Skip to content

When you choose to publish with PLOS, your research makes an impact. Make your work accessible to all, without restrictions, and accelerate scientific discovery with options like preprints and published peer review that make your work more Open.


A BRCA Journey

Geneticist and science writer Ricki Lewis describes a personal journey relating to the BRCA1 “breast cancer gene”, to coincide with the publication of Jane Gitschier’s PLOS Genetics interview with Mary-Claire King – King’s work underlay the discovery of BRCA1. This story echoes another from Theo Bloom, which featured earlier this week in Biologue. Ricki Lewis writes the DNA Science Blog on the PLOS Blogs Network.

The day after PLOS Genetics asked me to blog in relation to Jane Gitschier’s interview with Mary-Claire King, my friend Lisa’s daughter Maya got her BRCA1 test results. It had been a confusing eight months.

When Dr. King’s group published their 1990 paper identifying the chromosomal home of the breast cancer susceptibility gene BRCA1, Maya was two years old. In each of her cells, one chromosome 17 bore a mutation in BRCA1, but she wouldn’t know that for 22 years.

In late 2012, Maya’s younger brother Justin sent a DNA sample to 23andMe “for fun”. Included in the company’s genotyping panel were the BRCA founder mutations, seen mostly among eastern European Jews (the Ashkenazim). Justin had one, 5382insC BRCA1: an insertion of a cytosine, disrupting the gene’s reading frame. Presumably, so did his mother, because she’s Jewish; her husband Eric not.

Lisa came to me, panicked, for an unofficial genetic counseling session. She talked as I sketched the pedigree. Like most families, a few people on each side had cancers, mostly older folks. When I blogged about that conversation in February, we assumed Lisa had passed the mutation to her son and possibly her daughter.

A whirlwind of emotions engulfed the family. Lisa was petrified, Justin confused, Eric disbelieving and Maya enraged.

23andMe confirmed Justin’s result and offered to test the rest of the family, for free. Lisa’s nurse practitioner, unfamiliar with BRCA testing, sent a blood sample to Myriad Genetics on my suggestion. Eric didn’t get tested, assuming Lisa was the one transmitting the mutation. But then Lisa’s test came back from Myriad, and she didn’t have the mutation.

By Courtesy of Tom Ellenberger, Washington University School of Medicine in St. Louis. [Public domain], via Wikimedia Commons
Had someone in Eric’s family tree converted from Judaism? Apparently not – 23andMe’s genotyping of Justin had included ancestry information, and he was half Ashkenazi, Lisa’s half. Had the son’s founder mutation arisen spontaneously? If so, then the father wouldn’t have a mutation and Maya needn’t worry. Eric finally took Myriad’s test, the family still thinking there’d been a lab error. There hadn’t been. Eric indeed had the mutation typically found in Ashkenazi Jews.

What, exactly, is the cancer risk to Eric, Justin, or Maya (if she had the mutation), given that they have an Ashkenazi mutation resident on a non-Ashkenazi chromosome 17? Their pedigree didn’t look like those from the early-onset families that pepper Mary-Claire King’s 1990 paper, which was festooned with the solid symbols indicating mutations in every generation, with an average age of onset under 45.

Since then, risk estimates for developing BRCA1 cancers have declined as studies included people of more diverse genetic backgrounds. A meta-analysis from 2003 considered 22 studies involving 8,139 cases unselected for family history, which for BRCA1 gave a 65% cumulative risk to age 70 of breast cancer and 39% of ovarian. The wide ranges – for breast 44%-78%, and for ovarian 18%-54% – suggested input from other genes. The highest lifetime risk published, 87% from 1994, reflects the early emphasis on high-risk families necessary to nail down the gene.

Might the Ashkenazi mutation in a non-Ashkenazi chromosome confer a lower risk, due to different surrounding gene variants? What might be the effect of those variants coming from Lisa’s chromosome 17? This was a cis/trans situation – would two genes that interact on the same chromosome do so if on different ones?

Clearly Myriad and 23andMe wondered this, because they requested more information. 23andMe’s algorithm came up with, for Justin’s 5382insC mutation, a “lifetime risk of breast cancer for women is increased from 12% to about 60% and risk of ovarian cancer is increased from less than 2% to about 40%”, and warns of elevated risk of breast cancer in men and prostate cancer. I don’t know whether that factored in the unusual circumstances.

But Maya didn’t care whether her risk was 87% or 60% or 40%; she just cared that it was pretty high. Reading Angelina Jolie’s “My Medical Choice” in the New York Times in May pushed her off her couch of indecision and she took the test, from Myriad, in July.

Last week, Maya learned she had not inherited the mutation. At lunch soon after, she was like a different person. All smiles, chatty. But the cloud fleetingly returned when I asked if she regretted her brother’s taking the test, and would she take another type of genetic test?

“Absolutely not,” she said, referring to taking another test of any kind. She wishes the entire 8 months of stress had never happened. And while acknowledging that the test could have saved her life, she resents a view of genetic testing as what her brother once called “fun”. More extensive, in-person genetic counseling might have dampened the family’s angst and helped in decision-making.

My friend’s family’s journey with BRCA testing illustrates many things: people’s reactions to genetic information are unpredictable; people have a right to choose not to know genetic information; and there’s a lot we still don’t know, such as specific gene-gene interactions. I hope that delivery of genetic test results can keep up with the coming demands from exome and genome sequencing.


The author has declared that no competing interests exist.

Leave a Reply

Your email address will not be published. Required fields are marked *

Add your ORCID here. (e.g. 0000-0002-7299-680X)

Back to top