Noninvasive detection of early nasopharyngeal carcinoma based on three gene methylation analysis in bilateral nasal swab samples processed automatically

doi:10.21203/rs.3.rs-5029369/v1

Background

Efforts have been made to improve the performance of nasopharyngeal carcinoma screening strategies, including EBV related biomarkers. However, the advances achieved still remain imperfect. DNA methylation occurs early in cancer development and can be used as potential diagnostic biomarker. This study aimed to investigate the diagnostic performance of three methylated genes in patients with nasopharyngeal carcinoma (NPC).

Methods

255 nasopharyngeal swabs and 35 plasma samples were collected from patients with newly diagnosed or treated NPC and healthy controls. Methylation-specific PCR (MSP) was carried out to evaluate the methylation status of three genes (SEPTIN9, RASSF1A, and H4C6) in the swabs and plasma samples. Methylation rates, sensitivity, and specificity of the candidate genes were calculated. Furthermore, the detectability of methylated genes in paired swabs and plasma was compared.

Results

In the nasopharyngeal swabs of patients with newly diagnosed NPC, the detection rate of methylated SEPTIN9, RASSF1A, and H4C6 was 88.2%, 92.9% and 71.8%, while the rate decreased to 54.3%, 42.9% and 45.7% by blood plasma, respectively. The sensitivity of detecting methylated SEPTIN9, RASSF1A, and H4C6 to differentiate between untreated NPC, and treated NPC plus healthy controls was 88%, 93%, and 72%. Methylated RASSF1A showed the best classification accuracy (AUC = 0.956). The detection rate of the target methylated genes was significantly lower in the paired plasma samples.

Conclusion

The detection of RASSF1A methylation through non-invasive nasopharyngeal cavity swab sampling demonstrates significant potential in NPC diagnosis.

Nasopharyngeal carcinoma

Non-invasive

Early diagnosis

Methylation-specific PCR

NPC showed more frequent methylated genes compared to normal individuals, suggesting their utility for NPC identification. RASSF1A methylation may be a diagnostic biomarker in swab samples that could be used for the clinical diagnosis of NPC and even for NPC screening from the perspective of cost and automation.

Posting statement

The paper is being posted on a non-commercial preprint server, including the DOI and a link to the article.

Nasopharyngeal carcinoma (NPC) is a malignancy originating from the nasopharyngeal epithelium. The stage of the tumor is a significant influencing factor of prognosis, with higher-stage NPC having poorer outcomes¹.Early detection and diagnosis are particular important in improving the cure rate and reducing recurrence of NPC.

Since infection with the highly pathogenic Epstein-Barr virus (EBV) is one of the most important etiologic factors of NPC, plasma EBV antibody and EBV DNA testing have been considered for the early detection of NPC. The two approaches are based on enzyme linked immunosorbent assay (ELISA) EBV antibody tests and quantitative polymerase chain reaction (qPCR), detecting the antibodies responses to EBV antigens and circulating free EBV DNA. Both methods have been proven to be effective for NPC screening in high-incidence populations. However, these biomarkers have unsatisfactory positive predictive values (PPV), leading unnecessary procedures and anxiety for subjects with false positive results. Additionally, the level of circulating free DNA fragments in the blood is influenced by tumor stage and biological factors², which may lead to limited value of plasma EBV DNA in detecting patients with early clinical stage disease. New modalities and effective biomarkers for NPC screening in needed to reduce false positives and improve diagnostic accuracy.

DNA methylation is an epigenetic mechanism of gene silencing of tumor suppressors, which has been recognized as promising targets for the creation of robust diagnostic and predictive biomarkers. It was reported that DNA methylation is a highly stable molecular feature in NPC³. Compared to the plasma, nasopharyngeal swab sampling can possibly bring higher DNA concentrations, improving the sensitivity of tumor marker tests. In fact, some genes hypermethylated in NPC tissues have been reported. For example, our previous study found that Septin 9, which is commonly used in the diagnosis of colorectal cancer, has significant difference in methylation levels between NPC tissues and benign ones⁴. In a meta-analysis, RASSF1A methylation is closely associated with the development of NPC⁵. H4C6, also known as HIST1H4F, is considered as a pan-oncogene and has been shown to be highly methylated in a variety of cancerous tissues including head/neck tumors but hypomethylated in benign tissues⁶. Therefore, this study investigated the diagnostic value of three methylated genes (SEPTIN9, RASSF1A, and H4C6) in nasopharyngeal swabs for NPC.

