Preoperative diagnosis of BRCA1/2 mutation impacts decision-making for risk-reducing mastectomy in breast cancer patients

Decision to undergo risk-reducing mastectomy (RRM) needs to consider several factors, including patient’s preference, surgeon’s preference, family history, and genetic predisposition. The aim of this study was to examine whether preoperative diagnosis of BRCA1/2 mutation status could influence surgical decision-making in newly diagnosed breast cancer patients. We retrospectively reviewed ipsilateral breast cancer patients with BRCA1/2 mutation who underwent primary surgery between January 2008 and November 2019 at a single institution in Korea. Of 344 eligible patients, 140 (40.7%) patients were aware of their mutation status ‘prior to surgery’, while 204 (59.3%) did not. Contralateral RRM rate was significantly higher in the group with BRCA1/2 mutation status identified ‘prior to surgery’ compared to the group with mutation status identified ‘after surgery’ [45.0% (63/140) vs. 2.0% (4/204)] (p < 0.001). Reduced turnaround time of BRCA1/2 testing (p < 0.001) and the use of neoadjuvant chemotherapy (p < 0.001) were associated with BRCA1/2 mutation status identified prior to surgery. Although not statistically significant, higher incidence of developing contralateral breast cancer for BRCA1/2 mutation carriers who underwent ipsilateral surgery-only compared to those who underwent contralateral RRM was observed [12.1% (95% CI: 7.7–17.7%)] (p = 0.1618). Preoperative diagnosis of BRCA1/2 mutation could impact surgical decision-making for breast cancer patients to undergo risk-reducing surgery at the time of initial surgery.

A celebrity's announcement of having a BRCA1 mutation and undergoing bilateral risk-reducing mastectomy (RRM) has influenced breast cancer patients and women with high-risk breast cancer. The so-called "Angelina's effect" has impacted public's health awareness of hereditary breast cancer and increased rates of genetic counseling and tests for BRCA1/2 mutations 1-3 . This has also led to an increase in the rate of RRM among women with inherited BRCA1/2 mutation. BRCA1/2 mutation carriers who have developed breast cancer have a significant higher risk for ipsilateral breast recurrence with a median follow-up ≥ 7 years compared to non-carriers 4 . These patients also have an elevated risk of developing contralateral breast cancer, with a cumulative incidence up to 83% for BRCA1 and 62% for BRCA2 carriers [5][6][7] . Although the efficacy of contralateral RRM for women with BRCA1/2 mutation has been controversial, many studies have suggested a marked decrease of cancer in the opposite breast [8][9][10] . Therefore, newly diagnosed breast cancer patients with BRCA1/2 mutation can consider bilateral mastectomy as a treatment option regarding risk-reducing measures. Decision to undergo RRM needs to consider several factors such as patient's preference, surgeon's preference, subtype of breast cancer, clinical stage, family history, and genetic predisposition [11][12][13] .
In this current study, we investigated whether the timing of identification of BRCA1/2 mutation might affect surgical decision making. Some studies have shown that genetic diagnosis prior to surgery is associated with RRM 12,14,15 . With growth in the number of genetic testing and the development of sequencing technologies, we further investigated factors that enabled patients to obtain knowledge of BRCA1/2 mutation status prior to www.nature.com/scientificreports/ initial surgery. In addition, we analyzed oncologic outcomes of BRCA1/2 mutation carriers with breast cancer according to types of surgery.

