An integrated multi-omics analysis identifies prognostic molecular subtypes of non-muscle-invasive bladder cancer.
Lindskrog SV, Prip F, Lamy P, Taber A, Groeneveld CS, Birkenkamp-Demtröder K, Jensen JB, Strandgaard T, Nordentoft I, Christensen E, Sokac M, Birkbak NJ, Maretty L, Hermann GG, Petersen AC, Weyerer V, Grimm MO, Horstmann M, Sjödahl G, Höglund M, Steiniche T, Mogensen K, de Reyniès A, Nawroth R, Jordan B, Lin X, Dragicevic D, Ward DG, Goel A, Hurst CD, Raman JD, Warrick JI, Segersten U, Sikic D, van Kessel KEM, Maurer T, Meeks JJ, DeGraff DJ, Bryan RT, Knowles MA, Simic T, Hartmann A, Zwarthoff EC, Malmström PU, Malats N, Real FX, Dyrskjøt L. · 2021
Read An integrated multi-omics analysis identifies prognostic molecular subtypes of non-muscle-invasive bladder cancer. by Lindskrog SV, Prip F, Lamy P, Taber A, Groeneveld CS, Birkenkamp-Demtröder K, Jensen JB, Strandgaard T, Nordentoft I, Christensen E, Sokac M, Birkbak NJ, Maretty L, Hermann GG, Petersen AC, Weyerer V, Grimm MO, Horstmann M, Sjödahl G, Höglund M, Steiniche T, Mogensen K, de Reyniès A, Nawroth R, Jordan B, Lin X, Dragicevic D, Ward DG, Goel A, Hurst CD, Raman JD, Warrick JI, Segersten U, Sikic D, van Kessel KEM, Maurer T, Meeks JJ, DeGraff DJ, Bryan RT, Knowles MA, Simic T, Hartmann A, Zwarthoff EC, Malmström PU, Malats N, Real FX, Dyrskjøt L. (2021) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc8052448.txt.
Introduction
Lindskrog and co-authors sort non-muscle-invasive bladder cancer into molecular subtypes tied to outcome. Tumors that look alike under the microscope behave very differently. The team ran an integrated multi-omics analysis on 834 patients. Transcript data did the first sorting. Chromosome stability and immune infiltration added further layers.
Four transcriptomic classes emerged, labeled 1, 2a, 2b, and 3. Both the classes and chromosomal instability predicted outcome independently. Class 2a tumors carried p53 disruption and APOBEC mutations, and fared worst. Class 2b tumors drew heavy immune infiltration. More immune cells there meant fewer recurrences. The team then validated the classes in 1,228 further samples. A single-sample classifier makes the scheme usable in future trials. Surveillance and treatment could finally be tuned by subtype.
Catalog record
| CREATOR | Lindskrog SV, Prip F, Lamy P, Taber A, Groeneveld CS, Birkenkamp-Demtröder K, Jensen JB, Strandgaard T, Nordentoft I, Christensen E, Sokac M, Birkbak NJ, Maretty L, Hermann GG, Petersen AC, Weyerer V, Grimm MO, Horstmann M, Sjödahl G, Höglund M, Steiniche T, Mogensen K, de Reyniès A, Nawroth R, Jordan B, Lin X, Dragicevic D, Ward DG, Goel A, Hurst CD, Raman JD, Warrick JI, Segersten U, Sikic D, van Kessel KEM, Maurer T, Meeks JJ, DeGraff DJ, Bryan RT, Knowles MA, Simic T, Hartmann A, Zwarthoff EC, Malmström PU, Malats N, Real FX, Dyrskjøt L. |
|---|---|
| YEAR | 2021 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 107 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
An integrated multi-omics analysis identifies prognostic molecular subtypes of non-muscle-invasive bladder cancer By Sia Viborg Lindskrog, Frederik Prip, Philippe Lamy, Ann Taber, Clarice S Groeneveld, Karin Birkenkamp-Demtröder, Jørgen Bjerggaard Jensen, Trine Strandgaard, Iver Nordentoft, Emil Christensen, Mateo Sokac, Nicolai J Birkbak Published in: Nature Communications DOI: 10.1038/s41467-021-22465-w ABSTRACT Abstract The molecular landscape in non-muscle-invasive bladder cancer (NMIBC) is characterized by large biological heterogeneity with variable clinical outcomes. Here, we perform an integrative multi-omics analysis of patients diagnosed with NMIBC (n = 834). Transcriptomic analysis identifies four classes (1, 2a, 2b and 3) reflecting tumor biology and disease aggressiveness. Both transcriptome-based subtyping and the level of chromosomal instability provide independent prognostic value beyond established prognostic clinicopathological parameters. High chromosomal instability, p53-pathway disruption and APOBEC-related mutations are significantly associated with transcriptomic class 2a and poor outcome. RNA-derived immune cell infiltration is associated with chromosomally unstable tumors and enriched in class 2b. Spatial proteomics analysis confirms the higher infiltration of class 2b tumors and demonstrates an association between higher immune cell infiltration and lower recurrence rates. Finally, the independent prognostic value of the transcriptomic classes is documented in 1228 validation samples using a single sample classification tool. The classifier provides a framework for biomarker discovery and for optimizing treatment and surveillance in next-generation clinical trials. Subject terms: Cancer genomics, Cancer microenvironment, Bladder cancer, Computational biology and bioinformatics INTRODUCTION Urothelial non-muscle-invasive bladder cancer (NMIBC) represents the most common type of bladder cancer. Patients with NMIBC experience a high likelihood of disease recurrence (50–70%) and progression to muscle-invasive bladder cancer (MIBC; up to 20%, depending on stage and grade)1. Consequently, although 5-year survival rates are favorable (>90%), most patients must undergo lifelong cystoscopic surveillance and multiple therapeutic interventions, making bladder cancer the most expensive cancer to treat2. Clinically, high-risk NMIBC is treated with adjuvant intravesical instillations of Bacillus Calmette–Guérin (BCG) after surgery to eradicate residual disease and hence reduce the frequency of recurrence and progression1. Despite similar clinical and histopathological characteristics, tumors show large differences in disease aggressiveness and response to therapy, emphasizing the urgent need for further delineation of clinically useful biomarker tests to facilitate and improve patient surveillance and treatment3. Earlier studies of NMIBC