Variants of Uncertain Significance: Navigating the Gray Zone of Genomic Interpretation

Every human genome contains millions of positions where the DNA sequence differs from a reference standard. The vast majority of these differences are harmless—natural variation accumulated over generations. But a growing subset of variants identified through clinical sequencing falls into a deeply uncomfortable middle ground: they are neither clearly benign nor clearly pathogenic. Scientists and clinicians refer to these as variants of uncertain significance, or VUS, and their proliferation represents one of the most consequential interpretive challenges in modern genomic medicine.

For a patient who has undergone genetic testing in search of answers—perhaps following a cancer diagnosis, a concerning family history, or an unexplained rare disease—receiving a VUS result can feel like no answer at all. The letter from the laboratory confirms that something unusual was found in the genome, but offers no clear guidance on what to do about it. That ambiguity, for many patients, is its own burden.

The Scale of the Problem

The challenge is not trivial in scope. As whole-exome and whole-genome sequencing have become more accessible and clinically routine across the United States, the volume of VUS findings has grown accordingly. A single whole-exome sequencing run may reveal dozens of variants that cannot be confidently classified. Databases such as ClinVar, maintained by the National Center for Biotechnology Information, contain hundreds of thousands of variants, and a substantial proportion remain in the uncertain category.

The root cause is a knowledge asymmetry. Sequencing technology has advanced at a pace that far exceeds the accumulation of functional and clinical evidence needed to interpret findings. A variant may alter a protein's amino acid sequence, but whether that alteration disrupts the protein's function in a meaningful way—and whether that disruption is sufficient to cause disease—requires extensive experimental and epidemiological data that often does not yet exist.

This gap is particularly pronounced for variants in genes associated with rare diseases, where affected populations are small and published literature is sparse. It is also acute in non-European populations, whose genomic diversity has historically been underrepresented in the reference datasets that underpin variant classification.

How Variants Get Classified

The American College of Medical Genetics and Genomics, along with the Association for Molecular Pathology, established a five-tier classification system that laboratories use to categorize variants: pathogenic, likely pathogenic, uncertain significance, likely benign, and benign. These designations are not permanent. As evidence accumulates, a VUS may be reclassified in either direction—upgraded to likely pathogenic or downgraded to likely benign.

Classification decisions draw on multiple lines of evidence. Population frequency data helps establish how common a variant is across large cohorts; a variant present in many apparently healthy individuals is less likely to be severely pathogenic. Computational tools predict the functional impact of amino acid changes based on evolutionary conservation and protein structure modeling. Functional assays—laboratory experiments that directly test a variant's effect on protein activity—provide more direct evidence but are labor-intensive and not available for most genes.

Co-segregation studies, which examine whether a variant tracks with disease across multiple family members, can be particularly informative, though they require access to extended family data that is not always obtainable. Case-control studies comparing variant frequency between affected and unaffected individuals offer population-level evidence, but again, rare variants in small patient populations make such analyses statistically challenging.

Closing the Knowledge Gap

Several research initiatives are working systematically to reduce the proportion of variants that remain unclassified. The ClinGen consortium, funded by the National Institutes of Health, coordinates expert panels that review evidence for specific genes and variants, establishing authoritative classifications that laboratories can adopt. These panels bring together disease specialists, laboratory scientists, and bioinformaticians to evaluate evidence rigorously and transparently.

High-throughput functional genomics approaches—sometimes called saturation genome editing or multiplexed assays of variant effect—are beginning to generate functional data at scale. These methods can test the impact of thousands of variants simultaneously, producing datasets that dramatically accelerate classification. Researchers have applied these approaches to genes such as BRCA1 and BRCA2, producing functional evidence for variants that had languished in uncertain status for years.

Artificial intelligence and machine learning models trained on large variant databases are also improving the accuracy of computational predictions. While no algorithm can substitute for experimental evidence, refined predictive models help prioritize which variants warrant urgent functional follow-up and which are more likely to prove benign.

What This Means for Patients

For individuals who have received a VUS finding, clinical guidance emphasizes that uncertain does not mean dangerous. Medical management decisions should not be based on a VUS alone; rather, clinical context—family history, personal health history, and the gene in question—should guide any follow-up. Genetic counselors play an indispensable role in this conversation, helping patients understand what a VUS does and does not imply.

Patients are also encouraged to remain engaged with the testing laboratory or their healthcare provider over time, because reclassification does occur. Some laboratories proactively notify ordering clinicians when a VUS is reclassified; others require that patients or providers request updated interpretations periodically. Advocacy groups representing specific disease communities sometimes maintain registries that track reclassifications and alert affected families.

Sharing genetic data with research databases, with appropriate consent, accelerates the evidence accumulation that enables reclassification. Participation in research registries is one concrete way that individuals carrying VUS findings can contribute to a collective effort that may ultimately benefit themselves and others.

Looking Forward

The VUS problem will not be resolved overnight, but the trajectory is encouraging. As sequencing datasets grow, as functional genomics methods mature, and as international data-sharing frameworks become more robust, the proportion of variants that can be confidently classified will increase. The scientific community's commitment to reducing diagnostic uncertainty reflects a broader principle central to the mission of genomic medicine: that the value of genetic information is only fully realized when it can be interpreted with confidence and communicated with clarity.

For now, the gray zone remains a real and significant feature of the genomic landscape. Acknowledging its existence—and the active effort required to navigate it—is the first step toward ensuring that patients who seek answers from their DNA receive the most accurate and actionable interpretation that science can currently provide.