Advancing gastrointestinal cancer diagnosis and treatment: a narrative review of circulating tumor DNA in gastrointestinal malignancies

Introduction

Gastrointestinal (GI) cancers encompass a diverse group of malignancies that collectively account for a significant portion of the global cancer burden (1-3). These cancers are frequently diagnosed at advanced stages, limiting treatment options and contributing to high mortality rates (4-6). Despite advances in diagnostic and therapeutic modalities, many GI cancers are detected late, resulting in limited treatment options and poor prognoses (4,5). Current diagnostic methods, such as endoscopy and imaging, are often invasive, expensive, and suboptimal for early-stage disease detection or identifying minimal residual disease (MRD) following treatment (5).

The magnitude of this problem is evident when looking at individual cancer sites. Esophageal cancer, though relatively uncommon, remains highly lethal. In the USA, it accounts for just over 1% of new cancer diagnoses but is expected to cause more than 16,000 deaths in 2025. Five-year survival is less than 25%, mainly because most patients present with advanced disease. Symptoms usually prompt diagnostic evaluation, and while tools like barium swallow can suggest abnormalities, confirmation still requires invasive endoscopy with biopsy (3). Unlike colorectal cancer (CRC), there are no established screening programs for average-risk individuals, limiting opportunities for early detection.

Gastric cancer tells a similar story. More than 30,000 Americans will be diagnosed in 2025, and survival remains bleak once the disease has spread. Although outcomes can be favorable when diagnosed early, only about a third of patients are diagnosed at a localized stage (3). Cancers of the small intestine, while rare, add to this burden. They are often unpredictable in presentation, and survival varies widely by stage at diagnosis, highlighting the ongoing challenge of timely detection.

Pancreatic cancer stands out as one of the deadliest GI malignancies (3). Despite accounting for only a roughly 3% of new cancer diagnoses, pancreatic cancer is among the leading causes of cancer-related deaths, responsible for nearly 8% of all cancer-related deaths (3,7). The 5-year survival rate is just over 13%, and more than half of patients present with distant disease. Unlike CRC, no reliable screening test exists, and the pancreas is anatomically difficult to assess. Even with modern imaging such as computed tomography (CT), magnetic resonance imaging (MRI), or endoscopic ultrasound, early detection remains elusive (3,7).

Hepatic and intrahepatic biliary cancers also contribute disproportionately to mortality. In 2025, Surveillance, Epidemiology, and End Results (SEER) projected hepatic and intrahepatic biliary cancers to be the cause of more than 30,000 deaths in the USA, with overall survival (OS) around 22% (3). While nearly half of cases are diagnosed at a localized stage, outcomes decline sharply with more advanced disease.

Finally, CRC remains one of the most common and preventable GI cancers. More than 150,000 new cases and 52,000 deaths are projected in 2025, yet survival varies dramatically depending on stage (3). Colonoscopy continues to serve as the gold standard for screening, offering both detection and polyp removal, but its invasiveness, preparation requirements, and small procedural risks limit its uptake. Other approaches, such as flexible sigmoidoscopy, CT colonography, and stool-based tests, offer alternatives but come with trade-offs—ranging from restricted visualization to the need for follow-up colonoscopy when results are abnormal (3). These strategies have undoubtedly saved lives, but they remain imperfect.

Many cancers still go undetected until advanced stages, and current methods lack the sensitivity to identify MRD or predict relapse with precision. Circulating tumor DNA (ctDNA) can provide real-time insight into tumor biology through a simple blood test, potentially bridging this gap. These advancements have opened new avenues for ctDNA in cancer diagnosis, treatment selection, monitoring, and prognosis. However, while ctDNA has shown significant promise in CRC, its clinical utility in non-colorectal GI cancers remain underexplored. There is a need to consolidate and evaluate the available evidence on ctDNA’s role in GI cancers, including gastric cancer, hepatocellular carcinoma (HCC), pancreatic ductal adenocarcinoma (PDAC), CRC and esophageal cancer.

Accordingly, the objective of this review is to evaluate the clinical utility of ctDNA in GI cancers, including gastric cancer, HCC, PDAC, and esophageal cancer. By focusing on ctDNA’s role in early detection, treatment monitoring, MRD assessment, and prognosis, this review seeks to highlight ctDNA’s potential to transform cancer management. In addition, it will address existing challenges and propose future research directions to optimize ctDNA-based strategies for improving patient outcomes. The clinical applications of ctDNA across specific GI cancers are summarized in Tables S1-S4. We present this article in accordance with the Narrative Review reporting checklist (available at https://tgh.amegroups.com/article/view/10.21037/tgh-25-28/rc).

