ABSTRACT
Purpose
This study aimed to identify breast cancer‐specific circulating tumor DNA (ctDNA) methylation markers that correspond to tissue DNA methylation.
Methods
Using The Cancer Genome Atlas (TCGA) database, we selected breast cancer‐specific DNA methylation markers. The methylation and expression patterns of candidate genes were analyzed in breast cancer cell lines and tissue samples. We also assessed the methylation status in ctDNA obtained from breast cancer patients and examined associations with the clinicopathological features.
Results
Among candidate genes with breast cancer‐specific methylation patterns, USP44, ZNF454, and GPRC5B were selected. The methylation status and expression of selected genes varied by molecular subtype of cancer in the cell line. In tissue samples, expression of all three genes was generally lower in breast cancer than in controls. ctDNA methylation patterns showed no significant change before and after treatment for each candidate gene. Correlations between gene expression and DNA methylation status or clinicopathological characteristics in cancer tissues differed among genes.
Conclusion
Further studies are needed for clinical application of liquid biopsy using methylation analysis for ctDNA according to individual characteristics for breast cancer.
Keywords: breast cancer, ctDNA, differential methylation, DNA methylation, liquid biopsy
We selected breast cancer‐specific DNA methylation markers using The Cancer Genome Atlas (TCGA). We analyzed the methylation and expression patterns of candidate genes in breast cancer cell lines, ctDNA, and tissue samples. The relationship between ctDNA methylation markers and the clinicopathological features of breast cancer patients was investigated.
1. Introduction
Breast cancer is one of the most prevalent malignancies affecting women in Korea and worldwide [1, 2]. Early detection and diagnosis are important for improving outcomes in breast cancer patients. With recent advances in diagnostic techniques and treatments, both early diagnosis rates and the survival rates of breast cancer patients have increased [3, 4]. Nevertheless, there are still patients who show different treatment responses according to tumor heterogeneity and molecular biological characteristics of breast cancer. Therefore, the need for research on diagnostic methods and treatments suitable for individual cancer characteristics cannot be overlooked.
Tissue biopsy remains the gold standard for cancer diagnosis, as it involves the direct collection of tumor tissue. Despite its diagnostic value, tissue biopsy is invasive, often unsuitable for repeated testing, and sometimes impractical depending on the tumor's location. Due to these limitations, interest in liquid biopsy has increased [5, 6]. Liquid biopsy is a method of diagnosing cancer or using it to treat cancer by analyzing biomarkers such as circulating tumor cells (CTCs) and circulating tumor DNA (ctDNA) in the blood and possibly in other body fluids of patients [5, 7]. Specifically, ctDNA has emerged as a promising noninvasive cancer biomarker due to its ability to reflect the genetic and epigenetic landscape of tumors [8, 9, 10].
Recent studies have focused on detecting and analyzing CTCs and ctDNA, aiming to integrate liquid biopsy into clinical practice for cancers [5, 6, 11]. For instance, in June 2016, the US FDA approved a blood‐based test to detect EGFR mutations in non‐small cell lung cancer, enabling targeted therapy based on plasma DNA results [6]. Numerous studies have reported correlations between CTCs or ctDNA and cancer outcomes, further supporting the potential of liquid biopsy in clinical settings [5, 10, 12, 13, 14].
Despites the advantages of liquid biopsy, its clinical application in breast cancer remains limited due to a lack of standardized testing methods and protocols [15]. Further research is required to establish validated and reproducible approaches for clinical application.
It is well known that epigenetic mechanisms play an important role in the development and progression of cancer. Among various epigenetic modifications, DNA methylation is a crucial regulatory mechanism often dysregulated in cancer, leading to changes in gene expression that contribute to tumorigenesis [16]. While the methylation of cancer‐related genes has been extensively studied, the application of these findings to liquid biopsy is still emerging. Notably, DNA methylation analysis of ctDNA has shown promise in the diagnosis of liver and lung cancers [17, 18], but research on ctDNA methylation in breast cancer is insufficient.
The aim of this study was to discover breast cancer‐specific ctDNA methylation markers that are consistent with tissue DNA methylation. We analyzed the methylation and expression patterns of candidate genes in breast cancer cell lines and tissue samples. Additionally, we compared the methylation status of these genes in ctDNA obtained from breast cancer patients at different stages of treatment. The relationship between ctDNA methylation markers and the clinicopathological features of breast cancer patients was investigated, thereby assessing the potential of these markers for clinical applications in disease monitoring and prognosis.
2. Materials and Methods
2.1. Selection of Candidate Genes in the TCGA Database
To select cancer‐specific DNA methylation markers for breast cancer, we analyzed publicly available databases including DNA methylation and gene expression data derived from The Cancer Genome Atlas (TCGA) (https://portal.gdc.cancer.gov/). We used a web‐based tool, the SMART (Shiny Methylation Analysis Resource Tool) App (http://www.bioinfo‐zs.com/smartapp), to analyze DNA methylation and its association with other omics data [19]. The difference in beta value between tumor and adjacent normal tissues was set at ≥0.4, and FDR‐adjusted p‐value was <0.01. The data of 493 breast cancer tissues that show differentially methylated CpG sites compared to normal tissue were obtained. Gene expression data were combined with these data to identify differentially expressed genes. Finally, the three most relevant genes that have an inverse correlation between methylation and expression and show significant negative correlation coefficients were selected.
