BBI's Ivan Woo emphasizes a point on a poster summarizing the BARD1 paper during an event November 21 sponsored by the UW Medicine's Division of Medical Genetics.
[EDITOR'S NOTE: BBI’s Ivan Woo, a member of the Starita Lab, is the first author on a new preprint paper contributing to the understanding of how changes in the BARD1 gene affect cancer risk.]
What led you to choose this topic for the research?
An early memory of mine was going to the hospital to see my grandmother while she was being treated for breast cancer. While I was too young to understand, years later, learning that she had cancer helped spark fundamental questions about understanding how the cells that build us can be corrupted to cause disease.
Alongside this, as other family members became of age for cancer screening, they expressed great concern about how our family history of cancer might impact their own disease risk. These dual influences helped build my interest in understanding the genetics behind disease. In the Starita Lab, when I was given the opportunity to launch an independent project, BARD1, a gene associated with breast cancer, seemed like a clear choice due to my family’s experience with breast cancer.
What was the most surprising or unexpected finding from your research?
Protein translation begins with a start codon. If you disrupt the start codon, protein translation begins only after you reach another start codon in the mRNA transcript; this likely produces a truncated, non-functional protein product. However, with BARD1, we found that you can disrupt the start codon and functional protein is still produced. This occurs because after the first start codon, there is another start codon at position 26 that can begin translation of the transcript, rescuing function.
This is a very neat result. Using the start codon at position 26, amino acids 1–25 of BARD1 are missing, but the protein still functions, suggesting that those first 25 amino acids are dispensable. This reveals that for BARD1, genetic variants that introduce a premature stop codon before position 26 or disrupt the first start codon, are unlikely to contribute to disease risk.
The paper states: “Variant effect maps also powerfully inform genomic medicine.” How can clinicians use the paper’s conclusion in their work with patents?
Today, our ability to interpret genome sequencing results in the clinic is limited by variants of uncertain significance (VUS), genetic variants with an unclear impact on disease risk. Consequently, what frequently happens is a patient can get genetic testing, but this then returns a VUS—preventing their provider from using this sequencing information to inform their care. Ideally, if a genetic variant is identified after genetic testing, we already know if that variant is unlikely to contribute to disease (benign) or disease-causing (pathogenic).
For patients with identified pathogenic variants, their provider can guide them toward enhanced screening, preventative measures, or even personalized treatments. But all of this cannot happen without resolving VUS. With the BARD1 paper, the goal is to take a big chunk out of this VUS problem for BARD1. Working with the clinical genetics testing company, Ambry Genetics, in our paper, we were able to reclassify 95 percent of existing BARD1 VUS that have been cataloged by Ambry. This means that after our work has been peer-reviewed and officially published, Ambry will be able to use the data we’ve generated to return more informative genetic testing results to patients and their providers.
Next steps? What additional research do you believe should be conducted to build on the paper?
BARD1 has been linked to a handful of different cellular functions. In our paper, we were able to link BARD1’s role in error-free DNA repair to cancer risk as well as resolve the amino acids that appear to be required for this specific role. Currently, for breast cancer patients with BARD1-deficient tumors, treatment with Poly (ADP-ribose) Polymerase (PARP) inhibitors improves prognosis. Since we have teased out which specific residues of BARD1 are required for this role, additional work that directly investigates how we might be able to break these residues in cancer cells could be of therapeutic value.
Also, patients that have loss-of-function BARD1 variants have been traditionally ascribed an elevated risk for breast cancer. However, while preparing the manuscript, we found newer work linking BARD1 loss-of-function with elevated risk for neuroblastoma, a rare cancer of immature nerve cells that typically impacts young children. What is very interesting, is that BARD1’s risk associated with neuroblastoma seems to be significantly higher than the moderate risk associated with breast cancer. Consequently, there appears to be a difference in how loss-of-function BARD1 variants impact the different tissue types.
