Comprehensive evaluation and progress of colorectal cancer screening methods
- Open Access
- 01.12.2025
- Review
Erschienen in:
Abstract
Introduction
According to statistics from the International Agency for Research on Cancer (IARC) GLOBOCAN program, CRC is the third most common Malignancy and the second leading cause of Malignancy-related deaths worldwide in 2022. According to the data, it is estimated that there will be more than 1.9 million new cases of CRC (including anal cancer) and more than 900,000 deaths in 2022, representing about one-tenth of all malignant tumors [1]. There are significant differences in CRC morbidity and mortality between developed countries and countries with economies in transition, and the global burden of the disease has continued to increase in recent years. The incidence of CRC has declined in many high-prevalence countries over the past few decades, which is thought to be the result of population shifts towards healthier lifestyles (e.g., greater intake of fiber, such as fruits and vegetables) and the introduction of screening.
Early screening is central to the prevention and control strategy for CRC, with approximately 70% of patients with CRC being disseminated cases, and the development of CRC following an “adenoma-carcinoma” progression, which typically takes 5–10 years, providing a window of time for early detection and screening of cancer and precursor lesions [2]. In addition, CRC prognosis is highly correlated with diagnostic staging, with 5-year survival rates of TNM stage I CRC patients reaching more than 90%, compared with only about 12% for TNM stage IV [3]. Therefore, early diagnosis and treatment are key to improving the prognosis and reducing the burden of disease in patients with CRC. Screening can reduce the incidence of CRC in the population by detecting and removing precancerous lesions. At the same time, screening can facilitate early diagnosis and timely treatment of CRC, thereby reducing mortality and improving survival.
Anzeige
Effective screening methods can be used as part of a broader population-based screening service [4]. Existing screening methods, including fecal immunochemical testing (FIT), multi-target stool DNA test (mt-sDNA), and colonoscopy, each have their own limitations and challenges. FIT, although easy to perform, may result in false positives, and further colonoscopy is required to confirm the diagnosis based on a positive result [5, 6]. While colonoscopy, as the gold standard for CRC diagnosis, is highly accurate, it is highly invasive, involves some discomfort and risk, and is more demanding in terms of medical resources. In addition, there are some challenges in CRC screening, such as insufficient awareness of screening, uneven distribution of medical resources, and selection of screening technologies and protocols. Emerging screening technologies, such as blood-based DNA methylation, protein markers, and fecal DNA testing, have demonstrated high sensitivity and specificity, but are currently costly and not yet widely used for population screening. In the future, these emerging technologies are expected to provide more options for CRC screening, but further research is needed to provide health economic evidence and optimize screening protocols (Fig. 1).
CRC screening is critical and has its own advantages and disadvantages in improving patient survival and reducing mortality. Therefore, this article reviews the progress of research on biomarker and imaging screening methods based on fecal, blood and urine samples.
Literature search methods
This study conducted a systematic literature review to identify studies related to CRC screening methods published from January 2000 to March 2025. The databases searched included PubMed, Web of Science, and Embase. The core search terms were: CRC screening, early diagnosis, FIT, mt-sDNA, colonoscopy, circulating tumor DNA (ctDNA), biomarkers, microbiome, urine metabolites, and imaging techniques. Specific biomarkers (e.g., miRNA, Septin9, N1, N12-diethylamine) and methodological approaches (e.g., CT colonography) were also included. Inclusion criteria encompassed: original studies, meta-analyses, systematic reviews, and clinical guidelines; human subjects; focus on screening sensitivity/specificity, cost-effectiveness, or technological innovation; and English-language literature. Exclusion criteria included: animal studies, case reports, editorials, non-English-language literature, and studies unrelated to screening diagnosis. After deduplication, the literature was screened based on titles/abstracts. To ensure comprehensiveness, we manually supplemented the search by reviewing the references of key reviews and guidelines. Data extraction focused on study design, sample size and performance metrics.
Colorectal cancer screening methods
Questionnaire
The development of CRC has its specific risk factors, such as family history of CRC, history of intestinal polyps, lifestyle habits, etc. Existing high-risk factor questionnaires based on CRC risk factors, which are more frequently used in clinical practice, include the High Risk Factor Questionnaire (HRFQ) survey and the Asia-Pacific Colorectal Tumor Screening Score (APCS), which allows for the identification of high-risk populations by answering the questionnaires. A systematic evaluation and meta-analysis examined the performance of different CRC screening strategies, including the Risk Assessment Questionnaire (RAQ) and the Fecal Immunochemical Test (FIT) as the primary screening method. The results of the study showed that the incidence of CRC was significantly higher in the high-risk group compared to the low-risk group, highlighting the importance of questionnaires in identifying high-risk individuals [7].
