Skip to main content
NCBI home page
Future Science OA logoLink to Future Science OA
. 2025 Nov 18;11(1):2590381. doi: 10.1080/20565623.2025.2590381

The diagnostic value of the liquid biopsy in the colon cancer

Mona Mlika a,b,✉, Mohamed Majdi Zorgati c, Imen Ben Ismail b,d, Sarra Cheikhrouhou b, Chadli Dziri b
PMCID: PMC12629340  PMID: 41251116

Abstract

Introduction

The utility of liquid biopsy in the screening and the follow-up of many cancers and mainly CRC has been reported by many authors. Our aim was to assess its diagnostic yield as a surrogate of tissue biopsy in CRC.

Methods

The authors performed a meta-analysis according to the PRISMA recommendations. Publication bias was evaluated using the Funnel plot followed by the Egger regression test. Meta-disc 2.0 was used. In order to assess the heterogeneity, the Cochrane Chi 2 test (Q-test), Tau2 and 95% prediction interval (PI) were calculated (CMA version 4).

Results

Our results revealed a high diagnostic potential of the liquid biopsy in CRC with a pooled SEN, SPE, PLR, NLR and DOR accounting respectively for 77%, 94.3%, 6.85, 0.26 and 26.467. The AUC accounted for 0.97. Based on the prediction interval analysis, we observed an important heterogeneity. The meta-regression analysis revealed the sample nature (plasma), the biomarker used (cfDNA) and the intake of a preoperative treatment as potential moderators.

Conclusion

Even if our results highlighted a high diagnostic potential of liquid biopsy in the diagnosis of CRC, the heterogeneity observed reinforces the need to harmonize the use of particular samples, the biomarker assessed and to make attention to the treatment intake before surgery.

Keywords: Cancer, diagnostics, biostatistics, biomarkers, chemotherapy, personalized medicine, proteomics

PLAN LANGUAGE SUMMARY

Many researchers have looked into using liquid biopsy as a way to detect and monitor different types of cancer, especially colon cancer. In this study, we wanted to see how well liquid biopsy could work compared to the more traditional method, which involves taking a sample of the tumor directly.

To do this, we gathered and analyzed results from a number of previous studies. We made sure the studies were good quality and looked for any signs that the results might be biased. We then used different tools to figure out how accurate liquid biopsy is and whether the studies showed consistent results or not. We also checked if certain factors, like the type of sample or whether the patient had treatment before surgery, made a difference in the results. We found that liquid biopsy seems to work well for detecting colon cancer. However, there was quite a bit of variation between the different studies. This might be because they used different types of samples, looked at different markers in the blood, or included people who had already started treatment.

Even though liquid biopsy looks promising for diagnosing colon cancer, the differences between studies suggest we need to be more consistent about how and when the test is used—like making sure the same kind of sample is taken, the same blood markers are checked, and knowing whether someone has had treatment before surgery.

ARTICLE HIGHLIGHTS

  • Liquid biopsy has been assessed in the screening and early diagnosis of colon cancer but was insufficiently studied as a surrogate to tissue biopsy

  • Using predictive intervals in addition to the I2 is of major importance to assess the heterogeneity when dealing with the diagnostic value of liquid biopsy

  • Our results revealed a high diagnostic potential of the liquid biopsy in CRC with a pooled SEN, SPE, PLR, NLR and DOR accounting respectively for 77%, 94.3%, 6.847, 0.259 and 26.467. The AUC accounted for 0.97.

  • The meta-regression analysis including as moderators the sample used (plasma sample or not), the technique used (NGS or not), the number of the biomarkers assessed (single biomarker or not), the biomarkers assessed (cfDNA or not), the tumor stage (studies including early and late stages or not), the race (European race or not), the fact of including healthy patients and the presence or absence of a pre-operative treatment, highlighted the sample nature (plasma), the biomarker nature (cfDNA) and the intake of a pre-operative treatment as potential moderators

  • Even if our results seem promising, they have to be taken with caution because of the small dataset and the quality of the studies.

  • The heterogeneity observed in this study reinforces the need to harmonize the use of particular samples, the biomarker assessed and to make attention to the treatment intake before surgery.