2.2. Study Participants and Sample Collection
We enrolled patients who were diagnosed with breast cancer and planned treatment at Daegu Catholic University Hospital, Daegu, Republic of Korea, and a total of 38 patients were included in the analysis. Ethics approval for the study was obtained from the Institutional Review Board at the Daegu Catholic University Hospital (CR‐20‐198). Breast cancer tissues for the study were collected from parts of the resected cancer tissue immediately after surgical resection of the breast cancer, placed in RNA‐later (AM7021, Invitrogen), and stored at −80°C. Eleven patients who underwent surgical excision for benign breast disease were enrolled as controls. All tissue samples of the participants were fixed in formalin and embedded in paraffin (FFPE), and then stained with hematoxylin and eosin for further evaluation. For ctDNA analysis, 10–20 mL of blood was collected in Streck Cell‐Free DNA BCT (Streck, La Vista, NE, USA) from the participants at a time, once before treatment and once at 6 months after treatment. Whole blood samples were centrifuged at 1600 × g for 10 min for phase separation. Plasma was collected and submitted to a second centrifugation step at 3000 × g for 10 min to remove platelets and cell debris. Upper phase was collected with cryogenic tube and stored at −80°C until further processing.
2.3. Cell Line Cultures
A panel of breast cancer cell lines was selected to represent various subtypes of breast cancer. These included cell lines such as MCF‐7, luminal A subtype; BT‐474, Luminal B subtype; SKBR3, HER2‐positive subtype; and BT‐20, MDA‐MB‐231, and BT549, triple‐negative subtype. Most of the breast cancer cell lines (MCF‐7, BT‐474, SKBR3, BT‐20, MDA‐MB‐231) were obtained from the Korean Cell Line Bank (Seoul, Republic of Korea), and an additional breast cancer cell line, BT549, was purchased from the American Type Culture Collection (Manassas, VA, USA). White blood cells (WBCs) were cultured as the control group. Prior to DNA and RNA extraction, cells were washed with cold phosphate‐buffered saline to remove any residual media and then processed immediately to minimize degradation of nucleic acids.
2.4. Quantitative Reverse Transcription‐Polymerase Chain Reaction (qRT‐PCR)
RNA was extracted from cultured cells using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and High Pure RNA isolation kit (Roche, Mannheim, Germany) according to the manufacturer's protocol. The quantity of the isolated RNAs was assessed using an ND‐2000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE, USA). A total of 1 µg of RNA was used to synthesize cDNA using the Transcriptor First‐Strand cDNA Synthesis Kit (Roche, Mannheim, Germany) according to the manufacturer's protocol. RT‐PCR was performed using the Roche LightCycler 480 II system (Roche, Mannheim, Germany). PCR products were verified by agarose gel electrophoresis, and relative mRNA levels were normalized to housekeeping genes. Data were analyzed using the comparative Ct method (ΔΔCt). Primers were designed by using Primer Premiere version 6.25 software (Premier Biosoft International, Palo Alto, CA, USA). Primers for each gene used in this study are described in Table S1.
2.5. Tissue Microarrays (TMAs) and Immunohistochemistry (IHC)
Representative tumor cores were selected from the FFPE tumor blocks and used to construct TMAs according to the method described in our previous study [20]. Immunohistochemical analysis was performed using the Bond Polymer Refine Detection system (Leica Microsystems, Mount Waverley, Victoria, Australia). The primary antibodies used in this study were as follows: USP44 (LS‐C336416, LSBio), ZNF454 (Cat. No. NBP1‐81948, Novus Biologicals), GPRC5B (Cat. No. PA5‐32853, Invitrogen), ER (1:100, clone 6F11; Novocastra), PR (1:100, clone 16; Novocastra), HER2 (1:250, A0485; Dako), Ki‐67 (1:200, MM1‐L; Novocastra), Bcl‐2 (1:4, clone 124; Dako), p53 (1:200, BP53.12; Zymed), and epidermal growth factor receptor (EGFR) (1:100, clone EGFR.25; Novocastra).
2.6. ctDNA Isolation
For ctDNA isolation, frozen plasma samples were thawed in the refrigerator at 4°C. Once thawed, a new centrifugation was performed at 16,000 × g for 5 min at 4°C to ensure removal of impurities in the supernatant. ctDNA was isolated using QIAamp Circulating Nucleic Acid Kit (Qiagen, Hilden, Germany) according to the manufacturer's instruction. The concentration of the isolated ctDNA was quantified using an ND‐2000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE, USA).
2.7. Methylation‐Specific PCR (MSP)
Methylation status of cancer cell lines and ctDNA was analyzed using MSP.
Genomic DNA from cultured cell lines was extracted using Blood & Cell Culture DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instruction. Sodium bisulfite modification of 100 ng genomic DNA was performed using the EZ DNA Methylation‐Gold Kit (Zymo Research, Orange, CA, USA) according to the manufacturer's protocol.
MSP was performed using the Roche LightCycler 480 II system (Roche, Mannheim, Germany). MSP products were analyzed by agarose gel electrophoresis and visualized by ethidium bromide staining. Data with the comparative Ct method were analyzed to compare relative methylation values.
