bp-VI-RADS combined with normalized apparent diffusion coefficient enhances diagnostic efficacy for muscle-invasive bladder cancer

Introduction

Bladder cancer is one of the most common malignant tumors of the urinary system. In 2022, approximately 614,000 new cases of bladder cancer were diagnosed worldwide, making it the ninth most prevalent cancer globally, and approximately 220,000 bladder cancer-related deaths were reported worldwide, making it the eleventh leading cause of cancer-related mortality globally (1). Based on tumor infiltration into the muscular layer, bladder cancer is primarily classified as non-muscle-invasive bladder cancer (NMIBC) or muscle-invasive bladder cancer (MIBC). NMIBC can be treated with localized therapies such as the transurethral resection of bladder tumor (TURBT) and intravesical instillation, resulting in a favorable prognosis. Conversely, MIBC accounts for approximately 25% of cases, is aggressive, has a poor prognosis, and requires radical cystectomy (RC) and chemotherapy (2,3). Although TURBT serves as the gold standard for the local staging of bladder tumors, its limitations, such as specimen cauterization or the absence of the muscle layer, lead to understaging in approximately 25% of MIBC cases, delaying appropriate therapeutic interventions (4,5).

Due to its excellent soft-tissue resolution, magnetic resonance imaging (MRI) has become a crucial tool for the local staging of bladder tumors since the establishment of the multiparametric magnetic resonance imaging (mp-MRI)-based Vesical Imaging-Reporting and Data System (VI-RADS) in 2018 (6,7). The mp-MRI protocol includes T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced (DCE) imaging. However, intravenous contrast agent administration poses limitations for patients with renal insufficiency or allergies. Biparametric magnetic resonance imaging (bp-MRI), which comprises only T2WI and DWI, reduces economic costs, shortens scan times, enhances patient comfort, and avoids contrast-related issues. Multiple studies have demonstrated that bp-MRI has a diagnostic accuracy comparable to that of mp-MRI in detecting MIBC (8-10).

The quantification of the apparent diffusion coefficient (ADC), which reflects the tissue microstructure (e.g., cellular density and membrane integrity), could be used to assess tumor aggressiveness in bladder cancer (11). ADC maps can reveal areas of restricted diffusion, and in bladder malignancies, ADC values are associated with tumor grade and a higher T stage (12,13). Research has shown that the reliability of the ADC is influenced by inconsistencies in MRI acquisition parameters, such as b values (e.g., 0–2000 s/mm²) and magnetic field strengths (1.5T vs. 3.0T), making ADC values incomparable across studies and limiting their application within the VI-RADS framework (11,14). However, multiple studies have shown that the normalized apparent diffusion coefficient (nADC), which normalizes tumor ADC values against ADC measurements from normal tissues, addresses the effect of acquisition parameters on ADC measurements (15-18). The current VI-RADS employs a five-category assessment based on the likelihood of muscle invasion. Under the VI-RADS, scores of 1 and 2 strongly suggest NMIBC, and scores of 4 and 5 strongly suggest MIBC; however, a score of 3 represents an indeterminate category for which appropriate treatment recommendations cannot be provided (7,19,20). This diagnostic uncertainty remains unresolved in both the mp-MRI and bp-MRI approaches. Further, the VI-RADS relies primarily on tumor morphology for assessment and lacks quantitative metrics. Integrating quantitative indicators into the VI-RADS could significantly enhance diagnostic performance and thus holds substantial clinical significance. Therefore, this study aimed to evaluate the utility of nADC in bp-MRI for diagnosing muscle-invasive bladder cancer, particularly in VI-RADS lesions with a score of 3. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1877/rc).

Methods

Ethical approval

This retrospective study was approved by the Ethics Committee of Nanjing Drum Tower Hospital of Nanjing University (No. 2022-452-02). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The Institutional Ethics Committee waived the requirement of informed consent, as this was a retrospective study that evaluated anonymized data.

Patients

Patients with clinically suspected bladder cancer based on imaging findings, prior to pathological confirmation, were included in the study. The data of 265 patients with clinically suspected bladder cancer who underwent MRI between January 2021 and March 2023 were retrospectively evaluated. The exclusion criteria were as follows: (I) pathologically confirmed non-urothelial tumors or inflammatory changes; (II) prior surgery or neoadjuvant therapy before MRI examination; (III) a lack of pathological reports (no biopsy or surgery); (IV) inconclusive MRI findings (e.g., poor bladder filling, motion artifacts, or metal artifacts); and/or (V) indeterminate T staging due to cauterization or an absence of muscle layers. Ultimately, 172 patients were included in the study (Figure 1).

