Update on BRCA1/2: Technical and Policy Questions Arise as Testing Options Expand

As one of the most significant predictors of hereditary breast and ovarian cancer, the BRCA1/2 genes have become the poster child for genetic testing. In the past 18 months, the floodgates for testing options have opened, as companies seek to enter the diagnostic market in the wake of the US Supreme Court’s June 2013 ruling that “naturally occurring” human genes are a “product of nature” and cannot be patented, breaking Myriad Genetics’ monopoly on BRCA1/2 gene testing.

Yet even as BRCA1/2 testing becomes more widely available in oncology practice, fresh questions about interpreting the results of these assays and a debate about which individuals should be screened for mutations have arisen. Genetic alterations are central to the development of cancer and, though most are acquired over a person’s lifetime, between 5% and 10% are inherited germline mutations that can increase the lifetime risk of particular cancer types. Mutations associated with more than 50 hereditary cancer syndromes have been identified to date, and testing for such mutations can help to assess an individual’s cancer risk.

Thousands of BRCA1/2 mutations have been identified to date throughout the coding regions of these genes. BRCA1/2 alterations are generally classed as pathogenic on the basis of observed frequencies in cases of cancer compared with healthy controls. Though sporadic BRCA1/2 mutations are thought to be relatively rare, inherited germline mutations are more common. Overall, the prevalence of pathogenic BRCA 1/2 mutation carriers in the general population is estimated to be between 1 out of every 400 persons and 1 out of 800 persons. There are a number of founder mutations, which are high-frequency mutations that are particular to a specific population such as the 185delAG, 5382insC, and 6174T mutations, that are found at a frequency of 1 out of 40 persons in the Ashkenazi Jewish population. BRCA1/2 mutations are autosomal dominant, meaning an individual only needs to inherit a single defective copy of the gene from a parent to have an increased cancer risk.

BRCA1/2 mutations are particularly significantly associated with increased risk of breast and ovarian cancer. A genetic link between chromosome 17 (where BRCA1 resides) and breast cancer risk was first uncovered in the 1990s, and no predictor of risk of developing these cancer types has proved to be as powerful.

Compared with the general population, in whom the currently estimated lifetime risk of breast and ovarian cancer are 12% and 1.4%, respectively, individuals with inherited BRCA1/2 mutations have a substantially increased risk; by age 70, the risk of breast cancer is estimated to be 55% to 65% for BRCA1 carriers and 45% for BRCA2 carriers, whereas the risk of ovarian cancer is 39% for BRCA1 carriers and 11% to 17% for BRCA2 carriers.

Studies have shown that BRCA1/2 mutations are associated with increased risk of several other cancer types as well, with pancreatic and prostate cancers being the best studied. Both cancer types display a lower frequency of familial BRCA1/2 mutations than breast or ovarian cancers, and appear to be more closely associated with BRCA2.

More recently, studies have implicated BRCA2 in the development of lung cancer. Using four genomewide association data sets of more than 11,000 individuals of European descent, researchers found that BRCA2 mutations significantly increased the risk of non—small cell lung cancer, specifically squamous cell carcinoma, particularly among smokers in whom the risk may be doubled.

Testing Options Multiply

The association between breast cancer risk and BRCA1/2 mutations fueled a race to develop a means to identify mutation carriers at increased risk of breast and ovarian cancer. The victor was Myriad Genetics, a Utah-based company that subsequently patented the genes and began to offer commercial sequencing tests. Its monopoly on BRCA1/2 gene testing lasted close to 15 years, but came to an end in 2013 with a landmark ruling from the US Supreme Court. Historically, patents have been granted on genes, with some estimates suggesting that as much as 20% of the human genome may be noted on patents.

However, many believed that patenting genes was unconstitutional and invalid. In 2009, the American Civil Liberties Union and Public Patent Foundation filed a lawsuit against Myriad Genetics on behalf of researchers, genetic counselors, patients, and others to argue that human genes were products of nature and should not be patented. The Supreme Court ruled unanimously in favor of this position, and the door was opened for other companies to begin offering BRCA1/2 testing.

