Patterns of Mutation Enrichment in Metastatic Triple-Negative Breast Cancer   

Show less

César H Saravia, Claudio Flores, Luis J Schwarz, 

Leny Bravo, Jenny Zavaleta, Jhajaira Araujo, Silvia Neciosup, Joseph A Pinto

First Published August 9, 2019 Research Article 

Triple-negative breast cancer (TNBC) is a heterogeneous disease with aggressive biology and complex tumor evolution. Our purpose was to identify enrichment patterns of genomic alterations in metastatic triple-negative breast cancer (mTNBC).

Methods:

Genomic data were retrieved (mutations and copy number variations) from 550 primary TNBC tumors from the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) and The Cancer Genome Atlas (TCGA) data sets and 58 mTNBC tumors from “Mutational Profile of Metastatic Breast Cancers” and “The Metastatic Breast Cancer Project.” Statistical analysis of microarray data between primary and metastatic tumors was performed using a chi-square test, and the percentage of mutation enrichment in mTNBC cases was estimated. P-values were adjusted for multiple testing with Benjamini-Hochberg method with a false-discovery rate (FDR) <.05. In addition, we identified dominant hallmarks of cancer in mTNBC.

Results:

Seven genes with mutations were enriched in mTNBC after correcting for multiple testing. These included TTN, HMCN1, RELN, PKHD1L1, DMD, FRAS1, and RYR3. Only RPS6KB2 amplification was statistically significant in mTNBC; on the contrary, deletion of the genes TET1, RHOA, EPHA5, SET, KCNJ5, ABCG4, NKX3-1, SDHB, IGF2, and BRCA1 were the most frequent. The molecular alterations related to the hallmark of “genetic instability and mutation” were predominant in mTNBC. Interestingly, the hallmark of “activating immune destruction” was the least represented in mTNBC.

Conclusion:

Despite the study limitations, we identified recurrent patterns of genomic alterations with potential contribution to tumor evolution. Deletions were the aberrations more commonly found in mTNBC. Several molecular alterations are potentially targetable.

Keywords Triple-negative breast cancer, hallmarks of cancer, biomarker, copy number variation, mutation, amplification, deletion

Introduction

Triple-negative breast cancer (TNBC) corresponds to a group of heterogeneous and highly aggressive tumors characterized by the lack of expression of estrogen receptor (ER), progesterone receptor (PR), and HER2.¹ Despite the biological aggressiveness, these tumors present up to 30% of pathological complete responses after neoadjuvant chemotherapy, conferring good prognosis to patients, while those cases with residual diseases have the worse outcomes.²

Triple-negative breast cancers can be classified based on their gene-expression profiling in six distinct subtypes: basal-like 1 (BL1), basal-like 2 (BL2), immunomodulatory (IM), mesenchymal (M), mesenchymal stem-like (MSL), and a luminal androgen receptor (LAR) subtype;³ however, the subtypes IM and MSL have an mRNA background from immune and stromal cells, respectively, redefining the TNBC subtypes to only four groups: BL1, BL2, M, and LAR.⁴

In a recent study, Bareche et al, reported that the TNBC subtypes have diverse molecular patterns with substantial differences in activating the hallmarks of cancer, while the BL1 subtype is genetically unstable, LAR tumors have a higher mutational burden. Mesenchymal and MSL subtypes have an increased angiogenesis, and IM present higher expression of immune checkpoint genes.⁵

Triple-negative breast cancer has an intriguing molecular biology. These tumors present ≈80% of mutations in TP53, in sharp contrast to the luminal subtype (10%-30%); one consequence of the aforementioned is the high genomic instability in TNBC associated with a wide and continuous spectrum of clonal evolution distinct to other breast cancer subtypes.^6,7 Single-cell sequencing studies showed the patterns of clonal evolution of TNBC, where preexisting resistant cells undergo natural selection after chemotherapy accompanied by transcriptional reprogramming.⁸ In this way, copy number variations (CNVs) seem to be an early event during tumor evolution, present even before chemotherapy exposition.⁹

We conducted this study to identify patterns of enrichment of gene alterations through metastases in TNBC and then evaluate how these changes activate specific hallmarks of cancer during tumor evolution.

