RPS 407 - Kidney and bladder imaging

Author Block: E. Messina, A. Dehghanpour, A. Borrelli, S. Lucciola, C. Catalano, V. Panebianco; Rome/IT
Purpose: The validate a new bladder cancer (BCa) diagnostic pathway focusing on magnetic resonance imaging (MRI). Moreover, to identify clinical, pathological, and radiological characteristics, including the Vesical Imaging-Reporting and Data System (VI-RADS) score, individually demonstrating a significant correlation with muscle-invasive BCa (MIBC).
Methods or Background: Prospective multicentres study (four centres). Patients with BCa suspicion underwent MRI before trans-urethral resection of bladder tumour (TURBT). Two experienced uro-radiologists in each centre independently analysed MRIs. According to VI-RADS, a cutoff of ≥ 3 or ≥ 4 was assumed to define MIBC. TURBT and/or cystectomy reports were compared with preoperative VI-RADS scores to assess MRI accuracy for discriminating between non-muscle-invasive versus MIBC. Performance was assessed by ROC curve analysis. Univariable and multivariable logistic regression models were implemented.
Results or Findings: A final cohort of 251 patients was enrolled (median age 73 [IQR: 64, 77.5]). MRI showed sensitivity, specificity, PPV, NPV, and accuracy for MIBC detection ranging from 83-93%, 80-92%, 67-81%, 93-96%, and 84-89% for the more experienced readers. The area under the curve (AUC) was 0.95 (0.91-0.99). In the multivariable logistic regression model, haematuria (p = 0.02), tumour size (p=0.03), the VI-RADS score, using both a cutoff of 3 and 4 (p<0.0001), and concomitant hydronephrosis (p=0.02) were the variables correlating with a bladder cancer staged as ≥ T2. The inter-reader agreement was substantial (k=0.85).
Conclusion: VI-RADS score has been validated as a reliable preoperative tool predicting BCa muscle-invasiveness. Moreover, if integrated as an MRI biomarker with clinical and imaging data, VI-RADS can be integrated into a customised diagnostic approach to select patients more precisely, with the goal of reducing the risk of diseases’ understaging and enhancing clinical outcomes.
Limitations: High-volume referral centres, reduced reproducibility limit this study.
Funding for this study: Not applicable for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Institutional Ethics Committee approved this study.

Author Block: E. Yuan¹, Y. Chen¹, B. Song¹, J. Yao¹, H. Yin², H. Zhang²; ¹Chengdu/CN, ²Beijing/CN
Purpose: The study aims to develop and temporally validate the radiomic models for identifying pathological T3a upstaging in renal cell carcinoma.
Methods or Background: A total of 1003 patients with renal cell carcinoma in our hospital were retrospectively enrolled. The patients were assigned to development dataset (n = 729) and temporal validation dataset (n = 274). The pathological T3a status of each patient was pathologically confirmed. Radiomic features of the lesion and the 5 mm peritumoural area were extracted from the manually labeled portal venous phase CT images. Tumour model, peritumour model, and combine model were developed using tumour, peritumour, and both features. The modeling pipelines were internally validated using 1000*100 nested cross validation, and the final models were temporally validated in an independent temporal validation set. The model in each experiment was evaluated by using metrics of discrimination, calibration, and clinical utility. The frequencies of features being selected were also analysed.
Results or Findings: The tumour shape based feature "sphericity" was the most frequently selected feature with the largest coefficient in the final models. In internal validation, the tumour pipeline and combine pipeline showed similar discrimination (AUC = 0.8362±0.0327 vs. 0.8347±0.0331), calibration, and clinical utility, while the peritumour pipeline showed slightly but signification poor discrimination (AUC = 0.7994±0.0369), calibration, and clinical utility than the other two. In temporal validation, the tumour model, peritumour model, and combine model showed similar discrimination (AUC = 0.8457 vs. 0.8489 vs. 0.8426), calibration, and clinical utility.
Conclusion: The radiomic models showed favorable performance in predicting pathological T3a upstaging preoperatively in renal cell carcinoma patients.
Limitations: This is a single centre study. The models needs further multicentre and prospective validation.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The written consent was waived because of the retrospective design.

