Skip to main content
NCBI home page
Cancers logoLink to Cancers
. 2023 Aug 3;15(15):3959. doi: 10.3390/cancers15153959

Methylated Circulating Tumor DNA in Blood as a Tool for Diagnosing Lung Cancer: A Systematic Review and Meta-Analysis

Morten Borg 1,†, Sara Witting Christensen Wen 2,3,*,†, Rikke Fredslund Andersen 4, Signe Timm 2,3, Torben Frøstrup Hansen 2,3, Ole Hilberg 1,3
Editor: Samir M Hanash
PMCID: PMC10417522  PMID: 37568774

Abstract

Simple Summary

Lung cancer is the leading cause of cancer-related deaths and has a poor prognosis. Early detection could improve survival for this large patient group. Certain genes are more frequently changed by methylation in cancer cells compared to healthy cells. This methylated tumor DNA is present in the blood in small quantities and has been suggested as a diagnostic biomarker in many diseases, including lung cancer. The aim of the present literature review was to identify and collate the current evidence on methylated circulating tumor DNA in blood samples as a diagnostic tool for lung cancer. A systematic collection and presentation of the existing evidence will aid future research in this field.

Abstract

Lung cancer is the leading cause of cancer-related deaths, and early detection is crucial for improving patient outcomes. Current screening methods using computed tomography have limitations, prompting interest in non-invasive diagnostic tools such as methylated circulating tumor DNA (ctDNA). The PRISMA guidelines for systematic reviews were followed. The electronic databases MEDLINE, Embase, Web of Science, and Cochrane Library were systematically searched for articles. The search string contained three main topics: Lung cancer, blood, and methylated ctDNA. The extraction of data and quality assessment were carried out independently by the reviewers. In total, 33 studies were eligible for inclusion in this systematic review and meta-analysis. The most frequently studied genes were SHOX2, RASSF1A, and APC. The sensitivity and specificity of methylated ctDNA varied across studies, with a summary sensitivity estimate of 46.9% and a summary specificity estimate of 92.9%. The area under the hierarchical summary receiver operating characteristics curve was 0.81. The included studies were generally of acceptable quality, although they lacked information in certain areas. The risk of publication bias was not significant. Based on the findings, methylated ctDNA in blood shows potential as a rule-in tool for lung cancer diagnosis but requires further research, possibly in combination with other biomarkers.

Keywords: circulating tumor DNA, ctDNA, methylated tumor DNA, liquid biopsy, lung cancer

1. Introduction

Lung cancer is the leading cause of cancer-related deaths worldwide [1]. The survival rate of lung cancer patients is strongly correlated with the stage of the disease at diagnosis [2]. Early detection is critical for improving patient outcomes, as it increases the likelihood of successful treatment and long-term survival [3]. Unfortunately, a large proportion of lung cancer cases are diagnosed at an advanced stage [4], and hence, there is a desire for effective screening methods for lung cancer [5].

Current screening methods using computed tomography (CT) scans have resulted in a stage shift towards early-stage disease and reduced mortality [6,7]. Low-dose CT screening is now implemented in several countries, among them the US [8,9]. However, several limitations are present, including high costs, radiation exposure, and low adherence to follow-up scans [10,11,12]. Furthermore, the management of screen-detected pulmonary nodules is time-consuming [13] and associated with anxiety and depression among patients [14].

In recent years, there has been a growing interest in the use of circulating tumor DNA (ctDNA) as a non-invasive diagnostic tool for cancer [15,16]. ctDNA is released into the bloodstream by tumor cells and can be detected and quantified through molecular techniques [17]. Many assays are based on polymerase chain reaction (PCR), such as real-time or quantitative PCR and, in recent years, also digital PCR [18]. Other methods for quantitating ctDNA include sequencing techniques such as pyrosequencing or next-generation sequencing (NGS) methods [19]. Methylated ctDNA, which refers to circulating DNA with methyl-groups added to CpGs in a cancer-specific manner, has been shown to be a promising biomarker for cancer diagnosis and prognosis [20].

Several studies have investigated the diagnostic accuracy of methylated ctDNA in the blood of a variety of genes for lung cancer diagnosis [21,22,23]. The results have shown large variations in sensitivity and specificity and also in the choice of biological specimen and assay type [24]. An up-to-date systematic review and meta-analysis of the available evidence is necessary to provide a more comprehensive understanding of the diagnostic performance of methylated ctDNA for the diagnosis of lung cancer.

Hence, the aim of this systematic review and meta-analysis on methylated ctDNA for lung cancer diagnosis is to systematically identify and analyze all relevant studies that have investigated the diagnostic accuracy of methylated ctDNA in blood for lung cancer detection.

In total, 33 studies were eligible for inclusion in this review. The sensitivity and specificity of methylated ctDNA varied across studies, with a summary sensitivity estimate of 46.9% and a summary specificity estimate of 92.9%. The area under the hierarchical summary receiver operating characteristics curve was 0.81.

2. Materials and Methods

The review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement [25]. The study was registered on the Prospero website (registration number CRD42022361536, University of York, York, UK) on 1 October 2022.

2.1. Search Strategy and Screening

The electronic databases MEDLINE, Embase, Web of Science, and Cochrane Library were systematically searched for articles. The search string contained three main topics: (I) Lung cancer, (II) blood, and (III) methylated circulating tumor DNA. All relevant Subject Headings and free text terms were included in each topic and combined with the Boolean operator ‘OR’. The three main topics were combined with the Boolean operator ‘AND’. The detailed search string for each database can be accessed in the Supplementary Materials. We did not include the topic ‘Diagnosis’ in the search because preliminary searches showed that relevant studies would then be omitted. The searches contained no restrictions regarding language, article type, or date of publication since such restrictions could potentially limit study retrieval. The searches were conducted between 16 November 2022 and 12 December 2022. Forward and backward citation searches were performed on all included studies using the Web of Science to extract reference lists and citations. These searches were performed on 24 April 2023.

All references identified from each database were imported into Covidence (Covidence, Melbourne, Australia), with duplicates automatically removed. The initial screening of the title and abstract required only one vote and was performed by one of the three main authors (MB, SW, or OH).

2.2. Eligibility Criteria

Full-text review of all potentially relevant studies was performed independently by two reviewers (MB and SW) in a blinded manner using the Covidence software. In case of disagreement, the study was discussed between the main reviewers until consensus. The inclusion criteria were as follows: (I) Adults diagnosed with lung cancer or undergoing diagnostic work-up for lung cancer; (II) plasma or serum sample collected for quantitative analysis of methylated circulating tumor DNA; (III) tumor cytology or histopathology as a reference standard; (IV) outcome reported as diagnostic sensitivity and specificity or a contingency table with enough data to calculate these diagnostic measures for each reported target. The exclusion criteria were: (I) Case-reports, conference papers, editorials, notes, and literature reviews; (II) studies not in English; (III) pure in silico analyses.

2.3. Data Extraction and Quality Assessment

A data extraction form was developed in Covidence and pilot tested with seven studies. The final data collection form included the first author’s last name, publication year, geographic region, study aims, study design, type of blood sample, analysis method, type of reference standard, number of cases, and number of controls. Study outcomes included gene names, sensitivity and specificity, the area under the curve (AUC), and a contingency table of true positives, false positives, true negatives, and false negatives. Quality assessment was performed using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool [26].

Both data extraction and quality assessment were performed independently by two reviewers (MB and SW), and any disagreements were discussed until consensus with the option to consult a third reviewer (OH) in case of persisting disagreement. Data extraction issues regarding the analytical methodology were supervised and settled by an expert (RFA).

2.4. Statistical Analysis

Data were synthesized in the form of a summary table, including all relevant studies. If diagnostic sensitivity and/or specificity were not reported, these values were calculated using the following formulas: Sensitivity = true positive/(true positive + false negative). Specificity = true negative/(true negative + false positive). The STATA command metandi [27] was used for hierarchical summary receiver operating characteristics (HSROC) plots and for calculating summary effect measures. This method is based on a two-level mixed-effect logistic regression model with independent binomial distribution. The STATA command midas [28] was used for forest plots, including the 95% confidence interval (CI) for each study estimate and for Deek’s funnel plot and corresponding test for assessing the risk of publication bias. The level of significance was set at 0.05. STATA BE version 17 (StataCorp LLC, College Station, TX, USA) was used for statistical analyses. Microsoft Excel (Microsoft Corporation, Redmond, WA, USA) was used for forest plot graphics.

