What's new in WHO fifth edition – urinary tract

Abstract

The fifth edition of the WHO Blue Book on urological tumours, specifically in the bladder chapter, represents a refinement and update in the classification of bladder tumours building on the aggregate major changes made in previous editions. Progress in the molecular underpinnings of urothelial tumours, particularly with promising stratifiers for more precision‐based treatment approaches, have been made. Special attention has been paid to burning questions in bladder pathology, such as grading, heterogeneous lesions, inverted tumours and substaging. The concept of neuroendocrine tumours will be explained precisely.

Influence of lamina propria invasion extension on T1 high‐grade non‐muscle‐invasive bladder cancer in patients undergoing BCG or radical cystectomy

“…In this study, it was not reported in 28% of cases. The heterogeneity of T1 substaging systems has resulted in a lack of consensus regarding the adoption of a single standardised method [13].…”

Grade Heterogeneity in High-Grade Urothelial Carcinomas: Does It Have an Impact on the Survival of Patients With Intermediate/High-Risk Nonmuscle-Invasive Bladder Cancer Who Received Adequate Adjuvant Bacillus Calmette-Guérin Therapy?

“…14 To reduce interobserver variability among pathologists in the histological grading of urothelial carcinomas with grade heterogeneity, the WHO’s latest edition recommended a 5% HG component threshold for the diagnosis of HG UC. Accordingly, all papillary lesions with 5% or more HG areas are considered HG, while those with HG areas less than the threshold of 5% are considered LG, and it is recommended to report them as “LG-UC with focal HG component.” 1…”

“…Nonmuscle-invasive bladder cancer (NMIBC) can be categorized into papillary urothelial neoplasm of low malignant potential, noninvasive papillary carcinoma of low grade (LG), and high grade (HG) based on the 2016/2022 WHO classification. 1 In addition to the grade variable, various factors such as tumor size and the presence of carcinoma in situ (CIS) determine the patient’s risk group, and the risk group affects the treatment decisions. 2 In the European Organization for Research and Treatment of Cancer (EORTC) scoring system, tumor grade appears to be the second most significant factor determining progression risk after CIS.…”

Seeing the random forest through the decision trees. Supporting learning health systems from histopathology with machine learning models: Challenges and opportunities

“…However, the need to use them may partially explain why most histopathology classifications are updated recurrently, even if some of the tissue morphological changes used in their diagnostic criteria remain unchanged. [101][102][103][104][105][106][107][108] Similar to what was stated by Dr. Juan Rosai more than 2 decades ago, there are still no techniques more cost-effective, flexible, and rapidly informative than the morphological assessment of tissues by pathologists in clinical settings. 96 However, some techniques whose original role was to support or complement morphological histopathology classifications (e.g., immunohistochemistry and omics-based studies) 9,77,90,110,158 are now redefining some of them 103,107,109 and in a few specific cases, replacing them for treatment purposes, 162,163 somewhat reminiscent of how microscopic morphologic assessments once started to improve the recognition and prediction capabilities of gross descriptions and clinical findings.…”

“…That is because histopathology classifications usually change over time (i.e., are "moving targets"), but most diagnostically relevant cells/tissue components do not. [101][102][103][104][105][106][107][108][109][110][111] Also, as some of these cells/tissue components are listed in the different diagnostic criteria of many related diseases, these models could accelerate the development of tools that can support pathologists in more than one narrow task (e.g., by facilitating the recognition of several diseases). 112 In addition, the need to develop "explainable" algorithms would be less relevant and could gain regulatory agencies' approval easier if their scope is to be used as decision-support tools to improve pathologists' workflows (and not replace them in clinical settings by making diagnoses directly).…”