A Classification and Evaluation Method for BRCA1 / 2 Gene Variations

Abstract

The present invention discloses a method for classifying and evaluating BRCA1 / 2 gene mutations, comprising the following steps: (1) annotating the coding region and amino acid changes of BRCA1 / 2 gene mutations according to the rules of HGVS for a fixed transcript to obtain evidence, the transcript of BRCA1 being NM_007294 and the transcript of BRCA2 being NM_000059; (2) classifying the evidence obtained in step (1) according to the improved ACMG guidelines into population data, computer prediction data, functional research data, co-segregation data, allele data, database data and phenotype data, and then grading according to pathogenicity as PVS, PS, PM and PP, and grading according to benign conditions as BA, BS and BP; (3) comprehensively classifying and evaluating the obtained gene mutations based on the scores of the classification and grading of each evidence. The evaluation of the present invention is accurate and easy to implement.

Description

Technical Field

[0001] The present invention belongs to the technical field of gene mutation evaluation, and particularly relates to a classification evaluation method for BRCA1 / 2 gene mutations. Background Art

[0002] BRCA1 and BRCA2 genes play a crucial role in the homologous recombination repair pathway (HRR). In vivo, the HRR pathway is the main repair mechanism for DNA double-strand damage. If BRCA1 / 2 genes mutate and become inactivated (pathogenic mutations), it may lead to the loss of the repair function of the HRR pathway. Further, cells may develop cancer due to the accumulation of DNA damage. Clinically, pathogenic mutations in BRCA1 / 2 have also been proven to be related to various hereditary tumors. Studies have shown that the risk of breast cancer in women carrying pathogenic variants of BRCA1 / 2 is as high as 45-65% in their lifetime, and the risk of ovarian cancer is as high as 11-39%, which is much higher than that of non-carriers. A similar situation also occurs in male prostate cancer. In addition, clinical trials have shown that breast cancer or ovarian cancer patients carrying pathogenic variants of BRCA1 / 2 genes can benefit from the treatment of PARP inhibitors. Currently, various PARP inhibitors such as olaparib, niraparib, and talazoparib have been approved by the FDA or NMPA for marketing. Therefore, the detection of BRCA1 / 2 mutations has important clinical significance.

[0003] Classifying and interpreting the detected mutations is a key step in the BRCA1 / 2 mutation detection process. According to the classification systems of the International Agency for Research on Cancer (IARC), the American College of Medical Genetics and Genomics (ACMG), and the Evidence-based Network for the Interpretation of Germline Mutant Alleles (ENIGMA), BRCA1 / 2 mutations can be classified into the following 5 categories from high to low risk: category 5 - pathogenic, category 4 - likely pathogenic, category 3 - variant of uncertain significance, category 2 - likely benign, category 1 - benign. Among them, categories 4 or 5 have clear guiding significance in clinical practice. Categories 1 and 2 theoretically do not affect the function of genes, while category 3 is a variant with insufficient evidence or conflicting evidence.

[0004] BRCA1 / 2 genes are tumor suppressor genes and do not have hotspot areas. At the same time, since the total length of the gene coding region is 15Kb, BRCA1 / 2 mutations are not only numerous but also widely distributed. Currently, more than 30,000 BRCA1 / 2 gene mutations have been reported, and a large number of new mutations are reported every year. A large number of these mutations cannot be accurately classified due to insufficient evidence. In the most commonly used Clinvar database (https: / / www.ncbi.nlm.nih.gov / clinvar / ), a large number of variants of unclear significance are recorded. These situations have brought considerable challenges to the classification of BRCA1 / 2. Therefore, a clear, convenient and effective method is needed to classify BRCA1 / 2 mutations. At present, many institutions have formulated standardized classification rules for BRCA1 / 2 mutations, including the "Standards and Guidelines for Classification of Sequence Variants" (2015 edition) formulated by ACMG. [9] The "BRCA1 / 2 Gene Mutation Classification Standard" (version 1.1) formulated by ENIGMA and the "Chinese Expert Consensus on BRCA Data Interpretation" issued by the Chinese Medical Association in 2017, these rules have their own shortcomings, so there is no unified and recognized classification standard in the industry.

