Universal genetic testing of patients with newly diagnosed breast cancer — ready for prime time?

Current genetic testing guidelines may overlook patients with actionable mutations in high risk breast and ovarian cancer predisposition genes

The discovery of the BRCA1 and BRCA2 genes just over 25 years ago1 ushered in a new era of genetic testing for patients diagnosed with breast and/or ovarian cancer. A new field of practice in familial cancer emerged that has continued to evolve at an accelerating pace over the intervening years. With improvements in technology, changing patient attitudes and striking new clinical data, genetic testing may have now arrived at another defining moment — as a routine investigation for virtually all patients with newly diagnosed breast cancer (universal testing), a notion that was unimaginable a quarter of a century ago.

When first introduced, germline genetic testing in familial breast cancer was an expensive and cumbersome process, typically taking several months and restricted to detecting only specific types of pathogenic changes in the two BRCA genes. This limited testing was necessarily targeted to individuals who had a significant chance of harbouring an abnormality. Patients were selected on the basis of syndromic features, such as a strong family history, early age of onset, ancestry (eg, Ashkenazi), distinctive tumour phenotypes, and male breast cancer.2

The anticipated benefits of detecting a BRCA1 or BRCA2 mutation for individuals with breast cancer and their families in those early days were very quickly realised. Genetic testing frequently provides a watershed moment that allows an individual’s risk to be calibrated and managed. For example, it can inform decisions on breast‐conserving surgery with radiotherapy versus mastectomy, or can point to the need for risk‐reducing bilateral salpingo‐oophorectomy to mitigate ovarian cancer risk (given that effective screening remains elusive). For patients with metastatic disease, it can also reveal additional treatment options, such as platinum‐based chemotherapy3 or poly(ADP‐ribose) polymerase (PARP) inhibitor therapy, to which BRCA1‐ and BRCA2‐deficient tumours are especially vulnerable.4

Importantly, mutation detection in a proband spotlights a family‐specific mutation that can then be used to manage risk within the extended family. With effective genetic counselling, cascade (predictive, or pre‐symptomatic) testing of multiple at‐risk family members can ensue. Those who test positive can be offered rigorous risk management strategies that have been shown to improve outcomes and reduce mortality. For relatives with a negative predictive test, the result can provide reassurance and relief from the requirement for expensive and anxiety‐provoking screening. Mutation detection thus has the potential to deliver long term public health benefits through the identification and management of people harbouring high risk germline mutations who have not yet developed cancer.

In Australia, genetic testing guidelines recommend offering genetic testing to breast cancer patients with a ≥ 10% probability of harbouring a mutation.5 Algorithms, such as BRCAPRO (BayesMendel Lab; https://projects.iq.harvard.edu/bayesmendel/brcapro), CanRisk (https://www.canrisk.org/) and the Manchester Score (https://www.eviq.org.au/additional‐clinical‐information‐cancer‐genetics/3246‐manchester‐scoring‐system‐with‐guide‐to‐patho), are available to estimate the likelihood of an individual carrying a BRCA1 or BRCA2 mutation, and the online CanRisk tool has been updated to incorporate newly discovered breast cancer predisposition genes.6 In 2017, this approach to identify candidates for breast cancer genetic testing was endorsed by the introduction of a Medicare rebate (item 73296) to cover testing ordered by a specialist for a patient with breast cancer (as well as ovarian, fallopian tube or primary peritoneal cancer), where a suitable quantitative algorithm places the patient at greater than 10% likelihood of having a pathogenic or likely pathogenic gene variant identified.7

Fast forward to the present day, and the landscape is now radically transformed, bringing into question the value of following the 10% probability cut‐off for offering genetic testing. Multigene panel tests that include a suite of additional high risk genes (such as PALB2, TP53, PTEN, CDH1 and STK11) are now readily available (Box)5 and funded by the Medicare rebate.7 The increasingly rapid turnaround time for genetic testing means that patients with newly diagnosed breast cancer can undergo genetic testing in a time frame to allow the result to influence management decisions. At the same time, the cost of testing has plummeted and is now about one‐tenth of the cost of limited BRCA1 and BRCA2 testing in the mid to late 1990s.

In recent developments, the PARP inhibitor olaparib was found to improve outcomes for BRCA1 and BRCA2 mutation carriers with relapsed high grade epithelial ovarian, fallopian tube or primary peritoneal carcinoma,9 resulting in Pharmaceutical Benefits Scheme listing of the drug and extension of Medicare Benefits Scheme funding to genetic testing to determine eligibility for treatment (item 72395). This approach has now been expanded to include maintenance therapy for BRCA1 and BRCA2 mutation carriers who have responded to first line platinum‐based therapy.10 As a result, genetic testing is now essentially a prerequisite for all patients with high grade epithelial ovarian cancer, as it helps to identify patients suitable for targeted therapy. Here, the clinical utility of genetic testing has trumped any genetic testing algorithm.

