Bladder Cancer Study Challenges WHO’s 5% Rule for Grading Mixed Tumors

One of the most stubborn problems in bladder cancer pathology has just received a fresh dose of quantitative evidence. A retrospective study of 180 patients with non-muscle-invasive urothelial carcinoma, published in Pathology and Oncology Research, suggests that the threshold used to decide when a mixed-grade tumor should be called high-grade may be set far too low. The research, led by Gökhan Veli Alkan of Ordu University with colleagues from Hacettepe University and Ankara Etlik City Hospital, found that survival declined as the proportion of high-grade tissue within a tumor increased, and that the statistically meaningful tipping point sat at 27.5 percent — more than five times the 5 percent cut-off proposed by the 2022 WHO classification of urinary and male genital tumours.

The grading of urothelial carcinoma, the most common cancer of the bladder, has long been a battleground of competing classification schemes. The two-tier system that divides tumors into low-grade and high-grade categories was introduced in the 1998 WHO/ISUP consensus report and carried through the 2004, 2016 and 2022 WHO editions. Yet accumulating molecular evidence has led some researchers to argue that the older three-tier system from the 1973 classification correlates more strongly with both clinical outcomes and genetic alterations, and proposals for hybrid three- and four-tier schemes have followed. One of the central arguments for revisiting the system is reproducibility: pathologists asked to grade the same slide do not always agree, and the disagreement is worst when a single tumor contains visibly different grades of tissue.

That phenomenon, known as intratumoral grade heterogeneity, is far from rare. The 2022 WHO classification estimates that roughly one-third of urothelial carcinomas display it, while individual studies have reported that between 5 and 30 percent of non-invasive tumors contain both low-grade and high-grade components. When a pathologist encounters such a mixed tumor, the question of which component should determine the final grade is not academic. Under the 2004 WHO rules, non-invasive carcinomas with any grade heterogeneity were placed in the high-grade category, but several studies have since shown that the behavior of these tumors can resemble low-grade disease more than high-grade disease, raising the possibility that some patients are being labeled — and treated — more aggressively than their biology warrants.

The 2022 WHO classification attempted to standardize the answer with a numerical rule: if a low-grade tumor contains at least 5 percent high-grade component, it should be graded as high-grade; if the high-grade fraction falls below 5 percent, the tumor should be reported as low-grade with a minor high-grade component. The classification itself acknowledged that this recommendation rested on limited evidence, and the 2023 ISUP consensus report similarly concluded that current data support an intermediate behavior for low-grade tumors with focal high-grade areas, without reaching agreement on what proportion should define focal. The new Turkish study was designed precisely to supply the missing evidence, asking whether the size of the high-grade component carries prognostic weight and, if so, where the clinically meaningful threshold lies.

The researchers combed the archives of Ankara Dışkapı Training and Research Hospital for transurethral resection of bladder tumor specimens diagnosed between January 2019 and August 2022. All hematoxylin and eosin slides were re-evaluated according to the 2022 WHO criteria, with the proportion of low- and high-grade components and the presence of lamina propria invasion recorded for each case. The team excluded tumors with more than 90 percent high-grade component, tumors with no high-grade component at all, muscle-invasive cancers, carcinoma in situ, specific histological subtypes and recurrent carcinomas, leaving 180 analyzable cases after four were dropped for missing clinical data. The cohort was overwhelmingly male — 156 of 180 patients, or 86.7 percent — with a mean age of 65.9 years and a mean high-grade proportion of 33.7 percent, ranging from 2 to 80 percent.

The clinical picture that emerged was nuanced. Recurrence occurred in 48 patients, or 26.7 percent of the cohort, while lamina propria invasion was present in a quarter of cases. The two-year survival rate was 73.3 percent, with a mean overall survival of 14.6 months. Strikingly, the proportion of high-grade component showed no significant association with recurrence: patients who relapsed had a mean high-grade fraction of 37.6 percent versus 32.3 percent in those who did not, a difference the authors deemed statistically insignificant. Survival, however, told a different story. Patients who survived less than two years carried a mean high-grade proportion of 46.4 percent, compared with 29.1 percent among those who survived longer — a significant difference — and they were also older, averaging 72.6 years against 63.4 years.