Author Block: G. Brembilla, M. Cosenza, T. Russo, G. Basile, D. Raggi, C. Mercinelli, F. Montorsi, A. Necchi, F. De Cobelli; Milan/IT
Purpose: The aim of this study was to investigate the diagnostic accuracy and reproducibility of the neoadjuvant chemotherapy VI-RADS (nacVI-RADS) in assessing MIBC response to neoadjuvant immunotherapy.
Methods or Background: Two hundred and twenty bladder MRI scans from 110 patients who underwent pre- and post-immunotherapy MRI prior to radical cystectomy (RC) were retrospectively reviewed by two readers using nacVI-RADS scores. The diagnostic accuracy of nacVI-RADS was assessed using histopathology of RC specimens as the standard of reference, relative to two endpoints: complete pathologic response (ypT0) and downstaging (ypT1). The threshold for MRI positivity was considered nacVI-RADS 3; for downstaging, an additional threshold of nacVI-RADS 4 was also tested. Interobserver agreement was assessed with agreement coefficient 1 (AC1), Cohen’s k coefficient and percentages of agreement.
Results or Findings: Complete pathologic response was observed in 42% of patients. Complete radiologic response (i.e.: nacVI-RADS 1-2) was reported in 35.5% and 49.9% by reader 1 and reader 2, respectively. nacVI-RADS 3 had a sensitivity and NPV for >ypT0 residual disease of 82% (95%CI: 74-88) and 74% (95%CI: 64-83), respectively; for >ypT1 residual disease, sensitivity and NPV were 97% (95%CI: 90-99) and 97% (95%CI: 91-99), respectively. For the assessment of residual >ypT1 disease, a potentially better tradeoff between sensitivity, specificity and NPV was obtained for higher positivity thresholds of MRI. nacVI-RADS had an AUC for ypT0 of 79.6%, for ypT1 of 86.3%. Interreader agreement was substantial regardless the MRI positivity threshold (AC1 0.65-0.68; k 0.66-0.67).
Conclusion: nacVI-RADS criteria showed good accuracy and reproducibility in assessing MIBC response to neoadjuvant treatments.
Limitations: Single centre, experienced readers limit this study.
Funding for this study: Funding for this research was provided by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, and Associazione Italiana per la Ricerca sul Cancro (AIRC), Grant number: MFAG: 2017 Id.20617.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Ethics Committee of the Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy.

Author Block: J. Han, Y. Tao, Y. Zhang; Zhuhai/CN
Purpose: The aim of this study was to develop and validate a diagnostic model based on contrast-enhanced CT for identifying clear cell renal cell carcinoma (ccRCC) in small renal masses (SRMs).
Methods or Background: This retrospective multicentre study enrolled patients with pathologically confirmed SRMs. Data from three centres were used as training set (n=229), with data from one centre serving as an independent external validation set (n=81). Univariate and multivariate logistic regression analyses were used to screen independent risk factors for ccRCC and build the classification and regression tree (CART) model. Three radiologists were asked to diagnose the SRMs in each case independently based on professional experience and re-evaluated using the CART model.
Results or Findings: A total of 71% (220/310) of SRMs were ccRCC. Enhancement pattern, early dark cortical band (EDCB), the ratio of lesion to normal cortex attenuation (L/C) in the corticomedullary phase, non-enhancing phase L/C and sex were used to develop the CART model. In the testing cohort, the AUC and accuracy of the CART model were 0.903, 85.1%. The accuracy of radiologists was 67.9%,58.0%, and 56.8%. With the CART model support, the accuracy of radiologists improved to 86.4%,84.0%,79.0%. Interobserver agreement was significantly improved with the use of model aids (0.323 vs 0.654, P＜0.001).
Conclusion: The CART diagnostic model can identify ccRCC in SRMs and help radiologists make the diagnosis, potentially reducing the number of unnecessary biopsies.
Limitations: First, since it was a retrospective study, the existing selection bias may affect the results, and further prospective verification is needed in future work. Second, the data were collected from different institutions, and the scanning protocols were different. To minimise the impact of internal and external factors on the results, the quantitative data were standardised, and an independent external validation set was established.
Funding for this study: This work was supported in part by the National Key Research and Development Program of China under Grant Nos. 2023YFE0204300, in part by the National Natural Science Foundation of China under Grant No: 81801809, 82371917, 81830052, 81971691, 12126610, 62371476; in part by the Guangzhou Technology Program of Agriculture and Social Development of Key Research and Development Scheme under Grant No: 2023B03J1237, in part by the Basic and Applied Basic Research Foundation of Guangdong Province under Grant No: 2020A1515010572, and in part by the Zhuhai Basic and Applied Basic Research Foundation under Grant No: ZH22017003200001PWC.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the institutional review boards of the Fifth Hospital of Sun Yat- sen University.