[0005] ACMG's 2015 revised "Standards and Guidelines for Classification of Sequence Variants" is a relatively widely used rule guide in the industry. The classification rules are essentially a weighting system for evidence, which divides the types of evidence into eight categories (population data, computational prediction data, functional data, co-segregation data, new data, allele data, other databases, and other data), and divides them into seven levels according to their strength (benign independent evidence BA, benign strong evidence BS, benign supporting evidence BS, very strong pathogenicity evidence PVS, strong pathogenicity evidence PS, moderate pathogenicity evidence PM, pathogenicity supporting evidence PP). The collected relevant evidence is weighted, and then the variants are classified and judged according to the classification criteria. ACMG's classification rules are relatively comprehensive, with clear evidence classification and strong enforceability, which are suitable for the process of classification judgment rules. However, the ACMG classification rules are a universal rule for genetic genetic variations of all monogenic genetic diseases. There are many inapplicable places when it comes to the implementation level of genes; secondly, the definition of some evidence in the ACMG classification rules is vague, which may lead to subjective judgments in the implementation process. Therefore, there is much room for improvement in the ACMG classification rules.

[0006] ENIGMA formulated the "BRCA1 / 2 Gene Mutation Classification Standard" in 2015 and revised it in 2017. This is a relatively recognized classification rule in the industry specifically for the BRCA1 / 2 gene. The ENIGMA standard describes in detail the conditions required to divide the target mutation into 1-5 categories. The ENIGMA standard is tailor-made for the classification and interpretation of BRCA1 / 2 gene variations. The classification conditions are detailed and clear, and there are relatively few places where users need to make subjective judgments. However, in the ENIGMA standard, a multifactorial likelihood model is uniformly used to score various clinical evidence (such as family evidence, functional experimental evidence, allele evidence, etc.), and this model is not applicable in many cases, and there is a certain threshold for users, which makes the enforceability of the ENIGMA standard relatively difficult.

[0007] In 2017, the Pathology Branch of the Chinese Medical Association issued the "Chinese Expert Consensus on the Interpretation of BRCA Data", which mainly uses the ENIGMA standard as a framework, but lacks attention to some clinical evidence in the classification process, such as evidence of family co-segregation and evidence of disease-healthy population controls.

[0008] The essence of the technical solution disclosed in CN 110957006A is an upgraded and optimized version of the ACMG guidelines classification rules. It has modified some of the ACMG evidence details and added and reduced some evidence modules to make it more suitable for the classification and interpretation of BRCA1 / 2 gene mutations. However, there are still many shortcomings. First, its optimization is mainly aimed at LOF (loss of function) mutations, which may be more effective in judging this type of mutation, but the classification effect for other types of mutations (for example, missense mutations, synonymous mutations) is not obvious; secondly, there are still many conditions in the optimized rules that need to rely on the user's subjective judgment, so there is still a lot of room for optimization. There is reason to believe that using this method for mutation classification will still result in a large number of mutations that cannot produce an accurate classification result. Summary of the invention

[0009] The purpose of the present invention is to overcome the defects of the prior art and provide a classification and evaluation method for BRCA1 / 2 gene mutations.

[0010] The technical solution adopted by the present invention is as follows:

[0011] A classification and evaluation method for BRCA1 / 2 gene variation comprises the following steps:

[0012] (1) Annotate the coding region and amino acid changes of BRCA1 / 2 gene variants according to the HGVS rules for a fixed transcript to obtain evidence. The transcript of BRCA1 is NM_007294, and the transcript of BRCA2 is NM_000059;

[0013] (2) Classify the evidence obtained in step (1) according to the improved ACMG guidelines into population data, computer prediction data, functional research data, co-segregation data, allele data, database data, and phenotype data, and then classify them according to pathogenicity levels as PVS, PS, PM, and PP, and classify them according to benign conditions as BA, BS, and BP; the above improvements include at least one of the optimization of item PVS1 in the PVS module, the optimization of the PM module, the optimization of items PP1, PP3 to PP5 in the PP module, the optimization of item BA1 in the BA module, the optimization of item BS1 in the BS module, the optimization of items BP4 and BP6 to BP7 in the BP module, and the optimization of the combined usage of the evidence module;

[0014] (3) Classify and evaluate the obtained gene variants as pathogenic, likely pathogenic, of uncertain significance, likely benign, and benign based on the scores of the classification and grading of each evidence.

[0015] In a preferred embodiment of the present invention, the optimization of item PVS1 in the PVS module includes at least one of the following:

[0016] Include the variant G>non-G of the base at the 3'-end of the BRCA1 / 2 gene exon in the PVS1 evidence item; the base at the 3'-end of the exon is also a conserved sequence in mRNA splicing, and multiple variants occurring at the exon end of the BRCA1 / 2 gene have been verified to cause abnormal mRNA splicing. This optimization can improve the effectiveness of the variant classification rule

[0017] For variants in the classical splicing region and the last base at the 3'-end of the exon, exclude variants that are known or predicted to result in in-frame RNA isoforms. This optimization can improve the execution efficiency and accuracy of variant classification.

