In 1990, chromosome 17q21.2 was identified to be related to FBC (30), and the loss of 17q heterozygosity is frequently detected in familial breast and ovarian tumors (31). Using positional cloning, the BC susceptibility gene BRCA1 was found to be located on chromosome 17 at the 17q21 position (32). In 1995, the BRCA2 gene was identified on chromosome 13q12.36 (7). BRCA1 contains 24 exons and encodes a protein of 1,863 amino acids. The exon contains three mutation domains: a central N-terminal RING fingerprint domain (exons 2–7), a C-terminal BRCT domain (exons 16–24), and exons 11–13. The N-terminal RING fingerprint domain of BRCA1 binds to BRCA1-associated RING domain protein 1 (BARD1) (33), and the C-terminal BRCT domain binds to the phosphorylated protein (34, 35). BRCA2 contains 27 exons and encodes a protein of 3,418 amino acids. The N-terminal of BRCA2 contains the transcriptional activation domain, the middle section includes eight conserved motifs called BRC repeats that bind to RAD51, and the C-terminal contains the DNA-binding domain, two nuclear localization signals, and one TR2 (C-terminal RAD51 binding site) (36). BRCA1 and BRCA2 participate in RAD51-mediated homologous recombination (HR) for DNA repair. In the case of DNA double-strand damage, BRCA1 can be accurately located and phosphorylated at the damage site, and BRCA2 forms a complex with RAD51 to move it from the site of synthesis to the site of DNA damage processing (37). Furthermore, PALB2 acts as a bridge between BRCA1 and BRCA2. Its N-terminal coiled-coil motif binds to the coiled-coil motif encoded by exon 11 of BRCA1, and its C-terminal WD-40 repeats bind to the N-terminal of BRCA2 to form the BRCA1/PALB2/BRCA2 complex ( Figures 4 , 5 ). This complex is critical for HR after DNA double-strand breaks (38). In the absence of BRCA1 and BRCA2, HR is suppressed. DNA damage repair is alternatively carried out through the non-homologous end-joining pathway, which is more error-prone and leads to genome instability (7, 39). As a result, BRCA1 and BRCA2 play an essential role in maintaining genome integrity (40).

As tumor suppressor genes, mutations of BRCA1 (MIM 113705) and BRCA2 (MIM 600185) are closely related to the development of BC (41, 42). Carriers of these mutations have a 10–20 times higher risk of developing BC than those without BRCA mutations (43). Approximately 16%–25% of the FBC cases harbor harmful variants of BRCA1 and BRCA2 (44, 45). Although many genetic variants of BRCA1 and BRCA2 have been recorded, approximately 53%–55% of the variants occur in only one family. The most common BRCA1 variants were 185delAG (16.5%), 5382insC (8.8%), and missense variant C61G (1.8%). Meanwhile, the most frequently reported BRCA2 variants include 6174delT (9.6%), K3326X (2.6%), 3036del4 (0.9%), and 6503delTT (0.8%) (46). Notably, the mutation spectrum of BRCA1/BRCA2 varies significantly depending on geographic origin or ethnicity (47). For example, in China, the mutation rate of BRCA2 in FBC was higher than that of BRCA1 in the Shandong Province (48). However, other studies including different Chinese regions, such as Shanghai and the Henan Province, showed that the mutation rate of BRCA1 in FBC is higher than that of BRCA2 (49). The variant hotspots of BRCA1 in the Henan cohort were A3113G and A3780G, which was first reported in this population (50). Moreover, in the participants from Shanghai, two other new splice site variants in the BRCA1 gene (IVS17-1G>T, IVS21+1G>C) were discovered (51). BRCA2 gene mutations dominated in FBCs of the eastern Shandong population, and three BRCA2 gene variants, 2001delTTAT, 4099C to T, and 5873C to A, were discovered for the first time in this population (52). A new BRCA1 missense variant, c.5191C>A, was identified in the Taiwanese population, but whether it is pathogenic remains inconclusive (53). Moreover, the recurrent variant of BRCA1, 1100delAT, was found in the Shanghai, Jinan, Qingdao, and Shenyang populations (54), while the BRCA1 c.470_471delCT and c.981_982delAT variants were considered to be recurrent variants in the Hong Kong population (55). Additionally, racial differences have an impact on gene mutations. For example, a Singapore study that studied individuals with a personal or FH of familial breast/ovarian cancers (FBOCs) with the BRCA1 c.442-22_442-13del variant, found that this variant was more common in patients of Chinese origin. The study also implied that the BRCA1 c.442-22_442-13del variant could be a founding variant in Chinese individuals of ancient southern Han descent (56).