Methods

In this review, we analyze findings from recent studies to evaluate the clinical utility of ctDNA across GI malignancies. We searched PubMed/MEDLINE, Google Scholar, ClinicalTrials.gov, and the Cochrane Library between October 1, 2024, and August 5, 2025. Eligible studies were restricted to English-language publications, with no geographic limitations applied. The search strategy incorporated terms related to “circulating tumor DNA”, “ctDNA”, “liquid biopsy”, and GI malignancies (e.g., colorectal, pancreatic, hepatocellular, and esophageal cancers) and so on. The overall search strategy is summarized in Table 1, with the detailed PubMed search strategy in Table 2. The studies identified through this process were synthesized and are presented in Tables S1-S4, which outline the clinical applications of ctDNA across GI cancers.

Clinical applications of ctDNA in GI malignancies

ctDNA, a component of cell-free DNA (cfDNA), has emerged as a promising biomarker in oncology. This “liquid biopsy” approach provides real-time insights into tumor biology, enabling early detection, treatment response monitoring, and identification of actionable mutations (3-5). ctDNA refers to small fragments of DNA released into the bloodstream by tumor cells, carrying tumor-specific mutations and epigenetic alterations (3-5). Its potential to guide personalized therapy and predict relapse has positioned ctDNA as a critical tool in modern oncology (3). Advances in sequencing technologies, such as next-generation sequencing (NGS) and digital droplet polymerase chain reaction (ddPCR), have enabled the sensitive and specific detection of ctDNA, even at low concentrations (3).

Tumor-informed assays, such as Signatera, further enhance this approach by utilizing a patient’s tumor tissue to identify specific mutations, creating a personalized “tumor fingerprint” (3). These assays detect ctDNA carrying the same mutations, providing a highly sensitive measure of MRD and enabling early detection of recurrence months to years before clinical relapse, with demonstrated sensitivities of >95% and specificities of 99.7% (3). In parallel, tumor-naïve assays detect ctDNA without prior tumor tissue analysis, screening for a panel of common cancer-related mutations and epigenetic alterations (3). Techniques like CAncer Personalized Profiling by deep Sequencing (CAPP-Seq) and cell-free Methylated DNA ImmunoPrecipitation and high-throughput sequencing (cfMeDIP-seq) analyze somatic mutations and DNA methylation, enabling sensitive ctDNA detection even at low levels (3). Tumor-naïve assays demonstrate strong correlations between mutation- and methylation-based ctDNA detection in a range of malignancies, including head and neck squamous cell carcinoma (HNSCC). In HNSCC, pretreatment ctDNA positivity has been significantly associated with inferior overall survival (OS) [hazard ratio (HR) =7.5; P=0.03], with analogous prognostic implications observed in other cancer types (3). Additionally, lack of ctDNA clearance post-treatment strongly correlates with disease recurrence (3). Tumor-naïve assays are particularly valuable in low ctDNA tumor-informed methods. By providing information on tumor burden and potentially targetable mutations, these assays also aid in biomarker discovery for early-stage cancers (3). These applications, including evidence from GI cancers where ctDNA has been evaluated for prognostication and disease monitoring, are organized in Tables S1-S4.

Esophageal cancer

ctDNA has demonstrated promise in the management of esophageal cancer, serving as a non-invasive biomarker for diagnosis, treatment selection, disease monitoring, and prognosis (3,6). For diagnosis, ctDNA enables the detection of tumor-specific mutations, such as TP53 and HER2, which can identify esophageal cancer earlier than traditional imaging methods (3). This is particularly valuable for patients who may not be candidates for invasive procedures like endoscopy (3). Studies evaluating ctDNA as a biomarker for treatment response in esophageal cancer are outlined in Table S1. By the time esophageal cancer is typically diagnosed, multi-modal therapy with chemotherapy, radiation, and surgery is required for cure. Early detection through ctDNA could improve patient outcomes and quality of life by enabling early intervention and de-escalated therapy (3,7).

Standard treatment for early-stage esophageal cancers involves a neoadjuvant approach with chemotherapy or chemoradiotherapy. This approach is favored with evidence for improved pathologic complete response (pCR) rates and OS compared with surgery alone. Monitoring response to therapy with conventional imaging has been challenging, highlighting the potential benefit of using ctDNA. In the neoadjuvant setting, ctDNA can provide oncologists with real-time insights into treatment response and residual disease, helping to guide decisions about adjuvant therapy or surgical intervention. For example, in a meta-analysis by Zhang et al., ctDNA positivity after neoadjuvant treatment correlated with poorer outcomes (3). Key studies on ctDNA for monitoring response to neoadjuvant therapy are outlined in Table S1.

