RPS 1607 - Imaging of renal lesions

Author Block: B. Stensby Breen¹, H. Søgaard Christensen², J. Brøndum Frøkjær², O. Graumann³, M. Kleivane¹, A. Negard¹, I. Mjåland Salte¹, J. Borgbjerg¹; ¹Oslo/NO, ²Aalborg/DK, ³Aarhus/DK
Purpose: To assess the agreement of maximum tumor diameter and tumor nearness between non-contrast MRI and contrast-enhanced CT for SRM surveillance, and to compare diagnostic confidence in tumor delineation across modalities.
Methods or Background: Active surveillance (AS) of small renal masses (SRMs) typically relies on contrast-enhanced CT for serial size measurement. Non-contrast MRI protocols may provide an alternative.
We retrospectively identified 50 patients (mean age, 62.9 years; 44% women) who underwent both contrast-enhanced CT and axial T2-weighted MRI within 3 months as part of AS protocol. Nine radiologists from five hospitals independently assessed maximum tumor diameter (axial plane) and tumor nearness to the collecting system (TN, three-tier scale) and subjectively graded their diagnostic confidence in delineating the tumor contour (DC, 5-point scale) in 50 CT and MRI scans across four randomized reading sessions. Observer agreement for diameter assessment was quantified using the limits of agreement with the mean (LOAM), whereas TN and DC was analyzed using Gwet’s AC2.
Results or Findings: Mean tumor diameters were 19.9 mm for CT and 19.5 mm for MRI (mean difference, 0.4 mm; p=0.051). Reproducibility LOAMs were ±2.9 [2.7-3.2] mm (CT) and ±3.1 [2.9-3.4] mm (MRI), with overlapping confidence intervals. TN ratings and agreement were similar for CT and MRI (p=0.61; AC2=0.67 each), whereas DC ratings were higher for CT (p<0.001).
Conclusion: Non-contrast axial T2-weighted MRI demonstrated agreement comparable to contrast-enhanced CT for tumor diameter and tumor nearness, while diagnostic confidence favored CT without affecting agreement.
Limitations: A variety of MRI systems with a span in scan parameters.
Retrospective study design with patients from a single center.
Renal mass sample size is relatively small, albeit comparable to previous studies.
Maximum axial tumor diameter was evaluated to remain consistent with clinical standards and previous research.
Funding for this study: Akershus University Hospital
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the Norwegian Regional Committees for Medical and Health Research Ethics (reference 285666) and by the institutional data protection officer with a waiver for informed patient consent.

Author Block: G. Imperiale, G. Brembilla, C. Rodolfi, F. Cei, A. Larcher, G. Rosiello, U. Capitanio, F. De Cobelli; Milan/IT
Purpose: Evaluate the frequency of Clear Cell Likelihood Score (CCLS) features of Small Renal Masses (SRMs) on multi-parametric Magnetic Resonance (mpMRI) and investigate additional imaging characteristics to improve diagnostic accuracy and better define Renal Cell Carcinoma (RCC).
Methods or Background: Preoperative multiparametric mpMRI of patients with SRMs subject to surgery were retrospectively evaluated by two radiologists with intermediate experience blinded to the histological diagnosis (gold standard). CCLS features evaluated were: T2 signal intensity, microscopic fat, cortex vascularity, segmental enhancement inversion (SEI), arterial-delayed enhancement ratio (ADER >1.5); additionally, T2-heterogeneity, peritumoral-pseudocapsule and heterogeneity were assessed.
Results or Findings: 80 patients with 33 clear cell RCCs (41.25%), 25 papillary RCCs (31.25%), 9 oncocytomas (11.25%), 9 chromophobe RCCs (11.25%) and 4 fat-poor angiomyolipomas (5%) were evaluated. Clear cell RCCs (ccRCC) showed T2-hyperintensity (75,76%), heterogeneous signal (81,82%), pseudocapsule (72,73%), microscopic fat (54,55%) and marked hypervascular enhancement (96,97%), with heterogeneous pattern (81,82%). Papillary RCCs (pRCC) were predominantly T2-hypointense (64%), with lower frequency of pseudocapsule (52%), reduced vascularity (64% hypovascular) and rare heterogeneous pattern of enhancement (24%).