Anzeige
CRC screening methods based on fecal testing
Guaiac-based fecal occult blood test (gFOBT) and fecal immunochemical test (FIT)
The fecal occult blood test (FOBT) is a noninvasive CRC screening test that has been widely used in many screening programs and is recommended by current CRC screening guidelines [8, 9]. Overall, there are two types of FOBTs that use different mechanisms to identify fecal occult blood. Traditionally, guaiac-based FOBT (gFOBT) is used to detect the peroxidase activity of heme, which is less reliable due to the disadvantages of low sensitivity and specificity, high false-negative and false-positive rates, and high influence of environmental factors (e.g., diet, medications). In randomized trials and observational studies using different FOBTs, the sensitivity of gFOBT for detecting CRC was 31%−79% and the specificity was 87%−98% [10].
Another type of FOBT is FIT, which uses antibodies to specifically detect human hemoglobin in feces, which Makes FIT less likely to result in false-positive results due to composition. A meta-analysis of FIT by Lee et al. determined that the one-time assay had a sensitivity of 79% (95% CI 0.69–0.86) and specificity of 94% (95% CI 0.92–0.95) [11]. However, the sensitivity and specificity of FIT for detecting advanced adenomas were lower, estimated at 25–56% and 68–96%, respectively. Furthermore, although FIT has a higher detection rate of CRC and advanced adenomas compared to gFOBT, it may have a lower sensitivity for right-sided colonic lesions. In addition, FIT positivity is highly specific for lower gastrointestinal bleeding [12]. Compared to gFOBT, FIT has been shown to have better analytical and clinical sensitivity as well as higher detection of CRC and its precursors, improving compliance and cost-effectiveness [13].
In summary, gFOBT is a cost-effective means of mass CRC screening, but has major limitations. In the United States, most gFOBTs have been replaced by FIT, but gFOBTs are still used as the primary tool for CRC screening in high-risk populations in some countries. While both gFOBT and FIT detect occult blood, a key indicator of CRC and advanced adenomas, they primarily reflect bleeding events and have inherent limitations in sensitivity, particularly for early-stage lesions and sessile serrated polyps. To enhance the secondary prevention of CRC, a promising future direction involves individually combining either gFOBT or FIT with other non-invasive screening biomarkers. Such multimodal approaches aim to leverage complementary biological signals to improve overall sensitivity, specificity, and the detection of precancerous states beyond what fecal hemoglobin detection alone can achieve (Fig. 2).
Fig. 1
CRC screening methods
Fig. 2
Stool-based CRC screening methods
Fig. 3
Blood-based CRC screening methods. This figure schematically illustrates key biomarkers in peripheral blood utilized for CRC screening and diagnosis, including CTCs, exosomes, ctDNA, miRNAs, and proteins. Specifically, CTCs directly reflect tumor cell shedding and metastatic potential. ctDNA released from necrotic tumor cells enables tumor-specific characteristics revelation through analytical interrogation. Exosomes serve as intercellular communication vehicles transporting tumor-specific miRNAs, proteins, and other molecular cargo, while cell-free proteins collectively constitute critical diagnostic targets in liquid biopsy approaches
Multi-target stool DNA test (mt-sDNA)
Multi-target stool DNA test (mt-sDNA) mainly relies on the combined detection of DNA methylation and DNA mutations. SFRP2 gene methylation was the first reported CRC-associated fecal DNA methylation marker, and the results of a meta-analysis showed that it has high sensitivity (71%) and specificity (94%) with an AUC of 0.94 [14]. In addition, studies have shown that the waveform protein gene (VIM) methylation has a high specificity for diagnosing CRC, but the sensitivity varies widely between studies [15]. In addition, gene methylation of NDRG4, APC, ATM, BMP3, SFRP2, and MLH1 have been successively reported to be used for early diagnosis of CRC [15‐17]. Compared with DNA methylation, although there are fewer studies related to DNA Mutations, there is also evidence that they have some diagnostic value for CRC. It is worth mentioning that the combination of multiple Markers can enhance the diagnostic sensitivity and specificity of CRC. Based on the evidence of a cross-sectional study conducted in a general risk population of 4,482 cases, Ahlquist et al. found that the combined detection of APC and KRAS Mutations as well as VIM methylation resulted in a sensitivity of 58% for the diagnosis of CRC, and the combination of BMP3 and NDRG4 resulted in a sensitivity of 98% and a specificity of 90% [18]; VIM combined with SFRP2, with a sensitivity of 92.50% and specificity of 91.20% [19]; And multi-targeted fecal DNA test (Cologuard) consisting of KRAS mutations, BMP3 and NDRG4 promoter methylation, combined by β-actin and hemoglobin assay, with a sensitivity of 92.3% and specificity of 86.60% [19], and is the first CRC screening test approved by the U.S. Food and Drug Administration and promoted nationwide.
