BRCA1/2 associated cancer susceptibility: a clinical overview

A diagnosis of cancer is very often accompanied by several questions both from the patients and the physicians. “Why me?” is a question routinely phrased by most patients, reflecting not only a psychological distress but also a genuine search for an explanation on the causes of their disease. The largest percentage of them may not receive a concrete answer for this question. Still, for a small, but not a negligible portion, the cause will be concerned with an inherited cancer predisposition. In many, but not all families, this predisposition is clinically evident, as a number of cancer diagnoses in a family or an individual. However, as families are growing smaller in modern generations, inherited predisposition may be occult.

Genetic cancer predisposition varies in penetrance, i.e. the percentage of the individuals who will develop cancer, after inheriting a specific genetic alteration. On that basis, penetrance is classified as high, intermediate and low, conferring a relative increase in cancer risk of more than five to ten times, two to five times and less than two-fold respectively, in comparison to the average population’s cancer risk [1].

More specifically, 5–10% of breast cancer and over 15% of ovarian cancer diagnoses are caused by heritable mutations in certain genes; BRCA1/2 are the most frequently identified mutated genes. Their role as cancer susceptibility genes has been established since 1993–1994, and there has been an exponential increase in knowledge about them since then [2–4].

BRCA1 & 2 genes, located in chromosomes 17q21 and 13q12–13 respectively, function normally as a DNA repair genes, through participation in the homologous recombination mechanism [2]. Functionally significant mutations result in defective homologous recombination repairing. Prevalence of such mutations is estimated at 1:400 for general population, though inter-population variations may exist. Ashkenazi Jews have a prevalence of 1:40, justifying perhaps a discussion for populationwide genetic screening.

BRCA1/2 mutations are inherited in an autosomal dominant manner; de novo mutations are extremely rare, accounting for less than 1% of positive tests. Sanger sequencing is typically used to diagnose point mutations, whereas larger rearrangements (deletions/ duplications) are diagnosed with techniques such as MLPA. The spectrum of genetic variations for BRCA1 and 2 genes varies among populations, however a complete evaluation for both types of mutations, requires combination of both techniques for all populations. An exception probably concerns Ashkenazi Jewish families in which three specific mutations account for over 99% of all BRCA1/2 identified [3].

Studies performed in Greek population have described some interesting findings. More than 50% of all breast cancer patients with mutated BRCA1/2 seem to carry one of the six specific mutations [5]. Another study of 403 unselected triple negative breast cancer patients reported a prevalence of 16% mutations in the BRCA1 gene [6] much in keeping with international studies. Concerning ovarian cancer, a study including both the familial and non-familial ovarian cancer subjects identified a prevalence of 10% BRCA1/2 mutations in the overall study population [7].

Identification of mutation carriers is routinely based on clinical information, typically through a recent diagnosis of a BRCA-associated cancer in an individual (most frequently breast or ovarian) and of course, the availability of cancer family history. Histopathologic information may be crucial in this evaluation, since breast and ovarian cancers arising in this setting tend to have certain histopathological features. BRCA1-associated breast cancers are most frequently (80%) triple negative, whereas BRCA2 is associated with a sporadic phenotype, i.e. ER positive; interestingly HER-2 positivity is rather rarely expressed in BRCA-related breast cancers, in contrast with Li-Fraumeni syndrome, where over 60% are HER2 positive [12]. Ovarian cancer is mostly high-grade serous in phenotype.

The course from the suspicion of cancer susceptibility to the confirmation of a relevant germline mutation has to go through genetic counselling, which is the cornerstone of genetic evaluation and subsequent cancer risk management. It is usually provided by oncologists (or other cancer physicians) with special training or genetic counsellors. Consideration for genetic testing should lead to pre-test genetic counseling, and testing results should be also addressed during post-test counseling.

The questions and issues vary widely for individuals and families who receive genetic consultation, and they need to be addressed primarily. Age, prior diagnoses of cancer, family and marital status, priorities, values, and socio-economic issues are some of the different variables among individuals and families who seek genetic guidance. Most importantly, a person who will get tested is aware beforehand of the various outcomes, their meaning, their implications and their limitations, and so must be prepared for the next steps.

The recommended components of a typical cancer genetic counseling session, according to the US National Society of Genetic Counselors [13] include:

Setting a mutual agenda for the session

A pre-test genetic counselling session structure, proposed at the Genetic Oncology Training Program of the Hellenic Society for Medical Oncology is as follows, with an average duration of 90 minutes.

Identifying the proband’s motives, predominant questions, worries and priorities for genetic evaluation and testing

At the end of a genetic counselling session, the counselee must be in position to provide his consent for genetic testing, after having received and processed adequate information for a number of issues (Table 2) [14]. This can be a challenging task for both parts of counselling, since the amount of information to be passed can be very large and difficult to process. Often it can become even more complicated by the psychosocial burden of a recent or a potential cancer diagnosis in the counselee’s family or of a diagnosis for himself/herself.

Studies of the psychological impact of genetic testing have demonstrated a temporary increase of distress levels for individuals with deleterious mutations, which usually returns to the level before testing, while negative or inconclusive testing is associated with a progressive decrease in distress levels. Psychosocial assessment is part of the genetic counselling, and referral for specialist’s support may be appropriate for some individuals.

Comparative studies have determined that genetic counselling for cancer predisposition through telephone contact may be as effective and psychologically reassuring as face-to-face counselling. This makes access easier for patients who would need to travel a long distance to receive their counselling. The only limitation with distance counselling would be physical findings, such as skin/mucosal lesions, indicative of rare syndromes (such as Cowden’s, Peutz-Jeghers, etc.).

