Variant Allele Frequency in Cancer NGS Reports

What VAF measures, why percentages vary and what they cannot tell you alone

Key takeaways

• VAF means variant allele frequency or variant allele fraction in this report context.
• It measures supporting reads at a specific location, not the amount of cancer in the body.
• Tumor content, copy-number changes, tumor diversity and technical factors can influence the value.
• A value near 50% does not prove that a variant is inherited.
• Clinical significance depends on the exact alteration and evidence, not simply a higher or lower percentage.

What variant allele frequency measures

Variant allele frequency describes the proportion of usable sequence reads supporting a particular variant at a location. It is a laboratory measurement, not a direct percentage of cancer cells, a stage or proof of inherited risk.

NGS generates sequence reads from the material being analysed. At a particular location, some usable reads may support the reference sequence and others the reported variant. VAF expresses the proportion supporting that variant under the laboratory's analytic rules.

Term	Meaning in this discussion
Read depth	The number of usable sequence reads covering a location under the assay’s rules.
VAF / allele fraction	The proportion of those reads supporting the variant.
Tumor content	An estimate of tumor cells within a specimen; not identical to VAF.
Limit of detection	The validated detection capability under specified assay conditions.
Evidence tier	The clinical significance of a finding; not a category calculated from VAF alone.

A simple example of the calculation

Imagine that 100 usable reads cover a location and 20 support a particular variant. In this deliberately simplified example, the VAF is 20 divided by 100, multiplied by 100: 20%. This describes the reads being counted. It does not show that exactly 20% of cells are cancer cells.

Clinical assays apply quality filters, alignment rules and other processing steps; some also account for duplicate molecules. The figure is therefore an educational explanation, not a way to audit a clinical pipeline from a PDF alone. If a reported percentage appears inconsistent with the displayed counts, ask the issuing laboratory which denominator and filtering rules were used.^[1][2]

Why VAF is different from tumor content

A tissue specimen usually contains a mixture of tumor and non-tumor cells. Both can contribute DNA. A variant present in tumor cells may therefore be diluted by DNA from other cells. That is one reason a pathologist's tumor-content estimate and the sequencing allele fraction need not be the same.

Even in a tumor-rich specimen, VAF does not convert directly into a percentage of tumor cells. Copy-number changes and variation between tumor-cell groups affect the relationship.^[3] Oncquest's CGP 550+ catalogue calls for tumor content to be assessed by its pathologist; that specimen assessment is different from the sequencing allele percentage. Read both in their own context rather than doubling VAF to estimate tumor content.^[9]

Copy-number changes and tumor diversity

The number of gene copies can change in cancer. Extra copies or loss of a copy can affect the balance between variant and reference reads. A tumor can also contain groups of cells with different alterations, so a finding present in only one group may have a different allele fraction from another finding in the same specimen.

These possibilities explain why a high or low VAF is not automatically good or bad. They are contextual explanations, not diagnoses that can be assigned from the percentage alone. The laboratory and treating team interpret VAF with the other genomic findings, specimen assessment and technical limitations described in the report.^[2]

Why read depth and quality matter

A percentage based on very few reads carries different uncertainty from one supported by many reliable observations. But more reads do not, by themselves, correct every problem. Poor-quality material, sequence artefacts or difficulty analysing a region can still affect a result. Laboratories use validated quality criteria and reporting thresholds.

Read the assay's limitations and any comments specific to the sample. Average coverage is not the same as coverage at every individual position. If the relevant region was inadequately assessed, a negative result may not exclude a low-level variant there. The report should be interpreted within the test's validated performance, not against a universal depth number taken from another assay.^[1][2]

Low VAF and the limit of detection

The limit of detection describes what the assay can reliably detect under specified conditions. It is not identical for every method, specimen or alteration class. A laboratory may be able to detect some kinds of low-level variants more reliably than others.

A low VAF result is not automatically an error, and absence of a reported low-level variant is not proof that no such variant exists. Ask whether the finding meets the assay's validated reporting criteria and whether any additional confirmation is needed. Do not apply a 1%, 5% or other cutoff from a different test without understanding the method and clinical purpose.^[1][2]

Why 50% does not prove inherited origin

A heterozygous inherited variant may appear near 50% in an appropriate non-tumor sample, but a tumor-only result is more complicated. Tumor composition and genomic changes can produce overlapping allele fractions for somatic and germline findings. VAF therefore cannot reliably establish origin by itself.^[3]

When inherited risk is suspected, the clinical assessment considers the exact variant, gene, personal history, family history and relevant guidance. A validated germline test using an appropriate non-tumor specimen may be needed, with consent and genetic counselling. Relatives should not interpret their own risk from a percentage in someone else's tumor report.^[4]

Comparing results from different times or specimens

It is tempting to interpret a falling VAF as treatment success or a rising VAF as progression. That conclusion cannot safely be drawn from two isolated numbers. Tissue site, tumor content, specimen type, assay method and quality can all change between tests. A blood-based assay and a tissue panel also sample different biological material.