Author Block: L. Lofino, F. Schwartz, M. Al Tarhuni, A. Abadia, F. Ria, E. Samei, D. Marin; Durham, NC/US
Purpose: The study aims to compare the absolute CT attenuation errors of cystic renal lesions and abdominal organs on virtual non-contrast images (VNC) between photon-counting (PCCT) and energy-integrating detector CT (EID).
Methods or Background: In this HIPAA compliant, IRB-approved retrospective study, multiphase CT scans from one PCCT and two EID dual-source dual-energy CTs were retrieved. A total of 56 BMI-matched patients (26 women, 30 men; mean age 58.5 ± 15.3 years; mean BMI 29.0 ± 6.8 kg/m2), were included: 16 for PCCT and 20 each per EID systems. Attenuation measurements of abdominal organs were recorded on VNC and True Noncontrast (TNC) datasets. Furthermore, attenuation measurements of 16 cystic renal lesions (eight for PCCT and eight for EID) were compared on VNC and TNC. Absolute CT attenuation errors were calculated and compared between PCCT and EID for the entire population and a subset of 20 obese patients using paired t-tests. Absolute CT attenuation errors were also compared for all cystic renal lesions and for renal lesions <1 cm, separately.
Results or Findings: PCCT yielded significantly lower absolute CT attenuation errors than EID on VNC for spleen (2.6 ± 6.2 vs 8.0 ± 10.3) and pancreas (4.4 ± 1.8 vs 7.7 ± 9.7) for all patients and for liver, spleen and pancreas in the obese patient cohort (P<0.05). Furthermore, PCCT yielded significantly lower absolute CT attenuation errors compared to EID for all cystic renal lesions (2.0 ± 1.3 vs. 12.0 ± 8.9) and for renal lesions <1 cm (1.4 ± 0.9 vs. 19.1 ± 6.8), P<0.01.
Conclusion: PCCT has significantly lower absolute CT attenuation errors for abdominal organs and cystic renal lesions in VNC images, compared to two dual-energy EID CTs.
Limitations: Single centre, retrospective study with limited number of cases.
Funding for this study: No funding was required for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethics comittee approved the study.

