Validation of Preoperative Risk Grouping of the Selection of Patients Most Likely to Benefit From Neoadjuvant Chemotherapy Before Radical Cystectomy

Abstract

Introduction

The aim of this study was to validate the value of preoperative patient characteristics in prognosticating survival after radical cystectomy (RC) to guide treatment decisions regarding neoadjuvant systemic treatment.

Methods

We evaluated a single cohort of 449 consecutive patients treated with RC for bladder cancer. Patients treated with neoadjuvant therapy were excluded from the study cohort (n = 24). Patients were stratified based on preoperative characteristics into 2 risk groups. The high-risk group included patients harboring clinically non–organ-confined disease (≥ cT3), hydroureteronephrosis, lymphovascular invasion, or variant histology (micropapillary, neuroendocrine, sarcomatoid, or plasmacytoid variants on transurethral resection). The low-risk group included patients with cT2 disease without any of the aforementioned features. Survival expectancies after surgery were evaluated using competing risk and Kaplan-Meier analyses.

Results

We identified 153 (44.6%) low-risk and 190 (55.4%) high-risk patients. The majority of high-risk patients had only 1 high-risk feature (n = 111; 58.4%); the most common high-risk feature was preoperative hydroureteronephrosis (n = 107; 56.3%). The majority of low-risk patients were upstaged at time of RC (n = 118; 70.6%), whereas a pathologic downstage occurred only in 27 high-risk patients (14.2%). Cancer-specific mortality-free rates at 5 years after RC were 77.4% versus 64.4% for low-risk versus high-risk patients, respectively.

Conclusions

We confirm that preoperative risk features can stratify patients with muscle-invasive bladder cancer into differential risk groups regarding survival. Decision-making regarding neoadjuvant systemic therapy administration is likely to be improved by integrating clinical stage, lymphovascular invasion, variant histology, and hydroureteronephrosis.

Introduction

Bladder cancer (BCa) is the fourth most common cancer in men and the eleventh in women, with an estimated incidence of 74,000 in 2015 in the United States only.¹ Radical cystectomy (RC) with bilateral pelvic lymph node dissection represents the standard for muscle-invasive bladder cancer and for very high-risk non–muscle-invasive tumors.² Despite adequate surgery, up to 50% of patients experience disease recurrence and eventually die from their disease, supposedly due to micro metastases.3, 4, 5, 6 To improve the outcomes of RC in these patients, neoadjuvant chemotherapy (NAC) has been tested in prospective randomized clinical trials, SWOG-8710⁷ and BA06 30894.⁸ Both reported an improvement in survival for patients treated with NAC composed of methotrexate, vinblastine, doxorubicin, and cisplatin before RC versus those who underwent RC alone. However, this benefit was more evident in patients with clinical stage ≥ T3 disease, with a survival gain of 42 months versus 19 months in patients with clinical stage T3/T4 versus clinical T2 stage.9, 10 On the other hand, toxicity grade 3 and 4 were reported in 35% and 37%, respectively, of patients who underwent NAC.⁷

These results have led to a debate regarding which patients are most likely to benefit most from NAC, so that it can be avoided in those who are not likely to benefit. To address these needs, Culp et al¹¹ proposed a risk stratification based on preoperative parameters such as presence of hydroureteronephrosis, histological evidence of lymphovascular invasion (LVI), variant histology (defined as micropapillary or neuroendocrine variants), or clinical T3 or T4a diseases. Whereas each of these variables have been associated with worse prognosis, their combination in a risk grouping remains to be validated externally. Therefore, our aim was to externally validate the impact of this preoperative risk grouping for the prediction of survival after RC using a tertiary referral center database.