Diagnostic Value of Smoothelin and Vimentin in Differentiating Muscularis Propria from Muscularis Mucosa of Bladder Carcinoma

Deep learning for histopathological segmentation of smooth muscle in the urinary bladder

BACKGROUND

Histological assessment of smooth muscle is a critical step particularly in staging malignant tumors in various internal organs including  the urinary bladder. Nonetheless, manual segmentation and classification of muscular tissues by pathologists is often challenging. Therefore, a fully automated and reliable smooth muscle image segmentation system is in high demand.

METHODS

To characterize muscle fibers in the urinary bladder, including muscularis mucosa (MM) and muscularis propria (MP), we assessed 277 histological images from surgical specimens, using two well-known deep learning (DL) model groups, one including VGG16, ResNet18, SqueezeNet, and MobileNetV2, considered as a patch-based approach, and the other including U-Net, MA-Net, DeepLabv3 + , and FPN, considered as a pixel-based approach. All the trained models in both the groups were evaluated at pixel-level for their performance.

RESULTS

For segmenting MP and non-MP (including MM) regions, MobileNetV2, in the patch-based approach and U-Net, in the pixel-based approach outperformed their peers in the groups with mean Jaccard Index equal to 0.74 and 0.79, and mean Dice co-efficient equal to 0.82 and 0.88, respectively. We also demonstrated the strengths and weaknesses of the models in terms of speed and prediction accuracy.

CONCLUSIONS

This work not only creates a benchmark for future development of tools for the histological segmentation of smooth muscle but also provides an effective DL-based diagnostic system for accurate pathological staging of bladder cancer.

Bicentric Retrospective Analysis of en Bloc Resection and Muscularis Mucosae Detection Rate in Non-Muscle Invasive Bladder Tumors: A Real-World Scenario

Introduction

For risk stratification of non-muscle invasive bladder cancer (NMIBC), the depth of stromal invasion can be further classified, where the lamina muscularis mucosae (MM) serves as a reference structure. While the overall identifiability of MM in standard transurethral specimens is low, en bloc resection may help in identification and overall orientation. The aims of this study were to report the detection rate of MM in en bloc resected bladder tumors (ERBT) and to provide real-world information on tissue stability and preservation of en bloc architecture during recovery and processing for histopathologic evaluation.

Methods

Thirty-four ERBT specimens were histologically re-evaluated with regard to MM detectability and structure as well as the presence of en bloc architecture and further histologic features. Associations with tumor size and energy source and within histologic parameters were assessed by standard Pearson's chi-squared analyses and Cramér’s V effect size testing (V).

Results

The first parameter assessed was MM detection rate. In 19 out of 34 samples (56%) MM was detectable: scattered in 9 cases (26%), interrupted in 8 cases (24%) and continuous in 2 cases (6%). The second parameter assessed was preservation of en bloc architecture. In 11 out of 34 samples (32%), en bloc architecture could not be confirmed, and these samples served as a reference group for the detection of MM. Preservation of en bloc architecture was associated with an increased MM detection rate (MM in en bloc preserved 16/23, 70% vs. non-preserved 3/11, 27%; p = 0.020; V = 0.398) and with tumor size (p = 0.005; V = 0.595). Medium-sized tumors (1.1–2 cm) were best preserved. The choice of energy source did not show relevant association with en bloc architecture (p = n.s.).

Conclusions

In line with recent publications, ERBT increases the MM detection rate considerably. However, a third of the ERBT specimens lost en bloc architecture during sample recovery and processing. Tumor size is a relevant factor, with optimal architecture preservation between 1 and 2 cm. Optimizing resection techniques, recovery, transport, and diagnostic processing of ERBT samples is warranted to verify the diagnostic value of MM-based substaging.

Analysis of Tumor Depth Invasion With Anti-Smoothelin Antibody in Equivocal Transurethral Resection of Urinary Bladder Tumor Surgical Specimens

OBJECTIVE

To examine the expression and value of the smoothelin marker in control cases, to standardize the working method, and to analyze its application in pathologic staging process of problematic transurethral resection of bladder tumor (TURBT) cases.

MATERIAL AND METHODS

Immunohistochemical (IHC) staining was performed on tumor-free bladder wall sections, tumor-free large bowel sections, TURBTs with unequivocal tumor stage, and TURBTs with equivocal stage. The IHC staining of muscularis mucosa (MM), muscularis propria (MP), and blood vessels was evaluated semiquantitatively.

RESULTS

Smoothelin IHC staining pattern ranged from negative (30% to 67% cases) to 2+ (0% to 15% cases) in MM and from 1+ (10% to 50% cases) to 3+ (9% to 48% cases) in MP. When compared on the same slide, the smoothelin expression of MP showed a stronger staining intensity than the one of the MM in all the analyzed cases. Blood vessel muscle cells stained in a constant intensity as the MM (r = 0.9808; r = 0.9604). Smoothelin determined restaging of 33% of the problematic TURBT cases.

CONCLUSION

Smoothelin is an IHC marker that shows differential staining between coexistent MM and MP; however, variations in staining intensity and pattern may occur, aspects that can be influenced by different technique variables. We recommend using this marker as a diagnostic tool in problematic TURBT cases only when there is sufficient experience in control cases with this antibody.