The goal posts are similarly shifting towards universal genetic testing for patients with newly diagnosed breast cancer. A new landmark study (OlympiA)11 has found that adjuvant PARP inhibitor therapy is effective in certain BRCA1 and BRCA2 mutation carriers following chemotherapy. Benefit was observed for BRCA1 and BRCA2 mutation carriers with triple negative breast cancers measuring > 2 cm in size and/or with node‐positive disease, or who had residual disease after neoadjuvant chemotherapy. For BRCA1 and BRCA2 mutation carriers with oestrogen receptor‐positive and human epidermal growth factor receptor 2‐negative breast cancer, benefit was observed in patients with at least four involved lymph nodes, or a high residual disease score after neoadjuvant chemotherapy. At the time of reporting, the 3‐year invasive disease‐free survival was 86% in mutation carriers who received olaparib versus 77% in the placebo group (hazard ratio, 0.58 for invasive disease recurrence or death), and there was a trend to improved overall survival.11

These findings resulted in an immediate update of the American Society of Clinical Oncology consensus guidelines, which now recommend that BRCA1 and BRCA2 mutation carriers with early‐stage, human epidermal growth factor receptor 2‐negative breast cancer at high risk of recurrence be offered adjuvant olaparib after completion of adjuvant or neoadjuvant chemotherapy and local treatment.12 On a cautionary note, median follow‐up was only 2.5 years, and mature survival and safety data are awaited. Nevertheless, it seems likely that similar recommendations will be incorporated into clinical practice in Australia. Once this eventuates, determining BRCA1 or BRCA2 mutation status will become an imperative for a large proportion of breast cancer patients, just as it has for ovarian cancer.

There are other compelling reasons to know the mutation status in the setting of a new diagnosis of early breast cancer. Mutation carriers are more likely to choose mastectomy with contralateral prophylactic mastectomy, to mitigate their significant risk of developing another breast cancer. A person’s risk of contralateral breast cancer will depend on the specific germline mutation, their age at diagnosis and length of time in follow‐up. For example, the risk of developing contralateral breast cancer over the next 10 years for BRCA1 versus BRCA2 mutation carriers is about 23% and 16%, respectively. Over a 20‐year period, the risk is 40% and 26%, respectively.13 Bilateral salpingo‐oophorectomy can also be used as a form of endocrine therapy for people with oestrogen receptor‐positive breast cancer, in addition to its ovarian cancer risk reducing effect.

The prevalence of BRCA1 and BRCA2 mutations in the population is likely to be somewhere between 0.25% and 0.50%; however, the proportion of carriers identified in patients with a breast cancer diagnosis would be expected be significantly higher, regardless of family history. A recent United States population‐based study of more than 32 000 participants with breast cancer estimated the prevalence of BRCA1 and BRCA2 pathogenic variants to be 2.14%, increasing to 5.03% (95% confidence interval, 4.79–5.27%) if 12 established breast cancer predisposition genes were included.8 While 5% falls short of the historical 10% cut‐off, the lower detection rate must be weighed against the greater clinical utility of determining the germline mutation status and the reduced cost of testing. Notably, germline mutations may be as high as 10.4% (5.0% for BRCA1 and BRCA2 alone) in patients diagnosed with metastatic breast cancer, underscoring potential utility in that setting, given that PARP inhibitor therapy has been approved for BRCA1 and BRCA2 mutation carriers.14

There are challenges to be considered when contemplating widespread testing of newly diagnosed patients with breast cancer. Interpreting sequencing results is frequently complex, and patients with genetic variants of uncertain significance (VUS) of unknown actionability will be identified. To meet this challenge, the International Agency for Research on Cancer has proposed a system that subdivides variants into five classes15 (Supporting Information). Class 4 and 5 variants are considered pathogenic and clinically actionable. Class 3 variants are considered uncertain in nature and not clinically actionable. These are the most problematic. Class 1 and 2 variants (likely benign) rarely pose clinical challenges. Variant classification is being enabled through large international collaborations that pool global experience using various experimental approaches and modelling. Despite this, a strategy that expands testing to individuals with a lower chance of harbouring a mutation will inevitably increase the rate of VUS identification — a problem that grows as additional genes are added to a testing panel.

A widespread rollout of genetic testing will need to accommodate the increased burden of actionable pathogenic variants as well as all nuances of variants of uncertain significance, which may come at some psychological costs for patients. A risk exists of overtreatment of patients who are found to have VUS or mutations in lower penetrance risk genes,7,16 and caution and appropriate education of specialists and the broader medical community will be required.

We are arguably well placed in Australia to address these challenges through the well developed network of coordinated, state‐based familial cancer centres that have been established to provide genetic counselling for patients and advice to treating clinicians. Already genetic testing has begun to enter into the practice of many oncologists and breast surgeons through mainstreaming programs, where a breast team member provides genetic counselling and orders the panel test. Genetic services are engaged once a mutation or problematic VUS is identified. Mainstreaming was first introduced at the Royal Marsden Hospital, London,17 and has been successfully applied in the Australian setting.18

Universal testing of patients with newly diagnosed breast cancer is being explored in a research study underway in Melbourne (https://www.transbcr.org.au/magic). Close liaison between the Familial Cancer Centre, molecular diagnostics laboratory and breast oncology team (including through multidisciplinary team meetings) has proven essential. Successful application of universal genetic testing is likely to be an iterative process. Breast oncology teams will need to be progressively upskilled, the familial cancer workforce will need to be adapted and appropriately resourced, and strategies for explaining genetic test results to patients optimised. In many ways, these challenges are very similar to those faced when BRCA1 and BRCA2 genetic testing was first introduced in the late 1990s.

In conclusion, when we consider the rapid progress that has occurred since the discovery of the first breast cancer genes, it appears that we have reached the juncture where multigene panel testing has become relevant for virtually all patients with newly diagnosed breast cancer. A once expensive and highly specialised test is now relatively low cost and accessible, and critically, has demonstrable immediate clinical significance.