The management guidelines for carriers of pathogenic variants in BRCA1/2 with breast cancer suggest that a less aggressive surgical approach is possible, as even though the risk of contralateral breast cancer is high, there is no major increased risk of ipsilateral breast tumor recurrence when undergoing conservation compared with the noncarrier population.1
In this journal, Shubeck et al. show that this approach appears reasonable when tested in the real world.2 However, multiple provisos must be considered. First, close to 70% (65/99) of the patients who had breast-conserving surgery did not know they had a pathogenic variant. They chose this approach without knowledge of the pros and cons of the approach. If their pathogenic variant status had been known, their providers would likely have recommended a bilateral mastectomy (appropriately when this study was done), so, in essence, this was the blind leading the blind. But what is interesting is that 43.4% (43/99) of these patients chose bilateral mastectomy at a later date, either when their pathogenic variant status was made known to them or when the cancers recurred. It is possible that, with complete information initially and with time to think, patients might not necessarily have chosen conservation. In an ideal world, patients should know their pathogenic variant status long before they develop cancer and would either have been managed by risk-reducing mastectomy (markedly decreasing the risk of breast cancer) or by intensive screening (markedly decreasing the stage of detected cancers and thus decreasing the morbidity of treatment). But, in addition, patients aware of their pathogenic variant status would have had time to think about what they would do if cancer developed. It is not fair to tell an often very young patient that: (1) you have cancer, (2) it is advanced, (3) you have a pathogenic variant that likely caused that cancer, and (4) you are at high risk of another breast cancer; then ask the patient to make a decision about the preferred surgery. We owe it to our patients not to put them in this situation.
While the guidelines suggest that breast conservation is noninferior, the question remains whether that is what the well-informed and prepared patient would want. We need to find ways to better inform patients about their pathogenic variant status long before they develop cancer.
In the current study, only 80 of the 395 patients undergoing surgery knew their pathogenic variant status more than 3 months before surgery (Melissa L. Pilewskie, MD, personal communication, March 2022). This means that 79.7% (315/395) of the patients in this study did not know their pathogenic variant status until after they were diagnosed with breast cancer. This is, unfortunately, in line with existing estimates that, after 25 years of commercial cancer genetic testing, around 90% of carriers remain unaware of their status and thus not managed by the guidelines.3,4 This leads to cancers that could possibly have been prevented and to cancers found at a higher stage than necessary. This paper documents that unfortunate outcome.
Warner et al. have shown that carriers undergoing magnetic resonance imaging (MRI) screening have a node-positive rate of about 13%.5 However, in the current study, the node-positive rate for invasive cancers is about 43.5% (148/340),2 much higher than expected with guideline-based screening but similar to the 40% node-positive rate found in Warner et al.’s control group (i.e., BRCA pathogenic variant carriers undergoing routine screening). It should also be noted that the 10-year breast cancer-specific survival (BCSS) rate reported in this study (92.2% following breast-conserving surgery and 88.8% following mastectomy) may be misleading.2 Assuming that all breast cancer deaths were in patients with invasive cancer, this means that the 10-year BCSS for invasive cancer was about 76.6% following breast-conserving surgery and 68.3% following mastectomy, far below what would be expected in a screened population. This is likely explained by the inadequate screening these patients received because they were not known to be carriers.
It is understandable that we, as physicians, would like to offer conservation. This, in some ways, makes us feel better about the fact that we missed an opportunity to find patients’ pathogenic variant status when it had the potential to prevent their cancers or find them earlier. So, while we agree that conservation is reasonable for these patients, it should only be done with the patient fully informed and preferably having had time to think. Patients should know their pathogenic variant status long before they develop cancer and should have already experienced several rounds of MRI screening and perhaps a biopsy or two, to really understand what intensive screening involves. They should be aware that the risk of contralateral cancer at 10 years is about 13.9% and added to that is about a 7.3% risk of locoregional recurrence (as identified in the current study).2 Together, this means they have about a 21.2% risk of experiencing breast cancer again within the next 10 years if they choose conservation. They should be aware that, if they are BRCA1 carriers, they will likely undergo chemotherapy again. And they should be aware that some studies have shown an improved survival after 20 years with bilateral mastectomy.6,7,8 Under these circumstances, the patients’ choice of conservation could be reasonable.
If we really want to make these patients’ lives better, we should identify them as pathogenic variant carriers long before cancers develop. It seems there are several barriers to the identification of these carriers, including (but not limited to) no or incomplete family histories being obtained, a lack of knowledge by providers as to who needs testing, the requirement by certifying bodies or insurers for a genetic counseling visit before testing,9,10 guidelines that miss about half of the carriers,11 and poor uptake of cascade testing of relatives.
Whitworth et al. found that, when a major insurer changed their policy to require genetic counseling before testing, the test cancellation rate rose from 3.8% to 37.7%, disproportionately causing cancelations among African Americans and individuals of Latin ancestry.12 The rate of positive tests remained about the same, showing that the cancelations were somewhat random relative to risk. This cancelation rate was related to the additional medical encounter required, which many patients do not comply with or lack access to. Our team, like others, has found that the minority of patients (for us, 37%) referred for genetic counseling complied (for a variety of administrative and personal reasons).13 We, George et al., and others have found that there was a long delay between the request for consultation and the actual visit, likely adding to the lack of compliance.13,14 In our study, if a patient did not need urgent testing for a newly diagnosed cancer, the average time to the consultation was over 200 days.13 In addition, George et al. estimated the cost of counseling as over $400 per patient (when the facility and administrative costs were included.14
Approaches to increase testing include offering testing without pretest counseling (with counseling occurring only after a positive test is identified or for those needing more pretest reassurance or information). While many certifying bodies have taken it as a given that pretest counseling is useful, this has not held up under scrutiny. Gronwald et al. have shown that women in Poland could be tested just by returning a postcard found in a women’s magazine, with no major issues found.15 Swisher et al. have reported the MAGENTA study, where distress at 3 months in the no-counseling arms was noninferior to the mandatory pre- and posttest counseling arm, and distress, anxiety, depression, and decisional regret were similar across all arms. Moreover, counseling led to less testing, with the test completion rate being only 60.6% in the mandatory pre- and posttest counseling arm versus 86.4% in the no-counseling arm.16
The approach of no pretest counseling is being put into practice by Narod et al., who offer to test any Canadian who comes to their website,17 and by Offit et al., who offer to test Ashkenazi Jewish who come to their website.18 Population testing without pretest counseling is being offered to all patients over age 18 years by Geisinger (with the return of results for 76 selected genes including BRCA1/2 and Lynch genes),19 by the Medical University of South Carolina (with the return of results for BRCA, Lynch genes, and familial hypercholesterolemia), and others. This approach simplifies the testing process by not requiring that patients meet guidelines and removes the onerous task of filling out the genetic testing request form needed to obtain insurance coverage.
Taken together, these less labor-intensive approaches can help identify large numbers of pathogenic variant carriers before they develop cancer, allow them to benefit from intensive screening or risk-reducing surgery, and perhaps most importantly, give patients time to think. Then, when they develop cancer, they can make an informed decision.
In summary, the new guidelines and the data presented by Shubeck et al. both strongly suggest that conservation is a reasonable approach for BRCA carriers. However, if we are to have a real impact on population health, we must find carriers at scale, by either population testing or other approaches (such as point-of-care testing and simplification of the testing process). Societies and certifying bodies need to change their guidelines to support all providers in undertaking genetic testing, and providers need to incorporate genetic testing into their normal practice. The time has come to bring the promise of the Human Genome Project20 to all patients.