Figure 1 Flowchart showing patient selection for the study. MIBC, muscle-invasive bladder cancer; MRI, magnetic resonance imaging; NMIBC, non-muscle-invasive bladder cancer; RC, radical cystectomy; TURBT, transurethral resection of bladder tumor.

MRI protocol

All the bladder MRI data were acquired using 3.0T MRI scanners (uMR 770, United Imaging Healthcare, Shanghai, China; Ingenia, Philips Healthcare, Best, The Netherlands) with a 32-channel body coil (in vivo). The data for each patient included axial T2-weighted turbo spin-echo images, and axial diffusion-weighted spin-echo echo-planar images (Table S1). ADC maps were automatically generated and uploaded using post-processing software.

VI-RADS assessment

All the images, including the axial, sagittal, and coronal T2WI and DWI sequences, were evaluated. All the images were independently reviewed by two board-certified abdominal radiologists (A and B), specializing in genitourinary imaging, each with over five years of experience in the MRI-based diagnosis of bladder tumors, and proficiency in the bp-MRI scoring system. The reviewers were blinded to the pathological results and prior bp-MRI reports, but had access to routine clinical data preceding surgery (aligning with real-world clinical workflows). For patients with multiple tumors, the tumor with the highest VI-RADS score was selected for analysis. If the initial VI-RADS scores differed between the two reviewers, the final score was determined by a third reviewer (C) with over 10 years of experience. Similarly, the two radiologists (A and B), who were blinded to the clinical outcomes, independently reviewed the mp-MRI scans of the subset of patients that underwent mp-MRI (n=69), after excluding the DCE sequence. A third senior radiologist (C) adjudicated cases with discrepant scores.

ADC measurements

On the ADC maps, genitourinary radiologists (B and C) manually delineated regions of interest (ROIs) using the polygon tool within the three-dimensional Slicer medical imaging platform (accessible at: https://www.slicer.org/), focusing on visually identified areas with minimal ADC values. ADC measurements were performed in a double-blinded manner: operators were unaware of the clinical staging and histological grading of the tumors. To ensure ADC measurement accuracy, the ROIs excluded necrotic regions, microvessels, and artifacts, and were drawn at the maximal cross-sectional level of the tumor. ROIs for pedunculated tumors were delineated to exclude the stalk to avoid interference from non-tumor tissues, such as fibrous connective tissue, capillaries, and edema, and to minimize inter-reader variability in ADC measurements (21). For multifocal tumors, the lesion with the highest VI-RADS score and visually lowest ADC value was measured. Three distinct ROIs (>20 mm² each) were placed at visually identified areas of minimal ADC in each tumor to derive the mean apparent diffusion coefficient (ADC_mean) value (Figure 2).

Figure 2 Examples of the ROI delineation for tumors, urine, and the obturator internus muscle in two patients. Color dots denote ROI delineation: tumor/urine (B,C) and obturator internus muscle (E,F). (A-C) Patient 1: axial T2-weighted image and ADC map of a broad-based lesion (2.7 cm × 2.5 cm) in the right posterior bladder wall. The inner wall signal is focally interrupted at the lesion base. bp-MRI score: VI-RADS 3. Measured ADC: 0.799×10⁻³ mm²/s; nADC_urine =0.237; nADC_oim =0.646. Pathology revealed MIBC. (D-F) Patient 2: axial T2-weighted image and ADC map of a pedunculated lesion (3.3 cm × 2.7 cm) in the left lateral bladder wall. The detrusor muscle appears uninterrupted beneath the stalk. bp-MRI score: VI-RADS 2. Measured ADC: 0.857×10⁻³ mm²/s; nADC_urine =0.281, nADC_oim =0.718. Pathology revealed NMIBC. ADC, apparent diffusion coefficient; bp-MRI, biparametric magnetic resonance imaging; MIBC, muscle-invasive bladder cancer; nADC_oim, obturator internus muscle normalized apparent diffusion coefficient; nADC_urine, urine-normalized apparent diffusion coefficient; NMIBC, non-muscle-invasive bladder cancer; ROI, region of interest; VI-RADS, Vesical Imaging-Reporting and Data System.