1. Introduction

Colorectal cancer (CRC) is a major public health problem all over the world. It is the third common cancer and the most frequent gastro-intestinal cancer [1]. Its positive diagnosis is based on the microscopic exam of tissue samples with the adenocarcinomas being the most frequent histologic subtype. The screening of CRC is mainly based on stool-based and endoscopic tests [2,3]. Molecular approach has been mainly reported in the recent decade [4]. Many manuscripts reported the utility of liquid biopsy in the screening, the early diagnosis and the follow-up of CRC [2,4–6]. When dealing with liquid biopsy, we deal with biomarkers. The major biomarkers reported in the literature are mainly represented by the auto-antibodies, the circulating tumor cells, the exosomes, the micro-RNA, the proteomics and the metabolomics [7]. The only application of liquid biopsy that is commonly used in clinical practice is the identification of predictive biomarkers in metastatic patients [5]. No consensus has been reached concerning its diagnostic use. Besides, many techniques are used to detect the various mutations including next-generation sequencing (NGC), digital PCR (dPCR), quantitative PCR (qPCR).

In this manuscript, our aim was to assess the diagnostic value of the liquid biopsy as a potential surrogate to tissue biopsy in the diagnosis of CRC based on its SEN, SPE, predictive intervals and heterogeneity.

2. Material and methods

2.1. Data sources and bibliographic research

This meta-analysis was performed according to the PRISMA recommendations [8]. The authors adhered to PRISMA checklist. The bibliographic research was performed on Pubmed, Embase and Cochrane Databases during a 10-year-period (from 2015 to 2025) with limitations to English language. The key-words used for the research of medical titles (MeSH) for the bibliographic research consisted of: liquid biopsy OR exosomes OR circulating tumor cells OR circulating DNA OR micro-RNA AND diagnosis OR SEN OR SPE OR ROC curve AND CRC.

According to the inclusion criteria, the titles and the summaries, all the publications were reviewed and assessed.

2.2. Selection criteria of the different studies

In order to be included in this meta-analysis, the following criteria had to be included in the manuscripts: The manuscripts included were original manuscripts consisting of cohort studies or case-control studies, the manuscripts assessed the diagnostic value of the different biomarkers in the liquid biopsy, the manuscripts contained sufficient data including true negatives (TN), true positives (TP), false negatives (FN) and false positives (FP). The major exclusion criteria were as follow: the studies that don’t include the gold standard test (the microscopic exam), technical reports, case reports, commentaries or letters to editors.

2.3. Data extraction and quality assessment

Two authors (MM, IBI) assessed independently all the manuscripts and the data of the manuscripts selected. The authors’ names, the publication year, the markers, the number of the biomarkers used, the pre-operative treatment intake, the inclusion of healthy patients (potential true negatives), the true positives (TP), false positives (FP), true negatives (TN), the false negatives (FN) and the informations needed to assess the manuscripts’ quality. The latter was based on the revised version of the QUADAS-2 [9]. The most pertinent criteria were assessed by both authors and are represented in Table 1. In case of disagreement, the authors discussed their different points of view during programmed meetings. The manuscripts included were assessed according to the four domains: the selection of the patients, the flow and timing, the index test and the reference standard. The quality of the manuscripts was assessed using the QUADAS and every criteria was rated 1 for yes, −1 for no and 0 for not mentioned. The selection criteria elements of the patients were based on the patients selection (maximum score =2), the flow and timing was rated according to the period between the reference test and the index test (maximum score= 1), the reference standard domain was rated according to the ability of the test to classify the patients, the use of the reference test regardless of the index test, the independence between the standard test and the index test and the conditions of interpretation of the reference test for all patients (maximum score =4) and the index test domain was rated based on the condition of interpretation of the index test and the details about unconclusive results (maximum score = 4). According to the quality elements chosen through the QUADAS list, the maximum score rated was 11.

Table 1.

The QUADAS criteria of the different manuscripts included.

  Patient selection
 Flow and timing Reference standard
 Index test
 Total score
  Was the spectrum of patient representative of those who will receive the test in practice? Were withdrawals from the study explained? Is the time period between reference standard and index test short enough to be reasonably sure that the target condition did not change between the 2 tests? Is the reference standard likely to correctly classify the target condition? Did patients receive the same reference standard regardless of the index test? Was the reference standard independent of the index test? Were the reference standard results interpreted without knowledge of the results of the index test? Did the whole sample or a random selection of the sample, receive verification using a reference standard of diagnosis? Were the index test results interpreted without knowledge of the results of the reference test? Were the same clinical data available when test results were interpreted as would be available when the test is used in practice? Were uninterruptible/ intermediate test results reported? Total score
Hao YJ, et al [11] 1 0 1 1 1 1 1 1 −1 1 −1 6
Abdelhady SA, et al. [12] 1 0 1 1 1 1 1 1 −1 1 −1 6
Sayagués JM, et al [38]. 1 0 1 1 1 1 0 1 −1 1 −1 5
Vallejos PA, et al [39]. 1 0 1 1 1 1 0 1 −1 1 −1 5
Overs A, et al [40]. 1 0 1 1 1 1 1 1 −1 1 −1 6
Kranenbarg RA, et al [41]. 1 0 1 1 1 1 1 1 −1 1 −1 6
Lygre KB, et al [13]. 1 0 1 1 1 1 1 1 −1 1 −1 6
Open in a new tab

2.4. Publication bias

Publication bias was evaluated using the Funnel plot followed by the Egger regression test (CMA version 4) [10].