Recently, a study including 21,216 unselected patients with BC and 6,434 healthy controls from 19 medical centers throughout 11 Chinese provinces identified 1,958 BRAC1/2 variants through panel-based sequencing, of which 532 (27.2%) were pathogenic variants, and 858 (43.8%) were pathogenic variants of uncertain significance. The remaining 568 variants (29.0%) were benign. A total of 268 mutations in the BRCA1 gene and 242 mutations in BRCA2 were found in Asian patients with BC, most of which were meaningless mutations. Among these variants, researchers found 13 types of high-frequency lesions: p.Cys328fs, p.Asn704fs, p.Ser1862fs, and p.Ile1845fs in BRCA1; p.Ala938fs, p.Gln1037*, p.Ser1722fs, p. Tyr1894*, p.Leu1908fs, p.Glu2198fs, p.Ser2378*, p.Pro2802fs, and p.Thr3033fs in BRCA2. Eight of these variants have not been reported as high-frequency variants in Caucasians (57). Of note, single nucleotide polymorphisms (SNPs), a single nucleotide substitution at a specific position in the genome, may also contribute to BRCA1 changes. For example, two pathogenic SNPs were found on the 11th exon of BRCA1, which may be related to early-onset BC in the Chinese population (58).

Although various BRCA1/BRCA2 mutations are currently being investigated, variants of unknown significance that occur in BRCA1/BRCA2 account for 12%–13% of the cases, the functions of which are still unclear. Therefore, more research is required to determine the clinical importance of variants of unknown significance in BC (59).

The tumor suppressor gene TP53, located on chromosome 17p13.1, is the most commonly mutated gene in patients with cancer (60). The TP53 gene encodes the cellular tumor antigen p53, an intracellular transcription factor that controls multiple tumor suppressive pathways (61, 62). TP53 is defined as the “guardian of the genome” because of its role in conserving stability by preventing genetic mutations. The germline loss of TP53 can quickly lead to the formation of spontaneous cancers (63, 64). Moreover, cancers with wild-type TP53 predict a good prognosis (65), while those with mutant TP53 predict a worse prognosis (66–68).

Notably, about 30% of the BCs have TP53 mutations (60), most occurring in exons 5–8 (69). It has been reported that about 5% of the patients with BC with a positive FH and wild-type BRCA1 and BRCA2 carried a mutation in either CHEK2 or TP53 (70). In patients with Li-Fraumeni syndrome (LFS) with TP53 mutations, the risk of developing BC under 45 years of age is 18–60 times higher than that of the general population (71). One study of 150 patients with familial and early-onset BC revealed that the deletion variant of TP53 (643_660del18) appeared to have occurred only in the Chinese population (72).

The PTEN gene is located on chromosome 10q23.31, which encodes phosphatidylinositol 3,4,5-triphosphate 3-phosphatase. This has lipid phosphatase activity and works antagonistically to the PI3K signaling pathway. It also negatively regulates the mitogen-activated protein kinase (MAPK) pathway through its protein phosphatase activity (73). PTEN mutation is also associated with Cowden syndrome, an autosomal dominant genetic disease that increases the lifetime risk of BC. In PTEN variant carriers, the lifetime risk of BC is estimated to be 67%–85% (74).