[0018] More preferably, the optimization of item PVS1 in the PVS module further includes: limiting the loss-of-function variants occurring before amino acid position 1855 of BRCA1 gene or position 3309 of BRCA2 gene to be included in item PVS1. The original ACMG rule for all genetic genes only recommended cautious interpretation of LOF variants at the 3'-end position of the gene without giving specific positions. This optimization can clarify the positions, unify the rules, and avoid the influence of artificial subjective factors.

[0019] In a preferred embodiment of the present invention, the optimization of the PM module includes: deleting the three evidence classifications of PM1 / PM3 / PM6 in the original ACMG guidelines. The description of PM1 for hotspots is not applicable to the BRCA1 / 2 genes. Based on the collection and analysis of existing BRCA1 / 2 variant data, there are no hotspots in BRCA1 / 2; PM3 is only applicable to recessive genetic genes; PM6 is only suitable for diseases with a relatively high gene penetrance and a relatively low age of onset, so none of them are applicable to the classification of BRCA1 / 2 gene variants. This optimization can improve the execution efficiency of variant classification.

[0020] Further preferably, the optimization of the PM module further includes: adding the gnomAD database including exome and genome as a population frequency reference database in item PM2. This database is currently the largest global population variant database. Adding this database can improve the effectiveness and execution efficiency of variant classification.

[0021] In a preferred embodiment of the present invention, the optimization of items PP1 and PP3 to PP5 in the PP module includes at least one of the following:

[0022] Determining the number of gametes required for different levels of evidence in item PP1: observing 3 - 4 gametes in more than one family, at this time PP1 evidence is applicable; observing 5 - 6 gametes in more than one family, at this time PP1_Moderate evidence is applicable; observing 7 gametes in more than two families, at this time PP1_Strong evidence is applicable; there is a segregation phenomenon between this variant and the disease in the pedigree. This optimization clarifies the applicable scope of pathogenic co-segregation evidence and improves the accuracy and convenience of classification.

[0023] Determining to use four algorithms of SIFT, Polyphen-2, PROVEAN, and MutationTaster for function prediction in item PP3, and using five algorithms of NNSplice, NetGene2, FSPLICE, MaxEntScan, and Human Splicing Finder for splicing prediction, and requiring that at least three algorithm conclusions be consistent before use; this optimization rule is clearer and easier to execute, and can prevent the situation where multiple prediction results in the original ACMG guidelines conflict and cannot be determined.

[0024] Determining the phenotypes of carriers in item PP4, requiring to meet at least 1 of the 5 characteristics described in the NCCN guidelines for hereditary breast / ovarian cancer, and requiring at least two carriers to be collected; this optimization is more precise than before, reduces the influence of a large number of subjective factors, and can improve the executability and effectiveness of variant classification.

[0025] The support database for determining reliable sources of reputation in item PP5 includes ClinVar, BRCA Exchange, and LOVD, and requires that there be no conflicts in the records among the databases, and only those with a reliability rating of two stars or above determined in ClinVar can be adopted. This optimization is more precise than before and can improve the execution efficiency and effectiveness of variant classification.

[0026] Further preferably, the evidence classification of item PP2 in the original ACMG guidelines is deleted. This item of evidence is not applicable to the classification of BRCA1 / 2 gene variants. This optimization can improve the execution efficiency of variant classification.

[0027] In a preferred embodiment of the present invention, the optimization of item BA1 in the BA module includes: reducing the threshold of the total population frequency to 1%. This optimization can greatly improve the effectiveness and accuracy of BRCA1 / 2 gene variant classification.

[0028] Further preferably, the optimization of item BA1 in the BA module also includes adding the gnomAD database including exome and genome as a reference database for population frequency; this database is currently the largest population variant database in the world, and adding this database can improve the effectiveness of variant classification.

[0029] In a preferred embodiment of the present invention, the optimization of item BS1 in the BS module includes modifying it to variants with an allele count AN exceeding 15,000 and an occurrence frequency greater than or equal to 0.1% or AN greater than 5,000 and an occurrence frequency greater than or equal to 0.5% in the records of the gnomAD database including exome and genome, the 1000 Genomes Project database, and the ExAC database. This optimization is easier to execute than before and can greatly improve the effectiveness of BRCA1 / 2 gene variant classification.

