value of >3.2 kPa (sensitivity 67%/specificity 96%) respectively >3.8 kPa (sensitivity 95%/specificity 88%), both with p<0.001. For CLD-detection, a combination of D <99 x 10-5 mm2/sec and LS >3.2 kPa increased the AUC from 0.89 with LS to 0.95. For CSPH-detection, a combination of D* <434 x 10-5 mm2/sec and LS >3.8 kPa increased the specificity (90%) with a slightly lower sensitivity (91%, p<0.001).
Conclusion: IVIM imaging parameters (D, PF and D*) as well as LS measured by MRE allow non-invasive screening for CLD and CSPH.
Limitations: This was a single center study and should be externally validated.
Funding for this study: This study received funding by the Swiss National Science Foundation (SNF), grant number 188591.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the cantonal ethics committee of Bern (Kantonale Ethikkommission Bern).

Author Block: P. Spiesecke, T. Fischer; Berlin/DE
Purpose: Superharmonic imaging is a useful B-mode ultrasound technology increasing spatial resolution. Currently, there is a novel technology available which produces B-mode images by a combination of different harmonics up to third harmonic using an Artificial Intelligence (AI) driven algorithm. The aim of the present study is the first evaluation of this technology.
Methods or Background: For this prospective study, overall 52 healthy test persons and patients were included. Standard and novel B-mode images of the gallbladder were captured in each subject – in each case by recording several combinations of additional parameters such as Dynamic Range and Speckle Reduction. For this purpose, a premium ultrasound system was used (Canon Aplio i800 including third-harmonic imaging). The images were manually segmented manually and were subjected to computer-aided analysis to analyze artifacts and edge sharpness. Additionally, Radiologists in different stages of training and subspecialisation rated the images by means of different parameters on a Likert scale.
Results or Findings: N = 26 data sets each with and without gallbladder pathology were included. In gallbladder B-mode ultrasound, ratings of the Third Harmonic Imaging-derived images showed a significant reduction of artifacts in the gallbladder lumen as well as a higher sharpness of interfaces. Subjective analysis revealed higher image quality in Third Harmonic Imaging compared to standard B-mode.
Conclusion: Our results suggest, that the AI-driven Third Harmonic Imaging can be a useful tool to increase the sharpness of interfaces and reduce the artifacts in gallbladder B-mode ultrasound.
Limitations: The present study is a single center study which first evaluates this novel AI-driven Third Harmonics Imaging B-mode ultrasound technology. Further studies are necessary to evaluate this technology more detailed.
Funding for this study: None.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Local ethics committee.