KRT20, KRT5, ESR1 and ERBB2 Expression Can Predict Pathologic Outcome in Patients Undergoing Neoadjuvant Chemotherapy and Radical Cystectomy for Muscle-Invasive Bladder Cancer

Abstract

Patients with muscle-invasive bladder cancer (MIBC) that underwent neoadjuvant chemotherapy (NAC) prior to radical cystectomy (RC) show improved overall survival, especially those with pathological complete response (pCR). The response to NAC according to molecular subtypes has been discussed. Molecular targets such as estrogen receptor (ESR1) and human epidermal growth factor receptor 2 (ERBB2) play an important role in breast cancer management and have also been associated with urothelial bladder cancer. Hence, the association of Keratin 20 (KRT20) Keratin 5 (KRT5), ESR1, and ERBB2 mRNA expression in MIBC at transurethral resection (TUR-BT) with pCR after NAC was analyzed retrospectively. Formalin-fixed paraffin-embedded tumour tissue samples from TUR-BT of 54 patients (42 males, 12 females, median age of 64) with MIBC were analyzed for KRT20, KRT5, ESR1, and ERBB2 mRNA expression. After NAC, RC was performed, and the specimens were evaluated for pCR. Statistical analyses comprised nonparametric and chi2 testing, partition models, and Spearman correlation analyses. After NAC, 22 out of 54 patients (40.7%) had pCR. Tumours with an elevated expression of markers associated with luminal differentiation (KRT20, ERBB2, ESR1) were associated with a higher chance of pCR (55% vs. 15.8%, p = 0.009). Elevated ERBB2 expression was positively correlated with luminal expression features such as KRT20, and negatively with basal characteristics such as KRT5. Patients with MIBC showing a high expression of ERBB2, ESR1, or KRT20 have a significantly higher chance of pCR following NAC. These findings might improve patient selection for NAC in MIBC.

Molecular Subtypes as a Basis for Stratified Use of Neoadjuvant Chemotherapy for Muscle-Invasive Bladder Cancer—A Narrative Review

“…Of the three complete responders (pT0N0), two expressed CK20 highly and two expressed KRT5/6 along the edge/margin, thus suggesting one case consistent with a GU subtype and two cases consistent with a Uro subtype [ 60 ]. Similarly, Jütte and colleagues applied another four-marker qPCR panel (KRT20, ESR1, ERBB2, KRT5) to TURB tissue from 54 patients treated with two to three cycles of GC followed by RC [ 53 ]. High expression of luminal markers (KRT20, ESR1, and ERBB2) and low expression of KRT5 was significantly associated with pathologic complete response ( p = 0.009).…”

“…More studies support a good response in luminal-like subtype(s), but results are generally weak and the luminal subtype had to be inferred from markers that are not validated surrogates for subtyping. The two exceptions are the studies by Sjödahl et al in which the IHC subtyping was pre-defined with clear results on both response and survival [ 36 ], and the study by Jütte et al, which found a statistically quite robust association between the luminal-like category and response [ 53 ]. Still, it is not surprising that more heterogeneous results are found with panel-based subtyping than with RNA-based studies.…”

Development of a MRI-Based Radiomics Nomogram for Prediction of Response of Patients With Muscle-Invasive Bladder Cancer to Neoadjuvant Chemotherapy

“…In the era of precision medicine, the accuracy of prediction can only be improved via comprehensively analyzing all useful information. Previous studies have shown that molecular subtype, mRNA expression analysis, and gene mutations have the potential to predict pCR of MIBC patients to NAC ( 8 , 9 , 29 ). However, the performance of biomarkers in predicting the efficacy of NAC is not promising.…”

Prediction of Response to Cisplatin-Based Neoadjuvant Chemotherapy of Muscle-Invasive Bladder Cancer Patients by Molecular Subtyping including KRT and FGFR Target Gene Assessment

“…Here, we have used RT-qPCR-based quantitation of predefined hallmark subtyping markers (KRT5/KRT20) [ 8 ], with proven prognostic impact in muscle-invasive and non-muscle-invasive bladder cancer [ 11 , 12 ] and which have been shown to have some predictive value in a finding cohort [ 14 ]. Moreover, we have integrated FGFR1 and FGFR3mRNA expression analysis into the predictive model to evaluate the impact of stromal interactions and therapeutic implications in view of pan-FGFR inhibitors entering the field.…”

“…However, the predictive role of these markers in MIBC patients receiving neoadjuvant chemotherapy is still unknown. Early studies indicate a tremendous decrease in KRT20 mRNA levels, when comparing matched TURB and cystectomy tissue samples before vs. after neoadjuvant chemotherapy [ 14 ]. The aim of the present study was to evaluate the predictive role of KRT20 in combination with potential, druggable resistance markers in the neoadjuvant situation and to prove their clinical usefulness.…”