The role of methylation quantification of circulating tumor DNA (ctDNA) as a diagnostic biomarker of Pheochromocytomas (PCCs) and Paragangliomas (PGLs)

Abstract

Objectives

Circulating tumor DNAs (ctDNAs) are fragments of malignant tissue DNA that can simply signify the real time genetic change and epigenetic modification of a solid tumor tissue. Pheochromocytomas (PCCs) and Paragangliomas (PGLs) are malinancy of adrenal gland tissue that have the possible diagnosis by ctDNAs. In this study the methylation quanifcation of three target genes RDBP, SDHB, and SDHC in the ctDNA of PCCs/PGLs patients were measured as a diagnostic biomarker.

Methods

The biological samples include blood and fresh frozen tissue of twelve PCCs/PGLs patients and blood of 12 non tumoral patients as controls were recruited. Semi quantification methylation status of RDBP, SDHB, and SDHC (two CpG lslands of each gene named 1 and 2) was assesed between PCCs/PGLs patients and controls by Methylation specific-high resolution melting (MS-HRM) technique.

Results

Between six candidate CpG island of RDBP, SDHB, and SDHC, promoter methylation quantification of SDHC1 and RDBP2 was expressively unsimilar in PCCs/PGLs compare to the controls. SDHC1 was hypermethylated in 49.93% of PCCs/PGLs cases vs. 8.33% of control samples, p-value: 0.026, area under curve AUC = 0.757, and RDBP2 in 74.9% of PCCs/PGLs cases vs. 25.0% of control samples, p-value: 0.032, AUC = 0.750.

Conclusions

Our result shows that the ctDNA hypermethylation of SDHC1 and RDBP2 have role in tumorgenesis of adrenal gland and can consider for diagnosis of PCCs/PGLs.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40200-024-01466-8.

Introduction

Pheochromocytomas (PCCs) and Paragangliomas (PGLs) are rare catecholamine-secreting tumors that originate from the chromaffin cells of the adrenal medulla, typically displaying as a benign tumor and infrequently as malignant [1]. PCC and PGL collectively referred to as PPGLs— are rare neuroendocrine tumors. PCCs do arise from the chromaffin cells in the adrenal medulla but PGLs arise from the extra-adrenal ganglia PCCs/PGLs are considered as the same entity given their similar origin and clinical presentations [2]. The annual incidence of PCCs/PGLs is about 0.8 per 100,000 people [3]. Nevertheless, this is probably an underestimation, because half of PCCs/PGLs might be diagnosed at autopsy [4]. PCCs/PGLs mostly happen in the 3rd and 4th decades of age, equally in both genders [1]. There is an extreme need to optimize biomarkers, mainly genetically and epigenetically to discriminate the PCCs/PGLs and plan the exact treatment strategy at the exact time.

Recently, the new aspect of Liquid Biopsy (LB) has brought an exceptional tumor marker as the real-time representation of the tumor. Numerous studies are focused on both genetic and epigenetic circulating tumor DNA (ctDNA) as the tumor markers [5]. However, the latest discoveries related to ctDNA genetic modification and circulating tumor cells (CTCs) in PCCs/PGLs are limited [6, 7].

More than genetic modifications, there are some epigenetic variations in DNA molecules that can change the gene expression of the tumoral cells with no change in the exact DNA sequences [8]. DNA hypomethylation in CpG islans is an usual feature of malignant tissue. It can make aberrant activation of germline-specific genes in malignant tissue and leads to show oncogenic properties. It can be the reason why oncogenes specificity are interesting therapeutic targets [9, 10]. DNA methylation patterns in some specific genes are reported to be linked to progression-free survival (PFS) and overall survival (OS) in several cancers [11].

The ctDNAs can be considered a potential non-invasive source of methylation change in PCCs/PGLs. In this study, we examined the methylation quantification of PCCs/PGLs in ctDNA as the tumor transformation biomarker.

Materials and methods

This study was run on 12 consecutive PCCs/PGLs patients (cases) and 12 non-cancerous patients (controls). All patients signed the informed consent and all procedures were under National Institute for Medical Science Development Ethics Committee (IR.NIMAD.REC.1397.452). The blood samples were gathered in EDTA containing vials before surgery for ctDNA analysis. Tumor tissues were captured through surgery as the origin of genomic DNA (gDNA) source. All tissues were surgically resected and immediately put in liquid nitrogen tank as the fresh frozen tissue for additional molecular testings.