Longitudinal monitoring of ctDNA can offer critical insights into MRD and early relapse (3). For example, Shaker et al. demonstrated that patients with detectable ctDNA following surgery had significantly worse recurrence-free survival (RFS) compared to those with undetectable ctDNA (3). Furthermore, ctDNA serves as an early indicator of relapse, with serial sampling demonstrating its ability to predict recurrence after surgery (3). Shaker et al. also demonstrated that ctDNA levels decrease significantly following surgical resection (3). These findings highlight ctDNA’s potential as a robust prognostic tool for risk stratification and outcome prediction in esophageal cancer. This illustrates that longitudinal ctDNA monitoring has potential prognostic value in esophageal cancer, with elevated levels being strongly correlated with higher tumor burden and more aggressive disease (3,8,9). Representative studies outlining the role of ctDNA in disease monitoring, including its prognostic value in esophageal cancer, are presented in Table S1.

In the advanced setting, ctDNA can identify targetable mutations with treatment implications, including HER2 amplification, to guide therapies like trastuzumab (3). Additionally, ctDNA can identify resistance mutations, such as PIK3CA mutation, allowing clinicians to potentially anticipate or adjust treatment strategies before clinical relapse occurs (3,6). ctDNA can assist with disease monitoring, with changes in ctDNA levels correlating with the effectiveness of chemotherapy or targeted therapies (10). Persistently elevated ctDNA levels may suggest treatment resistance or disease progression, prompting timely adjustments to therapeutic strategies (6). Studies illustrating ctDNA’s role in guiding treatment selection and monitoring resistance in advanced esophageal cancer are included in Table S1. This dynamic monitoring facilitates a more personalized approach, improving the likelihood of successful outcomes (3,11).

Hepatobiliary cancers

ctDNA has also shown promise as a non-invasive biomarker in the management of hepatobiliary cancers, including HCC, cholangiocarcinoma, and gallbladder cancer (12). ctDNA enables the detection of tumor-specific mutations and genomic alterations, such as TP53, IDH1/2, and FGFR2 fusions, which are frequently observed in hepatobiliary cancers (13). Studies have demonstrated that ctDNA can detect mutations associated with cholangiocarcinoma and HCC months before clinical symptoms or imaging findings become apparent (13). This early detection capability is critical for biliary cancers, which are often diagnosed at advanced stages due to their asymptomatic nature (13). Representative studies supporting this early detection role of ctDNA in biliary cancers is summarized in Table S2.

Additional benefits of ctDNA for this disease include targeted treatment identification and response monitoring assessment. FGFR fusions detected in ctDNA can inform the use of FGFR inhibitors in cholangiocarcinoma, while mutations in NTRK and RET may guide therapeutic strategies in HCC (12,13). Examples of ctDNA-driven approaches to therapy selection and monitoring in hepatobiliary cancers are summarized in Table S2.

ctDNA can also identify resistance mutations, enabling clinicians to adjust treatment regimens before clinical relapse occurs (12). The emergence of EGFR or MET amplifications in ctDNA has been associated with resistance to tyrosine kinase inhibitors (TKIs) in advanced HCC, prompting increased monitoring to detect recurrences early and facilitate timely switches to alternative therapies. Rising ctDNA levels during treatment can serve as an early warning of therapeutic failure and as an early indicator of relapse, with serial sampling demonstrating its ability to predict recurrence after surgical resection or liver transplantation (4). In cholangiocarcinoma, ctDNA has been used to detect MRD and predict recurrence months before radiographic evidence of disease progression (13). Studies supporting the use of ctDNA for resistance profiling, recurrence surveillance, and MRD detection are listed in Table S2.

Pancreatic cancers

Pancreatic cancer remains a highly aggressive cancer with poor survival rates, largely due to late diagnosis and early metastatic spread (14,15). The application of ctDNA as a non-invasive tool has made significant advancements in improving patient outcomes (16). ctDNA has demonstrated significant potential in detecting tumor-specific mutations, such as those in KRAS, TP53, and CDKN2A, which are frequently observed in pancreatic cancer (14,17). For instance, in the CodeBreaK 100 trial, KRAS mutations were detected in ctDNA in 48% of cases, even when tumors were not yet visible on imaging (18). Representative studies evaluating ctDNA for early detection and detection of tumor-specific mutations in pancreatic cancer are highlighted in Table S3.

