Translate this page into:
Multi-Cancer Early Detection (MCED) circulating tumour DNA testing (ctDNA) - the Oncology Beyond the Obvious (OBTO) interpretation
-
Received: ,
Accepted: ,
How to cite this article: Parikh PM, Vora A, Patkar N, Limaye S, Batra U. Multi-Cancer Early Detection (MCED) circulating tumour DNA testing (ctDNA) - the Oncology Beyond the Obvious (OBTO) interpretation. South Asian J Cancer. 2026;15:1-4. doi: 10.25259/SAJC_47_2026
On 3rd February 2026, the USA signed into law the Nancy Gardner Sewell Medicare Multi-Cancer Early Detection (MCED) Screening Coverage Act (the Miller-Meeks Bill), which was passed by both chambers of Congress.[1] The implication is that at a future date, Medicare will cover payment for these cancer tests in the general population. This has far-reaching implications, some of which we will discuss in this editorial.
The circulating tumour DNA (ctDNA) testing generally falls into three categories, viz. companion diagnostics for therapy selection, monitoring for minimal residual disease (MRD), and early detection. We will restrict to the last one, screening for early detection of multiple cancers. MCED “screening” tests are intended to be used for asymptomatic individuals, so that they can detect cancer signals across many different cancer types, from a single blood sample.
Three such tests are [Table 1]:
| Test category | Example products | Primary goal | Tissue required? |
|---|---|---|---|
| Therapy selection | FoundationOne CDx | Finding the right drug | No (blood only) |
| MRD / monitoring | Signatera, NeXT personal | Detecting early recurrence | Yes (initial biopsy) |
| Early detection | Galleri, shield | Screening healthy people | No (blood only) |
MCED: Multi-Cancer Early Detection, ctDNA: Circulating tumour DNA, MRD: Minimal residual disease.
Caris Detect (Caris Life Sciences): Uses whole-genome sequencing (WGS).[2]
Shield (Guardant Health): Primarily for colorectal cancer screening.[3]
Galleri (GRAIL): Detects a signal across 50+ cancer types using epigenomics/methylation patterns.[4]
The largest data currently available is from the Achieve 1 Study (Caris Detect). It evaluated over 3,000 subjects, including high-risk and symptomatic populations. A significantly higher sensitivity was achieved for early-stage cancers, as compared to earlier-generation liquid biopsies [Table 2].
| Cancer stage | Sensitivity (detection rate) (%) |
|---|---|
| Stage I | 56.8 |
| Stage II | 67.7 |
| Stage III | 79.0 |
| Stage IV | 98.6 |
| Specificity | 99.2 (in asymptomatic screening populations) |
It seems to be ticking all the important boxes, Broad Biological approach through ultra-deep WGS, plan to incorporate whole transcriptome sequencing (WTS) to boost sensitivity and AI Training (CODEai™ engine) on a database of over 1 million real-world molecular profiles and 50 billion molecular markers. Add this to their technology of simultaneously analysing multiple components of the sample by using what they call circulating nucleic acid sequencing (cNAS). This includes cell-free DNA (cfDNA), cell-free RNA (cfRNA), and buffy coat analysis (to reduce noise). Their ultra-deep sequencing allows detection of tumour DNA even when it makes up <0.01% of the total DNA in the sample. Because they are using WGS, it can detect Aneuploidy and Copy Number Alterations, features that are hallmarks of early-stage cancer that cannot be detected by tests limited to specific genes (tumour-informed) or a methylation approach. Thus, Caris Detect “brute force” sequencing offers higher sensitivity for Stage I breast and prostate cancers (>50%) compared to first-generation tests that relied solely on DNA methylation.
B. Shield™ (Guardant Health) was the first blood test to receive full approval (July 2024) from the Food and Drug Administration (FDA) as a primary screening option, but it is restricted to colorectal cancer (CRC). Its value is for the complacent population that does not take advantage of traditional free screening methods because they are invasive. The target population is average-risk adults aged ≥45 years. Its value is strengthened by being included in the National Comprehensive Cancer Network (NCCN) clinical practice guidelines.
Data from Shield’s ECLIPSE trial (over 20,000 patients) is summarised in Table 3.
