A CRISPR Approach for Accurately Classifying BRCA1 Variants

Alexander M. Castellino, PhD

September 13, 2018

Researchers used CRISPR editing technology to introduce mutations at every nucleotide in 13 exons of the BRCA1 gene (~4000 mutations) and then assigned each one a functional score that provides information that could be useful in assessing whether the mutation poses a clinical risk to a patient and that could influence clinical management.

The researchers assigned a function score to each single nucleotide variant (SNV) and checked for its clinical utility using known variants deposited in the ClinVar database.

SNVs with a function score of 0 (deemed "functional" in the assay) corresponded to benign annotations in ClinVar; those with a function score of -2 (deemed "nonfunctional") corresponded to a pathogenic annotation with a 98% specificity.

By introducing a mutation at nearly every nucleotide, the study has tested SNVs that have not yet been observed in patients. "As more and more patients undergo genomic sequencing, we believe more and more of these variants will be observed," Gregory M. Findlay, PhD, the lead author on the study, told Medscape Medical News.

"This is the first study of this magnitude and suggests that it is possible to determine with a high certainty if a variant is clinically actionable," he added.

The study was published online September 12 in Nature.

"Findlay and colleagues' approach represents a potential game-changer for assessing VUS [variants of uncertain significance]. But first, it will be crucial to collect further clinical data to validate the exciting findings of this paper," writes Stephen J. Chanock, MD, from the Division of Cancer Epidemiology and Genetics at the National Cancer Institute, Bethesda, Maryland, in an accompanying commentary.

Genetic Variants in BRCA1

BRCA1 and genetic variants in it have been extensively studied, and some variants are known to predispose individuals to breast or ovarian cancer. Some variants are classified as benign and pose no reason for concern; deleterious variants are associated with a high risk for cancer; VUS are those the significance of which are difficult to interpret.

These VUS "fundamentally limit the clinical utility of genetic information," the authors write.

"We focused on the BRCA1 gene because it has been extensively sequenced in people, and variants in it are important to interpret correctly," Findlay said.

A function score assigned to each variant allows clinicians to evaluate new BRCA1 variants that emerge in their clinical practice and may provide guidance on clinical management, Findlay explained to Medscape Medical News.

Saturation Genomic Editing of 13 Exons in BRCA1

The BRCA1 gene is required for homology-directed repair during DNA replication. By using a haploid human cell line that has one copy of each human gene, the cells require the BRCA1 gene to function.

Using CRISPR editing, the researchers accurately mutated each nucleotide in the 13 exons of the gene, one base at a time. This technique, called saturation genome editing, allowed mutations to be introduced in one exon at a time.

After the cells were grown in vitro for 11 days, the investigators used next-generation sequencing to determine the frequency at which each variant was present in the cell population.

Because BRCA1 is necessary for cell survival, cells with each variant either lived or died, Findlay explained. "By determining the frequency of the variants over time, we were able to assess the functional impact of each variant," he said.

A variant was scored as functional (suggesting that it is benign) (median score = 0) or nonfunctional (suggesting that it is pathogenic) (median score = -2). Most variants fell into either of these two classifications, Findlay explained. "The distribution of scores is continuous, with 0 and -2 being approximate medians that most function scores fall near to," Findlay said. In their study of 3893 BRCA1 SNVs, 21% scored as nonfunctional, 73% as functional, and only 6% scored as intermediate.

Findlay and colleagues then went to the ClinVar database, which houses known variants in the BRCA1 gene, to determine how they mapped on their scoring scale. Of 169 SNVs annotated as pathogenic in the database, 162 were designated as nonfunctional, two as functional, and five as intermediate.

Of 22 SNVs annotated as benign in the ClinVar database, 20 were designated as functional, one as nonfunctional, and one as intermediate.

Why the Study Is a Game Changer

Editorialist Chanock, who hailed the study as a game changer, commented: "VUS are currently piling up, because the rate at which new patient sequences for BRCA1 are being collected is greatly outstripping the accumulation of clinical information needed to classify variants."

Findlay indicated that their function score of each SNV along with its interpretation is freely available to all nonprofit cancer centers. "We are getting the data to clinicians who may benefit from it," he said.

"One of the reasons we are excited about this approach is that it promises to also work for a lot of other genes associated with cancer that have VUS," he said.

Before they can turn their attention to other genes, Findlay and colleagues will study other regions of the BRCA1 gene to identify loss-of-function mutations. They will then extend their study into BRCA2.

Not Yet Ready for Prime Time

Previous efforts have used computational approaches coupled with in vitro assays, but the pace of these approaches has been slow, Chanock notes. Once validated, this approach will be a major advance over previous efforts to study variants in the laboratory, he points out.

Chanock speculated that these findings will likely be incorporated to annotate BRCA1 variants that are a part of the International BRCA Challenge. However, he advocated taking a more cautious approach. "The backbone of genetic testing is the availability of sufficient clinical data to assign risk to a given variant," he said.

"The new assay should supplement, not supplant, these data. It might be tempting to make immediate use of the assay to interpret VUS identified during human genetic testing, particularly because SGE [saturation genome editing] has been used successfully in the past to identify targets for drug development. But in vitrodata alone should not be used as the basis for medical advice — at least until the approach has been clinically validated," Chanock opines.

However, Chanock believes that the approach of Findlay and colleagues can be applied to interrogate VUS for other genes among the 20,000 in the human genome, including P53 and RB — two tumor suppressor genes.

"But although developing these assays for all exons in cancer genes will take time and money, the dividends could be spectacular for cancer geneticists," Chanock notes.

The authors have disclosed no relevant financial relationships.

Nature. Published online September 12, 2018. Abstract, Commentary


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.