Additionally, ctDNA can identify high risk patients who may benefit from intensified treatment regimens (16). For example, Botta et al. reported that patients with detectable ctDNA had a median survival of 13.6 months compared to 27.6 months for those without detectable ctDNA (19). However, the specific timepoint of ctDNA detection (e.g., perioperatively within 2–12 weeks after surgery or during surveillance >12 weeks post-surgery) was not detailed in this study (14,19). Similarly, Labori et al. found that ctDNA detection in unresectable pancreatic ductal adenocarcinoma (uPA) was associated with significantly shorter OS and progression-free survival (PFS) (14).

Recent findings from the KRASCIPANC study highlight the prognostic utility of ctDNA kinetics in uPA (14,15). This study demonstrated that elevated baseline levels of cfDNA and KRAS-mutated ctDNA, as well as the persistence of KRAS-mutated ctDNA at day 28 following chemotherapy initiation, were strongly associated with reduced centralized disease control rates, shorter PFS, and poorer OS in multivariate analyses (14,15,17). A predictive score combining cfDNA levels at diagnosis (≥30 or <30 ng/mL) and the presence or absence of KRAS-mutated ctDNA at day 28 was identified as an optimal predictor of centralized disease control rate (cDCR) [odds ratio (OR) =30.7, 95% confidence interval (CI): 4.31–218, P=0.001], PFS (HR =6.79, 95% CI: 2.76–16.7, P<0.001), and OS (HR =9.98, 95% CI: 4.14–24.1, P<0.001). These findings underscore the potential of ctDNA kinetics as a robust biomarker for predicting therapeutic response and survival outcomes in uPA (14,15,17). Analyses linking ctDNA kinetics with therapeutic response and survival outcomes in uPA are detailed in Table S3.

Furthermore, the ARTEMIS-PC study provides additional evidence of the critical role of ctDNA, particularly variant allele frequency (VAF), in predicting clinical outcomes. This study demonstrated that ctDNA clearance, especially in relation to VAF, is a strong predictor of improved objective response, disease control, and PFS in patients with unresectable pancreatic cancer. VAF exhibited superior predictive performance for disease control [area under the curve (AUC): 0.84 at enrollment, 0.97 at weeks 4 and 8], outperforming traditional biomarkers like carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA 19-9). These findings suggest that VAF could be a valuable marker in the personalized management of pancreatic cancer, complementing existing predictive biomarkers and potentially guiding therapeutic decisions.

The detection of ctDNA in the portal or peripheral vein, especially after tissue mobilization during surgery has been associated with worse RFS and OS (20). A recent study from France investigating ctDNA kinetics in portal and peripheral blood before and after resection found that intraoperative ctDNA detection, particularly following tissue mobilization, was correlated with significantly worse RFS (HR =3.26, 95% CI: 1.26–8.45; P=0.01) and OS (HR =5.46, 95% CI: 1.65–18.01, P=0.002) (20). Additionally, although portal vein sampling did not improve ctDNA detection, tissue mobilization was shown to increase ctDNA levels by 2.5-fold (P=0.03) in peripheral blood (20). This further highlights the role of ctDNA as a potential marker for monitoring disease progression, even though its detection after surgery is associated with poorer prognosis.

Beyond prognosis, ctDNA can help identify the subset of patients with pancreatic cancer who may benefit from targeted therapy (15). For example, KRAS mutations, present in over 90% of pancreatic cancers, can guide the use of targeted therapies such as KRAS G12C inhibitors (15,21). In a clinical trial involving patients with advanced pancreatic cancer, KRAS G12C mutations identified in ctDNA led to the initiation of sotorasib, resulting in a median PFS of 6.7 months (21). Additionally, ctDNA can identify emerging resistance mutations, allowing clinicians to potentially adjust treatment regimens before clinical progression occurs. For instance, in a study monitoring ctDNA during chemotherapy, the emergence of KRAS pathway mutations in ctDNA was associated with treatment resistance, prompting a switch to alternative therapies and improving outcomes (14,15).