We used the nested PCR technique to increase the sensitivity and specificity of MSP for ctDNA. For MSP of ctDNA, isolated ctDNA from patients’ plasma samples was denatured and amplified using PCR. Then, 100 ng of genomic DNA was treated with sodium bisulfite using the EZ DNA Methylation‐Gold Kit (Zymo Research, Orange, CA, USA). Primers for MSP and nested PCR used in this study are described in Table S1.
2.8. Pyrosequencing
Pyrosequencing was used to analyze the methylation status of each gene in the patients’ tumor tissues. Genomic DNA was extracted from the FFPE tumor tissue sections using the ReliaPrep FFPE gDNA Miniprep System (Promega, Madison, WI, USA) following the manufacturer's instructions. The protocol of pyrosequencing was modified from the method described in our previous study [20]. Sodium bisulfite modification of 200 ng genomic DNA was performed using the EZ DNA Methylation‐Gold Kit (Zymo Research, Orange, CA, USA) according to the manufacturer's protocol. Primers were designed using the PyroMark assay design program version 2.0.1.15 (Qiagen), and the sequences are presented in Table S1. PCR using bisulfate‐treated DNA was conducted using a PCR premix (Enzynomics, Daejeon, Korea), and the quality and quantity of the PCR product were confirmed by performing electrophoresis using 2% agarose gel (SeaKem LE Agarose, Lonza, Rockland, ME, USA) with loading 2 µL of 20 PCR products. Pyrosequencing was performed using the Pyro Gold reagent kit and PyroMark ID instrument (Qiagen) as instructed by the manufacturer. The methylation index (MtI) of each gene in each sample was calculated as the average value of mC/(mC + C) for all examined CpG sites in target regions. All experiments included a negative control without a template.
2.9. Statistical Analysis
Statistical analyses were performed using SPSS version 25.0 software (IBM Corp., Endicott, NY, USA). The change in the level of ctDNA before and after surgery for breast cancer was analyzed using the repeated‐measures one‐factor analysis. The level of ctDNA and the ctDNA methylation status were compared for each gene. Correlation between DNA methylation status and gene expression in breast cancer cell lines was analyzed using correlation analysis. The association between DNA methylation status and gene expression in breast cancer tissues was analyzed using the Student's t‐test or the nonparametric Mann–Whitney U‐test. Correlation between ctDNA levels and histologic tumor size was analyzed using correlation analysis. Correlation between ctDNA levels and tumor stage was analyzed using logistic regression. The clinicopathological characteristics of patients were compared according to the DNA methylation status and gene expression in breast cancer tissues. A one‐sample Kolmogorov–Smirnov test was used to evaluate the normal distribution fit of continuous parameters. The χ 2 test or Fisher's exact test was used for categorical data, and Student's t‐test or the nonparametric Mann–Whitney U‐test was used for continuous variables. All tests were two‐sided, and a p‐value of <0.05 was considered to indicate a statistically significant difference.
3. Results
3.1. Selection of Differentially Methylated Breast Cancer‐Specific Genes Using TCGA Data
Analysis of TCGA DNA methylation databases for breast cancer identified several candidate genes with cancer‐specific methylation patterns. Among these, USP44, ZNF454, and GPRC5B were selected based on their differential methylation profiles in breast cancer tissues compared to normal tissues. Specific CpG islands, mean differential methylation, and the correlation coefficient between DNA methylation and gene expression for each gene are shown in Figure S1.
3.2. Methylation and Expression Status of Candidate Genes in Breast Cancer Cell Lines
MSP and qRT‐PCR analyses were conducted on USP44, ZNF454, and GPRC5B genes in various breast cancer cell lines. The expression of the candidate genes was different according to the type of cancer cell lines (Figure 1). USP44 was highly expressed only in MDA‐MB‐231, with low expression in the other cell lines, while ZNF454 was highly expressed only in BT‐549 and showed decreased expression in the other cell lines. GPRC5B did not differ significantly depending on the cell line type.
FIGURE 1.
Expression of the candidate genes in breast cancer cell lines.
The methylation status of the candidate genes was also different according to the type of cancer cell line (Figure 2). For USP44 gene, MCF‐7, BT‐474, SKBR3, and BT‐20 showed weak methylation. BT549 and MDA‐MB‐231 as well as control showed unmethylation. For ZNF454 gene, MCF‐7, BT‐474, and BT‐20 showed methylation, but SKBR3 and BT549 showed unmethylation. For GPRC5B gene, BT‐474, SKBR3, BT‐20, and MDA‐MB‐231 showed weak methylation, but MCF‐7 and BT549 showed unmethylation.
FIGURE 2.
Methylation status of the candidate genes in breast cancer cell lines.
Methylation and expression of USP44 gene in MDA‐MB‐231 and those of ZNF454 gene in BT‐549 showed an inverse correlation.
3.3. Methylation and Expression Status of Candidate Genes in Breast Cancer Tissues
The positive expression rate in breast cancer tissues was 65.7% for USP44, 71.4% for ZNF454, and 77.1% for GPRC5B (Table S2). We expected that the results of analysis using tissues would show lower expression of cancer‐specific genes in cancer tissues than control group. The expression of USP44, ZNF454, and GPRC5B in breast cancer tended to be lower than that of the control group (71.4%, 78.6%, and 85.7%, respectively), but the findings did not show statistically significant results.
