We have investigated the current practice of genetic testing of newly diagnosed BC patients in one regional and one university hospital in South Eastern Norway where genetic testing has been mainstreamed into regular cancer care and is offered directly to the patient by the surgeon or oncologist. For the two hospitals combined, 75% of BC patients who fulfilled the NBCG criteria were offered testing. However, there were differences between the hospitals, as 63% of eligible patients were offered testing at the regional hospital (IH) and 90% at the university hospital (Ahus). Surprisingly, 52% of the patients who were not offered testing even though they fulfilled one or more of the criteria and were younger than 50 years at time of diagnosis. As many as 95% of the patients who were offered testing, wanted to be tested.
Our observations indicate that genetic testing is a health service that BC patients want. In contrast, in the DNA-BONus study, only 45.4% of BC patients who were offered testing completed the test. In this study, all breast and ovarian cancer patients diagnosed between 2012 and 2015 at hospitals in the western part of Norway were offered BRCA testing as part of a research project [24]. We cannot rule out that there are some selection biases in our study. Apart from this the difference might be explained by the increasing awareness of hereditary breast and ovarian cancer among cancer patients during the last years, and especially after Angelina Jolie shared her story in May 2013 [25].
Family history based criteria for testing are complex, open to interpretation, and may be difficult to use for both clinicians and patients. In addition, several studies have reported that a significant number of mutation positive BC patients have no suggestive family history of cancer [21, 26, 27]. We have previously suggested that testing should be offered to all BC patients below 60 years [21], and this is now included in the NBCG criteria [13]. An age limit for testing is not open to interpretation, and setting the threshold at 60 years would identify mutation carriers with a sensitivity similar to the family history based criteria [21]. With this in mind, it was surprising for us to see that as many as 52% of the patients who had not been offered testing even if they fulfilled the criteria, were young at time of diagnosis (below 50 years). At the regional hospital, 66% of women diagnosed with BC below 50 years were offered testing compared to 93.5% at the university hospital. Studies have shown that 5–10% of BC patients below 50 years have a BRCA mutation [21, 28]. Young carriers have many years ahead of them with a risk of contralateral BC and OC. They are the ones who will benefit the most from cancer prevention and hence, genetic testing. It is therefore important that routines for genetic testing ensure that these patients have access to this health service.
At Ahus and IH, 57% of all tests were requested by the surgeon at time of diagnosis, and 42% by the oncologist during chemotherapy. It is not stated in the guidelines when genetic testing should be done, but because the great majority of patients see the surgeon first, there is an understanding that it is the surgeons who bear the main responsibility. It is therefore surprising that 42% of the patients were offered testing by their oncologist. There are good arguments for discussing testing with the patient either before or after primary surgery. If a mutation is detected before surgery, the patient may choose bilateral mastectomy directly, and will be spared for one extra surgical procedure. On the other hand, at time of diagnosis, the main objective is to remove the tumor, and the patient may prefer to have more time to adjust to the new information of her mutation status and then choose to undergo prophylactic surgery. We argue that there is a need for definitions and guidelines regarding when genetic testing should be offered, and also to ensure that patients who do not undergo chemotherapy are offered testing.
In five of the journals it was noted that the patient had been tested previously, or that their relatives had been tested. None of these patients were offered a new test. The techniques and sensitivity of the genetic tests for BRCA mutations have changed dramatically since the genes were discovered, from testing only for a few founder mutations in BRCA1 to sequencing and copy number analysis of both genes. Many previously tested patients should therefore be offered a new test. Moreover, even if the patient’s relatives have been tested and not found to be mutation carriers, these could be phenocopies. The patient may still carry a BRCA mutation, and should be offered testing.
