The effect of neoadjuvant chemotherapy on pathologic downstaging and survival for patients with muscle-invasive bladder cancer with mixed histological features.

Abstract

e15019 Background: Urothelial carcinoma of the bladder (UCB) with mixed histological features (MH) is thought to portend a worse prognosis when compared to pure UC, based on reports of decreased survival for pure non-UC tumors. While patients with muscle-invasive UCB have been shown to derive a survival benefit from treatment with neoadjuvant chemotherapy (NC), the response of patients with MH is unclear. We sought to examine the pathologic and clinical outcomes of treatment with NC of patients with MH versus those with pure UC. Methods: Between 1990-2011, 318 patients were identified with clinical stage T2-T4a, N0-Nx, M0-Mx UCB who had either NC (+/- radical cystectomy) (n=67) or radical cystectomy (RC) alone (n=251). Among those given NC, 58% received methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC) and 24% received gemcitabine/cisplatin. 265 patients had pure UC and 53 patients had MH, defined as UC with concomitant squamous or glandular differentiation on final pathology (biopsy histology used if pT0 or RC not performed). Endpoints included downstaging to pT0 and overall survival. Multivariable Cox regression was used to estimate the effect of histological type (MH vs. pure UC) on all-cause mortality within each treatment arm (NC vs. RC only) and to test for treatment-by-histological type interaction. Results: The rate of downstaging to pT0 was significantly higher in NC treated patients (34.3%) compared to those with RC alone (4.4%, p<0.01). When stratified by histology, downstaging to pT0 was not significantly different for those treated with NC (37.0% for pure UC vs. 23.1% for MH, p=0.34). After adjusting for age, pathologic stage, and positive surgical margins on multivariate Cox regression, there was no difference in survival outcome for those with MH vs. pure UC (HR=1.5, p=0.5), nor for those treated with NC vs. RC only (HR=2.8, p=0.1). Conclusions: Prior evidence on the benefits of NC for patients with MH is mixed, but our data suggest that there is improvement in rate of pT0 on final pathology in those treated with NC, regardless of histology. As NC is not detrimental to the overall survival of MH patients, its administration should be continued.