Medicare-funded cancer genetic tests: a note of caution

Clinicians need appropriate education and support in keeping pace with the genomics revolution

Media headlines stating that genetic testing for patients with a high risk of breast and ovarian cancer are now free are somewhat misleading. Clinical genetic testing for heritable, germline mutations (pathogenic variants) in two major genes (BRCA1 and BRCA2) that are associated with a high risk of breast and ovarian cancer came into Australian practice in the mid-1990s, and were offered free of charge (but not under Medicare) to appropriate patients in public clinics. Until now, testing, which has proven clinical utility,1 has mostly been offered through a network of family cancer clinics and genetics services that provide expert genetic counselling and testing of these genes in the context of familial breast and ovarian cancer.

Testing is appropriate when there is at least a 10% chance of identifying a gene mutation responsible for the personal or family history of cancer. When the chance of detecting a mutation is less than 10%, self-funded BRCA1 and BRCA2 testing has been available in public or private genetic services. To help decide whether testing is appropriate, there are several algorithms to calculate the mutation probability (Box 1).

The change referred to in the media is that these tests are now being mainstreamed and they can be ordered for selected patients, with a new Medicare benefit, by non-genetic specialists in either public or private practice. While the proven benefits to selected families underscore the importance of broadening the availability of appropriate genetic testing, it is also essential that any clinician who orders breast cancer genetic testing understands the complexities and implications generated.

In general, the process of genetic testing in a family begins by testing an individual with cancer (usually with a high risk family history) and searching their BRCA1 and BRCA2 genes for a causative mutation. If a mutation is found, then other adult at-risk genetic relatives (male or female) can be offered predictive genetic testing for this family-specific mutation. Relatives who have inherited the family mutation are confirmed to be at higher risk of cancer (and will benefit from tailored screening and prevention), whereas relatives who have not will usually remain at the background risk for breast and ovarian cancer (depending on the cancer history on the other side of the family).

When a mutation is not found using the initial mutation search in an affected family member, this negative result is not truly negative, but rather is uninformative. No predictive test can be done for family members if that is the case. There are many families with a strong family history for which there must be some genetic cause, and yet it cannot be found. Limitations in knowledge and technology need to be understood, and individuals from these families remain at potentially high risk. Such uninformative results require careful counselling so that families (and their clinicians) are not falsely reassured by the “no mutation found” result.

Only 5% of female breast cancers, 15% of invasive epithelial ovarian cancers and up to 14% of male breast cancers are related to BRCA1 or BRCA2 mutations, thus, most patients with breast cancer do not need, nor will they benefit from, a genetic test. eviQ — a Cancer Institute NSW repository of evidence-based cancer treatment protocols and information — has published nationally agreed protocols for genetic testing.2 These outline criteria that aim to achieve a maximum pick-up rate of mutations while avoiding unnecessary expense to the health system.

New funding model for genetic testing: Medicare

Since 1 November 2017, any specialist or consultant physician (in public or private practice) can order genetic testing for breast and ovarian cancer under certain criteria, covered by the Medicare Benefits Schedule (MBS) (Box 2).

Item 73296

Below, there are several notes of importance in regards to ordering this test.

Calculation of eligibility for the Medicare benefit

tbox1

Eligibility is based on a quantitative algorithm giving the person being tested an over 10% chance of having a pathogenic mutation identified if testing is done. Requestors need to be familiar with the freely available algorithms (Box 1).

Interpretation of genetic variants

Small differences in DNA sequence account for the variation we see between individuals, including differences in hair, eye and skin colour; height; body shape; and susceptibility to disease. If a laboratory identifies a variant in a gene, it needs to determine if it is a harmless change (benign: a polymorphism) or harmful (disease causing: a mutation). However, the human genome is complex and classification of variants is not always straightforward.

Although there is a proposed standardised reporting system6 (with five classes of variants based on the degree of likelihood of pathogenicity), in Australia there is no nationally consistent approach to reporting variants. With the new MBS item numbers, it is likely that smaller laboratories, less familiar with the interpretation of sequence variants, may offer testing.