Tissue and ctDNA extraction

Fresh frozen tissues of PCCs/PGLs were removed and kept in liquid nitrogen for a maximum of 1 month, then tissue DNA (gDNA) extraction was done by DNeasy Blood & Tissue Kit (Qiagen, Netherlands, Cat No: 69,504) [12]. For ctDNA extraction, about 4–6 ml blood samples were collected from both PCCs/PGLs and control groups and the ctDNA extraction was done within two hours. For ctDNA extraction, firstly the plasma was detached by centrifugation of blood at 2800 rpm for fifteen minutes by means of the Ficoll separation method. Then, plasma was moved to a sterilized tube and ctDNA extracted according to the NORGEN Plasma/Serum Cell-Free Circulating DNA Purification Midi Kit protocol (Canada, Cat No: 55,600). Lastly, the concentration and purity of DNA (no RNA, protein ) were determined by optical density in 260 and 280 nm by Thermo Scientific™ NanoDrop™ spectrophotometers 2000c (Thermo Fisher Scientific Inc). Both extracted ctDNA and gDNA were stored at the freezer − 80 °C for additional molecular testings.

Bisulfite modification

For methylation quantification analysis ctDNA and its counterpart gDNA from each candidate PCCs/PGLs and ctDNA from controls were treated by the “EZ DNA Methylation-Gold™ Kit” (Zymoresearch, USA, Cat No: D5005) according to the manufacturer’s procedure. For the methylation high resolution melting analysis (HRM), six promoter regions of three target genes RDBP, SDHB, and SDHC were selected by specific primers (Table 1).

Table 1.

Specific primer sequences for methylation quantification through MS-HRM analysis of six selected promoter regions of RDBP, SDHB, and SDHC

Gene	Part	Forward Primer	Reverse Primer	Tann	Number of CpG sites in the amplicon
RDBP	a	5’ GGTAAGTTTTTTGTTTTTTAT 3’	5’ TTTAAATACATATAATTCA 3’	56 °C	15
	b	5’ GGATATAGTTTGGTTTAAG 3’	5’ ACATCTTTCTCCACTATTAC 3’	52 °C	9
SDHB	a	5’ GTTAGTGTTTTAGTGGATGT 3’	5’ AAACTCACCTACAAACAAAC 3’	57 °C	17
	b	5’ GGGAAGTTAAATGGGT 3’	5’ TCCACTAAAACCCACT 3’	55 °C	14
SDHC	a	5’ GTAATTAGTTAGGTAGAG 3’	5’ ACTAAATCACCTCAACA 3’	50 °C	14
	b	5’TAGATGTAGATTTTGAGTTA 3’	5’ACTCTACTAACTAATTTAC 3’	49 °C	6

The MS-HRM program includes following main parts: holding step (94°C for 20 min), going to 40 cycles of 94°C for 10 s, several annealing temperatures depending on the primer (varying from 45°C to 60°C) for 35 s, and extension time of 72°C for 35 s leading to the final step of the melting curve. The melting curve step was made by heating the samples to 90°C for 15 s and 60°C for one minute leading to 65°C for 15 s, and afterward uninterruptedly heated up to 95°C with the gaining of data through each 0.3^°C rises in temperature.

The mixture consisted of 10 µl of master mix (Amplicon, Cat No: A325406), 20 pmol of each primer, and 2µl (almost 10ng) of bisulfite modified DNA template in the whole volume of 20 µl. Moreover 0%, 25%, 50%, 75%, and 100% methylated controls were run in each reaction. Standard curves with identified methylation ratios were comprised in all assays by assuming the methylation ratio of the unknown target. MS-HRM tests were done by the ABI Step One Plus system in triplicates.

Statistical analysis

The hyper-methylation defined as 25% and 100% of methylation compared to controls and non-methylated when the methylation amount was near 0% compared to the control. The assessment was run between two main groups: the case group consisting of 12 PCCs/PGLs patients and the control including of 12 non-cancerous patients. The specificity and sensitivity of each target CpG site were assessed through a 2 × 2 table. The correspondence mutation and methylation of ctDNAs and their tissue equivalents were performed through spearman correlation analysis and ctDNA’s methylation of PCCs/PGLs patients Receiver Operating Characteristics (ROC) curves were constructed. All evaluates were done by Statistical Package for Science Software (SPSS) (version 17.0; SPSS Inc. Chicago, Illinois).

