Large numbers of individuals required to classify and define risk for a rare VUS in known cancer risk genes. B. H. Shirts1, A. G. Jacobsen1, G. P. Jarvik2, 3, B. L. Browning2 1) Laboratory Medicine, University of Washington, Seattle, WA; 2) Medical Genetics, University of Washington, Seattle, WA; 3) Genome Sciences, University of Washington, Seattle WA.

   Background: Up to half of unique genetic variants in genomic evaluations of familial cancer risk are rare variants of uncertain significance (VUS). Population studies indicate that as genomics enters the clinic additional rare variants will continuously be identified and that classification of rare VUS will be an ongoing issue. Methods: We modified standard power calculations to explore sample sizes necessary to classify and estimate relative risk for breast cancer and colon cancer variants responsible for varying levels of relative cancer risk. We required 80% power and tolerated a 10% false positive rate, since most variants clinically tested will be in known genes with high pretest probability of being deleterious based on in-silico results. We specifically evaluated variant population frequencies of 0.1%, 0.01% and 0.001%. Population-based studies were assumed to have equal numbers of cases and controls, and family-based studies were assumed to have equal proportions of first and second-degree relatives of carriers of the rare variant in question. We examined relative risk of breast cancer ranging from from 12 to 3 and relative risk of colon cancer ranging from 20 to 5. Results: Population-based samples necessary for classification of rare breast cancer variants ranged from 1,044 to 997,183 depending on relative risk and population frequency. Samples necessary for classification of rare colon cancer variants ranged from 661 to 425,633. The larger samples were necessary for rarer and less penetrant variants. Family based relative risk methods were robust to changes in variant population frequency and required at between 10 and 150 individuals, depending on penetrance. Even variants with moderately rare frequency (0.01%) and six-fold breast cancer risk required large population based samples or family based samples (27,387 and 31 respectively), samples larger than those used most large research studies. Conclusion: While the classification of the pathogenicity and associated risk of disease from variants is clinically important, we demonstrate that it may be unlikely that most rare missense variants will be classifiable in the near future and that accurate relative risk estimates may never be available for very rare variants. This knowledge may alter strategies for communicating information about variants of uncertain significance to patients.

You may contact the first author (during and after the meeting) at