Observational study finds a blood test detecting multiple cancers with high specificity — Evidence Review
Published in Proceedings of the National Academy of Sciences, by researchers from UCLA
Researched by
— the AI search engine for science
Table of Contents
A new study from UCLA reports that a cost-effective blood test called MethylScan can detect multiple cancers and liver diseases with high specificity and moderate sensitivity. Findings align with prior research showing that blood-based methylation profiling is a promising approach for early, multi-cancer detection.
- Several related studies confirm that cell-free DNA methylation analysis and other blood-based biomarkers can effectively detect and localize a range of cancers, often with high specificity and moderate sensitivity, supporting the results of the UCLA study 2 3 6 8 9.
- Unlike some existing tests that focus on mutations or require deep sequencing, MethylScan uses selective methylation enrichment to reduce background DNA and costs, building on recent advances in assay design and machine learning for better tissue-of-origin identification 7 8.
- The broader literature highlights both the clinical promise and ongoing challenges of multi-cancer early detection blood tests, including the need for large-scale validation and real-world population studies 2 6 12.
Study Overview and Key Findings
Early detection of cancer and other organ diseases remains a critical unmet need in medicine, as outcomes are significantly improved when conditions are caught early. Current liquid biopsy approaches often focus on a limited set of mutations or require deep sequencing, making them expensive and sometimes less informative for detecting multiple diseases simultaneously. The UCLA team developed MethylScan, a blood test that analyzes methylation patterns in circulating DNA, aiming for an affordable, broad-spectrum screening tool that can detect cancers, liver diseases, and signs of organ stress in a single assay.
| Property | Value |
|---|---|
| Organization | UCLA |
| Journal Name | Proceedings of the National Academy of Sciences |
| Authors | Dr. Jasmine Zhou, Dr. Wenyuan Li, Weihua Zeng, Shuo Li, Yonggang Zhou |
| Population | Patients with various cancers and liver diseases, healthy participants |
| Sample Size | n=1061 |
| Methods | Observational Study |
| Outcome | Detection of multiple cancers and liver diseases, tissue of origin |
| Results | MethylScan identified 63% of cancers with 98% specificity. |
Literature Review: Related Studies
To contextualize the new findings, we searched the Consensus research database (200+ million papers) for studies on blood-based cancer detection accuracy, methylation-based diagnostics, and biomarkers for early cancer detection. The following queries were used:
- blood test cancer detection accuracy
- MethylScan specificity cancer diagnosis
- early cancer detection blood biomarkers
| Topic | Key Findings |
|---|---|
| How effective are blood-based tests for early, multi-cancer detection? | - Blood-based methylation and multi-analyte tests (e.g., CancerSEEK, MCED, PanSeer) provide moderate to high sensitivity (51–95%) and high specificity (96–99.5%) for detecting multiple cancers, including early-stage disease 1 2 3 6 8 9. - Early-stage sensitivity remains a challenge, with most tests detecting 40–74% of early cancers 2 3 6 8 9. |
| What are the technical and economic barriers to widespread use of methylation-based assays? | - High sequencing depth and background DNA from blood cells increase costs and reduce sensitivity; targeted methylation enrichment and multimodal approaches can mitigate this 6 7 8 9. - Recent studies demonstrate that cost-effective, shallow sequencing with enrichment can achieve performance similar to deep sequencing at lower cost 7 8. |
| Can these tests reliably identify the tissue or organ of origin of cancer signals? | - Machine learning applied to methylation, fragmentomics, or RNA profiles can predict tissue of origin with 85–93% accuracy in positive cases 1 2 6 7 8 9. - Accurate localization is critical to guide follow-up diagnostics and is achievable with integrated multi-omics or advanced computational approaches 1 2 6 7 8 9. |
| What are the current limitations and future needs for blood-based early detection? | - Validation in large, prospective, diverse populations is needed before routine clinical adoption 2 6 12. - Sensitivity for early-stage and pre-cancerous lesions, as well as differentiation from benign disease, varies and requires further study 2 4 6 12. |
How effective are blood-based tests for early, multi-cancer detection?
Multiple studies have demonstrated that blood-based assays using cell-free DNA (cfDNA) methylation, mutations, or multi-analyte approaches can detect a variety of cancer types with high specificity and moderate sensitivity, especially for early-stage disease. The UCLA study's results are consistent with this literature, with MethylScan achieving 63% overall cancer sensitivity (55% for early-stage) at 98% specificity, paralleling the performance of other leading tests.
- CancerSEEK, an earlier multi-analyte blood test, achieved a median sensitivity of 70% across eight cancer types, with specificity above 99% 1.