For monitoring, ctDNA offers continuous, dynamic insights into treatment response and disease progression. In a study of 61 pancreatic cancer patients undergoing first-line chemotherapy, molecular assessment using ctDNA was performed to evaluate changes in KRAS-mutated ctDNA before and after treatment (14,15,17). The study classified ctDNA changes into five categories: molecular negative (mNT), molecular complete response (mCR), molecular partial response (mPR), molecular stable disease (mSD), and molecular progressive disease (mPD). Patients with progressive disease after initial CT imaging showed significantly worse therapeutic outcomes compared to 47 patients with disease control. Within the disease control group, patients with detectable ctDNA (mPD + mSD) had significantly shorter median survival (13.2 months) compared to those without detectable ctDNA (mCR + mNT; 21.7 months, P=0.003). Multivariate analysis revealed that the presence of ctDNA was the only independent prognostic factor (P=0.04), suggesting its utility as a sensitive marker for monitoring treatment response and predicting outcomes (22). Studies assessing ctDNA for monitoring treatment response and disease progression in pancreatic cancer are outlined in Table S3.

Colorectal cancers

The role of ctDNA in managing patients with CRC is well established, with ongoing clinical trials further elucidating its potential. ctDNA has become an indispensable tool for guiding personalized therapy and improving patient outcomes (2,23).

ctDNA can assist in CRC diagnosis and disease monitoring by aiding in the detection of tumor-specific mutations, such as those in APC, KRAS, and TP53, which are frequently observed in CRC and can be detected earlier than a tumor may be seen on colonoscopy, an invasive procedure that can be onerous on patients (2,6). Representative studies of ctDNA in CRC diagnosis and MRD detection are highlighted in Table S4. The BESPOKE CRC study validated the utility of ctDNA in identifying MRD, showing that 15.6% of patients were ctDNA-positive at the post-operative MRD time point, with significantly worse disease-free survival (DFS) compared to ctDNA-negative patients (HR =20.8, 95% CI: 10.0–43.4, P<0.001) (24). Similarly, the GALAXY study reinforced the prognostic value of ctDNA in CRC, demonstrating that ctDNA positivity during the MRD window was strongly associated with inferior DFS (HR: 15.75, 95% CI: 12.59–19.68, P<0.001) (25). Notably, a landmark analysis at the 3-month time point revealed that patients who achieved sustained ctDNA clearance had significantly better outcomes compared to those with transient clearance (HR: 32.57, 95% CI: 9.94–106.76, P<0.001) (25). Key studies supporting ctDNA-guided treatment selection and resistance monitoring in CRC are illustrated in Table S4. Additionally, a retrospective analysis of 120 patients with HCC demonstrated that ctDNA positivity within the MRD window (up to 12 weeks post-resection or liver transplantation) was strongly prognostic for recurrence, with a hazard ratio of 12 (95% CI: 3.7–38, P<0.001) (26). These findings underscore the ability of ctDNA to identify patients who may benefit from intensified adjuvant treatment, further supporting its role in early diagnosis and risk stratification.

In CRC, mutations in KRAS, NRAS, and BRAF detected in ctDNA can guide the use of targeted therapies, such as EGFR inhibitors or BRAF/MEK inhibitors (25). Additionally, ctDNA can identify emerging resistance mutations, such as KRAS G12C or EGFR extracellular domain mutations, allowing clinicians to adjust treatment regimens before clinical progression occurs (25). In the BESPOKE CRC study, ctDNA-guided adjuvant therapy significantly improved outcomes for MRD-positive patients, with those receiving adjuvant chemotherapy (ACT) showing a median DFS of 18.7 months compared to 6.7 months in the observation group (HR =3.9, 95% CI: 1.3–11.5, P=0.01). In contrast, no benefit of ACT was observed in MRD-negative patients (HR =1.1, 95% CI: 0.3–3.9, P=0.89), reinforcing the importance of ctDNA in tailoring treatment decisions (24). This approach not only spared low-risk patients from unnecessary treatment but also ensured that high-risk patients received timely and appropriate therapy.

For monitoring, ctDNA serves as a dynamic biomarker, enabling real-time evaluation of treatment response and tracking disease progression (27,28). Persistently elevated ctDNA levels may suggest treatment resistance or disease progression, prompting timely adjustments to therapeutic strategies (29,30). In BESPOKE CRC, patients achieving clearance at 12 weeks post-surgery had longer DFS compared to those who remained ctDNA⁺ (median DFS: 24.2 vs. 13.8 months; HR =0.4, 95% CI: 0.1–1.0, P=0.045). However, even patients with ctDNA clearance had worse DFS than those who were ctDNA⁻ at both time points (HR =22.5, 95% CI: 6.8–75.0, P<0.001), underscoring the importance of sustained ctDNA negativity (24). Findings on ctDNA as a tool for disease monitoring treatment response, treatment response assessment and disease progression are shown in Table S4.