The controversy hasn’t ended there, however. Myriad has since filed a number of lawsuits against companies for patent infringement, contending that the Supreme Court only removed five claims out of its patents and that it still has more than 500 remaining claims across 24 patents. While it remains to be seen whether any of these lawsuits hold water, some believe they have succeeded in scaring away smaller competitors. One example is Gene By Gene, which settled with Myriad and is only able to market its BRCA1/2 tests outside North America.

Regardless of the legal fracas surrounding BRCA1/2 testing, the development of newer sequencing technologies and the emergence of other BRCA1/2 testing companies have created a broader range of testing options.

Myriad’s original BRCA1/2 tests used Sanger sequencing, which is limited in its ability to detect all variants within a gene. They and other companies offering BRCA1/2 testing are now making use of next-generation sequencing and other state-of-the-art techniques that are able to deliver faster and more accurate results at a lower cost and have the potential to identify all BRCA1/2 variants.

Although the BRCA1/2 genes are the most strongly associated with breast and ovarian cancer risk, there are many other genes that convey increased susceptibility to these and other tumor types, which may also be present. As such, companies are now beginning to offer rationally designed screens that look for a range of different genetic variants.

In an effort to maintain its competitive edge, Myriad Genetics also has numerous novel testing platforms under development. Among them is myChoiceHRD, which will offer a DNA-based readout of the degree to which a tumor has lost the ability to repair damaged DNA—the HRD stands for homologous repair deficiency—which includes analysis of BRCA1/2 genes in addition to other genes involved in DNA repair.

Applications of BRCA1/2 Testing

Regardless of which test is performed, the principal application is the same: a prognostic tool for patients with family or personal history of breast and ovarian cancer.

Currently, it is recommended that women with family members who have had breast or ovarian cancer talk with their healthcare provider and undergo genetic risk assessment and counseling to identify those at increased risk of hereditary breast and ovarian cancers.

A number of mathematical and predictive models have been developed to assist; the most commonly used include BOADICEA, BRCAPRO, and Myriad I and II. Individuals who screen positive at this stage may then be recommended for BRCA1/2 testing. A BRCA1/2 test generally provides one of three results:

• Positive—The individual has a BRCA1/2 mutation with a known association with cancer
• Negative—The individual has no known BRCA1/2 mutation
• Variant of unknown significance (VUS)— An abnormality was found, but it has not been evaluated for its potential association with cancer risk

A positive BRCA1/2 test result has several potential therapeutic implications. Patients without cancer can choose to be closely monitored for early detection or to undergo bilateral prophylactic mastectomy or salpingo-oophorectomy. The latter has been shown to reduce the risk of breast and ovarian cancer and improve overall survival.

Meanwhile, in patients who already have cancer or eventually develop cancer, having a BRCA1/2 mutation can impact treatment efficacy. BRCA1/2 mutant cancers have been shown to be exquisitely sensitive to DNA-damaging therapeutic agents as a result of synthetic lethality—the concept derived from yeast genetics that a single mutation in either of two genes is insufficient to cause cell death, but a combination of both mutations is lethal. Under normal circumstances, backup DNA repair pathways can repair the damage induced by DNA-damaging chemotherapeutics (eg, platinum-based) and radiation therapy, but the double hit in cells that also have a defective BRCA1/2 pathway overwhelms the cancer cell. This concept has also driven the development of molecularly targeted therapies against the poly(ADP)ribose polymerase 1 (PARP1) enzyme, which repairs single-strand breaks in DNA, known as PARP inhibitors.

BRCA1/2 testing companies are now also creating tests that aid in the development of these therapeutic strategies. In a study presented at the 2014 ASCO meeting, MyChoiceHRD accurately predicted response to cisplatin and paclitaxel in triple-negative breast cancer (TNBC).

