# Blood-cell variants in liquid biopsy

Source: https://onco.cc/terms/blood-cell-variants-liquid-biopsy/  
OnCo record `blood-cell-variants-liquid-biopsy` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

A DNA change found in plasma may come from blood cells rather than a tumor. Establishing where a change came from is part of interpreting a liquid biopsy.

## Summary

A primary study compared plasma cell-free DNA, matched white blood cells and tumor samples and found substantial contributions from clonal hematopoiesis to plasma variants. Matched blood-cell sequencing helped attribute origin. A technically real somatic variant is not necessarily tumor-derived, and origin cannot be assigned from the gene name alone. This measurement problem is distinct from assessing the future health implications of a blood-cell clone.

## Fields

- Kind: Term
- Last checked: 2026-11-10
- Also known as: clonal hematopoiesis and plasma variant origin; white blood cell sequencing for cfDNA interpretation

## Notes

- How common it is: in paired plasma and white-blood-cell sequencing, 53.2% of cfDNA mutations in 124 people with metastatic cancer and 81.6% in 47 people without cancer had features of clonal haematopoiesis, and variants that could not be separated from white-blood-cell DNA kept an explicit ambiguous category (Razavi 2019). Blood-cell clones also fall into gene groups: DNMT3A, TET2 and ASXL1 with age (Jaiswal 2014) and TP53, PPM1D and CHEK2 selected by radiation, platinum and topoisomerase II inhibitors (Bolton 2020). A gene often mutated in tumours is therefore not proof of tumour origin.
- Attribution rule: matched white-blood-cell sequencing is the standard comparison, but its gene coverage, depth and detection rules matter, a clone below its sensitivity can stay unseen, and subtracting every blood match is not a complete method because a variant can be present in blood and tumour. In a fictional case the same variant V is detected in plasma and blood cells and not in the sampled tumour. That supports a blood-cell contribution, and tumour non-detection is not a biological zero because it is bounded by coverage, sensitivity, timing and the region sampled.

## Sources

- Matched plasma, blood-cell and tumor sequencing study (2019): https://doi.org/10.1038/s41591-019-0652-7
- Cancer therapy shapes the fitness landscape of clonal hematopoiesis: https://doi.org/10.1038/s41588-020-00710-0
- Age-Related Clonal Hematopoiesis Associated with Adverse Outcomes: https://doi.org/10.1056/NEJMoa1408617

## Connected records

- terms: [Analytical versus clinical validation](https://onco.cc/terms/analytical-vs-clinical-validation/), [Circulating tumour DNA (ctDNA)](https://onco.cc/terms/ctdna/), [Evidence from measurement to patient benefit](https://onco.cc/terms/measurement-to-benefit-evidence/), [Variant allele frequency (VAF)](https://onco.cc/terms/vaf/)
- technologies: [Liquid biopsy (ctDNA)](https://onco.cc/technologies/liquid-biopsy/)
- pathways: [Clonal haematopoiesis (CHIP)](https://onco.cc/pathways/clonal-haematopoiesis/)

---
JSON: https://onco.cc/api/v1/entities/blood-cell-variants-liquid-biopsy.json