Study design and samples

This is a case control study with samples retrospectively collected from June 2023 till September 2023. The study flowchart is shown in Fig. 2. Three groups of subjects were screened: patients with newly diagnosed NPC, patients with complete regression of tumor and healthy controls. Patients with NPC were recruited at Sun yat-sen University Cancer Center (SYSUCC), and the healthy controls were recruited at physical examination center. This study was approved by the Human Ethics Committee of the SYSUCC, and all participants provided informed consent.

Nasopharyngeal swabs were collected from all subjects via blind brushing method using nasopharyngeal swabs and accompanying preservation solution containing guanidine salt (Guangdong Jingquan Medical Technology Co., Ltd). In fact, this kind of swab has been successfully used for sampling during the pandemic of Covid-19. The flocked swab was inserted via the nose until the nasopharyngeal cavity was reached. Subsequently, the brush was rotated several times and quickly removed. Immediately after sampling, the brush tip was cut and placed in tubes containing 3 mL of preservation solution. Bilateral sampling was performed and swabs from the same patient were put in one tube. Furthermore, 5ml of blood samples were collected in 35 patients and the plasma was isolated.

DNA extraction, bisulfite conversion and methylation-specific PCR (MSP)

Due to the relatively large sample size of 3mL of preservation solution containing cells from nasopharyngeal swabs, as well as the tedious process of methylation detection, we here applied the AutoPure-12 (Guangdong Bright-Innovation BioMed Co., Ltd., Shunde, China) system to process cell preservation solution. As shown in Figure. 1, the cassettes withinAutoPure-12contain various reagents that are used for DNA extraction and bisulfite modification. This platform can maximumly process 3.5 mL of liquid sample and 12 specimens at a time. The whole procedure including DNA extraction, bisulfite treatment, and purification/recovery of bisDNA can be done automatically within 3 h, resulting in a 100 uL of bisDNA solution. Since plasma contains abundant proteins, the high viscosity leads to the aggregation of magnetic bead even use of proteinase K for protein digestion. Therefore, for plasma samples, the process of DNA isolation, bisulfite conversion and purification/recovery of bisDNA remains manually.

Methylation fluorescence quantitative kit (Guangdong Bright-Innovation BioMed Co., Ltd.) was two-channel assay consisting of each target gene along with an internal control ACTB gene. The ACTB gene served as a reference gene and samples with a cycle threshold (Ct) value of ACTB above 35 were deemed “low quality” and excluded. Ct value of methylated gene less than 40 was judged as positive. The methylation level of the target gene was determined using the difference between two Ct values (ΔCt = Ct ACTB - Ct target gene), and a methylation score was calculated as 2^ΔCt x 100.

Epstein–Barr virus DNA load in plasma

DNA from plasma samples was extracted using the Nucleic Acid Extraction Kit (Sansure Biotech Inc.) following the protocol recommended by the manufacturer. The EBV DNA load was measured by real-time qPCR using the EBV DNA PCR Reagents Kit (Sansure Biotech Inc.).

Sample size calculation

This study is a diagnostic accuracy test evaluating the accuracy of methylated genes in diagnosing NPC. Based on our preliminary experiments, the anticipated sensitivity and specificity of methylated RASSF1A is 80% and 90%, respectively. With an allowable error of 0.1 and a two-sided α of 0.05, the sample size is calculated with PASS software to be N = 70 cases. The study goal was to recruit at least 78 patients with NPC, anticipating a 10% sampling or assay failure. Actually 85 untreated NPC were recruited in this study.

Statistical analysis

Descriptive statistical analysis was conducted for clinical and demographic characteristics. The one-way ANOVA and χ2 tests were used to analyze continuous and categorical variables, respectively. The sensitivity, specificity, Youden index, positive predictive value, and negative predictive value were calculated to evaluate the performance of methylation status in NPC diagnosis. The receiver-operating-characteristic (ROC) curves were plotted based on the gene methylation score and the area under the curve (AUC) was calculated. The Mc Nemar test was used in comparison of the methylation status of genes in paired plasma and swab samples. Statistical analyses were performed using RStudio (publicly available). Statistical significance was defined as the conventional P value of < 0.05 (two-sided test).

