|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Precedent (Australian Lawyers Alliance) |
GENOMICS AND CAUSATION
A TEMPLATE FOR EVERY QUESTION?
By Dr Ken Maclean
Nowhere in medical negligence is the intersection between genomic diagnostics and causation more apparent than in the determination of intrinsic and extrinsic factors in cerebral palsy (CP), autism spectrum disorder (ASD) and/or intellectual disability (ID) in relation to perinatal asphyxia, fetal injury or illness in early childhood. This article focuses upon the application of genomic technologies, the robust nature of genomic pathology standards and the ultimate challenge: the clinical interpretation of potentially causative genetic variants in the judicial context. It builds upon my previous article, ‘Genomics: A new frontier in medical negligence’, published in Precedent 174.
The application of genomics to arguments of legal causation in both the prosecution and defence of medical negligence and civil liability claims is one that creates a new paradigm for many within the legal profession: that genetic disorders mimic, trigger or modulate much of what, for decades, was generally accepted to be the result of injury alone. What lies ahead are questions as to identifiable genomic contributions to multifactorial diseases, such as cancer and heart disease, foreseeable responses to injury or illness, unexplained deaths and, perhaps most daunting of all, how to rationally decipher the role of genetic factors in acquired psychiatric illnesses.
From a legal standpoint, the application of emerging genomics technologies to causation in the civil courts is one for which the same standard applies: on the balance of probabilities, but for the breach in duty of care, the person would not have suffered harm. From a practical standpoint, the debate centres on the utility, process, standards and limitations of genomic diagnostic pathology and its translation into a clinical genetic diagnosis such as in the determination of legal causation for fetal disorders, birth injuries and disease susceptibility.
NEURODEVELOPMENTAL DISORDERS
CP, ASD and ID are distinct but overlapping childhood neurodevelopmental disorders (NDDs) for which causation is often at issue. CP, in its pure form, is a non-progressive early-onset brain disorder affecting movement and posture[1] ASD and ID are primarily cognitive disorders.[2] ASD is characterised by impairment of communication and social reciprocity and a pervasive pattern of restricted and repetitive behaviours. ID and its correlate in younger children – global developmental delay (GDD) – are manifest by varying degrees of impairment of language, communication, socialisation, behaviour and motor skills. Each disorder has a potentially debilitating impact upon daily living skills, and academic and vocational attainments, as well as independent living.[3]
ASD and ID are determined according to criteria of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR),[4] an abridged compendium of neuropsychiatric diagnoses. Definitive diagnosis is usually based on a comprehensive multidisciplinary clinical and psychometric evaluation that incorporates standardised assessments of IQ together with questionnaire-based diagnostic evaluations of both everyday abilities (adaptive function) and autism-specific symptomatology. By contrast CP is a neurological diagnosis established from clinical and neuroimaging findings.[5]
There is an extensive overlap between CP, ASD and ID. Each can be associated with ‘atypical’ features such as brain malformation, other organ involvement or unexpected clinical elements.[6] These provide valuable insights into the timing and possible causation, which is encompassed by the term ‘disease aetiology’ in clinical medicine. An aetiological diagnosis does not alter the DSM-5-TR categorisation or clinical diagnosis of CP;[7] indeed the clinical categorisation retains clear utility in the clinic, where it is central to understanding disease impact, management and prognosis, and in the court, in considering short and long care requirements, likely outcomes and costs of compensable damages.[8] Aetiological considerations in CP range from a high likelihood of an environmental cause in an extremely premature infant,[9] infant with brain haemorrhage before or after birth, or postnatally acquired infection,[10] such as acute bacterial meningitis, to probable genetic aetiology in the infant with congenital anomalies and a distinctive appearance. Congenital cytomegalovirus (CMV)[11] and its genetic mimic Aicardi-Goutières syndrome[12] – a condition that elucidates the mechanisms by which congenital CMV causes fetal brain injury – highlights the importance of considering genetic causation even with proven in utero CMV exposure. Analogous to CP and perinatal asphyxia, the presence of a consistent pattern of clinical and laboratory findings that fulfils consensus criteria for congenital CMV does not preclude genetic causation. The greatest and most common challenge is in the middle ground: the term infant with no identified risk factors outside of a complicated delivery.