Methods

Data sets for primary and metastatic tumors

We retrieved genomic data regarding mutations and CNV of 550 primary triple-negative breast cancer (pTNBC) tumors from the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) and The Cancer Genome Atlas (TCGA) data sets and 58 metastatic triple-negative breast cancer (mTNBC) tumors from the “Mutational Profile of Metastatic Breast Cancers” and “The Metastatic Breast Cancer Project.” The genomic data of the cohorts was retrieved from http://www.cbioportal.org/.

Identification of enriched genes in metastatic tumors

Differences in proportions of alterations (mutations and CNV) between primary versus metastatic tumors were evaluated by chi-square test, and the percentage of enrichment of alteration in each gene was estimated using the following formula

$$\mathrm{E}\mathrm{n}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{h}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}=\frac{\%\mathrm{i}\mathrm{n}\mathrm{m}\mathrm{T}\mathrm{N}\mathrm{B}\mathrm{C}-\%\mathrm{i}\mathrm{n}\mathrm{p}\mathrm{T}\mathrm{N}\mathrm{B}\mathrm{C}}{\%\mathrm{i}\mathrm{n}\mathrm{p}\mathrm{T}\mathrm{N}\mathrm{B}\mathrm{C}}\times 100$$

P-values were adjusted for multiple testing with Benjamini-Hochberg method considering a false-discovery rate (FDR) <.05.

Analysis of hallmarks of cancer

The influence of altered genes enriched in metastatic conditions in the hallmarks of cancer was analyzed through normalized pointwise mutual information (NPMI) of the Cancer Hallmarks Analytics Tool (CHAT).¹⁰The analysis was conducted on an online platform (http://chat.lionproject.net/).

Protein-protein interaction network

We conducted an analysis for protein-protein interaction of the altered genes using the online platform STRING (Search Tool for the Retrieval of Interacting Genes/Proteins, version 10.5, STRING Consortium©) available at http://string-db.org. The analysis was performed with a medium confidence level (0.4). A green line indicates activation; red line, inhibition; blue line, binding; pink line, posttranslational modifications; and yellow line, expression.

Results

Mutations enriched in mTNBC

In total, 135 mutations were present in both primary and metastatic tumors, and 20 genes had a P-value < .05 in the analysis using chi-square test. After adjusting for multiple comparisons, only mutations in seven genes were statistically significant in mTNBC. These genes included TTN, HMCN1, RELN, PKHD1L1, DMD, FRAS1, and RYR3 (Table 1).

Table 1. Mutations enriched in mTNBC.

Table 1. Mutations enriched in mTNBC.

View larger version

Copy number variations enriched in mTNBC

In total, 661 amplifications were present in both, the primary and metastatic TNBC. Chi-square test identified 17 genes statistically different between the two groups. Only RPS6KB2 was significantly enriched after adjusting for multiple comparisons. On the contrary, 344 genes with deletions were identified, and from them, 94 presented relevant differences between primary and metastatic tumors. Finally, after adjusting for multiple comparisons, 76 deletions were significant in mTNBC including TET1, RHOA, EPHA5, SET, KCNJ5, NKX3-1, ABCG4, SDHB, IGF2, BRCA1, SESN3, PMS2, CSF1R, CD74, TSC1, SKP2, PLCG2, PIK3C2G, EIF4A2, ZNF331, FNBP1, ASXL3, PDGFRB, U2AF2, PBRM1, KMT2A, EPHB3, IL7R, DKK1, NUP214, SETD2, DDX6, NAV3, RPS14, ETV5, ABL1, MAP3K13, FLT4, FANCM, TOP1, TFPT, SLC34A2, FLI1, ETS1, NUP98, SPEN, CARS, EPHA2, MKI67, DNAH12, ARHGEF12, TEK, PRDM2, EPHA8, CDC42, CBL, DDX10, TICAM1, KDM4C, MEF2A, NCKIPSD, ZBTB16, EPHB2, FZD10, CHEK1, PAX7, ATM, RICTOR, AKT1, EGFR, AHNAK2, ZNF300, ACVR1B, IGF1R, PHOX2B, and LRP1B (Table 2).

Table 2. Copy number variations enriched in TNBC.

Table 2. Copy number variations enriched in TNBC.