Conclusion: MRI characterization of the major renal tumor histotypes confirms established features from the CCLS, such as hypervascularity and T2-hyperintensity in ccRCC and hypovascularity and T2-hypointensity in pRCC. It also identified T2-signal intensity and enhancement heterogeneity and the presence of a pseudocapsule as discriminators. These findings may aid radiologists in refining non-invasive differentiation among renal tumor subtypes and may be added to the CCLS in order to increase specificity.
Limitations: These include the retrospective single center design. Confirmation of features identified vs gold standard in this preliminary evaluation should be confirmed on larger cohorts.
Funding for this study: This study received no external funding.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the Ethics Committee.

Author Block: O. M. S. M. Sherif, A. Mousa, M. Zaky, A. Abdelhamid, M. E. Abou El-Ghar, T. A. El-Diasty; Mansoura/EG
Purpose: We aimed prospectively to assess effectiveness of Clear Cell Likelihood Score in small renal masses and its role in predicting tumor grade.
Methods or Background: The detection of small renal masses has increased due to recent widespread of cross-sectional imaging . Clear cell renal cell carcinoma (ccRCC) is the most common subtype with rapid progression and metastasis.
We utilized MRI Clear Cell Likelihood Score to investigate its ability to distinguish and grade different subtypes.
Results or Findings: In total, 103 patients with small solid renal masses of stage T1a (≤4 cm) were identified. Mean tumor size was 3.4 ± 0.6 cm. According to our study results, the clear cell Likelihood Score (ccLS) had sensitivity of 75.6 %, specificity of 93.5%, PPV of 88.6 %, NPV of 85.3 % and accuracy of 86.4 % in diagnosing ccRCC using a ccLS threshold of 4 and 5. As regard the assessment of ccLS threshold of 1 or 2 in excluding ccRCC pathological subtype, our study found that out of 29 patients with ccLS 1 or 2, there was only 1 ccRCC case with false result (3% false positive). It was also noted that there is significant relation between Arterial-to-delayed-enhancement-ratio (ADER) value and the grade of the ccRCC. The median interquartile range (IQR) of ADER parameter was statistically significant higher in grade II compared to grade I (Median was 1.6 and 0.9 respectively) and much higher in grade III compared to grades I and II (Median was 2.9) with P value < 0.001.
Conclusion: This ccLS showed promising efficacy in prediction and exclusion of ccRCC subtype. Moreover, it aids in predicting the ccRCC grade.
Limitations: Few cases of fat poor angiomyolipoma was interpreted as a high risk for being clear cell carcinoma.
Funding for this study: Self funding.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Mansoura University Research Ethics Committee

Author Block: E. Jambon, A. Crombé, G. Margue, N. Grenier, F. H. Cornelis, J-C. Bernhard; Bordeaux/FR
Purpose: Purpose: Small renal solid tumors (SRST) are increasingly being detected incidentally, posing diagnostic challenges. Up to 20% of SRST result in non-contributive biopsies. The aim was to assess the diagnostic accuracy of multiparametric MRI (mpMRI) in differentiating malignant from benign SRST in patients with suspected renal malignancy without no evident signs of malignancy.
Methods or Background: Methods or Background: This is a prospective multicentric French study. A cohort of 387 patients in 18 centers with non-hereditary, SRST between 1.5 and 4 cm in diameter was enrolled between November 2018 and May 2022. mpMRI protocols included T1w, T2w, diffusion-weighted imaging, and dynamic contrast-enhanced sequences. Two radiologists performed blinded readings with a centralized review in case of discordance. The primary endpoint is the negative predictive value (NPV) of a dichotomized Likert scale score, targeting a 98% NPV.
Results or Findings: Results or Findings: Among 287 patients analyzed for the primary endpoint, mpMRI achieved high sensitivity (97.4%) and PPV (83.2%) in detecting malignant tumors, but specificity (9.8%) and NPV (45.5%) remained low, limiting its ability to rule out malignancy. Likert scores varied significantly across histological subtypes, with oncocytomas and chromophobe RCCs often misclassified as malignant. In clear cell RCC (n = 121), higher Fuhrman grades were significantly associated with extensive diffusion restriction (p < 0.001) and T2 hypointensity (p = 0.001). The median Likert score increased with tumor grade (p = 0.004), suggesting that certain MRI features may reflect histological aggressiveness in this subtype.