To calculate the nADC, the ADC values were obtained for the bladder urine and bilateral obturator internus muscles (OIMs). To avoid partial volume effects from bowel motion, urine ROIs were positioned centrally in homogeneous areas of the bladder lumen, while muscle ROIs were placed centrally in homogeneous regions of the obturator internus at the level of the femoral necks (Figure 2). The following formula was used to calculate the nADC:nADC = ADC_mean (tumor)/ADC (reference), where the reference was either urine or the OIM.

Statistical analysis

The statistical analyses were performed using SPSS (version 25.0; IBM Corp., USA) and R (version 4.2.2; R Foundation, Austria). Statistical significance was defined as a two-tailed P<0.05. The Shapiro-Wilk test was used to assess the normality of the quantitative variables. For the normally distributed variables, the independent t-test was used; for the non-normally distributed variables, the Wilcoxon rank-sum test was used. For the categorical variables, Fisher’s exact test or the Chi-squared test was used as appropriate. Binary logistic regression was used to model muscle invasion status via the bp-MRI VI-RADS combined with the urine-normalized apparent diffusion coefficient (nADC_urine), obturator internus muscle normalized apparent diffusion coefficient (nADC_oim), or ADC_mean. Receiver operating characteristic (ROC) curves were used to evaluate diagnostic performance, and DeLong tests were used for comparisons. Optimal diagnostic thresholds (Youden indices) were applied to the VI-RADS 3 subgroup. The interobserver agreement for the ROIs used intraclass correlation coefficients (ICCs); the VI-RADS agreement used weighted kappa.

Results

In the final patient cohort (n=172), all patients underwent TURBT or RC within 3 weeks of MRI examination. The specimens were evaluated by pathologists, who had access to each patient’s basic clinical information but remained blinded to their bp-MRI and the VI-RADS scoring results (to reflect composite clinical workflows). The patients had a median age of 69 years (interquartile range, 62–78 years), and 34 (19.8%) were female and 138 (80.2%) were male. Pathology confirmed 118 (68.6%) cases of NMIBC and 54 (31.4%) cases of MIBC. TURBT was performed in 110 (64.0%) patients, and RC was performed in 62 (36.0%) patients (Table 1). The weighted kappa value for the bp-MRI VI-RADS score between the two reviewers was 0.905, indicating almost perfect agreement. Interobserver consistency was high for the ADC measurements across tumor ROIs 1–3 (ICC, 0.835–0.864), urine (ICC, 0.856), and the OIM (ICC, 0.798).

Diagnostic performance of the bp-MRI VI-RADS

The statistical analysis revealed the following distribution of VI-RADS scores: 1 (4.1%, 7/172), 2 (55.2%, 95/172), 3 (15.1%, 26/172), 4 (19.2%, 33/172), and 5 (6.4%, 11/172). All the VI-RADS 1 tumors were NMIBC, and all the VI-RADS 5 tumors were MIBC. Muscle invasion occurred in 9 of 95 VI-RADS lesions with a score of 2, 8 of 26 VI-RADS lesions with a score of 3, and 26 of 33 VI-RADS lesions with a score of 4 (Table 1). Individual VI-RADS scores from Readers A and B were recorded, with a third reader (Reader C) adjudicating 13 cases due to disagreements (Tables S2,S3).

Comparisons of the bp-MRI VI-RADS, ADC and nADC between MIBC and NMIBC

The patients with MIBC had a significantly lower median ADC value than the patients with NMIBC (0.731×10⁻³ vs. 0.952×10⁻³ mm²/s). Similarly, the MIBC patients had significantly lower median nADC_urine (0.23 vs. 0.30) and median nADC_oim (0.58 vs. 0.77) values than the NMIBC patients. In terms of the imaging scores, the MIBC patients had a significantly higher median VI-RADS score than the NMIBC patients (4.00 vs. 2.00). All these between-group differences were P<0.001 (Table 2).

Spearman’s correlation test was used to assess the relationship between the ADC values/nADC values and MIBC. The nADC_urine had the largest absolute Spearman’s correlation coefficient with MIBC (rho =−0.51, P<0.001), followed by the nADC_oim (rho =−0.49, P<0.001) and the ADC value (rho =−0.48, P<0.001). A very strong correlation was found between the ADC values and the nADC metrics (rho =0.95–0.97, P<0.001). No statistically significant correlations were found between the tumor ADC/nADC and reference tissues (urine and OIM) (rho ≤0.18, P<0.05). The nADC (nADC_urine/nADC_oim) showed significant but weak negative correlations with the urine and OIMs (rho =−0.24 to −0.34, P<0.001). These results are shown in Figure 3.