PALB2 is located on chromosome 16p12.2 and encodes a protein that interacts with BRCA2 as a functional partner. Thus, PALB2 affects the nuclear localization and stability of BRCA2 and can also act as a bridge between BRCA1 and BRCA2 (38, 75). The biallelic PALB2 mutation causes a new Fanconi anemia subtype, FA-N, also called FANCN (76, 77). Recent studies have found that germline mutations of PALB2 exist in families with BC, indicating that PALB2 may be a tumor suppressor for FBC (78, 79). Moreover, individuals with PALB2 mutations have a 2.3 times higher risk of developing BC (43). One study conducted in Finland showed that PALB2 c.1592delT is a founder variant, which causes truncated protein products with functional defects that cannot support BRCA2 to complete DNA repair. As such, females with this PALB2 mutation have a four times higher risk of developing BC (80). By age 50, the cumulative risk of BC in women with such mutations is estimated to be 14%, and by age 70, it increases to 35% (81). Zhang et al. also identified three harmful variants (c.3271delC, c.103C>T, and c.3035C>T) of PALB2 in 305 cases of FBC in China, and the mutation rates were all 0.33% (82). In addition, studies have found that women with PALB2 mutations from families with a history of BC have a greater risk of BC than those with no FH of BC (83).

The CDH1 gene is located on chromosome 16q22.1 and encodes the tumor suppressor, E-cadherin (a transmembrane calcium-dependent protein involved in cell-cell adhesion) (84–87). CDH1 mutation is closely related to lobular breast cancer, and the lifetime risk of developing BC in those with a CDH1 mutation is approximately 39%. Moreover, somatic and epigenetic changes in the CDH1 gene are frequently detected in sporadic tumors, including BC, and are associated with worse survival rates (88, 89). Pathogenic mutations of CDH1 are also the leading cause of hereditary diffuse gastric cancer (HDGC) (90), whose first clinical manifestation could be lobular breast cancer (91).

The STK11 gene is located on chromosome 19p13.3 and encodes a serine/threonine kinase that regulates many physiological processes, including energy metabolism and cell polarity (92). Most importantly, it is estimated that 32%–54% of the STK11 gene mutation carriers under 60 years of age have a high risk of developing BC during their lifetime. The risk of developing BC for these carriers is 8% at the age of 40, increasing to 32% at the age of 60. Compared with the general population, the risk of BC in this subpopulation is seven times higher (93).

The NF1 gene is located on chromosome 17q11.2 and encodes neurofibromin, a cytoplasmic protein that regulates multiple critical signaling pathways, such as the Ras-cAMP pathway (94). Neurofibromin can increase the guanosine triphosphate (GTP) hydrolytic rate of Ras and, thus, plays a tumor-suppressive role by reducing Ras activity (95). Females under the age of 50 with NF1 gene mutations have an up to five times higher risk of BC morbidity and mortality (96), and the risk was high among women under 40 years of age (97).

Cell cycle checkpoint kinase 2 (CHEK2) is involved in DNA damage and replication checkpoint responses and has been widely considered a BC-sensitive factor (98). The CHEK2 gene is located on chromosome 22 and is critical for cell cycle regulation. The pluripotent kinase CHEK2 is important for DNA damage response by causing cell cycle arrest or apoptosis. CHEK2 phosphorylates the TP53 tumor suppressor protein and prevents its degradation during DNA damage, leading to G1 cell cycle arrest (99). In addition, CHEK2 can induce cell apoptosis independent of TP53 by phosphorylating the tumor suppressor, promyelocytic leukemia protein (100). CHEK2.1100delC is a protein truncation variant first found in a family with LFS (101). Moreover, CHEK2.1100delC can be detected in 5.1% of the non-carriers of the BRCA1 or BRCA2 mutations with FH in northern Europe (102), and such mutation in females can increase the risk of developing BC by 2–3 times, and by 10 times in males (103, 104). A study based on a Chinese population with the CHEK2 mutation in 74 patients with BC with FH and 50 control subjects identified that the missense variant of CHEK2 in these cases was 1111C>T (His371Tyr) instead of CHEK2.1100delC, which implies that the CHEK2.1100delC variant was relatively rare in the Chinese population and that CHEK2 c.1111C>T mutation might be related to the genetic susceptibility of BC (105). Another study screened the CHEK2 coding sequence of 118 cases of Chinese FBC negative for BRCA1 and BRCA2 mutations and confirmed that the incidence of CHEK2 c.1111C>T in FBC was higher than that in non-selective BC (4.24% vs. 1.76%) (106).