Conclusion: Conclusion: MpMRI remains limited in its ability to distinguish benign from malignant tumors based on the predefined criteria used in this study. However, it may offer a noninvasive means of assessing tumor grade in clear cell renal cell carcinoma.
Limitations: Predefined criteria
Funding for this study: National funding (PHRC)
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The trial protocol (ClinicalTrials.gov identifier: NCT03470285) was approved by the CPP Sud-Méditerranée 5 (Committee for the Protection of Persons – French equivalent of an Ethical Committee) and the authorized by the authorisarion from ANSM (the French National Agency for the Safety of Health Products).

Author Block: J. Liu, J. Zhao; Lanzhou/CN
Purpose: To examine the imaging features and clinicopathologic features of TFE3-rearranged renal cell carcinoma and highlight key imaging findings for clinical decision-making.
Methods or Background: The imaging and clinicopathological data of 49 pathologically confirmed cases of TFE3-rearranged renal cell carcinoma from multiple institutions between January 2018 and December 2024 were retrospectively analyzed.
Results or Findings: Of the 49 patients, 17 were male and 32 were female; aged 44 ± 15 years (range: 10-82 years) , 22 were asymptomatic and 27 were symptomatic. Tumor size averaged 7.0 (range: 2.0–14.9) cm. Imaging showed 28 round-like and 21 lobulated lesions; 18 had distinct margins, while 31 had indistinct margins. A pseudocapsule was present in 36 cases, 24 of which were incomplete. Additionally, 42 tumors showed uneven density with necrosis or cystic degeneration; there were 25 cases of hemorrhage, 27 cases of irregular calcification, Peritumoral vessels were noted in 23 cases. Imaging modalities included CT (44 cases: 41 with both plain and enhanced scans, 3 with plain only), MRI (13 cases with plain and enhanced scans), and ultrasonography (15 cases, including 5 with contrast enhancement). Treatment included partial nephrectomy (21 cases), total nephrectomy (22 cases) , and biopsy (6 cases).With a median follow-up of 30 months (range: 1–72 months) until December 2024, 6 patients experienced tumor recurrence, 14 developed metastasis, and 7 died, while 5 patients were lost to follow-up.
Conclusion: TFE3-rearranged renal cell carcinoma is a rare renal tumor with distinct imaging and clinical features that can inform preoperative diagnosis, treatment, and prognosis.
Limitations: This is a retrospective single-center study whose results have not been validated in an independent cohort.
Funding for this study: No.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: J. Wei, X. Zhang, J. Liu, J. Zhou; Lanzhou/CN
Purpose: To investigate the value of clinical-radiologic features and whole-tumor histogram analysis derived from non-contrast CT in differentiating fat-poor angiomyolipoma (fp-AML) from chromophobe renal cell carcinoma (ChRCC).
Methods or Background: This retrospective study enrolled 89 patients with pathologically confirmed fp-AML (n=36) or ChRCC (n=53). Clinical, imaging, and pathological data were collected. Demographic data and conventional CT features (sex, age, symptoms, location, shape, margin, attenuation on non-contrast CT, tumor volume) were recorded. Using FireVoxel software, volumetric regions of interest were drawn on axial non-contrast CT images to extract histogram parameters (minimum, maximum, mean, standard deviation, variance, coefficient of variation, skewness, kurtosis, entropy, and percentiles). Statistical analyses included the Chi-square test, Mann-Whitney U test, and independent samples t-test. Multivariable logistic regression was used to identify independent predictors for differentiation and to construct clinical-radiologic and histogram models. Receiver operating characteristic (ROC) curve analysis was performed to evaluate diagnostic performance.
Results or Findings: Multivariable analysis identified non-contrast CT attenuation (OR=1.178, 95% CI: 1.087-1.277, P<0.001), tumor volume (OR=0.952, 95% CI: 0.928-0.976, P<0.001), mean attenuation (OR=0.764, 95% CI: 0.589-0.946, P=0.015), skewness (OR=0.135, 95% CI: 0.046-0.396, P<0.001), and the Perc.75 (OR=1.522, 95% CI: 1.169-1.983, P=0.002) as independent predictors for differentiating fp-AML from ChRCC. Both models demonstrated good performance. The clinical-radiologic model achieved an AUC of 0.905 (95% CI: 0.843-0.967), with a sensitivity of 77.8% and specificity of 90.6%. The histogram model achieved an AUC of 0.964 (95% CI: 0.932-0.996), with a sensitivity of 88.9% and specificity of 75.5%.
