Genetic testing in the acute setting: a round table discussion

Abstract

Genetic testing has historically been performed in the context of chronic disease and cancer diagnostics. The timelines for these tests are typically measured in days or weeks, rather than in minutes. As such, the concept that genetic information might be generated and then used to alter management in the acute setting has, thus far, not been feasible. However, recent advances in genetic technologies have the potential to allow genetic information to be generated significantly quicker. The m.1555A>G genetic variant is present in one in 500 individuals and predisposes to profound hearing loss following the administration of aminoglycoside antibiotics. These antibiotics are used frequently in cases of neonatal sepsis and it is estimated that approximately 180 neonates in the UK are at risk of antibiotic induced hearing loss each year because of this genetic change. Knowledge of this variant in the acute setting would allow clinicians to prescribe alternative antibiotics. The Pharmacogenetics to Avoid Loss of Hearing study will implement a genetic point of care test (POCT) for the m.1555A>G variant within two major UK based neonatal intensive care units. This represents the first trial of a genetic POCT aimed at altering management in the acute setting. This round table discussion outlines the novel ethical issues faced in the development of this trial and the legal barriers to implementation. We ask five stakeholders to provide their opinions on this trial and their perspectives on the concept of genetic testing in the acute setting.

Trial registration number

ISRCTN-13704894.

Introduction

To date, the clinical delivery of genetic testing has been in the context of chronic disease, aiding the specific diagnosis of the thousands of rare inherited disorders, the classification of cancer subtypes and prediction of response to cancer treatment. The results of these tests also allow clinicians, including geneticists and genetic counsellors to inform patients and their family members about their individual risks of developing specific disorders. The time between sample collection and the delivery of a clinically relevant result is typically measured in days and weeks rather than in minutes. That genetic information could be generated in real time and be used to influence treatment decisions in the acute setting has, thus far, not been feasible. However, genetic technologies have been developed in recent years which have the potential to provide clinically actionable results in a more rapid timeframe.

Ethical debate around genetic testing has, for the most part, been held in the context of current genetic testing approaches. It is widely accepted that individuals should be counselled prior to genetic testing, considering the nature of the test; the potential outcomes, including the psychological effects and need for further investigations or screening; the potential for unexpected findings revealing predisposition to diseases for which the initial testing was not undertaken, and the impact that any result might have on other family members.^{1 2} This is a detailed, technical and individualised process which should be undertaken by genetic counsellors, clinical geneticists or trained healthcare professionals. However, if genetic test results can be generated rapidly, in the context of an acute admission, this approach becomes more difficult to follow.

Aminoglycoside antibiotics such as gentamicin are commonly used around the world, approximately 7 million babies receive them each year. Their broad spectrum of activity coupled with their relative low cost make them a desirable agent for a range of indications including as a first line agent in neonatal sepsis.³ It is well known that aminoglycoside antibiotics given in high doses or over a protracted period can lead to deafness. However, certain families have a predisposition to this side-effect whereby just a single dose of aminoglycoside, particularly in early childhood, can cause a profound and irreversible hearing loss.⁴ It was recognised in 1993 that these families carry a specific genetic variant (mutation) called m.1555A>G in their mitochondrial DNA, affecting a gene called RNR1.⁵ The m.1555A>G variant is present in approximately one in 500 individuals and takes a minimum of 2 days to test for using traditional genetic testing techniques.⁶

When neonates present with sepsis, UK guidelines advise that individuals should be treated with a combination of a beta-lactam and gentamicin, with the first dose of antibiotic given within 1 hour (NICE CG149). As such, traditional sequencing approaches cannot be used to identify those carrying the m.1555A>G variant within a clinically relevant timeframe. However, a point-of-care-test (POCT) has been recently developed that can detect the presence of the variant from a buccal swab in approximately 25 min. In principle, this would allow neonates to be tested for the variant in the neonatal intensive care unit (NICU) before they receive their dose of antibiotic, thus allowing tailored antibiotic prescription.

Although it has been possible to develop this rapid sequencing approach, the implementation of this technology within ‘real world’ clinical practice requires examination. Could a genetic test really be integrated into practice without affecting the normal flow of care? The Pharmacogenetics to Avoid Loss of Hearing (PALOH) trial was designed to find an answer to that question. The trial involves testing for the m.1555A>G variant, via the new POCT, in every neonate admitted to two large UK based NICUs over a 6-month period. The test will be integrated into the admission procedure at each centre, allowing tailored antibiotic prescribing to take place based on the result.