Results

Demographic patients data of PCCs/PGLs and controls are presented in Table 2. The age of all patients ranged from 26 to 60. Among all 12 PCCs/PGLs tumor size varied from 0.5 to 6 cm and three patients (25%) were defined as the homogeneous and hypoechoic solid lesion with well-defined borders. PCCs/PGLs patients showed considerably upper systolic blood pressure (SBP), and diastolic blood pressure (DBP), and also significantly lower weight and body mass index (BMI).

Table 2.

Demographics of PCCs/PGLs (cases) and non-cancer patients (controls)

Variables	PCCs/PGLs cases (n = 12)	Controls (n = 12)	P-value
Age (years)	41.25 (± 10.532)	42.42 (± 11.828)	-
Gender
Female	8 (66.7%)	8 (66.7%)	-
Male	4 (33.3%)	4 (33.3%)	-
Weight (kg)	61.58 (± 5.299)	74.00 (± 11.201)	0. 002
Height (cm)	165.50 (± 9.200)	165.32 (± 9.55)	0.370
BMI	23.68 (± 3.884)	27.05 (± 3.833)	0.043
SBP (mm/Hg)	13.72(± 1.190)	11.50 (± 1.167)	< 0.001
DBP (mm/Hg)	10.36(± 1.68)	7.91 (± 1.37)	0.001
Educated (After High School)	7 (58.3%)	10 (83.3%)	0.146
Malignant	5(41.6%)	-	-

BMI body mass index, SBP: systolic blood pressure, DBP: diastolic blood pressure

For each run of MS-HRM ten wells were allocated to the methylation controls 0%, 25%, 50%, 75%, and 100% and test samples were run in triplicate. The melting curve was considered for additional analysis to determine the methylation status (Fig. 1). The methylation semiquantification of each sample was assessed by comparing the normalized melt curve.

Fig. 1.

The MS-HRM for studying the methylation status of six CpG sites of three target genes

The methylation status of all six target regions in case and controls are presented in Table 3; Fig. 2. Among six promoter regions within three candidate genes, the SDHC1 (the promoter region harboring initial ATG code) and RDBP2 methylation statuses were hypermethylated (more than 25% methylation) in PCCs/PGLs patients compared to the control group, p-value 0.026 and 0.03, respectively.

Table 3.

Methylation pattern of six target promoter regions in PCCs/PGLs and normal cases

Promoter Region	Methylation	PCCs/PGLs patients Number (percentage)	Control (percentage)	P-value
SDHB1	0-12.5% methylated (non-methylated)	8 (66.6%)	7 (58.3%)	0.886
	12.5 ≤, <25% methylated	1 (8.33%)	3 (25.0%)
	25≤, <50% methylated	1 (8.33%)	0 (0.0%)
	50≤, <75% methylated	1 (8.33%)	1 (8.33%)
	75–100% methylated	1 (8.33%)	1 (8.33%)
SDHB2	0-12.5% methylated (non-methylated)	3 (25.0%)	3 (25.0%)	0.507
	12.5 <, ≥ 25% methylated	1 (8.33%)	2 (16.6%)
	25≤, <50% methylated	2 (16.6%)	5 (41.6%)
	50≤, <75% methylated	3 (25.0%)	1 (8.33%)
	75–100% methylated	3 (25.0%)	1 (8.33%)
SDHC1	0-12.5% methylated (non-methylated)	4 (33.3%)	9 (75.0%)	0.026*
	12.5 ≤, <25% methylated	2 (16.6%)	2 (16.6%)
	25≤, <50% methylated	2 (16.6%)	1 (8.33%)
	50≤, <75% methylated	1 (8.33%)	0 (0.0%)
	75–100% methylated	3 (25.0%)	0 (0.0%)
SDHC2	0-12.5% methylated (non-methylated)	7 (58.3%)	5 (41.6%)	0.750
	12.5 ≤, <25% methylated	0 (0.0%)	2 (16.6%)
	25≤, <50% methylated	1 (8.33%)	2 (16.6%)
	50≤, <75% methylated	0 (0.0%)	1 (8.33%)
	75–100% methylated	4 (33.3%)	2 (16.6%)
RDBP2	0-12.5% methylated (non-methylated)	11 (891.6%)	10 (83.3%)	0.987
	12.5 ≤, <25% methylated	1 (8.33%)	1 (8.33%)
	25≤, <50% methylated	0 (0.0%)	1 (8.33%)
	50≤, <75% methylated	0 (0.0%)	0 (0.0%)
	75–100% methylated	0 (0.0%)	0 (0.0%)
RDBP1	0-12.5% methylated (non-methylated)	3 (25.0%)	7 (58.3%)	0.032*
	12.5 ≤, <25% methylated	0 (0.0%)	2 (16.6%)
	25≤, <50% methylated	2 (16.6%)	0 (0.0%)
	50≤, <75% methylated	3 (25.0%)	3 (25.0%)
	75–100% methylated	4 (33.3%)	0 (0.0%)