Conclusion: Clinical-radiologic features and whole-tumor histogram analysis based on non-contrast CT can help noninvasively differentiate fp-AML from ChRCC preoperatively, providing valuable guidance for clinical management.
Limitations: This study is limited by its retrospective, single-center design, and the findings have not been validated in an independent cohort.
Funding for this study: Natural Science Foundation of Gansu Province (22JR11RA060)
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Medical Ethics Committee of the Second Hospital of Lanzhou University (approval number: 2025A-585) and informed consent was waived.

Author Block: J. Wei, X. Zhang, J. Liu, J. Zhou; Lanzhou/CN
Purpose: To investigate the association of multiphase computed tomography(CT) features, and R.E.N.A.L. score with Leibovich risk stratification in localized renal cell carcinoma (RCC).
Methods or Background: This retrospective study analyzed 265 pathologically confirmed RCC patients. Clinical data, CT semantic features, and R.E.N.A.L. scores were collected. Patients were classified into low-risk (n=124) and intermediate-high-risk (n=141) groups per Leibovich criteria. Clinical characteristics included symptoms, albumin level, neutrophil-to-lymphocyte ratio (NLR), systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), and tumor extracellular volume fraction (ECV). CT semantic features included maximum tumor diameter, location, shape, margin, cystic change/necrosis, calcification, enhancement homogeneity, perinephric neovascularity, and R.E.N.A.L. score. Statistical analyses involved χ² tests, t-tests, Mann-Whitney U tests. Univariate and multivariate logistic regression analyses were employed to identify independent predictive factors for Leibovich risk stratification.
Results or Findings: Patients in the intermediate-high-risk group were more likely to have clinical symptoms, higher SII, and lower albumin levels compared to the low-risk group (P < 0.05). Tumor shape, margin, cystic change/necrosis, calcification, enhancement homogeneity, and perinephric neovascularity significantly differed between the two groups (P < 0.05). The maximum tumor diameter was larger in the intermediate-high-risk group, while the parenchymal phase enhancement ratio was lower (P < 0.05). Perinephric neovascularity was more frequent in the intermediate-high-risk group (P < 0.05). The R.E.N.A.L. score significantly differed between groups, R.E.N.A.L. score complexity was significantly higher in the intermediate-high-risk group (P<0.05). Multivariate analysis identified cystic change/necrosis, perinephric neovascularity, and R.E.N.A.L. score as independent predictors of intermediate-high Leibovich risk stratification in RCC.
Conclusion: Tumor cystic/necrotic change, perinephric neovascularity, and R.E.N.A.L. score are independent preoperative biomarkers for predicting intermediate-high Leibovich risk in RCC.
Limitations: This study is limited by its retrospective, single-center design and by the lack of xternal validation
Funding for this study: Natural Science Foundation of Gansu Province (22JR11RA060)
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Medical Ethics Committee of the Second Hospital of Lanzhou University (approval number: 2025A-585) and informed consent was waived.

Author Block: G. Imperiale, G. Brembilla, C. Rodolfi, F. Cei, A. Larcher, G. Rosiello, U. Capitanio, F. De Cobelli; Milan/IT
Purpose: Systematically evaluate the concordance of enhancement patterns between MRI and CT in small renal masses. This comparison aims to support the feasibility and broader application of the Clear Cell Likelihood Score (CCLS) in clinical contexts where contrast-enhanced MRI may be impractical, contraindicated or where optimization of healthcare resources is desirable.
Methods or Background: Patients with both preoperative CT and MRI were included, with histopathology serving as the reference standard. Two radiologists with intermediate experience, blinded to histology, independently assessed the enhancement patterns on MRI and CT; evaluated parameters included vascularity (hyper-, iso-, hypovascular), presence of segmental enhancement inversion (SEI) and arterial-delayed enhancement ratio (ADER >1.5). Lastly, CCLS was assessed both on MRI alone and on CT-derived enhancement pattern combined with MRI features.