View larger version

Influence of alterations in hallmarks of cancer

The analysis of the influence of enriched alterations in mTNBC on the hallmarks of cancer showed that the hallmark “genome instability and mutation” was extensively overrepresented followed by “sustaining proliferative signaling.” Paradoxically, the hallmark “avoiding immune destruction” was the least represented (Figure 1).

Figure 1. Hallmarks of cancer enriched in mTNBC.

Values of normalized pointwise mutual information (NPMI) were estimated using the software Cancer Hallmarks Analytical Tool.

Abbreviations: AID, avoiding immune destruction; EGS, evading growth suppressors; GIM, genome instability and mutation; IA, inducing angiogenesis; IM, Invasion and metastasis; RCD, resisting cell death; RI, replicative immortality; SPS; sustaining proliferative signaling; TPI, tumor promoting inflammation.

Protein-protein interaction between genes presenting alterations enriched in mTNBC

We observed that there was an absence of interaction between several genes; however, three clusters were identified. The first group involves genes related to DNA repair, where ATM is the central node. The second group has genes related to cell metabolism, such as AKT1, RICTOR, TSC1 among others. Tyrosine kinase proteins (receptors and no receptors) were part of the third cluster, containing genes such as EGFR, PDGFR, IGF1R ABL1 and ephrin (Eph) receptor subfamily (Figure 2).

Figure 2. Prediction of protein-protein interaction among gene alterations significantly enriched in mTNBC.

Discussion

Clonal evolution of TNBC is an intriguing phenomenon where patterns of evolution using single-cell sequencing have been previously unveiled; however, there are several unanswered questions about the key mechanisms favored during metastases.^8,11

In this work, we found molecular mechanisms that are enriched in the metastatic setting. The main limitation of the study is we evaluated changes in mutational rates between primary versus metastatic cohorts rather analyze a longitudinal cohort. Although single-cell sequencing could be a better approach to evaluate clonal evolution, our approach comprehends genes, hallmarks, and pathways significantly altered in mTNBC.

The evaluation of new activating mutations in cancer is challenging because of the high rate of false-positive results.¹² In our study, mutations in the TTN gene (whose germline mutations are associated to familial restrictive cardiomyopathy) were the most frequently mutated gene in mTNBC, although it is commonly described as a false-positive finding. TTN encodes for the giant sarcomeric filament Titin.¹³ Involvement of TTN in cancer remains unclear and several studies suggested it has not a participation in tumorigenesis or cancer progression while other studies suggest a direct involvement in cancer-related pathways.^14-16 TTNalterations are more recurrent in TNBC than other breast cancer subtypes.¹⁷

Regarding HMCN1, although its specific function remains unknown, it has been related to cancer cell invasion and metastasis. Interestingly, the intratumor heterogeneity of HMCN1 is associated with the prognostic of breast cancer.¹⁸ On the contrary, RELN is involved in cell migration where a low expression of this gene is associated with poor outcome in breast malignancies.^19,20

Amplifications were less observed than gene deletions. The RPS6KB2 gene amplification was enriched in mTNBC. RPS6KB2 encodes the S6K2 protein, an effector from the mammalian target of rapamycin (mTOR) signaling pathway, promoting protein synthesis and cell proliferation. In altered states, S6K2 produces aberrant mTORC1 function, thus inducing tumorigenesis.²¹ A high number of deletions (n = 76) were found in mTNBC. It describes the greater importance of deletions over amplifications. In our analysis, gene deletions were around six-fold higher in mTNBC than in pTNBC (pTNBC: 0.4%-1.3% vs mTNBC: 3.4%-8.6%; Table 2). Our data contrast with the results of Gao et al,⁹ whose work suggests that CNVs are an early event during tumor evolution. Deletion of key genes are important for tumor progression and metastases. In our work, TET1, a gene that encodes a cytosine demethylase, was the most frequently deleted in mTNBC (8.6%). TET1 expression decreases cell invasion and tumor formation in human breast cancer cell line xenografts, inhibiting DNA methylation of tissue inhibitors of the metalloproteinase family proteins.²²

The hallmarks of cancer describe 10 biological capabilities from tumors acquired during their evolution.²³ In mTNBC, genes whose alterations were more recurrent are related to the hallmark “genomic instability and mutation,” this is explained by the high incidence of TP53 and BRCA1 mutations in pTNBC.³ On the contrary, the hallmark “avoiding immune destruction” was the least represented among gene alterations in mTNBC (Figure 1). Our results suggest that the activation of this hallmark is an early event that does not need to be significantly increased during metastases.