2.5. Heterogeneity

In order to assess the heterogeneity, we calculated the I2, the Cochrane Chi 2 test (Q-test), Tau2 and 95% predictive interval (PI) (CMA version 4 and meta-Disc 2.0) [10].

2.6. Reasons for heterogeneity

Reasons for heterogeneity were investigated by testing interactions between relevant factors termed moderators (The nature of the sample used, the use of cf DNA as a biomarker, the preoperative treatment for the patients, the use of the Next-Generation sequencing technique, the inclusion of healthy patients, the use of a single biomarker, the inclusion of early and late stages of CRC and the European race) and effect size.

2.7. Sub-group analysis

A subgroup analysis was performed taking into account the TN as the event rate because we focused on the SPE.

The prediction intervals comparing the studies using the plasma as a sample versus those using the whole blood or the serum, the studies including patients receiving a preoperative treatment versus those without a preoperative treatment and the studies using cf DNA versus those using other biomarkers, were interpreted.

3. Results

Using the MeSH words, 14791 publications were initially identified, then 204 manuscripts were analyzed according to the titles and abstracts. After analyzing the summaries, the letters and the manuscripts lacking the informations needed, 7 manuscripts were retained (Figure 1). The general characteristics of the different studies were represented in Table 2.

Figure 1.

Figure 1.

Open in a new tab

PRISMA 2020 flow diagram of the meta-analysis.

Table 2.

The general characteristics of the studies included.

Authors TP FP TN FN Technique Markers Stage Race Sample TT Including normal patients
Hao YJ, et al. 2024 [11] 68 1 1 44 Platform (CytoSCM imaging@cellenvision) microfluidic chip CTCs 5 stage 0, 77 stage 1, 72 stage II, 69 stage III, 43 stage IV Asian Plasma 27 preoperation TT, 83 non-preoperation TT No
Hao YJ, et al. 2024 [11] 21 1 1 91 Platform (CytoSCM imaging@cellenvision) microfluidic chip CTM 5 stage 0, 77 stage 1, 72 stage II, 69 stage III, 43 stage IV Asian Plasma 27 preoperation TT, 83 non-preoperation TT No
Abdelhady SA, et al. 2020 [12] 29 1 29 1 Immunoassay kit from CUSABIO from DAS (Italy) GP73 15 stage I, 10 stage II, 5 stage III, 0 stage IV Arab Blood Not mentionned Yes
Abdelhady SA, et al. 2020 [12] 28 1 29 2 Immunoassay kit from CUSABIO from DAS (Italy) GP73 + ACE 15 stage I, 10 stage II, 5 stage III, 0 stage IV Arab Blood Not mentionned Yes
Sayaguès JM, et al. 2023 [38] 26 1 1 25 Cobas cfDNA (Roche, USA), Real-time PCR KRAS, NRAS, BRAF, PIK3CA 51 stage IV European Plasma Before TT No
Vallejos PA, et al. 2023 [39] 15 1 10 27 NGS Exosomes 14 stages I, II, 28 stage IV Plasma 24 preopratory TT, 15 no Yes
Overs A, et al. 2021 [40] 17 1 10 6 ctDNA by crystal digital PCR hypermethylated WIF1 and NPY genes 2 stage I, 20 stage IV European Plasma not mentionned Yes
Kranenbarg RAM, et al. 2021 [41] 19 1 20 1 Quantitative multi-plexed nested PCR cfDNA 8 gene subset 20 stage IV European Plasma Not mentionned Yes
Lygre KB, et al. 2023 [13] 47 1 1 3 NGS ctDNA Stage I European Plasma No preoperative TT No
Lygre KB, et al. 2023 [13] 27 1 1 7 ddPCR ctDNA Stage I European Plasma No preoperative TT No
Open in a new tab

Three studies were duplicated. The first one was performed by Hao YJ, et al [11]. It was duplicated because the authors used 2 different biomarkers including the circulating tumor cells and the circulating tumor microemboli. The second study was performed by Abdelhady SA, et al. It was duplicated because the authors used two different strategies. In the first strategy, they used one biomarker (GP73) and in the second strategy, they used 2 biomarkers (GP73 + ACE) [12]. The third study duplicated was the study performed by Lygre KB, et al [13]. The authors assessed the use of cf DNA but they used 2 different techniques consisting of the next-generation sequencing and the digital PCR. The different studies included 617 patients.

