Survival and progression in high grade tumour subset of G2 and G3 pT1 bladder transitional cell carcinoma

Results Of the 44 patients with G3 tumours, 20 are alive versus 22 of the 61 patients with high grade tumour subset of G2 ( P = 0.04). Of the 44 patients with G3 tumours, 13 progressed versus 12 of the 61 patients with high grade tumour subset of G2 ( P = 0.02). In multivariate analysis, G3 was a significant predictor of tumour progression ( P = 0.05) and marginally non-significant predictor of poor patient survival ( P = 0.056). Conclusions A notable difference in survival and progression between high grade tumour subset of G2 and G3 is observed. Keywords Bladder Cancer Grade Progression Superficial Survival Abbreviations BCG Bacillus Calmette-Guerin CIS carcinoma in situ G2 grade 2 G3 grade 3 HGPUC high grade papillary urothelial cancer ISUP International Society of Urological Pathologists TCC transitional cell carcinoma TURBT transurethral resection of bladder tumour WHO World Health Organization Introduction The old 1973 World Health Organization (WHO) grading system classified urothelial transitional cell carcinoma of the urinary bladder into 3 grades. Grade 1 tumours have the least degree of anaplasia compatible with the diagnosis of malignancy. Grade 3 (G3) applies to tumours with the most severe degree of cellular anaplasia and grade 2 (G2) lies between these two. 1 In the 1998 World Health Organization (WHO)/International Society of Urological Pathologists (ISUP) grading system for urothelial transitional cell carcinoma (TCC), a subset of G2 in the old 1973 WHO 3 tier system was added to G3 to form one high grade group, high grade papillary urothelial cancer (HGPUC). 2 The attempt to split G2 tumours of the old grading system into high and low grades was made with the intent to eliminate the vagueness and inter-observer variability associated with the old 1973 WHO grading system for urothelial TCC, which ended by the pathologist placing the majority of these tumours into the G2 group. 3 The 1998 WHO/ISUP consensus classification is designed to replace the popular and most commonly used 1973 WHO grading system and forms the basis of the 2004 WHO grading system, the latest system recommended for clinical use by the WHO. 4 In this report, we evaluate for any difference in survival and progression between high grade tumour subset of G2 and G3 in a cohort of pT1 (tumour invading lamina propria) bladder cancer. Previous work assessed for such differences between these tumour grades only in pTa bladder tumours. 3,5 Materials and methods Patients under study Between January 1 st 1991 and December 31 st 2003, 105 patients were diagnosed with pT1 TCC bladder tumours fulfilling the 1998 WHO/ISUP, HGPUC criteria. This comprised 85 men and 20 women with mean age at time of diagnosis of 69.4 years (range 37 to 90 years). Patient and tumour details are illustrated in Table 1 . Inclusion criteria All high grade tumour subset of G2 and G3 pT1 bladder TCC, primary and secondary that had HGPUC histological features ( n = 105). Exclusion criteria – Tumours that were pT1G2 not fulfilling HGPUC criteria. – Tumours with lamina propria invasion but no muscle included to determine muscle invasion. – Tumours that were found under staged in the early cystoscopic examination following transurethral resection of bladder tumour. – T1 tumours managed by radical cystectomy as the primary treatment modality and were found under-staged following postoperative pathological evaluation of the radical cystectomy specimen. Treatment modalities All patients were initially treated by TURBT with superficial and deep tumour resection. A follow-up cystoscopy early after the diagnosis of pT1 was performed to out rule under staging together with resection of any apparent residual tumour. Four treatment modalities were identified. – TURBT alone in 39 patients. – Intravesicle Bacillus Calmette-Guerin (BCG-medac ® , medac GmbH, Gesellschaft fur Klinische, Germany) following TURBT for 6 weeks in 24 patients Dose of intravesicle BCG was 120 mg. – Intravesicle Epirubicin (Epirubicin- Baxter ® , Baxter Healthcare, IL, USA) following TURBT for 6 weeks in 25 patients. Dose of Epirubicin used was 50 mg. – Early radical cystectomy in 17 patients. Intravesicle BCG or Epirubicin was retained in the bladder for 2 h and the treatment course would begin within the first 3 weeks following TURBT for patients receiving BCG and the first week for patients receiving Epirubicin. The decision on the type of management was a product of individualised doctor-patient consultation which included a discussion about each treatment modality reporting outcomes, side effects and impact on quality of life. Patients preferring conservative management for high grade pT1 tumours and associated carcinoma in situ (CIS) were offered BCG rather than Epirubicin. Follow up and recurrence management in patients treated conservatively The frequency of follow up cystoscopy was every 3 months during the first 2 years, every 4 months in the third year, and every 6 months thereafter up to 5 years and then yearly. With any disease recurrence this schedule was recommenced. Recurrences were managed with a consistent protocol. – In patients treated by TURBT alone ( n = 39) and continued to recur at a lower stage and grade were continued to be offered TURBT. – Patients who recurred following treatment with intravesicle BCG ( n = 24) were offered another cycle of 6 weeks BCG. – Patients who recurred following