Mean methylation frequency of the USP44 and ZNF454 genes was 42.5 ± 16.7% and 37.6 ± 16.7%, respectively, in breast cancer tissues. There was no significant correlation between methylation and expression status of candidate genes in breast cancer tissues.
3.4. Comparison of ctDNA Levels and ctDNA Methylation Status According to Cancer Treatment
The mean level of ctDNA in breast cancer patients in this study was 3.3 ± 1.7 (range 0.2–7.6) ng/mL (Table 1). The control group showed a relatively higher level of circulating cell‐free DNA (cfDNA) compared to ctDNA in breast cancer patients, with 5.3 ± 1.2 (range 3.7–7.9) ng/mL (p = 0.001). Positive methylation rate of ctDNA was 60.0% for USP44, 8.6% for GPRC5B, and 0% for ZNF454. There was no association between ctDNA methylation status and gene expression in breast cancer tissues.
TABLE 1.
cfDNA levels in patients.
| Patient | Time of blood collection | Number of blood samples | Plasma volume (mL), mean (range) | Total cfDNA concentration (ng), mean (range) | Mean concentration of cfDNA per 1 µL (ng/µL), mean (range) |
|---|---|---|---|---|---|
| Breast cancer | Initial | 38 |
10.4 ± 2.4 (5–14) |
32.3 ± 15.3 (2.1–79.6) |
3.3 ± 1.7 (0.2–7.6) |
| Follow‐up (posttreatment) | 35 |
5.3 ± 1.1 (3–10) |
30.8 ± 6.9 (20.5–57.6) |
6.1 ± 1.8 (2.8–11.2) |
|
| Benign breast disease | Initial | 11 |
5.2 ± 0.7 (3.5–6) |
27.4 ± 6.3 (20.3–42.9) |
5.3 ± 1.2 (3.7–7.9) |
When comparing changes in the mean level of ctDNA before and after treatment, the ctDNA levels were measured to be higher after treatment compared to before treatment, but there were no statistically significant differences. In the comparison of changes in the methylation pattern of candidate genes in ctDNA before and after treatment, there were no significant changes for each candidate genes.
To correlate ctDNA levels with tumor burden, we analyzed the association between ctDNA levels and histologic tumor size or tumor stage, but there was no significant correlation.
3.5. Comparative Analysis of the Association of Gene Expression and DNA Methylation Status With Clinicopathological Characteristics
We analyzed the association between gene expression and DNA methylation status in cancer tissues and clinicopathological characteristics (Tables 2 and 3). The results showed that GPRC5B expression was associated with fibrotic focus in breast cancer tissue (p = 0.020). Methylation status of the USP44 gene was associated with lymph node involvement, molecular subtype, and HER2 overexpression (p = 0.001, p = 0.029, and p = 0.047, respectively). In addition, methylation status of the ZNF454 gene was associated with lymph node involvement and HER2 overexpression (p = 0.023 and p = 0.013, respectively).
TABLE 2.
Comparative analysis of the association between candidate gene expression in breast cancer tissues and clinicopathological characteristics of breast cancer patients.
| Clinicopathologic variables | USP44 | ZNF454 | GPRC5B | ||||
|---|---|---|---|---|---|---|---|
| Positive expression, n (%) | p‐value | Positive expression, n (%) | p‐value | Positive expression, n (%) | p‐value | ||
| Age (years) | <50 | 11 (78.6) | 0.282 | 12 (85.7) | 0.252 | 11 (78.6) | 0.869 |
| ≥50 | 12 (57.1) | 13 (61.9) | 16 (76.2) | ||||
| Menopausal status, n (%) | Premenopausal | 10 (76.9) | 0.463 | 11 (84.6) | 0.259 | 10 (46.9) | 0.981 |
| Postmenopausal | 13 (59.1) | 14 (63.6) | 17 (77.3) | ||||
| Tumor size (cm) | ≤2 | 14 (77.8) | 0.122 | 12 (66.7) | 0.711 | 15 (83.3) | 0.443 |
| >2 | 9 (52.9) | 13 (76.5) | 12 (70.6) | ||||
| Histologic grade, n (%) | I | 2 (33.3) | 0.182 | 4 (66.7) | 0.940 | 2 (33.3) | 0.018* |
| II | 14 (73.7) | 14 (73.7) | 16 (59.3) | ||||
| III | 7 (70.0) | 7 (70.0) | 9 (90.0) | ||||
| Nodal involvement, n (%) | Negative | 21 (70.0) | 0.313 | 21 (70.0) | 0.647 | 23 (76.7) | 0.869 |
| Positive | 2 (40.0) | 4 (80.0) | 4 (80.0) | ||||
| Initial stage, n (%) | I | 14 (77.8) | 0.157 | 12 (66.7) | 0.795 | 15 (83.3) | 0.590 |
