Background

Early-onset breast cancer (EOBC) represents a distinct clinical and biological entity affecting women aged 18–45 years, regardless of family history [1]. Defining the threshold age that is considered "early-onset" for getting breast cancer fairly differs in different scientific studies, guidelines, or protocols [2, 3]. Although breast cancer (BC) occurs commonly in elders (50 years old or older) and only 12% of the new diagnoses of breast cancer in the United States are found in females younger than 45 years [4], EOBC has recently been undertaken as a research priority because of several objectives: BC in young women is more likely to be heritable than in older women and is usually more aggressive, harder to treat, and associated with a poor clinical outcome. Furthermore, EOBC patients inevitably experience certain long-term survivorship issues, such as contraception, management of perimenopause and menopause symptoms, and reproductive options (e.g., fertility preservation) [1]. Current evidence indicates that breast cancer is the top cancer-related cause of death in women aged < 45 years [5].

EOBC may be an independent predictor for "Hereditary Breast and/or Ovarian Cancer (HBOC) Syndrome" and fulfills the criteria for genetic testing of hereditary cancer-related genes (i.e., BRCA1/2) [6]. The probability of detecting a disease-causing variant in HBOC-related genes is maximized in an EOBC patient with a family history of multiple affected individuals on the same side [7, 8]. While the likelihood of identifying HBOC is about 80% when four or more BC patients diagnosed under age 60 are present in the same family [9], it is about 20% when a BC patient has only one more affected relative [10, 11].

BRCA1 and BRCA2 P/LP variants have been demonstrated to be related to an actionable genetic predisposition to BC. Thus, genetic testing of BRCA1/2 has got to be taken into consideration in order to ensure personalized breast cancer surveillance, appropriate risk reduction options, and therapeutic indications [12, 13]. In this study, we aim to present the BRCA1/2 mutational landscape of a Turkish cohort with EOBC and compare the hormonal profiles between BRCA P/LP variant carriers and the others.

Methods

Patients

From June 2020 to May 2022, unrelated subjects with EOBC who had applied to the department of medical genetics at Isparta or Eskişehir City Hospitals for genetic testing were recruited. All included individuals were also subjected to a detailed examination, including medical and family history and laboratory assessments, including the hormone receptor status of their tumors.

Genetic analysis

Genomic DNA was extracted from peripheral blood using the QIAamp DNA Blood Mini QIAcube Kit (Qiagen, Hilden, Germany) according to the standard procedures. Next-generation sequencing (NGS) technology was used to screen for disease-causing variants in BRCA1/2 genes. For NGS analysis, all coding exons and exon–intron junctions of BRCA1/2 were sequenced on the MGISEQ-200 platform (BGI, Shenzhen, Guangdong, China) using a custom-designed NGS panel (Twist Biosciences, San Francisco, CA, USA). All detected disease-causing variants were also confirmed by Sanger sequencing.

Patients found to be negative for BRCA P/LP variants were further tested for the detection of large genomic rearrangements (LGRs) in BRCA genes by the multiplex ligation-dependent probe amplification (MLPA) method. SALSA® MLPA Probemixes (P002 for the BRCA1 gene, P045 and P090 for the BRCA2 gene) were used for this purpose following the manufacturer's recommendations (MRC-Holland, Amsterdam, the Netherlands).

Variant interpretation

Each variant was annotated according to the Human Genome Variation Society (HGVS) nomenclature. NM_007300.4 (for BRCA1) and NM_000059.4 (for BRCA2) were used as the reference sequences.

To predict the effects of novel variants, in silico tools (PolyPhen2 [http://genetics.bwh.harvard.edu/pph2/], SIFT [https://sift.bii.a-star.edu.sg], MutationTaster [https://www.mutationtaster.org], CADD [https://cadd.gs.washington.edu], and MCAP [http://bejerano.stanford.edu/mcap/]) and population databases (1000G [http://www.internationalgenome.org] and gnomAD [https://gnomad.broadinstitute.org/]) were employed, and all identified variants were classified according to the American College of Medical Genetics and Genomics (ACMG) criteria [14].

Statistical analysis

Analyses were performed using IBM® SPSS Statistics version 28.0.1.1 (IBM Corp., Armonk, NY, USA). We used median, minimum and maximum values, mean, and standard deviation (SD) as descriptive statistics. Chi-square and Mann–Whitney U tests were utilized to compare differences between categorical variables. A P value of < 0.05 was considered statistically significant.

Results

Variants in BRCA1 and BRCA2 genes

Totally, 67 unrelated individuals with EOBC were included in this study, of whom 53 (79.10%) patients were BRCA-negative and 14 (20.90%) patients had BRCA P/LP variants. Among the BRCA-positive group, nine (64.28%) were BRCA2-positive and five (35.72%) were BRCA1-positive (Fig. 1).

Fig. 1
figure 1

BRCA1/2 P/LP variant status of all included patients

Full size image

All variants detected in the BRCA1 and BRCA2 genes are described in Table 1. Six BRCA1 variants were detected, all of which had been previously identified. Two novel BRCA2 variants, including a missense variant of uncertain significance (VUS) and a large genomic rearrangement (LGR), were identified (Fig. 2), together with 12 variants previously reported.