The demonstrated existence of genetic forms of CP that have an onset in early childhood (hereditary spastic paraplegias),[13] and the recognition clinically, epidemiologically and genomically that CP is simply a clinical descriptor rather than an aetiological diagnosis, has led to calls for genomic testing to be offered to all children with CP.[14] The underlying tenet in perinatal asphyxia is that the presence of a primary genetic disorder known to cause CP, ASD or ID increases the likelihood of fetal maladaptation with the physiological stresses that are intrinsic elements of labour and delivery. Effectively there is a continuum in fetal adaptive response, reserve and compensatory ability that in some instances may be independent and others highly interdependent on the events of labour and delivery in the genesis of a neurodevelopmental disability in a child. Proof of concept predates the genomic era. The value of genomic testing is that it replaces hundreds of individual tests with a single high-yield test.[15]
Genomic testing has shown that rare disease is common.[16] Technological and analytic advances have accelerated the ability to pinpoint meaningful genetic variants. The global agglomeration of linked genomic and clinical datasets into curated open-access databases has improved the quality, robustness and direct utility of test reports that are issued by genomic pathology laboratories. What was once considered the exception in CP, ASD and ID is that individually rare and ultra-rare diseases are collectively common and readily discoverable in up to a quarter of unselected cases.[17] DNA testing is routine, as are orders for genomic testing in children with moderate to severe neurodevelopmental disability.
Gene lists for conditions such as ASD and ID continue to expand with at least 1,500 ‘high confidence’ genes now associated with ID/ASD.[18] Not surprisingly these genes play an important role in brain development and neuronal function.[19] Genomic testing in CP has shown convergence with prior clinical understanding – that genetic error in factors affecting blood flow, oxygenation and essential cell functions in areas of the brain susceptible to hypoxic-ischaemic brain injury at birth may cause CP in infants – with or without a history of overt fetal distress.
Errors in an individual gene may occur anywhere within the region of the 10,000 plus nucleotides that might make up an average-sized gene. It is the case in children with severe cognitive disability that the causative gene errors are usually highly disruptive errors of genes critical for normal brain development and that these errors are typically de novo meaning a new genetic change arising just in the affected individual. This is identified by many litigants – that the disorder has arisen in the plaintiff in the absence of any relevant family history, albeit with a different inference.
Knowledge of how high-impact genetic variants segregate in families is often lacking in the litigated context, where DNA testing is more likely to be restricted to the affected individual. Incorporation of parental DNA samples, which is usual in clinical testing, informs whether a gene variant associated with a condition such as ID/ASD is present in other family members and if so, what the impact is on other family members. Arguably, inclusion of parental samples may simplify the diagnostic testing process and clarity of the result obtained by the courts.
The increasingly granular level of knowledge of individual genes and specific gene errors is one that is accompanied by a nascent knowledge of how genetic and cellular pathways link together to give rise to disease. A practical example is the differentiation of ‘ASD-predominant’ genes from ‘ID genes’ albeit that this is controversial.[20] In contrast with the scientific knowledge, it is often the case that the clinical elements, which are harder to curate than a discrete genetic variant, are not as well defined. The consequence of a lack of detail in database submissions and publications is that it makes the process of applying the knowledge to individual cases particularly challenging.
THE FOLBIGG INQUIRY AND BEYOND
Accompanying the increasing confidence in the quality of the pathology results that are reported has been a shift in emphasis onto understanding what individual genetic variants mean for individual cases in a specific clinical context. Interpreting a genomic pathology report is not simply a matter of applying a clinical viewpoint. It is a position derived from multiple expert inputs in the process of clinical and laboratory evaluation. This may involve the use of computational prediction programs, detailed analyses of genomic databases and the genetic literature, networked outreach, testing of other family members or functional studies that seek to establish how a genetic variant affects the cell or a model organism such as a mouse or zebrafish. All emanating from the primary pathology report, its clinical interpretation and the importance of appropriate evaluation to determine direct relevance.