Finally, our data of protein-protein interaction indicates that at least three gene clusters with alterations are prevalent in mTNBC (Figure 2). These clusters involve genes related to DNA repair, in concordance with the overrepresentation of the hallmark of “genomic instability and mutation.” In addition, evidence of increased metabolism is observed through the enrichment of alterations in genes participating in the mTOR signaling pathway. The third cluster of genes involves well-known tyrosine kinase receptors such as EGFR, PDGFR, and others.

The presence of genes encoding Eph receptors in the interactions is interesting. In our analysis, we observed that the loss of this family of receptors is common in mTNBC. Although typically, Eph receptors promote cell division and are commonly overexpressed in human tumors and have a dual role in tumor promotion and suppression.^24,25

In conclusion, despite the limitations of this study, we could identify genes and potential mechanisms involved in TNBC evolution through metastasis. Our analysis identified that deletions were the most frequently enriched alterations and also suggests that some biological mechanisms are potentially targetable.

Funding:
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The publication of this article was supported by a Grant of Oncosalud-AUNA.

Declaration of Conflicting Interests:
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Contributions
Study design: C.H.S, C.F, L.J.S and J.A.P. Data collection, data assembly and data preprocessing: C.H.S, L.B, J.Z, J.A. Statistical Analysis: C.F. and S.N. Data interpretation: all authors. Writing of Manuscript: All authors. Preparation of tables and figures: C.H., L.B, J.Z and J.A. All authors reviewed and approved the manuscript.