Patients Characteristics

After excluding samples which failed quality control or PCR assays, 255 swabs were collected from 85 patients with untreated NPC, 97 patients with complete responded NPC, and 73 healthy subjects without tumor disease (Fig. 2). Paired plasma samples were obtained in 35 patients with untreated NPC. The majority of the patients had advanced stage (III-IV). Tumor stage was balanced between the untreated and treated group. The plasma EBV DNA was highest in patients with newly diagnosed NPC, while it was undetectable in all of the healthy controls (Table 1).

Table 1

Characteristics of the subjects with untreated NPC, treated NPC and healthy controls.
Characteristics	Subjects
No.(%)/Median(Range)	untreated NPC (n = 85)	treated NPC (n = 97)	healthy CTL (n = 73)	p value
Gender				< 0.001
Male	56 (65.9)	68 (70.1)	29 (39.7)
Female	29 (34.1)	29 (29.9)	44 (60.3)
Age (mean (SD))	47.29 (12.72)	46.53 (11.64)	29.29 (5.70)	< 0.001
T stage				0.058*
T1-2	10 (11.8)	23 (23.7)	NA
T3-4	75 (88.2)	74 (76.3)	NA
N stage				0.921*
N0-1	37 (43.5)	44 (45.4)	NA
N2-3	48 (56.5)	53 (54.6)	NA
M stage				1*
M0	79 (92.9)	91 (93.8)	NA
M1	6 (7.1)	6 (6.2)	NA
Stage				0.343*
stage1-2	6 (7.1)	12 (12.4)	NA
stage3-4	79 (92.9)	85 (87.6)	NA
plasma EBV DNA** (cp/ml)	238.50 [0.00, 1325.00]	198.50 [20.50, 2755.00]	0.00 [0.00, 0.00]	< 0.001
plasma EBV DNA**				< 0.001
=0	23 (27.4)	8 (25.0)	21 (100)
> 0	61 (72.6)	24 (75.0)	0 (0)
SEPTIN9 methylation				< 0.001
No	10 (11.8)	55 (56.7)	45 (61.6)
Yes	75 (88.2)	42 (43.3)	28 (38.4)
RASSF1A methylation				< 0.001
No	6 (7.1)	94 (96.9)	73 (100.0)
Yes	79 (92.9)	3 (3.1)	0 (0.0)
H4C6 methylation				< 0.001
No	24 (28.2)	91 (93.8)	60 (82.2)
Yes	61 (71.8)	6 (6.2)	13 (17.8)
SEPTIN9 methylation score	0.39 [0.04, 3.35]	0.00 [0.00, 0.03]	0.00 [0.00, 0.01]	< 0.001
RASSF1A methylation score	0.22 [0.02, 4.12]	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	< 0.001
H4C6 methylation score	0.06 [0.00, 0.56]	0.00 [0.00, 0.00]	0.00 [0.00, 0.00]	< 0.001
*chi-square test between patients with untreated NPC or treated NPC.
**available data in 137 subjects

Diagnostic performance of SEPTIN9, RASSF1A and H4C6 methylation

The detection rate and methylation scores are shown in Table 1. In swabs of newly diagnosed NPC, the highest methylated gene detection rate was for RASSF1A (92.9%), and the lowest for H4C6 (71.8%). Methylated RASSF1A (mRASSF1A) was detected in 3 (3.1%) patients after treatment, and was undetectable in all of healthy controls. Sensitivity and specificity of detectable mRASSF1A distinguishing between untreated NPC, and treated NPC plus healthy controls were 93% and 98%, respectively (Table 2). The sensitivity of the mSEPTIN9 and mH4C6 was 88% and 72%, respectively, and the specificity was 59% and 89%. Besides, Table S1 showed the performance of these methylated markers distinguishing patients with NPC from heathy controls. Both of the sensitivity and specificity of the mRASSF1A were the highest.