Nowhere is this better exemplified in real time than the Folbigg Inquiry prompted by a petition to the Governor of NSW from over 90 pre-eminent scientists, clinicians and academics, who collectively identified the relevance of a genomic pathology test result to causation arguments which was not available at the time of the initial trial.[21] The inquiry has heard from a wide range of experts – clinical geneticists, genomic pathologists, research scientists and clinicians – in fields directly related to the question of genetic causation.[22] Much of the inquiry has centred on the analysis, clinical and functional interpretation of a single nucleotide variant (SNV) present in two of the four deceased infants.[23] There has been consensus amongst the expert opinions that a previously unreported change in the CALM2 genetic sequence, classified by the reporting pathologists as a variation of uncertain significance (VUS), has created a change in the ‘Calmodulin 2’ protein that is true and real and of clear importance to the case. The divergence arises, not whether it is possible, rather whether it is likely that this change has caused the sudden cardiac death of the two infants who inherited the gene change from their mother. That the children had intercurrent illnesses highlights the potential relevance of environmental factors as a possible trigger for a cascade of events leading to fatal arrhythmia in the presence of a genetic susceptibility. The distinction between VUS and ‘likely pathogenic' (likely disease causing) is central to what may be fundamentally differing opinions on causation.
Similarly high-profile criminal cases in the UK that exemplified the need to consider genetic causation include those of Sally Clark,[24] Angela Canning[25] and Donna Anthony,[26] whose convictions for the murder of their own infants were overturned on appeal between 2003–5, and Trupti Patel, who was acquitted of the same charges in 2003.[27] These matters not only garnered attention towards genetic diagnosis. They led to a paradigm shift on expert evidence in unexplained child deaths.[28] The counterpoint to the question of an inflicted injury, such as asphyxiation, is the possibility of a genetic cause identifiable only by genomic sequence and the need to consider this alongside other causes of death. Nearly 20 years after the judgment in R v Cannings,[29] the Folbigg Inquiry shows the application of genomic science to the courts while increasingly routine is ‘still at the frontiers of knowledge’.[30]
The concept of host-response or gene-environment interactions is one that is well established in toxic torts.[31] In clinical paediatrics, an understanding of genetic epilepsy disorders has helped determine why something as routine as childhood immunisation might trigger the onset of seizures and a seemingly inexorably progressive decline in mental function from intractable epilepsy.[32] Dravet syndrome is an archetypal ‘epileptic encephalopathy’ in which fever or illness triggers drug-resistant seizures and injury. Prior to genetic and genomic sequencing, diagnosis was not possible, nor could treatments be as precisely tailored. The ability to detect an underlying genetic disorder is of critical relevance in considering causation at law: be it the proximate cause of an immunisation or an illness or the underlying disorder that is highly likely to become manifest given the appropriate trigger. To this end, the question and supposition arises as to whether the child with a primary genetic form of CP, ASD or ID is more likely to exhibit fetal distress during delivery as a consequence of their disorder or as a consequence of the delivery, or whether both may be the case.
Even in strongly genetically determined disorders such as familial breast and ovarian cancers, it is clear that factors beyond the individual genetic change influence the occurrence and nature of a cancer.[33] The genetic variant or mutation may be a necessary part of a causative cascade but is not sufficient of itself to cause cancer. The other events can be chance genetic events related to the primary mutation, the effect of other genes (polygenic) or result from specific exposures. A tool used in genomic research that has been introduced clinically and commercially is a polygenic risk score,[34] which looks at multiple genetic factors in combination to determine a likelihood of cancer, heart disease, diabetes, etc. While best considered a tool in development, like many genomic diagnostic tests, it is quickly moving into the clinical domain.
PSYCHIATRIC GENETICS
Perhaps the most challenging area of all is that of psychiatric and behavioural genetics. Early iterations of legal and clinical DNA testing in this area were flawed[35]. These focused on common dichotomous variants in a handful of genes known to play a role in key biochemical pathways in the brain using data from small ‘underpowered’ studies. They endeavoured to correlate complex human behaviours such as risk-taking or antisocial behaviour (DRD4 dopamine receptor)[36] love and fidelity (vasopressin/oxytocin receptors)[37] or response to childhood trauma (serotonin transporter 5HTT)[38] to discrete genetic variants present throughout the wider population. The rush to prematurely draw conclusions in the face of low quality-evidence, limited knowledge and poorly nuanced assumptions on the genetics of common disorders is of little merit. Even large, well-powered studies – prior to genomic sequencing – produced conflicting information as to which of the 20,000 genes might be important in common disorders such as depression, anxiety, PTSD, OCD and ADHD. Somewhat surprisingly it has been commercial as well as national DNA biobanks, for example 23andMe and the UK Biobank, with sample numbers in the hundreds of thousands that have begun to better elucidate ‘high-confidence’ genes for common psychiatric disorders.
