VAF-tumor content graph: a simple visual framework for interpreting hereditary cancer variants and supporting genetic counseling in tumor-only sequencing.

Graphical plot of variant allele frequency against tumour content, with theoretical reference lines derived from the Knudson two-hit model, to discriminate germline from somatic origin.

Comprehensive genomic profiling by tumor-only sequencing detects pathogenic or likely pathogenic variants in hereditary cancer susceptibility genes whose germline or somatic origin is hard to establish, complicating communication in clinic. The authors propose a variant allele frequency-tumour content graph incorporating theoretical reference lines based on the Knudson two-hit model, applied retrospectively to patients who had both tumor-only and paired tumour-normal panels between 2018 and 2025, plotting variants recommended for disclosure by the institutional expert panel. Among 103 patients, 35 had confirmed germline pathogenic or likely pathogenic variants; among BRCA1/2 variants, loss of heterozygosity was seen in 12 of 22 hereditary breast and ovarian cancer-associated tumours and in 2 of 5 non-associated tumours, and among other susceptibility genes 4 of 8 cases carried two variants distributed along the theoretical germline and somatic lines. Overall, 18 of 35 cases (51%) showed patterns consistent with the two-hit model, and in one patient with mismatch repair deficiency and two with POLE variants, multiple variants clustered along the somatic line.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The graph formalises reasoning molecular biologists already perform mentally, and its value is chiefly pedagogical: it helps explain in tumour board, or to the patient, why a high variant allele frequency does not prove germline origin. But 51% of patterns consistent with the two-hit model means that one time in two the graph does not settle the question, a limitation the authors underplay: it never removes the need for germline confirmation on a blood sample. No diagnostic performance is quantified, neither sensitivity nor predictive value, which rules out using it to filter germline testing indications.

Analysis by Dr Thibaut Benquey