ctDNA also has significant prognostic value in CRC, with elevated levels strongly correlated with worse outcomes and more aggressive disease (31,32). Higher baseline ctDNA levels are predictive of reduced OS, while longitudinal monitoring provides critical insights into MRD and early relapse (31,32). For instance, in the study by Zhou et al. and O’Sullivan et al., postoperative ctDNA detection was strongly associated with worse RFS (HR =13.0; P<0.001) (31,32). Patients with detectable ctDNA following surgery or systemic therapy had significantly worse RFS compared to those with undetectable ctDNA (31,32). These findings underscore ctDNA as a robust prognostic tool for risk stratification and outcome prediction in CRC (9).

Challenges and considerations

Despite the promising role of ctDNA in the management of non-colorectal GI malignancies (NCGIs), several challenges persist that must be addressed to enable its broader clinical application (3,4,13).

One of the primary practical challenges in ctDNA analysis is its technical complexity and time-sensitive nature (13). The short half-life of cfDNA, approximately 16 minutes, poses significant scalability issues, as blood samples must be processed within 6 hours or stored in specialized stabilization tubes to prevent degradation or contamination (33). This rapid degradation is critical to address, as delays in processing can lead to leukocyte lysis, diluting the ctDNA fraction and compromising the accuracy of results (3,4,33).

The sensitivity and specificity of ctDNA detection are further influenced by the choice of sequencing methods and the inherent biological variability of tumors (3,4). NGS offers high sensitivity for detecting a broad range of mutations but is costly and requires complex bioinformatic analysis (4). In contrast, ddPCR is highly specific but limited to detecting a predefined set of mutations, making it less suitable for comprehensive genomic profiling (4). Tumor heterogeneity adds another layer of complexity, as different tumor clones shed varying amounts of ctDNA, potentially leading to false negatives or incomplete molecular profiles (3,4). Additionally, patient-specific factors, such as liver or kidney dysfunction, can alter ctDNA clearance rates, further impacting detection sensitivity (3,4).

Standardization of pre-analytical and analytical protocols is another critical challenge that must be addressed to ensure the accuracy and reproducibility of ctDNA testing (3,4). Currently, there is no consensus on key parameters, such as the optimal timing for sample collection, the required plasma volume, or the preferred sequencing methods (4). For example, in PDAC, ctDNA detection rates vary significantly depending on whether samples are collected before or after surgery or systemic therapy (19). Studies addressing these differences are summarized in Table S2. Similarly, in esophageal cancer, the lack of standardized protocols has resulted in wide variability in sensitivity and specificity across studies, limiting the comparability of results (34,35). Relevant data on ctDNA monitoring in this setting are compiled in Table S2.

Cost is another significant barrier to the widespread adoption of ctDNA testing in clinical practice (3,4). While ctDNA testing is generally more cost-effective than traditional tissue biopsies, the initial setup costs for advanced sequencing technologies can be prohibitive (4). Personalized tumor-informed ctDNA assays, particularly in PDAC, are especially expensive due to the need for whole-exome sequencing and germline controls (19). In esophageal cancer, the cost of ctDNA testing varies depending on the sequencing platform, with more sensitive methods like NGS being significantly more expensive than targeted PCR-based approaches (34). Future economic studies comparing ctDNA with other diagnostic and monitoring methods will be essential to determine its feasibility for routine clinical use. Beyond cost considerations in high resource settings, the global rollout of ctDNA technologies will likely be delayed in low and middle-income countries, where cancer incidence is rising sharply but infrastructure, technical expertise, and reimbursement mechanisms remain limited. Addressing these disparities will be critical to ensure that advances in liquid biopsy do not widen the global gap in cancer outcomes.

The clinical utility of ctDNA in NCGIs malignancies continues to be validated through ongoing research (3,4). While numerous studies have demonstrated its potential for early detection, treatment monitoring, and prognosis, most evidence comes from small, retrospective cohorts, limiting the generalizability of findings (3,4). These studies are organized in Tables S1-S4 and synthesize the current evidence base across GI cancers. Larger, prospective studies are needed to confirm the role of ctDNA in rountine clinical practice and to establish evidence-based guidelines for its use (3,4). Furthermore, integrating ctDNA with other biomarkers, such as circulating tumor cells (CTCs) or exosomes, and combining it with advanced imaging techniques could enhance its diagnostic and prognostic accuracy, though this multimodal approach requires further investigation (3).