3. Results

3.1. Search Results and Eligibility

The study selection process is illustrated by a PRISMA flow chart (Figure 1). Searches performed in MEDLINE, Embase, Web of Science, and Cochrane Library resulted in 15,211 records. Forward and backward citation searching identified a further 2425 references. Screening for duplicates removed 5738 records, leaving 11,898 records for title and abstract screening. The authors identified 241 records for full-text review, and 33 studies fulfilled all eligibility criteria and were included in the review.

Figure 1.

Figure 1

Open in a new tab

PRISMA flow chart.

3.2. Characteristics of Included Studies

Of the thirty-three studies included in this review, fifteen studies originated from Asia, twelve from the EU, five from Northern America, and one from Russia. The majority of the studies were of a case-control design (29/34), while the remaining five were cohort studies. We did not identify any randomized clinical trials. The cases were compared to matched controls in nine studies, unmatched healthy controls in thirteen studies, unmatched patients with benign diseases in five studies, and the remaining six studies were made up of the non-cancer patients from the cohort studies and one study with a control group consisting of healthy subjects, benign diseases and prostate cancer [29]. The number of cases reported ranged from 13 [30] to 188 [29], while the number of controls ranged from 11 [31] to 155 [29]. The detailed study characteristics can be viewed in Table 1.

Table 1.

Study characteristics.

Study ID Region Study Design Cases Histology Stage Controls Cohort Number of Cases Number of Controls Reference Standard
Usadel, 2002 [32] Northern America Case-control study Retrospectively selected cases LUSC 35/99 * (35%), LUAD 47/99 (48%), other 17/99 (17%) I 53/99 * (54%), II 23/99 (23%), III 17/99 (17%), IV 6/99 (6%) Unmatched healthy controls Training 71 serum, 33 plasma (15 matched) 50 Histopathology or cytology.
Ostrow, 2009 [30] Northern America Case-control study Retrospectively selected cases LUSC 6/13 (46%), LUAD 1/13 (8%), other 6/13 (46%) Not reported Matched on certain characteristics Training 13 24 Tumor tissue biopsy/histopathology
LUSC 7/70 (10%), LUAD 47/70 (67%), other 16/70 (23%) I 49/70 (70%), II 2/70 (3%), III 10/70 (14%), IV 4/70 (6%), no stage 5/70 (7%) Validation 70 23 with nodules + 80 smokers with no nodules
Zhang, 2010 A [33] China Case-control study Retrospectively selected cases LUSC 36/78 (46%), LUAD 30/78 (38%), other 12/78 (15%) I–II 58/78 (74%), III–IV 20/78 (26%) Unmatched healthy controls Training 78 50 Histopathology or cytology
Zhang, 2010 B [34] China Case-control study Retrospectively selected cases LUSC 36/78 (46%), LUAD 30/78 (38%), other 12/78 (15%) I–II 58/78 (74%), III–IV 20/78 (26%) Unmatched healthy controls Training 78 50 Tumor tissue biopsy/histopathology
Begum, 2011 [35] Northern America Case-control study Retrospectively selected cases LUSC 26/76 (34%), LUAD 36/76 (47%), other 14/76 (18%) I 41/76 (54%), II 17/76 (22%), III 11/76 (14%), IV 5/76 (7%), unknown 2/76 (3%) Matched on certain characteristics Training 76 30 Histopathology or cytology
Kneip, 2011 [29] EU Case-control study Retrospectively selected cases LUSC 38/188 (20%), LUAD 31/188 (16%), SCLC 15/188 (8%), other/unknown 104/188 (55%) I 37/188 (20%), II 29/188 (15%), III 53/188 (28%), IV 42/188 (22%), unknown 27/188 (14%) Combination of healthy, benign and prostate cancer Training 188 155 Histopathology or cytology
Ponomaryova, 2011 [23] Other: Russia Case-control study Retrospectively selected cases LUSC 34/52 (65%), LUAD 18/52 (35%) I–II 25/52 (48%), III–IV 27/52 (52%) Unmatched healthy controls Training 52 26 Histopathology or cytology
Vinayanuwattikun, 2011 [36] Other: Asian country Case-control study Retrospectively selected cases NSCLC, not further described The whole cohort was described as ‘advanced’. Matched on certain characteristics Training 38 52 Tumor tissue biopsy/histopathology
Balgkouranidou, 2014 A [37] EU Case-control study Retrospectively selected cases LUSC 23/44 # (52%), LUAD 20/44 (45%), missing 1/44 (2%) I 14/44 # (32%), II–III 29/44 (66%), missing 1/44 (2%) Unmatched healthy controls Training 48 24 (same used for training and validation) Histopathology or cytology
LUSC 24/74 (32%), non-squamous 50/74 (68%) IV 74/74 (100%) Validation 74 24 (same used for training and validation)
Powrozek, 2014 [38] EU Case-control study Retrospectively selected cases LUSC 20/70 (29%), LUAD 20/70 (29%), SCLC 23/70 (33%), other 7/70 (10%) I 0/47 € (0%), II 7/47 (15%), III 23/47 (49%), IV 17/47 (36%) Matched on certain characteristics Training 70 100 Not described
Gao, 2015 [39] China Cohort study Diagnostic work-up for LC LUSC 23/58 (40%), LUAD 18/58 (31%), SCLC 2/58 (3%), other 15/58 (26%) All were early-stage lung cancer (T1a–T2a) Non-cancer participants who underwent diagnostic work-up Training 58 plasma
40 serum		 31 with benign disease, 23 healthy Histopathology or cytology
Balgkouranidou, 2016 B [22] EU Case-control study Retrospectively selected cases LUSC 21/44 # (48%), LUAD 22/44 (50%), missing 1/44 (2%) I 14/44 # (32%), II–III 29/44 (66%), missing 1/44 (2%) Unmatched healthy controls Training 48 49 (same used for training and validation) Tumor tissue biopsy/histopathology
LUSC 24/74 (32%), non-squamous 50/74 (68%) IV 74/74 (100%) Validation 74 49 (same used for training and validation)
Powrozek, 2016 [40] EU Case-control study Retrospectively selected cases LUSC 20/65 (31%), LUAD 22/65 (34%), SCLC 19/65 (29%), other 4/65 (6%) I 0/46 (0%), II 7/46 (15%), III 22/46 (48%), IV 17/46 (37%), limited 9/19 (47%), extensive 10/19 (53%) Unmatched healthy controls Training 65 95 Tumor tissue biopsy/histopathology
Powrozek, 2016 [41] EU Case-control study Retrospectively selected cases LUSC 30/70 (43%), LUAD 25/70 (36%), SCLC 15 (21%) I 8/55 # (15%), II 12/55 (22%), III 19/55 (35%), IV 16/55 (29%) Unmatched healthy controls Training 70 80 Surgery specimen/histopathology
Aslam, 2017 [42] Other: Asian country Case-control study Retrospectively selected cases LUSC 19/34 (56%), LUAD 7/34 (21%), other 8/34 (24%) Not reported Matched on certain characteristics Training 34 34 Tumor tissue biopsy/histopathology
Hulbert, 2017 [43] Northern America Cohort study Diagnostic work-up for LC LUSC 26/150 (17%), LUAD 121/150 (81%), other 3/150 (2%) I 136/150 (91%), II 14/150 (9%), III 0/150 (0%), IV 0/150 (0%) Non-cancer participants who underwent diagnostic work-up Training 125 50 Surgery specimen/histopathology
Ooki, 2017 [44] Northern America Case-control study Retrospectively selected cases LUAD 43/43 (100%) I 43/43 (100%) Matched on certain characteristics Training 43 LUAD 42 (same used for training and validation) Histopathology or cytology
LUSC 40/40 (100%) I 40/40 (100%) Validation 40 LUSC 42 (same used for training and validation)
Nunes, 2019 [45] EU Case-control study Retrospectively selected cases LUSC 42/129 (33%), LUAD 65/129 (50%), SCLC 19/129 (15%), other 3/129 (2%) I 15/129, II 11/129, III 27/129, IV 76/129 Non-cancer participants who underwent diagnostic work-up Training 129 28 Histopathology or cytology
Villalba, 2019 [46] EU Case-control study Retrospectively selected cases LUSC 38/89 (43%), LUAD 51/89 (57%) I 8/89 (9%), II 8/89 (9%), III 19/89 (21%), IV 52/89 (58%), missing 2/89 (2%) Matched on certain characteristics Training 89 25 Surgery specimen/histopathology
Yang, 2019 [31] China Cohort study Diagnostic work-up for LC LUSC 12/39 (31%), LUAD 25/39 (64%), other 2/39 (5%) I 39/39 (100%) Non-cancer participants who underwent diagnostic work-up Training 39 11 Surgery specimen/histopathology
Chen, 2020 [47] China Cohort study Diagnostic work-up for LC LUSC 22/163 (13%), LUAD 139/163 (85%), other 2/163 (1%) I 163/163 (100%) Non-cancer participants who underwent diagnostic work-up Training 163 83 Surgery specimen/histopathology
Huang, 2020 [48] China Cohort study Diagnostic work-up for LC LUSC 15/104 (14%), LUAD 53/104 (51%), SCLC 3/104 (3%), other 1/104 (1%), missing 32/104 (31%) I 48/104 (46%), II 15/104 (14%), III 20/104 (19%), IV 21/104 (20%) Unmatched patients with benign diseases Training 104 36 with benign disease, 50 healthy Surgery specimen/histopathology
LUSC 4/19 (21%), LUAD 14/19 (74%), other 1/19 (5%) I 12/19 (63%), II 4/19 (21%), III 3/19 (16%) Validation 19 11
Li, 2020 [49] China Case-control study Retrospectively selected cases LUSC 24/48 (50%), LUAD 18/48 (38%), other 6/48 (13%) I–II 15/48 (31%), III–IV 33/48 (69%) Unmatched healthy controls Training 48 51 Histopathology or cytology
Wen, 2020 [50] EU Case-control study Retrospectively selected cases LUAD 48/48 (100%) III 3/48 (6%), IV 45/48 (94%) Unmatched healthy controls Training 48 100 Histopathology or cytology
Xu, 2020 [51] China Case-control study Retrospectively selected cases LUSC 28/302 (9%), LUAD 236/302 (78%), SCLC 32/302 (11%), other 6/302 (2%) I 68/302 (23%), II 62/302 (21%), III 72/302 (24%), IV 100/302 (33%) Matched on certain characteristics Training 302 153 Not described
Mastoraki, 2021 [52] EU Case-control study Retrospectively selected cases LUSC 19/48 (40%), LUAD 28/48 (58%), other 1/48 (2%) I–II 28/48 (58%), III–IV 13/48 (27%), missing 7/48 (15%) Matched on certain characteristics Training 48 early stage 60 (same used for training and validation) Histopathology or cytology
Not available IV 91/91 (100%) Validation 91 stage IV 60 (same used for training and validation)
Park, 2021 [53] Other: Asian country Case-control study Retrospectively selected cases Not available Not available Unmatched healthy controls Training 64 64 Tumor tissue biopsy/histopathology
Szczyrek, 2021 [54] EU Case-control study Diagnostic work-up for LC LUSC 34/101 (34%), LUAD 52/101 (51%), SCLC 8/101 (8%), other 7/101 (7%) IA–IIIA 27/101 (27%), IIIB–IV 66/101 (65%), missing 8/101 (8%) Unmatched healthy controls Training 101 45 Tumor tissue biopsy/histopathology
Kim, 2022 [55] Other: Asian country Case-control study Diagnostic work-up for LC LUSC 30/72 (42%), LUAD 31/72 (43%), other 11/72 (15%) I 41/72 (57%), II 26/72 (36%), III 3/72 (4%), IV 2/72 (3%) Unmatched patients with benign diseases Training 72 61 Surgery specimen/histopathology
Palanca-Ballester, 2022 [56] EU Case-control study Retrospectively selected cases LUSC 13/44 (30%), LUAD 31/44 (70%) I 4/44 (9%), II 7/44 (16%), III 3/44 (7%), IV 30/44 (68%) Unmatched patients with benign diseases Training 44 39 Other: Histopathology or cytology
Vo, 2022 [57] Other: Asian country Case-control study Retrospectively selected cases Not available I 2/30 (7%), II 8/30 (27%), III 15/30 (50%), IV 5/30 (17%) Unmatched healthy controls Training 30 27 Other: Histopathology or cytology.
Zeng, 2022 [58] China Case-control study Retrospectively selected cases LUSC 58/121 (48%), LUAD 63/121 (52%) I–II 78/121 (64%), III–IV 43/121 (36%) Unmatched patients with benign diseases Training 121 121 Surgery specimen/histopathology
Zhang, 2022 [59] China Case-control study Retrospectively selected cases LUSC 8/23 (35%), LUAD 10/23 (43%), SCLC 5/23 (22%) I–II 2/23 (9%), III–IV 21/23 (91%) Unmatched patients with benign diseases Training 23 56 Histopathology or cytology
Open in a new tab