Author Block: A. Dehghanpour, E. Messina, A. Borrelli, S. Lucciola, C. Catalano, V. Panebianco; Rome/IT
Purpose: Recently a novel scoring system named nac-VIRADS (neoadjuvant chemotherapy VI-RADS) was proposed to assess radiological response to chemotherapy in MIBC (muscle-invasive bladder cancer) patients using mpMRI.
The primary aim of this study was to validate the performance of nac-VIRADS scoring system and assess its inter-reader agreement. Additionally, to investigate the radiological and clinicopathological features independently correlating with response to systemic therapy.
Methods or Background: This prospective multicenter study included patients with non-metastatic MIBC who underwent trans-urethral resection of bladder tumour (TURBT) and/or repeated TUR, followed by neoadjuvant chemotherapy (NAC), radical cystectomy (RC), and extended pelvic lymph node dissection. Patients underwent pre- and post-treatment multiparametric MRI. Radiological response was evaluated by two experienced urogenital radiologists using nac-VIRADS scoring system. The reference standard was histopathologic reports and tumour regression grade (TRG) from RC.
Nac-VIRADS performance was assessed by means of ROC analysis, deriving sensitivity, specificity, PPV, NPV, and accuracy. Interreader agreement was determined with Cohen’s k statistics. Univariable and multivariable analyses were implemented.
Results or Findings: Fifty-five patients were included. n=13 were classified as nac-VIRADS 1-2, n=14 as nac-VIRADS 3, n=22 as nac-VIRADS 4, and n=6 as nac-VIRADS 5. Overall, nac-VIRADS score showed a high diagnostic performance in concordance with histological results expressed by TRG. Diagnostic performance showed a range of 86%-92% sensitivity, 71%-79% specificity, 80%-82% (PPV, and 79%-83% NPV. AUC was 0.92 (95% CI: 0.83–1) for the more experienced reader. Inter-reader agreement was almost perfect (K = 0.82).
Conclusion: nac-VIRADS scoring system offers a reliable and reproducible approach, employing a clear algorithm, to assess response to systemic therapy in MIBC patients. Its implementation in a standardised pathway has the potential to significantly influence therapeutic decision-making and enhance overall patient survival.
Limitations: Limited sample size limits the scope of this study.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the Institutional Ethical Committee.

Author Block: C. O. Page, D. Ap Dafydd, C. Kelly-Morland, W. Cazzaniga, D. Nicol, D. Levine, S. J. Withey; London/UK
Purpose: The aim of this study was to investigate the feasibility of utilising preoperative contrast-enhanced CT to estimate renal split function, compared to standard-of-care nuclear medicine studies (DMSA and MAG3) in preoperative planning for nephrectomy.
Methods or Background: Retrospective analysis was conducted on patients with either DMSA- or MAG3-derived estimates of split renal function (NM-SRF) and a contemporaneous contrast-enhanced CT scan. Patients with hydronephrosis were excluded. Using reconstruction software with density thresholding (Syngo.via, Siemens Healthineers), renal volumes and mean attenuation values were calculated. The product of these from each kidney was used to calculate a CT-derived estimated split renal function (CT-SRF). The estimated percentage renal function from the right kidney from NM-SRF and CT-SRF were compared using Pearson correlation coefficient. The estimated remaining renal function following nephrectomy using NM-SRF and CT-SRF was compared to actual GFR values obtained >3 months post-operative. The difference between estimated and true post-operative GFR values were compared using Student’s T-test.
Results or Findings: n=49 patients (mean age 65.8 ± 9.3 years; 73.4% male) were included in the initial analysis. The Pearson correlation coefficient between CT-SRF and NM-SRF was 0.835, indicating a very strong correlation (p=<0.0001)
For n=28 patients, post-operative renal function (>3 months following the date of nephrectomy) was available. CT-SRF and NM-SRF underestimated post-operative GFR by 12.1(±11.9) and 10.5(±12.2) ml/min/1.73 m2, respectively. There was no significant difference in performance between these tests (p=0.604). Time to calculate CT-SRF was 157±68 seconds.
Conclusion: CT-SRF was a time-efficient and feasible way of estimating split renal function with comparable performance to standard-of-care NM-SRF in this small retrospective cohort.
Limitations: Larger prospective studies are required to assess whether it is possible to use the available CT data to avoid the additional radiation exposure, hospital attendance, and cost associated with DMSA and MAG3 testing.
Funding for this study: No funding was obtained for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Local institutional board review approved this study.