Methods
We conducted a retrospective review of breast cancer patients tested for BRCA1/2 mutations between January 2008 and November 2019 who underwent curative surgery at Samsung Medical Center. Patients were eligible for this study if they underwent upfront surgery or surgery following neoadjuvant chemotherapy. Genetic testing was conducted at any time in relation to their diagnosis of breast cancer or surgery. They were identified to carry a deleterious BRCA1/2 mutation. We excluded patients who had a history of previous breast cancer, bilateral breast cancer, stage IV disease, those had not undergone a curative breast surgery, and male patients. Patients with breast cancer were routinely transferred to a dedicated genetic counselor for pre-test counseling and BRCA1/2 testing on the first visit to our outpatient clinic if the following criteria were met and patients agreed. Criteria for BRCA1/2 testing were as follows: family history of breast and/or ovarian cancer in first-or second-degree relative, diagnosis of breast cancer at 40 years old or younger, bilateral breast cancer, and multiple primary cancers including breast cancer. Post-test counseling was provided to individuals to inform results of testing and discuss treatment options.
We collected information from registered medical records, including age at breast cancer diagnosis, types of breast surgery, administration of neoadjuvant chemotherapy, pathologic staging, tumor characteristics, types of adjuvant therapy, and patient outcome. Patients were categorized as 'BRCA1/2 mutation status known prior to surgery' if their date of BRCA1/2 testing result was earlier than the date of surgery and 'BRCA1/2 mutation status known after surgery' if their results were yielded on the day of surgery or after surgery. We additionally calculated turnaround time (TAT) of each testing defined by the time interval between the submission of the patient's peripheral blood and delivery of the verified result.
BRCA1/2 testing was performed using genomic DNA isolated from peripheral blood. The sequencing method was initially Sanger sequencing. It was then transitioned to next generation sequencing after September 2016. We classified 'pathogenic' or 'likely pathogenic' results as positive mutation for BRCA1/2, while results of 'benign' , 'likely benign' , or 'uncertain significance' were classified as negative mutations 16 .
Comparison of patients with BRCA1/2 mutation identified prior to surgery and those after surgery was performed using the Chi-Square test or Fischer's exact test for categorical variables and Wilcoxon rank sum test for continuous variables. Multivariable analysis using logistic regression with backward variable selection was conducted to identify factors associated with BRCA1/2 mutation status known prior to surgery. Kaplan-Meier curves were used to analyze recurrence-free survival (RFS) and overall survival (OS). Log-rank tests were used to compare survival outcomes between patients who underwent ipsilateral surgery only and patients who underwent contralateral RRM. Cumulative incidence function plot was used to analyze the 5-year cumulative incidence of contralateral breast cancer. Gray's test was used to compare outcomes between the two groups. All statistical analyses were executed using SAS version 9.4 (SAS Institute, Cary, NC, USA) and R 3.6.1 (Vienna, Austria; http:// www.R-proje ct. org/). A P-value of < 0.05 was considered statistically significant.
All study procedures were conducted in accordance with the Declaration of Helsinki and were approved by the Institutional Review Board (IRB) of Samsung Medical Center, South Korea (IRB File No. 2020-12-145). The requirement for informed consent was waived by the IRB of Samsung Medical Center because of the retrospective nature of this study.

Results
Patient characteristics. Among 3950 breast cancer patients who were tested for BRCA1/2 mutation between January 2008 and November 2019 in Samsung Medical Center, 486 patients were BRCA1/2 mutation carriers. A total of 344 patients were eligible for the final analysis. Of the final cohort, 140 (40.7%) patients had their BRCA1/2 mutation status identified prior to surgery (Group 1), while 204 (59.3%) were aware of their mutation status after surgery (Group 2) ( Table 1). Women in Group 1 were younger (37.5 years vs 42.0 years, p < 0.001) with shorter TAT of BRCA1/2 mutation testing compared to Group 2 (21 days vs 29 days, p < 0.001). Almost two thirds (60.7%) of subjects in Group 1 received neoadjuvant chemotherapy as primary treatment, whereas 89.2% of patients in the Group 2 underwent surgery as primary treatment (p < 0.001). Higher clinical T stage and N stage were observed in Group 1 than in Group 2. Additionally, subjects in Group 1 were less likely to be treated with adjuvant chemotherapy (42.9% vs 73.4%) and radiotherapy (60.7% vs 80.8%). Proportions of BRCA1/2 mutations, histology, and biologic subtype were not significantly different between the two groups. Out of 14 patients with HER2 positive receptors, 12 received anti-HER2 therapy, and the remaining two with ductal carcinoma in situ did not receive such therapy.
Types of surgery according to preoperative knowledge of BRCA1/2 mutations. As shown in Fig. 1, contralateral RRM was performed for 45.0% (63/140) of patients with BRCA1/2 mutation identified prior to surgery. The rate was significantly (p < 0.001) lower for patients with BRCA1/2 mutation identified after surgery, of which only 2.0% (4/204) chose contralateral RRM as initial surgery whereas 72.5% (148/204) elected ipsilateral partial mastectomy and 25.5% (52/204) elected ipsilateral total mastectomy. Out of the whole cohort, 13 patients eventually underwent bilateral mastectomy after their initial surgery, including 8 patients whose BRCA1/2 mutation were identified after the initial surgery, 1 patient who was a known BRCA1/2 mutation carrier prior to initial surgery, and 4 who developed contralateral breast cancer.
Factors associated with BRCA1/2 mutation status known prior to surgery. Multivariable Fig. 3. A definite increase in the BRCA1/2 mutation carriers was observed in 2014. Since then, pro-