| Metric | Performance (%) | Notes |
|---|---|---|
| CRC sensitivity | 83.1 | Detects roughly 8 out of 10 colorectal cancers |
| Specificity | 89.6 | About a 10% chance of a “false positive” |
| Advanced adenoma | 13.0 | Key limitation: It is poor at detecting pre-cancerous polyps compared to colonoscopy |
| Adherence rate | 95.0 | Key strength: Real-world 2025- 2026 data show patients are much more likely to complete this test than a stool kit or colonoscopy |
CRC: Colorectal cancer
Since Shield is approved for Medicare coverage (65+ population with minimal out-of-pocket costs), it is now being used as a “blood-first” approach, especially for people who have historically refused colonoscopies or failed to return stool-based kits. It is to be remembered that this test is not for high-risk people, and a positive test should be immediately followed by a confirmatory “diagnostic” colonoscopy. It will be interesting to know what impact the Shield Multi-Cancer Detection (MCD) will achieve, since its updated technology has been expanded to detect 10+ different cancers [methylation partitioning cell-free DNA (mp-cfDNA)] by looking for genetic mutations. Guardant’s “secret sauce” is the use of a proprietary “probe” set that targets over 20,000 genomic regions known to be differentially methylated. It is important to remember that fragmentomics is based on the analysis of the length of the DNA, tumour DNA fragments being shorter than healthy DNA. While early-stage tumours (Stage I and II) usually do not shed enough DNA for mutations to be detected, the methylation patterns of cancer give a more robust and easily detectable signal.
GALLERI: The “broad signal” specialist
Based on GRAIL’s technology, the Galleri ctDNA test focuses on DNA methylation profiling of more than 100,000 regions involved in about 50 different cancers. It was validated in the CCGA (Circulating cell-free Genome Atlas) study. Its specificity is said to be high (more than 98%). Its limiting factor is that it is better for late-stage detection. Based on modelling projections, it was expected to reduce stage III and IV cancers by 50% and reduce deaths by one-fifth in as little as 5 years. The National Health Service (NHS) Galleri trial was launched with great expectations, which were dampened significantly when their February press release confirmed that the primary endpoint was not met, even when the study consisted of 142,000 subjects. The investigators have announced an extension of six months to the follow-up in the hope that it will make the statistical results more robust. We will know soon enough when the full results are presented at the American Society of Clinical Oncology (ASCO) annual meeting in June. But the likelihood of meeting the primary endpoint statistical significance is bleak.
Table 4 compares the specific laboratory workflows for these three tests. Technical comparison of the three MCED approaches are shown in Table 5.
| Feature | Galleri (GRAIL) | Caris Detect (Caris) | Shield (Guardant) |
|---|---|---|---|
| Primary method | Targeted methylation | Ultra-deep WGS | Methylation partitioning |
| DNA coverage | Over 100,000 specific methylation regions | The entire genome (whole genome sequencing) | ~20,000 targeted genomic regions |
| Analyte focus | Epigenomic: DNA methylation patterns (tags) | Multi-Modal: mutations, aneuploidy (DNA), and transcripts (RNA) | Epigenomic: focuses on colorectal-specific DNA methylation |
| Noise filtering | High-specificity machine learning filters | White blood cell (WBC) sequencing to subtract ageing noise (CHIP) | Algorithmic filtering of non-tumour methylation signals |
| Biological pillars | Focuses on “cancer signal of origin” (CSO) | Uses “nine pillars” of biology, including the whole transcriptome | Optimised for high sensitivity in a single organ (colon) |
| Sequence depth | High-depth targeted sequencing | Extreme ultra-deep coverage across 3 billion bases | Targeted high-depth sequencing |
| Primary indicator | Methylation-based “cancer signal” | Complex genomic/transcriptomic “molecular signature” | Binary methylation score (positive/negative) |
WGS: Whole genome sequencing, DNA: Deoxyribonucleic acid, RNA: Ribonucleic acid, CHIP: Clonal haematopoiesis of indeterminate potential, CSO: Cancer signal of origin.
| Feature | Galleri (GRAIL) | Caris Detect (Caris) | Shield (Guardant) |
|---|---|---|---|
| Primary method | Targeted methylation (epigenetics) | Whole-Genome Sequencing (WGS) + WTS | Targeted methylation partitioning (mp-cfDNA) |
| Detection signal | Chemical “tags” (methylation patterns) | Massive genomic deletions, insertions, and expressions | Organ-specific chemical tags (colorectal optimised) |
| Biological insight | Identifies “how the DNA is packaged” | Analyses the entire genetic blueprint (brute force) | Pinpoints specific “high-volume” cancer signals |
| Key capability | Cancer signal of origin (CSO) (where the cancer is) | Exceptional sensitivity for early-stage (stage I) disease | Optimised for primary screening (high cancer detection) |
| Aneuploidy detection | (Chromosomal gains/losses) primary focus | (Mutations, deletions, aneuploidy) primary focus | (Methylation focus) limited |
| CHIP correction | High (algorithmic filtering) | Very high (directly sequences the patient's white blood cells) | High (algorithmic filtering) |
| Multi-modal analytes | DNA only (methylation) | Multi-OMIC (DNA + RNA [whole transcriptome]) | DNA only (methylation + mutations) |
| Proprietary process | High-depth targeted sequencing | cNAS™ (circulating nucleic acid seq) | mp-cfDNA™ (methylation partitioning) |
MCED: Multi-cancer early detection, DNA: Deoxyribonucleic acid, CSO: Cancer signal, CHIP: Clonal haematopoiesis of indeterminate, RNA: Ribonucleic acid.