Figure 3 Correlation analysis of the ADC and nADC with MIBC and reference tissues. *, P<0.05 (statistically significant correlation); **, P<0.01 (highly statistically significant correlation); ***, P<0.001 (extremely statistically significant correlation). ADC, apparent diffusion coefficient; ADC_mean, mean ADC; ADC_oim, obturator internus muscle ADC; ADC_urine, urine ADC; MIBC, muscle-invasive bladder cancer; nADC, normalized ADC; nADC_oim, obturator internus muscle nADC; nADC_urine, urine nADC; rho, Spearman’s rank correlation coefficient.

Model construction and validation

We first evaluated baseline diagnostic models, including the bp-MRI VI-RADS and three ADC-based parameters. As an imaging reference standard, the bp-MRI VI-RADS showed good diagnostic performance for MIBC at the optimal diagnostic threshold (Youden index, VI-RADS ≥4): area under the curve (AUC) 0.87, accuracy 0.86, sensitivity 0.69, and specificity 0.94. Thus, this model can help clinicians to effectively detect MIBC. Among the ADC metrics, the nADC_urine showed the best performance with an AUC of 0.82 and an accuracy of 0.78. At the cut-off value of 0.259, the sensitivity was 0.78 and the specificity was 0.80. Conversely, the ADC_mean (AUC, 0.80; cut-off value, 0.822×10⁻³ mm²/s) and nADC_oim (AUC, 0.80; cut-off value, 0.694) had slightly lower accuracies of 0.75 and 0.77, respectively. The ADC_mean achieved a sensitivity of 0.77 and a specificity of 0.70, while the nADC_oim achieved a sensitivity of 0.75 and a specificity of 0.81 (Figure 4 and Table 3). These results indicate that standalone ADC parameters have moderate diagnostic utility.

Figure 4 ROC curve analysis of the VI-RADS and ADC-based composite models for bladder cancer staging. Model A, Model combining the ADC_mean and bp-MRI VI-RADS; Model O, Model combining the nADC_oim and bp-MRI VI-RADS; Model U, Model combining the nADC_urine and bp-MRI VI-RADS. ADC, apparent diffusion coefficient; ADC_mean, mean apparent diffusion coefficient; AUC, area under the curve; bp-MRI, biparametric magnetic resonance imaging; nADC_oim, obturator internus muscle normalized apparent diffusion coefficient; nADC_urine, urine-normalized apparent diffusion coefficient; ROC, receiver operating characteristic; VI-RADS, Vesical Imaging-Reporting and Data System.

To increase model diagnostic accuracy, three composite models were constructed by binomial logistic regression analysis. These models consistently outperformed individual methods, demonstrating higher AUC and accuracy values. The results were as follows: Model A (ADC_mean + VI-RADS): AUC: 0.91, accuracy: 0.82, sensitivity: 0.81, specificity: 0.83; Model U (nADC_urine + VI-RADS): AUC: 0.93, accuracy: 0.90, sensitivity: 0.92, and specificity: 0.85 (this model had the highest AUC among all the models, as well as high sensitivity and specificity); Model O (nADC_oim + VI-RADS): AUC: 0.92, accuracy: 0.88, sensitivity: 0.91, specificity: 0.81 (this model performed well but was slightly inferior to Model U). Overall, the composite models significantly surpassed single parameters, with all AUCs exceeding 0.90 (Table 3).

Significant differences were observed between the VI-RADS model and all composite models (P<0.01), confirming that integrating ADC metrics with the VI-RADS substantially improved model diagnostic accuracy for bladder cancer staging. Among the composite models, Model U showed the best performance (AUC 0.93, sensitivity/specificity ≥0.90), followed by Model O. Model A exhibited the lowest performance but still had an AUC >0.90. Model U showed statistically significant advantages over Model O (P=0.046) and Model A (P<0.01). The difference between Model O and Model A approached statistical significance (P=0.056) (Table 4).