[0030] Further preferably, the evidence classification of item BS2 in the original ACMG guidelines is deleted. This item of evidence is only applicable to diseases with early complete penetrance and is therefore not applicable to the BRCA1 / 2 gene. This improvement can enhance the execution efficiency of variant classification.

[0031] In a preferred embodiment of the present invention, the optimization of items BP4 and BP6 to BP7 in the BP module includes at least one of the following:

[0032] For the functional prediction in BP4, four algorithms, namely SIFT, Polyphen-2, PROVEAN, and MutationTaster, are used. For the splicing prediction, four algorithms, namely NNSplice, NetGene2, FSPLICE, and MaxEntScan, are used. And it is required that the conclusions of at least three algorithms be consistent before they can be used. This optimization rule is clearer, easier to execute, and can prevent the situation where multiple prediction results in the original ACMG guidelines conflict and cannot be determined.

[0033] For the support database for determining reliable reputation sources in BP6, it includes ClinVar, BRCA Exchange, and LOVD. It is required that the records between the databases do not conflict, and the reliability determined in ClinVar is two stars or above before it can be adopted. This optimization is more precise than before and can improve the execution efficiency and effectiveness of variant classification.

[0034] In BP7, three conditions are determined: a) For synonymous mutations, they occur in non-conserved regions and there is no substantial evidence to prove that they will cause abnormal mRNA splicing; b) For missense mutations, they have the same amino acid change but different base changes as the known missense mutations clearly classified as class 1 variants, and there is no substantial evidence to prove that this variant will cause abnormal mRNA splicing; c) This evidence classification can only be used in combination with BP4 during the classification determination of the final variant, cannot be used in combination with other evidence items, nor can it be used independently, otherwise it is invalid.

[0035] Further preferably, the two evidence items BP1 / BP5 are deleted. These two evidence items do not conform to the genetic characteristics of the BRCA1 / 2 genes and related diseases, so they are deleted. This optimization can improve the execution efficiency of variant classification.

[0036] Further preferably, the optimization of the combined usage of evidence modules includes that in the final variant classification, evidence items belonging to the same type of data classification cannot be used simultaneously, with the exception of BP4 + BP7, and these two pieces of evidence are allowed to be used simultaneously. This optimization, combined with the optimization of item BP7, can greatly enhance the effectiveness of the classification method, and it is more significant for novel variants (especially synonymous mutations) for which valid clinical evidence or functional experiment evidence cannot be collected. Further, this optimization may be achieved using other descriptions, such as: 1) For synonymous mutations occurring in non-conserved regions, and multiple splicing prediction algorithms consistently predict no impact on mRNA splicing (splicing prediction affects the reference BP4 determination method), in the absence of more other valid evidence, it can be directly determined as 2 - Suspected Benign; 2) For missense mutations, having the same amino acid change but different base changes as known missense mutations clearly classified as class 1 variants, and multiple splicing prediction algorithms consistently predict no impact on mRNA splicing (splicing prediction affects the reference BP4 determination method), in the absence of more other valid evidence, the locus can be classified as 2 - Suspected Benign according to BP4 + BP7.

[0037] In a preferred embodiment of the present invention, the improvement in step (2) further includes the optimization of the PS module.

[0038] Further preferably, the optimization of the PS module includes deleting the PS2 evidence classification in the original ACMG guidelines. This piece of evidence is only suitable for diseases with high gene penetrance and low onset age, and thus is not suitable for the classification of BRCA1 / 2 gene variants; this optimization can improve the execution efficiency of variant classification.

[0039] The beneficial effects of the present invention are as follows: The present invention provides a more accurate, efficient, and easier-to-execute and automated BRCA1 / 2 gene variant classification and evaluation method to address the problems of strong subjectivity, weak executability, and low classification efficiency in existing variant classification and evaluation methods, and can more accurately evaluate the pathogenic risks of different BRCA1 / 2 gene variants to make suggestions on the frequency of clinical examinations. Brief Description of the Drawings

[0040] Figure 1 It is a schematic diagram of the principle of Embodiment 1 of the present invention. Detailed Embodiments

[0041] The technical solutions of the present invention will be further described and illustrated below through specific embodiments in combination with the drawings.