Studies have found that mt-sDNA has a sensitivity of 91.9% for stages I-III CRC, significantly higher than the sensitivity of FIT (29.7%) evaluated in the same cohort. In addition, sensitivity increases with lesion size and the grading of adenomas and sessile serrated polyps [20]. MT-sDNA assay has been shown to be more sensitive and specific than FOBT for early detection of CRC and advanced adenomas. Compared to FOBT, Barnell et al. were more sensitive in detecting CRC (90.0% vs. 42.0%) and advanced adenomas (70.6% vs. 19.6%) with mt-sDNA. In addition, it showed excellent specificity in the detection of CRC (94.0% vs. 90.0%) [21].
Anzeige
Fecal RNA markers and protein markers
MicroRNAs (miRNAs) are small non-coding RNAs that are involved in the regulation of gene expression by binding to mRNAs and are highly conserved and stable [22]. Previous studies have preliminarily demonstrated that some miRNA molecules in feces (e.g., miR-21, miR-92a, miR-29b, miR-135, and miR-20a [23‐25]) are differentially expressed in patients with CRC and in healthy populations, and could potentially be used for the early diagnosis of CRC (Fig. 2). As observed in many recent studies, the expression of specific miRNAs is altered during the early stages of CRC development, providing potential avenues for prevention and early detection of CRC [26‐28]. Pardini et al. [26] evaluated the miRNA profiles in fecal samples from two European cohorts of patients using a Machine learning model, and found that 25 miRNAs were differentially expressed in CRC patients were differentially expressed, and a panel of five miRNAs consisting of miR-607-5p, miR-6777-5p, miR-4488, miR-149-3p, and miR-1246 was isolated with the highest discriminatory power, with AUCs ranging from 0.86 ± 0.01 to 0.96 ± 0.01 for the models containing this panel. The expression profile of the fecal miRNA profile was comprehensively evaluated as a biomarker for non-invasive CRC diagnosis.
Table 1
Part of biomarkers for CRC stool
Biomarker | Stool | References | ||
|---|---|---|---|---|
Sensitivity (%) | Specificity (%) | |||
gFOBT | 31.0–79.0 | 87.0–98.0 | [10] | |
FIT | 79.0 | 94.0 | [11] | |
mt-sDNA | SFRP2 | 71.0 | 94.0 | [14] |
SFRP2, TFPI2, NDRG4, BMP3 | 94.0 | 55.0 | [48] | |
NDRG4 | 28.57 | 97.5 | [49] | |
COL4A2、TLX2 | 91.0 | 98.0 | [50] | |
NDRG4, BMP3 | 98.0 | 90.0 | [51] | |
NDRG4, BMP3, mutation KRAS, hemoglobin | 92.0–98.0 | 87.0–90.0 | [51] | |
MLH1, VIM, MGMT | 75.0 | 86.5 | [52] | |
SDC2 | 81.0 | 95.0 | [53] | |
VIM | 38.0–81.0 | 86.9 ~ 100.0 | [15] | |
miRNA | miR-20a | 36.0 | 86.8 | [54] |
miR-21 | 71.1 | 71.5 | [55] | |
miR-29a | 85.0 | 61.0 | [56] | |
miR-92a | 71.6 | 73.3 | [55] | |
miR-135 | 78.0 | 68.0 | [57] | |
miR-149-3p | – | – | [26] | |
miR-607-5p | ||||
miR-1246 | ||||
miR-4488 | ||||
miR-6777-5p | ||||
Protein | FC | 92.8 | 41.7 | [58] |
M2-PK | 80.3 | 95.2 | [31] | |
MMP-9 | 72.2 | 95 | [59] | |
Microbiota | Fusobacterium nucleatum | 70.0 | 79.0 | [60] |
pks + Escherichia coli | 93.3 | 73.3 | [61] | |
clbA + bacteria | 81.5 | 56.4 | [62] | |
‘m3’ from a Lachnoclostridium | 62.1 | 78.5 | [63] | |
Fusobacterium nucleatum/Bifidobacterium | 84.6 | 92.3 | [64] | |
Table 2
Part of biomarkers for CRC blood screening
Sensitivity (%) | Specificity (%) | References | ||
|---|---|---|---|---|
CTCs | 78-85.3 | 75.4–78.5 | [81] | |
ctDNA | Septin9 | 47–87 | 89–98 | |
ALX4 | 83.3 | 70 | [87] | |
SFRP2 | 63.8 | 97.3 | [88] | |
SDC2 | 69–87 | 95.2 | [89] | |
BCAT1 | 47.3–64.9 | 96.5 | [90] | |
IKFZ | 48-67.6 | 95.1 | ||
VIM | 59 | 93 | [92] | |
miRNA | miR-21 | 77 | 83 | [93] |
miR-29a | 96.1 | 31.6 | [77] | |
miR-215b | 93.2 | 32.6 | ||
miR-145 | 85.4 | 43.8 | ||
miR-21-5p | 76.1 | 84.4 | Nearly all papers | |
miR-92a-3p | 74.8 | 81.5 | [55] | |
miR-135b-5p | 93.1 | 72.7 | [94] | |
miR-223-3p | – | – | [95] | |