Given the acute context in which the PALOH trial is taking place, it was recognised that it would not be practicable for informed consent to be gained prospectively. Newborns admitted to NICU are acutely unwell and this is likely to represent one of the most stressful time in a parent’s life. As such, it was recognised that asking for consent to perform this genetic test as part of a study, immediately after birth, would be entirely inappropriate. Following discussion with several stakeholders, including parents, neonatologists and geneticists, it was decided that an opt-out consenting model should be sought. The neonates would be tested for the variant as part of the admission process via a buccal swab, the variant details would be used to inform prescribing and data would be subsequently gathered and analysed as part of the feasibility trial. Parents will be given information about the trial at an appropriate point on NICU, and they would have the right to ask for their child’s data to be removed from the study at that stage. This provides a mechanism to opt-out at the analysis stage, although by this point their child will have already had tailored antibiotics prescribing based on the presence or absence of the variant.

To our knowledge this represents the first example of an opt-out consenting model being used in a trial involving genetic testing. Genetic exceptionalism might lead ethics committees to be particularly cautious when considering consent mechanisms when genetic testing is being performed. The concern being that any genetic result may have relevance not simply for the individual tested, but for the wider family. The m.1555A>G variant lies within the mitochondrial, rather than the nuclear, genome meaning that it is inherited down the maternal lineage. As such, many maternal relatives of the proband will also carry the variant, conferring the same risk of antibiotic-induced hearing loss to those individuals.

In the summer of 2019, an application for ethical approval for the PALOH trial was submitted to the National Health Service Research Ethics Committee (REC), contesting that the potential benefits warranted approval of an opt-out consenting model. After consideration, the REC approved the PALOH trial design, pending Human Research Authority (HRA) approval, making it the first trial of a genetic POCT to alter management in the acute setting. During the preparation of the ethics application, it was presumed that having a REC committee approve an opt-out consenting model would be the biggest hurdle; therefore, this was viewed as a significant moment in the project. However, in the week following the REC meeting the PALOH study team received notice that the HRA were withholding approval as they had concerns that the study was in breech of the Human Tissue Act (2004) (HT Act).

The HT Act was written in the wake of the Alder Hey organs scandal to regulate the removal, storage and use of human tissue. Section 45 and Schedule 4 of the HT Act set out the provisions regarding DNA analysis. An individual has committed an offence under section 45 if they:

1. Are in possession of bodily material intending that DNA in the material be analysed without qualifying consent.

2. The results of the analysis of DNA be used for something other than an excepted purpose.

For the purposes of the PALOH trial, although the REC approved the use of an opt-out consenting model, the HRA felt that this did not meet the criteria for ‘qualifying consent’.

The PALOH trial was initially due to start recruiting in Autumn of 2019, however, this question from the HTA threatened not only the future of PALOH, but the whole concept of acute genetic testing. It was strongly felt that the intended purpose of the HTA was not to prevent implementation studies such as PALOH, however, in 2004 the landscape of genetic technology was fundamentally different from today. The Human Genome Project was completed in 2003 and represented the culmination of a 13-year international collaboration to map a single human genome costing approximately $3 billion.⁷ Now, a genome can be sequenced in under 24 hours for approximately $500 in a single laboratory. This great leap forwards exemplifies how it is reasonable to assume that, in 2004, legislators may not have envisaged that genetic data might be generated in real time via a POCT to alter management in an acutely unwell individual.

Following the HRA’s intervention, the PALOH trial was temporarily suspected until a resolution could be identified. Input from the Human Tissue Authority and legal experts was sought in the drafting of the response to the HRA. The opinions and perspectives of key stakeholders are presented below.

This type of POCT technology and its implementation in the PALOH trial represents a new paradigm for how genetics might be applied in the acute setting. The ability to identify clinically relevant genetic information by the bedside may offer tangible patient benefit, but it also raises a number of ethical and legal questions. This round-table discussion asks experts and patients to consider the acceptability of POCT genetics in the acute setting. We will explore the ethical challenges raised by testing prior to informed consent, the burden of responsibility placed on the practitioner and the wider societal impacts technology like this might begin to have as genetics becomes a routine part of everyday healthcare. By documenting the ethical and legal issues precipitated by PALOH, and our approach towards these issues, we hope to support future investigators aiming to implement genetic data to improve patient outcomes in the acute setting.