to
Close
close

Member Login


User ID  
Password

Reporting

1. Introduction

The goal of a genomics report is to convey accurate, interpretable and succinct information that is relevant to patient care. In the Massively parallel sequencing era, this simple statement is becoming increasingly difficult to put into practice. This chapter aims to provide guidelines and establish principles that should assist in the preparation of a genomics report.

Approaches to genomic analysis vary in terms of the technology and methodology used as well as the breadth of genetic variation that is interrogated; the analysis may yield information about a single class of genetic variation or may extend to encompass all sequence and structural variants. This presents a number of challenges for the clinical laboratory when preparing a report based on Genomic data. The issue of clinical validity and utility is an important concept to keep in mind when formulating the report, although addressing this is beyond the scope of this document, and is already well covered by existing legislation in NPAAC standard publications (validation of in-house IVDs and nucleic acid testing).

A key issue in reporting genomic tests is that variants of known or possible pathogenicity may be identified which may be unrelated to the primary clinical indication for the test. Such incidental findings are inevitable in high-resolution genomic studies utilising Massively Parallel Sequencing techniques which interrogate a greater proportion of the human genomic sequence, presenting difficulties both for laboratories in reporting such variants and for clinicians receiving unsolicited information. The potential for false positive results is also amplified by the increasing number of genes interrogated, and the low prevalence of some of the disorders which may be the subject of investigation by genomic sequencing.

A second key issue is that a large number of variants of uncertain clinical significance can be identified and the reporting of this information requires careful management to minimise potential harm while providing the maximum available, relevant information, for clinical management.

It is essential that laboratories producing reports of genomic tests have clearly defined, evidence based protocols for classifying the clinical significance of detected genetic variants and addressing incidental findings. This protocol should define, prior to the analytical result being available, which outcomes will be reported and which will not and that these protocols are available to requestors of Genomic tests.

It is also essential that these results are reported clearly, consistently and unambiguously, using established nomenclature guidelines such as those available from the Human Genome Variation Society (http://www.hgvs.org/mutnomen/) and relevant standardised reporting formats such as the RCPA Guidelines for reporting molecular genetic tests to medical practitioners 2009. It must be recognised that laboratory reports may be read by both experts and non-experts, and may be stored for years in a patient’s medical record.

This chapter provides a guide for the reporting of NGS results in the clinical context. It has been developed in the interests of ensuring the analytical and clinical validity of genomic reports, the consistency and clarity of reporting, thereby assisting in the production of a report that is accurate, interpretable, succinct and relevant to patient care.

2. Published Guidelines

A number of national and international professional bodies have issued policy statements regarding the clinical application of genomic sequencing that include guidelines for the reporting of genomic testing. It is advised that these are consulted to provide a broader overview of the issues relating to the reporting of genomic data.

Published Guidelines for the Clinical application of Genomics that include recommendations on the reporting of massively parallel sequencing data:

3. The Clinical Context of Testing

3.1 The laboratory should have explicit requirements regarding the level and quality of clinical information that must be provided when a genomic test is requested.

The Massively Parallel Sequencing Genomics report must include the context in which testing has been requested. This assists with correct interpretation of the report, and re-statement of the clinical context for testing is currently an NPAAC requirement. Detailed and informative clinical information is also critical for the laboratory to produce a report that presents its conclusions in the most appropriate clinical context.

The clinical interpretation of genomic analyses are improved when testing laboratories are provided with discriminating clinical details.

Bioinformatics pipelines are capable of providing interpretative comment on identified variants. Reporting laboratories should ensure that they understand the limitations of informatics interpretations and provide adequate review of automatically generated interpretations in the clinical context of testing.

3.2 The laboratory should maintain active dialog with requesting clinicians regarding requests, interpretation, and reporting of genomic tests.

Strengthening laboratory liaison with requesting clinicians is essential in any rapidly developing field of medical testing. The interpretation and reporting of genomic test results would benefit from a team approach, whereby clinical laboratory scientists, pathologists, clinical geneticists, and other medical professionals are involved in genomic data interpretation.

This liaison should extend from requesting of the test and defining the most appropriate genomic testing through to interpretation of the report.

Astute clinical assessment and laboratory-clinician consultation can focus genomic analysis on specific genomic regions, potentially enabling diagnosis of monogenic disorders, or clarify differential diagnoses.

4. Variant Reporting

4.1 The laboratory should consistently classify genomic variants according to their clinical significance.

Variants that are classified as benign or even likely benign could potentially not be included in a report, with reports confined to variants classified as pathogenic or likely pathogenic (EuroGentest 2014 final draft; ACMG Standards and Guidelines 2015).  Local policy can dictate which variant categories are to be reported; however, a record of all variants identified should be maintained by the testing laboratory and should be readily accessible for review and may be disclosed upon request to a clinician. The report should clearly state the laboratory's reporting policy indicating which classes of variants have been reported, highlighting the possible existence of variants which may not appear on the report.

