Abstract

Introduction
Vesical Imaging-Reporting and Data System (VI-RADS) has been shown to have good diagnostic efficacy in differentiating muscular and non-muscular invasive bladder cancer^1,2. However, the original intention of VI-RADS and previous studies have only focused on the target population of patients with primary bladder cancer^1-3. Post-treatment patients, including those with recurrence, were not included in the evaluation. However, bladder cancer has a high risk of recurrence and progression^4,5, and recurrent patients are also faced with staging problems in deciding which surgery to have next or systemic treatment. Therefore, we expanded the target population by including patients after treatment into the cohort, and further evaluated the diagnostic performance of VI-RADS. The premise of using VI-RADS is that the bladder space occupying is bladder cancer. We excluded patients who had localized thickening of the bladder wall after surgery, as it is difficult to differentiate between inflammatory and neoplastic lesions^3,6. The remaining patients, including post-treatment patients, were included prospectively to assess the scope of VI-RADS application.
Materials and Methods
From January 2020 to October 2020, we prospectively performed preoperative multiparametric magnetic resonance imaging (mpMRI) on patients with suspected bladder cancer, including post-treatment patients. All data were acquired on a 3T MRI scanner (DiscoveryTM MR750, GE Healthcare, Milwaukee, WI). According to VI-RADS, T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced (DCE) imaging were obtained. The entire cohort was divided into the primary bladder cancer group and the post-treatment group, and the recurrence group was separated from the post-treatment group. VI-RADS score were performed on the whole group and subgroups, respectively. A score of 3 or greater was defined as imageological muscle infiltration. The diagnostic efficacy was assessed by calculating the sensitivity, specificity, positive (PPV) and negative predictive values (NPV) for the whole cohort of patients and all subgroups, using postoperative pathological reports as standard reference. The receiver operating characteristic (ROC) curve analysis was performed, and the area under the curve (AUC) was used to assess the performance of VI-RADS score for predicting muscle-invasive tumor in each group. Z test was performed to compare the difference of AUC between groups. Two-tailed P value less than 0.05 was considered significant. The study design is shown in Figure 1.
Results
A total of 73 patients were enrolled. For the whole group, a score of 3 or greater for VI-RADS to predict muscle invasive tumors resulted in the sensitivity, specificity, PPV, and NPV of 90.0% (95% confidence interval [CI]: 68.3–98.8), 86.8% (95% CI: 74.7–94.5), 72.0% (95% CI: 50.6–87.9) and 95.8% (95% CI: 85.7–99.5), respectively. The AUC value was 0.945 (95% CI: 0.865–0.985). For the primary group, the sensitivity, specificity, PPV, and NPV were 87.5% (95% CI: 47.3–99.7), 91.2% (95% CI: 76.3–98.1), 70.0% (95% CI: 34.8–93.3), and 96.9% (95% CI 83.8–99.9), respectively. For the post-treatment group, the sensitivity, specificity, PPV, and NPV were 91.7% (95% CI: 61.5–99.8), 79.0% (95% CI: 54.4–93.9), 73.3% (95% CI: 44.9–92.2), and 93.7% (95% CI 69.8–99.8), respectively. The AUC of the primary group was 0.936 (95% CI: 0.815–0.988) and the AUC of the post-treatment group was 0.947 (95% CI: 0.803–0.995). For the recurrence group, the sensitivity, specificity, PPV, and NPV were 88.9% (95% CI: 51.8–99.7), 85.7% (95% CI: 57.2–98.2), 80.0% (95% CI: 44.4–97.5), and 92.3% (95% CI: 64.0–99.8), respectively. The AUC value was 0.944 (95% CI: 0.763–0.997). No significant differences were observed in AUC between the primary group and the post-treatment group (p = 0.870), and in AUC between the primary group and the recurrence group (p = 0.912). Sensitivity, specificity and ROC results in all groups are summarized in Figure 2.
Discussion
In the entire cohort including patients with bladder cancer after treatment, VI-RADS still performed well in identifying muscle infiltration. In the primary group, post-treatment group and recurrence group, all AUC values of VI-RADS were greater than 0.90. Our results demonstrated that there were no statistically significant differences in diagnostic efficacy of VI-RADS between the primary group and the post-treatment group, and between the primary group and the recurrence group. These results indicated that that post-treatment changes in the bladder wall do not completely limit the applicability of VI-RADS for these patients.
Conclusions
Our results showed that VI-RADS does not need to be limited to patients with incipient BC. Patients after treatment and with recurrent disease may benefit from it as well under certain conditions. Overall, this study demonstrated that VI-RADS can be considered an effective preoperative imaging staging tool for a wider range of BC patients and should be investigated further.

Acknowledgements

No acknowledgement found.