- Methylation-based MCED tests have reported overall sensitivities of 51–74% and specificities above 97%, with early-stage sensitivity in the 40–74% range 2 6 8.
- The PanSeer assay detected cancer up to four years before clinical diagnosis in 95% of participants who later developed cancer, with 96% specificity 3.
- Early detection rates for specific cancers (e.g., colorectal) can be higher (up to 83%) when the test is optimized for a single cancer type, but multi-cancer tests generally have lower sensitivity for early-stage disease 4.
What are the technical and economic barriers to widespread use of methylation-based assays?
A major technical challenge for cfDNA assays is the abundance of background DNA from normal blood cells, which can obscure rare cancer signals and necessitate high sequencing depth, driving up costs. The UCLA team's approach of selectively removing unmethylated DNA and enriching for solid-organ methylation patterns is in line with recent innovations to make these assays more affordable and practical.
- Targeted methylation enrichment (e.g., cfMethyl-Seq, SPOT-MAS) and shallow genome-wide sequencing have enabled high specificity and moderate sensitivity with significantly reduced sequencing requirements 7 8.
- Multimodal approaches that combine methylation, fragmentomics, and copy number profiling have shown that meaningful sensitivity can be achieved at lower cost, which is critical for population-level screening 7 9.
- The cost-effectiveness of these new approaches (e.g., $20 per sample in the UCLA study) could accelerate adoption in routine care 7 8.
Can these tests reliably identify the tissue or organ of origin of cancer signals?
One of the most clinically valuable features of methylation-based cfDNA tests is their ability to pinpoint the likely tissue or organ source of a cancer signal, guiding follow-up imaging or biopsies. The UCLA study echoes this, with MethylScan reliably identifying tissue of origin for cancer and distinguishing among liver disease subtypes.
- Machine learning models trained on methylation, fragmentomics, or RNA signatures can localize the tissue of origin in 85–96% of true positive cases 1 2 6 7 8 9.
- Accurate localization is particularly valuable in multi-cancer early detection settings, where the source of an abnormal test result needs to be rapidly narrowed for further diagnostic workup 1 2 6.
- The ability to distinguish between benign and malignant conditions—such as viral hepatitis versus metabolic liver disease—has also been demonstrated by methylation profiling 6 8.
What are the current limitations and future needs for blood-based early detection?
Despite promising results, substantial work remains before cfDNA methylation tests become standard practice for broad population screening. Limitations include the need for prospective, real-world validation, improved sensitivity for early-stage and pre-cancerous lesions, and robust differentiation from benign diseases.
- Most published studies, including the UCLA work, are observational or case-control in design; large, prospective population studies are needed to assess clinical utility and real-world performance 2 6 12.
- Sensitivity for early-stage cancers and precancerous lesions is lower than for advanced disease, and some tests have limited ability to distinguish cancer from benign or inflammatory conditions 2 4 6.
- Ongoing development of non-invasive biomarker panels and integration of multi-omics data may improve early detection rates, but clinical implementation will require addressing issues of cost, accessibility, and follow-up protocols 11 13 15.
Future Research Questions
While the UCLA study and related research underline the promise of cfDNA methylation analysis for early, multi-disease detection, several key questions remain. Future work should address clinical validation in diverse populations, test performance in real-world screening, and technical optimization for even greater sensitivity and specificity.
| Research Question | Relevance |
|---|---|
| How does MethylScan perform in large, prospective population screening settings? | Real-world validation is critical to determine clinical utility, as past case-control studies may not reflect population-level performance or logistical challenges 2 6 12. |
| Can cfDNA methylation-based tests improve early-stage and precancerous lesion detection? | Sensitivity for early-stage cancers and pre-cancerous lesions remains a limitation; improved methods or combined biomarker panels could address this gap 2 4 6 8. |
| What is the impact of background DNA removal and shallow sequencing on test accuracy and cost-effectiveness? | Technical innovations like selective methylation enrichment could make widespread screening feasible, but their impact on accuracy and cost in diverse settings needs further study 7 8. |
| How accurately can multi-disease blood tests distinguish benign from malignant or inflammatory conditions? | Distinguishing between cancer, benign disease, and inflammatory conditions is essential to minimize false positives and unnecessary follow-up 4 6 8. |
| What are the long-term clinical outcomes and cost-benefit of multi-cancer early detection blood tests in routine care? | Understanding the impact on patient survival, quality of life, health system costs, and screening adherence is crucial for policy and clinical adoption 11 13 15. |