[0042] Embodiment 1

[0043] As Figure 1As shown, a classification and evaluation method for BRCA1 / 2 gene mutations is mainly based on the rules in the ACMG guidelines as a framework, and upgrades and optimizes the rules therein to make them more accurate and easier to execute, including the following steps:

[0044] (1) Annotate the coding region and amino acid changes of BRCA1 / 2 gene mutations according to the rules of HGVS for a fixed transcript to obtain evidence. The transcript of BRCA1 is NM_007294, and the transcript of BRCA2 is NM_000059;

[0045] (2) Classify the evidence obtained in step (1) according to the improved ACMG guidelines into population data, computer prediction data, functional research data, co-segregation data, allele data, database data, and phenotype data, and then classify them according to pathogenicity levels as PVS, PS, PM, and PP, and classify them according to benign conditions as BA, BS, and BP; The evidence classification and grading tables are shown in Tables 1 and 2:

[0046] Table 1 Classification Table of Pathogenic Evidence

[0047]

[0048]

[0049]

[0050] Table 2 Classification Table of Benign Evidence

[0051]

[0052] Note: *, Evidence classified into the same data category cannot be used simultaneously.

[0053] Table 3 Known or predicted classical splicing region mutations in the BRCA1 / BRCA2 genes can lead to in-frame RNA isoforms, which may have a repair effect on gene function. Such mutations cannot be used as PVS1 evidence.

[0054]

[0055] Table 4 Known or predicted exon-end mutations (G>non-G) in the BRCA1 / 2 genes can lead to in-frame RNA isoforms, which may have a repair effect on gene function. Such mutations cannot be used as PVS1 evidence.

[0056] gene exon BRCA1 2 / 3 / 6 / 9 / 10 / 11 / 16 / 18 / 19 / 20 / 22 BRCA2 1 / 3 / 8 / 9 / 11 / 12 / 22 / 26

[0057] (3) Classify and evaluate the obtained gene mutations as pathogenic, likely pathogenic, of uncertain significance, likely benign, and benign according to the scores of the classification and grading of each evidence. The specific interpretation rules for the classification and evaluation are shown in Table 5:

[0058] Table 5 Final Classification and Judgment Rules

[0059]

[0060] The following Examples 2 to 7 are specific examples of applying the classification and evaluation method of Example 1.

[0061] Example 2

[0062] Patient information: Female, 35 years old, breast cancer

[0063] Sample type: Paraffin wax roll + whole blood (paired samples)

[0064] Detected mutation: chr17:41215886C>T, BRCA1 gene, mutation abundance is 42.83%. Confirmed as a somatic mutation according to the comparison of whole blood samples;

[0065] Interpretation and annotation process:

[0066] 1) Annotate with the NM_007294 transcript of the BRCA1 gene, and the mutation result is c.5152+5G>A: p.?, which occurs in intron 18 (non-coding region) of the BRCA1 gene, 5 bp away from the exon edge, but not in the classical splicing region;

[0067] 2) Query the population database, and this mutation is not recorded in the ExAC, gnomAD, and 1000 Genomes databases (PM2);

[0068] 3) Query the public mutation database, and this mutation is reported as Pathogenic / Likely_pathogenic in the ClinVar database (ID = 55427, reliability is two stars) (PP5);

[0069] 4) Query relevant experimental reports. In a case study (PMID: 27886673), through in vitro analysis of the patient's RNA, it was proved that this mutation would cause the deletion of the entire exon 18, resulting in the damage of the important BRCT functional domain; in another in vitro experimental study (PMID: 30209399), the researcher generated the mutation by gene editing and then transfected it into the HAP1 cell line for functional determination, and it was proved that this mutation was Loss of Functional (PS3);

[0070] 5) The mutation was predicted using splicing prediction algorithms, and the results showed that three algorithms (FSPLICE, MaxEntScan, NNsplice) predicted it to affect splicing (PP3);

[0071] 6) Considering all the above evidence, the final classification result of this mutation is Class 4 - Suspected pathogenic, and the corresponding evidence items are PS3 + PM2 + PP5 + PP3.

