Pharmacogenetics
Individual feedback
3.44 In some cases, researchers provide individual feedback to patients. In others, researchers elect to offer individual test results to patients who request the information. (If data have been anonymised, individual feedback is of course not possible without additional samples being obtained.) There is no clear guidance on this matter in the UK. The MRC observed in 2001 that there is no consensus on the question of individual feedback and requires only that researchers should decide what level of feedback they will provide and inform patients as part of the process of obtaining consent.35 In other countries, there is a similar degree of flexibility. For example, in France, the overall results of a research project must be communicated to participants. Depending on the type of research conducted and on the results obtained, ethical review committees may also approve the feedback of individual results.36 We support the view of the Human Genetics Commission that the feedback of the overall results of research should be promoted.37
3.45 As already noted, in the UK, research participants are able to request access to information about themselves under the DPA, provided their data have not been anonymised. However, there is some uncertainty about how far this right of access extends. If the information requires specialist knowledge and processes to be conveyed to the individual in a meaningful form, it may be argued that the holder of the information does not have to make it available. Section 8(2) of the DPA states that where the data provided are ‘expressed in terms which are not intelligible without explanation, the copy must be accompanied by an explanation of these terms’, but there is uncertainty as to how far this obligation extends.
3.46 Information of immediate clinical relevance to a research participant would be passed automatically to the individual, usually through his or her physician.38 But the definition of immediate clinical relevance is not straightforward. A simple rule would be to convey information about test results outside the normal range, but in pharmacogenetic studies, this is not likely to apply. A more useful definition might be information which would reasonably be thought to indicate the presence of or significantly increased susceptibility to a disease, or which might affect the current treatment of the participant. The feedback of information that is not of immediate clinical relevance but which still has implications, or may do in the future, for the health or treatment of the participant, raises difficult issues.
3.47 Arguments against providing individual feedback in such circumstances tend to focus on the uncertain nature of research results. An important point is that tests carried out in a research setting are not as reliable or as stringent as those conducted in clinical practice. Tests would need to be repeated in a clinical setting to provide results that could form the basis for treatment. In addition, the information generated may be difficult to interpret and the relevance to the patient hard to estimate, especially if research is exploratory or at an early stage. Explaining the relevance of the data might require specialist expertise that would be difficult and costly to provide to the thousands of participants in a clinical trial. It could be argued that passing on individual results of research whose findings have not been replicated is irresponsible, since participants are unlikely to be informed subsequently if later research comes to a different conclusion. It is also unclear whether participants in clinical trials involving pharmacogenetics will necessarily have an interest in receiving individual feedback.39
3.48 However, proponents of individual feedback argue that while information may be uncertain and complex, and while not all participants may be interested in receiving feedback, a decision about whether or not the information is given should be in the hands of the individual, not the researchers. From this perspective, allowing researchers to determine whether individual feedback is provided, is unjustified paternalism which assumes that participants are not capable of understanding or dealing with information about themselves. While not recommending the unrestricted provision of individual feedback, the MRC has said that ‘participants have a right to know individual research results that affect their interests, but should be able to choose whether to exercise that right.’40
3.49 While we are sympathetic to the view that patients should have the opportunity to receive useful and validated information about their medical treatment, we consider that only on rare occasions will such information be obtained as part of research in pharmacogenetics. In the atypical cases in which a clinical trial is likely to produce validated and clinically useful data regarding individual participants, we recommend that all participants should be offered the opportunity to receive individual feedback of such data as part of the process of obtaining consent. As far as possible, the nature and implications of the information to be obtained should be explained to participants. We recognise that decisions about whether data that may be obtained in the course of research are likely to be clinically useful, and assessments of when findings can be said to be sufficiently well validated, will be complex. We therefore recommend that researchers should explain their decisions regarding the provision of individual feedback to the relevant research ethics committee.
Footnotes35 Medical Research Council (2001) Human Tissue and Biological Samples for Use in Research – Operational and Ethical Guidelines (London: MRC).
36 Article L.1122-1 of the Code of Public Health, as modified by the law of March 4, 2002.
37 Human Genetics Commission (2002) Inside Information: Balancing interests in the use of personal genetic data (London: Department of Health), para. 5.51.
38 The storage of genetic information obtained in the course of research in medical records might be thought problematic if third parties such as insurers were able to obtain access to it. However, the Association of British Insurers has issued the following statement on this matter: ‘Insurers are only interested in the results of genetic tests where the results have been communicated to patients as part of a clinical diagnostic process and then, only if the test has been approved by the Genetics and Insurance Committee (GAIC) (or is one of the tests submitted to GAIC by December 2000). Research projects rarely, if ever, produce tests that meet these criteria, and in these circumstances, insurers would not be interested in any test results that the projects did produce. In addition, whatever the circumstances, insurers do not take account of genetic test results that are made available to policyholders after their policy has been taken out.’ Joint statement on genetic research and insurance produced by the UK Forum on Genetics and Insurance, 24 April 2001.
39 Corrigan O (2003) Response to Nuffield Council on Bioethics Consultation, Cambridge. In a recent phase I clinical trial conducted with pharmacogenetic testing, none of the 23 volunteer subjects were interested in knowing their CYP2D6 status.
40 Medical Research Council (2001) Human Tissue and Biological Samples for Use in Research – Operational and Ethical Guidelines (London: MRC).