Abstract

Abstract Introduction: Using RNA-seq (HiSeq) transcriptome data in The Cancer Genome Atlas (TCGA) bladder cancer (BCa) dataset two molecular subtypes, basal and luminal, were identified in muscle invasive BCa (MIBCa). Recurring reported markers of the basal subtype are KRT5, KRT6A, KRT14 high and KRT20, GATA3, UPK3A, FOXA1 low. The luminal subtype is the opposite pattern of expression. Double-Negative (DN) subtype was defined as no expression of the 7 markers. Luminal, basal and DN subtypes were reported as predictors of better, poor and worst prognosis, respectively, based on overall survival (OS). EGFR is defined as a squamous differentiation marker associated with poor prognosis. The objective was to assess the clinical significance of these subtypes in TCGA data with validation in datasets from Oncomine. Methods: TCGA dataset containing 407 MIBCa patients was accessed through Xena Browser. The dataset included patient demographics, clinical parameters, OS, recurrence-free survival (RFS) and the transcript levels (log2(nor._count+1) of basal and luminal markers. All available follow-up data was included (27.9±28.49; max: 166 months). 25 BCa datasets (n=360 MIBCa patients) containing these same variables were accessed through Oncomine. High and low levels were stratified by median. Association of individual basal and luminal markers and the subtypes with clinical and outcome variables was analyzed by univariate and multivariate analyses. Kaplan-Meier analysis was performed to stratify patients into risk groups for OS and RFS. Results: In TCGA data, neither basal nor luminal markers levels significantly correlated with metastasis or lymphovascular invasion (LVI); P = 0.2 to 0.9. Only KRT5 significantly but inversely correlated with lymph node (LN) positivity; P=0.011. EGFR levels did not correlate with metastasis, LVI, or LN positivity; P > 0.05. In TCGA dataset 77 (18.9%), 50 (12.3%) and 14 (3.4%) expressed basal, luminal and DN subtypes, respectively; 266 (66.4%) patients did not conform to any group. In univariate or multivariate analyses, the subtypes also did not correlate with metastasis, LVI or LN status. Only luminal subtype associated with better OS; P=0.003. However no subtype significantly correlated with RFS. In KM analysis no subtype stratified patients regarding RFS (P>0.2). Oncomine BCa datasets validated these results. Conclusion: TCGA and Oncomine datasets show that the majority of MIBCa tissues express a mixed pattern of basal and luminal markers. Furthermore, basal, luminal or DN subtypes do not associate with clinical parameters or prognosis of MIBCa patients. Citation Format: Daley S. Morera, Daniel Belew, Andre R. Jordan, Vinata B. Lokeshwar. Molecular subtypes in muscle invasive bladder cancer: Evaluation of clinical significance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4614.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call