Table 1 All detected variants in this study
Full size table
Fig. 2
figure 2

Images of the novel variants identified in our study. Arrows indicate the site of mutation. a BRCA2(NM_000059.4): c.6664 T > G (p.Tyr2222Asp) variant in a heterozygous state is demonstrated by the IGV image. b Exon 6 heterozygous deletion of BRCA2 is illustrated by the Coffalyser.Net™ software

Full size image

Clinico-pathological features and mutational profiles

Table 2 summarizes the clinico-pathological findings of all cases. The median age at diagnosis for BRCA2 P/LP variant carriers was moderately higher than for BRCA1 P/LP variant carriers (40 vs. 37), although not statistically significant. Patient age groups were similar in BRCA1-positive vs. BRCA2-positive and BRCA-positive vs. BRCA-negative (P = 0.405 and P = 0.418, respectively).

Table 2 Baseline characteristics and clinico-pathological data of the cohort
Full size table

There was no statistically significant difference observed between BRCA-positive vs. BRCA-negative or BRCA1-positive vs. BRCA2-positive for the family history status (P = 0.075 and P = 0.648, respectively).

The percentages of estrogen receptor (ER)-positive and progesterone receptor (PR)-positive patients were 69% (45/65) and 60% (39/65), respectively. Conversely, our cohort was more likely to have human epidermal growth factor receptor 2 (HER2)-negative tumors (62% [36/58]). We did not observe any statistically significant association between hormone receptor status (ER, PR, and HER2), either between BRCA-positive vs. BRCA-negative or BRCA1-positive vs. BRCA2-positive. Of 61 subjects with an available hormonal status of breast cancer, 15% (9/61) were triple-negative (TN), and among them, only three patients were found to be BRCA P/LP variant carriers.

Discussion

In this study, we identified a Turkish cohort composed of 67 unrelated patients with EOBC, and their clinical, pathological, and genetic characteristics are listed in Additional file 1: Table S1.

The identification of clinically actionable P/LP variants in EOBC patients is especially pivotal given the risk of second primary malignancies, the need for proper surveillance, potential reproductive decision-making, and segregation testing of at-risk relatives that provides early diagnosis and prevention of the disease by determining pre-symptomatic variant carriers. Additionally, uncovering the mutational landscape of breast cancer in this age group may help to optimize therapeutic management; for example, knowledge of BRCA1/2 status may play a central role in both surgical decision-making and systemic treatment decisions [15].

We observed in our study group that about 21% of EOBC patients have P/LP variants in BRCA genes. The diagnostic yield of BRCA1/2 genetic testing deviates from other similar studies because many other independent factors, such as the size of the study groups, inclusion criteria, or referral/ascertainment bias, influence the results automatically. For example, while Akdeniz Odemis et al. [16] detected a 16.3% diagnostic yield in a total of 1202 EOBC patients, Biancolella et al. [17] calculated their diagnostic yield as 23.52% in a cohort of 51 EOBC women with or without a positive family history from Burkina Faso [16, 17]. Another study from Turkey found that BRCA1/2 genetic testing provided a diagnostic yield of 17.54% in 171 individuals with EOBC [18].

Out of the 20 variants observed in our study, two (10%) were novel. Missense alterations have been the most detected type of variant in BRCA2, and the majority of them are classified as VUS. As of September 2022, 6.188 BRCA2 VUS variants had been cataloged in ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/), of which 5.539 (89.51%) were listed as missense variants. Moreover, of the total 6.559 BRCA2 missense variants, 5.539 (84.44%) were VUS. While vigorous scientific efforts to validate the functional significance of VUS are being made [19,20,21,22,23,24], a large number of BRCA2 variants remain unclassified. The rate of receiving a VUS report has also been significantly associated with ethnic origin [25, 26]. Thus, the identification of VUS variants in BRCA genes is a significant clinical challenge in terms of risk assessment.

According to the NCCN® Guideline for Breast, Ovarian, and/or Pancreatic Cancer Genetic Assessment (Version 1.2024), VUS variants should not be used to revise the patient's medical management. In this instance, screening and risk reduction strategies should be recommended on the basis of personal and family history. Additionally, testing relatives for a VUS should not be done for clinical reasons unless there are conflicting interpretations of the data.

Specific recurrent variants in BRCA1/2 genes have been delineated in certain populations, such as the Ashkenazi-Jewish, French-Canadian, Brazilian, Italian, Icelandic, and Polish populations [27,28,29,30]. For Turkey, several studies reported that the c.5329dup (also known as c.5266dup in alternate nomenclature) variant in BRCA1 is the most common detected variant, which is possibly attributed to a founder effect [31,32,33,34,35]. In our study, we also detected the c.5329dup variant in an EOBC patient with a positive family history.

While the majority of germline disease-causing variants in BRCA1/2 are small-scale, LGRs are also defined in a notable proportion of patients originating from distinct populations [36,37,38,39,40]. A small number of studies from Turkey report different LGRs in BRCA genes with a variable frequency [31, 35, 41]. Although the prevalence of rearrangement in the BRCA1 gene is found to be higher [42,43,44,45], we herein reported two LGRs in BRCA2, one of which is novel (deletion of exon 6) detected in a 32-year-old affected individual. It is evident that LGRs in BRCA genes comprise a significant proportion of the Turkish population. For this reason, a targeted and affordable genetic testing strategy that also includes the analysis of LGRs should be developed in Turkey for the molecular characterization of high-risk individuals.

Conclusion

In conclusion, enhanced knowledge about the mutational spectrum of druggable genes can aid clinicians in the management of EOBC patients. This study contributes to existing literature by extending the molecular basis of BRCA1/2 genes.