[0072] Example 3

[0073] Patient information: Female, 65 years old, pelvic tumor

[0074] Sample type: Paraffin wax roll

[0075] Detected mutation: chr13:32954310delT, BRCA2 gene, mutation abundance is 8.04%, confirmed as a somatic mutation; Interpretation and annotation process:

[0076] 1) Annotated with the NM_000059 transcript of the BRCA2 gene, and the mutation result was found to be c.9256 + 28delT: p.?, occurring in intron 24 (non-coding region) of the BRCA2 gene, 28 bp away from the exon edge, and not in the classical splicing region;

[0077] 2) Querying the population database showed that the frequencies in the AFR populations of the ExAC, gnomADe, and gnomADg databases were 0.001281 (12 / 9370), 0.00104 (16 / 15386), and 0.001265 (11 / 8694) respectively, and there was no record in the Thousand Genomes database (BS1);

[0078] 3) Querying the public variant database showed that this variant was reported as Likelybenign in the ClinVar database (ID = 55427, reliability is one star); not classified in BRCA exchange;

[0079] 4) Prediction using prediction algorithms showed that 3 algorithms (FSPLICE, NetGene2, NNsplice) predicted it not to affect splicing (BP4);

[0080] 5) Considering all the above evidence, the final classification result of this mutation is Class 2 - Suspected benign, and the corresponding evidence items are BSI + BP4.

[0081] Example 4 (Comparative example)

[0082] Patient information: Female, breast cancer

[0083] Sample type: Whole blood

[0084] Detected variant: chr13: 32945237G>C, BRCA2 gene, confirmed as germline variant, heterozygous

[0085] Interpretation and annotation process of the method of the present invention:

[0086] 1) Annotate using the NM_000059 transcript of the BRCA2 gene, and the mutation is annotated as c.8632G>C, occurring at the last base of exon 20 of the BRCA2 gene, from G>non-G, giving evidence of PVS1;

[0087] 2) Querying the population database shows that this variant is not recorded in the 1000G, ExAC, and gnomAD databases, giving evidence of PM2;

[0088] 3) Querying the public variant database shows that this variant is reported as Uncertain_significance in the ClinVar database (ID = 918105, reliability is one star); not classified in BRCA exchange;

[0089] 4) Querying the literature reports, no other experimental evidence is found;

[0090] 5) Combining the above evidence, the final classification result according to Example 1 is category 4 - suspected pathogenicity, and the corresponding evidence is PVS1+PM2.

[0091] Annotation and interpretation process according to the original ACMG method:

[0092] 1) Annotate using the NM_000059 transcript of the BRCA2 gene, and the mutation is annotated as c.8632G>Cp.(E2878Q), belonging to a missense mutation, occurring at the last base of exon 20 of the BRCA2 gene, not in the classical splicing region, not giving evidence;

[0093] 2) Querying the population database shows that this variant is not recorded in the 1000G, ExAC, and gnomAD databases, giving evidence of PM2;

[0094] 3) Querying the public variant database shows that this variant is reported as Uncertain_significance in the ClinVar database (ID = 918105, reliability is one star); not classified in BRCA exchange, not giving evidence;

[0095] 4) Prediction using the prediction algorithm shows that 3 software (MutationTaster, PolyPhen-2, SIFT) predict it as a harmful mutation, and 1 software (PROVEAN) predicts it as a harmless mutation, giving evidence of PP3;

[0096] 5) Searching the literature reports, no other experimental evidence was found;

[0097] 6) Based on the above evidence, the classification result according to the original ACMG method is Class 3 - of uncertain significance, and the corresponding evidence is PM2+PP3.

[0098] Verification method:

[0099] The Minigene platform was used to verify the mRNA splicing effect of this mutation. The specific experimental method refers to the previous report (PMID: 20721748). The results showed that the mutation could lead to BRCA2 exon20 skipping (complete exon splicing deletion). After prediction, the protein translation result after the BRCA2 c.8632G>C mutation was p.W2830Kfs*13, which could lead to premature termination of protein translation, confirming that it was a pathogenic variant and consistent with the judgment result of this method.

[0100] Example 5

[0101] Patient information: Female, 45 years old, ovarian cancer

[0102] Sample type: Paraffin slide

[0103] Detected variant: chr17:41256926A>T, BRCA1 gene, confirmed as a somatic mutation, mutation abundance 47.3%

[0104] Annotation and interpretation process

[0105] 1) Annotating with the NM_007294 transcript of the BRCA1 gene, the mutation result was c.260T>Ap.L87*, which occurred in exon 6 of the BRCA1 gene and was a nonsense mutation. It might cause the 87th amino acid of the gene-encoded protein to change from leucine to a stop codon (and before the 1855th amino acid), giving evidence of PVS1;

[0106] 2) Searching the population database showed that this variant was not recorded in the 1000G, ExAC, and gnomAD databases, giving evidence of PM2;

[0107] 3) Searching the public variant database showed that it was recorded as Pathogenic in the BRCAExchange database and reported as Pathogenic in the ClinVar database (ID = 266284, reliability three stars), giving evidence of PP5;

[0108] 4) Search relevant literature reports. In a research experiment (PMID: 30209399), the authors verified the function of this variant through saturated genome editing. The result showed that this variant belongs to loss of function;

[0109] 5) Based on the above evidence, the final classification result according to Example 1 is Class 5 - pathogenic variant, and the corresponding evidence is PVS1 + PS3 + PM2 + PP5.