The ATM gene is located on chromosome 11q, which encodes a phosphatase essential for substrate phosphorylation involved in DNA repair and cell cycle regulation (107). The ATM gene mutation rate in the general population is about 1%, and ATM heterozygotes have an increased risk of developing BC (108), especially women over 50 years of age (109). The lifetime risk of developing BC in patients with ATM monoallelic mutations is about 17%–52% (110, 111). A recent meta-analysis showed that the fully deleterious variants of ATM could cause a BC risk 2–4 times higher than that of the general population (112).

The NBN gene is located on chromosome 8q21.3 and encodes the Nibrin protein, responsible for interaction with DNA repair proteins involved in DNA double-strand break signaling (113, 114). Carriers of NBN monoallelic mutations have a significantly increased risk of BC, with an estimated odds ratio of 3.1 (115). Moreover, the truncated c.657del5 variant of NBN is also regarded as a high-risk factor for BC (116).

Proteins encoded by the RAD51 gene are important for DNA double-strand break repair. Seven RAD51 paralogs have been identified in mammals, including RAD51, RAD51B, RAD51C, RAD51D, XRCC2, XRCC3, and DMC1 (117). Based on the present knowledge, mutations in RAD51C and RAD51D are closely related to carcinogenesis (118, 119). For example, RAD51C and RAD51D gene mutations can be detected in patients with FBOC (120). One study from China conducted a genetic analysis of 273 patients with BRCA1/2-negative FBC and identified four previously unknown amino acid substitution variants in the RAD51C gene, the 4C>G (R2G) located in exon 1, 635G>A (R212H), and 644A>G (D215G) in exon 4, and 882G>C (Q294H) in exon 6. The R212H variant was unlikely to be pathogenic, as it only existed in healthy individuals. However, the R2G and D215G variants were suggested to be pathogenic to Chinese women after analysis using the SIFT, PolyPhen, and PMut algorithms (121). Moreover, a recent study showed that the protein-truncating variants in RAD51C and RAD51D are related to FBC. The estimated relative risks of BC associated with RAD51C and RAD51D mutations were 1.99 and 1.83, respectively. Therefore, they were classified into the moderate-risk category based on the current National Institute for Health and Care Excellence guidelines (57, 122).

MLH1, MSH2, MSH6, and PMS2 encode DNA mismatch repair (MMR) proteins responsible for DNA mismatch repair (123, 124). Mutations in these genes can cause Lynch syndrome. Several studies have found that MMR gene mutations frequently exist in patients with BC (125), but the association between Lynch syndrome and BC is unclear (126).

With the widespread application of high-throughput sequencing, a large number of genes related to BC risk have been identified, such as BAP1, PPM1D, and ABRAXAS1 (127, 128). However, their exact connection with BC and its penetrance remains unclear.

HBOC is an inherited disorder that was first reported in the 1970s (144). The clinical characteristics of HBOC include young age of onset, multiple family members with BC or OC, or both, or one family member with both BC and OC, or bilateral BC (145). The most common genetic changes associated with HBOC are mutations in the BRCA1 and BRCA2 genes (30). Approximately 60% of the typical families with HBOC harbor BRCA mutations (70). In China, this proportion was approximately 15.8% (146). Finally, because BRCA gene mutations are common in FBC, some patients with HBOC exhibit similar symptoms as patients with FBC.

