NHMRC Public Consultations

Skip Navigation and go to Content
Visit NHMRC website

Section 3 (Chapters 3.1 & 3.5), Glossary and Revisions to Section 5 National Statement on Ethical Conduct in Human Research, 2007 submission

This submission reflects the views of
Organisation Name: 
Garvan Institute of Medical Research
Personal Details
Specific Comments
2. Chapter 3.1

Chapter 3.1 The elements of research

Overall comment - Encapsulates a broad range of projects and circumstances very well.

There is a problem with one element. Internally, we are divided on the issue of return of results.

One view holds that we should respect an expressed desire NOT to know the results of a genomic test. This is based on the principle of autonomy.

The contrary view is that such a decision cannot be informed for all future possibilities, owing to a) the complexity of the data and programmatic realities of consent; b) changing utility with time. The second view holds that we should find a form of wording that leaves open the option to return results which are in the opinion of those with a duty of care to participants now of such clinical significance that there is a duty of care to inform the participant. In the event of a medical issue arising in a participant or relative that was foreshadowed by the result of the study, a legal complaint would be difficult to refute on the basis that declining to receive informaton was invalid on the basis of informed consent.

Practically, the number of subjects who would decline on this basis—with an emphasis on the duty of care to the participant and a rigorous process—is small. And these individuals probably lack the tolerance of inherent ambiguity to participate in open-ended research.

2. Chapter 3.1

3.1.60  Consent from participants has not been mentioned in this section.  Presumably consent from participants would be sought before data sharing was considered.

2. Chapter 3.1

3.1.63 ‘researchers’ rather than ‘researcher’

3. Chapter 3.5

Chapter 3.5 Genomic research

Overall comments

  • There is no distinction between somatic and germline data. 
  • The need for validation of research findings is justified.  However the difference between confirmation (which should be performed to the best of the researcher’s ability prior to informing the participant)  and validation (which can only be performed with a new sample in a clinical setting and therefore necessarily involves the participant) needs to be stated clearly.
  • The roles of biobanks/cohorts and researchers in returning results needs to include consultation between the parties.
  • Significant change in not needing to confirm participant’s choice not to be informed.
3. Chapter 3.5

3.5.12 This explanation will need to be added to all PICFs

3. Chapter 3.5

3.5.17 This is a significant change from the current requirement in 3.5.2(c) which states that ‘where participants or relatives prefer not to receive genetic information that is important for their health, they should be advised that they will be approached to confirm this decision when the results of the research are available.’  This new section now puts the onus on participants to let researchers know if they change their mind about being informed rather than the researcher recontacting the participant and confirming the decision.

The validity of this approach is open to question. One view is that the patient cannot be informed about a refusal to have results returned at the outset (I can flesh out this argument if you like). Therefore, in the event of a finding that reaches a pre-specified threshold, this result should be communicated to the subject to make an informed choice about a matter of major clinical significance. How this is done is a technical matter that is easily resolved.

The alternate view is that it is simplest (ie facilitates research) to remove from the researcher the onus to follow up on a ‘no’ choice, also partly on the ethical basis of autonomy.

If the intent is to foster exploration of the clinical implications and opportunities of genomic research, the former position is possibly slightly better. However, the reality is that very few people elect ‘no’. 

3. Chapter 3.5

3.5.26(a) states that ‘Researchers should not share genomic information unless sharing data is consistent with the consent that has been obtained for the research projects or for clinical purposes

It is unclear what ‘clinical purposes’ is trying to indicate here. Leaving it vague is probably no bad thing, or even important given the pace of change in knowledge that would affect the definition in an individual case. Creating mechanisms that allow for real-time and defensible evaluation of individual results is therefore essential.

3. Chapter 3.5

3.5.37 states that ‘Over time there may be substantive change in the understanding of the significance of research results or findings.  In such cases, researchers have a responsibility to provide the research cohort with the opportunity for each participant to re-consider their decision related to receiving results or findings during the period of the research project.’