*If the p-value is less than 0.05, we reject the null hypothesis that there’s no difference between the two groups and conclude that a significant difference does exist. *If at least one targeted promoter region was methylated more than 25% the final methylation status of the target gene was hypermethylated

Fig. 2.

Comparison of methylation quantification in six target promoter regions of SDHB, SDHC, and RDBP genes between PCCs/PGLs cases and non-cancerous patients. PCC is an abbreviation of PCCs/PGLs

The maximum value for kappa observed was 0.919 between SDHB2 and SDHC1 and 0.805 between RDBP2 and SDHC2 (Table 4).

Table 4.

Observed agreements between six CpG sites of SDHB, SDHC, and RDBP2 in plasma ctDNA

	RDBP1	RDBP2	SDHB1	SDHB2
SDHC1	0.567	0.668	0.715	0.919 *
SDHC2	0.624	0.824 *	0.805	0.649

According to possible interpretation of Kappa, Poor agreement = Less than 0.20, Fair agreement = 0.20 to 0.40, Moderate agreement = 0.40 to 0.60, Good agreement = 0.60 to 0.80 and, Very good agreement = 0.80 to 1.00. *The highest kappa agreement score.

ROC curves were plotted based on the percentage of methylated reference (PMR) values in PCCs/PGLs and the control group to indicate if any of the targeted six promoter regions in circulating plasma ctDNA can be considered as PCCs/PGLs specific diagnostic markers (Fig. 3).

Fig. 3.

Receiver Operating Characteristic (ROC) curves for the DNA methylation markers (as ranked by p-value), using the current collection of circulating plasma ctDNA

The sensitivity and specificity of target CpG sites were defined based on the capacity to correctly classify subjects with methylated promoters into PCCs/PGLs. The area under the curve (AUC) ranged from 0.455 in RDBP1 to 0.757 in SDHC1 and then 0.750 in RDBP2 (Table 5).

Table 5.

AUC, Sensitivity & Specificity Analysis for six targeted promoter regions of PCCs/PGLs from patients ctDNA

The ctDNA promoter Locus	AUC (95% CI)	Sensitivity	Specificity
SDHB1	0.476 (0.240–0.712)	25%	83.3%
SDHB2	0.622 (0.388–0.855)	66.6%	41.6%
SDHC1	0.757 (0.558–0.956)	50%	91.6%
SDHC2	0.483 (0.242–0.723)	41.6%	58.3%
RDBP1	0.455 (0.220–0.690)	21%	91.6%
RDBP2	0.750 (0.549–0.951)	75%	75%

Discussion

Several genetic mutations are considered important molecular diagnostic biomarkers in PCCs/PGLs. Our study pioneering represents the role of specific genes DNA methylation as PCCs/PGLs potential diagnostic tool. Herein, we report SDHC1 hypermethylation in 49.93% of PCCs/PGLs cases vs. 8.33% of control samples, p-value: 0.026; and RDBP2 promoter region 74.9% hypermethylation in cases vs. 25.0% in controls, p-value: 0.03. These two target regions have AUC 0.750 in RDBP2 and 0.757 in SDHC1 promoters.

Changes in gene expression patterns that trigger the cell to malignancy can be the result of the epigenetic change [13, 14]. The only study so far in which the ctDNA was challenged as PCCs/PGLs tumor genetic indicator was done by Wang et al. in 2018 and to the best of our knowledge, our study is the only one that seeks methylation of ctDNA [6]. Nowadays it has become common knowledge that ctDNA as the liquid biopsy main component has great potential to non-invasively show the tumor genetic status [15–18].