Characteristics of all studies included in the review. Studies are arranged according to year of publication, starting with the oldest study and then alphabetically if more studies were published in the same year. Study cohorts used in more than one publication are labeled with A, B, etc. Independent training and validation cohorts are reported in separate rows. Study ID consists of the first author’s last name and the year of publication, followed by the reference. Region refers to the geographical region in which the study was performed. LC, lung cancer; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; NSCLC, non-small cell lung cancer; SCLC, small cell lung cancer. * A subset of patients had blood samples collected, but data on histology and stage were only reported for the whole patient cohort. # The number of plasma samples was reported as n = 48; however, in the table reporting clinicopathological data for BRMS1 methylated, only n = 44 patients were reported. € The detailed disease stage was only reported for the NSCLC patients.

3.3. Sample Type and Analysis Method

The analysis details are summarized in Table 2. The majority of the studies (28/33) used plasma for detecting ctDNA; three studies tested both plasma and serum [32,35,39], one study used serum only [44], and one study used citrate plasma [59]. Usadel, 2002 [32], analyzed the APC marker on 15 paired samples and reported a higher sensitivity for plasma versus serum, although not significantly different (93% versus 40%, respectively, p = 0.08). Gao, 2015 [39], analyzed all markers on both plasma and serum, and they reported higher sensitivity in the serum cohort (RASSF1A 43.1% versus 52.5% and APC 24.1% versus 42.5% for plasma versus serum, respectively). The main method applied by the studies was quantitative methylation-specific PCR (QMSP, 27/33); three studies used digital PCR [46,50,56] and two studies used sequencing-based assays [55,59]. One study employed a method where the target DNA was increased by PCR and then quantified by Surface Enhanced Raman Spectroscopy [49].

Table 2.

Sample type and analysis method.

