Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Meta-Analysis
. 2020 Feb;29(2):509-519.
doi: 10.1158/1055-9965.EPI-18-1359. Epub 2019 Dec 23.

The Impact of Stroma Admixture on Molecular Subtypes and Prognostic Gene Signatures in Serous Ovarian Cancer

Affiliations
Meta-Analysis

The Impact of Stroma Admixture on Molecular Subtypes and Prognostic Gene Signatures in Serous Ovarian Cancer

Matthew Schwede et al. Cancer Epidemiol Biomarkers Prev. 2020 Feb.

Abstract

Background: Recent efforts to improve outcomes for high-grade serous ovarian cancer, a leading cause of cancer death in women, have focused on identifying molecular subtypes and prognostic gene signatures, but existing subtypes have poor cross-study robustness. We tested the contribution of cell admixture in published ovarian cancer molecular subtypes and prognostic gene signatures.

Methods: Gene signatures of tumor and stroma were developed using paired microdissected tissue from two independent studies. Stromal genes were investigated in two molecular subtype classifications and 61 published gene signatures. Prognostic performance of gene signatures of stromal admixture was evaluated in 2,527 ovarian tumors (16 studies). Computational simulations of increasing stromal cell proportion were performed by mixing gene-expression profiles of paired microdissected ovarian tumor and stroma.

Results: Recently described ovarian cancer molecular subtypes are strongly associated with the cell admixture. Tumors were classified as different molecular subtypes in simulations where the percentage of stromal cells increased. Stromal gene expression in bulk tumors was associated with overall survival (hazard ratio, 1.17; 95% confidence interval, 1.11-1.23), and in one data set, increased stroma was associated with anatomic sampling location. Five published prognostic gene signatures were no longer prognostic in a multivariate model that adjusted for stromal content.

Conclusions: Cell admixture affects the interpretation and reproduction of ovarian cancer molecular subtypes and gene signatures derived from bulk tissue. Elucidating the role of stroma in the tumor microenvironment and in prognosis is important.

Impact: Single-cell analyses may be required to refine the molecular subtypes of high-grade serous ovarian cancer.

PubMed Disclaimer

Conflict of interest statement

The authors disclose no potential conflicts of interest.

Figures

Figure 1:
Figure 1:. Variability in stromal proportions may present challenges in gene expression analysis of bulk tissue.
A) If two tumors of the same molecular subtype have different proportions of stromal cells, the observed average gene expression in the tumor with more stromal cells will show an increase in stroma-associated gene expression with a concurrent decrease in tumor-associated gene expression. This apparent decrease in tumor gene expression may result in false discovery of molecular subtypes. B) Different proportions of stromal cells will increase noise in analysis, which may cause researchers not to discover underlying molecular subtypes.
Figure 2:
Figure 2:. AOCS molecular subtypes can be distinguished by stromal and tumor gene expression.
Analysis of gene expression profiles of microdissected epithelial tumor cells and stroma from four C1 tumors identified a 688-gene ovarian tumor-stroma gene signature (FDR-adjusted p<0.01). The signature contained 461 genes that were over-expressed in microdissected stroma and 227 genes with increased expression in epithelial tumor cells, and these distinguished ovarian tumor and stroma in an independent MGH dataset (Figure S2). The heatmap shows the tumor and stromal gene expression profiles in the AOCS’s HGSOC tumors and microdissected C1 stroma and tumor samples. C1 microdissected stroma samples clustered with C1 subtype tumors, while the C1 microdissected tumor samples clustered with C4 subtype tumors. All of these analyses were done using unsupervised hierarchical cluster analysis. Colors of the heatmap were scaled (z-score values across rows).
Figure 3:
Figure 3:. Overlap between genes that are differentially expressed between the AOCS and TCGA molecular subtypes and tumor-stroma genes sets.
Black and white lines indicate genes that were or were not strongly differentially expressed (limma, moderated t-statistics, p<0.001 after FDR correction), respectively, between each pair of A) AOCS and B) TCGA molecular subtypes. The color bar indicates whether these genes were also differentially expressed in laser capture microdissected C1 tumor (yellow) and stroma (green) at p<0.05, or not significantly differentially expressed between C1 tumor and stroma (orange). Stromal and tumor genes with cutoff p<0.05 rather than 0.01 were used in order to explore the extent of stromal gene enrichment in these subtypes. C) Correlation between average expression of AOCS and TCGA molecular subtypes, which was calculated using the intersection of all differentially expressed genes between any pair of AOCS or TCGA subtypes. Mes, Imm, Diff, and Pro are the TCGA molecular subtypes mesenchymal, immunoreactive, differentiated, and proliferative, respectively. D) There was a large overlap between genes expressed in C1 microdissected stroma and genes that were differentially expressed either between C1 vs. other AOCS subtypes, or between the TCGA mesenchymal subtype vs. other TCGA subtypes.
Figure 4:
Figure 4:. Stromal content’s association with overall survival.
These forest plots examine association with overall survival across ovarian cancer studies (n=16) using A) the AOCS tumor and stromal gene signatures or B) only the genes that were specific to tumor-associated stroma from the MGH study. The two TCGA gene expression datasets are n=510 Affymetrix HT_HG-U133A and n=256 Illumina HiSeq RNA sequencing. C) Kaplan-Meier curves of pathologist assessment of percent stroma (cutoff of 30%) in high stage TCGA tumors (HR 1.01 for each percent increase in stromal content, CI 1.0–1.02, p=0.024). A 30% stroma threshold is shown since it is between the original TCGA consortium tumor purity threshold of at least 80% and the later reduced threshold of 60% tumor cells.

Similar articles

Cited by

References

    1. Tothill RW, Tinker AV, George J, Brown R, Fox SB, Lade S, et al. Novel molecular subtypes of serous and endometrioid ovarian cancer linked to clinical outcome. Clin Cancer Res Off J Am Assoc Cancer Res. 2008;14:5198–208. - PubMed
    1. Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474:609–15. - PMC - PubMed
    1. Waldron L, Haibe-Kains B, Culhane AC, Riester M, Ding J, Wang XV, et al. Comparative meta-analysis of prognostic gene signatures for late-stage ovarian cancer. J Natl Cancer Inst. 2014;106. - PMC - PubMed
    1. Riester M, Wei W, Waldron L, Culhane AC, Trippa L, Oliva E, et al. Risk prediction for late-stage ovarian cancer by meta-analysis of 1525 patient samples. J Natl Cancer Inst. 2014;106. - PMC - PubMed
    1. Waldron L, Riester M, Birrer M. Molecular subtypes of high-grade serous ovarian cancer: the holy grail? J Natl Cancer Inst. 2014;106. - PubMed

Publication types

Substances