biology addressed gene expression for classification of aggressiveness, resulting in the identification of two major molecular subtypes4–6. When considering the whole spectrum of disease stages, five subtypes of bladder cancer were identified; in particular, three subtypes (Urothelial-like, genomically unstable, and a group of infiltrated cases) were associated with NMIBC7. In a more recent study of 460 NMIBC patients, we reported three gene expression-based classes (class 1–3; UROMOL2016 classification system) with different clinical outcomes and molecular characteristics8. Differences in biological processes, such as cell cycle, epithelial-mesenchymal transition (EMT), and differentiation, were observed. Furthermore, mutations in well-known cancer driver genes, i.e., TP53 and ERBB2, were primarily found in high-risk class 2 tumors, together with enrichment for APOBEC-related mutational processes. Analysis of genomic alterations in NMIBC has revealed complex genomic patterns underlying bladder carcinogenesis. Activating mutations in FGFR3 and PIK3CA and chromosome 9 deletions have been identified as early disease drivers9–11. Recently, van Kessel et al. showed that NMIBC at high risk for progression could be further subdivided into good, moderate, and poor progression risk groups based on mutations in FGFR3 and methylation of GATA212. Hurst et al. assessed 160 tumors for genome-wide copy number alterations (CNAs) using array-based comparative genomic hybridization (CGH). The study included 49 high-grade T1 tumors that separated into three major genomic subgroups, one of which contained the majority of tumors showing disease progression13. In a more recent study, the same group analyzed CNAs in 140 Ta tumors and identified two major genomic subtypes (GS1 and GS2). GS1 tumors showed no or very few CNAs, while tumors in GS2 showed more alterations and a high frequency of chromosome 9 deletions14. Exome sequencing of 28 Ta tumors revealed that GS2 tumors had a higher mutational load with enrichment for APOBEC-related mutations compared to GS1 tumors. Furthermore, comparing 79 of the samples to transcriptional subtypes showed that the tumors were primarily classified as the Urothelial-like A subtype (Lund Taxonomy). Application of the UROMOL2016 classification system showed that GS2 tumors with higher genomic instability were enriched for the class 2 subtype14. However, additional refinement of these genomic studies is required to determine optimal predictors of disease aggressiveness and outcome. The tumor microenvironment has also been linked to prognosis in NMIBC. A high infiltration of cytotoxic T lymphocytes (CTLs) is associated with better prognosis in many cancer types, including MIBC15,16. In contrast, high infiltration of tumor infiltrating-lymphocytes (TILs) has been associated with progression in NMIBC17,18. Furthermore, the presence of tumor-associated macrophages and mature tumor-infiltrating dendritic cells has been related to progression of NMIBC19. The impact of regulatory T cells (Tregs) is conflicting, since high levels of Treg infiltration has been associated with both a favorable20 and unfavorable prognosis of bladder cancer21,22. The impact of immune cell infiltration on disease outcome and association with molecular subtypes and genomic alterations in NMIBC needs to be further studied. Overall, our understanding of the molecular landscape of NMIBC is still incomplete, and integrative multi-omics analysis is needed to obtain further knowledge of biological processes contributing to disease aggressiveness, recurrence, and progression. This should ultimately lead to biomarker-based optimized surveillance and therapy modalities for patients with NMIBC. Here, we report an integrative multi-omics analysis of NMIBC tumors from a total of 834 patients included in the UROMOL project. With this analysis, we delineate genomic and transcriptomic predictors of outcome in NMIBC, and present an online tool for the classification of independent samples (http://nmibc-class.dk). RESULTS Clinical, pathological, and molecular information Patients were enrolled in the UROMOL project, a European multicenter prospective study of NMIBC. The initial reports from the UROMOL project included only transcriptomic analysis6,8. We have now performed an integrated multi-omics analysis and have expanded the work to a larger NMIBC patient series with updated follow-up that is essential to acquire insight into the implications for patient management. In total, 862 tumors (613 Ta, 238 T1, 11 carcinoma in situ (CIS)) were analyzed in this study. Median follow-up for patients without progression was 49 months and 10.3% progressed to MIBC. A detailed summary of clinical and histopathological information and the analyses performed is provided in Supplementary Table 1. Delineation of transcriptomic classes in NMIBC We analyzed bulk RNA-Sequencing (RNA-Seq) data from 535 patients (397 Ta, 135 T1, 3 CIS; an expansion of the 460 NMIBC patient cohort previously analyzed8). Using unsupervised consensus clustering of gene-based expression values restricted to the 4000 genes with highest variation across the dataset23, we identified four transcriptomic classes which partially overlapped with the previous UROMOL2016 classes 1–3: high-risk class 2 was further subdivided into two subclasses, named class 2a and 2b for continuity (Fig. 1a, b). Classes showed significantly different progression-free survival (PFS, p = 6.6 × 10−5; log-rank test; Fig. 1c): patients with class 2a tumors had the worst outcome, followed by patients with class 2b tumors. Patients with class 1 tumors had the best recurrence-free survival (RFS, p = 0.025; log-rank test; Fig. 1d). Multivariable Cox regression analysis revealed that high-risk class 2a and 2b were independently associated with worse PFS and RFS when adjusted for the clinical EORTC risk