ORCID iDs
César H Saravia  https://orcid.org/0000-0002-4734-3145

Jhajaira Araujo  https://orcid.org/0000-0002-9639-8070

Joseph A Pinto  https://orcid.org/0000-0002-7744-1635

References

1.	Dent, R, Trudeau, M, Pritchard, KI. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007;13:4429-4434. doi:10.1158/1078-0432.CCR-06-3045. Google Scholar | Crossref | ISI
2.	Carey, LA, Dees, EC, Sawyer, L. The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res. 2007;13:2329-2334. doi:10.1158/1078-0432.CCR-06-1109. Google Scholar | Crossref | ISI
3.	Lehmann, BD, Bauer, JA, Chen, X. Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest. 2011;121:2750-2767. doi:10.1172/JCI45014. Google Scholar | Crossref | ISI
4.	Lehmann, BD, Jovanović, B, Chen, X. Refinement of triple-negative breast cancer molecular subtypes: implications for neoadjuvant chemotherapy selection. PLoS ONE. 2016;11:e0157368. doi:10.1371/journal.pone.0157368. Google Scholar | Crossref
5.	Bareche, Y, Venet, D, Ignatiadis, M. Unravelling triple-negative breast cancer molecular heterogeneity using an integrative multiomic analysis. Ann Oncol. 2018;29:895-902. doi:10.1093/annonc/mdy024. Google Scholar | Crossref
6.	Cancer Genome Atlas Network . Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61-70. doi:10.1038/nature11412. Google Scholar | Crossref | ISI
7.	Shah, SP, Roth, A, Goya, R. The clonal and mutational evolution spectrum of primary triple-negative breast cancers. Nature. 2012;486:395-399. doi:10.1038/nature10933. Google Scholar | Crossref | ISI
8.	Kim, C, Gao, R, Sei, E. Chemoresistance evolution in triple-negative breast cancer delineated by single-cell sequencing. Cell. 2018;173:879-893.e13. doi:10.1016/j.cell.2018.03.041. Google Scholar | Crossref
9.	Gao, R, Davis, A, McDonald, TO. Punctuated copy number evolution and clonal stasis in triple-negative breast cancer. Nat Genet. 2016;48:1119-1130. doi:10.1038/ng.3641. Google Scholar | Crossref
10.	Baker, S, Ali, I, Silins, I. Cancer Hallmarks Analytics Tool (CHAT): a text mining approach to organize and evaluate scientific literature on cancer. Bioinformatics. 2017;33:3973-3981. doi:10.1093/bioinformatics/btx454. Google Scholar | Crossref
11.	Patel, N, Weekes, D, Drosopoulos, K. Integrated genomics and functional validation identifies malignant cell specific dependencies in triple negative breast cancer. Nat Commun. 2018;9:1044. doi:10.1038/s41467-018-03283-z. Google Scholar | Crossref
12.	Lawrence, MS, Stojanov, P, Polak, P. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature. 2013;499:214-218. doi:10.1038/nature12213. Google Scholar | Crossref | ISI
13.	Peled, Y, Gramlich, M, Yoskovitz, G. Titin mutation in familial restrictive cardiomyopathy. Int J Cardiol. 2014;171:24-30. doi:10.1016/j.ijcard.2013.11.037. Google Scholar | Crossref
14.	Cannataro, VL, Gaffney, SG, Townsend, JP. Effect sizes of somatic mutations in cancer. J Natl Cancer Inst. 2018;110:1171-1177. doi:10.1093/jnci/djy168. Google Scholar | Crossref
15.	Stephens, PJ, Tarpey, PS, Davies, H. The landscape of cancer genes and mutational processes in breast cancer. Nature. 2012;486:400-404. doi:10.1038/nature11017. Google Scholar | Crossref | ISI
16.	Rajendran, BK, Deng, C-X. Characterization of potential driver mutations involved in human breast cancer by computational approaches. Oncotarget. 2017;8:50252-50272. doi:10.18632/oncotarget.17225. Google Scholar | Crossref
17.	Ademuyiwa, FO, Tao, Y, Luo, J, Weilbaecher, K, Ma, CX. Differences in the mutational landscape of triple-negative breast cancer in African Americans and Caucasians. Breast Cancer Res Treat. 2017;161:491-499. doi:10.1007/s10549-016-4062-y. Google Scholar | Crossref
18.	Kikutake, C, Yoshihara, M, Sato, T, Saito, D, Suyama, M. Intratumor heterogeneity of HMCN1 mutant alleles associated with poor prognosis in patients with breast cancer. Oncotarget. 2018;9:33337-33347. doi:10.18632/oncotarget.26071. Google Scholar | Crossref
19.	Stein, T, Cosimo, E, Yu, X. Loss of reelin expression in breast cancer is epigenetically controlled and associated with poor prognosis. Am J Pathol. 2010;177:2323-2333. doi:10.2353/ajpath.2010.100209. Google Scholar | Crossref
20.	Castellano, E, Molina-Arcas, M, Krygowska, AA. RAS signalling through PI3-kinase controls cell migration via modulation of reelin expression. Nat Commun. 2016;7:11245. doi:10.1038/ncomms11245. Google Scholar | Crossref
21.	Nardella, C, Lunardi, A, Fedele, G. Differential expression of S6K2 dictates tissue-specific requirement for S6K1 in mediating aberrant mTORC1 signaling and tumorigenesis. Cancer Res. 2011;71:3669-3675. doi:10.1158/0008-5472.CAN-10-3962. Google Scholar | Crossref
22.	Hsu, C-H, Peng, K-L, Kang, M-L. TET1 suppresses cancer invasion by activating the tissue inhibitors of metalloproteinases. Cell Rep. 2012;2:568-579. doi:10.1016/j.celrep.2012.08.030. Google Scholar | Crossref
23.	Hanahan, D, Weinberg, RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646-674. doi:10.1016/j.cell.2011.02.013. Google Scholar | Crossref | ISI
24.	Chen, J, Zhuang, G, Frieden, L, Debinski, W. Eph receptors and ephrins in cancer: common themes and controversies. Cancer Res. 2008;68:10031-10033. doi:10.1158/0008-5472.CAN-08-3010. Google Scholar | Crossref
25.	Xi, H-Q, Wu, X-S, Wei, B, Chen, L. Eph receptors and ephrins as targets for cancer therapy. J Cell Mol Med. 2012;16:2894-2909. doi:10.1111/j.1582-4934.2012.01612.x. Google Scholar | Crossref | ISI

View Abstract