Author Block: M. Poida, M. Paniagua González, M. Camarena Gea, C. Gerlotti Slusnys, P. López Gómez, J. De Luis Yanes, C. Calles Blanco, E. De Miguel Campos, A. Garcia Perez; Madrid/ES
Purpose: The R.E.N.A.L. nephrometry score was developed to classify renal masses based on their complexity (low, intermediate, and high) and guide decision-making in cases that could benefit from partial nephrectomy. Although initially designed for this purpose it might also be useful for predicting outcomes in percutaneous ablation of renal masses-a treatment that has emerged as an alternative to surgery and has shown promising results in small renal masses. In this study, we aim to examine the cases of renal masses treated with ablation in our institution.
Methods or Background: We reviewed cross-sectional studies prior to the intervention and retrospectively applied the R.E.N.A.L. score to a total of 81 renal masses treated at our institution between January 2020 and June 2023, obtaining two groups based on the complexity of the lesions: low (N=44) and intermediate-high (N=37). Additionally, we analysed all available follow-up studies to date to identify complications and signs of tumour recurrence.The Fisher’s exact test was used to test for an association of R.E.N.A.L. score with complications and recurrences.
Results or Findings: The immediate complications (the majority being hematomas at the site of intervention) occurred in 3 cases of low complexity (6.8%) and in 9 cases of intermediate-high complexity (24.3%). There was a significant association between the complexity group and the immediate complication rate (p<0.033). Tumoural recurrences were seen in 8 cases of low complexity (18.1%) and in 3 cases of intermediate-high complexity (8.1%). No significant association has been found (p=0.331).
Conclusion: The R.E.N.A.L. scoring system could be a useful tool for predicting outcomes in percutaneous ablation for renal masses.
Limitations: Due to a lack of cases in the high complexity group, we had to fuse it with the intermediate complexity group for statistical analysis.
Funding for this study: No funding was obtained for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: No ethical approval was required for this study.

Author Block: S. Bodard; Paris/FR
Purpose: Estimation of glomerular function is necessary to diagnose kidney diseases. However, the study of glomeruli in the clinic is currently done indirectly through urine and blood tests. A recent imaging technique called Ultrasound Localisation Microscopy (ULM) has appeared. It is based on the ability to record continuous movements of individual microbubbles in the bloodstream. The aim of this study was to use ULM for glomeruli visualisation. Although ULM improved the resolution of vascular imaging up to tenfold, the imaging of the smallest vessels had yet to be reported.
Methods or Background: We acquired ultrasound sequences from living humans and then applied filters to divide the data set into slow-moving and fast-moving microbubbles. We performed a double tracking to highlight and characterise populations of microbubbles with singular behaviours. We decided to call this technique “sensing ULM” (sULM).
Results or Findings: In this study, we report the observation of microbubbles flowing in the glomeruli in living humans. We present a set of analysis tools to extract quantitative information from individual microbubbles, such as remanence time or normalised distance.
Conclusion: As glomeruli play a key role in kidney function, it would be possible that their observation yields a deeper understanding of the kidney. It could also be a tool to diagnose kidney diseases in patients. More generally, it will bring imaging capabilities closer to the functional units of organs, which is a key to understand most diseases, such as cancer, diabetes, or kidney failures.
Limitations: The spatial resolution of CEUS, i.e. between 70 and 150 μm, the localisation error of 43±4 μm - that increases with depth due to divergent ultrasound beam - and the presence of efferent and afferent arterioles in the same track as glomeruli leaded us to a rough PDW estimation with sULM
Funding for this study: This study was funded by the European Research Council under the European Union Horizon H2020 program (ERC Consolidator grant agreement No 772786-ResolveStroke).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The Ethics Committee of the French Society of Radiology approved this study (CERF, reference number CRM-2203-240).

