Validation of clinical T stages and of prognostic negative markers in patients with muscle invasive bladder cancer: data in the Swedish National Bladder Cancer Registry vs. data from a detailed research database.

Abstract

Background: A previously published study at Norrland University Hospital, Umeå, Sweden, found that in 29.5% of patients with urinary bladder cancer (UBC) who underwent cystectomy, incorrect cT-stage (clinical T-stage) was registered in the Swedish National Register of Urinary Bladder Cancer (SNRUBC). Tumor in bladder diverticulum (TIBD) and tumor-associated hydronephrosis (TAH) were common causes for misclassification. Our aim was to further investigate cT-staging, as well as pathoanatomical markers, in the SNRUBC, compared to detailed data from medical records in a larger, retrospective multicenter cohort. Our secondary objective was to describe the frequency of pathoanatomical markers in pathology reports (PAD) after transurethral resection of the bladder (TURb): variant histology (VH), concomitant carcinoma in situ (CIS), lymphovascular invasion (LVI) and perineural invasion (PNI).

Methods: Medical records of 630 patients planned for radical cystectomy in the years 2009-2022 in the Northern Healthcare Region, Region of Gävleborg and Region of Västmanland were reviewed. Factors impacting risk of misclassification of cT-staging were identified through logistic regression. In TURb pathology reports, all comments on pathoanatomical markers were identified. For each pathoanatomical marker, respectively, comments were then registered as positive or negative. The absence of a comment on a marker was registered as "not commented".

Results: A total discrepancy rate of 36.5% was found between validated cT-staging and the SNRUBC, of which 13.3% were upstaged from

Conclusions: The SNRUBC has a significant prevalence of misclassification of cT-staging with a large proportion due to TAH and TIBD. Misclassification of VH and CIS is also common. Improved guidelines could increase consistency. Total rates of recorded pathoanatomical markers in TURb-reports are low.