KRT13-expressing epithelial cell population predicts better response to chemotherapy and immunotherapy in bladder cancer: Comprehensive evidences based on BCa database

https://doi.org/10.1016/j.compbiomed.2023.106795Get rights and content

Highlights

  • KRT13-expressing epithelial cell population was identified.

  • Novel molecular mechanism affecting ICB therapy outcomes was demonstrated.

  • Prognostic model using neural network was constructed.

  • A user-friendly and open-access web application named BCa database was developed.

Abstract

Neoadjuvant chemotherapy (NAC) prior to surgery and immune checkpoint therapy (ICT) has revolutionized bladder cancer (BCa) treatment. Patients likely to benefit from these therapies need to be accurately stratified; however, this remains a major clinical challenge. In the present study, single-cell RNA sequencing was used to evaluate the predictive ability of an epithelial cell population highly expressing keratin 13 (KRT13) to assess therapeutic response in BCa. The presence of KRT13-enriched tumors indicated favorable outcomes after NAC and superior response to ICT in patients with BCa. Furthermore, KRT13 population characteristics appeared to be closely related to changes in the immune microenvironment in the vicinity of this cell population. We constructed a prognostic model using an artificial neural network based on the gene signatures in the KRT13 population; the model demonstrated strong robustness and superiority. Additionally, a user-friendly and open-access web application named BCa database was developed for researchers to study BCa by mining the connective map database.

Introduction

Owing to the recent rapid advancements in the field of bladder cancer (BCa) treatment, neoadjuvant chemotherapy (NAC) and immunotherapy have opened new avenues of research and treatment; immune checkpoint therapy (ICT) in particular has revolutionized advanced BCa management [1,2]. In clinical settings, current biomarkers do not provide any clinical guidance regarding the use of NAC or ICT, thereby hindering the further development of related therapies [3,4]. Although the consensus molecular classification system has been proposed for predicting therapeutic response; however, it does not provide compelling evidence [4,5]. Therefore, new biomarkers for predicting therapeutic efficacy in BCa are greatly needed. The progressive development of high-throughput sequencing technology and emerging studies using single-cell RNA sequencing to analyze BCa have helped provide an initial understanding of intratumoral heterogeneity [[6], [7], [8], [9]]. Evaluation of characteristics related to cancer cells was hypothesized to be relatively better via scRNA seq. Therefore, this can ensure that unwanted interferences that may arise from stromal cells may be avoided, which might be beneficial for the development of more effective predictive and prognostic tools.

In this study, an epithelial cell population characterized by high expression of keratin 13 (KRT13) was demonstrated in BCa via scRNA-seq; this increased KRT13 expression showed a positive correlation with clinical outcomes after neoadjuvant chemotherapy and ICT. Besides, a prognostic artificial neural network (ANN) model was established on the bases of the overexpressed genes in the KRT13 population. To ensure data availability and contribute to the biomedical community, a user-friendly, open-access database named Bca database (http://202.114.71.211:8888/Bca/) has been developed in the present study for users to search and download data.

Section snippets

BCa datasets and resources

scRNA-seq data of patients with BCa—downloaded from GEO (accession number, GSE130001)—included 12,098 cell samples, each with 27,010 genes. BCa Illumina Hi-Seq counts from The Cancer Genome Atlas (TCGA) were downloaded from the Genomic Data Commons (GDC) data portal, and corresponding clinical annotation including survival information was accessed via the TCGA Clinical Data Resource. Affymetrix array data corresponding to a trial of neoadjuvant cisplatin-based chemotherapy in BCa was downloaded

Identification of KRT13-expressing epithelial cells

A total of 4003 single cells in the GSE130001 dataset were clustered into four major clusters after quality control. Cluster-specific genes were used to annotate cell types with classic markers described in previous studies as follows: epithelial cells (EPCAM+, KRT18+); endothelial cells (VWF+); two types of fibroblasts (COL3A1+, FGF7+)—iCAFs (PDGFRA+); and myo-CAFs (mCAFs) (RGS5+) (Fig. 1a and b and Table S1). Unsupervised clustering of the epithelial compartment alone revealed clusters that

Discussion

In the present study, the comprehensive profiling of BCa at the single cell level was performed; the outcomes of which enabled us to elucidate the constituents of current molecular subtypes and to derive molecular signatures with higher resolution and greater therapeutic relevance. All epithelial phenotypes and the KRT13 phenotype were identified subsequently. KRT13, a 54 kDa type-I keratin, was mainly expressed in the suprabasal layers of non-cornified stratified squamous epithelia [32]. The

Ethics approval and consent to participate

The present study was approved by the Clinical Research Ethics Committees of the participating institutions.

Funding

This work was supported by the Zhongnan Hospital of Wuhan University Science, Technology and Innovation Seed Fund (znpy2019077) and Translational Medicine and Interdisciplinary Research Joint Fund of Zhongnan Hospital of Wuhan University (Grant No. ZNJC202217 and ZNJC202232).

Authors' contributions

S. L., X. W., and L. C. designed this research. D. Y., C. L., and L. S. organized the processing flow. D. Y. and C. C. completed the whole analytic process of this study. D. Y., S. W., X. T., L. M., and D. W. organized and presented the results. D. Y. contributed to the writing of the manuscript. All authors read and approved the final manuscript.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used for the analyses during the current study available from the corresponding.

Declaration of competing interest

The authors declare that they have no conflict of interest.

Acknowledgements

Not applicable.

References (45)

  • A.T. Lenis et al.

    Bladder cancer: a review

    JAMA, J. Am. Med. Assoc.

    (2020)
  • J.L. Pfail et al.

    Immunotherapy in non-muscle-invasive bladder cancer: current status and future directions

    World J. Urol.

    (2021)
  • H. Wang et al.

    Single-cell analyses reveal mechanisms of cancer stem cell maintenance and epithelial-mesenchymal transition in recurrent bladder cancer

    Clin. Cancer Res.

    (2021)
  • H. Lai et al.

    Single-cell RNA sequencing reveals the epithelial cell heterogeneity and invasive subpopulation in human bladder cancer

    Int. J. Cancer

    (2021)
  • Z. Chen et al.

    Single-cell RNA sequencing highlights the role of inflammatory cancer-associated fibroblasts in bladder urothelial carcinoma

    Nat. Commun.

    (2020)
  • X. Yan et al.

    Novel T-cell signature based on cell pair algorithm predicts survival and immunotherapy response for patients with bladder urothelial carcinoma

    Front. Immunol.

    (2022)
  • IMvigor 210 trial

    Lancet Oncol.

    (2015)
  • J.H. Hoffman-Censits et al.

    IMvigor 210, a phase II trial of atezolizumab (MPDL3280A) in platinum-treated locally advanced or metastatic urothelial carcinoma (mUC)

    J. Clin. Oncol.

    (2016)
  • X. Qiu et al.

    Reversed graph embedding resolves complex single-cell trajectories

    Nat. Methods

    (2017)
  • S. Aibar et al.

    SCENIC: single-cell regulatory network inference and clustering

    Nat. Methods

    (2017)
  • D.A. Barbie et al.

    Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1

    Nature

    (2009)
  • R.D. Gill

    Multistate life-tables and regression models

    Math. Popul. Stud.

    (1992)
  • Cited by (0)

    View full text