Author Block: S. Reischl, M. Makowski, A. Sauter; Munich/DE
Purpose: Neoadjuvanttherapy, including radioimmunotherapy (RIT), is considered for locally advanced bladder cancer (BC) patients that are not candidates for primary surgery. Evaluating tumor response to neoadjuvant therapy is crucial for determining the feasibility of bladder-sparing strategies. This study aims to evaluate the utility of multiparametric magnetic resonance imaging (mpMRI) using the Vesical Imaging-Reporting And Data System (VI-RADS) criteria in evaluation of tumour response to RIT.
Methods or Background: This study was designed as secondary analysis of a tertiary-center, prospective study including patients with locally advance BC treated by neoadjuvant RIT and followed by cystectomy. MpMRI was performed pre- and post-RIT and the imaging protocol followed the VI-RADS guidelines, including T2-weighted, diffusion-weighted and contrast-enhanced sequences. Semi-quantiative scoring image analysis was performed by two radiologists. Inter- and intrareader agreement were assessed using Kendall's τ. Histopathological examination served as ground truth for the evaluation of the diagnostic accuracy of mpMRI in post-RIT mpMRI.
Results or Findings: Fifteen patients underwent post-RIT mpMRI and subsequent cystectomy. Intrareader agreement (τ=0.74; p<0.001 and τ=0.53; p<0.015) was superior to interreader agreement (τ=0.38; p<0.081) post-RIT. Of nine patients with residual tumour only one case of carcinoma in situ (pTis) was scored false-negative, while of five cases with muscle-invasive residual tumour (≥pT2) only one patient was scored false-negative (VI-RADS≤2) for muscle invasion, in each case by one reader.
Conclusion: This study suggests VI-RADS criteria in mpMRI can be valuable tools for evaluating tumour response in BC patients after RIT. These findings open the door to expanding bladder-sparing strategies for a broader patient population, pending validation in larger studies.
Limitations: This study was restricted by its limited cohort size.
Funding for this study: This study was funded by inhouse ressources the Department of Radiology of the TUM.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the ethical review board of our institution.

Author Block: B. K. Alabri, I. Alsalmi; Muscat/OM
Purpose: The main goal of this study is to evaluate the utility of standard deviation and mean Hounsfield Unit (HU) values in plain CT scans to distinguish between solid/complex cystic lesions and benign cystic lesions in the kidneys. The hypothesis is that there is a statistically significant difference in HU standard deviation between cystic and solid renal masses.
Methods or Background: A retrospective analysis of 239 cases over the past 13 years was conducted, focusing on histologically proven renal masses for patients who underwent a plain CT scan of the abdomen prior to biopsy or resection of a renal lesion, with exclusion criteria applied for cases of angiomyolipomas and lesions smaller than one cm. Attenuation values were assessed through region of interest (ROI) measurements, ensuring comprehensive coverage while avoiding areas of calcifications and necrosis. Renal MRI data from the same period were used as a control group for simple cysts.
Results or Findings: The study included 239 cases, dividing renal masses into simple cystic (n=115, 48.1%) and solid (n=124, 51.9%) groups based on CT scans. Average Hounsfield Units (HU) in simple cysts were much lower (mean=7.10) than in solid lesions (Mean=37.35) with a p-value <0.001. Standard deviation (SD) values followed a similar pattern, being lower in simple cysts (Mean=12.85) compared to solid lesions (Mean=16.61) with a p-value <0.001. HU was more effective than SD in distinguishing them. A scoring system combining HU and SD effectively differentiated between cystic and solid renal lesions, showing potential clinical use.
Conclusion: Combining Hounsfield unit and standard deviation values improves renal lesion characterisation on CT scans, offering potential for more precise clinical decisions and patient management.
Limitations: Single-centre retrospective nature, raising possible issues with regard to generalisation of results and selection bias and sample size limit this study.
Funding for this study: No funding was provided for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: The study is retrospective and hence, no ethical approval was sought.