Long-term Follow-up of Bladder Cancer Patients with Disseminated Tumour Cells in Bone Marrow

Abstract

BackgroundThe clinical relevance of polymerase chain reaction (PCR)–based techniques for detection of disseminated tumour cells (DTCs) in the bone marrow of bladder cancer (BCa) patients is still under debate, as data on long-term follow-up analysis have not yet been published. ObjectiveThe aim of the present prospective study was to assess the prognostic significance of DTCs detected by cytokeratin-20 (CK20) reverse-transcriptase PCR in bone marrow from BCa patients undergoing radical cystectomy (RC). Design, setting, and participantsBone marrow samples from 51 BCa patients with high-risk non–muscle-invasive or muscle-invasive urothelial carcinoma were drawn from the anterior iliac crest prior to RC. CK20-positive cells in bone marrow were detected by qualitative RT-PCR. MeasurementsBCa patients with CK20 status were analysed with respect to the end points tumour progression and cancer death. A multivariate Cox regression analysis was performed to determine independent prognostic factors for progression-free survival (PFS), tumour-specific survival (TSS), and overall survival (OS). Results and limitationsCK20-positive cells were detected in 16 of 51 (31.4%) BCa patients of all stages. BCa patients with CK20-negative status displayed a 7-yr PFS rate of 64% versus 35.2% for CK20-positive patients (p=0.007). TSS was significantly shorter in the CK20-positive group, with a 7-yr survival rate of 46.9% compared to CK20-negative patients with 70.2% (p=0.012). The 7-yr OS rate of 37.5% for CK20-positive patients was significantly <65.7% in the CK20-negative group (p=0.006). A subgroup analysis of lymph node–negative patients (pN0) discriminated by CK20 status revealed significant differences in PFS, TSS, and OS. In a multivariate analysis, CK20-status provides independent prognostic information with respect to all three survival end points. ConclusionsBCa patients with positive CK20 status in bone marrow represent a high-risk subgroup reflected by an unfavourable outcome in the long-term analysis.