The classification of a variant as benign or pathogenic must be based on a secure evidence-base such that significant reclassification in the future is unlikely without additional and convincing functional data. However, the observation of a variant of unknown significance may require a fresh evaluation of that variant to reflect new information that may be available.

To ensure consistency, it is highly recommended that a reporting laboratory maintain an in house database of variants that is professionally curated to a standard acceptable for clinical use (RCPA Clinical database standards document reference) and submits variant data to a clinical standard external database.

4.2 The laboratory should have a clear policy regarding the reporting of variants and this policy must be readily available to referrers.

Irrespective of how genomic variants are classified, there is likely to be a substantial number of variants of ‘unknown significance’ for which there is no relevant evidence to assist interpretation, i.e. there is no evidence base on which to determine clinical significance for the condition under investigation.

Reporting laboratories should be aware that there is a possibility of potential over-interpretation of results of uncertain significance based on a limited understanding of contextual information. As such, reporting laboratories must minimise the potential for readers of genomic test reports to misconstrue the clinical significance of certain clinical categories of genomic findings and the ensuing anxiety/harm that this may bring to patients (and their families). The report should be transparent in how it has reached its conclusions, and include information relating to how the findings may influence subsequent clinical judgment, including suggestions for further testing, if necessary.

4.3 Assessing the Available Evidence

The laboratory’s interpretation of clinical significance should be based on sound evidence. Peer-reviewed literature and clinical or near clinical quality databases could be regarded as high quality primary evidence in the assessment of clinical significance for a particular genomic variant. The RCPA Standards for Clinical Databases of Genetic Variants document is a useful guide for measuring the quality of a clinical database. Functional studies on genetic variants are occasionally performed by clinical laboratories to evaluate clinical significance. Examples include RNA studies to determine the effect of a variant on RNA splicing events, and protein functional studies to determine protein activity. Caution must be exercised with in-house studies which have not been subject to peer review processes to ensure that appropriate controls are included and that the results are analytically valid. Furthermore, in assessing protein function, care must be taken in determining which functions of a multi-functional protein are relevant to the disease state and therefore should be assessed.

When evaluating literature the quality of the publication should be taken into account. Critical review of literature cited in reports is a requisite competency skill for any laboratory geneticist. This should include consideration of statistical significance requirements in case/control and comprehensive familial series.

When utilising genomic databases consideration should be given to the purpose of each database and the processes used for the classification of variants submitted to the database.

It should be noted that merely appearing in a database with a classification of pathogenic does not constitute final proof that a particular variant should be reported as pathogenic in the clinical context of the report. It is more likely that a database entry will provide the starting point for the collection of evidence of proof of pathogenicity and that multiple databases and sources will be required before arriving at this conclusion.

In general, classifications should be based on multiple independent lines of evidence such as in vitro or in vivo functional data, segregation with disease, algorithmic prediction of protein function or RNA transcription events and disease/normal population variant frequencies.

Care should be taken to avoid over-interpreting early reports showing enrichment of a variant in “affected” populations, particularly those that have not been verified in replication studies. This should also apply to variants found to confer low relative risk and predictive power for common (multifactorial) diseases and traits.

4.4 Variant Interpretation

4.4.1 The laboratory professionals who provide clinical interpretations of genomic variants should understand, have access to, and utilise up-to-date resources to aid them in their task.

4.4.2 The laboratory should develop a protocol for the assessment of variant pathogenicity. See above section 4.1

Novel and rare genetic variants pose the greatest interpretational challenge owing largely to the continued lack of high-quality, large-scale control data. Additionally, there is often a lack of information on rare variants for evidence based assessment.

The task of assessing seemingly novel or rare variants is challenged by the accelerating pace of discovery of rare variants associated with phenotypic abnormality, as well as a growing number of clinically significant variants recognised to have incomplete penetrance and variable expression.

The potential influence of genetic background, in particular modifier sequence variants is a confounding factor and should be considered where there is any evidence to suggest there may be a modifier effect.

Report writers should be aware of the possibility that the initial diagnosis arrived at in the laboratory might be incomplete and that additional clinically significant genetic variants (that have gone undetected or remain unclassified due to lack of an evidence base at the time of reporting) may underlie the patient’s non-specific or variable phenotype.

The laboratory geneticist should not over-interpret the genomic analysis result, especially where a variant of ‘uncertain’ significance is concerned, in recognition that these findings do not necessarily indicate a diagnosis for the patient. Reported assertions regarding variant pathogenicity should ideally include any mitigating clinical information such as the inheritance pattern, clinical context, and phenotype.