| II | 9 (60.0) | 11 (73.3) | 10 (66.7) | ||||
| III | 0 (0) | 1 (100) | 1 (100) | ||||
| IV | 0 (0) | 1 (100) | 1 (100) | ||||
| Molecular subtype, n (%) | Luminal A | 8 (57.1) | 0.355 | 7 (50.0) | 0.125 | 8 (57.1) | 0.136 |
| Luminal B | 13 (76.5) | 14 (82.4) | 15 (88.2) | ||||
| HER2 | 0 (0) | 1 (100) | 1 (100) | ||||
| Basal‐like | 2 (66.7) | 3 (100) | 3 (100) | ||||
| Lymphovascular invasion, n (%) | Negative | 18 (66.7) | 0.827 | 19 (70.4) | 0.799 | 20 (74.1) | 0.648 |
| Positive | 5 (62.5) | 6 (75.0) | 7 (87.5) | ||||
| ER, n (%) | Negative | 2 (50.0) | 0.594 | 4 (100) | 0.303 | 4 (100) | 0.553 |
| Positive | 21 (67.7) | 21 (67.7) | 23 (74.2) | ||||
| PR, n (%) | Negative | 3 (42.9) | 0.200 | 5 (71.4) | 1.000 | 5 (71.4) | 0.648 |
| Positive | 20 (71.4) | 20 (71.4) | 22 (78.6) | ||||
| HER2 overexpression, n (%) | Negative | 19 (67.9) | 0.670 | 19 (67.9) | 0.644 | 21 (75.0) | 0.546 |
| Positive | 4 (57.1) | 6 (85.7) | 6 (85.7) | ||||
| Ki‐67, n (%) | <14% | 9 (56.3) | 0.279 | 9 (56.3) | 0.132 | 10 (62.5) | 0.105 |
| ≥14% | 14 (73.7) | 16 (84.2) | 17 (89.5) | ||||
| Tumor‐infiltrating lymphocytes, n (%) | Negative | 12 (66.7) | 0.277 | 12 (66.7) | 0.629 | 14 (77.8) | 0.539 |
| Positive | 6 (100) | 5 (83.3) | 6 (100) | ||||
| Fibrotic focus, n (%) | Negative | 13 (92.9) | 0.050 | 10 (71.4) | 0.939 | 14 (100) | 0.020* |
| Positive | 5 (50.0) | 7 (70.0) | 6 (60.0) | ||||
Abbreviations: ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; PR, progesterone receptor
*Statistically significant (p < 0.05).
TABLE 3.
Comparative analysis of the association between candidate gene methylation status in breast cancer tissues and clinicopathological characteristics of breast cancer patients.
| Clinicopathologic variables | USP44 methylation | ZNF454 methylation | |||
|---|---|---|---|---|---|
| Mean levels (%) | p‐value | Mean levels (%) | p‐value | ||
| Age (years) | <50 | 39.06 ± 13.17 | 0.227 | 33.08 ± 15.82 | 0.201 |
| ≥50 | 44.77 ± 13.83 | 40.47 ± 17.01 | |||
| Menopausal status, n (%) | Premenopausal | 38.64 ± 13.61 | 0.202 | 33.65 ± 16.32 | 0.293 |
| Postmenopausal | 44.75 ± 13.51 | 39.83 ± 16.90 | |||
| Tumor size (cm) | ≤2 | 45.03 ± 12.98 | 0.284 | 41.07 ± 13.19 | 0.217 |
| >2 | 40.07 ± 14.27 | 34.12 ± 19.41 | |||
| Histologic grade, n (%) | I | 38.99 ± 18.27 | 0.970 | 32.99 ± 11.10 | 0.173 |
| II | 37.06 ± 14.50 | 44.25 ± 11.68 | |||
| III | 37.84 ± 21.39 | 44.86 ± 17.25 | |||
| Nodal involvement, n (%) | Negative | 45.65 ± 11.77 | 0.001* | 40.39 ± 15.49 | 0.023* |
| Positive | 27.04 ± 12.04 | 23.62 ± 17.15 | |||
| Initial stage, n (%) | I | 45.03 ± 12.98 | 0.288 | 41.07 ± 13.19 | 0.105 |
| II | 41.29 ± 14.06 | 37.00 ± 18.53 | |||
| III | 18.45 | 5.88 | |||
| IV | 42.18 | 16.25 | |||
| Molecular subtype, n (%) | Luminal A | 39.38 ± 12.76 | 0.029* | 37.80 ± 13.44 | 0.149 |
| Luminal B | 47.35 ± 11.12 | 37.69 ± 17.50 | |||
| HER2 | 58.92 | 69.80 | |||
| Basal‐like | 27.72 ± 19.25 | 25.39 ± 19.46 | |||
| Lymphovascular invasion, n (%) | Negative | 43.13 ± 14.24 | 0.641 | 38.52 ± 15.92 | 0.541 |
| Positive | 40.52 ± 12.14 | 34.35 ± 20.16 | |||
| ER, n (%) | Negative | 35.52 ± 22.15 | 0.531 | 36.49 ± 27.30 | 0.891 |
| Positive | 43.43 ± 12.51 | 37.73 ± 15.61 | |||
| PR, n (%) | Negative | 34.64 ± 15.70 | 0.088 | 35.10 ± 19.80 | 0.666 |
| Positive | 44.46 ± 12.71 | 38.20 ± 16.25 | |||
| HER2 overexpression, n (%) | Negative | 40.14 ± 13.74 | 0.047* | 33.97 ± 14.87 | 0.013* |
| Positive | 50.97 ± 10.17 | 50.27 ± 17.62 | |||
| Ki‐67, n (%) | <14% | 38.57 ± 11.76 | 0.099 | 35.48 ± 13.46 | 0.482 |
| ≥14% | 46.11 ± 14.58 | 39.48 ± 19.37 | |||
| Tumor‐infiltrating lymphocytes, n (%) | Negative | 48.20 ± 12.45 | 0.808 | 40.97 ± 16.74 | 0.842 |
| Positive | 46.82 ± 9.77 | 42.50 ± 13.98 | |||
| Fibrotic focus, n (%) | Negative | 49.00 ± 12.03 | 0.578 | 42.28 ± 14.24 | 0.740 |
| Positive | 46.24 ± 11.52 | 40.04 ± 18.52 | |||
Abbreviations: ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; PR, progesterone receptor
*Statistically significant (p < 0.05).