The verdict on MCED in 2026
Any screening blood test that is expected to be used in the general population is a massive exercise, with financial, logistic and policy implications.[5,6] Hardly any screening program has resulted in all-cause mortality. If the ctDNA testing for MCED is to be successful, it has to cross several hurdles. Is the increase in the diagnosis of early-stage cancers an overdiagnosis? Will it shift to the early stage of detection consistently across multiple cancer types? What happens in cases of prostate and thyroid cancers, where MCED typically has lower sensitivity? Will earlier diagnosis only enrich for patients with subclinical micrometastasis? Are commercial and financial implications smothering evidence-based medicine? NHL Galleri results be confirmed by data from other MCED trials? (More than 200,000 patient data exists from Symplify, Pathfinder, Pathfinder2, CancerSEEK, Detect-A Reflection, Reach, Summit, and Strive trials).
On the other hand, we should not lose sight of the potential benefits. The positive predictive value across all cancers is more than 60%. It can help detect cancers that are not amenable to other screening options (73% of all MCED-detected cancers do not have any recommended screening today – prostate, stomach, pancreas, gall bladder, renal, etc). ctDNA detection is associated with poor biology of the cancer rather than the stage. It can predict tissue of origin (>90%) in patients who are true positives, and the tumour origin can be verified. The approach can be a useful adjunct for high-risk populations.
Several tests developed and validated in India are also knocking on the doors of the regulatory authorities.[7-12] We trust that there is no bias or discrimination against us; that they take decisions based purely on the robustness of the science and data; and that these tests provide a meaningful difference to patients all over the world.
In conclusion, MCED trials address important unmet needs, they are based on good scientific rationale, end points are practical though ideally we would want to wait for proof of mortality reduction, the clinically significant of the intervention is not clear at present, it is not practice changing in its present form and it needs to be a clear winner in the Western world (USA and European Union) before it can be considered for other countries across the globe.
References
- Prevent Cancer Foundation celebrates landmark passage of MCED screening coverage act. Available from: https://preventcancer.org/news/prevent-cancer-foundation-celebrates-landmark-passage-of-mced-screening-coverage-act/. [Last accessed 2026 Apr 16]
- [Google Scholar]
- Caris Life Sciences finalizes ACHIEVE-1 study results reinforcing the superior sensitivity and specificity of Caris Detect. Available from: https://www.carislifesciences.com/about/news-and-media/caris-life-sciences-finalizes-achieve-1-study-results-reinforcing-the-superior-sensitivity-and-specificity-of-caris-detect/. [Last accessed 2026 Apr 14]
- [Google Scholar]
- A cell-free DNA blood-based test for colorectal cancer screening. N Engl J Med. 2024;390:973-83.
- [CrossRef] [PubMed] [Google Scholar]
- Galleri cancer tests: controversial NHS trial fails to meet key goal. BMJ. 2026;392:s364.
- [Google Scholar]
- Estimating the incremental population health impact of a multi-cancer early detection (MCED) test to complement existing screening among elevated risk populations with multiple cancer risk factors: A mathematical modeling study. BMC Health Serv Res. 2024;24:1584.
- [CrossRef] [PubMed] [Google Scholar]
- Multi-cancer early detection tests: Attributes for clinical implementation. Cancer Biomark. 2025;42:18758592241297849.
- [CrossRef] [PubMed] [Google Scholar]
- Liquid biopsy-A pivotal test to help navigate clinical decisions at a precision center in India. J Liq Biopsy. 2025;9:100323.
- [CrossRef] [PubMed] [Google Scholar]
- Circulating tumor DNA profiling for non-invasive genomic analysis in Indian lung cancer patients: A real-world experience. J Liq Biopsy. 2025;8:100300.
- [CrossRef] [PubMed] [Google Scholar]
- Liquid biopsy for detection of pancreaticobiliary cancers by functional enrichment and immunofluorescent profiling of circulating tumor cells and their clusters. Cancers (Basel). 2024;16:1400.
- [CrossRef] [PubMed] [Google Scholar]
- Accurate prostate cancer detection based on enrichment and characterization of prostate cancer specific circulating tumor cells. Cancer Med. 2023;12:9116-27.
- [CrossRef] [PubMed] [Google Scholar]
- Accurate screening for early-stage breast cancer by detection and profiling of circulating tumor cells. Cancers (Basel). 2022;14:3341.
- [CrossRef] [PubMed] [Google Scholar]
- CancerSpot: A multi-cancer early detection test using targeted methylation sequencing In: Proceedings of the American Association for Cancer Research Annual Meeting 2025. Chicago, IL. Philadelphia (PA): AACR; 2025. p. :A25-30. Cancer Res 2025;85(Suppl 1):Abstract 2343. Available from: https://aacrjournals.org/cancerres/article/85/8_Supplement_1/2343/758441/Abstract-2343-CancerSpot-a-multi-cancer-early [Last accessed 2026 Apr 17]
- [CrossRef] [Google Scholar]