Analysts should also provide insight into the extent of what remains unknown and the challenge this brings to the task of clinically interpreting genomic findings.

The report should include or cite the evidence which justifies the conclusion regarding the clinical significance of identified genomic variants.

4.5 Systematic Review of Variant Interpretations

4.5.1 The reporting laboratory must have a written protocol for the review of variants and this protocol must be available to referrers.

Interpretation of the clinical significance status of a genomic variant may change in the light of new information.

A key question requiring consideration is how (validated/issued) reports that require modification in light of new information, with their attendant clinical risks, should be dealt with. In practical terms, responsibility for a specific patient lies primarily with the physician with an ongoing patient relationship.

Recommendations that call for continued review of variants (including ‘benign’, ‘likely benign’, ‘unknown’ and ‘uncertain’ significance findings), which collectively may number in the tens of thousands per sample in the case of whole exome sequencing and whole genome sequencing, have significant resource implications for laboratories, particularly for professional time. Scheduled review based solely on defined time intervals also raises the possibility of re-issuing clinically inappropriate reports performed in isolation from the referring clinic.

It is advised that laboratories have a formal process for evaluating new evidence, re-interpreting previous, individual patient results, re-contacting referrers, and contributing to patient reviews, where required. However, a bi-directional flow from the clinic through continued review of patient files can also contribute to the timely review of variants.

5. Incidental Findings

Genomic analysis will inevitably detect clinically significant variants, which are unrelated to the clinical features that prompted testing. The issues associated with detection of these remain under discussion, but their solutions will no doubt involve an emphasis on counselling and education before testing is performed, informed consent with a clear explanation of the current limits of testing and interpretation, maintenance of privacy and confidentiality, and sensitivity to culture within families, their heritage, and their communities.

Commonly encountered examples of unsolicited findings detected during genomic testing include:

  • Detection of consanguinity and incest, where this was not known to ordering clinicians and families;
  • Detection of carrier status for autosomal recessive disorders unrelated to the clinical indication for genomic testing;
  • Detection of variants involving highly penetrant genes associated with dominant, adult-onset conditions.

5.1 Laboratories performing genomic testing should have clear policies in place for disclosure of incidental findings.

It is advisable that clinicians and patients be informed of these policies and the types of incidental findings that will be reported.

Clinicians may give patients the option of not receiving certain results. While these policies should be in place, exceptional circumstances may arise which need to be handled judiciously on a case-by-case basis through laboratory-clinician consultation.

In keeping with the principles of good laboratory practice, uncertainty associated with reporting incidental findings is usually best managed with input from a medical genetic specialist and/or the referrer.

Reporting of incidental findings should be limited to variants that are unequivocally classified as pathogenic or likely pathogenic and are deemed reportable according to the laboratory’s policy.

6. The Genomic Test Report

Clinical Genomic reports should follow the general principles for reporting of genetic tests as described in the RCPA guidelines (Guidelines for reporting molecular genetic tests to medical practitioners, 2009) and be consistent with ISO 15189 as a minimum requirement. However, the genomics report is likely to need to convey a much larger amount of information both in terms of test data and in terms of descriptive information about the test and particularly its limitations.

The challenge is to produce a report that contains all the relevant information for accurate interpretation of the report while avoiding information overload and possible distraction from the actual result required for patient care. One approach to solving this problem is to provide a report with a prominent one page patient summary containing all class 4 and 5 variants relevant to the request, that is supplemented with further information relating to the actual test (limitations, bioinformatics pipeline description and metrics) and variants other than class 4 and 5 if appropriate to the request. Such multi page reports would need to comply with all relevant standards and guidelines for reporting including page number, report date and the inclusion of patient demographics on each page to ensure unambiguous linking of the entire report.

A consistent approach to reporting genomic findings is important, particularly for families dispersed across state or national boundaries. Crucially, those responsible for reporting should appreciate that interpretive difference may influence medical management and patient choices. Even if a report is directed to the expert requesting clinician, it is also important to note that reports may be included in medical records and hence be read by non-experts involved in the patient’s care. Hence, every effort should be made to ensure that the report is succinct, clear and interpretable by as wide a range of relevant clinicians as possible.

6.1 Key requirements of a Genomics Report

The minimum suggested content for a report is described below. The following list is not a recommendation for the structure of the report which could be ordered to provide a one page summary of important results relevant to patient care followed by additional pages detailing the test and any further results that may be appropriate.