There was no association between ctDNA level, ctDNA methylation status of candidate genes, and clinicopathological characteristics in this study.
Clinicopathological characteristics of breast cancer patients included in this study are shown in Table S3.
4. Discussion
This study investigated the methylation and expression status of USP44, ZNF454, and GPRC5B genes in breast cancer tissues, cell lines, and ctDNA. Our findings provide insights into the complex epigenetic landscape of breast cancer and the potential utility of ctDNA methylation analysis as a noninvasive biomarker, though certain challenges remain.
Using TCGA data, we identified USP44, ZNF454, and GPRC5B as breast cancer‐specific candidate genes with differential methylation profiles. These results align with previous research on the role of DNA methylation in cancer progression [16, 21]. The methylation and expression profiles observed in breast cancer tissues and cell lines highlight the intrinsic heterogeneity of breast cancer. Although methylation levels were significantly altered in breast cancer tissues, the correlation between methylation and gene expression varied by gene and sample type, suggesting context‐specific regulatory mechanisms. For example, USP44 and ZNF454 exhibited an inverse correlation between methylation and expression in certain breast cancer cell lines, but this pattern was not consistently observed in breast cancer tissues.
Our analysis revealed variability in methylation and expression across breast cancer cell lines. For instance, USP44 showed high expression in MDA‐MB‐231 cells and weak methylation in several other cell lines. Similarly, ZNF454 was highly expressed in BT‐549 cells and exhibited methylation in MCF‐7 and BT‐20 cell lines. These findings suggest that methylation and expression patterns may depend on intrinsic properties of different breast cancer subtypes, aligning with the heterogeneity observed in breast cancer [22].
The detection of ctDNA in breast cancer patients represents an emerging frontier in liquid biopsy technologies. Our study found that the mean ctDNA level was lower than the cfDNA levels in the control group, diverging from reports that cancer patients typically exhibit elevated ctDNA levels due to tumor shedding [23]. Recent advancements in ctDNA analysis, such as Guardant360, GRAIL's methylation‐based Galleri test, and DELFI's fragmentation‐based profiling, have demonstrated the feasibility of detecting cancer‐specific ctDNA alterations with high sensitivity and specificity [24, 25, 26]. Furthermore, studies have demonstrated the feasibility of using ctDNA methylation for detecting breast cancer, with multiple cfDNA fragmentomic features, including methylation patterns, showing promise in distinguishing cancer patients from healthy individuals [27]. Additionally, methylation‐based ctDNA subtyping has been shown to detect the presence of breast cancer and discriminate IHC subtypes with high accuracy, highlighting its potential in noninvasive cancer diagnostics [28]. However, the limited methylation positivity of ZNF454 and GPRC5B in ctDNA in our study suggests that further optimization is needed to reliably capture tumor‐specific epigenetic signatures.
The genes USP44, ZNF454, and GPRC5B, investigated in this study, have been implicated in various cellular processes, and emerging evidence suggests their involvement in breast cancer. Dysregulation of USP44 can lead to chromosomal instability and has been associated with tumorigenesis in various cancers, including breast cancer [29]. ZNF454 is a member of the zinc finger protein family, which is involved in DNA binding and transcriptional regulation [30]. While the specific role of ZNF454 in breast cancer remains under investigation, zinc finger proteins, in general, have been shown to influence cancer progression [30]. GPRC5B is a member of the G protein‐coupled receptor (GPCR) family. GPCRs have been implicated in various aspects of cancer biology [31]. While direct evidence linking GPRC5B to breast cancer is limited, the involvement of GPCRs in cancer suggests a potential role for GPRC5B in breast cancer biology. In our study, the results showed an association between these genes and clinicopathological features, which highlight the distinct roles of GPRC5B, USP44, and ZNF454 in the pathophysiology of breast cancer and their potential as biomarkers. Our study found a significant association between GPRC5B expression and fibrotic focus in breast cancer tissue. While the exact role of GPRC5B in fibrotic focus remains unclear, GPRC5B may contribute to the remodeling of the extracellular matrix and stromal interactions, which are pivotal in breast cancer progression. Further studies are warranted to elucidate these mechanisms. Additionally, the methylation status of USP44 and ZNF454 genes demonstrated significant associations with key clinicopathological characteristics, such as lymph node involvement and HER2 overexpression. These findings align with previous reports suggesting that DNA methylation can modulate gene expression in a subtype‐specific manner, influencing tumor invasiveness and metastatic potential [32, 33]. HER2 overexpression, a hallmark of aggressive breast cancer subtypes, is known to alter epigenetic landscapes, including DNA methylation [34]. The observed associations suggest that USP44 and ZNF454 methylation might serve as biomarkers for HER2‐driven breast cancer. While these findings provide insights into the potential roles of these genes, their clinical significance requires further validation.
