Science Editor

Harboring a mutation in her BRCA1 or BRCA2 genes sharply increases a woman's risk of breast and ovarian cancer. Unless it doesn't.

The problem with extrapolating from sequence to risk is a familiar one: "There are so many variants of BRCA1 and BRCA2," Shyam Sharan told BioWorld Today. Some definitely raise a woman's risk; others are known to be harmless. But when a woman gets her BRCA genes sequenced to determine her personal risk, perhaps after a family member has developed breast cancer, "the problem comes in 10-13 percent of cases, when they find variants of unknown clinical significance."

Such variants of unknown significance present an obvious dilemma for genetic counselors - and with 200,000 women having their BRCA1 and 2 genes sequenced to date, and the number constantly increasing, by now 800 of them have been cataloged for BRCA1 and 1,100 for BRCA2.

In the Sept. 21, 2009, issue of the Journal of Clinical Investigation, Sharan and his colleagues from the National Cancer Institute reported a new method that can help to determine whether mutations in BRCA1 are likely to be harmless, or warrant preventive steps.

The paper follows the 2008 publication of a similar assay for the BRCA2 gene out of Sharan's lab. An accompanying commentary in JCI called the new method "an elegant approach to evaluating . . . BRCA1 missense alleles" and said it has the potential to allow "the power of molecular biology [to] be harnessed to separate the harmful variants from the harmless, allowing both patients and physicians to make appropriate clinical decisions."

In their study, Sharan and his colleagues took advantage of the fact that the BRCA1 gene is crucial for the survival of cultured mouse embryonic stem cells.

The authors knocked out one copy of the BRCA1 gene in a mouse embryonic stem cell line, and made the other copy conditional. In cell culture, they then knocked out the second copy and simultaneously replaced it with the normal human BRCA1 gene, or with one of several mutations - some harmless, some known to raise breast cancer risk and some variants of unknown significance.

Sharan and his team first tested several variants for outright lethality to mouse embryonic stem cells, and found that while several variants that do not increase the risk for breast cancer rescued up to 50 percent of stem cells, there were no survivors among the cells whose own BRCA1 copy was replaced by variants that raise cancer risk.

The authors then went on to test whether they could find more subtle effects of BRCA1 mutations with their method as well. Even if a BRCA1 variant is able to rescue stem cells, Sharan explained, "that does not mean that the BRCA1 is perfectly normal," because functions may be affected that, although they are not critical to survival, do impair cell functioning.

The authors challenged rescued cells with different DNA damaging agents, and found that it was possible to identify variants that do not kill stem cells outright, but do have harmful effects.

In total, Sharan and his team reported testing 13 variants, including five that are known to be harmful, one that is known to be harmless and seven variants of unknown significance.

Seven of the variants, including all of the known harmful variants, were unable to rescue mouse embryonic stem cells whose own copy of BRCA1 was missing. Six variants rescued from 10 percent to 50 percent of embryonic stem cells; of those, three, all of them variants of unknown significance, had some negative effects on the stem cells during more detailed testing.

The work has obvious practical implications - but also an obvious stumbling block: Though under challenge, the patents for BRCA1 and 2 genes currently are held by Myriad Genetics. (See BioWorld Today, May 14, 2009.)

Sharan acknowledged that the IP situation might be a future stumbling block. Nevertheless, he noted, the National Cancer Institutes are "very interested" in licensing and finding industrial partners for further development. And as for the IP situation, he said, "We have to clear a lot of hurdles before we even get to that."