Report Details

  • Reporting laboratory details
  • Title of report
  • Report status and Report authorisation
  • Issue date and time of report

Patient identification

  • Name
  • Date of Birth
  • Unique laboratory identifier
  • Gender
  • Ethnicity if relevant to testing

Patient diagnosis context

  • Clinical details on request
  • Specimen
  • Type (blood, tissue and site, fluid)
  • Secondary specimen identifier (Block number, referring laboratory identifier)

Test description

  • Test category
  • Purpose of test (e.g to assist in the diagnosis of … or the exclusion of…)
  • Genes tested list
  • Methodology used including confirmation of variants by an orthogonal method if performed
  • Limitations to test including any remaining uncertainty where it exists

Result summary

  • Inheritance model used for sequencing data analysis if relevant
  • Gene name using HGNC approved gene symbol
  • Zygosity
  • cDNA nomenclature utilising standardised nomenclature (HGVS recommended)
  • Protein nomenclature utilising HGVS recommended nomenclature
  • Genomic coordinates utilising HGVS recommended nomenclature based on an LRG where available and a RefSeqGene record if not available.
  • Reference sequences including genome build or reference sequence version
  • Variant reporting policy for the reporting laboratory that complies with relevant guidelines

Interpretive comment

  • Variant classification as class 1 (benign) through to class 5 (pathogenic)
  • Narrative comment indicating the relevance of the identified variants to the reason for the test request
  • If applicable the need for follow up or confirmatory testing should be indicated on the report

6.2 Laboratories should include recommendations for appropriate follow-up in reports.

In situations in which further genetic studies may be warranted (e.g. parental testing, segregation analysis, testing of other tissues), these recommendations should be included in the test report.

6.2.1 Examples/resources

  • Caris NSCLC example report
  • Foundation One Sample Report for Cancer Related Genes

7. Internal Laboratory Databases

7.1 Laboratories should establish an internal database of genomic findings.

This could serve the purpose of identifying common genomic variants specific to a patient population and/or recurrent false-positive calls associated with a particular genomic platform.

The curation of an internal laboratory- and platform-specific list of common benign variants can assist with the interpretation process. This could aid with the process of systematic review of variant interpretations.

Any handling of data derived from medical testing should comply with the relevant regulatory and legislative requirements.

8. Sharing Genomic Data

8.1 Laboratories are strongly encouraged to submit genotypic data from genomic testing to appropriate clinical databases to facilitate consistency of interpretation across laboratories.

Genomic testing would benefit from the availability of clinically vetted, regularly updated databases of annotated variants that ideally would include population frequencies and referenced clinical relevance for each variant. Thus, there is a need for consolidation of the various genotype-phenotype databases available into a commonly available and perhaps centralized clinical- grade resource that is publically accessible.

The availability of phenotypic information is essential for the investigation of genotype-phenotype relationships both at the individual patient level and at the global level. Integrating the phenotype information into these clinical- grade resource databases will greatly assist with interpretation of variants identified from genomic analyses. Laboratories should be encouraged to contribute phenotypic information whilst being aware of the issues posed by privacy concerns, data complexity and lack of uniform methods for collection of phenotypic data.

9. Reporting of Somatic Variants

The reporting of somatic variants in oncology should follow the recommendations described previously for germ line variants, with the following additional considerations. If the variants are known to predict a likely response to specific therapeutic agents then the relationship between the presence or absence of the variant and the agent must be included on the report. Due to the heterogeneous nature of tumours the percentage of the variant allele that is considered significant to recommend or not recommend use of the agent should be included on the report along with the uncertainty of measurement for the variant. The percentage of tumour cells in a solid tumour sample or blasts in haematological malignancy must also be included in the report as contextual information.

10. Summary Comment

The utility of a genomics report can be increased by the preparation of a standardised report that adheres to established guidelines. However, of equal importance is the need to ensure that clinicians, genetic counsellors, or others who read these reports have the necessary training and support to optimise interpretation of genomic test results for patient care. In part, this can be achieved by greater collaboration between requestors of genomic tests and the laboratories performing genomic tests however, formal education of requestors in the interpretation of genomics reports is likely to further increase the utility of genomics reports.

Specific training for genomic testing needs to be incorporated into existing professional development programmes for laboratory geneticists, clinical geneticists and other clinical specialities (e.g cardiology, obstetrics, neurology) and primary care medical practitioners that may be likely to request genomic testing. With the advent of direct to consumer testing both in Australia and abroad, genomic education of the general public should also be addressed, and may be feasible by the technological advances made in mobile applications.

It is suggested that individual laboratories reporting genomic tests play an active role in developing and implementing a continuing professional development program focussed on genomic testing by establishing structured in-house training programs for their staff and potential referrers who are seeking to extend their professional competency into this arena. Such a programme should focus on the tests offered by the individual reporting laboratory.

10.1 Resources

Contact form for feedback and comments:

 

*Required

*Required

*Required

 Security code