Table 2

The performance of variables for distinguishing between untreated NPC, and treated NPC plus heathy controls.
Variable	Sensitivity	Specificity	Youden index	NPV	PPV
SEPTIN9 methylation	0.88	0.59	0.47	0.91	0.52
RASSF1A methylation	0.93	0.98	0.91	0.97	0.96
H4C6 methylation	0.72	0.89	0.61	0.86	0.76
plasma EBV DNA	0.73	0.55	0.27	0.56	0.72

ROC curves were constructed by the result of methylated gene detection (Fig. 3). ROC curve analysis showed AUCs of 74.9%, 96.5% and 77.0% by detecting methylated SEPTIN9, RASSF1A and H4C6 to differentiate between untreated NPC and heathy controls. And it also showed AUCs of 73.5%, 95.6% and 80.3% to distinguish between untreated NPC, and treated NPC plus heathy controls by detecting those genes. Additionally, methylation of RASSF1A showed the highest NPV (97%) and PPV (96%). The plasma EBV DNA was negative in 23 (27.4%) patients with untreated NPC, 8 (25%) patients with treated NPC and 21 (100%) healthy controls. The sensitivity and specificity of qPCR detectable plasma EBV DNA for distinguishing between untreated NPC, and treated NPC plus healthy subjects were 73% and 55%, respectively.

Methylation difference in paired nasopharyngeal swabs and plasma

Paired plasma and swab samples were obtained from 35 patients. The Mc Nemar test showed significant lower detection rate of methylated genes in the plasma samples (Table 3). Methylated SEPTIN9, RASSF1A and H4C6 was detected in 19 (54.3%), 15 (42.9%) and 16 (45.7%) plasma samples, respectively. Moreover, the methylation score was compared between the plasma and swab samples. Significant higher scores of methylated SEPTIN9 (P = 0.04) and RASSF1A (P = 0.02) were found in the swabs (Fig. 4).

Table 3. Identification of methylated genes in the matched swab or blood samples (paired chi-square test).

Detection of methylated SEPTIN9

		blood
		negative	positive	overall	P value
swab	negative	0	2	2	<0.01
swab	positive	16	17	33	<0.01
	overall	16	19	35

Detection of methylated RASSF1A

		blood
		negative	positive	overall	P value
swab	negative	3	0	3	<0.001
swab	positive	17	15	32	<0.001
	overall	20	15	35

Detection of methylated H4C6

		blood
		negative	positive	overall	P value
swab	negative	3	3	6	<0.01
swab	positive	16	13	29	<0.01
	overall	19	16	35

Early diagnosis of NPC is quite challenging due to its anatomical peculiarity and the lack of obvious clinical signs in the early stages. At present, nasopharyngeal endoscopy combined with pathological examination of suspected cases remains the gold standard way for the diagnosis of NPC, although assessment of clinical symptoms and family history provides good clues which sometimes are not specific. However, this way is mainly applicable to patients with suspected NPC and is not suitable for early screening and especially for large-scale screening due to it is costive, invasive and inconvenient. In addition, although the current laboratory blood tests (e.g., EBV related testing) can provide high specificity and sensitivity, they still have low positive predictive values (PPVs) and produce more false-positive results, which leads to repeated nasal endoscopic examination, biopsy and long-term follow-up. DNA methylation is an epigenetic mechanism of gene silencing of tumor suppressors, which has been recognized as promising targets for the creation of robust diagnostic and predictive biomarkers, more important was that the sample can be collected with noninvasive nasopharyngeal swabs.

In the present study, we investigated the value of methylated SEPTIN9, RASSF1A and H4C6 in distinguishing patients with NPC from normal controls. The three genes indicated a good potential to distinguish NPC from nasopharyngitis and health controls in a previous study using a home-made qMS-PCR kit⁷. In this study, we used commercial kits containing the qMS-PCR assays of the three genes. We found that the methylation status and score of RASSF1A have high sensitivity and specificity in diagnosing NPC. Moreover, the methylation level in nasopharyngeal swab samples was higher than that in the plasma, suggesting that swab might be a better modality in NPC screening. In our pilot study, the sensitivity of bilateral nasal swab sampling was significantly higher than that of unilateral nasal swab sampling (data not shown). Because if nasopharyngeal malignancy is one-sided, and unilateral sampling will miss the contralateral tumor nidus.