Discussion
In the present study, concomitant contralateral RRM was significantly more prevalent in breast cancer patients whose BRCA1/2 mutations were identified prior to surgery. We found that most patients underwent either ipsilateral partial mastectomy or ipsilateral total mastectomy when their mutation status was unknown at initial surgery. This is line with earlier studies reporting that preoperative genetic testing impacts surgical decision making in favor of RRM 12,14,15,17,18 . We additionally demonstrated several factors that enabled patients to gain knowledge about their BRCA1/2 mutation status prior to initial surgery. Although not statistically significant, www.nature.com/scientificreports/ we observed a higher risk of developing contralateral breast cancer for BRCA1/2 mutation carriers who underwent ipsilateral surgery-only compared to those who underwent contralateral RRM. Preoperative testing and diagnosis of BRCA1/2 mutation provided opportunities for patients to engage in risk reducing measures at the time of primary surgery. Short TATs of sequencing for BRCA1/2 mutation allowed patients to receive their results before the initial surgery. A study has assessed TATs of BRCA1/2 testing among ovarian cancer patients and revealed that amount of time it takes from initial genetic counseling to receipt of testing results is mainly driven by sample processing time, with median TATs varied among countries (2.7 weeks, 14.6 weeks, and 5.9 weeks in the US, Italy, and Spain, respectively) 19 . In our study, the median TAT in the group that mutation status was known prior to surgery was 21 days, whereas the median TAT in the group that mutation status was identified after surgery was 30 days. Traditionally, genetic testing to identify BRCA1/2 mutation has been conducted by Sanger sequencing. Sanger sequencing analysis generally requires 4 to 6 weeks of TAT in South Korea, making it difficult for patients to obtain genetic information preoperatively 20 . The development of next-generation sequencing (NGS) and its application for detecting genetic mutations have significantly shortened the analysis time and lowered the expenditure. Our institution adopted NGS for identifying BRCA1/2 mutations in late 2016. We have observed reduced TATs since then, proven by decreasing line graph shown in Fig. 3. Our previous retrospective study included 164 BRCA1/2 mutation carriers who underwent curative breast cancer surgery between 2004 and 2015 and identified only 15 patients with known BRCA mutation prior to surgery 15 . The current study expanded the inclusion period until late 2019. The number of patients with known BRCA1/2 mutation status prior to surgery was increased up to 140. Recent increase of patients with BRCA1/2 mutation status identified prior to surgery could be explained by the application of NGS on BRCA1/2 mutation testing which reduced TATs and increased the use of neoadjuvant chemotherapy.
Our study demonstrates that undergoing neoadjuvant chemotherapy has great odds of gaining knowledge of BRCA1/2 mutation status prior to surgery. It allows the time it takes for genetic testing to yield results while patients are treated. However, out of 107 patients who underwent neoadjuvant chemotherapy, 22 patients did not receive results preoperatively. This was mainly due to late identification of patients with high risk of hereditary breast cancer or delayed prescription of BRCA1/2 mutation testing.
Trained personnel, training programs, and formal accreditation processes are essential for accessibility to genetic counselling and testing. South Korea has implemented a national program for genetic counselling and genetic testing through national healthcare system, which has eventually increased proportions of genetic testing uptake to breast cancer incidence of 7.7%, the highest in Asia 21 . A prior study has indicated that early detection and genetic referral of patients who meet guidelines for BRCA1/2 testing are important for receiving testing results prior to surgery 12 . According to the American BRCA Outcomes and Utilization of Testing (ABOUT) Study 22 , lack of physician recommendation is the most commonly reported reason for not receiving genetic counseling. In our clinic, genetic counseling and testing are offered to patients with high risk of hereditary breast cancer on the first day of visit following breast cancer diagnosis. A concept of rapid genetic counseling and testing (RGCT) in newly diagnosed breast cancer patients is being emphasized. In order to maximize the likelihood that test results can be used for surgical decision, surgeons need to provide genetic testing at the point of care. They could even delay primary surgery slightly for a small subset of high-risk patients [23][24][25] . www.nature.com/scientificreports/ In this study, contralateral RRM demonstrated no survival benefit for BRCA1/2 mutation carriers with breast cancer. This is in line with a meta-analysis by Valanchis 4 , stating that there is no significant difference in breastcancer-specific survival between BRCA1/2 carriers who undergo contralateral RRM and breast conserving therapy. However, mortality benefit from preventing contralateral second primary cancer is likely to be apparent after an extended period 26,27 . Metcalf et al. 26 have noted a 48% reduction of death from breast cancer in BRCA1/2 mutation carriers who undergo contralateral RRM compared to those who undergo ipsilateral mastectomy after a 20-year follow-up.