[0110] Example 6

[0111] Patient information: Female, 35 years old, unknown tumor type

[0112] Sample type: Paraffin-embedded tissue

[0113] Detected variant: chr17:41245915delC, BRCA1 gene, confirmed as a somatic mutation with a mutation abundance of 28.50%; Annotation and interpretation process

[0114] 1) Annotate using the NM_007294 transcript of the BRCA1 gene. The mutation result is c.1633delGp.V545*, which occurs in exon 11 of the BRCA1 gene. It belongs to a deletion mutation, resulting in the 545th amino acid of the gene-encoded protein changing from valine to a stop codon (and occurring before the 1855th amino acid), giving evidence of PVS1;

[0115] 2) Querying the population database shows that this variant has no record in the 1000G, ExAC, and gnomAD databases, giving evidence of PM2;

[0116] 3) Querying the public variant database shows that this variant has no record in the ClinVar and BRCA Exchange databases;

[0117] 4) Querying literature reports, no other experimental evidence was found;

[0118] 6) Based on the above evidence, the final classification result according to Example 1 is Class 4 - likely pathogenic, and the corresponding evidence is PVS1 + PM2.

[0119] Example 7

[0120] Patient information: Female, 70 years old, fallopian tube cancer

[0121] Sample type: Paraffin slide

[0122] Detected variant: chr17:41258506delT, BRCA1 gene, confirmed as a somatic mutation with a mutation abundance of 76.3%; Annotation and interpretation process

[0123] 1) Annotated with the NM_007294 transcript of the BRCA1 gene, the mutation result was found to be c.179delA p.Q60Rfs*9, occurring in exon 5 of the BRCA1 gene, which may cause the 60th amino acid of the gene-encoded protein to change from glutamine to arginine and premature termination at the 68th position (and before the 1855th amino acid), providing evidence of PVS1;

[0124] 2) Querying the population database showed that this variant was not recorded in the 1000G, ExAC, and gnomAD databases, providing evidence of PM2;

[0125] 3) Querying the public mutation database showed that this variant was recorded as Pathogenic in the BRCAExchange database and reported as Pathogenic in the ClinVar database (ID = 54353, reliability is three stars), providing evidence of PP5;

[0126] 4) Querying the literature reports, no other experimental evidence was found;

[0127] 5) Considering the above evidence, the final classification result according to Example 1 was Class 4 - Suspected Pathogenic, and the corresponding evidence was PVS1 + PM2.

[0128] As described above, it is only a preferred embodiment of the present invention, and thus the scope of implementation of the present invention cannot be limited thereby. That is, equivalent changes and modifications made according to the scope of the present invention patent and the content of the specification should still fall within the scope covered by the present invention.