Some validated monitoring approaches use molecular measurements in specific settings, but a routine diagnostic CGP value should not automatically be repurposed as a monitoring test. Ask the treating team whether the results are comparable and whether they were obtained within a suitable monitoring strategy. Clinical assessment and other investigations remain important.^[5]

Putting the percentage back into context

• What is the exact variant, and what evidence supports its clinical significance?
• How many usable reads support the finding, and are there quality qualifications?
• What was the specimen's tumor content, and are copy-number changes relevant?
• Does the result meet the assay's reporting criteria?
• Is inherited origin a genuine concern requiring separate assessment?
• If compared with another result, are the specimen and method sufficiently comparable?

Verified Oncquest offering facts

1

Report measurement fields

Oncquest's CGP sample reports include allele-percentage or related measurement fields alongside other variant details.

Why it matters for you

Those fields need the gene, exact change, read depth and quality context to be meaningful. A bare number copied from the report cannot explain whether the finding is important, whether it is inherited or whether a medicine is suitable.

2

OncqSol CGP offerings

Oncquest's molecular diagnostics and gene-sequencing services include the tissue-based OncqSol CGP 270+ (SMO10765) and CGP 550+ (OMO1034) offerings.

Why it matters for you

In these reports, an allele percentage is one measurement within a wider interpretation. It needs to be read alongside the specimen assessment, exact alteration and laboratory comments, not as a standalone cancer score.

For a question about VAF on an Oncquest report, contact 0124 6650000 or life@oncquestlabs.com to ask how to obtain laboratory clarification. Have the report reference, test code and exact variant notation available. Ask which quality and specimen factors affect the percentage, then review its clinical significance with the oncologist. A larger number alone does not establish stage, prognosis or treatment benefit.^[7]

Frequently asked questions

FAQs

Does 20% VAF mean 20% cancer cells?

No. It means that about 20% of usable reads at that location support the variant under the assay’s rules. Cell mixture, gene-copy changes and tumor diversity prevent a universal conversion to cancer-cell percentage.

Is a high VAF always worse?

No. VAF is not a severity or stage scale. The clinical importance depends on the exact alteration and its context.

Does a VAF near 50% mean the variant is inherited?

No. Tumor-only values overlap between possible somatic and germline findings. Appropriate confirmation and genetic counselling may be needed.

Can a very low VAF still be real?

Yes, but interpretation depends on the assay’s validated detection limits and quality evidence. Ask whether the result meets reporting criteria and whether confirmation is appropriate.

Does a lower VAF after treatment prove response?

Not by itself. Different samples, tumor content and methods can change the percentage. Use only a clinically appropriate monitoring strategy and the treating team’s interpretation.

Is VAF the same as TMB?

No. VAF is a proportion of reads supporting a particular variant. TMB summarises mutation burden across a defined amount of assessed DNA.

Sources

1. Oncquest — CGP 270+ Sample Report Supplied SMO10765 example reviewed for structure and methodology only; individual findings are not reproduced.
2. Oncquest — CGP 550+ Sample Report Supplied OMO1034 example reviewed for structure and methodology only; individual findings are not reproduced.
3. AMP / ASCO / CAP — Somatic Variant Reporting Consensus Li et al., Journal of Molecular Diagnostics (2017); evidence tiers and interpretation principles.
4. National Cancer Institute — Genetic Testing Fact Sheet Inherited-risk testing, consent and genetic counselling.
5. National Cancer Institute — Biomarker Testing for Cancer Treatment General educational context on methods, interpretation and limitations.
6. Oncquest — Company Overview Oncology focus, specialised diagnostic disciplines and laboratory network.
7. Oncquest — Patient Services and Enquiries Test search, centre search, online report access and published enquiry contacts.
8. Oncquest — OncqSol CGP 270+ Official test code, NGS method and tissue specimen requirements.
9. Oncquest — OncqSol CGP 550+ Official test code, specimen requirements and pathology assessment.