treatment with intravesicle Epirubicin ( n = 25) were offered another cycle of 6 weeks Epirubicin. The decision for radical cystectomy was based either on demonstration of local progression or recurrent high grade disease with perceived risk of progression. Progression was defined as development of muscle invasive disease (≥T2) or metastasis (M+). Follow up for patients treated with early radical cystectomy Post operatively; patients were seen 3 weeks after discharge then at 3 monthly intervals during the first year, 6 monthly intervals during the second year then annually thereafter. Routine clinical and biochemical evaluation were done. Imaging investigations were tailored according to symptomatic manifestations. Local or distant recurrences were managed by chemotherapy, radiotherapy or by keeping terminally ill patients comfortable. Histopathological grading Tumours were originally graded according to the 1973 WHO 3 tier system. Haematoxylin and Eosin (H&E) slides for those tumours were reviewed and re-graded. Re-grading was done by 2 of the authors acting together to reach a consensus view according to the 1998 WHO/ISUP consensus classification criteria without knowledge of clinical outcome. This resulted in 44 G3 and 61 high grade tumour subset of G2, meeting the new 1998 WHO/ISUP pathological criteria for HGPUC. 2 Tumours with mixed patterns of G3 and high grade tumour subset of G2 were grouped under the G3 category. Statistical methods Fischer Exact and Rank-Sum tests were used to compare demographic differences between grades. Kaplan–Meier methods were used to determine survival outcomes at different time points. Wilcoxon (Breslow) tests were used to compare equality of survival between groups. Cox proportional Hazards models were used for multivariate analysis. A P -value ≤0.05 was deemed to be a significant result. All of the analysis was conducted using Stata (Version 8, College Station, Texas, USA). Results Follow-up Median follow up time was 4 years (range 0.7–13) with 40% of patients followed up for a minimum of 5 years. Patient survival Of the 105 patients under study 61 are alive. Of the 44 patients with G3 tumours, 20 are alive versus 22 of the 61 patients with high grade tumour subset of G2 ( P = 0.04). Fig. 1 illustrates the actuarial disease specific survival rates for the 2 tumour grades. Metastasis as a cause of death occurred in 11 patients with G3 tumours versus 8 in patients with high grade tumour subset of G2 ( P = 0.02). Tumour progression Of the 105 tumours under study 25 progressed. Of the 44 G3 tumours, 13 progressed versus 12 of the 61 high grade tumour subset of G2 ( P = 0.02). Fig. 1 illustrates the actuarial progression free survival rates for the 2 tumours. There was a trend for G3 tumours to have worse progression free survival rates than the high grade tumour subset of G2 after controlling for treatment modality chosen. This was statistically non-significant owing to smaller numbers compared following breakdown according to treatment modality chosen. Deferred radical cystectomy The incidence rate of deferred cystectomy per annum in the G3 and high grade tumour subset of G2 tumours initially managed by conservative treatment modalities ( n = 88) was 12.8% and 5.4% respectively ( P = 0.08). Multivariate analysis results Results are illustrated in Table 2 . Discussion We assessed the survival and progression patterns for high grade tumour subset of G2 and G3 were the same, as these 2 grades are summed as one homogenous pathological grade, HGPUC, in the 1998 WHO/ISUP consensus classification. A recent report raised concerns that the new 1998 WHO/ISUP grading system may reduce the diagnostic accuracy of urine cytology after combining the high grade tumour subset of G2 and G3 into one group. Curry and Wojcik 6 evaluated 100 urine cytology specimens, with urine cytology positive in 11 of 26 in G1 tumours, 28 of 61 in G2 tumours and 12 of 13 in G3 tumours. After the reclassification to the new grading system, there was 1 papilloma, 12 papillary urothelial neoplasms of low malignant potential (PUNLMP) lesions, 50 low-grade urothelial carcinomas, and 37 high-grade carcinomas (HGPUC). The cytology was positive in 28 of 37 HGPUC tumours. The authors concluded that the accuracy of urinary cytology for the detection of high-grade lesions has decreased, because the criteria for high-grade lesions is lowered, placing most of the previous WHO G2 tumours into the current high-grade carcinoma category. Our results showed substantial difference in both survival and progression patterns between high grade tumour subset of G2 and G3. The overall survival and progression values were lying in between these 2 grades. There is evidence that supports our results. Holmang 3 reported on 363 patients with pTa bladder tumours of whom 108 had HGPUC noted a statistically significant worse progression rate for G3 than high grade tumour subset of G2. Samaratunga 5 compared the 1973 WHO and the 1998 WHO/ISUP grading systems in 134 patients with pTa bladder cancer and similarly noted a higher progression rate for G3 than the high grade tumour subset of G2 at 45 month follow-up period. In another report by Holmang 7 a significantly poorer survival rate was noted with G3 when compared to the high grade tumour subset of G2 despite advanced tumour stage (pT3) in upper urinary tract transitional cell cancer. In this report by Holmang, tumour grade added little prognostic value in pT1 tumours. However the numbers tested for pT1 