2. UroToday Home
  3. Recent Abstracts
  4. Urologic Oncology
  5. Bladder Cancer

Performance of a Modified 3D-T2-Weighted MRI Technique in Characterizing Bladder Cancer - Expert Commentary

Accurate diagnosis and tumor staging are critical for treatment decisions in bladder cancer (BC). Cystoscopic biopsy or transurethral resection of bladder tumor (TURBT) is invasive and has been associated with incorrect staging. New multiparametric magnetic resonance imaging (MRI) protocols have been optimized to evaluate muscle invasiveness and allow for more in-depth diagnostics.


For instance, the 3D-T2-weighted-SPACE sequence has a higher resolution than conventional T2-weighted imaging and provides quantitative radiomics data. Due to practical concerns such as long scanning time and patient discomfort, the compressed sensing (CS) technique was developed to accelerate image acquisition. In a new study, Li et al. evaluated the performance of the 3D-CS-T2-weighted-SPACE sequence.

The study cohort consisted of 108 patients with BC who were randomized into the training cohort (n = 75) and a validation cohort (n = 33). There were no significant differences in clinical characteristics between the two cohorts. To construct the radiomics model, 1,781 features were extracted from each sequence. Features found to be consistent across graders with intraclass correlation coefficients (ICC) of at least 0.75 were subsequently selected, and redundant features were removed. The LASSO logistic regression algorithm was used to construct the models, and the number of features remaining after screening was three in the T2-weighted model, three in the 3D-T2-weighted-SPACE model, and five in the 3D-CS-T2-weighted-SPACE model. The 3D-CS-T2-weighted-SPACE model obtained the highest AUC (0.87; 95% CI, 0.73 - 1.00). Qualitatively, the models had no significant differences in image quality. The investigators subsequently developed a nomogram that included only clinical data, radiomics data, or a combination of both. The decision curve showed that the radiomics and combined nomogram models yielded higher net benefits than the clinical model.

The reduction in time associated with the 3D-CS-T2-weighted-SPACE model did not reduce performance and is, therefore, suitable for clinical applications. This tool can potentially augment more invasive characterization staging if validated in larger cohorts.

Written by: Bishoy M. Faltas, MD, Director of Bladder Cancer Research, Englander Institute for Precision Medicine, Weill Cornell Medicine

References:

  1. Li, S., Fan, Z., Guo, J. et al. Compressed sensing 3D T2WI radiomics model: improving diagnostic performance in muscle invasion of bladder cancer. BMC Med Imaging. Published online June 17, 2024.
Read the Abstract