Claims

1. A classification and evaluation method for BRCA1 / 2 gene mutations, characterized in that: It includes the following steps: (1) Annotate the coding region and amino acid changes of BRCA1 / 2 gene mutations according to the HGVS rules for fixed transcripts to obtain evidence. The transcript of BRCA1 is NM_007294, and the transcript of BRCA2 is NM_000059; (2) Classify the evidence obtained in step (1) according to the improved ACMG guidelines into population data, computer prediction data, functional research data, co-segregation data, allele data, database data, and phenotypic data, and then classify them according to pathogenicity levels as PVS, PS, PM, and PP, and classify them according to benign conditions as BA, BS, and BP; the above improvements include at least one of the optimization of item PVS1 in the PVS module, the optimization of the PM module, the optimization of items PP1, PP3 to PP5 in the PP module, the optimization of item BA1 in the BA module, the optimization of item BS1 in the BS module, the optimization of items BP4, BP6 to BP7 in the BP module, and the optimization of the combined usage of the evidence module; (3) Classify and evaluate the obtained gene mutations as pathogenic, likely pathogenic, of uncertain significance, likely benign, and benign based on the scores of the classification and grading of each evidence; The above optimization of item PVS1 in the PVS module includes at least one of the following: Include the mutation G>non-G at the 3'-end base of the BRCA1 / 2 gene exon in the PVS1 evidence item; For mutations in the classical splicing region and the last base at the 3'-end of the exon, exclude mutations that are known or predicted to result in in-frame RNA isoforms; The above optimization of the PM module includes: deleting the three evidence classifications of PM1 / PM3 / PM6 in the original ACMG guidelines; The above optimization of items PP1, PP3 to PP5 in the PP module includes at least one of the following: Determine the number of gametes required for different levels of evidence in item PP1: Observe 3-4 gametes in more than one family, and at this time, PP1 evidence is applicable; Observe 5-6 gametes in more than one family, and at this time, PP1_Moderate evidence is applicable; Observe 7 gametes in more than two families, and at this time, PP1_Strong evidence is applicable; Determine that in item PP3, SIFT, Polyphen-2, PROVEAN, and MutationTaster algorithms are used for function prediction, and NNSplice, NetGene2, FSPLICE, MaxEntScan, and Human Splicing Finder algorithms are used for splicing prediction, and it is required that at least three algorithm conclusions are consistent before use; Determine the phenotypes of carriers in item PP4, requiring at least 1 of the 5 characteristics described in the NCCN guidelines for hereditary breast / ovarian cancer, and requiring at least two carriers to be collected; The support databases for determining reliable reputation sources in PP5 include ClinVar, BRCA Exchange, and LOVD. It is required that there are no conflicts in the records between the databases, and only when the reliability determined in ClinVar is two stars or above can it be adopted; The above optimization of item BA1 in the BA module includes: reducing the threshold of the total population frequency to 1%; The above optimization of item BS1 in the BS module includes modifying it to variants with the number of alleles AN exceeding 15,000 and the occurrence frequency greater than or equal to 0.1%, or AN greater than 5,000 and the occurrence frequency greater than or equal to 0.5% in the records of the gnomAD database (including exome and genome), the 1000 Genomes Project database, and the ExAC database; The above optimization of items BP4, BP6 to BP7 in the BP module includes at least one of the following: In item BP4, SIFT, Polyphen-2, PROVEAN, and MutationTaster are used to determine functional prediction, and NNSplice, NetGene2, FSPLICE, and MaxEntScan are used for splicing prediction. It is required that the conclusions of at least three algorithms are consistent before it can be used; The support databases for determining reliable reputation sources in item BP6 include ClinVar, BRCA Exchange, and LOVD. It is required that there are no conflicts in the records between the databases, and only when the reliability determined in ClinVar is two stars or above can it be adopted; In item BP7, three conditions are determined: a) For synonymous mutations, they occur in non-conserved regions and there is no substantial evidence to prove that they will cause abnormal mRNA splicing; b) For missense mutations, they have the same amino acid change as the known missense mutations clearly classified as class 1 variants but different base changes, and there is no substantial evidence to prove that this variant will cause abnormal mRNA splicing; c) This item of evidence can only be used in combination with BP4 during the classification determination of the final variant, and cannot be used in combination with other evidence items or independently, otherwise it is invalid.

2. The classification and evaluation method according to claim 1, wherein: The above optimization of item PVS1 in the PVS module also includes: limiting the loss-of-function variants occurring before amino acid position 1855 of the BRCA1 gene or position 3309 of the BRCA2 gene to be included in item PVS1.

3. The classification and evaluation method according to claim 1, characterized in that: The above optimization of the PM module also includes: adding the gnomAD database (including exome and genome) as the population frequency reference database in item PM2.

4. The classification and evaluation method according to claim 1, wherein: The evidence classification of PP2 in the original ACMG guidelines was deleted.

5. The classification and evaluation method according to claim 1, characterized in that: The above optimization of item BA1 in the BA module also includes adding the gnomAD database (including exome and genome) as the population frequency reference database.

6. The classification and evaluation method according to claim 1, wherein: The evidence classification of BS2 in the original ACMG guidelines was deleted.

7. The classification and evaluation method according to claim 1, characterized in that: Evidence items BP1 and BP5 were deleted.

8. The classification and evaluation method according to claim 1, wherein: The above optimization of the combined usage of evidence modules includes that in the final variant classification, evidence items belonging to the same type of data classification cannot be used simultaneously, and BP4 + BP7 is an exception, and these two items of evidence are allowed to be used simultaneously.

9. The classification and evaluation method according to claim 1, wherein: The improvement in step (2) also includes the optimization of the PS module, and the optimization of the PS module includes deleting the PS2 item evidence classification in the original ACMG guidelines.

Images