Subgroup stratification was performed for the VI-RADS cases with a score of 3 (n=26) based on the optimal diagnostic thresholds (the Youden index) from the ROC curve analysis. The results were as follows: nADC_urine =0.259, nADC_oim =0.694, and ADC_mean =0.822×10^–3 mm²/s (Table 3). The following subgroups were defined: VI-RADS 3U (nADC_urine), VI-RADS 3O (nADC_oim), and VI-RADS 3A (ADC_mean). Among the 26 patients with a VI-RADS score of 3, the classification outcomes for MIBC versus NMIBC using the aforementioned thresholds are summarized in Table 5. Notably, VI-RADS 3U demonstrated optimal performance, achieving 100% sensitivity and a specificity of 77.8%. All eight MIBC cases were classified correctly. Among the 18 NMIBC patients, 4 were classified as having MIBC (false positives). VI-RADS 3O showed slightly inferior performance: 1 MIBC case was misclassified as NMIBC, and 5 NMIBC cases were misclassified as MIBC, with both the sensitivity and specificity exceeding 70%. VI-RADS 3A exhibited the lowest sensitivity (50%) and specificity (66.7%), with only 4 MIBC cases correctly classified and 12 NMIBC cases correctly identified as true negatives.

Discussion

This study evaluated the efficacy of the bp-MRI-based VI-RADS in distinguishing between MIBC and NMIBC, with a focus on the diagnostic enhancement provided by nADC values as potential biomarkers of tumor aggressiveness.

Decreased ADC values reflect increased tumor malignancy, as increased cellular density and tighter cell packing at higher tumor grades significantly reduce the extracellular space. This structural change restricts free water diffusion in tissues, manifesting as hyperintensity on DWI and correspondingly lower ADC values calculated from DWI. Although ADC values are correlated with tumor aggressiveness (a lower ADC typically indicates greater invasiveness) and hold potential for preoperative assessment, technical variations across studies, such as differences in MRI scanner models and protocol parameters (e.g., b-value settings), have led to non-normalized measurements. This heterogeneity limits the utility of standalone ADC measurements as reliable preoperative predictors, necessitating normalization methods (e.g., reference tissue calibration) to improve reproducibility (12,13,21).

In our study, both the ADC values (0.731×10⁻³ mm²/s) and nADC metrics (nADC_urine, 0.23; nADC_oim, 0.58) were significantly lower in the MIBC group than the NMIBC group (ADC, 0.952×10⁻³ mm²/s; nADC_urine, 0.30; nADC_oim, 0.77). The composite model integrating the nADC_urine with the VI-RADS (Model U) achieved optimal performance (AUC 0.93), which is consistent with previous studies demonstrating that reduced nADC values are correlated with T stage and can serve as biomarkers of bladder cancer aggressiveness (15,17,22,23).

Wang et al. normalized tumor ADC values against those of the bladder urine, gluteus maximus, and OIMs, and reported that the nADC was superior to the raw ADC for histological grading, with the nADC_urine yielding the best diagnostic performance (22). Our results confirmed that combining the nADC with the VI-RADS improve the diagnostic accuracy over the ADC-VI-RADS models, which aligns with Liu et al.’s findings that combining the nADC_iliopsoas with the VI-RADS resulted in an AUC of 0.91 (23).

Critically, tumor nADC metrics are independent of MRI protocol parameters, enabling cross-study comparability of quantitative values (24,25). We evaluated two reference tissues, of which, the tumor nADC_urine showed superior performance. The ADC of the tumor was not significantly correlated with that of the reference tissue (urine/OIM; P<0.001; rho ≤0.11), while the ADC of the urine and OIM were moderately correlated (rho =0.33, P<0.001), supporting both as viable references.

Studies of other organs have proposed diverse reference tissues (e.g., myometrium, cervix, white matter, and normal peripheral/transition zones) (17,26,27). However, intrinsic variations in reference tissues, such as fluid composition, inflammation, tumor hemorrhage, or fibrosis, highlight the need for standardized reference selection to ensure consistency and reliability.

The VI-RADS has become a key clinical tool for preoperative staging. In our cohort of 172 patients, the bp-MRI VI-RADS showed excellent diagnostic performance at a threshold ≥4 for MIBC (AUC 0.87, sensitivity 0.69, specificity 0.94), ensuring high accuracy. For bladder cancer patients requiring frequent MRI surveillance, bp-MRI—despite omitting some parameters—reduces the scan time, economic burden, and contrast-induced renal risk while maintaining comparable quality to mp-MRI (8-10). A major challenge for the VI-RADS remains the subjectivity in interpreting borderline cases, particularly tumors with a VI-RADS score of 3. Multiple metrics (e.g., ADC values, contact length, and tumor morphology) have been proposed to address this issue (12,28,29).