When a member of a family with a known BRCA1/2 mutation is tested for this mutation and is found negative, then it is considered as ‘true negative’, meaning that it does not carry the mutation considered responsible for this family’s phenotype. True negative individuals are considered as at average population risk for breast and ovarian cancers. This information is of great value and significance for a relative of a deleterious mutation carrier, since her/his medical care and surveillance is that of average risk population and very importantly, it is news received with great relief.

In families without a known BRCA1/2 mutation, most tests will return with uninformative results, mostly negative and at times, VUS. These results mean that the individual’s (and the family’s phenotype) is not explained by a BRCA mutation and should be discussed during post-test genetic counseling. In that session, implications of this test are reviewed and further testing options are presented and discussed. Until a couple of years ago, genetic testing was concluded unless a phenotype indicative of another high penetrance gene (such as Cowden’s, Li-Fraumeni, Peut-Jeghers, etc.) was considered worthwhile testing. This has been changing with the emerging and expanding availability of multiplex gene testing, through multi-gene panels.

It is now increasingly frequent for both counseling oncologist and counselees to pursue second-line testing for a group of genes related to breast and/or ovarian cancer. Such panels are now commercially widely available and they are considered as a cost-effective alternative to traditional single-gene testing strategy. Many of these panels include intermediate penetrance genes, whose clinical meaning and utility has severe limitation; for many genes, proper management is unclear, due to little amount of published evidence; other better studied genes, such as PALB2 or CHEK2, are accompanied by acceptable estimations on cancer risk for mutation carriers [32, 33]. Unfortunately it remains highly unclear whether true negative relatives run elevated risk and how severe it is. Due to that, counseling healthy, negative members of such a family remains highly problematic.

Another downside of panel testing has to do with unexpected results, where a deleterious mutations is detected in a gene that is irrelevant with the phenotype, which prompted the testing (e.g. APC in a family with breast and ovarian cancer diagnoses etc.). Though rare, the tested individual should be advised and warned on this possibility.

It is almost certain that cancer risk in pathogenic BRCA1/2 mutation carriers is subject to modifying factors. Recent data from a very large study (iCOGS) indicate the potential use of a common polymorphism panel, which can classify carriers in percentiles; top 5–10% of them are almost certain to develop breast or ovarian cancer, while the lowest 5–10 percentiles, may have as low as 50% lifetime risk for breast cancer and 10–15% for ovarian cancer [34, 35]. Validation of this panel could lead to additional testing for the refinement of risk estimates for carriers, leading to a more accurate and better justified clinical decisions and management. PARP inhibitors are, as abovementioned, a new class of anti-neoplastic agents. Promising studies are under way, investigating their role not only in breast and ovarian cancer, but also in other associated cancers such as pancreatic and prostate. What is more interesting is the testing not only of BRCA1/2 in germline or tumour DNA, but its potential substitution from HRD assays. Such assays test the tumour for potential deficiency in its homologous recombination mechanisms, which may be subsequently exploited by PARP inhibition. So, it is possible in the near future that genomic testing may be the first step to identify tumour HRD, prompting subsequently a germline test with a panel including all relevant genes (BRCA1/2, ATM, BARD2, CHEK2, BRIP1, RAD51C, RAD51D etc), thus directing the testing flow from genomics to genetics.

Recently, M.C. King, a leading investigator of the identification of BRCA1 and 2 genes, at a provocative keynote lecture, while receiving the 2014 Lasker award, addressed the proposal for a population-wide BRCA testing [36]. Despite the significant controversies, risks and problems associated with such a public health strategy, an extensive discussion was kicked off due to that proposal. An important argument is the fact that approximately one-fourth of mutation carriers have no relevant family history, thus having no possibility of being warned that they are carriers and take appropriate protective measures. Population data supporting it come from Ashkenazi Jewish communities, where the prevalence of the mutation is very high (1:40). Moreover, three specific mutations are identifiable with a much less costly test than full gene sequencing and rearrangement testing. In contrast, most populations would require a much more expensive testing, with a much greater number of tests needed to identify the mutation carrier (average mutation prevalence is 1:400). That would require an explosive increase in clinical-genetics infrastructures to provide appropriate counselling. Even if compliance of deleterious mutation carriers in cancer surveillance and/or risk reduction interventions could be high enough to ensure cost-efficacy of the strategy, a large number would be found with VUS, making clinical management difficult. Besides, a colossal number of women would have uninformative (negative) results, leaving the question open on whether further testing would be required. A 2013 study of an oncology department in Greece identified the proportional of women fulfilling NCCN guidelines criteria, being over 40%, while the percentage of the ones who were offered and underwent counselling and testing was miniscule [37]. A recently published study performed within a large US health insurance system, reported that among women who got BRCA genetic testing, the majority did not receive genetic counselling (61.8%), with only 38.2% having been counselled by a genetics clinician. The primary reason for not getting counselling is the lack of referral by the physician who requested the testing. Women who did receive counselling reported a better level of understanding and satisfaction about their tests [38]. This result is rather surprising, contrasting relevant guidelines. It is the author’s opinion that for many countries, effort and resources should be primarily allocated to the proper adoption of testing and management guidelines, until at least the proportion of women who get tested corresponds to the vast majority of those who fulfill testing criteria, and their further management is according to these guidelines.