3.1. Quality assessment of the different studies

According to the methodology described in the methods section, the mean score attributed to the different studies accounted for 5.71/11. The major concerns were about the conditions of interpretation of the index test and the description of the intermediate results.

3.2. The diagnostic performance

The Spearman correlation coefficient accounted for −0,576 (p = 0,082), indicating the absence of the threshold effect. The different diagnostic indices were as follow: the pooled SEN and SPE accounted for 0,77 (CI 95% [0.55–0.902] and 0.943 (CI 95% [0.836–0.982]), pooled PLR and NLR accounted for 6.847 (CI 95% [2.879–16.285]) and 0.259 (95% CI [0.113–0.593]) and the pooled DOR accounted for 26.467 (95% CI [5.372–130.396]). The area under curve of the symmetric SROC accounted for 0.971 suggesting a high diagnostic potential of the biomarkers in CRC. The I2 of the SEN and SPE accounted respectively for 0.919 and 0.385. As we focused on the liquid biopsy as a diagnostic tool, we put emphasis on its specificity. When focusing on the specificity, the prediction interval accounted for [0.459–0.997]. This result highlighted a heterogeneity (Figure 2).

Figure 2.

Figure 2.

Open in a new tab

The Forest plot of the specificity highlighting a pooled specificity of 0.943. The event rate was the true negative. The prediction interval accounted for [0.459–0.997] putting emphasis on an important heterogeneity.

3.3. The publication bias and the meta-regression analysis

The publication bias assessment test revealed no potential publication bias (Egger test P = 0,292, Funnel plot symmetrical) (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

No potential publication bias was present based on the Egger test (P = 0,292) and the Funnel plot.

In order to assess the sources of heterogeneity concerning the specificity, we performed a meta-regression analysis including the sample used (plasma sample or not), the technique used (NGS or not), the number of the biomarkers assessed (single biomarker or not), the biomarkers assessed (cfDNA or not), the tumor stage (studies including early and late stages or not), the race (European race or not), the fact of including healthy patients and the presence or absence of a pre-operative treatment consisting of chemotherapy or radiation therapy. The statistical analysis revealed that the sample nature (plasma), the biomarker nature (cfDNA) and the intake of a pre-operative treatment were moderators with a significance accounting respectively for: p = 0.020, p = 0.043 and p = 0.0178. The Table 3 illustrates the different results concerning the potential moderators.

Table 3.

The meta-regression analysis putting emphasis on the plasma sample, the pre-operative treatment and the use of cfDNA biomarker as a moderators.

Potential moderators Significance (p)
Plasma sample 0.020
Pre-operative treatment 0.0178
The biomarker (cfDNA) 0.0439
The technique used (NGS) 0.0806
The inclusion of healthy patients 0.0976
The use of a single biomarker 0.5708
The stage of CRC (studies including all stages) 0.9359
The European race 0.6856
Open in a new tab

3.4. The subgroup analysis

A subgroup analysis was performed taking into account the TN as the event rate because we focused on the specificity.

The prediction interval of the studies using the plasma as a sample and those using the whole blood or the serum as sample accounted for 0 highlighting the importance of the sample used as a modulator. The same fact was observed between the studies including patients receiving a preoperative treatment and those without a preoperative treatment putting emphasis on the importance of the preoperative treatment as a modulator between the different studies.

The subgroup analysis taking into account the cf DNA illustrated a prediction interval reaching 0 highlighting the importance of differentiating the studies using the cf DNA as a biomarker in comparison to those using other biomarkers. The Figure 4 illustrates the different results of the subgroup analysis.

Figure 4.

Figure 4.

Open in a new tab

The subgroup analysis taking into account the plasma sample, the preoperative treatment intake and the use of cfDNA as a biomarker highlighted the homogeneity inside each group with a prediction interval reaching 0 for each group.

4. Discussion

4.1. Main findings

Our results revealed a high diagnostic potential of the liquid biopsy in CRC with a pooled SEN, SPE, PLR, NLR and DOR accounting respectively for 77%, 94.3%, 6.847, 0.259 and 26.467. The AUC accounted for 0.97. We focused on the specificity because of the diagnostic nature of the test that we assessed. Based on the prediction interval analysis of the specificity, we observed an important heterogeneity.