In our subgroup analysis of lesions with a VI-RADS score of 3, the nADC_urine demonstrated superior diagnostic capability for detecting MIBC compared with the nADC_oim and ADC_mean (AUC, 0.896 vs. 0.854 vs. 0.778). Notably, the nADC_urine achieved 100% sensitivity, correctly identifying all MIBC cases. This highlights the potential of the nADC_urine as a quantitative parameter for resolving diagnostically challenging MIBC/NMIBC classifications. Further, the progressive change in the nADC values was correlated with increasing confidence in predicting tumor invasion depth. Composite Model U, which combines ADC metrics with the VI-RADS achieved optimal performance and outperformed Model O (P<0.01) and Model A (P=0.046), increasing the AUC from 0.87 (the VI-RADS alone) to 0.93 while maintaining high specificity and sensitivity (both ≥0.85). Integrating the quantitative nADC with other clinical and imaging features into future VI-RADS iterations could enable precise patient stratification and optimized treatment planning.

This study explored the quantitative threshold for MIBC diagnosis; however, as a single-center, retrospective study, its specific threshold value lacks immediate clinical applicability and external validation, particularly given the limited cohort of only 26 patients with a VI-RADS score of 3. Thus, it cannot yet be recommended for clinical decision-making. Second, while Readers B and C demonstrated good interobserver agreement in measuring the ADC_mean values for tumors and reference tissues through ROI delineation—suggesting objectivity in ADC quantification—this consistency may be influenced by reader experience and training. Variations may exist among readers with different skill levels, necessitating independent validation with additional readers. Further, the dataset was acquired from two different MRI scanners. Although nADC values help mitigate interscanner parameter variations, the disproportionate distribution of data between the two scanners might have introduced device selection bias. Further, due to the limitations of the TURBT procedure, such as insufficient resection depth or missed tumors, the number of patients with NMIBC might have been underestimated, which could affect the stability of the model evaluation. Thus, the conclusions warrant further validation through prospective, multicenter studies with larger cohorts.

Conclusions

In conclusion, our study demonstrates that the integration of nADC_urine significantly enhances the diagnostic performance of bp-MRI-based VI-RADS in preoperatively identifying MIBC. As a reproducible quantitative biomarker, nADC_urine effectively mitigates the technical heterogeneity inherent in multi-scanner protocols, offering superior reliability over raw ADC measurements. Specifically, the nADC_urine-augmented VI-RADS model provides high-precision stratification for indeterminate VI-RADS category 3 lesions, potentially reducing diagnostic uncertainty and optimizing surgical decision-making. While these findings underscore the clinical utility of quantitative bp-MRI, prospective multicenter trials are warranted to standardize these thresholds for routine clinical practice.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1877/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1877/dss