Results or Findings: The cohort consisted of 47 patients, with 20 clear cell RCCs (42.6%), 12 papillary RCCs (25.5%), 9 oncocytomas (19.1%), 4 chromophobe RCCs (8.5%) and 2 fat-poor angiomyolipomas (4.3%).
Enhancement type was concordant in 41 cases (87.2%). Concordance for SEI was observed in 42 cases (89%) and for ADER in 44 cases (94%). Overall, the CCLS was consistent in 85% of the cohort between MRI only and CT-MRI combined.
Conclusion: MRI and CT showed a high level of concordance in the characterization of enhancement patterns in small renal masses, supporting the potential role of CT as a complementary tool when MRI with contrast is not feasible. Importantly, as CT is usually performed before MRI in clinical practice, sparing contrast in MRI could be reasonable, with CT-MRI combined CCLS representing a viable option.
Limitations: This study is limited by its single-center design, small cohort and histotype imbalance favoring clear cell and papillary RCCs. Findings should be considered preliminary, with validation in larger, multicenter populations warranted.
Funding for this study: This study received no external funding.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The study was approved by the Ethics Committee.

Author Block: M. Liu, S-C. Lin, Y-T. Lin, W-H. Chen, Y-J. Liu, C-H. Liao; Taichung/TW
Purpose: This study aimed to develop an automated radiomics-based machine learning framework using multiple MRI weightings to distinguish clear cell renal cell carcinoma (ccRCC) from non-clear cell renal cell carcinoma (ncRCC). Accurate preoperative identification of RCC subtypes is critical for selecting optimal surgical strategies and guiding targeted therapies, especially for patients unable to undergo contrast-enhanced imaging.
Methods or Background: RCC accounts for 85% of malignant kidney tumors, with ccRCC comprising ~75% of cases. MRI provides diverse weighted sequences (T2WI, T1WI, CE-T1WI, In/Out phase, DWI/ADC) useful for RCC differentiation. Radiomics enables extraction of thousands of quantitative features from tumor ROIs, which can be processed with machine learning. This study employed AutoML via TPOT to identify optimal feature sets and classifiers. MRI data from 76 patients (49 ccRCC, 27 ncRCC) between 2014–2023 were analyzed. Tumor ROIs were segmented in 2D, 3D, and combined formats; features were extracted and reduced through feature selection, followed by AutoML training with 10-fold cross-validation.
Results or Findings: Among classifiers, LinearSVC and SGD achieved the best performance. In 3D analyses, AUCs reached 0.81–0.86 across image sets. Importantly, models trained without contrast-enhanced images (C−) still attained robust accuracy (AUC up to 0.81–0.86 when combining 2D and 3D data). This demonstrates that non-contrast MRI radiomics can yield reliable subtype classification, potentially avoiding gadolinium-related risks.
Conclusion: The AutoML-based radiomics approach successfully differentiated ccRCC from ncRCC using multiparametric MRI, achieving high performance even without CE-T1WI. This method offers a clinically valuable tool for preoperative planning, particularly in patients with renal dysfunction or contrast contraindications.
Limitations: The study’s retrospective design and modest cohort size, especially the smaller ncRCC subgroup, may affect generalizability. Larger multicenter datasets and prospective validation are needed to confirm robustness and clinical applicability.
Funding for this study: There is no funding for this study
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Institutional Review Board (IRB) of Taichung Veterans General Hospital (TCVGH)

Author Block: N. Giordano¹, F. Urraro², V. Patanè¹, M. Piscopo¹, A. Russo¹, S. Cappabianca¹, A. Reginelli¹; ¹Naples/IT, ²Roma/IT
Purpose: To assess the diagnostic accuracy of contrast-enhanced ultrasound (CEUS) in differentiating renal cell carcinoma (RCC) from angiomyolipoma (AML) in small (≤3 cm), hyperechoic renal masses and to identify predictive enhancement patterns in routine clinical practice.