4.2. Discussing heterogeneity

Liquid biopsy has been reported and tested mainly as a screening test including other consensual methods consisting of stool-based tests, the multitarget stool DNA test and the endoscopic tests consisting of flexible sigmoidoscopy, colonoscopy, colon capsule endoscopy and the molecular tests. All these screening tests have been reported with a high SEN. Some authors reported that the analysis of liquid biopsy as a diagnostic tool revealed a high heterogeneity whose clinical significance has yet to be defined [5].

In order to assess the sources of heterogeneity, we performed a meta-regression analysis including the sample used (plasma sample or not), the technique (NGS or not), the number of the biomarkers assessed (single biomarker or not), the biomarkers assessed (cfDNA or not), the tumor stage (studies including early and late stages or not), the race (European race or not), the fact of including healthy patients in the cohort and the presence or absence of a pre-operative treatment consisting of chemotherapy or radiation therapy. The meta-regression analysis revealed the sample nature (plasma), the biomarker used (cfDNA) and the intake of a preoperative treatment as potential modulators.

Many biomarkers were described in CRC including genomic biomarkers consisting of chromosomal instability involving different pathways (Adenomatous Polyposis Coli gene (APC), gene of the protein 53 (TP53), Kristen rat sarcoma gene (KRAS), proto-oncogene B-Raf (BRAF), phosphatidylinositol 3-kinase (PIK3CA), mothers against decapentaplegic homolog 4 (SMAD4)), DNA methylation indicators especially in the early detection of CRC including N-myc downstream-regulated gene 4 (NDRG4), bone morphogenetic protein 3 (BMP3) and septin 9 (SEPT9) [2,14,15], transcriptomic biomarkers (miRNAs, circular RNA, long non coding RNA [2], circRNAs including has_circ_001978, has_circ_105039, has_circ_103627 and circ_0124554) and proteomic biomarkers [6]. The most frequently assessed biomarkers are represented by circulating tumor cells (CTC) and cfDNA. The major limitations of CTC and DNA consist of the low amounts of circulating tumor cells and circulating tumor DNA in samples. CTC are circulating tumor cells from the primary tumor or from a metastatic site. cfDNA is the free genetic material released from tumor cells during tumor turnover.

The utility of using cfDNA has been reported by some authors, besides, some authors reported that they are unlikely released in pre-invasive lesions [3]. The screening of preoperative CTC using CellSearch system has been reported to be correlated to the stage [16]. High CTC count has also been reported as correlated to a poor prognosis by some authors [17]. The screening of mutations of driver genes to detect CRC has been reported as unavailable by some authors because of evidence suggesting that non-transformed-cells such as epidermal and blood cells can carry genetic alterations in driver genes [18–23]. In this study, the subgroup analysis revealed the cfDNA as a moderator. The levels of cfDNA or mutant ctDNA has been reported as correlated with a shorter overall survival by some authors [24–27]. cfDNA and ctDNA levels have been reported as correlated with the tumor burden. The ability of the tumor cells to shed DNA in the circulation might reflect the aggressiveness of the disease [5]. Dying cells release fragmented DNA into the circulation and the cfDNA originating from tumor cells carry tumor-related alterations that can be detected using NGS techniques and PCR-based methodologies [3]. The most frequently reported technique to assess CRC was the NGS-based techniques. The NGS-based techniques have been reported as useful techniques to detect somatic mutations in the cfDNA from early CRC. This is due to the possibility of assessing multiple loci in a single reaction. Many techniques have also been reported such as the TEC-seq technology [24,28]. Methylation studies in the cfDNA have also been reported in the literature with a meta-analysis reported by Yan C, et al [29]. This meta-analysis included 14 studies enrolling 9870 patients. Pooled SEN and SPE accounted for 66 and 91% without taking into account the tumor stage and the different kits used [30]. We also assessed the technique used as a possible covariate but the meta-regression analysis proved that the NGS-based techniques weren’t potential moderators. This fact may be explained by the lack in the tests’ standardization. Some tests such as the CancerSEEK test which take into account both mutations in the cf DNA and circulating protein markers, the authors reported high SEN and SPE in early disease stages [31].