Funding: This study was supported by fundings for Clinical Trials from The Affiliated Drum Tower Hospital, Medical School of Nanjing University (No. 2022-LCYJ-PY-16), National Natural Science Foundation of China (Nos. 82473307 and 82172691), and Jiangsu Provincial Social Development Project (No. BE2023659).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1877/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This retrospective study was approved by the Ethics Committee of Nanjing Drum Tower Hospital of Nanjing University (No. 2022-452-02). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Since this was a retrospective study and anonymised data were evaluated, informed consent was waived by our Institutional Ethics Committee.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. van der Heijden AG, Bruins HM, Carrion A, Cathomas R, Compérat E, Dimitropoulos K, et al. European Association of Urology Guidelines on Muscle-invasive and Metastatic Bladder Cancer: Summary of the 2025 Guidelines. Eur Urol 2025;87:582-600. [Crossref] [PubMed]
3. Richters A, Aben KKH, Kiemeney LALM. The global burden of urinary bladder cancer: an update. World J Urol 2020;38:1895-904. [Crossref] [PubMed]
4. Kulkarni GS, Hakenberg OW, Gschwend JE, Thalmann G, Kassouf W, Kamat A, Zlotta A. An updated critical analysis of the treatment strategy for newly diagnosed high-grade T1 (previously T1G3) bladder cancer. Eur Urol 2010;57:60-70. [Crossref] [PubMed]
5. Pastuszak A, Zdrojowy R, Poletajew S, Adamowicz J, Krajewski W. Technical developments in transurethral resection of bladder tumours. Contemp Oncol (Pozn) 2019;23:195-201. [Crossref] [PubMed]
6. Panebianco V, Narumi Y, Altun E, Bochner BH, Efstathiou JA, Hafeez S, Huddart R, Kennish S, Lerner S, Montironi R, Muglia VF, Salomon G, Thomas S, Vargas HA, Witjes JA, Takeuchi M, Barentsz J, Catto JWF. Multiparametric Magnetic Resonance Imaging for Bladder Cancer: Development of VI-RADS (Vesical Imaging-Reporting And Data System). Eur Urol 2018;74:294-306. [Crossref] [PubMed]
7. Woo S, Panebianco V, Narumi Y, Del Giudice F, Muglia VF, Takeuchi M, Ghafoor S, Bochner BH, Goh AC, Hricak H, Catto JWF, Vargas HA. Diagnostic Performance of Vesical Imaging Reporting and Data System for the Prediction of Muscle-invasive Bladder Cancer: A Systematic Review and Meta-analysis. Eur Urol Oncol 2020;3:306-15. [Crossref] [PubMed]
8. Delli Pizzi A, Mastrodicasa D, Marchioni M, Primiceri G, Di Fabio F, Cianci R, Seccia B, Sessa B, Mincuzzi E, Romanelli M, Castellan P, Castellucci R, Colasante A, Schips L, Basilico R, Caulo M. Bladder cancer: do we need contrast injection for MRI assessment of muscle invasion? A prospective multi-reader VI-RADS approach. Eur Radiol 2021;31:3874-83. [Crossref] [PubMed]
9. Liu P, Cai L, Jiang L, Chen H, Cao Q, Bai K, Bai R, Wu Q, Yang X, Lu Q. Comparative diagnostic performance of VI-RADS based on biparametric and multiparametric MRI in predicting muscle invasion in bladder cancer. BMC Med Imaging 2025;25:60. [Crossref] [PubMed]
10. Aslan S, Cakir IM, Oguz U, Bekci T, Demirelli E. Comparison of the diagnostic accuracy and validity of biparametric MRI and multiparametric MRI-based VI-RADS scoring in bladder cancer; is contrast material really necessary in detecting muscle invasion? Abdom Radiol (NY) 2022;47:771-80. [Crossref] [PubMed]
11. Bollineni VR, Kramer G, Liu Y, Melidis C, deSouza NM. A literature review of the association between diffusion-weighted MRI derived apparent diffusion coefficient and tumour aggressiveness in pelvic cancer. Cancer Treat Rev 2015;41:496-502. [Crossref] [PubMed]
12. Zhang Z, Xiao W, Wang Y, Zhang W, Luo M. Exploring the potential of the combined diagnostic model of ADC value and bp-MRI VI-RADS in the evaluation of muscle invasion in bladder Cancer. Abdom Radiol (NY) 2025;50:3100-7. [Crossref] [PubMed]
13. Kobayashi S, Koga F, Kajino K, Yoshita S, Ishii C, Tanaka H, Saito K, Masuda H, Fujii Y, Yamada T, Kihara K. Apparent diffusion coefficient value reflects invasive and proliferative potential of bladder cancer. J Magn Reson Imaging 2014;39:172-8. [Crossref] [PubMed]
14. Wáng YXJ. An explanation for the triphasic dependency of apparent diffusion coefficient (ADC) on T2 relaxation time: the multiple T2 compartments model. Quant Imaging Med Surg 2025;15:3779-91. [Crossref] [PubMed]
