Response classification after neoadjuvant chemotherapy in muscle-invasive bladder carcinoma is based on the TNM stage at radical cystectomy. We recently showed that histopathologic tumor regression grades (TRGs) add prognostic information to TNM. Our aim was to validate the prognostic significance of TRG in muscle-invasive bladder cancer in a multicenter setting. We enrolled 389 patients who underwent cisplatin-based chemotherapy before radical cystectomy in 8 centers between 2010 and 2016. Median follow-up was 2.2 years. TRG was determined in radical cystectomy specimens by local pathologists. Central pathology review was conducted in 20% of cases, which were randomly selected. The major response was defined as ≤pT1N0. The remaining patients were grouped into partial responders (≥ypT2N0-3 and TRG 2) and nonresponders (≥ypT2N0-3 and TRG 3). TRG was successfully determined in all cases, and interobserver agreement in central pathology review was high (κ=0.83). After combining TRG and TNM, 47%, 15%, and 38% of patients were major, partial, and nonresponders, respectively. Combination of TRG and TNM showed significant prognostic discrimination of overall survival (major responder: reference; partial responder: hazard ratio 3.5 [95% confidence interval: 1.8-6.8]; nonresponder: hazard ratio 6.1 [95% confidence interval: 3.6-10.3]). This discrimination was superior compared with TNM staging alone, supported by 2 goodness-of-fit criteria (P=0.041). TRG is a simple, reproducible histopathologic measurement of response to neoadjuvant chemotherapy in muscle-invasive bladder cancer. Integrating TRG with TNM staging resulted in significantly better prognostic stratification than TNM staging alone.
RC preceded by cisplatin-based NAC is the standard treatment for cT2-T4aN0M0 bladder cancer patients. NAC has shown to improve survival by an 8% at five years compared to surgery alone, without affecting surgical morbidity and complications’ rates. Final pathologic stage at RC represents the strongest predictor of long-term oncological outcomes, with 5-years cancer-specific survival rates ranging from 88% in pT0N0M0 disease to 30% in pT4N0M0 and even less in nodal positive disease. To date, pathologic response has been always evaluated using the TNM staging system, usually categorizing patients into complete responders (those with pT0N0 at final pathology), partial responder (pTa, pTis, pT1N0) and non-responders (³pT2 or N+ disease). However, not all the published literature adopted this classification of response, thus making difficult to compare the results from different series. Moreover, even if is presumed that patients with residual muscle-invasive disease did not benefit from NAC, this is not fully demonstrated and underline the need for better prognostic models.
Previously, Fleischmann et al., in a pilot study of 56 patients, demonstrated that TRG was able to predict overall survival in patients undergoing NAC and RC and, even, to outperform the predictive accuracy of final pathologic TNM classification. Voskuilen and colleagues provided an external validation of these findings by combining the predictive role of both TNM and TRG in a large multicenter cohort of bladder cancer patients. This trial has several strengths such as the large sample size, the multicentric nature of the study, the central pathologic review of the specimens (performed in the 20% of cases) and the reproducibility of the study (a high pathologic interobserver agreement of TRG was observed). Limitations of the study mainly regard the heterogeneity of the enrolled population. Actually, the inclusion of clinical node-positive patients remains questionable and may have influenced the results. The findings of this study may have several practical implications. First, the prediction of oncological outcomes after NAC and RC is of fundamental importance for decision-making regarding potential adjuvant systemic treatments and personalized follow up schemes. Second, pathologic response to NAC is usually used as a surrogate endpoint for cancer-specific survival to evaluate the efficacy of neoadjuvant therapies such as NAC and neoadjuvant immunotherapy. However, since several reports questioned the accuracy of TNM classification in chasing this goal, its combination with TRG may represent a better accurate tool readily available for clinical practice.
In this original article the authors aimed to validate the prognostic value of tumor regression grade (TRG) in patients with muscle-invasive bladder cancer undergoing neoadjuvant chemotherapy (NAC) and radical cystectomy (RC). Overall, 389 consecutive patients from eight different centers who underwent NAC and RC for cT2-4aN0-3M0 bladder cancer between 2010 and 2016 were enrolled. Based on final pathology, patients were divided into three groups: those who experienced a major response (£ypT1N0), partial responders (³ypT2N0-N3 and TRG 2), and non-responders (³ypT2N0-N3 and TRG 3). TGR was assessed in every pathologic specimen and a relatively high interobserver variability was observed. The combination of TRG and TNM classification provided to the following results: 47% of patients were categorized to be major responders, 15% partial responders and the remaining 38% non-responders. Combining TRG and TNM led to a significantly improved prognostic accuracy in terms of overall survival when compared to TNM alone. In conclusion, adding TRG to TNM classification may improve patients’ risk stratification and, therefore, decision-making regarding adjuvant therapies and follow up.