The Challenge of Interpretation

Ideally, BRCA1/2 testing produces a definitive “yes” or “no” result, but in reality the outcome is significantly more ambiguous.

The implications of a negative result, for example, depend upon the kind of test that is used and the family history available. For comprehensive analyses in which the whole genome is screened, a negative result means an individual has no known pathogenic mutation.

However, if single-site analysis is used, then although a specific heritable mutation can be ruled out, other mutations cannot. Furthermore, a result can really only be classed as a “true” negative if the individual being tested has relatives with known BRCA1/2 mutations. If no relatives have been tested or no mutations have been detected in tested relatives, then the result is seen as an “uninformative” negative.

BRCA1/2 testing is also clouded by a third type of result, the identification of VUS, which can complicate clinical management since it is unclear what the associated cancer risk may be. Reducing the VUS rate would vastly improve the utility of genetic testing, and large collaborative research efforts are under way to catalogue and evaluate unknown variants.

The ultimate goal is to determine the potential pathogenicity of these variants using a variety of different methods. The International Agency for Cancer Research convened a panel of experts who developed a five-tiered classification scheme for the probability of pathogenicity.

Reported VUS rates vary between BRCA1/2 testing companies; while newer companies have higher reported rates (4%-6%), many years of experience have allowed Myriad to accumulate a large proprietary repository of genetic information that, the company contends, affords its tests an extremely low rate (0.6% and 1.6% for BRCA1 and BRCA2, respectively).

Myriad stopped publicly sharing data in 2004 and has received much criticism for it. In response, numerous groups have organized public databases to expand the pool of publicly available genetic information.

The Free the Data movement, which is gaining significant popularity among patient groups, established the ClinVar database, whereas the National Institutes of Health has developed the Breast Cancer Information Core, among others.

In September, four prominent academic centers established the Prospective Registry of Multiplex Testing (PROMPT), an online registry for individuals and families who have undergone testing for inherited cancer-causing genetic mutations. The founding institutions are the Dana-Farber Cancer Institute, Mayo Clinic, Memorial Sloan Kettering Cancer Center, and Penn Medicine; commercial test providers, including Myriad Genetics, have agreed to offer information about the PROMPT registry to patients and providers.

Whom to Test?

A number of organizations have generated clinical guidelines for genetic testing, with the most widely followed being those generated by the National Comprehensive Cancer Network and the US Preventive Services Task Force.

Both organizations recommend that only individuals with a family history of breast or ovarian cancer undergo BRCA1/2 testing, while individuals without family history are not encouraged to undergo testing unless they have breast or ovarian cancer. Indeed, Medicare and many private medical insurers don’t cover testing outside this protocol.

Mary-Claire King, PhD, who discovered the genetic link between breast cancer and BRCA1/2 mutations, has become a vocal advocate for universal BRCA1/2 testing among all women over the age of 30 years. King is a professor of Medical Genetics and Genome Sciences at the University of Washington and recently was honored with the 2014 Lasker- Koshland Special Achievement Award in Medical Science.

She cites the growing number of studies showing that pathogenic BRCA1/2 mutations frequently occur in a significant proportion of women who do not have a strong family history of relevant cancers.

This includes a study in which more than 8000 healthy Ashkenazi Jewish men aged 30 years and older were screened for one of three BRCA1/2 mutations. The female relatives of the 175 men who were found to harbor one of these mutations were then tested, and 211 carriers were identified.

Almost half of these women reported little or no family history of relevant cancer. Other studies have shown that assessment of family history through the use of models such as BRCAPRO, as recommended by current guidelines, does not effectively target testing to high-risk patient populations. Not everyone supports this position, however. Other prominent researchers urge more caution, arguing that there are a plethora of other genes involved in breast and ovarian cancer risk and that there is still ambiguity surrounding testing results, which could lead to many women undergoing unnecessary testing and prophylactic procedures. They also argue that the study in question looked at individuals with a high prevalence of BRCA1/2 mutations and may not translate into the general population.