Other meta-analyses assessed the use of miRNAs in the screening of CRC. The moderators reported in the different meta-analyses consisted of the different sources (plasma vs serum vs whole blood), the testing technique and the reference test used [32]. In these different meta-analyses, in addition to the different goals and aims, the heterogeneity was assessed using the I-square and not the prediction interval. Many authors reported the higher fiability of the prediction interval in order to assess the heterogeneity among the different studies included in a meta-analysis [10,33,34].

In a meta-analysis performed by Zeng W, et al. [35], the authors put emphasis on the fact that using panels of miRNAs had a higher SEN and SPE than using single miRNA. Besides, the overlap between the different signatures was minimal. In this study, the meta-regression analysis didn’t highlight the number of biomarkers as a potential moderator. Other meta-analysis focused on the prognostic of the sampling site (peripheral blood, mesenteric portal blood or bone marrow) [36]. The authors reported that the detection of CTC in peripheral blood and not in mesenteric portal blood or bone marrow was correlated with poor prognosis. In our study, the meta-regression analysis and the subgroup analysis highlighted the importance of using plasma samples as a moderator.

In our study, we assessed the factor of using preoperative chemotherapy as a potential modulator. In some studies, the authors reported that patient receiving standard first-line combination chemotherapy had a significantly higher radiologic response when correlated with the reduction in ctDNA levels [37].

The meta-regression analysis didn’t highlight the stage as a modulator. This fact can be discussed because of the difference in five-year survival between stage I and stage IV that account for respectively 90% versus less than 10% [2].

4.3. Strengths and limitations of our study

The major advantage of this study was to assess the diagnostic value of the liquid biopsy as a surrogate to tissue biopsy. The gold standard for the diagnosis of CRC consists of tissue biopsy analysis. Its indisputable limitations are represented by the invasiveness, the lack of patient compliance, their incompetence in reflecting the dynamic heterogeneity of tumor and the impossibility for a long-term surveillance [11].

The major limitations of this meta-analysis were the small dataset and the quality of the studies included. In fact, the mean score attributed to the different studies accounted for 5.71/11 with the major concerns being about the conditions of interpretation of the index test and the description of the intermediate results. Besides, the most promising detection techniques are NGS, qPCR or dPCR-based techniques. The combination of CTCs, protein and mutation detection in our study was due to the limited number of studies (7) but it made an important compromise on the results of this work.

The heterogeneity observed in this study reinforces the need for future research focusing on harmonizing the use of particular samples, the biomarker assessed and making attention to the treatment intake before surgery.

5. Conclusion

This meta-analysis highlighted the high diagnostic potential of liquid biopsy as a surrogate to tissue samples but the importance of the heterogeneity which was assessed using the prediction interval put emphasis on the necessity of taking these results with caution. Besides, the main controversies related to CRC diagnosis using liquid biopsy are represented by the cost of the molecular techniques in comparison to the gold standard which consists of the microscopic exam.

Funding Statement

This paper was not funded.

Authors’ contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Mona Mlika. The first draft of the manuscript was written by Mona Mlika, Majdi Zorgati, Imen Ben Ismail and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