15. Nguyen TB, Ushinsky A, Yang A, Nguyentat M, Fardin S, Uchio E, Lall C, Lee T, Houshyar R. Utility of quantitative apparent diffusion coefficient measurements and normalized apparent diffusion coefficient ratios in the diagnosis of clinically significant peripheral zone prostate cancer. Br J Radiol 2018;91:20180091. [Crossref] [PubMed]
16. Elson A, Bovi J, Siker M, Schultz C, Paulson E. Evaluation of absolute and normalized apparent diffusion coefficient (ADC) values within the post-operative T2/FLAIR volume as adverse prognostic indicators in glioblastoma. J Neurooncol 2015;122:549-58. [Crossref] [PubMed]
17. Lura N, Wagner-Larsen KS, Ryste S, Fasmer K, Forsse D, Trovik J, Halle MK, Bertelsen BI, Riemer F, Salvesen Ø, Woie K, Krakstad C, Haldorsen IS. Tumor ADC value predicts outcome and yields refined prognostication in uterine cervical cancer. Cancer Imaging 2025;25:23. [Crossref] [PubMed]
18. Bajgiran AM, Mirak SA, Sung K, Sisk AE, Reiter RE, Raman SS. Apparent Diffusion Coefficient (ADC) Ratio Versus Conventional ADC for Detecting Clinically Significant Prostate Cancer With 3-T MRI. AJR Am J Roentgenol 2019;213:W134-42. [Crossref] [PubMed]
19. Bicchetti M, Simone G, Giannarini G, Girometti R, Briganti A, Brunocilla E, Cardone G, De Cobelli F, Gaudiano C, Del Giudice F, Flammia S, Leonardo C, Pecoraro M, Schiavina R, Catalano C, Panebianco V. A novel pathway to detect muscle-invasive bladder cancer based on integrated clinical features and VI-RADS score on MRI: results of a prospective multicenter study. Radiol Med 2022;127:881-90. [Crossref] [PubMed]
20. Pecoraro M, Cipollari S, Messina E, Laschena L, Dehghanpour A, Borrelli A, Del Giudice F, Muglia VF, Vargas HA, Panebianco V. Multiparametric MRI for Bladder Cancer: A Practical Approach to the Clinical Application of VI-RADS. Radiology 2025;314:e233459. [Crossref] [PubMed]
21. Li H, Liu L, Ding L, Zhang Z, Zhang M. Quantitative Assessment of Bladder Cancer Reflects Grade and Recurrence: Comparing of Three Methods of Positioning Region of Interest for ADC Measurements at Diffusion-weighted MR Imaging. Acad Radiol 2019;26:1148-53. [Crossref] [PubMed]
22. Wang HJ, Pui MH, Guo Y, Li SR, Liu MJ, Guan J, Zhang XL, Feng Y. Value of normalized apparent diffusion coefficient for estimating histological grade of vesical urothelial carcinoma. Clin Radiol 2014;69:727-31. [Crossref] [PubMed]
23. Liu P, Cai L, Yu R, Cao Q, Bai K, Zhuang J, Wu Q, Li P, Yang X, Lu Q. Significance of Normalized Apparent Diffusion Coefficient in the Vesical Imaging-Reporting and Data System for Diagnosing Muscle-Invasive Bladder Cancer. J Magn Reson Imaging 2024;60:1639-47. [Crossref] [PubMed]
24. Woo S, Kim SY, Cho JY, Kim SH. Preoperative Evaluation of Prostate Cancer Aggressiveness: Using ADC and ADC Ratio in Determining Gleason Score. AJR Am J Roentgenol 2016;207:114-20. [Crossref] [PubMed]
25. Gladwish A, Milosevic M, Fyles A, Xie J, Halankar J, Metser U, Jiang H, Becker N, Levin W, Manchul L, Foltz W, Han K. Association of Apparent Diffusion Coefficient with Disease Recurrence in Patients with Locally Advanced Cervical Cancer Treated with Radical Chemotherapy and Radiation Therapy. Radiology 2016;279:158-66. [Crossref] [PubMed]
26. Shankar JJ, Bata A, Ritchie K, Hebb A, Walling S. Normalized Apparent Diffusion Coefficient in the Prognostication of Patients with Glioblastoma Multiforme. Can J Neurol Sci 2016;43:127-33. [Crossref] [PubMed]
27. Wang X, Hielscher T, Radtke JP, Görtz M, Schütz V, Kuder TA, Gnirs R, Schwab C, Stenzinger A, Hohenfellner M, Schlemmer HP, Bonekamp D. Comparison of single-scanner single-protocol quantitative ADC measurements to ADC ratios to detect clinically significant prostate cancer. Eur J Radiol 2021;136:109538. [Crossref] [PubMed]
28. Doğan MB, Gündüz N, Kazan HÖ, Çakıcı MÇ, Yıldırım A, Erdem G. Contribution of lesion shape features to the VI-RADS for predicting muscle invasion in bladder tumors. Eur J Radiol 2025;187:112104. [Crossref] [PubMed]
29. Pastorino GL, Mercinelli C, Necchi A. The role of MRI in muscle-invasive bladder cancer: an update from the last two years. Curr Opin Urol 2025;35:165-70. [Crossref] [PubMed]