LFS was first discovered in 1969 as an autosomal dominant malignant tumor syndrome caused by a germline mutation in the TP53 tumor suppressor gene. The TP53 gene mutation can be detected in approximately 50%–70% of the LFS cases, which is much higher than its prevalence of 1% in BC cases (147, 148). Furthermore, patients with LFS may suffer from multiple cancers, including BC, brain tumors, soft tissue sarcoma, leukemia, osteosarcoma, adrenal cortical malignancies, and broncho-alveolar lung cancer. The lifetime risk of developing BC in patients with LFS is estimated to be 25%–79% (149, 150). Some patients with BC of LFS show familial aggregation. Therefore, the NCCN guidelines recommend that women with TP53 pathogenic variant/likely pathogenic variant undergo a clinical breast examination every 6–12 months beginning at the age of 20 and a breast magnetic resonance imaging (MRI) with contrast screening every year from 20–75 years (151).

HDGC is an autosomal dominant genetic disease caused by a CDH1 mutation. Patients with HDGC are susceptible to lobular breast cancer (LBC). The cumulative risk of LBC in women with CDH1 mutations is estimated to be 39%–52% by age 80 (143, 152, 153).

Ataxia telangiectasia is an autosomal recessive genetic disease that is closely related to ATM mutations. The clinical manifestations include eyelid telangiectasia, cerebellar ataxia, and immunodeficiency (140). Some studies have shown that heterozygous carriers of ATM mutations have an increased risk of BC. A meta-analysis including three cohort studies of relatives of patients with ataxia telangiectasia estimated that the relative risk of BC in patients with ataxia telangiectasia is approximately 18% at 80 years of age (112). Moreover, if one has a history of radiation exposure, the risk increases further (154).

Cowden syndrome is an autosomal dominant genetic disease, which is clinically characterized by hamartoma-like lesions, pathognomonic skin lesions, benign breast disease, early-onset BC, and thyroid cancer. This disease is closely related to the PTEN/MMAC1/TEP1 gene mutations. Furthermore, approximately 75% of the female patients with Cowden syndrome harbor multiple benign breast lesions, such as fibroadenoma, cystic lesions, and ductal hyperplasia (155). A French study estimated that the cumulative BC risk in patients with Cowden syndrome was between 25 and 85%, and the cumulative incidence of BC by the age of 70 was 77% (95% CI: 59%–91%) (156).

Peutz-Jeghers syndrome is a rare autosomal dominant genetic disease caused by mutations in the STK11 gene. The clinical manifestations of Peutz-Jeghers syndrome include hamartoma-like polyps in the gastrointestinal tract, melanin deposition in the skin and mucous membranes, pancreatic cancer, and mucocutaneous periorificial lentiginosis. The lifetime risk of patients with Peutz-Jeghers syndrome developing BC is 24%–54%, and the average age of onset is approximately 39 years (157, 158). The NCCN guidelines recommend that carriers of STK11 gene mutations undergo clinical breast examinations every 6 months and annual mammography and breast MRI examinations from the age of 25 (159).

Genetic counseling is a counseling process for relatives who have genetic diseases or are at risk of infection to provide disease occurrence and early detection or prophylactic intervention methods. Genetic counseling can prevent genetic diseases and provide insight regarding reproductive options and should be conducted by a well-trained professional counselor. The professional counselor conducts family investigation and analysis by evaluating the proband in the family and estimating the possibility of the disease in the offspring.

Genetic counseling for healthy family members of FBC is an important step in the screening process for high-risk populations and should be performed before blood is drawn to collect DNA to identify pathogenic mutations. However, not all individuals are required to receive genetic counseling. Eligible participants should meet one of the following criteria: 1. Personal medical history with genetic risk of i) early-onset BC (≤35 years old); ii) both BC and OC; iii) simultaneous cancers other than BC and OC; and 2. Significant FH of BC/OC: i) BC ≤50 years old; ii) bilateral BC; iii) at least three family members with ovarian, peritoneal, or tubal cancer; iv) at least one male family member with BC; v) multiple cases of BC in the family; vi) at least one primary cancer patient in the family with BRCA-related diseases; and vii) Nordic Jewish descent ( Figure 7 ) (4, 160).