References

  • 1.Siegel RL, Kratzer TB, Giaquinto AN, et al. Cancer statistics, 2025. CA Cancer J Clin. 2025;75(1):10–45. doi: 10.3322/caac.21871 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Zhang Y, Wang Y, Zhang B, et al. Methods and biomarkers for early detection, prediction, and diagnosis of colorectal cancer. Biomed Pharmacother. 2023;163:114786. doi: 10.1016/j.biopha.2023.114786 [DOI] [PubMed] [Google Scholar]
  • 3.Myint NNM, Verma AM, Fernandez-Garcia D, et al. Circulating tumor DNA in patients with colorectal adenomas: assessment of detectability and genetic heterogeneity. Cell Death Dis. 2018;9(9):894. doi: 10.1038/s41419-018-0934-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Emile SH, Barsom SH, Wexner SD.. An updated review of the methods, guidelines of, and controversies on screening for colorectal cancer. Am J Surg. 2022;224(1 Pt B):339–347. doi: 10.1016/j.amjsurg.2022.03.034 [DOI] [PubMed] [Google Scholar]
  • 5.Normanno N, Cervantes A, Ciardiello F, et al. The liquid biopsy in the management of colorectal cancer patients: Current applications and future scenarios. Cancer Treat Rev. 2018;70:1–8. doi: 10.1016/j.ctrv.2018.07.007 [DOI] [PubMed] [Google Scholar]
  • 6.Noothalapati H, Iwasaki K, Yamamoto T.. Non-invasive diagnosis of colorectal cancer by Raman spectroscopy: Recent developments in liquid biopsy and endoscopy approaches. Spectrochim Acta A Mol Biomol Spectrosc. 2021;258:119818. doi: 10.1016/j.saa.2021.119818 [DOI] [PubMed] [Google Scholar]
  • 7.Mlika M, Zorgati MM, Abdennadher M, et al. The diagnostic performance of micro-RNA and metabolites in lung cancer: A meta-analysis. Asian Cardiovasc Thorac Ann. 2024;32(1):45–65. doi: 10.1177/02184923231215538 [DOI] [PubMed] [Google Scholar]
  • 8.Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Syst Rev. 2021;10(1):89. doi: 10.1186/s13643-021-01626-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Whiting PF, Rutjes AWS, Westwood ME, et al. QUADAS-2: A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies; 2011. www.annals.org. [DOI] [PubMed]
  • 10.Borenstein M, Hedges L, Rothstein H. Introduction to Meta-Analysis; 2007. www.Meta-Analysis.com.
  • 11.Hao YJ, Chang LW, Yang CY, et al. The rare circulating tumor microemboli as a biomarker contributes to predicting early colorectal cancer recurrences after medical treatment. Transl Res. 2024;263:1–14. doi: 10.1016/j.trsl.2023.07.011 [DOI] [PubMed] [Google Scholar]
  • 12.Abdelhady SA, Attya H, Abdo M, et al. Clinical significance of golgi protein-73 as a diagnostic marker for Egyptian patients with colorectal cancer: Preliminary study. Cancer Rep. 2021;4(5):e1379. doi: 10.1002/cnr2.1379 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Lygre KB, Forthun RB, Høysæter T, et al. Assessment of postoperative circulating tumour DNA to predict early recurrence in patients with stage I–III right-sided colon cancer: prospective observational study. BJS Open. 2024;8(1):12–32. doi: 10.1093/bjsopen/zrad146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Massen M, Lommen K, Wouters KAD, et al. Technical considerations in PCR-based assay design for diagnostic DNA methylation cancer biomarkers. Clin Epigenetics. 2022;14(1):56. doi: 10.1186/s13148-022-01273-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Song L, Peng X, Zhang Y, et al. AMR-to-text generation with synchronous node replacement grammar. Association for Computational Linguistics (ACL); 2017, pp. 7–13. [Google Scholar]
  • 16.Bork U, Rahbari NN, Schölch S, et al. Circulating tumour cells and outcome in non-metastatic colorectal cancer: A prospective study. Br J Cancer. 2015;112(8):1306–1313. doi: 10.1038/bjc.2015.88 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Denève E, Riethdorf S, Ramos J, et al. Capture of viable circulating tumor cells in the liver of colorectal cancer patients. Clin Chem. 2013;59(9):1384–1392. doi: 10.1373/clinchem.2013.202846 [DOI] [PubMed] [Google Scholar]
  • 18.Gormally E, Vineis P, Matullo G, et al. TP53 and KRAS2 mutations in plasma DNA of healthy subjects and subsequent cancer occurrence: A prospective study. Cancer Res. 2006;66(13):6871–6876. doi: 10.1158/0008-5472.CAN-05-4556 [DOI] [PubMed] [Google Scholar]
  • 19.Fernandez-Cuesta L, Perdomo S, Avogbe PH, et al. Identification of Circulating Tumor DNA for the Early Detection of Small-cell Lung Cancer. EBioMedicine. 2016;10:117–123. doi: 10.1016/j.ebiom.2016.06.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Schwaederle M, Husain H, Fanta PT, et al. Oncotarget 9707 www.impactjournals.com/oncotarget Detection rate of actionable mutations in diverse cancers using a biopsy-free (blood) circulating tumor cell DNA assay. www.impactjournals.com/oncotarget/ [DOI] [PMC free article] [PubMed]
