Liquid Biopsy Biomarkers: Testing Blood Instead of Cutting Into Tumors

Liquid biopsies analyze tumor-derived material in blood or other fluids instead of tissue resection. Wan et al. (2017) review how circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), and extracellular vesicles each carry different clinical information.¹ The useful question is which analyte fits your intended use, not whether blood is always better than tissue.

What Each Analyte Measures

Circulating tumor DNA (ctDNA)

ctDNA fragments reflect somatic alterations from dying tumor cells. Razavi et al. (2019) showed that plasma cfDNA variant sources include tumor and non-tumor contributions, which matters for assay design and false-positive interpretation.²

Applications include mutation detection for targeted therapy, serial monitoring during treatment, and post-surgical minimal residual disease (MRD) assessment when papers report validated assays for those contexts.

CTCs and extracellular vesicles

CTCs are rare intact tumor cells; isolation and analysis are technically demanding. Exosomes and other vesicles carry proteins and nucleic acids that may complement DNA-only panels. Literature claims should be read per analyte and per indication; pooling heterogeneous assays obscures what replicates.

Cited Clinical Examples

Multi-cancer early detection (MCED)

Klein et al. (2021) reported clinical validation of a targeted methylation MCED test in an independent validation set, with performance characteristics developed in the broader CCGA program.³ In CCGA substudies, overall sensitivity for cancer signal detection was 51.5% with 99.5% specificity, with stage-dependent performance (Klein et al., 2021).

Schrag et al. (2023) conducted PATHFINDER, a prospective feasibility study of MCED testing in asymptomatic adults. A cancer signal was detected in 1.4% of participants; diagnostic evaluation confirmed cancer in 0.5%.⁴ PATHFINDER addresses workflow and diagnostic resolution; it is not a substitute for randomized screening trials.

Treatment monitoring

Dawson et al. (2013) showed that ctDNA levels tracked metastatic breast cancer burden and could change before imaging in their cohort.⁵ Serial monitoring claims must cite the specific assay, cancer type, and endpoint used in each study.

MRD and recurrence risk

Stover et al. (2018) associated ctDNA tumor fraction with outcomes in metastatic triple-negative breast cancer.⁶ Post-surgical MRD performance varies by tumor type, assay limit of detection, and time point; do not generalize one cohort's accuracy to all solid tumors.

Tie et al. (2016) reported that circulating tumor DNA detected molecular residual disease and predicted recurrence after colorectal cancer resection.⁷ Tie et al. (2022) randomized stage II colon cancer to ctDNA-guided versus standard adjuvant chemotherapy in the DYNAMIC trial (15% vs 28% chemotherapy use).⁸ Kotani et al. (2023) linked postsurgical ctDNA positivity to higher recurrence risk (HR 10.0) in the GALAXY arm of CIRCULATE-Japan.⁹ MRD literature should be read per tumor type, assay panel, and time point after surgery.

Abbosh et al. (2017) tracked early ctDNA dynamics in early-stage NSCLC using TRACERx, linking variant allele frequency trajectories to relapse risk.¹⁰ Early detection and MRD are different intended uses; pooling their performance metrics obscures what replicates.

Clonal hematopoiesis and false positives

Not every somatic variant in plasma comes from the tumor under study. Razavi et al. (2019) showed that plasma cfDNA carries contributions from hematopoietic clones and other non-tumor sources.² Age, prior chemotherapy, and CHIP prevalence affect how you interpret mutation-only panels in screening versus MRD settings.

Assay Design Choices That Change Performance

Liquid biopsy programs fail when teams copy a published sensitivity number without matching the assay architecture. Key variables include:

• Targeted versus genome-wide: Mutation panels trade breadth for depth; methylation MCED tests use different signal biology than amplicon ctDNA panels.
• Limit of detection (LOD): MRD claims depend on variant allele frequency cutoffs; two papers reporting "ctDNA positive" may use different LODs.
• Pre-analytics: Collection tube, centrifugation delay, and freeze-thaw cycles shift yield (Alix-Panabières & Pantel, 2021).
• Reference standard: Case-control MCED validation differs from prospective screening feasibility (Klein et al., 2021 vs. Schrag et al., 2023).

When scoping literature in Motif, record assay name, analyte, cancer type, stage, LOD, and comparator for each association. That metadata explains why two PMIDs report different sensitivities for "the same" clinical question.