During genetic counseling, genetic counselors can use family risk stratification tools, such as the Manchester Scoring System (MSS), Family History Screen-7, Pedigree Assessment Tool, Referral Screening Tool, and Ontario Family History Assessment Tool (Ontario-FHAT) to distinguish the participants who need follow-up consultation. Healthy populations with a high risk of BC need advanced genetic counseling or BRCA gene mutation testing (162, 163). For others with a low risk of BC, routine screening strategies are recommended (161). Following the BC screening strategy for the Chinese population (2021 version), the low risk population undergoes X-ray or B-ultrasound examinations every 1–2 years, while high-risk groups undergo annual X-ray and B-ultrasound examinations and breast MRI when necessary. Patients that fall between low and high risk undergo X-ray and B-ultrasound examinations every 1–2 years (164).

Genetic testing is conducive to the early detection of high-risk groups of BC, providing preventive measures, and improving prognosis (165). Currently, only two genes (BRCA1 and BRCA2) are routinely used for BC genetic testing, and the test subjects are usually limited to women with an FH of BC and OC. However, given that genetic testing is relatively expensive and may cause adverse socio-psychological effects, testing should target individuals that most likely have gene mutations (166, 167). For example, BRCA1/2 testing is only provided to individuals with a mutation probability greater than or equal to 10% in many regions such as British (168). Therefore, advanced genetic counseling or models should be used to predict the likelihood of BRCA1 and BRCA2 mutations before performing genetic testing. When a patient qualifies as a test participant, they should fully understand the advantages and disadvantages of genetic testing and provide fully informed and written consent before genetic testing is possible (5).

It is generally recommended that female test subjects younger than 18 years of age undergo BRCA genetic testing. The focus of mutation analysis depends on the FH. If there are family members with BC with precise genetic mutation sites or ethnic groups carrying specific genetic mutation sites, these sites can be tested first (169). If the test subject does not have these characteristics, the entire specific gene sequence must be tested to exclude other site mutations. It is worth noting that women who are clinically negative for BRCA mutations are also at risk for BC (170). Walsh et al. found that among high-risk American families for BC and whose BRCA1 and BRCA2 gene test results are negative, approximately 12% still carry genetic mutations in BRCA1, BRCA2, CHEK2, TP53, and PTEN (70). These false negatives can easily lead to a missed diagnosis in high-risk groups. In addition, most of the gene mutations in FBC are unrelated to BRCA1, such as ATM, CHEK2, and BARD1, in the homologous recombination pathway (171). Further studies have shown that the truncated variants of ATM and CHEK2 are more closely associated with estrogen receptor (ER)-positive BC, while BARD1, BRCA1, BRCA2, PALB2, RAD51C, and RAD51D are more closely related to ER-negative BC (29). Therefore, it is believed that both sequencing and global screening for rearrangements should be performed for women with high risk for BC who are BRCA1/2 negative. For families with only negative BRCA1/2 sequencing test results, multiplex ligation-dependent probe amplification testing can be performed.

If the result of genetic testing is positive, that indicates the detected mutant gene is the same as the mutant gene of the BC patient in the family; therefore, the test subject needs to continue the screening process. Subsequent BC screening includes regular breast self-examination and imaging examinations. If the BC patient is not a mutation carrier, the test subject will temporarily not be listed as a high-risk patient. Those participants whose tests are all negative will temporarily not be classified as high-risk groups, and the general population screening process is recommended (161).

