Case 2 - Unusual BRCA2 mutation - results

Index to results and discussion page

• What is the mutation?
• What is the nomenclature to describe the mutation?
• Is the mutation likely to be pathogenic?
• How would you report this case?

What is the mutation?

Examination of the sequence trace reveals that the mutation is not a straightforward one, but may be described most simply as a 9bp deletion associated with an upstream 2bp deletion/insertion:

          delins           9bp deletion

normal: AAGCAGTGAAGAATGCAGCAGACCCAGCTTA

mutant: AAGCAGGTAAGAATGCAGCTTA

A fair amount of debate took place in the Leeds lab as to whether the mutation was two separate events (as illustrated above) or whether it could have occurred as a larger deletion (TGAAGAATGCAGCAGACCCAGC) accompanied by an insertion (GTAAGAATGCAGC). Debate as to the mechanism is academic, but there are two important considerations. Firstly, if they are separate events, it is not possible from the evidence on this single individual to rule out the possibility that they are on opposite chromosomes. This would have important clinical consequences, but may be considered to be unlikely since it would probably require a coincidence, and each mutation alone may be pathogenic (see below). Secondly, whether they are separate events or not may influence how one describes the change using HGVS nomenclature (see below).

Back to top of page

What is the nomenclature to describe the mutation?

See Human Genome Variation Society (HGVS) web pages for guidance and examples on human mutation nomenclature.

In view of the debate over the mutation nomenclature in this case, the Leeds lab sought advice from Johan den Dunnen from HGVS. His reply was as follows:

'The answer is rather simple; description should be clear and unequivocal and not be influenced by what might have happened. I expect that a description of an insertion/deletion is rather complex so probably something like [c.8951_8952delinsGT; c.8964_8972del].'

In other words, clarity is most important. Nomenclature should describe the observed sequence and not necessarily describe how it arose.

Applying the principle of clarity will assist with similar situations when one has to decide whether to call a mutation as one or two events. When events are widely separated (e.g. several kb), nobody would have any difficulty seeing them as separate and use nomenclature accordingly. On the other hand, when they are only a single bp apart, most would probably see them as a single event and it would not cross anyone's mind to describe as two events: as in the following example.

                    delins 9bp deletion

normal: AAGCAGTGAAGAATGCAGCAGACCCAGCTTA

mutant: AAGCAGTGAAGAATGCGACTTA

This could be described as: c.8961_8962delinsGA; c.8964_8972del

But by applying the principle of simplicity and clarity, is probably better described as: c.8961_8972delinsGAC

In the intermediate zone, examination on a case by case basis will have to determine the clearest nomenclature!

Back to top of page

Is the mutation likely to be pathogenic?

The following is a list of checks one can consider in ascertaining the pathogenicity of a sequence variant. The results of checks in this case are indicated:

• Has the variant been reported before - is it on a mutation database, or has it been published, or is it a known SNP?

  • Not reported before (searching on pubmed; high-profile mutation reports; BIC mutation database.

• If it is a missence mutation, does it lead to a conservative or non-conservative amino acid change?

  • Not applicable.

• If it is a missence mutation, is the amino acid conserved in evolution (in orthologs)?

  • Not applicable.

• Does the variant create or destroy a splice site?

  • Yes, splice-site prediction software (http://www.fruitfly.org/seq_tools/) predicts that the mutant sequence activates a cryptic splice site (GAAGCAGTgAAGAAT to GAAGCAGGTAAGAAT; score 1.0) with a higher score than the native donor splice site (CCTTGAGGTGAGAGA; scone 0.98).

• Are RNA studies feasible in this case to back-up predictions?

  • Not performed in this case. But would determine the preference for donor splice site utilisation (with the caveat of possible tissue-to-tissue alternative splicing).

• What is the likely effect on the protein? If the variant results is a frame-shift, can you confirm it is likely to result in premature protein truncation?

  • If the activated cryptic splice site is utilised, a frame-shift is anticipated. Generally, a frame-shift results in premature termination. This is confirmed in this case. I have worked out that the protein nomenclature for this is p.2908Val_AlafsX5 (a stop codon is at the 5th position counting from the first changed amino acid caused by the frame-shift). But I may be wrong - if you disagree, please fill in the feedback form!

     

            exon 21   exon 22

         tat gaa gca ggg tta ttt cag tga

    2905  Y   E   A   G   L   F   Q   X

                      changed amino acids

     

  • If the predicted new splice site is not utilised, one still predicts two protein changes in this case: p.Val2908Gly and p.Asp2913_Ala2915del.

• Does the variant segregate with the disease / phenotype in the pedigree?

  • Not studied in this case, but has been suggested to clinicians if family samples can be obtained.

• Have functional studies been performed and published for this mutation? Or can they be performed for this case?

  • Not performed in this case.

  • Functional studies are generally beyond the scope of what a diagnostic lab can perform.

Back to top of page

How would you report this case?

A copy of the actual report for this case, reported in 2005, may be downloaded here (pdf)

The following points are included on the report:

• The clinical reason for testing is restated, including a reference to a previous report.

• The result of the test (exon 21 sequencing) is presented in tabular form using HGVS nomenclature.

• The result of the test is described in text.

• The possibility that the result could be caused by 'mutation' or variants on separate chromosomes is mentioned.

• Comments concerning the pathogenicity are listed:

  • The mutation is not on the BIC database and has not been published;

  • The mutation is predicted to create an alternative splice donor site;

  • The consequences of utilisation of the alternative splice donor site on the protein is not mentioned on this report, although some may argue that this would be helpful;

  • The result on the protein if the alternative splice site is not utilised is mentioned;

  • The possibility of testing further family members to determine if the mutation segregates with the phenotype is mentioned.

• Predictive testing is offered, subject to confidence in the pathogenicity.

• A summary statement is included.

Proceed to feedback on this case

Back to top of page