tumours were smaller than what we report. The G2 tumours in Holmang's report were that of the 1999 WHO grading system. These represent the high grade tumour subset of G2 in the old 1973 WHO grading system, similar to the G2 population in our report. In the present study, G3 was a predictor of poor patient survival and progression in univariate analysis. In the multivariate analysis, G3 continued to be a predictor of tumour progression ( P = 0.05) and marginally non-significant predictor of poor patient survival ( P = 0.056). Large tumour size, multiplicity and CIS were not significant predictors of poor patient survival and tumour progression with multivariate analysis. However 44/105 of the tumours under study were large (≥3 cm) and 41/105 were multiple (≥4). Also, there was a tendency to offer BCG or radical cystectomy for pT1G3 with CIS. The efficacy of these treatment modalities in improving survival and preventing tumour progression in patients with CIS is well documented. 8 Also, tumour multiplicity and size were reported to be more important predictors of tumour recurrence while for evolution of superficial bladder cancer to invasive disease, the anaplasia grade and T-stage were the most important. 9 We included patients with pT1 tumours treated with early radical cystectomy in the study population. We sought so, since patients diagnosed initially as pT1 tumours and were managed by early radical cystectomy and found under-staged in the post cystectomy pathological diagnoses were excluded. Also, recent reports showed comparable results for radical TURBT and radical cystectomy in muscle invasive disease in selected patients. 10,11 Another reason to include pT1 tumours managed by early radical cystectomy was that all tumours managed by conservative treatment approach had early cystoscopy after TURBT and if found under-staged, were excluded. Dalbagni 12 concluded that the risk of under staging in pT1 disease is negligible after restaging TUR. Within our study, the incidence of deferred cystectomy per annum in those patients treated initially with a bladder sparing approach is higher in pT1G3 tumours than the high grade tumour subset of G2, an indication of increased risk of tumour recurrence and progression for pT1G3. Similarly the incidence of mortality due to metastasis was higher in pT1G3. A significant P value was not reached in assessing individual treatment outcome for high grade tumour subset of G2 and G3 pT1 bladder tumours. This is due to the small numbers of patients compared following breakdown by group. However assessing individual treatment outcome was not the main intent of our study. This type of testing necessitates a bigger patient population than our study group. To summarise, we compared high grade tumour subset of G2 and G3 in pT1 bladder cancer and we noted a trend for poorer survival and progression rates for G3 than the high grade tumour subset of G2. This also applied to the rate of development of metastasis and the incidence of deferred cystectomy. Conclusions Based on our results, in pT1 bladder tumours, the high grade tumour subset of G2 and G3 are two separate entities, though they are summed together as one homogenous group in the 1998 WHO/ISUP grading system. References 1 J.I. Epstein The new World Health Organization/International Society of Urological Pathology (WHO/ISUP) classification for TA, T1 bladder tumors: is it an improvement? Crit Rev Oncol/Hematol 47 2003 83 89 2 J.I. Epstein M.B. Amin V.E. Reuter F.K. Mostofi The World Health Organization/International Society of Urological Pathology consensus classification of Urothelial (transitional cell) neoplasms of the urinary bladder Am J Surg Pathol 22 1998 1435 1448 3 S. Holmang P. Andius H. Hedelin K. Wester C. Busch S.L. Johansson Stage progression in Ta papillary urothelial tumors: relationship to grade, immunohistochemical expression of tumor markers, mitotic frequency and DNA ploidy J Urol 165 2001 1124 1128 4 A. Lopez-Beltran R. Rodolfo Montironi Non-invasive urothelial neoplasms: according to the most recent WHO classification Eur Urol 46 2004 170 176 5 H. Samaratunga D.V. Markov J.I. Epstein Comparison of WHO/ISUP classification of non-invasive papillary urothelial neoplasms for risk of progression Urology 60 2002 315 319 6 J.L. Curry E.M. Wojcik The effects of the current World Health Organization/International Society of Urologic Pathologists bladder neoplasm classification system on urine cytology results Cancer 96 2002 140 145 7 S. Holmang S.L. Johansson Urothelial carcinoma of the upper urinary tract: comparison between the WHO/ISUP 1998 consensus classification and WHO 1999 classification system Urology 66 2005 274 278 8 A.P. van der Meijden R. Sylvester W. Oosterlinck EAU guidelines on the diagnosis and treatment of urothelial carcinoma in situ Eur Urol 48 2005 363 371 9 W. Oosterlinck B. Lobel G. Jakse P.U. Malmstrom M. Stockle C. Sternberg European Association of Urology (EAU) Working Group on Oncological Urology Guidelines on bladder cancer Eur Urol 41 2002 105 112 10 H.W. Herr Transurethral resection of muscle-invasive bladder cancer: 10-year outcome J Clin Oncol 19 2001 89 93 11 E. Solsona I. Iborra J.L. Monros J. Casanova C. Calabuig Feasibility of transurethral resection for muscle infiltrating carcinoma of the bladder: long-term follow up of a prospective study J Urol 159 1998 95 98 12 G. Dalbagni H.W. Herr V.E. Reuter Impact of second transurethral resection on the staging of T1 bladder cancer Urology 60 2002 824 825