1. First identify the test that was performed
Check the intended use, specimen, genes and regions, panel version, variant classes, coverage and detection limit. A SNP array, clinical panel, exome, pharmacogenomic result and tumour profile cannot be read through one template. A tumour finding is usually somatic; a blood finding can be germline, mosaic or related to clonal haematopoiesis—origin needs separate assessment.
2. Five classes are not five risk levels
Inherited sequence variants are commonly classified as pathogenic, likely pathogenic, variant of uncertain significance, likely benign or benign. These describe evidence that a variant causes a specific condition—not the probability that a person will inevitably become ill. Pathogenic and likely pathogenic classifications may support decisions alongside other evidence; benign variants are not considered causal for the stated condition.
3. A positive result is not inevitable disease
Meaning depends on purpose: a result may support a diagnosis, identify carrier status, elevated risk, a pharmacogenomic phenotype or a tumour biomarker. Predictive conditions require consideration of penetrance, age, sex, family history and observed features. A report often cannot predict exact onset, severity or course.
4. A VUS is a question, not an answer
A VUS means that evidence is insufficient or conflicting. ACMG/AMP guidance states that a VUS should not be used in clinical decision-making. It does not confirm a syndrome and is not a basis for preventive surgery, medication changes or predictive cascade testing in healthy relatives. Phenotype, family segregation, functional evidence and later reclassification may sometimes clarify it.
5. A negative result may be informative or uninformative
When a familial pathogenic variant is known and reliably absent in a relative, the result may be informative. If the familial cause is unknown, ‘nothing found’ does not rule out disease: a variant may fall outside coverage, belong to another class, have a low allele fraction, lie in a gene not yet known or escape current interpretation. Care based on personal and family history does not automatically stop.
6. Check how the variant is recorded
A sequence finding should state the gene, transcript, HGVS description, zygosity, genome build, classification, applied criteria and interpretation date. A CNV needs coordinates and copy number; a fusion needs an exact assay-resolvable event; pharmacogenomics needs the alleles tested and phenotype-translation rules. A label such as ‘BRCA+’ or ‘risk gene’ is insufficient.
7. Confirmation and reanalysis are different
Confirmation verifies an important finding with a validated method and appropriate specimen, especially when the original test was consumer-facing or could change care. Reanalysis revisits existing data after databases, criteria or clinical information change. It does not fix physical blind spots in the original assay; sometimes a new test is required.
8. Seven questions after receiving a report
What exactly was found and by which method? For which relationship was the variant classified? Does the result fit the clinical or family question? Is confirmation needed? What was not examined? Could interpretation change over time? Which next step is guideline-supported—and is any action needed at all? These answers matter more than the generic word ‘positive’.