Receiver operating characteristic curve analysis sharpened the finding into a number. A high-grade component of at least 27.5 percent predicted shorter survival with an area under the curve of 0.682 and a p-value of 0.032, while an age threshold of 66.5 years was an even stronger predictor, with an area under the curve of 0.756. No significant cut-off values emerged for recurrence with respect to high-grade proportion, age or survival time. Kaplan-Meier analysis showed a parallel trend toward worse survival in patients above the 27.5 percent threshold, though it narrowly missed statistical significance with a log-rank p-value of 0.061. The study also confirmed the expected relationship between grade and invasion: tumors that had invaded the lamina propria contained a significantly higher high-grade proportion, 51.3 percent versus 27.8 percent, and occurred in significantly older patients.

These results dovetail with a growing body of literature suggesting that mixed-grade tumors occupy a genuine middle ground. Tumors with grade heterogeneity have been reported to behave clinically like low-grade cancers, and one analysis of their response to intravesical BCG therapy found higher response rates than in pure high-grade tumors, leading the authors to describe them as a favorable prognostic subtype. A recent evaluation of 160 papillary urothelial carcinomas with grade heterogeneity reported that, when the 5 percent threshold was applied, tumors classified as high-grade survived better than pure high-grade tumors but worse than pure low-grade tumors — intermediate behavior that sits awkwardly with a binary classification. Conversely, a 2025 study in Human Pathology found that mixed-grade papillary tumors with even a minor high-grade component fared significantly worse than pure low-grade tumors, underscoring that the high-grade fraction is not prognostically inert.

The authors of the new study are careful about what their number does and does not mean. The 27.5 percent cut-off is higher than the WHO’s 5 percent recommendation, and they argue that applying a higher threshold in routine practice could limit the subjectivity introduced by focal degenerative changes that can mimic high-grade features, potentially improving reproducibility among pathologists. At the same time, they emphasize that risk stratification incorporating tumor size, multiplicity and the presence of carcinoma in situ — factors highlighted in the European Association of Urology’s 2021 guidelines for non-muscle-invasive bladder cancer — remains essential for prognostic evaluation. Their own study did not include tumor multifocality or coexisting carcinoma in situ among the analyzed cases, and the absence of tumor size data is acknowledged as a limitation, as is the single-center, retrospective design with a modest sample size.

What the study delivers is a concrete, data-driven challenge to a rule that was always flagged as provisional. If a high-grade fraction must exceed roughly a quarter of the tumor before survival is measurably affected, then sweeping every tumor with a 5 percent high-grade component into the high-grade category may misclassify a substantial group of patients whose tumors behave in an intermediate fashion — exposing some to unnecessary intensive surveillance and treatment while the true high-risk minority still needs to be identified. The authors call for multicenter studies incorporating tumor size and other clinical and prognostic parameters to strengthen grading consistency and treatment decisions. Until such data arrive, the 5 percent rule stands, but pathologists and urologists now have a quantified reason to view it with healthy skepticism — and a specific number to test against.

Subject of Research: Prognostic significance of the high-grade component proportion in non-muscle-invasive bladder urothelial carcinoma with grade heterogeneity

Article Title: Prognosis and tumor grading in bladder urothelial carcinomas with intratumoral grade heterogeneity

Article References: Alkan, G. V., Doğan, K., Türk Alkan, M., & Önder, E. (2026). Prognosis and tumor grading in bladder urothelial carcinomas with intratumoral grade heterogeneity. Pathology & Oncology Research, 32, Article 1612553. https://doi.org/10.3389/pore.2026.1612553

Image Credits: AI Generated

DOI: 10.3389/pore.2026.1612553

Keywords: bladder cancer, urothelial carcinoma, tumor grading, grade heterogeneity, WHO classification, non-muscle-invasive bladder cancer, pathology, prognosis, survival, ROC analysis, lamina propria invasion, ISUP consensus