Despite notable findings, our study has limitations in correlation between ctDNA and tissue biomarkers, as well as in association between ctDNA levels or methylation status and clinicopathological characteristics. In addition, we were unable to obtain successful results of ctDNA methylation status in control group using MSP. This could be attributed to the complexity of ctDNA release mechanisms and their heterogeneous origins, which may not fully represent the tumor methylation landscape [23, 35]. Technologies like GRAIL and DELFI have addressed this issue by integrating genome‐wide methylation patterns or fragmentation profiles, enabling more comprehensive representation of the tumor methylome [25, 26]. Incorporating similar approaches in future research could enhance the detection of tumor‐specific methylation markers.
5. Conclusion
While ctDNA methylation analysis has significant potential, challenges such as variability in methylation patterns, limited correlation with tissue biomarkers, and the need for high‐sensitivity assays must be addressed. Future studies should focus on larger, multi‐institutional cohorts and advanced liquid biopsy technologies to explore and validate these findings.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supplementary Table 1: Primer sequences used for Polymerase Chain Reaction (PCR), methylation specific PCR (MSP) and pyrosequencing.
Supplementary Table 2: Methylation and expression status of candidate genes in breast tissues.
Supplementary Table 3: Patient characteristics.
Supporting File 2: ajco70015‐sup‐0002‐FigureS1.tif.
Supporting File 3: ajco70015‐sup‐0003‐SuppMat.docx.
Acknowledgments
This work was supported by the grant of Daegu Catholic University Medical Center (2020).
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.
References
- 1. Statistics Korea , “Statistics Korea,” http://kosis.kr.
- 2. Siegel R. L., Kratzer T. B., Giaquinto A. N., Sung H., and Jemal A., “Cancer Statistics, 2025,” CA: A Cancer Journal for Clinicians 75, no. 1 (2025): 10–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Trapani D., Ginsburg O., Fadelu T., et al., “Global Challenges and Policy Solutions in Breast Cancer Control,” Cancer Treatment Reviews 104 (2022): 102339. [DOI] [PubMed] [Google Scholar]
- 4. Barrios C. H., “Global Challenges in Breast Cancer Detection and Treatment,” Breast 62, no. Suppl 1 (2022): S3–S6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Mazzitelli C., Santini D., Corradini A. G., et al., “Liquid Biopsy in the Management of Breast Cancer Patients: Where Are We Now and Where Are We Going,” Diagnostics 13 (2023): 1241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sisodiya S., Kasherwal V., Khan A., et al., “Liquid Biopsies: Emerging Role and Clinical Applications in Solid Tumours,” Translational Oncology 35 (2023): 101716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Venetis K., Cursano G., Pescia C., et al., “Liquid Biopsy: Cell‐Free DNA Based Analysis in Breast Cancer,” Journal of Liquid Biopsy 1 (2023): 100002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Zhang X., Zhao D., Yin Y., et al., “Circulating Cell‐Free DNA‐Based Methylation Patterns for Breast Cancer Diagnosis,” npj Breast Cancer 7 (2021): 106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Huang J. and Wang L., “Cell‐Free DNA Methylation Profiling Analysis‐Technologies and Bioinformatics,” Cancers 11, no. 11 (2019): 1741. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Uehiro N., Sato F., Pu F., et al., “Circulating Cell‐Free DNA‐Based Epigenetic Assay Can Detect Early Breast Cancer,” Breast Cancer Research 18, no. 1 (2016): 129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Zhang K., Fu R., Liu R., and Su Z., “Circulating Cell‐Free DNA‐Based Multi‐Cancer Early Detection,” Trends in Cancer 10, no. 2 (2024): 161–174. [DOI] [PubMed] [Google Scholar]
- 12. Stergiopoulou D., Markou A., Strati A., et al., “Comprehensive Liquid Biopsy Analysis as a Tool for the Early Detection of Minimal Residual Disease in Breast Cancer,” Scientific Reports 13, no. 1 (2023): 1258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Nader‐Marta G., Monteforte M., Agostinetto E., et al., “Circulating Tumor DNA for Predicting Recurrence in Patients With Operable Breast Cancer: A Systematic Review and Meta‐Analysis,” ESMO Open 9, no. 3 (2024): 102390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Garcia‐Murillas I., Cutts R., Abbott C., et al., “Ultra‐Sensitive ctDNA Mutation Tracking to Identify Molecular Residual Disease and Predict Relapse in Patients With Early Breast Cancer,” Journal of Clinical Oncology 42, no. 16_suppl (2024): 1010. [Google Scholar]
- 15. Keup C., Kimmig R., and Kasimir‐Bauer S., “The Diversity of Liquid Biopsies and Their Potential in Breast Cancer Management,” Cancers 15, no. 22 (2023): 5463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Lakshminarasimhan R. and Liang G., “The Role of DNA Methylation in Cancer,” Advances in Experimental Medicine and Biology 945 (2016): 151–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Zhu Q., Xie J., Mei W., and Zeng C., “Methylated Circulating Tumor DNA in Hepatocellular Carcinoma: A Comprehensive Analysis of Biomarker Potential and Clinical Implications,” Cancer Treatment Reviews 128 (2024): 102763. [DOI] [PubMed] [Google Scholar]