The greater risk of developing contralateral breast cancer in BRCA1/2 mutation carriers compared to women with sporadic cancer supports the necessity of RRM for the unaffected side of breast for BRCA1/2 mutation carriers. The cumulative incidence for mutation carriers vs. sporadic cases is 20% vs. 2% at 5 years and 42% vs 9% at 12 years [28][29][30] . A previous case-control study conducted in our center has shown 4.0% of metachronous contralateral breast cancer recurrence in BRCA1/2-negative patients with risk factors for hereditary breast and/ or ovarian cancer 31 . However, the current study showed 8.4% of metachronous contralateral breast cancer recurrence in patients with BRCA1/2 mutations.
Several studies have concluded that contralateral RRM could reduce the risk among BRCA1/2 mutation carriers with a history of breast cancer 8,14,32,33 . Van Sprundel et al. 8 have reported a reduction of contralateral breast cancer risk by 91%, independent of the effect of bilateral prophylactic oophorectomy. Patients in our cohort who underwent contralateral RRM did not develop any contralateral breast cancer. The 5-year cumulative incidence of contralateral breast cancer patients who underwent ipsilateral surgery only was 12.1%. However, the difference in the risk of contralateral breast cancer between those two groups was insignificant. This might be due to a follow-up time as short as 15 months for the contralateral RRM group. The Korean National Health Insurance coverage for contralateral breast surgery in breast cancer patients with BRCA1/2 mutation carrier was initiated in March 2017 and the first patient in our cohort to receive contralateral RRM was in April 2014. These conditions might have resulted in a relatively short median follow-up time for the contralateral RRM group.
Decision to undergo RRM is a complex process, which involves weighing the benefits and harms. Of 140 patients with BRCA1/2 mutation status identified prior to surgery, 77 (55%) did not choose RRM as their initial surgery. Previous studies stated that reasons for not opting RRM include its lack of survival benefit, increased operative morbidity, potential to delay adjuvant therapy if complications occur, and possible poor cosmetic outcome 34,35 . Benefits for undergoing RRM include reducing the risk of contralateral breast cancer and possibilities of avoiding additional treatment. Patients who were unaware of their BRCA1/2 status and underwent partial mastectomy received radiotherapy. If they eventually undergo mastectomy and reconstruction afterwards, previous radiation could elicit complications. Of 30 patients who developed contralateral breast cancer in our cohort, 11 (36.7%) were triple-negative breast cancer. All 30 patients underwent subsequent curative surgery, 13 (43.3%) received chemotherapy, and 15 (50.0%) received radiotherapy. These additional treatments could have been spared if BRCA1/2 carriers with unilateral breast cancer underwent RRM at the time of initial surgery. Furthermore, although intensive surveillance could be opted for BRCA2 carriers due to low mortality, women electing for RRM want to avoid the lifelong distress related to intensive surveillance 36 .
This study has several limitations. First, this was a retrospective analysis of patients treated at a single institution, which could limit the generalizability of our findings. Second, the criterion used to categorize patients was 'date of BRCA1/2 testing result' , which could lead to overestimation of patients with BRCA1/2 mutation status known prior to surgery. Thirteen patients had no actual record about delivery of mutation status prior to surgery. However, all patients in the cohort received pre-test counseling. In addition, we routinely informed BRCA1/2 testing result even on the day before surgery. Third, factors that could affect surgical decision to undergo RRM were not analyzed. We demonstrated that reduced TATs and treating with neoadjuvant chemotherapy enabled patients to be aware of BRCA1/2 mutation status preoperatively. These factors could potentially affect surgical decision-making. However, their direct associations with increased RRM were not proved. Lastly, our oncologic outcomes have limitation for interpretation due to the short follow-up time for the ipsilateral surgery-only group and lack of adjustment for factors that affect outcomes such as age and use of adjuvant therapy. Further investigation on oncologic outcomes with an extended period of follow-up and adjustment for clinical variables is required. These limitations were countered by the significance of present study. We demonstrated that long TAT of gene testing as well as lack of systemized genetic counseling, delay in referral of patients to genetic counselor impeded patients' preoperative awareness of BRCA1/2 mutation status. As far as we know, it is the first to present such results in Asia.

Conclusions
In conclusion, our findings demonstrate that preoperative identification of BRCA1/2 mutation could affect surgical decision-making for breast cancer patients in favor of contralateral RRM. With the application of NGS for BRCA1/2 testing and the use of neoadjuvant chemotherapy, more patients are able to acquire their mutation status prior to initial surgery. Although undergoing contralateral RRM may not guarantee improved survival, it could reduce contralateral breast cancer risk, contribute to the reduction of additional future surgeries, unnecessary adjuvant radiation, and improve patient satisfaction. It is important that clinicians provide genetic counseling and testing at the time of diagnosis in order to inform patients of their BRCA1/2 mutation status prior to surgery and provide them with optimal treatment options.

Data availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.