Genetic Testing in Breast Cancer Offers Much Information but Poses Challenges in Interpretation

Get Permission

Louise E. Morrell, MD

You will gain information if you do a panel test, but not all findings will be actionable. Be prepared for variants of unknown significance, which average one or two per test.

—Louise E. Morrell, MD

For breast cancer patients with robust family histories, medical oncologists should be testing not only for BRCA1/2 mutations, but also for large duplications and deletions as well as for PALB2 mutations.

“These [findings] have proven utility in testing breast cancer patients,” said Louise E. Morrell, MD, Medical Director of the Lynn Cancer Institute, Boca Raton, Florida, in an update on germline testing for breast cancer at the 2015 New Orleans Summer Cancer Meeting.1

Large Duplications and Deletions in BRCA

Comprehensive BRCA testing includes complete sequencing of the BRCA1 and BRCA2 genes and an additional procedure to identify five common large rearrangements in the BRCA1 gene. BART—the ­BRACAnalysis Large Rearrangement Test—was introduced in 2006 and detects additional rare large genomic rearrangements in both BRCA1 and BRCA2. About 1% to 2% of individuals who meet the family history criteria for BRCA gene testing will have a mutation detected by the test. The guidelines of the National Comprehensive Cancer Network (NCCN) now recommend large rearrangement deletion and duplication testing to be included in BRCA gene testing.

“The original BRCA analysis [routine Sanger sequencing] looked for sequence errors, which is like going through sentences and finding errors. But if you cut out a whole paragraph, sequencing doesn’t find the error. This is what large-deletion analysis finds,” she explained.

In very high-risk families, 9% of all mutations are due to large duplications and deletions. This additional testing identifies an additional 2.3% of mutations in such families, though far less among patients with more modest risk (those with triple-negative disease or only one affected family member).

“This means that for patients with a [considerable] family history who are looking for an explanation for the many cancers [in their families], there is a 2.3% chance you will find a mutation testing for large rearrangement delection duplication when tests were negative before,” she said.

Beyond BRCA

“Today, oncologists will be testing for many abnormalities beyond BRCA, one being PALB2,” she said. While 10 times less frequent than BRCA, it is a genetically important factor.

PALB2 is a partner and localizer of BRCA2 and is part of the homologous DNA recombination-repair pathway. PALB2 accounts for 2.4% of familial aggregates of breast cancer and renders a 33% lifetime risk in carriers. This risk increases to 58% if the person has more than two affected first-degree relatives.

“We have learned that family history matters a lot with regard to risk associated with this mutation,” Dr. Morrelll said. “We now know that these genes and those we discover in the future should not be viewed in isolation. They need an accompanying familial phenotype to be most damaging.”

Information from 17,000 BRCA mutation carriers has been evaluated by the Consortium of Investigators of Modifiers of BRCA1/2 (CIMBA). This has led to the identification of risk modifiers for these mutation carriers, which enhances the accuracy of risk prediction for the individual patient.

“We are realizing that not all patients are the same,” she said. “This may have added value to the patient who has the mutation but doesn’t want her ovaries out yet.”

Understanding Panel Testing

Panel testing with next-generation sequencing is now offered by more than a dozen laboratories. Data from panel testing are being published, confirming the challenges with estimates of cancer risk and the problematic issues of variants of unknown significance and incidental findings. These observations appear relatively consistent across laboratories and institutions, she said.

Panels should always include the high-risk and intermediate-risk genes. High-risk alleles are rare-to–very rare and impose a very high relative risk (up to 10-fold) for breast cancer. In this category are TP53, BRCA1, BRCA2, PTEN, CDH1, and STK11. Intermediate-risk alleles are rare with essentially a two- to fourfold increase in the risk of cancer. They include BRIP1, ATM, PALB2, and CHEK2.

Panel testing reports “actionable findings,” which are findings that could “plausibly lead to a specific medical recommendation or intervention,” she said. It does not mean that taking such action has demonstrated clinical utility, such as improvement in survival.

Panels evaluate 10 times the number of genes as single-gene testing, but they also increase the number of incidental findings and greatly multiply the number of variants of unknown significance. A variant of unknown significance is an alteration that reflects only a minor change along a gene that may be 10,000 to 15,000 nucleotides long. For example, in the sentence, “The big red dog ran out,” a variant of unknown significance might cause it to read, “The big ned dog ran out,” she explained.

“Based on this minor change, one would not necessarily expect a variant of unknown significance to have significant health consequences, but the variants of unknown significance classification means that we do not yet know,” she said. “Determining the significance of variants is proving to be a major challenge to interpreting multiplex panel testing.”

The clinician can try to make sense of a variant of unknown significance by looking at its location within a gene and its protein functionality, cosegregation with disease in affected families, prevalence in a control population, and other aspects. Still, confusion reigns with variants of unknown significance, “and this is where we must dedicate considerable time, expertise, and resources in interpreting tests for our patients and families: the interpretaton is very challenging,” she remarked.

Other issues with panel testing create challenges for clinicians: lack of publication of the results of validation studies, lack of standardization for validation among labs, and unknown accuracy of panel tests. Labs can also differ in other ways: how they report results, their “transparency,” depth of coverage with next-generation sequencing, detection of large-deletion duplication, methods of defining variants of unknown significance, genes tested, and payment approaches.

Differences in interpretation may be the most confusing to clinicians and patients. “A patient can have a test at one lab, and a gene is called a mutation. Her sister can use a different lab, get the same finding, and it’s not called a mutation. We are not always getting the same interpretation of the same finding,” Dr. Morrell said.

When to Consider Panel Testing

“When should we test for a panel of cancer genes? That’s a question that is not yet answered,” she indicated.

Dr. Morrell said she opts for panel testing when the results can help families understand their strong family history and when the results are likely to be actionable. “Mostly, panel testing helps future generations,” she added.

She noted that 10% of breast cancer patients with a strong family history who undergo panel testing will have a deleterious mutation, of which about 6% will be due to BRCA1 or BRCA2 and 4% due to genes other than BRCA1/2. The panel test should include PALB2, which carries a lifetime risk of about 33% to 58%. For estrogen receptor–positive breast cancer patients, CHEK2, ATM, and TP53 should be included in the panel.

“You will gain information if you do a panel test, but not all findings will be actionable. Be prepared for variants of unknown significance, which can average one or two per test,” she said. Despite these issues, she added, “There’s value in collecting these data…. We should be doing all we can to gather this information.”

Dr. Morrell suggested that physicians consult the NCCN Clinical Practice Guidelines for Genetic/Familial High-Risk Assessment2 and the National Institutes of Health ­GeneReviews3 websites for information about unfamiliar genes that are reported in test results. ■

Disclosure: Dr. Morrell reported no potential conflicts of interest.


1. Morrell L: Next generation sequencing and inherited risk for cancer: BRCA, Lynch and beyond. 2015 New Orleans Summer Cancer Meeting. Dr. Peter A. Cassileth Memorial Lecture. Presented July 18, 2015.

2. National Comprehensive Cancer Network: Genetic/Familial High-Risk Assessment, Version 2.2015. Available at Accessed August 14, 2015.

3. Pagon RA, Adam MP, Ardinger HH, et al (eds): GeneReviews. Available at Accessed August 14, 2015.