Methods or Background: This retrospective single-center study included 104 patients with incidentally detected hyperechoic renal lesions ≤3 cm examined by CEUS between December 2021 and July 2024. Two blinded radiologists independently assessed wash-in/wash-out dynamics, peak intensity, homogeneity, and perilesional rim-like enhancement. Histopathology was the reference standard when available; lesions stable for ≥18 months on follow-up were considered benign. Diagnostic metrics were calculated, and multivariate logistic regression identified independent predictors. Interobserver agreement was assessed by intraclass correlation coefficient (ICC)
Results or Findings: Of 104 lesions, 80 were managed as AMLs, with 4 biopsied during follow-up (2 papillary RCCs, 2 AMLs). Twenty-eight lesions underwent biopsy, confirming 26 RCCs (53% papillary, 32% chromophobe, 15% clear cell) and 2 atypical AMLs. RCCs typically demonstrated heterogeneous hyperenhancement with rapid wash-out and perilesional rim-like enhancement, whereas AMLs showed iso- or homogeneous enhancement with delayed wash-out. Rapid wash-out achieved sensitivity of 87% and specificity of 94% (AUC 0.91), while perilesional rim-like enhancement reached specificity of 98% and PPV of 92%. Combined CEUS features yielded an AUC of 0.93. Interobserver agreement was good (ICC 0.75–0.9).
Conclusion: CEUS provides excellent diagnostic accuracy in distinguishing RCC from AML in small hyperechoic renal masses. Rapid wash-out and perilesional rim-like enhancement are independent predictors of malignancy, supporting CEUS as a reliable, radiation-free tool to guide clinical decision-making
Limitations: This was a retrospective, single-center study with limited sample size. Histological confirmation was not available for all AMLs, relying on imaging stability as a surrogate of benignity. Larger prospective multicenter studies are required to validate these findings.
Funding for this study: No external funding was received for this study
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approved by the local Ethics Committee of the University Hospital ‘Luigi Vanvitelli’, Naples, Italy (Protocol 158/i/2022)

Author Block: S. Li, Z. Li; Wuhan/CN
Purpose: Occult pathological T3a (pT3a) upstaging in cT1b–T2a clear cell renal cell carcinoma (ccRCC) correlated with poor prognosis and necessitates modifications in surgical planning. However, predicting it preoperatively remains challenging. Conventional imaging emphasizes intratumoral features, yet early invasion frequently manifests as subtle alterations in peritumoral tissue. This study aimed to evaluate the diagnostic value of peritumoral tissue and to test whether combining peritumoral and intratumoral radiomics improves prediction of occult pT3a.
Methods or Background: This multicenter retrospective analysis included 1240 patients initially staged as cT1b–T2a ccRCC. Tumor and peritumoral regions were segmented using nnU-Net, and 107 radiomic features were extracted. Logistic regression, support vector machine, decision tree, and random forest classifiers were trained using tumor-only, peritumor-only, and combined features. Model performance was assessed with AUC, sensitivity, specificity, and accuracy across internal and external cohorts, with DeLong tests for AUC comparisons. SHAP analysis was used to evaluate feature contributions and tumor–peritumor interactions.
Results or Findings: In the internal validation cohort, the Radiomics_Combined model achieved an AUC of 0.800, significant outperforming the Radiomics_peritumor model (AUC 0.709, p = 0.020) and showing a nonsignificant improvement over the Radiomics_tumor model (AUC 0.787, p = 0.798). Similar trends were observed in external validation cohorts. SHAP analysis identified 30 peritumoral features among the top 40 predictors, with peritumor_shape_Maximum3DDiameter ranked highest. SHAP dependence plots further demonstrated nonlinear tumor–peritumor interactions, indicating that the model captures cross-regional effects reflecting subtle morphologic changes associated with local invasion.
Conclusion: Peritumoral tissue contributes distinct and complementary signatures of early invasion. Integrating tumor and peritumoral features enables accurate, interpretable prediction of occult pT3a upstaging in cT1b–T2a ccRCC and may improve surgical dicision-making.
Limitations: Retrospective design.
Funding for this study: This study was supported by grants from the National Natural Science Foundation of China (No. 82371942).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The ethical approvals for this multi-center retrospective study were provided by the Institutional Review Board of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology; the Institutional Review Board of the First Affiliated Hospital of Nanjing Medical University; and the Institutional Review Board of Xiangyang Central Hospital.