Conclusions

In conclusion, ctDNA represents a transformative tool in managing NCGIs, offering non-invasive, real-time insights into tumor biology. While challenges remain, the continued advancement of ctDNA technology and its integration into clinical practice has the potential to significantly improve early detection, treatment selection, and patient outcomes in HCC, PDAC, and esophageal cancer (3,4,13).

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://tgh.amegroups.com/article/view/10.21037/tgh-25-28/rc

Peer Review File: Available at https://tgh.amegroups.com/article/view/10.21037/tgh-25-28/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tgh.amegroups.com/article/view/10.21037/tgh-25-28/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Singh A. Global burden of five major types of gastrointestinal cancer. Prz Gastroenterol 2024;19:236-54. [Crossref] [PubMed]
2. Zhang X, Xu Z, Shang L, et al. Global burden of colorectal cancer attributable to metabolic risks from 1990 to 2021, with predictions to 2046. BMC Cancer 2025;25:228. [Crossref] [PubMed]
3. American Cancer Society Cancer Facts & Figures2025. Available online: https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/2025-cancer-facts-figures.html
4. Ma L, Guo H, Zhao Y, et al. Liquid biopsy in cancer current: status, challenges and future prospects. Signal Transduct Target Ther 2024;9:336. [Crossref] [PubMed]
5. Mishra M, Ahmed R, Das DK, et al. Recent Advancements in the Application of Circulating Tumor DNA as Biomarkers for Early Detection of Cancers. ACS Biomater Sci Eng 2024;10:4740-56. [Crossref] [PubMed]
6. Fujiyoshi K, Sugihara R, Miyamoto N, et al. Comprehensive genetic profiling using tissue and blood in locally advanced tumors. Oncol Rep 2025;54:134. [Crossref] [PubMed]
7. Li T, Lin C, Wang W. Global, regional, and national burden of pancreatic cancer from 1990 to 2021, its attributable risk factors, and projections to 2050: a systematic analysis of the global burden of disease study 2021. BMC Cancer 2025;25:189. [Crossref] [PubMed]
8. Buonopane RI, Saldanha E, De Menezes JSA, et al. Circulating tumor DNA as a minimal residual disease assessment and recurrence risk in hepatocellular carcinoma: A systematic review and meta-analysis. J Clin Oncol 2025;43:e15055. [Crossref] [PubMed]
9. Nakamura Y, Watanabe J, Akazawa N, et al. ctDNA-based molecular residual disease and survival in resectable colorectal cancer. Nat Med 2024;30:3272-83. [Crossref] [PubMed]
10. Iden CR, Mustafa SM, Øgaard N, et al. Circulating tumor DNA predicts recurrence and survival in patients with resectable gastric and gastroesophageal junction cancer. Gastric Cancer 2025;28:83-95. [Crossref] [PubMed]
11. Wang B, Wang M, Lin Y, et al. Circulating tumor DNA methylation: a promising clinical tool for cancer diagnosis and management. Clin Chem Lab Med 2024;62:2111-27. [Crossref] [PubMed]
12. Wang MD, Diao YK, Yao LQ, et al. Emerging role of molecular diagnosis and personalized therapy for hepatocellular carcinoma. ILIVER 2024;3:100083. [Crossref] [PubMed]
13. Bardol T, Pageaux GP, Assenat E, et al. Circulating Tumor DNA Clinical Applications in Hepatocellular Carcinoma: Current Trends and Future Perspectives. Clin Chem 2024;70:33-48. [Crossref] [PubMed]
14. Labori KJ. Prognostic and monitoring potential of circulating tumor DNA in resectable pancreatic cancer. Transl Gastroenterol Hepatol 2025;10:37. [Crossref] [PubMed]
15. Jonnalagadda P, Arnold V, Weinberg BA. A Review of Circulating Tumor DNA (ctDNA) in Pancreatic Cancer: Ready for the Clinic? J Gastrointest Cancer 2025;56:50. [Crossref] [PubMed]
16. Labiano I, Huerta AE, Alsina M, et al. Building on the clinical applicability of ctDNA analysis in non-metastatic pancreatic ductal adenocarcinoma. Sci Rep 2024;14:16203. [Crossref] [PubMed]
