Impact of Molecular Subtypes in Muscle-invasive Bladder Cancer on Predicting Response and Survival after Neoadjuvant Chemotherapy

TLDR: Although cisplatin-based neoadjuvant chemotherapy improves patient outcomes, the results suggest that patients with basal tumors should be prioritized for NAC, and the first single-sample genomic subtyping classifier to subtype MIBC is discovered, which may be suitable for integration into routine clinical practice.

About: This article is published in European Urology.The article was published on 2017-10-01 and is currently open access. It has received 593 citations till now. The article focuses on the topics: Bladder cancer.

Figures

Fig. 2 – Kaplan–Meier estimates of OS according to molecular subtype in the non-NAC (left) and NAC (right) discovery datasets. (A) OS stratified according to the UNC subtypes. In the non-NAC setting (TCGA, left), patients with claudin-low and basal tumors had worse OS compared to patients with luminal tumors. The prognostic significance of each subtype changed in the context of NAC (right). While patients with claudin-low tumors still

Fig. 3 – Discovery and validation of the GSC. (A) Proposed GSC bladder cancer classes derived from a consensus of four models (UNC, MDA, TCGA, and Lund). Colors indicate each class: claudin-low (gray), basal (red), luminal-infiltrated (light blue), and luminal (dark blue). (B) GSC cross-validation scores for each sample in the NAC dataset. The vertical bands represent the probability of each sample belonging to each class. The bottom bars indicate the classes predicted by GSC as well as the consensus classes. (C) Performance of the GSC in the discovery (10-fold cross validation for model performance) and two independent validation cohorts (NAC validation cohort and non-NAC validation cohort). Across all the cohorts, GSC was able to predict all subtypes significantly with a high area under the curve (compared with consensus classes). AUC = area under the curve; CI = confidence interval; Cons. = consensus; GSC = genomic subtyping classifier; inf = infiltrated; MDA = MD Anderson Cancer Center; NAC = neoadjuvant chemotherapy; TCGA = The Cancer Genome Atlas; UNC = University of North Carolina.

Fig. 4 – Clinical significance of GSC with and without cisplatin-based NAC. (A) Kaplan–Meier plots for OS in non-NAC (left) and NAC (right) datasets stratified according to the classes predicted by GSC. (B) OS of the NAC dataset according to major pathological downstaging stratified by claudin-low (upper left), basal (upper right), luminal-infiltrated (lower left), and luminal (lower right) subtypes. GSC = genomic subtyping classifier; inf = infiltrated; NAC = neoadjuvant chemotherapy; OS = overall survival.

Fig. 1 – Molecular subtypes in pre-NAC TURBT samples in discovery NAC cohort. (A) Overview of four previously described bladder cancer subtyping methods. UNC divides the cancers in two classes, luminal and basal. Claudin [6_TD$DIFF]-low is a subtype within basal that has lost epithelial differentiation and has a high level of immune infiltration. The MDA subtyping method also divides into basal and luminal tumors, in addition to a third subtype with an

Table 2 – Results of Cox proportional hazard analysis of GSC’s ability to predict overall survival (NAC and non-NAC)