With the development of genetic testing, many new testing methods have emerged, such as multigene panel testing, which can detect multiple gene loci simultaneously. In a study of 35,000 women with BC, the 25-gene panel showed mutations most commonly in BRCA1, BRCA2, CHEK2, ATM, and PALB2 (172). Moreover, NCCN Guidelines 2017 stated that if more than one gene can explain hereditary cancer, polygenic testing may be required to be more effective and cost-effective. A recent study also demonstrated that population-based multigene panel testing is more cost-effective than individual BRCA1/BRCA2 testing (173). Moreover, a current study in China showed that multigene panel testing could increase the mutation detection rate in patients with a high risk of BC (137). Thus, although multigene panel testing still has shortcomings for practical applications (174), it may become an important technology in future FBC screening strategies.

From a technical point of view, it is practical to accurately detect BRCA1/2 mutations and other susceptible gene mutations in China. However, there is currently no uniform standard for the hot spots of FBC mutant genes in the Chinese population. Moreover, there is no uniform standard for genetic testing. The shortage of professionals to provide genetic counseling results in the imperfect prevention of uncertain genetic mutation carriers. Therefore, establishing an entirely professional team and standardized genetic counseling processes for FBC prevention and treatment are urgently needed.

Only 5%–10% of the BC susceptibility gene carriers will eventually develop BC due to gene variations. Predicting the risk of BC in individuals with FH is helpful for the clinical prevention and treatment of FBC. At present, many clinical models that combine the patient’s personal history, FH, and other factors to assess the risk of BC have been proposed (175), although no standardized prediction method has been widely accepted.

The Gail model is a statistical analysis model based on case-control data, which integrates the risk factors of BC, including BC history, age, age at menarche, age at first birth, number of first-degree relatives with BC, breast biopsy results, and race. Gail-1 in 1989 was initially used to predict the risk of invasive BC and carcinoma in situ in white women who underwent mammography each year (176). Gail et al. modified Gail-2 in 1990 to improve its predictive power (175). Since then, the Gail model has been validated in many different populations (white, African American, Hispanic, Asian American, American Indian, and Alaska Natives).

The Claus model estimates the lifetime risk of BC based on the FH of first- and second-degree relatives with BC and OC (177). The reference parameters included the age of onset of the paternal and maternal lines and information on cancer history (178).

The Couch (Penn I) model was initially used to predict the probability of BRCA1 mutations in 169 families with BC. The Penn II model is based on logistic regression analysis and integrates specific clinical features (such as bilateral BC) to predict the possibility of BRCA1/2 mutations in individuals or families. Studies have shown that the prediction accuracy of the Penn II model is higher than that of the Penn I and Myriad II models (179).

The BRCAPRO model is based on the Bayes theorem, the prevalence of BC and OC in first- and second-degree relatives, and the age of onset of the disease in family members to screen BRCA1/2 gene mutation carriers. The model was developed into a computer software in 2002 (180, 181). At present, the model is continuously updated, and its feasibility has been confirmed (182).

The Myriad model is based on 7,461 samples tested for BRCA1/2 mutations and 2,539 samples of detected mutations in three descendants of Ashkenazi Jewish ancestry (including FH, age of onset of FBOC, and presence of invasive cancer) to establish a model to predict the possibility of carrying mutations. The model has been publicly released at present, and the sample size is being expanded for research and updates (183).

The BOADICEA model was developed based on the complex isolation analysis of the occurrence of BC and OC in the combined data of 1,484 BC cases and 156 multi-case families (184). This model not only allows for simultaneous effects of both BRCA1 and BRCA2 but also allows for the effects of genetic modifiers and the multiplier effect of low penetrance genes on BC risk (185).

Evans et al. used whole-gene screening technology to screen the DNA samples of 422 non-Jewish patients with a history of BC/OC for BRCA1 mutations and performed BRCA2 screening for 318 subsets. After combining the screening results and FH, a simple scoring system, the Manchester scoring system, was designed to predict pathogenic mutations (186). In 2009 and 2017, MSS2 and MSS3, respectively, were released, where the pathological weight was considered in the scoring system, and the corresponding score was adjusted (187, 188). For subjects in Northwest England, a total MSS3 score of 15–19 is equal to the 10% threshold, and a score of 20 points is equal to the 20% threshold, but the threshold may need to be adjusted when applied to other populations.