Regulatory and Analytical Context

FDA-cleared or approved plasma tests exist for specific mutations and companion diagnostic uses (for example EGFR in NSCLC). Each approval carries a defined intended use, specimen type, and analyte. Literature review should map which papers use the same platform and cutoff as your planned assay.

Pepe et al. (2001) describe phased evaluation of diagnostic tests from technical feasibility through impact on patient outcomes.¹¹ Liquid biopsy programs that skip phase distinctions often over-interpret feasibility studies as proof of clinical utility.

Choosing the Right Analyte for Your Question

Clinical question	Typical analyte	Evidence to prioritize
Screen asymptomatic adults	Methylation or multi-analyte MCED	Case-control validation plus prospective feasibility (Klein et al., 2021; Schrag et al., 2023)
Guide targeted therapy	ctDNA mutation panel	Concordance with tissue and FDA-labeled intended use
Monitor treatment response	Serial ctDNA allele fraction	Longitudinal cohorts with imaging comparators (Dawson et al., 2013)
Post-surgical MRD	Tumor-informed or fixed panels	Tumor-type-specific recurrence endpoints (Tie et al., 2016; Tie et al., 2022; Kotani et al., 2023)

Read our blog on liquid biopsy market analysis for investment and commercial context. For validation after literature triage, read our blog on biomarker discovery and validation to learn more.

Alix-Panabières and Pantel (2021) emphasize that pre-analytical variables (collection tube, processing time, storage) affect liquid biopsy results. DOI: 10.1038/s41571-021-00486-5. Protocols that skip these details often fail to replicate.

Scoping Liquid-Biopsy Evidence in Motif

1. Ask a focused question (e.g., ctDNA MRD after colorectal resection, or MCED screening feasibility)
2. Search PubMed, PMC, and Europe PMC; record screening counts in search provenance
3. Extract associations with assay name, cancer type, stage, sensitivity/specificity, and PMIDs
4. Separate discovery cohort papers from independent validation studies
5. Cross-reference genes and variants to ClinVar and related sources
6. Export cited tables for assay validation plans or protocol backgrounds

Failure modes:

• Attributing CCGA case-control sensitivity to PATHFINDER feasibility outcomes
• Pooling studies that used different LOD thresholds or mutation panels
• Citing MCED screening papers as proof of MRD performance
• Treating Motif exports as analytical validation data

For precision-medicine context across modalities, read our blog on personalized medicine biomarker analysis to learn more.

References

1. Wan, J.C.M., et al. (2017). Liquid biopsies come of age: towards implementation of circulating tumour DNA. Nature Reviews Cancer, 17(4), 223-238. PMID: 28233803
2. Razavi, P., et al. (2019). High-intensity sequencing reveals the sources of plasma circulating cell-free DNA variants. Nature Medicine, 25(12), 1928-1937. PMID: 31792460
3. Klein, E.A., et al. (2021). Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set. Annals of Oncology, 32(9), 1167-1177. PMID: 34176681
4. Schrag, D., et al. (2023). Blood-based tests for multicancer early detection (PATHFINDER): a prospective cohort study. The Lancet, 402(10409), 1251-1260. PMID: 37805216
5. Dawson, S.J., et al. (2013). Analysis of circulating tumor DNA to monitor metastatic breast cancer. New England Journal of Medicine, 368(13), 1199-1209. PMID: 23484797
6. Stover, D.G., et al. (2018). Association of cell-free DNA tumor fraction and somatic copy number alterations with survival in metastatic triple-negative breast cancer. Journal of Clinical Oncology, 36(6), 543-553. PMID: 29283787
7. Tie, J., et al. (2016). Circulating tumor DNA detects MRD and predicts recurrence in stage II colon cancer. Sci Transl Med, 8(346), 346ra77. PMID: 27384348
8. Tie, J., et al. (2022). ctDNA-guided adjuvant therapy in stage II colon cancer. NEJM, 386(24), 2261-2272. PMID: 35657320
9. Kotani, D., et al. (2023). Molecular residual disease and efficacy of adjuvant chemotherapy in CRC. Nat Med, 29(1), 127-134. PMID: 36646802
10. Abbosh, C., et al. (2017). Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution. Nature, 545(7655), 446-451. PMID: 28420412
11. Pepe, M.S., et al. (2001). Phases of biomarker development for early detection of cancer. Journal of the National Cancer Institute, 93(14), 1054-1061. PMID: 11459867
12. Alix-Panabières, C., & Pantel, K. (2021). Liquid biopsy: from discovery to clinical application. Nature Reviews Cancer, 21(6), 374-388. DOI: 10.1038/s41571-021-00486-5