  • 21.Allenson K, Castillo J, San Lucas FA, et al. High prevalence of mutant KRAS in circulating exosome-derived DNA from early-stage pancreatic cancer patients. Ann Oncol. 2017;28(4):741–747. doi: 10.1093/annonc/mdx004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Genovese G, Kähler AK, Handsaker RE, et al. Clonal Hematopoiesis and Blood-Cancer Risk Inferred from Blood DNA Sequence. N Engl J Med. 2014;371(26):2477–2487. doi: 10.1056/NEJMoa1409405 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Martincorena I, Roshan A, Gerstung M, et al. High burden and pervasive positive selection of somatic mutations in normal human skin. Science. 2015;348(6237):880–886. doi: 10.1126/science.aaa6806 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med. 2014;6(224):224ra24. doi: 10.1126/scitranslmed.3007094 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Tabernero J, Lenz HJ, Siena S, et al. Analysis of circulating DNA and protein biomarkers to predict the clinical activity of regorafenib and assess prognosis in patients with metastatic colorectal cancer: A retrospective, exploratory analysis of the CORRECT trial. Lancet Oncol. 2015;16(8):937–948. doi: 10.1016/S1470-2045(15)00138-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.El Messaoudi S, Mouliere F, Du Manoir S, et al. Circulating DNA as a strong multimarker prognostic tool for metastatic colorectal cancer patient management care. Clinical Cancer Research. 2016;22(12):3067–3077. doi: 10.1158/1078 [DOI] [PubMed] [Google Scholar]
  • 27.Fan G, Zhang K, Yang X, et al. Prognostic value of circulating tumor DNA in patients with colon cancer: Systematic review. PLoS One. 2017;12(2):e0171991. doi: 10.1371/journal.pone.0171991 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Phallen J, Sausen M, Adleff V, et al. Direct detection of early-stage cancers using circulating tumor DNA. Sci Transl Med. 2017;9(403):12;23. doi: 10.1126/scitranslmed.aan2415 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Yang CK, Hsu HC, Liu YH, et al. EV-miRome-wide profiling uncovers miR-320c for detecting metastatic colorectal cancer and monitoring the therapeutic response. Cell Oncol (Dordr). 2022;45(4):621–638. doi: 10.1007/s13402-022-00688-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Xie L, Jiang X, Li Q, et al. Diagnostic value of methylated Septin9 for colorectal cancer detection. Front Oncol. 2018;8:247. doi: 10.3389/fonc.2018.00247 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Cohen JD, Li L, Wang Y, et al. Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science (1979). 2018;359(6378):926–930. doi: 10.1126/science.aar3247 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Yu W, Wang Z, Shen L, et al. Circulating microRNA-21 as a potential diagnostic marker for colorectal cancer: A meta-analysis. Mol Clin Oncol. 2016;4(2):237–244. doi: 10.3892/mco.2015.702 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Dziri C, Fingerhut A.. Up-to-date composition and critical appraisal of meta-analyses of comparative studies. Ann Laparosc Endosc Surg. 2025;10:4–4. doi: 10.21037/ales-24-26 [DOI] [Google Scholar]
  • 34.Borenstein M. Common mistakes in meta-analysis and how to avoid them. www.Meta-Analysis.com. [DOI] [PubMed]
  • 35.Zeng W, Tu Y, Zhu Y, et al. Predictive power of circulating miRNAs in detecting colorectal cancer. Tumour Biol. 2015;36(4):2559–2567. doi: 10.1007/s13277-014-2872-2 [DOI] [PubMed] [Google Scholar]
  • 36.Rahbari NN, Aigner M, Thorlund K, et al. Meta-analysis shows that detection of circulating tumor cells indicates poor prognosis in patients with colorectal cancer. Gastroenterology. 2010;138(5):1714–1726. doi: 10.1053/j.gastro.2010.01.008 [DOI] [PubMed] [Google Scholar]
  • 37.Tie J, Kinde I, Wang Y, et al. Circulating tumor DNA as an early marker of therapeutic response in patients with metastatic colorectal cancer. Ann Oncol. 2015;26(8):1715–1722. doi: 10.1093/annonc/mdv177 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Sayagués JM, Montero JC, Jiménez-Pérez A, et al. Analysis of circulating tumor DNA in synchronous metastatic colorectal cancer at diagnosis predicts overall patient survival. Int J Mol Sci. 2023;24(9):8438. doi: 10.3390/ijms24098438 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Vallejos PA, Gonda A, Yu J, et al. Plasma exosome gene signature differentiates colon cancer from healthy controls. Ann Surg Oncol. 2023;30(6):3833–3844. doi: 10.1245/s10434-023-13219-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Overs A, Flammang M, Hervouet E, et al. The detection of specific hypermethylated WIF1 and NPY genes in circulating DNA by crystal digital PCRTM is a powerful new tool for colorectal cancer diagnosis and screening. BMC Cancer. 2021;21(1):1092. doi: 10.1186/s12885-021-08816-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Klein Kranenbarg RAM, Vali AH, IJzermans JNM, et al. High performance methylated DNA markers for detection of colon adenocarcinoma. Clin Epigenetics. 2021;13(1):218. doi: 10.1186/s13148-021-01206-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Future Science OA are provided here courtesy of Taylor & Francis

RESOURCES