The move from candidate genes to clinical genetic diagnoses that stand up to robust analysis in psychiatry remains some time away, even with increasingly collaborative multinational approaches to the collection and evaluation of large scale genomic and clinical data. The exception is schizophrenia, which broadly parallels NDDs in its genetic and other causation.[39] It will take time for psychiatric genetic testing to evolve to a state of being broadly clinically applicable. The Folbigg Inquiry shows just how divided clinical opinion can be on genetic causation in what is a well-established disease model of sudden cardiac death due to an arrhythmic disorder with expert witnesses who are well versed in genetic testing and its clinical interpretation. There is a need to foster similar expertise across the diverse areas of psychiatric practice.
CONCLUSION
Overall, there is no doubt that genomic testing has a clear place in civil litigation as well as the criminal courts. Spearheaded by case law in obstetric and paediatric medicine, it is increasingly undertaken at a preliminary stage, during discovery or interlocutory proceedings as a foreseeable requirement to bring cases to trial or for determination of settlements at mediation. The identification of a ‘high impact’ genomic diagnosis where genetic causation is established or accepted will lead to abandonment, earlier settlement or a permanent stay of many cases. In most cases, genomic testing will be uninformative. What is critical to appreciate is that the existence of genetic variant and its link to a clinical diagnosis does not equate to proof of causation. Accurate clinical interpretation and understanding as to the range of consequences, which is increasingly done prior to trial, is critical to narrowing down what is at dispute and what might reasonably be predicted but for the interceding event. It remains to be seen as to what constraints are placed upon orders for genomic testing, determinations of merits prior to testing, admissibility of variants of uncertain significance and ultimately, its probative value in an ever-widening array of cases.
Dr Ken Maclean is a clinical geneticist in private practice with an active interest in reproductive genetics and its role in child health, genomic testing in litigation and health law.
[1] V Horber et al, ‘The role of neuroimaging and genetic analysis in the diagnosis of children with cerebral palsy’, Frontiers in Neurology, 2021, 11:628075.
[2] A Thurm et al, ‘State of the field: Differentiating intellectual disability from autism spectrum disorder’, Frontiers in Psychiatry, Vol. 10, 2019, 526.
[3] SS Kuo et al, ‘Developmental variability in autism across 17 000 autistic individuals and 4000 siblings without an autism diagnosis: Comparisons by cohort, intellectual disability, genetic etiology, and age at diagnosis’, JAMA Pediatrics, Vol. 176, No. 9, 2022, 915–923.
[4] Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR) 5th ed; American Psychiatric Association, 2013.
[5] M Bax, C Tydeman and O Flodmark, ‘Clinical and MRI correlates of cerebral palsy: The European Cerebral Palsy Study’, JAMA, 2006, Vol. 296, No. 13, 1602–8; SA Lewis et al, ‘Insights from genetic studies of cerebral palsy’, Frontiers in Neurology, Vol. 11, 2021, 625428.
[6] M Severino et al, ‘Definitions and classification of malformations of cortical development: Practical guidelines’, Brain, Vol. 143, No. 10, 2020, 2874–2894, with a correction in Brain, Vol. 143, No. 12, 2020, e108.
[7] AH MacLennan et al, ‘Genetic or other causation should not change the clinical diagnosis of cerebral palsy’, Journal of Child Neurology, Vol. 34, No. 8, 2019, 472–476.
[8] S Kularatna et al, ‘The cost of neurodevelopmental disability: Scoping review of economic evaluation methods’, ClinicoEconomics and Outcomes Research, Vol. 14, 2022, 665–682.
[9] F Schachinger and S Farr, ‘The effects of preterm birth on musculoskeletal health-related disorders’, Journal of Clinical Medicine, Vol. 10, No. 21, 2021, 5082.
[10] M Mynarek et al, ‘Incidence of invasive group B streptococcal infection and the risk of infant death and cerebral palsy: A Norwegian cohort study’, Paediatric Research, Vol. 89, No. 6, 2021, 1541–1548.
[11] LT Ong and SWD Fan, ‘The association between congenital cytomegalovirus infection and cerebral palsy: A systematic review and meta-analysis’, Journal of Paediatrics and Child Health, Vol. 58, No. 12, 2022, 2156–2162.
[12] DM d'Angelo et al, ‘Type I interferonopathies in children: An overview’, Frontiers in Pediatrics, Vol. 9, 631329.
