The disease sites of female genital cancers of BRCA1/2-associated hereditary breast and ovarian cancer: a retrospective study

Disease sites of female genital tract cancers of BRCA1/2-associated hereditary breast and ovarian cancer (HBOC) are less understood than non-hereditary cancers. We aimed to elucidate the disease site distribution of genital cancers in women with the germline BRCA1 and BRCA2 pathogenic variants (BRCA1+ and BRCA2+) of HBOC. For the primary disease site, the proportion of fallopian tube and peritoneal cancer was significantly higher in BRCA2+ (40.5%) compared with BRCA1+ (15.4%) and BRCA− (no pathogenic variant, 12.8%). For the metastatic site, the proportion of peritoneal dissemination was significantly higher in BRCA1+ (71.9%) than BRCA− (55.1%) and not different from BRCA2+ (71.4%). With one of the most extensive patients, this study supported the previous reports showing that the pathogenic variants of BRCA1/2 were involved in the female genitalia’s disease sites.


BRCA2
Background Disease sites of female genital tract cancers of hereditary breast and ovarian cancer (HBOC) are less understood than non-hereditary cancers. We aimed to elucidate the disease distribution of ovarian, fallopian tube, and primary peritoneal cancers in women with the germline BRCA1 and BRCA2 pathogenic variants (BRCA1+ and BRCA2+) of HBOC.

Materials and methods
The ethics review board of the Japanese HBOC Consortium approved the establishment of the database and the future publication of our analysis results on February 18, 2016, to investigate Japanese HBOC patients' characteristics. The registered subjects were any women who consecutively underwent blood BRCA1 and BRCA2 genetic testing and agreed with this study from 2016 to 2018. We carried out the genetic testing of germline BRCA1 and BRCA2 in the 80 medical institutions where genetic counseling by certified specialists was available. We accepted all genetic testing purposes, including clinical practice for diagnosing BRCA1/2-associated HBOC or decision of PARP inhibitors' indication or translational research. For clinical practice, we usually used the testing criteria which NCCN published during the study period [1]. To collect rough data on a large number of BRCA1/2-associated HBOC cancer patients, we did not set any exclusion criteria on age, family history, the modalities and intervals for image testing, and personal history of BRCA1/2-associated HBOC cancers. To confirm whether BRCA1/2 variants were pathogenic or not, we used the Myriad Genetic Laboratories database of the latest version at the time of testing in 90.1% of patients. We also carried out the other tests at the facilities of the investigators or other research institutes. We finally checked the latest information on ClinVar (the archival database at the National Center for Biotechnology Information [NCBI], https://www.ncbi.nlm.nih.gov/clinvar/) in 2020 and confirmed there were no critical changes for pathogenicity. We showed the pathogenic variants registered in our database elsewhere [2]. We retrospectively reviewed the data on patients in the BRCA− (no pathogenic variant), BRCA1+, and BRCA2+ groups, and investigated primary disease sites and metastatic sites. We performed all statistical analyses with the JMP® Pro software program, ver. 14.0.0 (SAS Institute, Cary, NC, USA), and compared the proportions of disease sites between each group with Pearson's χ 2 test or Fisher's exact test. We considered P values of less than 0.05 to be statistically significant.

Discussion
The results of our study suggest that female genital cancer in BRCA2+ less frequently originate from the ovary than BRCA1+ and BRCA−. The disease site distributions by BRCA status are similar to those of previous Japanese prospective study [3] where the proportion of fallopian tube and peritoneal cancer was 40% in BRCA2+, and 15% in BRCA1+, and 14% in BRCA−, and our retrospective study with a larger number of BRCA1/2 patients supported these results. Although the proportion is somewhat different, HBOC-related female genital cancer in BRCA2+ also less frequently originates from the ovaries than BRCA1+ and BRCA− in the USA [4] and Australia [5].
Although the frequencies of peritoneal dissemination of BRCA1+ and BRCA2+ are almost identical, the Fig. 1 The proportion of primary disease sites of female genital cancers according to BRCA variants proportion of peritoneal dissemination was significantly higher when comparing BRCA1+ (71.9%) to BRCA− (55.1%) but not significantly different between BRCA2+ (71.4%) and BRCA− (P = 0.06). These results might suggest that BRCA2+ lacked the power to detect statistical significance due to the small sample size, and we will need further analysis with the larger population. The prevalence of metastasis has been controversial. A previous retrospective study in Italy showed that metastatic sites were the same between BRCA1+/2+ and BRCA− [6]. In contrast, a recent report showed an increased incidence of visceral metastases in Scottish BRCA1/2-defective ovarian cancer patients [7], and we need a more extensive study to explore whether ethnicity is involved in this difference.
The previous prospective study showed that BRCA1+ and BRCA2+ consisted of 9.9% and 4.7% of Japanese women with newly diagnosed ovarian cancer [3]. In this study, the proportion of BRCA1+ was considerably high (37.3%), and the proportion of BRCA2+ was also higher than expected (8.3%). We guess this result suggests selection and institutional bias due to actively conducted genetic testing by medical genetics specialists.
In conclusion, although our study is retrospective, this study, with one of the most extensive patients, supported the previous reports showing that the pathogenic variants of BRCA1/2 were involved in the female genitalia's disease sites.

Availability of data and materials
The data that support the findings of this study are available in the previous article [2].

Ethics approval and consent to participate
The ethics review board of the Japanese HBOC Consortium approved the establishment of the database on February 18, 2016, to investigate Japanese HBOC patients' characteristics. Among 80 medical institutions of this study, two institutions did not permit to include patients without informed consent, while others permitted an opt-out option.

Consent for publication
The ethics review board of the Japanese HBOC Consortium approved the future publication of this study on February 18, 2016, to investigate Japanese HBOC patients' characteristics.