Bladder cancer grading using the four‐tier combination of the World Health Organization (WHO) 1973 and WHO 2004 classifications

Abstract

ObjectiveTo investigate the impact of grading in urothelial bladder cancer (UBC) stages Ta and T1, comparing the World Health Organization (WHO) grading classifications of 1973 (WHO73) and 2004 (WHO04) and a combination of these (WHO73/04).Patients and MethodsAll patients with primary Ta and T1 UBC in the Östergötland region, Sweden, between 1992 and 2007 were included. From 1992, we introduced a new programme for management and follow‐up of UBC, including prospectively performed registration of all patients, a systematic description of the location and size of all tumours, primary resection and intravesical treatment in the case of recurrence. All tumour specimens were retrospectively reviewed in 2008 and graded according to the WHO73 and WHO04. A combination of WHO73/04, Grade 1 (G1), Grade 2 low grade (G2LG), Grade 2 high grade (G2HG) and Grade 3 (G3) was analysed in relation to clinical variables and outcomes.ResultsThere were 769 patients with a median age of 72 years and a median follow‐up duration of 74 months. Recurrence was noted in 484 patients (63%) and progression in 80 patients (10%). Recurrence was more common in multiple tumours, larger tumours and in tumours of higher grade (G2LG, G2HG and G3). Progression was more common in tumours classified as larger, T1 and G2HG and G3. Notably, in tumours classified as G2HG, recurrence and progression were more common than in the G2LG group. Harrell's concordance index for the WHO73/04 was higher for recurrence and progression than in the WHO73 or WHO04.ConclusionIn the four‐tier combined WHO73/04 for urothelial cancer, we observed two G2 sub‐groups, G2HG and G2LG. There was a better outcome in the latter group, and the importance of G1 and G3 tumours could be fully evaluated. The WHO73/04 had greater accuracy for recurrence and progression than either the WHO73 or WHO04.

Head‐to‐head: how many categories for grading urothelial carcinoma?

“…Similarly, a retrospective multicentre study of 5145 primary Ta/T1 papillary NMIBC among 17 institutions showed that neither WHO 2004 nor 1973 grading predicted recurrence, but both systems were prognostic for progression to muscle‐invasive (T2) carcinoma with a 5‐year progression risk of 13.7 and 18.8% for WHO 2004 HG and WHO 1973 G3 bladder cancer, respectively (Figure 1). 20 Two recent separate studies (one Canadian, one Swedish) confirmed the importance of distinguishing G2‐HG from G3‐HG, as both found a statistically significant difference in progression risk between these two groups 13,18 . Both studies convincingly demonstrated superiority of the hybrid WHO 1973/2004 grading system over two‐tier WHO 2004 using area under the curve or Harrell's concordance index statistics using stage progression as outcome parameter.…”

“…Another advantage of the two‐tiered classification is that the two grades are well established with regard to their clinical impact. In contrast, a recent well‐powered study evaluated a four‐tiered system, and could not determine significant differences in progression between G1 and G2‐LG or G2‐LG and G2‐HG tumours 13 …”

“…20 Two recent separate studies (one Canadian, one Swedish) confirmed the importance of distinguishing G2-HG from G3-HG, as both found a statistically significant difference in progression risk between these two groups. 13,18 Both studies convincingly demonstrated superiority of the hybrid WHO 1973/2004 grading system over two-tier WHO 2004 using area under the curve or Harrell's concordance index statistics using stage progression as outcome parameter.…”

“…The recent study cited by Drs Comperat and van der Kwast reported a statistically significant difference in progression rate between G1-LG and G2-HG, but not between G2-LG and G2-HG. 13 This observation would support adoption of a four-tier grading system that also splits the WHO 2004 LG category.…”

“…In contrast, a recent well-powered study evaluated a four-tiered system, and could not determine significant differences in progression between G1 and G2-LG or G2-LG and G2-HG tumours. 13 The removal of the G2 category implies that tumours previously designated G2 now need to be classified as LG or HGwhich in some 'borderline' cases can be challenging, despite established criteria. However, increasing the number of grades is not a viable solution to this challenge, because an increase in the number of borders will only increase the number of 'borderline' cases.…”