17. Rompen IF, Habib JR, Wolfgang CL, et al. Anatomical and Biological Considerations to Determine Resectability in Pancreatic Cancer. Cancers (Basel) 2024;16:489. [Crossref] [PubMed]
18. Tsumura T, Doi K, Marusawa H. Precision Medicine of Hepatobiliary and Pancreatic Cancers: Focusing on Clinical Trial Outcomes. Cancers (Basel) 2022;14:3674. [Crossref] [PubMed]
19. Botta GP, Abdelrahim M, Drengler RL, et al. Association of personalized and tumor-informed ctDNA with patient survival outcomes in pancreatic adenocarcinoma. Oncologist 2024;29:859-69. [Crossref] [PubMed]
20. Maulat C, Canivet C, Cabarrou B, et al. Prognostic impact of circulating tumor DNA detection in portal and peripheral blood in resected pancreatic ductal adenocarcinoma patients. Sci Rep 2024;14:27296. [Crossref] [PubMed]
21. Eckhoff AM, Kanu E, Fletcher A, et al. Initial Report: Personalized Circulating Tumor DNA and Survival in Patients with Resectable Pancreatic Cancer. Ann Surg Oncol 2024;31:1444-6. [Crossref] [PubMed]
22. Watanabe F, Suzuki K, Aizawa H, et al. Circulating tumor DNA in molecular assessment feasibly predicts early progression of pancreatic cancer that cannot be identified via initial imaging. Sci Rep 2023;13:4809. [Crossref] [PubMed]
23. Ji J, Wang C, Goel A, et al. Circulating Tumor DNA Testing in Curatively Resected Colorectal Cancer and Salvage Resection. JAMA Netw Open 2024;7:e2452661. [Crossref] [PubMed]
24. Kasi PM, Aushev VN, Ensor J, et al. Circulating tumor DNA (ctDNA) for informing adjuvant chemotherapy (ACT) in stage II/III colorectal cancer (CRC): Interim analysis of BESPOKE CRC study. J Clin Oncol 2024;42.
25. Kataoka K, Mori K, Nakamura Y, et al. Survival benefit of adjuvant chemotherapy based on molecular residual disease detection in resected colorectal liver metastases: subgroup analysis from CIRCULATE-Japan GALAXY. Ann Oncol 2024;35:1015-25. [Crossref] [PubMed]
26. Abdelrahim M, Mejia A, Esmail A, et al. 181P Circulating tumor DNA (ctDNA) testing for recurrence and treatment response monitoring in hepatocellular carcinoma (HCC). Ann Oncol 2024;35:S82.
27. Krell M, Llera B, Brown ZJ. Circulating Tumor DNA and Management of Colorectal Cancer. Cancers (Basel) 2023;16:21. [Crossref] [PubMed]
28. Emiloju OE, Storandt M, Zemla T, et al. Tumor-Informed Circulating Tumor DNA for Minimal Residual Disease Detection in the Management of Colorectal Cancer. JCO Precis Oncol 2024;8:e2300127. [Crossref] [PubMed]
29. Puccini A, Martelli V, Pastorino A, et al. ctDNA to Guide Treatment of Colorectal Cancer: Ready for Standard of Care? Curr Treat Options Oncol 2023;24:76-92. [Crossref] [PubMed]
30. Loree JM, Bubie A, Eluri M, et al. Clinical utility of serial circulating tumor DNA (ctDNA) to identify acquired re-sistance to anti-EGFR antibodies in metastatic colorectal cancer (mCRC). J Clin Oncol 2025;42:158.
31. Zhou Q, Chen X, Zeng B, et al. Circulating tumor DNA as a biomarker of prognosis prediction in colorectal cancer: a systematic review and meta-analysis. J Natl Cancer Cent 2024;5:167-78. [Crossref] [PubMed]
32. O'Sullivan NJ, Temperley HC, Kyle ET, et al. Assessing circulating tumour DNA (ctDNA) as a prognostic biomarker in locally advanced rectal cancer: a systematic review and meta-analysis. Int J Colorectal Dis 2024;39:82. [Crossref] [PubMed]
33. Guo Q, Shan L, Luo J, et al. Novel strategies in liquid biopsy. Clin Chim Acta 2025;576:120385. [Crossref] [PubMed]
34. Zhang X, Eyck BM, Yang Y, et al. Accuracy of detecting residual disease after neoadjuvant chemoradiotherapy for esophageal squamous cell carcinoma (preSINO trial): a prospective multicenter diagnostic cohort study. BMC Cancer 2020;20:194. [Crossref] [PubMed]
35. Yang W, Fang Z, Wang X, et al. Abstract 6412: Combination of liquid biopsy and PET/CT enhances prediction of pathological response to neoadjuvant immunochemotherapy in patients with esophageal squamous cell carcinoma. Cancer Res 2024;84:6412.