Study ID Sample Type Analysis Method Assay Type How Was the Cut-Off Determined?
Usadel, 2002 [32] Plasma; Serum QMSP Single gene Not reported
Ostrow, 2009 [30] Plasma QMSP Single gene Defined by a training cohort and validated in an independent cohort
Zhang, 2010 A [33] Plasma QMSP Single gene Not reported
Zhang, 2010 B [34] Plasma QMSP Single gene Not reported
Begum, 2011 [35] Plasma; Serum QMSP Single gene Defined by a training cohort and validated in an independent cohort
Kneip, 2011 [29] Plasma QMSP Single gene Defined by a training cohort and validated in an independent cohort
Ponomaryova, 2011 [23] Plasma QMSP Single gene Defined by a training cohort
Vinayanuwattikun, 2011 [36] Plasma QMSP Single gene Defined by a training cohort
Balgkouranidou, 2014 A [37] Plasma QMSP Single gene Not reported
Powrozek, 2014 [38] Plasma QMSP Single gene Defined in a previous study
Gao, 2015 [39] Plasma; Serum QMSP Multiplex Defined by a training cohort
Balgkouranidou, 2016 B [22] Plasma QMSP Single gene Not reported
Powrozek, 2016 [40] Plasma QMSP Single gene Defined by a training cohort
Powrozek, 2016 [41] Plasma QMSP Single gene Not reported
Aslam, 2017 [42] Plasma QMSP Single gene Not reported
Hulbert, 2017 [43] Plasma QMSP Single gene Defined by a training cohort
Ooki, 2017 [44] Serum QMSP Single gene Defined by a training cohort and validated in an independent cohort
Nunes, 2019 [45] Plasma QMSP Multiplex Defined by a training cohort and validated in an independent cohort
Villalba, 2019 [46] Plasma Digital PCR Single gene Defined by a training cohort
Yang, 2019 [31] Plasma QMSP Single gene Defined by a training cohort
Chen, 2020 [47] Plasma QMSP Single gene Defined by a training cohort
Huang, 2020 [48] Plasma QMSP Not described Defined by a training cohort and validated in an independent cohort
Li, 2020 [49] Plasma PCR-SERS Single gene Defined by a training cohort
Wen, 2020 [50] Plasma Digital PCR Single gene Defined by a training cohort
Xu, 2020 [51] Plasma QMSP Multiplex Arbitrarily set at 90% specificity for both markers.
Mastoraki, 2021 [52] Plasma QMSP Single gene Defined by a training cohort and validated in an independent cohort
Park, 2021 [53] Plasma QMSP Single gene Defined by a training cohort
Szczyrek, 2021 [54] Plasma QMSP Single gene Defined by a training cohort
Kim, 2022 [55] Plasma Pyrosequencing Single gene Defined by a training cohort and validated in an independent cohort
Palanca-Ballester, 2022 [56] Plasma Digital PCR Single gene Defined by a training cohort
Vo, 2022 [57] Plasma QMSP Single gene Defined by a training cohort
Zeng, 2022 [58] Plasma QMSP Single gene Defined by a training cohort
Zhang, 2022 [59] Plasma or serum Pyrosequencing Single gene Not reported
Open in a new tab

Overview of the sample types and analysis methods employed by the included studies. Studies are arranged according to year of publication, starting with the oldest study and then alphabetically if more studies were published in the same year. Study cohorts used in more than one publication are labeled with A, B, etc. Study ID consists of the first author’s last name and the year of publication, followed by the reference. PCR, polymerase chain reaction; QMSP, quantitative methylation-specific polymerase chain reaction; SERS, surface-enhanced Raman spectroscopy.

3.4. Diagnostic Performance of Methylated Circulating Tumor DNA

The 33 included studies investigated a total of 40 different genes (see the comprehensive list in Supplementary Materials Table S1). The three most frequently used genes were SHOX2, RASSF1A, and APC, which were all reported in seven independent cohorts (Figure 2). The sensitivity ranged from 8% to 93%, while the specificity ranged from 69% to 100%. Since there was substantial heterogeneity between the studies regarding biological material and choice of analysis method, we chose not to include pooled measures for sensitivity and specificity in the forest plots. The complete contingency tables for all studies can be accessed in Supplementary Materials Table S2.

Figure 2.

Figure 2

Open in a new tab

Forest plots of biomarker sensitivity. Forest plots of sensitivity (left-hand panels) and specificity (right-hand panels) for the three most frequently investigated genes. SHOX2 (a,b) was evaluated by Kneip, 2011 [29], Huang, 2020 [48], Vo, 2022 [57], Xu, 2020 [51], and Zeng, 2022 [58]. RASSF1A (c,d) was evaluated by Begum, 2011 [35], Gao, 2015 [39], Li, 2020 [49], Nunes, 2019 [45], and Yang, 2019 [31]. APC (e,f) was evaluated by Usadel, 2002 [32], Begum, 2011 [35], Gao, 2015 [39], and Nunes, 2019 [45]. Dots represent the estimated effect size, and error bars illustrate the 95% confidence intervals.

Pooling all genes from all independent cohorts resulted in 94 unique data points, as visualized in the hierarchical summary receiver operating characteristics (HSROC) curve (Figure 3). The summary sensitivity estimate was 46.9% (95% CI 41.0–52.9%), and the summary specificity estimate was 92.9% (95% CI 90.3–94.8%), suggesting good diagnostic properties of methylated ctDNA in lung cancer detection, especially in specificity. The summary diagnostic odds ratio was 11.5 (95% CI 8.6–15.4). The area under the HSROC curve (AUROC) was 0.81 (95% CI 0.77–0.84).

Figure 3.

Figure 3

Open in a new tab

Hierarchical summary receiver operating characteristics curve. Hierarchical summary receiver operating characteristics (HSROC) curve of the 94 unique data points obtained from the 33 studies included in the meta-analysis. This is a graphical depiction of the random-effects model that includes estimates of the between-study variance. The open circles represent the study estimates, i.e., each target gene investigated in an independent cohort. The red square represents the summary point of all 94 data points, with the 95% confidence region outlined in yellow and the 95% prediction region outlined in gray. The solid green line is the summary HSROC curve.

3.5. Quality Assessment and Risk of Bias

The included studies were generally of acceptable quality; however, all studies lacked information in at least one area of the quality assessment (Figure 4). Only 4/33 studies explicitly stated that a consecutive or random sample of patients was enrolled, and only 3/33 studies avoided a case-control design. Consequently, patient selection posed the biggest potential risk for introducing bias. There was generally low concern that the included patients did not match the review question, with 3/33 studies considered as high-risk. One study had missing data on both cases and controls [29], the second study had a control group consisting entirely of males [23], and the third study included only adenocarcinoma patients [50].

Figure 4.

Figure 4

Open in a new tab

QUADAS-2. Stacked bar chart of the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) consensus judgments. Green diagonal stripes, yellow vertical stripes, and solid red areas represent the percentage of studies judged as low risk, unclear risk, and high risk of bias, respectively.

Only 2/33 studies stated that the index test was performed blinded to the results of the reference standard, while 10/33 studies described the use of a pre-specified threshold or a training-validation design. The majority of the studies (19/33) did not sufficiently describe how the index test was performed or interpreted and was, therefore, evaluated as unclear in regard to the risk of bias. All studies were considered to have index tests within the scope of the review question.

The choice, conduct and interpretation of the reference standard were largely considered in low risk of introducing bias, and we found the target condition defined by the reference standard to match the review question. One study [29] was deemed unclear in terms of reference standards.

The interval between the index test and reference standard was not appropriate in three studies where the sample storage time was too long, while the timing aspect was not sufficiently described in 18/33 studies. The complete quality assessments can be found in the Supplementary Materials (Tables S3–S7).

The majority of the studies (25/33) were funded by public or non-profit organizations; four studies did not report on funding, and four studies were entirely or partly funded by a company or corporate funding source. Likewise, 24/33 studies reported that the authors had no potential conflicts of interest; three studies did not report whether conflicts of interest were present, three studies reported conflicts not pertaining to the funding source, and three studies reported conflicts of interest involving study funding.

There was no significant risk of publication bias in the present systematic review (Deek’s Funnel Plot Asymmetry Test p = 0.30, see Supplementary Materials Figure S1).

4. Discussion

The current review identified 33 separate studies. The same cohorts were used in more studies, leaving 31 unique cohorts addressing the question of whether methylated ctDNA in the blood is useful for identifying lung cancer. In total, 40 separate genes were investigated, and the sensitivity and specificity among the three most frequently analyzed genes (SHOX2, RASSF1A, and APC) varied from 8% to 93% and 69% to 100%, respectively. The summary sensitivity was 46.9%, and the summary specificity was 92.9%.

The accuracy requirements for a test depend heavily on the condition and intended clinical setting. A test with high specificity and low sensitivity may be used to rule in a diagnosis, while the opposite scenario may be used to rule out a diagnosis [60]. The summary sensitivity estimate from the present meta-analysis is very modest, while the summary specificity estimate is moderate to good. However, some of the genes investigated in the individual studies have performance measures that indicate a potential for future clinical application, possibly combined in biomarker panels. Methylated ctDNA in blood seems, at present, best suited as a rule-in tool.

Most often, the studies used a case-control design (28/33 studies). This type of design bears clear advantages; the possibility to conduct a retrospective study with blood samples already at hand and the option of matching the control group to the case group on various parameters. However, there is a higher risk of introducing bias in a retrospective study design, although this is expected to be less of an issue since patient-reported data were not included. This is reflected in the study quality assessments, where the five cohort studies were all judged as having a low risk of bias in the patient selection area, which was true for only four of the case-control studies. Prospective cohort studies are more costly and time-consuming but offer a better level of evidence [61]. We did not identify any randomized clinical trials on this subject.