Table 3
Advantages and disadvantages of common CRC screening methods
Screening methods | Advantage | Disadvantage |
|---|---|---|
Questionnaire | Cheap | Time-consuming, laborious, low positive rate |
FOBT | Non-invasive | Susceptible to false positives due to diet and medications, low accuracy, low detection rate and poor sensitivity in progressive adenomas |
Stool marker testing | Non-invasive | Still in the research phase, expensive and less recognized |
Blood marker tests | Non-invasive | High false-positive and false-negative rates, lack of criteria, low sensitivity and specificity for diagnosing progressive adenomas and early carcinomas |
Urine marker testing | Non-invasive | Current clinical evidence is insufficient |
Endoscopy | The gold standard of diagnosis | Invasive examination, strict requirements for bowel preparation, high cost of examination, poor subjective feelings of patients |
Computed tomography colonography | Non-invasive | Presence of radiation, stringent bowel preparation requirements, high level of imaging physician reading required |
Anzeige
Fecal protein marker-based tests are gaining attention in CRC screening, especially fecal calprotectin (FC), M2-type pyruvate kinase (M2-PK) and matrix metallopeptidase-9 (MMP-9). A retrospective study showed that patients with right-sided CRC were more likely to have higher FC levels compared to left-sided CRC (92% vs. 74%) and that tumor stage III/IV and right-sided tumor localization were associated with FC levels ≥ 50 µg/g [29]. Another systematic evaluation and meta-analysis noted that CRC patients were more likely to have elevated FC levels than controls, nevertheless, the low specificity of FC limits its use in the diagnosis or screening of CRC [30]. M2-PK is expressed in tumor cells and is involved in the regulation of metabolism [31]. Several studies have demonstrated the satisfactory performance of M2-PK in detecting CRC. Its sensitivity ranges from 73 to 92.8%, while the specificity ranges from 83.3 to 95.2%, which is slightly lower compared to FOBT [32]. Fecal MMP-9 was able to differentiate between patients with CRC and those without polyps with high sensitivity and specificity. In addition, fecal MMP-9 was able to identify nearly 60% of patients with high-risk adenomas, suggesting that fecal MMP-9 may become a new non-invasive diagnostic marker [33].
Fecal microbial testing
Improvements in sequencing platforms and advances in histologic technologies have allowed for a growing body of research on the association between human gut microbes and cancer. Case-control studies, in vitro studies, and animal models from different regions and ethnic sources have shown that dysbiosis of the intestinal flora is associated with the process of CRC development, progression, and regression [34, 35]. Among them, Fusobacterium nucleatum (Fn), as well as some specific strains such as Escherichia coli and Bacteroides fragilis, play important roles in colorectal mucosal carcinogenesis. For example, Fn produces the adhesin FadA which is involved in the activation of the β-linker protein Wnt pathway and promotes the development of CRC [36]. Several studies have confirmed significant differences in the fecal microbiome of patients with different CRC stages [37, 38]. During the development of CRC, pathogenic gut bacterial species and numbers are increasing; for example, Fn is significantly enriched in the feces of healthy individuals, colorectal adenoma (CRA) patients, and CRC patients [39‐41]. In addition, metabolites of the gut microbiota, such as short-chain fatty acids (SCFAs) [42] and bile acids [43], have been suggested as potential markers for CRC screening. Changes in these metabolites may reflect gut microbiota-host interactions and the influence of the microbiota on host metabolism.
In addition, the role of gut fungi in CRC has been increasingly recognized. Cross-cohort metagenomic analyses revealed a significant enrichment of Aspergillus species, particularly Aspergillus rambellii, in CRC patients. This fungal proliferation potentially drives tumorigenesis through upregulating the D-amino acid metabolic pathway and promoting aberrant butyrate metabolism [44]. Experimental studies have demonstrated that Aspergillus rambellii enhances CRC cell proliferation, suppresses apoptosis, and significantly accelerates tumor growth in xenograft models [45]. Meanwhile, studies have identified that the abundance of 14 fungal biomarkers distinguishes CRC patients from healthy subjects, suggesting these microbial signatures may serve as potential diagnostic tools for CRC [46].
The role of the gut virome in CRC has garnered increasing scientific attention. Recent studies employing ultra-deep metagenomic sequencing have identified 14 viral signatures, which, when integrated with FIT, elevate the diagnostic sensitivity for CRC to 97.1% [47]. Although the functional characterization of viral signatures remains challenging, their potential in non-invasive screening has provided novel avenues for early CRC diagnosis (Table 1).
Anzeige
CRC screening methods based on blood testing
Circulating tumor cell assay
The existence of circulating tumor cells (CTCs) was first reported by Australian pathologist Thomas Ashworth in 1869, but CTCs really came into the limelight in the mid-1990s. Studies have shown that CTCs, which are clones derived from primary tumor cells, are present in the early stages of CRC genesis and play an important role in tumor formation, progression, and metastasis [65, 66]. CTCs are found in very low levels in the peripheral circulation of healthy individuals and patients with benign colorectal diseases, which Makes the detection of CTCs as a method of early screening for CRC possible (Fig. 3). A CellSearch platform study based on immunologic principles found that the sensitivity of CTCs in patients with stage I-IV CRC was 4.9%, 10.5%−20.7%, 8.3%−24.1%, and 18.8%−60.7%, respectively, with an overall sensitivity of 10.5% [67]. The CTC-Biopsy platform, which is based on the physical principle of screening CTCs, showed better data than the CellSearch platform, with the corresponding sensitivities of 12.5%, 31%, 74%, and 91.7%, respectively, and an overall sensitivity of 52.8% [68]. CTCs detection, as a form of “liquid biopsy”, offers advantages over traditional histological biopsies, including minimal invasiveness, repeatable testing, high specificity, and sensitivity. It provides real-time information on disease status in CRC patients, facilitating early screening, prognostic evaluation, and treatment response monitoring. While demonstrating potential in CRC screening, CTC detection still faces several limitations: false-negative results due to CTC heterogeneity, technical constraints of detection methods, subjectivity in interpretation, cost and operational complexity, unvalidated clinical utility, and potentially insufficient sensitivity and specificity compared to other screening modalities. These challenges need to be addressed through technological advancements and large-scale clinical studies to enhance the accuracy and practical applicability of CTC detection in CRC screening.
Circulating tumor DNA detection
Circulating tumor DNA (ctDNA) refers to free tumor tissue DNA in the blood, which can originate from tumor tissue necrosis, apoptosis release, lysis of circulating tumor cells, or direct secretion by tumor cells [69], and some of these DNAs have certain tumor specificity and can be used for tumor screening and early diagnosis [70, 71] (Fig. 3). Among those that have received more attention in recent years are methylation of the Septin9 gene, which is mainly associated with cytoplasmic division and cell cycle regulation [72]. The results of a screening study for the average risk of developing CRC in an asymptomatic population showed an overall sensitivity of 48% and an overall specificity of 92% for Septin9 gene methylation [73]. In addition to Septin9 gene methylation, successive small-sample clinical case-control studies have suggested that a series of other DNA methylations in blood have the potential to be used as CRC screening and early diagnostic markers, including ALX4, APC, SFRP2, SDC2, SPG20, etc [71, 74, 75]., but evidence from large-sample prospective cohort studies is still needed to further validate its diagnostic value. Besides, mutation screening of p53, KRAS, NRAS, BRAF and other genes in ctDNA can facilitate the identification of CRC-related biomarkers with high specificity and sensitivity [76].
Detection of RNA markers in blood
Both serum/plasma and fecal-based miRNAs can be used as biomarkers for CRC screening (Fig. 3). A case performed by Liu et al. [77] purified miRNAs from plasma samples and assessed the predictive ability of ten different miRNA sequences for CRC risk. Three of them, miR-29a, miR-125b, and miR-145, were significantly associated with CRC. Assessment of the predictive value of these miRNAs using ROC curve analysis revealed an increase in AUC from 0.61 to 0.71 after incorporation of these miRNAs into the model, suggesting that these miRNAs can be used to enhance CRC prediction models. Among the extensively studied miRNAs, miR-21-5p and miR-92a-3p and their cluster members (including miR-18a-5p, miR-29a-3p, and miR-20a-5p) were found to be the candidate genes to be focused on, and miR-135b-5p, miR-223-3p, miR-139-3p, and miR-4516 were found to have the potential to facilitate CRC detection.
In addition, long non-coding RNAs (lncRNAs), a class of non-coding RNAs with more than 200 nucleotides, have been found to be involved in regulating tumorigenesis and progression in recent years. Studies have shown that RP11, RPPH1, H19, SNHG11 and NEAT1 are significantly up-regulated in CRC tissues and can be considered as new biomarkers for early diagnosis of CRC [78].
Anzeige
Detection of soluble protein markers in blood
Some protein macromolecules such as antigens and cytokines in blood also have potential for CRC screening and early diagnosis (Fig. 3). Wilhelmsen et al. [79] found that a series of serum protein molecules represented by serum carcinoembryonic antigen (CEA) and CA19-9 had AUCs of 0.52 ~ 0.65 for CRC and high-risk adenomas. The findings of Rho et al. [80] showed that BAG family molecular chaperone regulator 4, interleukin-6 receptor subunit beta, vascular hemophilia von Willebrand factor, epidermal growth factor receptor, and a combination of CD44 glycoproteins are also promising for the diagnosis of early CRC (Table 2).
CRC screening methods based on urine testing
In recent years, the potential use of urinary metabolites in CRC screening has received widespread attention. In particular, metabolites such as N ~ 1, N~(12)-diacetylspermine (DiAcSpm), kynurenine, taurine, alanine, and 3-aminoisobutyric acid have been identified as promising biomarkers. Studies have shown that the concentration of DiAcSpm is significantly elevated in patients with CRC and that it is strongly associated with tumor progression and prognosis [96]. The increase of kynurenine, a tryptophan metabolite, in the urine of CRC patients may reflect the immunomodulatory processes in the tumor microenvironment [97]. A study using liquid chromatography-mass spectrometry to screen urine found that urinary diacetylspermine and kynurenine had a specificity of 80.0% and a sensitivity of 80.0% for CRC [98]. In addition, recent metabolomics studies have shown that amino acids such as taurine and alanine also exhibit significant changes in the urine of CRC patients and may be involved in the metabolic reprogramming of tumor cells [99]. In summary, the detection of these metabolites in urine provides a new research direction and clinical application prospect for early screening of CRC.
Colonoscopy
Colonoscopy and microscopic biopsy are now the gold standard for CRC screening, diagnosis and follow-up. Studies have shown that colonoscopy is associated with reduced morbidity and mortality from CRC. A systematic evaluation and meta-analysis showed that colonoscopy was associated with a 52% reduction in CRC incidence (relative risk RR: 0.48) and a 62% reduction in CRC mortality (RR: 0.38) [100]. As a primary screening tool, colonoscopy is recommended for average-risk adults beginning at age 45–50, with a screening interval of 10 years for individuals with normal findings. High-risk populations (e.g., those with a family history of CRC, genetic syndromes, or inflammatory bowel disease [IBD]) should initiate screening earlier (e.g., at age 40 or 10 years before the youngest affected relative) and undergo more frequent surveillance (e.g., every 1–5 years) [101, 102]. Following a positive non-invasive test, colonoscopy is essential as a confirmatory diagnostic procedure. Additionally, colonoscopy carries inherent risks of adverse events, including minor complications (such as abdominal pain, distension, and diarrhea) and serious events (e.g., perforation 0.05%, bleeding 0.3%−6.1%, and death estimated 0.007%−0.07%) [103, 104]. Nevertheless, due to its dual role in both prevention and early cancer detection, colonoscopy remains the primary screening option for high-risk groups and a validated alternative for average-risk individuals seeking definitive screening.
Radiographic examination
Computed tomography colonography
Computed tomography colonography (CTC) is a minimally invasive imaging technique used for CRC screening and diagnosis. It provides a detailed view of the entire colon by combining CT scanning with image reconstruction techniques. CTC is usually performed after the patient undergoes bowel preparation and colon dilatation, and uses 2D and 3D visualization techniques to examine the interior of the colon. According to a systematic evaluation and meta-analysis [105], the sensitivity of CTC in detecting CRC and adenomatous polyps was 88.8% and the specificity was 75.4%. It is worth noting that compared to colonoscopy, CTC has the advantage of higher patient compliance and can simultaneously examine extraintestinal lesions. However, its overall screening sensitivity and specificity for adenomas are relatively lower [106]. The limitations of CTC include the need for rigorous bowel preparation, high requirements for equipment and technical personnel, and a certain risk of radiation exposure. It is primarily suitable for patients who cannot undergo colonoscopy and is not the preferred screening method. Therefore, CTC is recommended as a feasible alternative to colonoscopy for specific clinical scenarios. Overall, CTC serves as an effective tool for CRC screening, opening new possibilities for early diagnosis and treatment.
Anzeige
Colonic capsule endoscopy
Compared with conventional colonoscopy, colonic capsule endoscopy (CCE) offers higher patient acceptance and lower risk of complications. According to a systematic evaluation and meta-analysis of studies [107], the accuracy of CCE in detecting colorectal tumors has been confirmed. For polyps > 6 mm in diameter, CCE had a sensitivity of 79%−96% and a specificity of 66%−97%. For polyps ≥ 10 mm in diameter, the sensitivity of CCE was 84%−97%, which was better than the results of CTC, and the CRC detection rate for completed CCE examinations was 93% (25/27). These results suggest that CCE is a safe and effective tool in CRC screening, with accuracy comparable to that of colonoscopy and even superior to CTC. In addition, the noninvasive nature of CCE makes it a promising tool for periodic screening, especially for high-risk populations who cannot tolerate conventional colonoscopy. However, there are some limitations of CCE in CRC screening, including inconsistency in completion rates, which depends on the booster used and may lead to compromised consistency and universality of screening; high requirements for bowel preparation, where inadequate bowel preparation may mask mucosal lesions and affect the results; the risk of capsule retention, especially in diseases such as Crohn’s disease or neoplasms that cause narrowing of the bowel; the finding of a lesions cannot be biopsied after detection, limiting its diagnostic capabilities; cost-effectiveness issues that make CCE not recommended for CRC population screening at this time; and technical limitations, including lower diagnostic sensitivity for CRC and the risk of operational failures and adverse events. These shortcomings suggest that although CCE offers a painless screening option, its technical challenges and cost-effectiveness need to be considered in practical applications to determine its applicability in CRC screening.
Clinical screening guidelines and implementation strategies
Current clinical guidelines stratify screening strategies based on individualized risk profiles. For average-risk populations (aged 45–75 years without familial or genetic predisposition), the U.S. Preventive Services Task Force (USPSTF) recommends: annual FIT, mt-sDNA every 3 years, colonoscopy every 10 years, or CTC every 5 years. Among these, FIT demonstrates optimal cost-effectiveness for large-scale population screening, particularly in resource-limited settings, while mt-sDNA and colonoscopy are prioritized in contexts where sensitivity outweighs cost considerations [102, 108]. For high-risk cohorts, such as individuals with first-degree relatives diagnosed with CRC/adenomas, IBD patients, or confirmed hereditary syndromes, colonoscopy remains the gold standard with intensified surveillance intervals (every 1–5 years). The American Cancer Society (ACS) specifically recommends initiating screening at age 40 or 10 years younger than the earliest familial diagnosis for hereditary risk groups, followed by biennial colonoscopies [108]. Such risk-adapted protocols aim to balance early detection efficacy against over-screening burdens, though challenges persist in adherence due to socioeconomic disparities and healthcare access limitations. Notably, emerging non-invasive biomarkers have not yet been formally incorporated into major guidelines pending large-scale validation. Future guideline updates should emphasize health economic evaluations and risk-customized algorithms to optimize precision and accessibility.
Discussion
This article reviews the research progress of CRC screening methods and systematically analyzes a series of tests ranging from traditional screening techniques to emerging molecular markers. As one of the malignant tumors with a high mortality rate worldwide, the key to the prevention and control of CRC lies in its early detection and treatment, and screening is an important means to achieve this goal. Although colonoscopy is regarded as the gold standard for diagnosis by virtue of its high accuracy, its invasive character limits its application in widespread screening. In contrast, non-invasive screening methods, such as FIT, are prized for their simplicity and high compliance, despite certain false-positive issues. In addition, Multi-targeted fecal DNA testing technology offers new possibilities for early detection of CRC by virtue of its high sensitivity and specificity, but the cost issue is still an obstacle to its wide application. Meanwhile, blood- and urine-based molecular Marker detection technologies, which open up new ways to realize non-invasive screening, are promising, although they are still in the research stage. Looking ahead, the development direction of CRC screening technology will be more diversified and precise. When evaluating the practical application value of these screening methods, cost, accessibility, and effectiveness are the core factors that need to be comprehensively considered. Currently, FIT remains the preferred method for population-based screening worldwide, especially in low- and middle-income countries and regions with limited resources, due to its low cost, ease of use, high accessibility, and relatively reasonable effectiveness. Colonoscopy, as the diagnostic gold standard, is highly invasive, operationally complex, and requires specialized resources and infrastructure. Its relatively high cost and potential risks significantly limit its accessibility as a large-scale initial screening tool. Emerging molecular Marker detection technologies show great potential. However, their high costs and relatively complex laboratory analysis requirements currently hinder their widespread adoption globally (Table 3). Quantitative assessment of the performance metrics of screening tests is of significant guidance for optimizing CRC screening strategies. Based on a CRC prevalence rate of 0.5% in the target population (i.e., 50 cases per 10,000 screened individuals), FIT with a sensitivity of 79% and specificity of 94% is projected to miss 10.5 cancer cases (false negatives) and generate 597 false positive results per 10,000 screened individuals, leading to unnecessary colonoscopies and associated healthcare burdens [11]; In contrast, mt-sDNA can reduce missed cases to 3.85 with its higher sensitivity (92.3%), but its lower specificity (86.6%) would significantly increase false-positive results to 1,333, thereby substantially increasing the demand for invasive examinations [19]; Plasma Septin9 gene methylation testing (sensitivity 48%, specificity 92%) would miss 26 cancer cases and produce 796 false-positive results at the same prevalence rate, further confirming its limitations as a screening tool [73]. False-negative results may lead to delayed clinical intervention, while false-positive results may cause psychological stress for patients and increase the risk of overtreatment. Therefore, optimizing screening strategies requires a comprehensive consideration of detection performance parameters, regional medical resource allocation, the prevalence characteristics of the target population, and risk stratification models to achieve cost-effectiveness maximization.
Advances in biomarker and Multi-omics technologies underscore the transformative potential of artificial intelligence and Machine learning in revolutionizing CRC screening analytics. Deep learning models, for instance, process endoscopic videos in real time, boosting polyp detection sensitivity above 90% and reducing small lesion miss rates [109]. Multi-center studies confirm AI-assisted colonoscopy lowers adenoma miss rates by 40% [110]. Furthermore, AI integrates multi-omics and clinical data to build high-precision non-invasive models (regression error < 0.0001), optimizing screening strategies [111]. This robust analytical capability efficiently identifies high-value biomarker combinations, enabling precise, personalized screening. And the development of individualized screening protocols that incorporate an individual’s genetic background, lifestyle habits, and gut microbiome characteristics will further improve the efficiency and effectiveness of screening. Strategies to improve screening adherence, such as the development of noninvasive, convenient, and user-friendly screening methods, are also important directions for future research. In addition, the advancement of interdisciplinary research will bring together research efforts in the fields of biology, medicine, computer science, and public health to promote the development of CRC screening technologies.
However, this study has methodological limitations that cannot be ignored. First, the performance evidence of emerging screening technologies is highly dependent on the experimental conditions designed in the original studies. Their sensitivity/specifi-city may fluctuate beyond clinically acceptable ranges due to non-standardized variables such as baseline risk in the population, sample collection protocols, and detection platform sensitivity thresholds. This systemic data gap makes it difficult to quantify the true boundaries of the technology’s sensitivity. These limitations highlight the fragility of the current evidence chain in this field, underscoring the urgent need for prospective, multicenter studies to establish population-stratified calibration models and standardized operational protocols.
Conclusion
This review underscores the critical role of CRC screening in reducing mortality through early detection. While colonoscopy remains the gold standard, its invasiveness and resource intensity limit widespread population application. Non-invasive methods offer promising alternatives with better compliance, though challenges remain regarding cost and sensitivity to precancerous lesions. Emerging blood- and urine-based biomarkers and microbiological features show potential for high specificity but require further validation. As technology advances and healthcare resources improve, future CRC screening will focus on enhancing efficiency, convenience, and cost-effectiveness. This evolution will prioritize individualization and precision, heavily leveraging artificial intelligence and big data analysis to optimize biomarkers and develop cost-effective, personalized screening strategies. The integration of these technologies aims to achieve earlier detection and treatment, ultimately reducing the global burden of CRC and improving patient survival rates and quality of life.
Declarations
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