[13] P Hedera, ‘Hereditary spastic paraplegia overview’, in MP Adam et al (eds), GeneReviews [Internet] University of Washington, Seattle, 15 August 2000 [Updated 11 Feb 2021] 1993–2023.
[14] AH Maclennan, ‘All cases of cerebral palsy warrant genomic screening’, Developmental Medicine & Child Neurology, 2021, Vol. 63, No. 12, 1369.
[15] S Srivastava et al, ‘Meta-analysis and multidisciplinary consensus statement: Exome sequencing is a first-tier clinical diagnostic test for individuals with neurodevelopmental disorders’, Genetics in Medicine, Vol. 21, No. 11, 2019, 2413–2421. Also see JM Savatt and SM Myers, ‘Genetic Testing in Neurodevelopmental Disorders’, Frontiers in Pediatrics, Vol. 9, 2021, 526779.
[16] T Maroilley and M Tarailo-Graovac, ‘Uncovering missing heritability in rare diseases’, Genes, Vol. 10, No.4, 2019, 275.
[17] MA Corbett et al, ‘Pathogenic copy number variants that affect gene expression contribute to genomic burden in cerebral palsy’, npj Genomic Medicine, Vol. 3, 2018, 33. Also see CL van Eyk et al, ‘Yield of clinically reportable genetic variants in unselected cerebral palsy by whole genome sequencing’, npj Genomic Medicine, Vol. 6, No. 74, 2021; JM Friedman, P van Essen P and CDM van Karnebeek, ‘Cerebral palsy and related neuromotor disorders: Overview of genetic and genomic studies’, Molecular Genetics and Metabolism, Vol. 137, No. 4, 2022, 399–419; HJ May et al, ‘Genetic testing in individuals with cerebral palsy’, Developmental Medicine and Child Neurology, Vol. 63, No.12, 2021, 1448–1455; and SA Lewis et al, above note 5; S Srivastava et al, ‘Meta-analysis and multidisciplinary consensus statement: Exome sequencing is a first-tier clinical diagnostic test for individuals with neurodevelopmental disorders’, Genetics in Medicine, Vol. 21, No. 11, 2019, 2413–2421. Also see JM Savatt and SM Myers, ‘Genetic Testing in Neurodevelopmental Disorders’, Frontiers in Pediatrics, Vol. 9, 2021, 526779.
[18] M Nakanishi, MP Anderson and T Takumi, ‘Recent genetic and functional insights in autism spectrum disorder’, Current Opinion in Neurology, Vol. 32, No. 4, 2019, 627–634.
[19] N Maia et al, ‘Intellectual disability genomics: current state, pitfalls and future challenges’, BMC Genomics, Vol. 22, No. 1, 2021, 909.
[20] B Trost et al, ‘Genomic architecture of autism from comprehensive whole-genome sequence annotation’, Cell, Vol. 185, No. 23, 2022, 4409–4427, at e18. Also see SM Myers et al, ‘Insufficient Evidence for "Autism-Specific" Genes’, American Journal of Human Genetics, Vol. 106, No. 5, 2020, 587–595; SM Myers et al, ‘Response to Buxbaum et al,’ The American Journal of Human Genetics, Vol. 107, No. 5, 2020, 1004; and FK Satterstrom et al, ‘Consortium large-scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism.’ Cell, Vol. 180, 2020; 568–584, at e23.
[21] Petition to Governor of New South Wales for Pardon of Kathleen Folbigg (2021) <https://www.science.org.au/files/userfiles/events/news/documents/petition-to-governor-of-nsw-for-pardon-of-kathleen-folbigg-05-03-21.pdf>.
[22] NSW Government, 2022 Inquiry into the convictions of Kathleen Megan Folbigg <https://2022folbigginquiry.dcj.nsw.gov.au>.
[23] M Brohus et al ‘Infanticide vs. inherited cardiac arrhythmias,’ EP Europace Vol. 23 No. 3, 2021 441-450. Correction published in EP Europace euad021.
[26] R v Anthony [2005] EWCA Crim 952.
[27] Cannings, above note 25; R Hill, ‘Reflections on the cot death cases’, Significance, Vol. 2, No. 1, 2005, 13–16.
[28] Attorney General, Review of infant death cases following the Court of Appeal decision in the Case of R v Cannings, London: Stationery Office, 2004; R Nobles and D Schiff, ‘Misleading statistics within criminal trials: The Sally Clarke case.’ Significance, Vol. 2, No. 1, 2005, 17–19. