Colorectal cancer (CRC) is a severe disease that poses a substantial threat to human health. According to the 2024 global cancer data (GLOBCAN 2024) given by the International Agency for Research on Cancer (IARC), the incidence of CRC is listed as the third most frequent, with around 935,173 people dying from CRC each year globally, making it the second biggest cause of cancer-related fatalities (1–3). The 5-year survival rate for CRC decreases significantly from 95% at stage I to just 6% at stage IV, indicating a strong correlation between the stage at diagnosis and the death rate. Early detection of CRC is critical to improving survival because it reduces the chance of death, stops the disease from spreading, and enhances the quality of life and long-term health outcomes (4). While fecal occult blood tests and colonoscopies are standard diagnostic tools, their invasive nature and the discomfort they cause often lead to patient aversion and subsequent diagnostic delays (5, 6). The search for dependable biomarkers has resulted in the use of blood-based markers like carcinoembryonic antigen (CEA) and carbohydrate antigens (CA 19–9 and CA 242) (7), and alpha-fetoprotein (AFP) (8); however, their diagnostic efficacy is marred by low sensitivity and specificity, particularly in benign conditions. This limitation necessitates the exploration of novel biomarkers that can offer more precise diagnostic and prognostic insights for high-risk CRC patients (9–12). Thus, searching for new biomarkers in CRC diagnosis and prognosis has attracted many researchers. There is considerable interest in non-invasive liquid biopsy (LB) as a supplementary and maybe alternative approach, attempting to address these limitations (13–16). LB is a process that extracts tumor-derived biomarkers from body fluids such as blood, urine, saliva, feces, or cerebrospinal fluid. This approach may identify a number of circulating tumor components, including circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), messenger RNA (mRNA), non-coding RNA, extracellular vesicles (EVs), and tumor-educated platelets (TEPs). LB is essentially a non-invasive diagnostic tool that aids in the early diagnosis and monitoring of cancer (17, 18).

Recent studies emphasize circulating cell-free DNA (cfDNA) analysis as a possible biomarker for many therapeutic applications in cancer, such as diagnosis, evaluation of tumor development, assessment of therapy response, and prognosis (19–21). cfDNA includes DNA that enters the bloodstream from the breakdown of cells or is actively secreted by normal and cancerous tissues. It contains important genetic details about the primary tumor’s structure. When collecting tumor biopsy is difficult, analyzing cfDNA in Blood can serve as an alternative for histopathological examination (22, 23). Furthermore, the amount of cfDNA is typically changed in malignancies, even in the early stages (24). Higher levels of blood cfDNA are reported to have a strong association with malignancy and poorer clinical outcomes. Over the last two decades, cfDNA analysis has become a promising technique for cancer diagnosis and prognosis (25, 26).

Arthrobacter luteus (ALU) fragments are primate-specific repeats that make up approximately 11% of the human genome (27). The ALU fragments are short interspersed nuclear elements (SINEs) that are common in the human genome. They are transposable elements that may be affected by environmental factors (28) and ALUs have wide-ranging influences on gene expression, affecting polyadenylation, splicing, and ADAR editing (27). They significantly influence the composition of cfDNA (26). These fragments are widely utilized as target molecules in LB applications due to their abundant presence throughout the human genome. Analyses focused on Alu-based cfDNA have demonstrated promising application potential in several key areas of cancer management (29): Numerous studies have employed major repetitive regions within the human genome, specifically ALU115 and ALU247, for quantifying cell-free DNA (cfDNA). ALU115 reflects total DNA content, encompassing both longer and shorter cfDNA fragments. In contrast, ALU247 specifically identifies tumor-derived DNA, focusing on longer cfDNA fragments associated with malignancy (30, 31). ALU115/247 has been intensively researched as a promising non-invasive diagnostic and prognostic marker in the area of LB, with a wide variety of quantitative and qualitative results (32, 33).

Thus, searching for new biomarkers in CRC diagnosis and prognosis has attracted many researchers. The present study aimed to carry out the first meta-analysis to quantitatively analyze the diagnostic and prognostic accuracy of ALU-based cfDNA in CRC patients.

This report has been organized following the PRISMA 2020 framework, as depicted in Figure 1 (34), additionally, we have incorporated the PRISMA 2020 Checklist (35) in Supplementary Table 1 to ensure comprehensive reporting and methodological transparency.

In recent years, LB has become increasingly common for cancer diagnosis. To determine the potential diagnostic and prognostic impact of ALU-based cfDNA, 17 papers were analyzed to identify ALU-based cfDNA sequences as a cancer biomarker. In this exploratory meta-analysis of 13 trials using 14 sets of data, the pooled analysis had a diagnostic sensitivity of 0.81 and a specificity of 0.90. These two typical characteristics verified the accuracy of ALU-based cfDNA as a viable noninvasive predictor for CRC detection.

The advent of Next-Generation Sequencing (NGS) has revolutionized the landscape of genomic analysis, offering a comprehensive tool that surpasses the capabilities of traditional qPCR systems, particularly in the context of ALU sequence coverage. NGS provides a hypothesis-free approach that does not require prior knowledge of sequence information, enabling the discovery of novel genes and the quantification of rare variants and transcripts with higher sensitivity (59). NGS has transformed the LB field for CRC through its comprehensive analysis capabilities. The ability of NGS to extensively map ALU sequences within cfDNA is vital for uncovering an array of biomarkers, offering a deeper understanding of CRC’s genetic intricacies. This enhanced level of detail is fundamental for the precise diagnosis and the prognostic evaluation of CRC. By encompassing a broader spectrum of ALU sequences, NGS ensures a more accurate and dependable measure of cfDNA’s biomarker potential, which may significantly improve the clinical management and outcomes for CRC patients (33, 60).

A recent meta-analysis on the use of circulating cfDNA as a diagnostic marker for CRC indicated that while the sensitivity of circulating cfDNA is unsatisfactory, its specificity remains acceptable for CRC diagnosis (61). Furthermore, another systematic review revealed elevated concentrations of ALU repetitive elements in cancer patients, whereas these concentrations were reduced in control groups, benign conditions, various cancer stages, and other diseases. The total ALU (115 and 247) sequence levels emerge as potential biomarkers for investigative purposes and cancer prognosis (32).

Our study’s pooled DOR of 40.58 indicated a good overall accuracy. Furthermore, the PLR was found to be 8.40, whereas the NLR was discovered to be 0.21. These data show that the meta-analysis’s good likelihood ratios may indicate the results’ robustness and accuracy. Furthermore, ROC is often used to define overall test performance, and AUC acts as a measuring indication; the AUC of SROC for ALU-based cfDNA was 0.90, showing that ALU-based cfDNA has a reasonably good accuracy for CRC detection. Heterogeneity is an important issue in meta-analysis. The I² test revealed high heterogeneity in our study’s trials. The threshold effect is typically the predominant source of heterogeneity in diagnostic meta-analyses. To evaluate the likely cause of variability, we used meta-regression to examine the features of included studies, including publication year, research location, kind of specimens, detection techniques, measuring objects, number of cases, and four major domains in QUADAS-2.

Finally, our analysis revealed that study quality played a significant role in the substantial heterogeneity, implying that the study design with high-risk biases of “Patient selection,” “Index Test,” “Reference Standard,” and “Flow and Timing” may be more likely than other characteristics to affect diagnostic accuracy. Heterogeneity may have increased due to other factors such as age, tumor type, metastasis, TNM staging, surgery method, and treatment regimen, which were not investigated in the current investigation due to a lack of data availability. Although publication bias can be a concern in meta-analyses, Deeks’ funnel plot asymmetry test found no such bias, indicating that our meta-analysis results for diagnosis and prognosis are credible. Furthermore, blood-based (plasma or serum) tests showed considerably better overall diagnostic accuracy and were created as a non-invasive diagnostic biomarker for CRC. On the other hand, the meta-analyses revealed that ALU-based cfDNA was a potential biomarker for predicting OS in CRC patients. In our prognostic analysis, which comprised four trials and six data sets, the pooled results showed that patients with low levels of ALU-based cfDNA had a 2.33 (HR = 2.33) times reduced chance of poor OS. Furthermore, there was substantial variability across the prognostic meta-analyses. A significant range in sample types was seen throughout the studies that contributed to the predictive results. However, several issues still limit its implementation in clinical practice. There is no consensus on the utility of cfDNA testing in cancer patients. We remain skeptical of any future suggestions for ALU-based cfDNA. A future study will assist in determining this. CEA and CA19–9 are commonly utilized clinical indicators for the diagnosis and prognosis of CRC. Higher CEA concentrations occur in only 5%-40% of CRC patients, while positive findings are frequently reported in cancer-free patients who suffer from benign conditions such as liver injury or inflammatory disorders (53, 62). As a result, we hope and aim to demonstrate that ALU-based cfDNA can be a therapeutically valuable surrogate marker.

Our study has several important strengths. First, this is the first study to comprehensively examine the association between the amount of ALU-based cfDNA and its importance in diagnosing and predicting outcomes in CRC patients. Both diagnostic and prognostic meta-analyses were performed based on a sufficient number of publications. Second, our meta-analysis has uncovered interesting results, paving the way for future research. However, there were also several limitations in our work. Firstly, all publications regarding the prognosis of ALU-based cfDNA have exclusively focused on evaluating OS rather than other endpoints such as DFS, PFS, and CSS. In addition, This meta-analysis was limited to the evaluation of univariate OS due to the available studies that could be included in the analysis and highlight the necessity for further investigations into the prognostic impact of ALU-based cfDNA using other time-to-event (TTE) end-points. Furthermore, little research has been conducted to determine the prognostic value of ALU-based cfDNA in CRC. Second, our research relied on data from published studies rather than individual patient data (IPD), which limited the ability to conduct a uniform analysis of all data. So, Obtaining raw data is crucial for survival prediction and analysis. Moreover, we did not extend the search to non-English publications, which may have resulted in bias because good results are more easily accepted by English-language journals. Even though progress has been made in spotting and predicting CRC, we still urgently need better markers to improve early detection and predict patient outcomes more accurately (63). Future research directions for ALU-based cfDNA in CRC include validation across diverse populations, longitudinal studies, integration with other biomarkers, standardization of detection techniques, clinical utility assessment, exploration of tumor-specific ALU-based cfDNA, prospective cohort studies, patient acceptance evaluation, cost-effectiveness analysis, and collaboration for consensus guidelines. These efforts will refine ALU-based cfDNA’s clinical applicability and address limitations.

The findings of this systematic review and meta-analysis have important implications for the clinical management of CRC. The results indicate that serum ALU-based cell-free DNA (cfDNA) is a promising biomarker that can aid in the diagnosis and prognosis of CRC. This non-invasive test could potentially be incorporated into routine clinical practice to supplement existing diagnostic tools and provide valuable prognostic information to guide treatment decisions.

The next steps should involve further large-scale, high-quality studies to validate the diagnostic and prognostic performance of ALU-based cfDNA testing in diverse patient populations. Additionally, research is needed to optimize the analytical methods and establish standardized protocols for the measurement of ALU-based cfDNA. Once these steps are taken, ALU-based cfDNA testing could become a valuable addition to the clinical armamentarium for managing patients with CRC, potentially leading to earlier diagnosis, more personalized treatment, and improved patient outcomes.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

MT: Data curation, Investigation, Writing – original draft, Writing – review & editing. LA: Conceptualization, Data curation, Project administration, Supervision, Validation, Visualization, Writing – review & editing. SM: Conceptualization, Data curation, Project administration, Supervision, Validation, Visualization, Writing – review & editing. AK: Conceptualization, Data curation, Project administration, Supervision, Validation, Visualization, Writing – review & editing. TJ: Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Software, Supervision, Validation, Visualization, Writing – review & editing.