The large majority of investigations were performed using plasma. The optimal biologic sample type for ctDNA detection has been a recurrent topic of discussion, but there is generally a consensus that plasma is preferable to serum [62]. This is partly due to a higher yield of ctDNA in plasma and partly due to a higher level of contamination with genomic DNA in serum [62,63]. Two studies included in the present review did a direct comparison between plasma and serum and found diverging results [32,39]. Usadel, 2002 [32], reported a non-significant difference in favor of plasma. Gao, 2015 [39], found serum to be superior in terms of sensitivity; however, they did not perform any statistical testing. This observed difference might simply be by chance since not all subjects were matched, and results from the 26 matched patients generated slightly different sensitivity and specificity.

To meet the inclusion criteria, the reference test for diagnosing lung cancer should involve tumor cytology or histopathology. However, some of the studies included in the analysis did not consistently specify whether the diagnosis was obtained through cytological or histological biopsy or through surgical resection. Despite the lack of detailed information on the diagnostic method in all studies, the histological type of lung cancer was described, indicating that the diagnosis was confirmed rather than solely based on clinical suspicion. Reporting both histology and stage of the lung cancer cases is crucial, and future research should consider reporting diagnostic sensitivity and specificity for these subgroups provided sufficient participant numbers.

One of the criteria for inclusion in the current study was the quantitative measurement of methylated DNA. In total, 27/33 studies performed QMSP. Probably this is because of the relatively low costs of equipment and assays and the ease of performance [64]. Both QMSP and digital PCR with MethyLight has demonstrated a strong correlation between the expected and observed methylation values, while NGS tended to overestimate the methylation level [18]. However, digital PCR is a much more sensitive analysis, as demonstrated by Yu and colleagues, who reported a 20-fold lower limit of detection with droplet digital PCR compared to conventional quantitative PCR [65]. A large, interlaboratory study concluded that droplet digital PCR could perform highly reproducible absolute quantification of a specific DNA target with a reported inter-laboratory difference of <12% [66]. Digital PCR is a more recently developed technology, which is reflected in the present review, where the oldest study performing digital PCR was from 2019 [46].

Pyrosequencing was employed by Zhang, 2022 [59] and Kim, 2022 [55]. The sequencing-based methods for quantifying methylated tumor DNA can be more time-consuming and costly compared to PCR-based methods. However, these methods can be used for genome-wide coverage with comparable sensitivity to digital PCR [67], while the PCR-based methods are limited to a single gene or smaller multiplex gene panels. The eligibility criteria specified that only studies reporting outcome data for single genes would be included in the review. This was the main reason why only two of the included studies performed sequencing since sequencing outcomes are often reported as large algorithms or multiplex panels. A recent review of the data on methylated ctDNA in ovarian cancer concluded that panels of methylation markers performed better than single genes [68]. The results from the present review may guide the choice of biomarkers to include in future multiplex panels for diagnosing lung cancer.

The studies originated from different parts of the world. The majority of the studies (15/33) originated from China or other Asian countries, while twelve originated from EU countries and five were from North America. These geographical differences may impact the applicability of the results since studies have shown that a significant proportion of lung cancer cases among East Asian populations occur in individuals who are non-smokers, in contrast to Caucasians [69], and that DNA methylation is highly affected by smoking [70]. The studies included in this meta-analysis have not consistently reported race and a direct comparison of the diagnostic accuracy of the respective genes in different races is not possible.

We applied wide criteria in the search strategy, searched the major databases and identified a very large number of potentially eligible studies. However, we can never be entirely certain that we have identified all relevant studies. Some studies may only have used the gene name and not any of the broader ctDNA terms in the title and abstract and thus would not be included in our search results. We did not search the gray literature or unpublished results.

This review and meta-analysis combine data from multiple studies, allowing for a larger sample size and increased statistical power. This can enhance the precision and reliability of the findings. By pooling data from various studies, the meta-analysis provides a comprehensive overview of the available evidence on the diagnostic accuracy of methylated DNA in lung cancer. It helps identify consistent patterns or trends across different studies. The meta-analysis incorporates data from multiple populations and settings, possibly providing a more generalizable estimate of the diagnostic accuracy. This enhances the applicability of the findings to different patient populations. However, caution is advised when interpreting the summary estimates, as the list of investigated genes is extensive and diverse.

The quality of the individual studies included in the meta-analysis varied. Limitations or biases in the design, conduct, or reporting of the primary studies can potentially affect the validity and reliability of the meta-analysis results. Given the high number of case-control studies with healthy individuals in the control group, the results are subject to spectrum bias. A more accurate effect estimate may be obtained in cohort studies or case-control studies, including patients with benign diseases. In addition, there is a risk of publication bias, as studies with positive or significant results are more likely to be published, while studies with negative or non-significant findings may remain unpublished. This can introduce a bias in the overall estimate of diagnostic accuracy. However, as described previously, no significant risk of publication bias was found using Deek’s funnel plot asymmetry test. Variations in study characteristics such as patient and control group populations, assay methods, cutoff values, and study designs can introduce heterogeneity across studies. This heterogeneity can affect the pooling of results and warrants caution in the generalizability of the findings.

5. Conclusions

In conclusion, methylated ctDNA from blood samples can be detected by various methods, but the summary sensitivity estimate implies that further improvements are needed before this type of biomarker is ready for testing in a randomized clinical trial setting, much less for clinical implementation. The detailed results presented here may aid the selection of genes for biomarker panels. Future studies should consider the following points: (I) Adopt a cohort design to reduce the risk of bias; (II) Follow best practice guidelines for the preanalytical work-flow as suggested by Meddeb and colleagues [62] to reduce variability; (III) Rigorously follow the Standards for Reporting Diagnostic accuracy studies (STARD) guidelines [71] since all studies had at least one unclear item in the QUADAS-2 evaluation; (IV) Emphasize the detection of early-stage lung cancer, as this is where the survival benefit is most significant, and provide specific results for lung cancer stage and histology whenever possible.

Acknowledgments

The authors would like to thank research librarian Mette Brandt Eriksen for assistance with databases and search strategy.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/cancers15153959/s1, Table S1: Comprehensive list of genes; Table S2: Contingency tables for all included studies; Table S3: Quality assessment QUADAS-2—Patient selection; Table S4: Quality assessment QUADAS-2—Index test; Table S5: Quality assessment QUADAS-2—Reference standard; Table S6: Quality assessment QUADAS-2—Flow and timing; Table S7: Quality assessment—Funding and conflicts of interest; Figure S1: Funnel plot.

Click here for additional data file. (275KB, zip)

Author Contributions

Conceptualization, M.B., S.W.C.W., T.F.H. and O.H.; methodology, M.B., S.W.C.W., R.F.A. and S.T.; formal analysis, M.B., S.W.C.W. and S.T.; investigation, M.B. and S.W.C.W.; data curation, M.B. and S.W.C.W.; writing—original draft preparation, M.B. and S.W.C.W.; writing—review and editing, R.F.A., S.T., T.F.H. and O.H.; supervision, R.F.A. and O.H.; project administration, S.W.C.W.; funding acquisition, O.H. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available in the article and the supplementary materials.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Funding Statement

This research was funded by the Danish Cancer Society via the Danish Comprehensive Cancer Centers: “Danish Research Center for Lung Cancer” and “ctDNA Research Center—The Danish Research Center for Circulating Tumor DNA Guided Cancer Management, Danish Cancer Society (grant no. R257-A14700)”.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

References

  • 1.Sung H., Ferlay J., Siegel R.L., Laversanne M., Soerjomataram I., Jemal A., Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J. Clin. 2021;71:209–249. doi: 10.3322/caac.21660. [DOI] [PubMed] [Google Scholar]
  • 2.Woodard G.A., Jones K.D., Jablons D.M. Lung Cancer Staging and Prognosis. In: Reckamp K.L., editor. Lung Cancer. Volume 170. Springer International Publishing; Cham, Switzerland: 2016. pp. 47–75. Cancer Treatment and Research. [DOI] [PubMed] [Google Scholar]
  • 3.O’Dowd E.L., Baldwin D.R. Early Diagnosis Pivotal to Survival in Lung Cancer. Practitioner. 2014;258:2–3, 21–24. [PubMed] [Google Scholar]
  • 4.Allemani C., Matsuda T., Di Carlo V., Harewood R., Matz M., Nikšić M., Bonaventure A., Valkov M., Johnson C.J., Estève J., et al. Global Surveillance of Trends in Cancer Survival 2000–14 (CONCORD-3): Analysis of Individual Records for 37 513 025 Patients Diagnosed with One of 18 Cancers from 322 Population-Based Registries in 71 Countries. Lancet. 2018;391:1023–1075. doi: 10.1016/S0140-6736(17)33326-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Verma N., Wu M., Altmayer S. Lung Cancer Screening: How We Do It and Why. Semin. Roentgenol. 2020;55:14–22. doi: 10.1053/j.ro.2019.10.004. [DOI] [PubMed] [Google Scholar]
  • 6.Potter A.L., Rosenstein A.L., Kiang M.V., Shah S.A., Gaissert H.A., Chang D.C., Fintelmann F.J., Yang C.-F.J. Association of Computed Tomography Screening with Lung Cancer Stage Shift and Survival in the United States: Quasi-Experimental Study. BMJ. 2022;376:e069008. doi: 10.1136/bmj-2021-069008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Singareddy A., Flanagan M.E., Samson P.P., Waqar S.N., Devarakonda S., Ward J.P., Herzog B.H., Rohatgi A., Robinson C.G., Gao F., et al. Trends in Stage I Lung Cancer. Clin. Lung Cancer. 2023;24:114–119. doi: 10.1016/j.cllc.2022.11.005. [DOI] [PubMed] [Google Scholar]
  • 8.Van Meerbeeck J.P., Franck C. Lung Cancer Screening in Europe: Where Are We in 2021? Transl. Lung Cancer Res. 2021;10:2407–2417. doi: 10.21037/tlcr-20-890. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Mazzone P.J., Silvestri G.A., Souter L.H., Caverly T.J., Kanne J.P., Katki H.A., Wiener R.S., Detterbeck F.C. Screening for Lung Cancer. Chest. 2021;160:e427–e494. doi: 10.1016/j.chest.2021.06.063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Rosenkrantz A.B., Xue X., Gyftopoulos S., Kim D.C., Nicola G.N. Downstream Costs Associated with Incidental Pulmonary Nodules Detected on CT. Acad. Radiol. 2019;26:798–802. doi: 10.1016/j.acra.2018.07.005. [DOI] [PubMed] [Google Scholar]
  • 11.Shi H.-M., Sun Z.-C., Ju F.-H. Understanding the Harm of Low-dose Computed Tomography Radiation to the Body (Review) Exp. Ther. Med. 2022;24:534. doi: 10.3892/etm.2022.11461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Lin Y., Fu M., Ding R., Inoue K., Jeon C.Y., Hsu W., Aberle D.R., Prosper A.E. Patient Adherence to Lung CT Screening Reporting & Data System–Recommended Screening Intervals in the United States: A Systematic Review and Meta-Analysis. J. Thorac. Oncol. 2022;17:38–55. doi: 10.1016/j.jtho.2021.09.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.MacMahon H., Naidich D.P., Goo J.M., Lee K.S., Leung A.N.C., Mayo J.R., Mehta A.C., Ohno Y., Powell C.A., Prokop M., et al. Guidelines for Management of Incidental Pulmonary Nodules Detected on CT Images: From the Fleischner Society 2017. Radiology. 2017;284:228–243. doi: 10.1148/radiol.2017161659. [DOI] [PubMed] [Google Scholar]
  • 14.Li L., Zhao Y., Li H. Assessment of Anxiety and Depression in Patients with Incidental Pulmonary Nodules and Analysis of Its Related Impact Factors. Thorac. Cancer. 2020;11:1433–1442. doi: 10.1111/1759-7714.13406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Casagrande G.M.S., Silva M.D.O., Reis R.M., Leal L.F. Liquid Biopsy for Lung Cancer: Up-to-Date and Perspectives for Screening Programs. Int. J. Mol. Sci. 2023;24:2505. doi: 10.3390/ijms24032505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Wen X., Pu H., Liu Q., Guo Z., Luo D. Circulating Tumor DNA—A Novel Biomarker of Tumor Progression and Its Favorable Detection Techniques. Cancers. 2022;14:6025. doi: 10.3390/cancers14246025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Dacic S. State of the Art of Pathologic and Molecular Testing. Hematol. Oncol. Clin. 2023;37:463–473. doi: 10.1016/j.hoc.2023.02.001. [DOI] [PubMed] [Google Scholar]
  • 18.Redshaw N., Huggett J.F., Taylor M.S., Foy C.A., Devonshire A.S. Quantification of Epigenetic Biomarkers: An Evaluation of Established and Emerging Methods for DNA Methylation Analysis. BMC Genom. 2014;15:1174. doi: 10.1186/1471-2164-15-1174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.McGinn S., Gut I.G. DNA Sequencing—Spanning the Generations. New Biotechnol. 2013;30:366–372. doi: 10.1016/j.nbt.2012.11.012. [DOI] [PubMed] [Google Scholar]
  • 20.Shields M.D., Chen K., Dutcher G., Patel I., Pellini B. Making the Rounds: Exploring the Role of Circulating Tumor DNA (CtDNA) in Non-Small Cell Lung Cancer. Int. J. Mol. Sci. 2022;23:9006. doi: 10.3390/ijms23169006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Walter R.F.H., Rozynek P., Casjens S., Werner R., Mairinger F.D., Speel E.J.M., Zur Hausen A., Meier S., Wohlschlaeger J., Theegarten D., et al. Methylation of L1RE1, RARB, and RASSF1 Function as Possible Biomarkers for the Differential Diagnosis of Lung Cancer. PLoS ONE. 2018;13:e0195716. doi: 10.1371/journal.pone.0195716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Balgkouranidou I., Chimonidou M., Milaki G., Tsaroucha E., Kakolyris S., Georgoulias V., Lianidou E. SOX17 Promoter Methylation in Plasma Circulating Tumor DNA of Patients with Non-Small Cell Lung Cancer. Clin. Chem. Lab. Med. 2016;54:1385–1393. doi: 10.1515/cclm-2015-0776. [DOI] [PubMed] [Google Scholar]
  • 23.Ponomaryova A.A., Rykova E.Y., Cherdyntseva N.V., Skvortsova T.E., Dobrodeev A.Y., Zav’yalov A.A., Tuzikov S.A., Vlassov V.V., Laktionov P.P. RARβ2 Gene Methylation Level in the Circulating DNA from Blood of Patients with Lung Cancer. Eur. J. Cancer Prev. 2011;20:453–455. doi: 10.1097/CEJ.0b013e3283498eb4. [DOI] [PubMed] [Google Scholar]
  • 24.Li L., Fu K., Zhou W., Snyder M. Applying Circulating Tumor DNA Methylation in the Diagnosis of Lung Cancer. Precis. Clin. Med. 2019;2:45–56. doi: 10.1093/pcmedi/pbz003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Page M.J., McKenzie J.E., Bossuyt P.M., Boutron I., Hoffmann T.C., Mulrow C.D., Shamseer L., Tetzlaff J.M., Akl E.A., Brennan S.E., et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Whiting P.F., Rutjes A.W.S., Westwood M.E., Mallett S., Deeks J.J., Reitsma J.B., Leeflang M.M.G., Sterne J.A.C., Bossuyt P.M.M., QUADAS-2 Group QUADAS-2: A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies. Ann. Intern. Med. 2011;155:529–536. doi: 10.7326/0003-4819-155-8-201110180-00009. [DOI] [PubMed] [Google Scholar]