- 18. Borg M., Wen S. W. C., Andersen R. F., Timm S., Hansen T. F., and Hilberg O., “Methylated Circulating Tumor DNA in Blood as a Tool for Diagnosing Lung Cancer: A Systematic Review and Meta‐Analysis,” Cancers 15, no. 15 (2023): 3959. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Li Y., Ge D., and Lu C., “The SMART App: An Interactive Web Application for Comprehensive DNA Methylation Analysis and Visualization,” Epigenetics & Chromatin 12, no. 1 (2019): 71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Jeong Y. J., Jeong H. Y., Bong J. G., Park S. H., and Oh H. K., “Low Methylation Levels of the SFRP1 Gene Are Associated With the Basal‐Like Subtype of Breast Cancer,” Oncology Reports 29 (2013): 1946–1954. [DOI] [PubMed] [Google Scholar]
- 21. Zhang M., Wang Y., Wang Y., et al., “Integrative Analysis of DNA Methylation and Gene Expression to Determine Specific Diagnostic Biomarkers and Prognostic Biomarkers of Breast Cancer,” Frontiers in Cell and Developmental Biology 8 (2020): 529386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Guo L., Kong D., Liu J., et al., “Breast Cancer Heterogeneity and Its Implication in Personalized Precision Therapy,” Experimental Hematology & Oncology 12 (2023): 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Wan J. C. M., Massie C., Garcia‐Corbacho J., et al., “Liquid Biopsies Come of Age: Towards Implementation of Circulating Tumour DNA,” Nature Reviews Cancer 17, no. 4 (2017): 223–238. [DOI] [PubMed] [Google Scholar]
- 24. Bauml J. M., Li B. T., Velcheti V., et al., “Clinical Validation of Guardant360 CDx as a Blood‐based Companion Diagnostic for Sotorasib,” Lung Cancer 166 (2022): 270–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Klein E. A., Richards D., Cohn A., et al., “Clinical Validation of a Targeted Methylation‐Based Multi‐Cancer Early Detection Test Using an Independent Validation Set,” Annals of Oncology 32, no. 9 (2021): 1167–1177. [DOI] [PubMed] [Google Scholar]
- 26. Cristiano S., Leal A., Phallen J., et al., “Genome‐Wide Cell‐Free DNA Fragmentation in Patients With Cancer,” Nature 570, no. 7761 (2019): 385–389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Pham T. M. Q., Phan T. H., Jasmine T. X., et al., “Multimodal Analysis of Genome‐Wide Methylation, Copy Number Aberrations, and End Motif Signatures Enhances Detection of Early‐Stage Breast Cancer,” Frontiers in Oncology 13 (2023): 1127086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Pasha N., Amenssag A., Hrebien S., et al., “Discriminating Subtypes in Advanced Breast Cancer With ctDNA Methylation Profiling,” Journal of Clinical Oncology 42, no. 16_suppl (2024): 1013. [Google Scholar]
- 29. Lou Y., Ye M., Xu C., and Tao F., “Insight Into the Physiological and Pathological Roles of USP44, a Potential Tumor Target (Review),” Oncology Letters 24, no. 6 (2022): 455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Zhao J., Wen D., Zhang S., and Jiang H., and Di X., “The Role of Zinc Finger Proteins in Malignant Tumors,” FASEB Journal 37, no. 9 (2023): e23157. [DOI] [PubMed] [Google Scholar]
- 31. Parija M., Adhya A. K., and Mishra S. K., “G‐Protein‐Coupled Receptor 141 Mediates Breast Cancer Proliferation and Metastasis by Regulating Oncogenic Mediators and the p‐mTOR/p53 Axis,” Oncotarget 14 (2023): 466–480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Sher G., Salman N. A., Khan A. Q., et al., “Epigenetic and Breast Cancer Therapy: Promising Diagnostic and Therapeutic Applications,” Seminars in Cancer Biology 83 (2022): 152–165. [DOI] [PubMed] [Google Scholar]
- 33. Garcia‐Martinez L., Zhang Y., Nakata Y., Chan H. L., and Morey L., “Epigenetic Mechanisms in Breast Cancer Therapy and Resistance,” Nature Communications 12 (2021): 1786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Singla H., Ludhiadch A., Kaur R. P., Chander H., Kumar V., and Munshi A., “Recent Advances in HER2 Positive Breast Cancer Epigenetics: Susceptibility and Therapeutic Strategies,” European Journal of Medicinal Chemistry 142 (2017): 316–327. [DOI] [PubMed] [Google Scholar]
- 35. Saha S., Araf Y., and Promon S. K., “Circulating Tumor DNA in Cancer Diagnosis, Monitoring, and Prognosis,” Journal of the Egyptian National Cancer Institute 34, no. 1 (2022): 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1: Primer sequences used for Polymerase Chain Reaction (PCR), methylation specific PCR (MSP) and pyrosequencing.
Supplementary Table 2: Methylation and expression status of candidate genes in breast tissues.
Supplementary Table 3: Patient characteristics.
Supporting File 2: ajco70015‐sup‐0002‐FigureS1.tif.
Supporting File 3: ajco70015‐sup‐0003‐SuppMat.docx.
Data Availability Statement
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.