Uncertainty of pathogenicity

About 10% of BRCA1 and BRCA2 tests result in the identification of a variant of uncertain significance (VUS) that cannot be clearly classified as either benign or disease causing. The VUS rates are higher for the other five, less well studied genes included in the MBS item descriptor. Although BRCA1 and BRCA2 are two of the most commonly analysed human genes, after 20 years of testing, there is still argument about the classification of some variants as to pathogenicity. The identification of VUS will continue to pose considerable challenges to both the laboratory and the clinician who orders a genetic test. In the face of a VUS, the ordering clinician should be able to explain the uncertainty, but can refer the patient to a clinical genetics service for expert assistance.

Even when a variant is clearly pathogenic, genetic counselling sessions can take a long time to address both the personal and familial implications of the result. The identification of a VUS adds complexity to these sessions and may cause confusion for clinicians and patients, particularly when a clinician has limited experience with the nuances of genetic testing. In fact, misinterpretation of BRCA1 or BRCA2 genetic test results has already resulted in inappropriate surgery (including bilateral mastectomy), with legal consequences.7,8 Genetic counsellors and genetics specialists have specific expertise in explaining the possible clinical, psychosocial and familial implications, and the uncertainties of a VUS to the patient, their family and their doctors.

Testing of other genes covered by the item number

The item number covers testing for some other genes (in addition to BRCA1 and BRCA2): PALB2, STK11, PTEN, CDH1 and TP53. Importantly, inclusion in the item number does not mandate testing these genes when BRCA1 or BRCA2 are tested.

PALB2 is a well established breast (probably not an ovarian) cancer-related gene. Except for one particular mutation (PALB2 c.3113G>A) associated with a higher risk, in most families, germline mutations in PALB2 are associated with only a moderate risk of breast cancer. Some variants in PALB2 are not associated with breast cancer at all and this is often misunderstood.

STK11, PTEN, CDH1 and TP53 should only be tested if there are clinical indicators that they might be involved. Testing outside the clinical syndromes described in Box 3 should be approached with caution, as experience with testing of these genes and managing the medical and psychosocial consequences requires specialised training.

While ultimately all of these genes (and others, such as mismatch repair genes) may be included in a “breast/ovarian cancer panel” of genes, the testing of multiple genes may uncover unclassified variants, variants outside the usual clinical context, variants unrelated to the current cancer, or unexpected important variants for which the patient has not been well prepared.

Item 73297

This item number would, at first glance, seem to be fairly straightforward as it covers the predictive test for the family-specific mutation.

A critical prerequisite for this testing is a copy of the genetic test report from the family member in whom the mutation was identified, so the local laboratory knows which specific genetic variant to test for and can confirm if the variant is disease causing and not a VUS or a benign variant.

Laboratory requirements for predictive testing

The item descriptor allows for a predictive genetic test to be ordered by any specialist or consultant physician. However, laboratories undertaking clinical testing need to be accredited by the National Association of Testing Authorities Australia and, therefore, governed by the requirements from the National Pathology Accreditation Advisory Council. The Council considers this predictive testing to be a level 2 DNA test (with potential to lead to complex clinical issues).9 In other words, it is a test “… for which specialised knowledge is needed for the DNA test to be requested, and for which professional genetic counselling should precede and accompany the test,” and “specific written consent and counselling issues are associated with this grouping”.9 While the laboratories may not need to sight the consent form, the clinician must indicate that consent has been given on the request form. The guidelines from the Human Genetics Society of Australasia (www.hgsa.org.au/documents/item/1574) recommend the use of a consent form; however, approved consent forms vary from state to state. Liaison with family cancer clinics may be helpful for ordering clinicians.

Item 73295

This test is relatively straightforward and not discussed here.

Conclusion

The new Medicare item numbers available for breast and ovarian cancer-related genetic testing are an important step forward in facilitating the transition to mainstream genetic testing coordinated by non-genetic health professionals. This article aims to highlight the benefits of mainstream testing, but also to alert non-genetic specialists about some important caveats regarding genetic testing (Box 4). Because genetic and genomic testing will soon be incorporated into many aspects of medical care, it is important that clinicians receive appropriate education and support to enable safe, evidence-based practice in keeping pace with the genomics revolution. This will be optimised if ordering clinicians develop a close relationship with a local family cancer clinic or clinical genetics service, which are ideally placed, and willing, to educate and support non-genetic practitioners in the transition to mainstream genetic testing.