The main objective of this study was to investigate to what extent BC patients who fulfill the NBCG criteria were offered genetic testing. We found that 75% of eligible BC patients were offered testing. Other studies have reported testing rates ranging from 15.3% to 60% [16–20]. The design of our study does not enable us to fully explain why the rates differ. However, the observed difference may at least partly reflect that some studies present data on BC patients diagnosed before 2010 when awareness of hereditary breast and ovarian cancer and availability of genetic testing was much more limited. The results may also reflect differences in organization of genetic testing of newly diagnosed BC patients, and different reimbursement schemes among countries. In Norway, all testing is covered by the national social security system, and is not dependent on insurance coverage, regional legislation or patient’s co-payment. As mentioned above, genetic testing is included in regular oncological care, and offered directly by the surgeon or the oncologist without referral to pre-test genetic counseling at a genetics department in South Eastern Norway where our study is performed. In several countries, genetic testing can only be requested by a medical geneticist or genetic counselor. Because genetics departments often have long waiting lists it has been argued that restricting testing to be done only after genetic counseling creates a barrier to genetic testing that contributes to keeping the rates of testing low [23]. With the increasing demand for genetic testing in diagnosis and treatment of BC and other cancers there is a need for alternative models that ensure both high access to testing and that the patient’s need for information is met [29].
As many as 90% of BC patients who were eligible for testing according to the NBCG criteria were offered testing at the university hospital (Ahus) and 63% at the regional hospital (IH). At IH, 66% of patients eligible for testing because of young age of onset and 59% of patients eligible for testing because of their family history of cancer were offered testing. The lower rates of testing therefore affected all patient groups. The numbers were 94% and 100% at Ahus for the two groups of patients. Our data do not provide systematic information on why there was a difference between the two hospitals. At the university hospital, the EPR had a standardized format with headings that included the term “heredity”, whereas the EPR at the regional hospital to a lesser degree seemed to have a set structure with pre-defined headings. Having a set structure with headings will remind the clinician of asking about family history and may also remind the clinicians of genetic testing, and the number of patients who had been asked about family history was somewhat higher for the university hospital than for the regional hospital (95% vs 78%). The observed difference may also be due to differences in awareness and traditions regarding genetic testing between the two hospitals. A clinician who more rarely discusses genetic testing with BC patients may have a higher threshold for bringing up the subject than a clinician who more regularly offers it to patients even if the patient fulfill the criteria.
We suspect that including genetic testing into routine care like it is done in South Eastern Norway contributed to the high rates of genetic testing observed in our study. However, our data indicate that even though testing has been mainstreamed into regular cancer care, it may not be equally available to all BC patients, and there may be need for efforts to raise awareness of genetic testing among clinicians across hospitals. Such efforts could include regular meetings with the regional medical genetics department. One could also establish national, standardized EPR formats that make it mandatory for the physician to include information on cancer in the patient’s family, and information on whether or not the patient has been offered genetic testing. At OUH, results of genetic testing are often included on the top of journal entries together with information about TNM and hormone receptor status.
The Cancer Registry of Norway includes a national quality register for breast cancer diagnostics and treatment. The register includes several parameters relevant to evaluate the quality of the health care provision in hospitals. Currently the register does not include information on whether the patients are offered genetic testing. We will argue that including this information could increase awareness of genetic testing as an integral part of breast cancer diagnostics and treatment, and thereby increase the rates of testing. This would also enable a quantitative evaluation of the delivery of this health care service on a local, regional and national level.
There are some limitations to our study. Our data are based on information from patient records, and we cannot exclude that there may be information relevant for the study that was not recorded. Some patients may have been offered but have declined testing, without it being noted. Family history could for instance have been asked, but not registered in the EPR. The response rate was 63.3% for the regional hospital and 65.7% for the university hospital. No second reminder was sent to the patients. We have only information about those who responded. Even if the response rate was similar, we cannot exclude that the dataset may be skewed either towards those who have been offered testing and/or have accepted testing, or towards those who were not offered testing and/or did not go through with testing. However, the two cohorts were similar in terms of age distribution, and also similar to the age distribution of all BC patients in Norway [30].