Blood-based circulating tumour DNA (ctDNA) assays are being actively explored as a non-invasive strategy for colorectal cancer (CRC) screening. While current screening strategies such as colonoscopy and stool-based tests have demonstrated effectiveness, their uptake remains suboptimal due to invasiveness and patient-related barriers.
This systematic review and meta-analysis aimed to evaluate the diagnostic accuracy of ctDNA-based blood tests in detecting colorectal cancer and advanced colorectal neoplasia in asymptomatic, average-risk populations, and to clarify their potential role in population-level screening.
The results of this study were published in Critical Reviews in Oncology/Hematology, Volume 222, in June 2026.
Title: Blood-based circulating tumour DNA (ctDNA) tests for colorectal cancer screening: Systematic review and meta-analysis of diagnostic accuracy
Authors: Fabio Carbone, Davide Ciardiello, Stefano Granieri, Nicola Fazio, Antonio Avallone, Paolo Delrio
Read more about Colon Cancer Success Rate on OncoDaily.
Study Design and Population
This work was designed as a systematic review and meta-analysis conducted in accordance with Cochrane Diagnostic Test Accuracy (DTA) principles. A comprehensive literature search was performed across PubMed, EMBASE, and the Cochrane Library up to July 2025.
Studies were eligible if they evaluated ctDNA-based blood assays in asymptomatic, average-risk populations and used colonoscopy with histopathological confirmation as the reference standard. This design ensured that included studies reflected real-world screening settings rather than enriched diagnostic cohorts.
The analysis focused on evaluating diagnostic performance across the following endpoints:
- Invasive colorectal cancer (CRC)
- Advanced precancerous lesions (APL)
- Advanced colorectal neoplasia (ACN), defined as CRC plus APL
The analysis focused on pooled sensitivity and specificity, along with additional diagnostic measures including likelihood ratios and area under the curve (AUC), calculated using a bivariate random-effects model.
From an initial 227 identified studies, 180 records underwent screening after duplicate removal, and 33 were assessed in full text. Ultimately, three prospective population-based studies met inclusion criteria and were included in the final analysis, representing a total of 36,381 participants. These included studies reflect large-scale screening populations and provide a relevant basis for evaluating real-world performance.
Key Findings
A total of 36,381 participants were included in the final analysis, comprising 4187 cases of advanced colorectal neoplasia (12%), 190 colorectal cancers (0.5%), and 3997 advanced precancerous lesions (11%).
Diagnostic performance
The diagnostic performance of ctDNA assays varied across clinical endpoints.
For advanced colorectal neoplasia, overall detection remained limited, with sensitivity around 0.16 (95% CI 0.14–0.18), despite consistently high specificity of approximately 0.91 (95% CI 0.89–0.92). Additional measures of diagnostic performance further supported these findings, confirming the modest discriminatory ability, with no significant heterogeneity observed.
For colorectal cancer detection, ctDNA testing demonstrated moderate sensitivity (~0.72) and high specificity (~0.91), indicating strong overall diagnostic performance. This was supported by favorable likelihood ratios (LR+ 7.70; LR− 0.30) and an AUC of 0.92, although substantial heterogeneity was observed.
When focusing on next-generation methylation-based assays, diagnostic performance improved, with sensitivity reaching 0.81 (95% CI 0.73–0.88) and consistently high specificity (~0.90), alongside a high AUC of 0.93 and reduced variability between studies.
In contrast, performance for advanced precancerous lesions remained limited, with sensitivity of 0.13 (95% CI 0.12–0.14) despite preserved specificity (~0.90). Overall, these results indicate poor discriminatory capacity for precursor lesions, as reflected by low likelihood ratios and a low AUC (0.17).
Stage-Specific Performance
Sensitivity increased with tumour stage, from 0.53 (95% CI 0.37–0.69) in stage I disease to approximately 0.89 in more advanced stages, including stage II and IV, while stage III reached 0.82 (95% CI 0.38–0.97). Across all stages, specificity remained around 0.90, although heterogeneity was consistently high (I² 77.8%–86.6%).
What Do These Results Mean in Practice?
When applied to a low-prevalence screening population (CRC prevalence ~0.5%), these estimates translate into a relatively small number of detected cases alongside a notable proportion of false-positive results. While the negative predictive value remains very high (~99.8%), the positive predictive value is low (~3.9%), underscoring the challenges of screening in this setting.
Evidence Quality and Limitations
Risk of bias was assessed as unclear in some domains, particularly patient selection and flow/timing, although applicability concerns were low. Certainty of evidence was moderate for CRC and ACN, and low for APL detection. Publication bias could not be assessed due to the small number of studies.
Expert Perspective
The clinical role of ctDNA in colorectal cancer screening continues to evolve, particularly regarding whether it should be positioned as a complementary tool or a future alternative to established strategies.
Davide Ciardiello, MD, PhD, medical oncologist at the Division of Gastrointestinal Medical Oncology and Neuroendocrine Tumours, European Institute of Oncology IRCCS, Milan, Italy, said:
“I believe that liquid biopsy can currently be a valid tool for increasing adherence to colorectal cancer screening. Currently, the method is highly sensitive for advanced lesions, but less so for precancerous lesions. For this reason, for now, it cannot replace colonoscopy.”
Addressing the future of ctDNA-based screening and the developments needed for broader implementation, he continued:
“However, in the future, with the use of more sensitive methods, it could become an interesting option, especially considering the non-invasivness and the possibility of repeating the test over the time.”
Conclusion
This meta-analysis demonstrates that ctDNA-based blood tests provide high specificity and clinically meaningful accuracy for invasive colorectal cancer detection, particularly with next-generation methylation assays.
However, their consistently low sensitivity for advanced precancerous lesions limits their role in cancer prevention. These findings position ctDNA as a complementary screening approach rather than a replacement for established methods.
Further technological refinement and large-scale validation studies are required to define the role of ctDNA testing within population-based colorectal cancer screening programmes.
Read the full article here.