Risk-reducing bilateral mastectomy dramatically lowers the risk for breast cancer compared with surveillance among BRCA1/2 mutation carriers, but whether that benefit extends to improved survival remains unclear.

Two recent studies that aimed to clarify this relationship came to different conclusions.

A systematic review and meta-analysis published in January in JAMA Surgery reported that risk-reducing bilateral mastectomy was associated with significantly lower overall and breast cancer-specific mortality. However, a large prospective cohort study, published a month later in the Journal of Clinical Oncology, found no overall or breast cancer-specific survival advantage for women who underwent surgery vs those who chose intensive imaging surveillance.

What can explain the apparent discrepancy? And how should clinicians counsel BRCA1/2 carriers who are weighing the pros and cons of prophylactic surgery vs long-term surveillance?

Survival or No Survival Benefit

In the meta-analysis, which pooled data from six observational studies, 2558 female patients received the risk-reducing surgery, and 3577 underwent observation or surveillance. Five analyses were single- or multicenter cohort studies (four were retrospective and one was prospective), and one was a prospective multicenter pseudo-randomized controlled trial, with mean follow-up times ranging from 41.6 months to 127 months across the studies.

Risk-reducing bilateral mastectomy was associated with a 63% lower risk for overall mortality than surveillance, according to a pooled analysis of four studies (hazard ratio [HR], 0.37; P < .001), and an 86% lower risk for breast cancer-specific mortality than surveillance, according to a pooled analysis of 2 studies (HR, 0.14; P = .002).

The researchers concluded that these findings support risk-reducing bilateral mastectomy as a “potentially life-extending intervention.”

However, the prospective cohort study, which followed 1205 BRCA1/2 carriers, found no overall or cancer-specific survival benefit among individuals who underwent the risk-reducing surgery.

In this study, women elected either risk-reducing bilateral mastectomy (n = 460) or imaging surveillance (n = 745). Median follow-up was slightly more than 8 years in the surgery group and almost 6 years in the surveillance group.

As expected, the overall annual incidence of breast cancer was much lower in the surgery group (0.15% vs 2.4%; a 94% reduction). However, the rate of breast cancer-specific deaths was similar in both groups — 0.43% (2 of 460) in the surgery group vs 0.54% (4 of 745) in the surveillance group (P = .36). Overall survival was also similar between groups.

“For women electing imaging surveillance over risk-reducing surgery, our results may offer reassurance that their breast cancer-specific survival and overall survival are unlikely to be compromised,” lead author Ashu Gandhi, MD, of the University of Manchester, Manchester, England, and colleagues wrote.

Understanding the Conflicting Findings

Gideon Meyerowitz-Katz, PhD, epidemiologist and senior research fellow at the University of Wollongong, Wollongong, Australia, noted that the meta-analysis included only six observational studies, one of which played an outsized role in the reported findings: a Dutch study from 2019, which comprised nearly half of the patient population assessed.

The Dutch study included almost 3000 BRCA1/2 carriers who were followed for a mean of about 10 years. Compared with surveillance, risk-reducing bilateral mastectomy was associated with a 60% reduced risk for overall mortality (HR, 0.40) and a 94% reduced risk for breast cancer-related mortality (HR, 0.06) in BRCA1 carriers, as well as a 55% reduced risk for overall mortality (HR, 0.45) in BRCA2 carriers. However, for BRCA2 carriers, the researchers could not estimate an HR because no women opting for bilateral risk-reducing mastectomy died due to breast cancer.

Meyerowitz-Katz explained that statistical estimates for mortality rely heavily on the total number of events or deaths in this case. Most studies in the meta-analysis recorded few deaths and wide CIs. The 2019 Dutch study, however, reported many more deaths in its surveillance group (81 among BRCA1/2 carriers vs 14 deaths in the surgery group).  Consequently, the estimate from the meta-analysis was largely driven by this one study, which effectively overshadowed findings in the other smaller studies, he explained.

Gandhi also highlighted the Dutch study’s reliance on self-reported data for the surveillance cohort, which could mean that some deaths went unreported over the follow-up period. Relying on self-reported data “lowers one’s confidence in the quality of the surveillance actually received by women,” Gandhi said.

Kathy D. Miller, MD, professor of oncology at Indiana University School of Medicine, Indianapolis, agreed that it’s not clear how well those who chose surveillance were surveilled and commented, more generally, that the meta-analysis included several studies with shorter follow-up periods.

In contrast, Gandhi explained, his team relied on an investigator-directed, intensive surveillance program, during which time each event, cancer diagnosis, and death was collected prospectively. Still, Meyerowitz-Katz noted that a survival difference between the two groups could still emerge with longer follow-up.

“Most people who get breast cancer get it much later in life,” Meyerowitz-Katz said. “So, in 20 years’ time, when the women who had surgeries in 2013 are in their sixties, then we’ll have a much better estimate of the effect.”

Overall, though, Miller noted that the low death rate in the new prospective study’s surveillance arm supports the safety of that approach, even with the inherent limitations of a nonrandomized study and shorter follow-up.

“Encouragingly, deaths from breast cancer were uncommon in both groups,” Miller said. “That is the reassuring message of the cohort study.”

What Should Clinicians Tell Patients?

Both surgery and surveillance should be presented to patients as reasonable choices, Miller said.

“Too many women with mutations are given the message that they ‘have to have’ or ‘need to have’ bilateral mastectomy,” she explained. Given that a survival benefit of surgery is unclear, “I don’t think that should be a major driver in patients’ decision-making.”

Instead, the focus should be on patient preference because both approaches “require different things of our patients,” Miller said. “We owe our patients a balanced discussion of both options.”

Despite uncertainty surrounding survival outcomes, Gandhi acknowledged an advantage to the surgical route. “One can reasonably argue that risk-reducing surgery saves the labor of being sick, as breast cancer incidence is reduced by surgery,” he said.

But until longer follow-up and more high-quality data are available on mortality, clinicians and patients will be relying on incomplete evidence.

“The most reliable thing that I can say is that the evidence is currently mixed, and it’s hard to draw a definite conclusion,” Meyerowitz-Katz said.

Funding for the study by Gandhi and colleagues was provided by the National Institute for Health and Care Research to the Manchester Biomedical Research Centre. Gandhi reported receiving research funding from Gilead Sciences. Cathal O’Reilly and colleagues reported having no conflicts of interest. Meyerowitz-Katz and Miller disclosed having no relevant financial relationships.