  • This is in conflict with new section 3.5.17.
  • This is difficult for indefinite research projects such as cohort studies
3. Chapter 3.5

3.5.38 What about ongoing cohort studies or biobanks?  Do they have an ongoing obligation to return information?

Suggest this is impossible in many cases (deceased individuals or cohorts), would violate the terms of consent, and cripple research activity. Must be prospective.

3. Chapter 3.5

Decision Tree

  • The use of the term pertinent findings is confusing.  Why not just use primary findings?  In this context are pertinent and secondary findings both of clinical significance?  With the definitions being used in the document, not all primary findings from genomic research are clinically significant.
3. Chapter 3.5 states ‘Researchers should recognise that, while standard practice for targeted genetic research is to return individual research results, standard practice for genomic research is to not return research findings to participants’

What is the evidence for this statement?  It seems to run counter to current practice?

3. Chapter 3.5 (a)….that researchers have an obligation to have a process in place for the return of findings that are of proven validity and of health significance….

  • The definition of validity in the glossary of the document is a judgement about the likely accuracy of the findings or results as measured by NATA accredited testing or its equivalent.
  • Throughout the entire section 3.5 it is unclear whether it is recommending that a result must be of proven validity before it is returned to participants or whether it is acceptable for participants to be informed of results whilst at the same time being told that the result must be confirmed in a clinical setting. 

There are some key points.  A result cannot be truly validated unless a new blood sample is provided in a clinical setting and testing performed in a clinical setting.  This is not possible without informing the participant of the need for further testing.  A result is not validated if DNA collected in the research setting is supplied to the clinical laboratory.  In some cases a NATA accredited test will not be available to confirm research findings. 

3. Chapter 3.5 (d) What about cohorts and biobanks?  Do they have an ongoing responsibility?

3. Chapter 3.5 states ‘Where unspecified collections by biobanks are involved, researchers should describe the role, if any, that any biobank involved in the collection, management or storage of any biospecimens used in genomic research will have in the return of findings.  Researchers should note that there is no general expectation that there is a role for a biobank in the return of findings of genomic research.’

  • Does this mean that biobanks don’t need ethically defensible plans as part of ethics approval?
  • How does this relate to cohort studies that are effectively biobanks for external researchers?

Researchers receive re-identifiable specimens from biobanks and cohort studies.  If information is to be returned to study participants then by necessity the biobank/cohort has to be involved to make re-identification possible. 

3. Chapter 3.5  It is unclear the intention of this statement especially the ‘including those of associated familial cancer centres or their equivalent’ section

3. Chapter 3.5 Biobanks and cohorts generally have conditions of access to materials that include arrangements with researchers about the return of information to the biobank/cohort.  The researcher would need to consult with the biobank/cohort with regard to the statement in this section.  

3. Chapter 3.5  Again the timing of validation is not made clear or the option for validation at different timepoints is not made clear. 

3. Chapter 3.5 (a)  This would seem to indicate that findings should be validated before they are assessed for clinical significance.  This is not practical – see comments in section above.  In the context of biobanks/cohorts any discussion about assessment of clinical significance of findings should be in consultation with the biobank/cohort who are likely to already have mechanisms in place.

3. Chapter 3.5 (b) This seems to be in conflict.  ..’recommendations for finding the necessary expertise for making these judgements, if not within the expertise of the research team – a process that must (ii) be independent of the research team and any managing clinicians….

3. Chapter 3.5 (c) again this seems to indicate that the finding must be validated before returning to participants. See comments section above.

3. Chapter 3.5 (a) In practical terms, what is expected of researcher as indicated in this section where it states ‘the iterative nature of consent (ie. multiple time points)?

3. Chapter 3.5 states ‘participants should only be notified after the test validity’. Again see comments section

3. Chapter 3.5  This relates to new section 3.5.17 also and is a marked change from section 3.5.2 (c)  in the current statement. See comments above.

3. Chapter 3.5  This introduces ambiguity, see section  Do researchers have to provide or recommend to participants that they seek appropriate services?  

5. Glossary

Genomic Research - The definition of genomic research includes single gene research to whole genome sequencing.  So how can it be policy not to return genomic data as stated in

5. Glossary

Mutations -  Does the term mutation or germline or somatic even appear in the text of these chapters?

Page reviewed: 10 July, 2018