Downregulated expression of RASSF1A has been associated with pathogenesis of cancer^8,9. Introducing RASSF1A exogenously has been shown to induce growth inhibition and apoptosis in NPC cell lines, suggesting its tumor suppressor function¹⁰. Previous studies with NPC tissue have implied that hypermethylation may contribute to the transcriptional inactivation of the RASSF1A gene¹¹. In addition, hypermethylation of RASSF1A was correlated with advanced tumor stage and the presence of lymph node metastasis in patients with NPC¹². Therefore, methylation tests of RASSF1A are potentially valuable in improving the sensitivity for NPC detection. Our results also showed that the diagnostic sensitivity of mRASSF1A was over 90%, suggesting that it may aid in the screening of NPC. mRASSF1A was detected in 3 of 97 patients with complete responded NPC. These 3 patients in complete remission were with Stage III or IVa NPC, and completed treatment at least 1 year ago. Up to the date of preparing this manuscript, there were no signs of recurrence or metastasis 9 months after the methylation tests. We will continue following these patients.

A swab involves applying a brush to collect specimens of NPC in a painless and non-invasive manner, which is a relatively simple process and can be completed quickly and easily. In comparison to plasma, nasal swab can be collected without specialized training from a diverse range of populations, making it an excellent option for self-sampling in NPC screening. After the sample has been collected, it can be sent to a lab for further testing and analysis. Swabs can be used to detect the same biomarkers that were usually obtained from plasma specimens, such as EBV DNA and microRNAs^5,13,14.Our findings indicated that the detection rate of methylated genes is significantly higher than that in the plasma for patients with NPC, probably because swab samples are close to the tumor lesion. In addition, the ingredients is relatively simple after elution of nasal swabs with preservation solution compared to plasma samples that containing abundant proteins. Thus, the nasopharyngeal epithelial cells within preservation buffer can be processed automatically for DNA extraction, bisulfite treatment, and purification/recovery of bisDNA, otherwise, manual operation would be very tedious.

We must admit that the present study is based on a retrospective cohort and lack of robust power to clarify the practice value in clinical situation. We have initiated a prospective clinical study and already got registered on clinicaltrials.gov (NCT06367049).

NPC showed more frequent methylated genes compared to normal individuals, suggesting their utility for NPC identification. RASSF1A methylation may be a diagnostic biomarker in swab samples that could be used for the clinical diagnosis of NPC and even for NPC screening from the perspective of cost and automation.

AUC: Area under the curve; EBV: Epstein-Barr virus; ELISA: Enzyme linked immunosorbent assay; mH4C6: Methylated H4C6; mRASSF1A: Methylated RASSF1A; mSEPTIN9: Methylated SEPTIN9; MSP:Methylation-specific PCR; NPC: Nasopharyngeal carcinoma; NPV: Negative predictive values; PPV: Positive predictive values; qPCR: Quantitative polymerase chain reaction; ROC: Receiver-operating-characteristic.

Ethics approval and consent to participate

We have initiated a prospective clinical study that has been registered on clinicaltrials.gov (NCT06367049) and approved by Ethics Committee of Sun Yat-Sen University Cancer Center (B2023-716-01).

Consent for publication

Not applicable.

Availability of data and materials

The data of this research was submitted to RDD (Research Data Deposit) platform (http://www.researchdata.org.cn), and was currently under review. A unique RDD identification code will be provided later.

All authors agree to share the data of this review, which can be obtained by contacting the corresponding authors. Email: [email protected]

Competing interests

All authors disclosed no relevant relationships.

Funding

This study was supported by GuangDong Basic and Applied Basic Research Foundation（2021A1515220021）

Authors' contributions

All the authors contributed in the preparation of this work. Q. Z. H. , D. H. and Z. X. were responsible for the collection of nasopharyngeal swabs. Q. Z. H. , L. L. X. and G. Q. L. collected clinical data from healthy controls and nasopharyngeal cancer patients. Z. S. took charge of the methylation detection. Q. Z. H. and C. S. Y. were responsible for data analysis, writing, edit and revision of the body text. H. Y. J. , Z. L. , and Q. Z. H. were drafted and revised the article. H. Y. J. , Z. L. , Q. Z. H. , C. S. Y. and Z. S. were responsible for the theme, final editing, and preparation of the manuscript for submission. H. Y. J. , Z. L. , C. M. Y. and H. P. Y. critically revised the manuscript. All authors read and approved the final manuscript.