Global DNA methylation array indicates three distinct clusters: M1–3. M1 comprises tumors with SDHx mutations and hypermethylation, M2 VHL-mutated tumors, and M3 tumors with NF1 and RET mutations and hypomethylation [11, 19]. Epithelial to Mesenchymal Transition (EMT) can be activated in metastatic PCCs/PGLs by SDHB gene mutations [20, 21]. It was shown by Astuti et al. that SDHB was hypermethylated in 21% of primary neuroblastomas and 32% of PCCs/PGLs [22]. Our results indicate 25.0% SDHB promoter region hypermethylation and more than 75% of CpG sites harboring initial ATG.

However, based on posterior microsatellite instability and hypermethylation promoter studies, there is still doubt that SDHB methylation can play a role once it is unlikely to be related to either tumor initiation or progression in neuroblastoma [23]. We have considered two upstream promoter regions of initial ATG with no difference in methylations between PCCs/PGLs and control patients. Opposing, it was described that EMT hypermethylation and stemness properties that can be the result of SDHB loss of function [24]. Also, SDHB mutation can be altered in CpG isaland methylation status by methyltransferase enzymes like MGMT [25–28].

Considering gene silencing by promoter methylation is an important epigenetic regulatory mechanism in primary steps of tumorigenesis, SDHC promoter hypermethylation might has a critical role in the development of PCCs/PGLs. Altered expression of SDHC has been reported several times as the consequence of both genetic and epigenetic changes [29–32]. The SDHC promoter hypermethylation can leads to SDHC inactivation so it could be the significant of epigenetic modifications and functional displays in the genetic evaluation of patients [33]. SDH-loss cells are selectively vulnerable to LDH genetic knock-down or chemical inhibition, suggesting that LDH inhibition may be an effective therapeutic strategy for SDH-loss and SDHC-loss transcriptional change correlate with baseline expression values in normal cells [34]. Our results indicate that the CpG sites far from initial ATG are hypermethylated in PCCs/PGLs and its methylation is in agreement with SDHB and RDBP. The RDBP gene is responsible for coding negative elongation factor E (NELF) as a complex that adversely controls the function of transcription by RNA polymerase II [35]. The first large-scale study of DNA methylation in metastatic PGL by de Cubas and colleagues supports that RDBP could be used for stratifying patients according to the risk of developing metastases [36].

Our study indicated RDBP promoter hypermethylation in PCCs/PGLs, in line with Backman et al. that has suggested hypermethylated RDBP in metastasizing PGLs regardless of mutational status [11]. RDBP hypermethylation should be further explored as malignancy and survival markers in patients with PCCs/PGLs [37–39]. RDBP methylation can trigger the PCCs/PGLs to be malignant and interestingly in 5 malignant patients of our study, the CpG sites in promoter regions harboring ATG were hypermethylated, similarly to Yong Joon Suh et al. findings [40]. RDBP methylation status can be the predictor of outcome in PCCs/PGLs and the potential of targeted therapy [37, 41]. Unlike SDHx which are methylated or mutated in several tumor types, the RDBP hypermethylation has been just reported in PCCs/PGLs and of the initial 86 candidate CpGs, from 47 genes, just RDBP was confirmed and could be used for stratifying patients according to the risk of developing metastases [36]. However, this type of test still has great limitations since it is only possible to be performed in centers with the technology available implicating in higher hospital structure and investment, which is not (yet) the reality of many centers all over the world.

Conclusions

In line with current literature, our results support both SDHC1 and RDBP2 hypermethylation in ctDNA of PCCs/PGLs as potential diagnostic and prognostic tools.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

SMKA and RH are principal investigators, LOR edited manuscript, MAP and ME analyses the data, SHN and SMT provide data and data curation, FKH wrote the manuscript.

Funding

National Institute for Medical. Research Development (NIMAD) Grant number 977079.

Data availability

All data are provided in the manuscript and additional will be provided by Dr. Fatemeh Khatami on request.

Declarations

Ethics approval and consent to participate

All procedures approved by National Institute for Medical Science Development Ethics Committee (IR.NIMAD.REC.1397.452).

Consent for publication

All patients signed the informed consent for publishing result of the study anonymously.

Competing interests

All authors declare there is not any conflict of interest for this publication.