  • 27.Harbord R.M., Whiting P. Metandi: Meta-Analysis of Diagnostic Accuracy Using Hierarchical Logistic Regression. Stata J. Promot. Commun. Stat. Stata. 2009;9:211–229. doi: 10.1177/1536867X0900900203. [DOI] [Google Scholar]
  • 28.Dwamena B. MIDAS: Stata Module for Meta-Analytical Integration of Diagnostic Test Accuracy Studies. EconPapers. 2007 [Google Scholar]
  • 29.Kneip C., Schmidt B., Seegebarth A., Weickmann S., Fleischhacker M., Liebenberg V., Field J.K., Dietrich D. SHOX2 DNA Methylation Is a Biomarker for the Diagnosis of Lung Cancer in Plasma. J. Thorac. Oncol. 2011;6:1632–1638. doi: 10.1097/JTO.0b013e318220ef9a. [DOI] [PubMed] [Google Scholar]
  • 30.Ostrow K., Loyo M., Greenberg A., Gaither-Davis A., Siegfried J., Hoque M., Bigbee W., Rom W., Sidransky D. Molecular Analysis of Plasma DNA for the Early Detection of Lung Cancer by Quantitative Methylation Specific PCR. Cancer Res. 2009;69:247. doi: 10.1158/1078-0432.CCR-09-3304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Yang Z., Qi W., Sun L., Zhou H., Zhou B., Hu Y. DNA Methylation Analysis of Selected Genes for the Detection of Early-Stage Lung Cancer Using Circulating Cell-Free DNA. Adv. Clin. Exp. Med. 2019;28:361–366. doi: 10.17219/acem/84935. [DOI] [PubMed] [Google Scholar]
  • 32.Usadel H., Brabender J., Danenberg K.D., Jeronimo C., Harden S., Engles J., Danenberg P.V., Yang S., Sidransky D. Quantitative Adenomatous Polyposis Coli Promoter Methylation Analysis in Tumor Tissue, Serum, and Plasma DNA of Patients with Lung Cancer. Cancer Res. 2002;62:371–375. [PubMed] [Google Scholar]
  • 33.Zhang Y.W., Miao Y.F., Yi J., Geng J., Wang R., Chen L.B. Transcriptional Inactivation of Secreted Frizzled-Related Protein 1 by Promoter Hypermethylation as a Potential Biomarker for Non-Small Cell Lung Cancer. Neoplasma. 2010;57:228–233. doi: 10.4149/neo_2010_03_228. [DOI] [PubMed] [Google Scholar]
  • 34.Zhang Y., Song H., Miao Y., Wang R., Chen L. Frequent Transcriptional Inactivation of Kallikrein 10 Gene by CpG Island Hypermethylation in Non-Small Cell Lung Cancer. Cancer Sci. 2010;101:934–940. doi: 10.1111/j.1349-7006.2009.01486.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Begum S., Brait M., Dasgupta S., Ostrow K.L., Zahurak M., Carvalho A.L., Califano J.A., Goodman S.N., Westra W.H., Hoque M.O., et al. An Epigenetic Marker Panel for Detection of Lung Cancer Using Cell-Free Serum DNA. Clin. Cancer Res. 2011;17:4494–4503. doi: 10.1158/1078-0432.CCR-10-3436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Vinayanuwattikun C., Sriuranpong V., Tanasanvimon S., Chantranuwat P., Mutirangura A. Epithelial-Specific Methylation Marker: A Potential Plasma Biomarker in Advanced Non-Small Cell Lung Cancer. J. Thorac. Oncol. 2011;6:1818–1825. doi: 10.1097/JTO.0b013e318226b46f. [DOI] [PubMed] [Google Scholar]
  • 37.Balgkouranidou I., Chimonidou M., Milaki G., Tsarouxa E.G., Kakolyris S., Welch D.R., Georgoulias V., Lianidou E.S. Breast Cancer Metastasis Suppressor-1 Promoter Methylation in Cell-Free DNA Provides Prognostic Information in Non-Small Cell Lung Cancer. Br. J. Cancer. 2014;110:2054–2062. doi: 10.1038/bjc.2014.104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Powrozek T., Krawczyk P., Kucharczyk T., Milanowski J. Septin 9 Promoter Region Methylation in Free Circulating DNA-Potential Role in Noninvasive Diagnosis of Lung Cancer: Preliminary Report. Med. Oncol. 2014;31:917. doi: 10.1007/s12032-014-0917-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Gao L., Xie E., Yu T., Chen D., Zhang L., Zhang B., Wang F., Xu J., Huang P., Liu X., et al. Methylated APC and RASSF1A in Multiple Specimens Contribute to the Differential Diagnosis of Patients with Undetermined Solitary Pulmonary Nodules. J. Thorac. Dis. 2015;7:422–432. doi: 10.3978/j.issn.2072-1439.2015.01.24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Powrozek T., Krawczyk P., Nicos M., Kuznar-Kaminska B., Batura-Gabryel H., Milanowski J. Methylation of the DCLK1 Promoter Region in Circulating Free DNA and Its Prognostic Value in Lung Cancer Patients. Clin. Transl. Oncol. 2016;18:398–404. doi: 10.1007/s12094-015-1382-z. [DOI] [PubMed] [Google Scholar]
  • 41.Powrozek T., Krawczyk P., Kuznar-Kaminska B., Batura-Gabryel H., Milanowski J. Analysis of RTEL1 and PCDHGB6 Promoter Methylation in Circulating-Free DNA of Lung Cancer Patients Using Liquid Biopsy: A Pilot Study. Exp. Lung Res. 2016;42:307–313. doi: 10.1080/01902148.2016.1214191. [DOI] [PubMed] [Google Scholar]
  • 42.Aslam M.S., Shaeer A., Abbas Z., Ahmed A., Gull I., Athar M.A. Cell-Free DNA Quantification and Methylation Status of DCC Gene as Predictive Diagnostic Biomarkers of Lung Cancer in Patients Reported at Gulab Devi Chest Hospital, Lahore. Technol. Cancer Res. Treat. 2017;16:758–765. doi: 10.1177/1533034616682155. [DOI] [Google Scholar]
  • 43.Hulbert A., Jusue-Torres I., Stark A., Chen C., Rodgers K., Lee B., Griffin C., Yang A., Huang P., Wrangle J., et al. Early Detection of Lung Cancer Using DNA Promoter Hypermethylation in Plasma and Sputum. Clin. Cancer Res. 2017;23:1998–2005. doi: 10.1158/1078-0432.CCR-16-1371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Ooki A., Maleki Z., Tsay J.-C.J., Goparaju C., Brait M., Turaga N., Nam H.-S., Rom W.N., Pass H.I., Sidransky D., et al. A Panel of Novel Detection and Prognostic Methylated DNA Markers in Primary Non-Small Cell Lung Cancer and Serum DNA. Clin. Cancer Res. 2017;23:7141–7152. doi: 10.1158/1078-0432.CCR-17-1222. [DOI] [PubMed] [Google Scholar]
  • 45.Nunes S.P., Diniz F., Moreira-Barbosa C., Constancio V., Silva A.V., Oliveira J., Soares M., Paulino S., Cunha A.L., Rodrigues J., et al. Subtyping Lung Cancer Using Dna Methylation in Liquid Biopsies. J. Clin. Med. 2019;8:1500. doi: 10.3390/jcm8091500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Villalba M., Exposito F., Pajares M.J., Sainz C., Redrado M., Remirez A., Wistuba I., Behrens C., Jantus-Lewintre E., Camps C., et al. TMPRSS4: A Novel Tumor Prognostic Indicator for the Stratification of Stage IA Tumors and a Liquid Biopsy Biomarker for NSCLC Patients. J. Clin. Med. 2019;8:2134. doi: 10.3390/jcm8122134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Chen C., Huang X., Yin W., Peng M., Wu F., Wu X., Tang J., Chen M., Wang X., Hulbert A., et al. Ultrasensitive DNA Hypermethylation Detection Using Plasma for Early Detection of NSCLC: A Study in Chinese Patients with Very Small Nodules. Clin. Epigenetics. 2020;12:828. doi: 10.1186/s13148-020-00828-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Huang W., Huang H., Zhang S., Wang X., Ouyang J., Lin Z., Chen P. A Novel Diagnosis Method Based on Methylation Analysis of SHOX2 and Serum Biomarker for Early Stage Lung Cancer. Cancer Control. 2020;27:1073274820969703. doi: 10.1177/1073274820969703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Li X., Yang T., Li C.S., Song Y., Wang D., Jin L., Lou H., Li W. Polymerase Chain Reaction—Surface-Enhanced Raman Spectroscopy (PCR-SERS) Method for Gene Methylation Level Detection in Plasma. Theranostics. 2020;10:898–909. doi: 10.7150/thno.30204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Wen S.W.C., Andersen R.F., Petersen L.M.S., Hager H., Hilberg O., Jakobsen A., Hansen T.F. Comparison of Mutated Kras and Methylated Hoxa9 Tumor-Specific Dna in Advanced Lung Adenocarcinoma. Cancers. 2020;12:3728. doi: 10.3390/cancers12123728. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Xu Z., Wang Y., Wang L., Xiong J., Wang H., Cui F., Peng H. The Performance of the SHOX2/PTGER4 Methylation Assay Is Influenced by Cancer Stage, Age, Type and Differentiation. Biomark. Med. 2020;14:341–351. doi: 10.2217/bmm-2019-0325. [DOI] [PubMed] [Google Scholar]
