US clarifies position on secondary findings in clinical genome sequencing

By Alison Hall

25 November 2014


Last week came the welcome and long-awaited news that the AGMG have finally updated their policy statement on analysing and reporting secondary findings from clinical genome-scale sequencing. Following the policy announcement on 1 April 2014, that the recommendations would be updated to provide for patients undergoing clinical sequencing to opt-out of receiving secondary findings en bloc, the updated ACMG Policy Statement clarifies:

  • The elements that should be discussed as part of the consent to clinical genome-scale sequencing, including that written consent should be obtained, and that laboratories will routinely analyse ‘the sequence of a set of genes deemed to be highly medically actionable so as to detect pathogenic variants that may predispose to a severe but preventable outcome’
  • That patients may opt out of such analysis provided that they understand the implications of so doing
  • That sequencing of the same set of genes should be routinely performed in children (but that parents should have a right of opt-out)
  • That the right of opt-out, is an all or nothing choice: it is not feasible to offer analysis and reporting of a subset of medically actionable genes.

Undoubtedly the publication of the ACMG’s initial policy statement on incidental findings in clinical exome and genome sequencing generated vigorous debate, and the updated statement acknowledges that in the intervening year, the ACMG Board were informed by two developments: the report from the Presidential Commission on Bioethical Issues regarding secondary findings, and a survey of ACMG members in January 2014. That the Presidential Commission’s report has been influential is self-evident, given that the recent ACMG Policy Statement now refers to ‘secondary’ rather than ‘incidental’ findings, and distinguishes between the two types of findings.

Views of US genetics professionals

The results of the ACMG survey have been published in the same issue of Genetics in Medicine and make interesting reading. Scheuner et al.’s paper provides a description of the survey and reports its findings. 29% of the 1687 ACMG members responded, and exploratory factor analysis was used to identify the underlying factors in responses. Four factors – best practices, patient preferences, guidance and informed consent explained 51% of the survey variance. The headline conclusions of the survey were that -

  • The ‘best practice’ of seeking and reporting secondary findings received general support, are regarded as consistent with medical standards being evidence based, and the benefits outweigh the harms, at least for adults
  • The benefits of analysing and reporting these secondary findings in children is more contested, as is the potential impact on health care resources
  • Most respondents agreed that patient preferences should be considered, including opting out of the reporting of secondary findings, and that addressing this in the consent process ‘was feasible and critical’.

The take-home messages from the survey highlight the importance of the informed consent process in ensuring that patients who are offered clinical genome-scale sequencing fully understand what is involved, and advise that more empirical research on the reporting of secondary findings is urgently needed.

The UK perspective

The PHG Foundation has been involved in an 18 month project (Realising Genomics in Clinical Practice) exploring the ethical, legal and social challenges likely to arise from implementing NGS diagnostic panels, WES and WGS into clinical settings in the NHS. We noted greater hesitancy about adopting WES and WGS as a first-line test in clinical practice in the UK than in the US and some reticence about generating findings that are difficult to interpret or are not related to the clinical reason for referral in an NHS setting, in part because of fears that existing resources, capacity and training are likely to be inadequate to deal with downstream patient management and treatment. The impact on health care resources was also an issue for respondents to the ACGM survey although it generated mixed opinions. In our own project, we concluded that in the NHS setting, a partial solution may be to utilise these technologies in a more targeted way to minimise the volume of incidental findings and variants of uncertain significance that are generated and reported.

Addressing the knowledge gap for clinical genome sequencing

The difference between these two approaches is likely to be partly a function of the health care settings and arrangements for funding and reimbursement: but it also points to residual uncertainty about the benefit of generating and reporting secondary findings in those undergoing sequencing, given that the secondary findings may be generated in individuals who have no family history of the disease in question. The ACGM survey emphasises the urgent need for empirical research into the successful implementation of seeking and reporting secondary findings, and this should include more evidence about the scientific and clinical validity and utility of the secondary findings in the ACGM list, and more extensive social sciences research that charts patient experiences. This should also include research on that subset of patients who are subsequently found to have false positive or false negative findings. This knowledge gap is something that the Genomics England Clinical Interpretation Partnerships should address as a matter of urgency, as the main phase of the 100,000 Genomes Project gets underway.

Putting patients centre-stage

Secondly, the change in the ACGM’s policy highlights the need for consent processes to become increasingly sensitive to patient preferences and to ensure that patients understand the potential for generating secondary or incidental findings and variants of unknown significance. This also involves emphasising how patients may be involved in disclosure decisions, including the potential for reanalysis and recontact, perhaps many years after an initial clinical encounter. Both these developments highlight how genomic sequencing technologies may catalyse more personalised approaches to healthcare, in which patients are encouraged to engage in a dialogue with health care professionals.

Genomics and policy newsletter

Sign up