Image courtesy of Pixaby
Scientists have enlisted the gene editing tool CRISPR in a hunt for cancer causing mutations, releasing into the open valuable data that could help doctors better advise their patients.
A new study lists almost 4,000 individual “misspellings,” or variants, in the “breast cancer gene” BRCA1 and how likely each one is to cause disease. The vast majority of variants—3,000 of them—are new to public databases containing genetic test results from people who had their BRCA1 sequenced. The work also tips the balance on hundreds of variants whose medical importance was unclear and that left thousands of women in limbo.
Most gene misspellings are harmless, but certain variants, like those in the BRCA1 gene and the related BRCA2 gene, can mean life or death. That’s because they greatly increase one’s chances of having breast and/or ovarian cancer. With this information, women may decide to undergo preventative surgery as famously publicized by Angelina Jolie and previously explained by Signals here. Men are affected too, with harmful BRCA1/2 variants also raising their risk of breast, pancreatic and prostate cancer.
BRCA1 is probably the most sequenced gene under the sun and genetic tests have revealed thousands of variants. These fall into three broad categories: pathogenic, or disease-causing variants (the most serious, but also the least common); benign, which have no ill effect on health (the most common ones); and, variants of uncertain significance, or VUS, for which there are not enough medical data to label them as either harmful or benign.
There are more than 2,500 VUS, according to the public database ClinVar. The VUS reportedly give physicians the most headaches because they do not know how to advise and reassure their patients who have just received a diagnosis of “I can’t tell you whether you’re going to get cancer or not.”
Despite the fact there is valuable genetic and medical data that could bring us closer to some answers, it isn’t in the public domain. Over the past 20 years, Myriad Genetics, the first company to offer the BRCA1 test, has collected data from 2 million people. In 2006, Myriad made this information proprietary and unavailable to cancer researchers – who aren’t too happy about it.
Which is partly why researchers from Washington University, led by Lea Starita and Jay Shendure, of the Brotman Baty Institute for Precision Medicine, decided to create all the possible BRCA1 variants themselves.
(To learn more about how BRCA1 works, read Sara Nolte’s Signals post here.)
To create BRCA1 variants, Gregory Findlay, a graduate student in the lab who spearheaded the work, changed every single DNA letter in parts of the gene that matter the most.
As is the case with most human genes, the DNA sequence of BRCA1 is split into protein-coding segments, which are separated by noncoding DNA. When a gene is switched on, its DNA is copied into a protein template from which the noncoding segments are excised. Findlay focused on the 13 coding segments, out of 22 total, and scrambled them in each possible way to create 3,893 variants—almost all possible variants.
Previously, researchers mainly studied the lab-made “processed” form of the gene, which contains only the coding parts. But with CRISPR, the Washington team was able make changes in the “organic” form of BRCA1, as it occurs in the genome, to approximate as much as possible the kinds of mutations that can appear in people.
They then put each variant into 20 million cells in a dish and let them grow for 11 days. Cells that have a functional BRCA1 protein would continue to divide and repair damage in their DNA that inevitably occurs with each round of cell division. On the other hand, faulty BRCA1 protein variants will cause the cells to die and there would be fewer cells left.
Strikingly, this simple experiment – did the cells with a specific BRCA1 variant live or die?— produced data for about 700 variants that accurately matched what’s known in real life. That is, variants that do not cause cancer also have no bearing on cell survival in the dish, whereas variants known to be harmful caused the cells to die.
Another 300 variants were previously in the VUS grey zone. Emboldened by how well their findings matched what’s already known, the researchers were able to label each VUS as either harmful or benign. They did the same for the 3,000 variants that are new—they have never before been seen in patients, at least not in those whose records are public (chances are Myriad has data on some of these variants).
The team made the findings publicly available to help physicians communicate their diagnosis when they encounter a patient with one of the above new variants. This matters because women who are told they carry a VUS frequently have a negative view of their test results and see themselves as being at a higher risk of cancer. In a small study, ten out of 19 participants had undergone preventative surgery compared to none of the five who interpreted the VUS as non-informative.
If disease-causing mutations could really be identified in the lab, and not just in patients, this could accelerate molecular diagnostics.
Researchers could for example find out ahead of time which variants are harmful, even before that variant was revealed in a genetic test. This prior knowledge would be especially powerful for rare variants, which exist in only a small number of people and where there’s understandably little medical data. It could open the door to medicine, or at least diagnostics, that is truly personalized.
Jovana Drinjakovic
Latest posts by Jovana Drinjakovic (see all)
- Canadian immunotherapy holds promise for patients with brain cancer - September 9, 2020
- Could stem cells be enlisted to battle COVID-19? - March 26, 2020
- Study reveals large differences at the molecular level between stem cells grown on different biomaterials - April 18, 2019
Comments