  • 52.Mastoraki S., Balgkouranidou I., Tsaroucha E., Klinakis A., Georgoulias V., Lianidou E. KMT2C Promoter Methylation in Plasma-Circulating Tumor DNA Is a Prognostic Biomarker in Non-Small Cell Lung Cancer. Mol. Oncol. 2021;15:2412–2422. doi: 10.1002/1878-0261.12848. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Park M., Lee J., Lee J., Hwang S. Alu Cell-Free DNA Concentration, Alu Index, and LINE-1 Hypomethylation as a Cancer Predictor. Clin. Biochem. 2021;94:67–73. doi: 10.1016/j.clinbiochem.2021.04.021. [DOI] [PubMed] [Google Scholar]
  • 54.Szczyrek M., Grenda A., Kuznar-Kaminska B., Krawczyk P., Sawicki M., Batura-Gabryel H., Mlak R., Szudy-Szczyrek A., Krajka T., Krajka A., et al. Methylation of DROSHA and DICER as a Biomarker for the Detection of Lung Cancer. Cancers. 2021;13:6139. doi: 10.3390/cancers13236139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Kim Y., Lee B.B., Kim D., Um S.-W., Han J., Shim Y.M., Kim D.-H. Aberrant Methylation of SLIT2 Gene in Plasma Cell-Free DNA of Non-Small Cell Lung Cancer Patients. Cancers. 2022;14:296. doi: 10.3390/cancers14020296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Palanca-Ballester C., Hervas D., Villalba M., Valdes-Sanchez T., Garcia D., Alcoriza-Balaguer M.I., Benet M., Martinez-Tomas R., Briones-Gomez A., Galbis-Caravajal J., et al. Translation of a Tissue Epigenetic Signature to Circulating Free DNA Suggests BCAT1 as a Potential Noninvasive Diagnostic Biomarker for Lung Cancer. Clin. Epigenetics. 2022;14:116. doi: 10.1186/s13148-022-01334-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Vo T.T.L., Nguyen T.N., Nguyen T.T., Pham A.T.D., Vuong D.L., Ta V.T., Ho V.S. SHOX2 Methylation in Vietnamese Patients with Lung Cancer. Mol. Biol. Rep. 2022;49:3413–3421. doi: 10.1007/s11033-022-07172-z. [DOI] [PubMed] [Google Scholar]
  • 58.Zeng S., Lin C., Huang Y. MiR-375 Combined with SHOX2 Methylation Has Higher Diagnostic Efficacy for Non-Small-Cell Lung Cancer. Mol. Biotechnol. 2022;65:1187–1197. doi: 10.1007/s12033-022-00604-y. [DOI] [PubMed] [Google Scholar]
  • 59.Zhang L.Y., Sun X.W., Ding Y.J., Yan Y.R., Wang Y., Li C.X., Li S.Q., Zhang L., Song H.J., Li H.P., et al. SERPINA1 Methylation Levels Are Associated with Lung Cancer Development in Male Patients with Chronic Obstructive Pulmonary Disease. Int. J. Chron. Obstruct. Pulmon. Dis. 2022;17:2117–2125. doi: 10.2147/COPD.S368543. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Power M., Fell G., Wright M. Principles for High-Quality, High-Value Testing. Evid. Based Med. 2013;18:5–10. doi: 10.1136/eb-2012-100645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Mokhles S., Takkenberg J.J., Treasure T. Evidence-Based and Personalized Medicine. It’s [AND] Not [OR] Ann. Thorac. Surg. 2017;103:351–360. doi: 10.1016/j.athoracsur.2016.08.100. [DOI] [PubMed] [Google Scholar]
  • 62.Meddeb R., Pisareva E., Thierry A.R. Guidelines for the Preanalytical Conditions for Analyzing Circulating Cell-Free DNA. Clin. Chem. 2019;65:623–633. doi: 10.1373/clinchem.2018.298323. [DOI] [PubMed] [Google Scholar]
  • 63.El Messaoudi S., Rolet F., Mouliere F., Thierry A.R. Circulating Cell Free DNA: Preanalytical Considerations. Clin. Chim. Acta. 2013;424:222–230. doi: 10.1016/j.cca.2013.05.022. [DOI] [PubMed] [Google Scholar]
  • 64.Taylor S.C., Nadeau K., Abbasi M., Lachance C., Nguyen M., Fenrich J. The Ultimate QPCR Experiment: Producing Publication Quality, Reproducible Data the First Time. Trends Biotechnol. 2019;37:761–774. doi: 10.1016/j.tibtech.2018.12.002. [DOI] [PubMed] [Google Scholar]
  • 65.Yu M., Carter K.T., Makar K.W., Vickers K., Ulrich C.M., Schoen R.E., Brenner D., Markowitz S.D., Grady W.M. MethyLight Droplet Digital PCR for Detection and Absolute Quantification of Infrequently Methylated Alleles. Epigenetics. 2015;10:803–809. doi: 10.1080/15592294.2015.1068490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Whale A.S., Devonshire A.S., Karlin-Neumann G., Regan J., Javier L., Cowen S., Fernandez-Gonzalez A., Jones G.M., Redshaw N., Beck J., et al. International Interlaboratory Digital PCR Study Demonstrating High Reproducibility for the Measurement of a Rare Sequence Variant. Anal. Chem. 2017;89:1724–1733. doi: 10.1021/acs.analchem.6b03980. [DOI] [PubMed] [Google Scholar]
  • 67.Postel M., Roosen A., Laurent-Puig P., Taly V., Wang-Renault S.-F. Droplet-Based Digital PCR and next Generation Sequencing for Monitoring Circulating Tumor DNA: A Cancer Diagnostic Perspective. Expert Rev. Mol. Diagn. 2018;18:7–17. doi: 10.1080/14737159.2018.1400384. [DOI] [PubMed] [Google Scholar]
  • 68.Terp S.K., Stoico M.P., Dybkær K., Pedersen I.S. Early Diagnosis of Ovarian Cancer Based on Methylation Profiles in Peripheral Blood Cell-Free DNA: A Systematic Review. Clin. Epigenet. 2023;15:24. doi: 10.1186/s13148-023-01440-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Park J.Y., Jang S.H. Epidemiology of Lung Cancer in Korea: Recent Trends. Tuberc. Respir. Dis. 2016;79:58–69. doi: 10.4046/trd.2016.79.2.58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Leenen F.A.D., Muller C.P., Turner J.D. DNA Methylation: Conducting the Orchestra from Exposure to Phenotype? Clin. Epigenetics. 2016;8:92. doi: 10.1186/s13148-016-0256-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Cohen J.F., Korevaar D.A., Altman D.G., Bruns D.E., Gatsonis C.A., Hooft L., Irwig L., Levine D., Reitsma J.B., Vet H.C.W.d., et al. STARD 2015 Guidelines for Reporting Diagnostic Accuracy Studies: Explanation and Elaboration. BMJ Open. 2016;6:e012799. doi: 10.1136/bmjopen-2016-012799. [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

Click here for additional data file. (275KB, zip)

Data Availability Statement

The data presented in this study are available in the article and the supplementary materials.

Articles from Cancers are provided here courtesy of Multidisciplinary Digital Publishing Institute (MDPI)

RESOURCES