Acknowledgements

Not applicable.

Zhang MX, Li J, Shen GP, et al. Intensity-modulated radiotherapy prolongs the survival of patients with nasopharyngeal carcinoma compared with conventional two-dimensional radiotherapy: A 10-year experience with a large cohort and long follow-up. Eur J Cancer. Nov 2015
Yuwono NL, Warton K, Ford CE. The influence of biological and lifestyle factors on circulating cell-free DNA in blood plasma. Elife. Nov 9 2021
Jiang W, Liu N, Chen XZ, et al. Genome-Wide Identification of a Methylation Gene Panel as a Prognostic Biomarker in Nasopharyngeal Carcinoma. Mol Cancer Ther. Dec 2015
Lyu JY, Chen JY, Zhang XJ, et al. Septin 9 Methylation in Nasopharyngeal Swabs: A Potential Minimally Invasive Biomarker for the Early Detection of Nasopharyngeal Carcinoma. Dis Markers. 2020
Ye M, Huang T, Ni C, Yang P, Chen S. Diagnostic Capacity of RASSF1A Promoter Methylation as a Biomarker in Tissue, Brushing, and Blood Samples of Nasopharyngeal Carcinoma. EBioMedicine. 2017
Dong S, Li W, Wang L, et al. Histone-Related Genes Are Hypermethylated in Lung Cancer and Hypermethylated HIST1H4F Could Serve as a Pan-Cancer Biomarker. Cancer Res. 2019
Qian T, Zhou Z, Zhang Q, Liou YL, Zhou H. SEPT9, H4C6, and RASSF1A methylation in nasopharyngeal swabs: A reflection of potential minimally invasive biomarkers for early screening of nasopharyngeal cancer. Medicine (Baltimore). 2023
Dubois F, Bergot E, Zalcman G, Levallet G. RASSF1A, puppeteer of cellular homeostasis, fights tumorigenesis, and metastasis-an updated review. Cell Death Dis. Dec 5 2019
Schmidt ML, Hobbing KR, Donninger H, Clark GJ. RASSF1A Deficiency Enhances RAS-Driven Lung Tumorigenesis. Cancer Res. May 15 2018
Wang T, Liu H, Chen Y, Liu W, Yu J, Wu G. Methylation associated inactivation of RASSF1A and its synergistic effect with activated K-Ras in nasopharyngeal carcinoma. J Exp Clin Cancer Res. Dec 30 2009
Lo KW, Kwong J, Hui AB, et al. High frequency of promoter hypermethylation of RASSF1A in nasopharyngeal carcinoma. Cancer Res. May 15 2001
Fendri A, Masmoudi A, Khabir A, et al. Inactivation of RASSF1A, RARbeta2 and DAP-kinase by promoter methylation correlates with lymph node metastasis in nasopharyngeal carcinoma. Cancer Biol Ther. Mar 2009
Zheng XH, Li XZ, Zhou T, et al. Quantitative detection of Epstein-Barr virus DNA methylation in the diagnosis of nasopharyngeal carcinoma by blind brush sampling. Int J Cancer. Jun 15 2023
Zhang PF, Zheng XH, Li XZ, et al. Nasopharyngeal brushing: a convenient and feasible sampling method for nucleic acid-based nasopharyngeal carcinoma research. Cancer Commun (Lond). Apr 12 2018

No competing interests reported.

TableS1.docx

Noninvasive detection of early nasopharyngeal carcinoma based on three gene methylation analysis in bilateral nasal swab samples processed automatically

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Novelty and Impact

Introduction

Materials and Methods

Study design and samples

DNA extraction, bisulfite conversion and methylation-specific PCR (MSP)

Epstein–Barr virus DNA load in plasma

Sample size calculation

Statistical analysis

Results

Patients Characteristics

Diagnostic performance of SEPTIN9, RASSF1A and H4C6 methylation

Methylation difference in paired nasopharyngeal swabs and plasma

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1