Nuclear morphometry in grading transitional cell bladder cancer compared with subjective histological grading.

A retrospective study was performed comprising 265 bladder cancer patients. The patients were clinically followed up for an average of 10 years. The initial tumour biopsies were subjected to morphometric analysis. The mean nuclear area (NA), the standard deviation of nuclear area (SDNA) and the mean area of the 10 largest nuclei (NA10) were measured using IBAS 1&2 image analyzer. The prognostic value of NA, SDNA, NA10, papillary, subjective histological grading (WHO) and clinical stage (UICC) was evaluated. The progress in T-category was related to histological grade (p less than 0.0001), non-papillar growth (p = 0.0023), SDNA (p = 0.0110) and NA10 (p = 0.0305), in that order. The same parameters in addition to NA predicted lymph node involvement and metastasis. Recurrence rate was significantly related to NA10 (p = 0.0250). Non-papillar growth (p = 0.002), clinical stage (p = 0.005), histological grade (p = 0.0120), NA (p = 0.0143), SDNA (0.0383) and NA10 (p = 0.0632) predicted recurrence-free period. Bladder cancer survival was related to clinical stage (p less than 0.0001), histological grade (p less than 0.0001), SDNA (p less than 0.0001), non-papillar growth (p less than 0.0001), NA (p = 0.0001) and NA 10 (p = 0.0001), in that order. Grade II tumours could be regrouped prognostically using NA (p = 0.006), SDNA (p = 0.033) and NA10 (p = 0.016) as classifiers. Clinical stage, NA and histological grade predicted bladder cancer survival in a multiparameter analysis. The results show that NA and SDNA are powerful prognosticators of survival. NA10 and SDNA predict progression better than NA. The multiparameter analysis identified clinical stage, histological grade and NA as the most important prognosticators of survival.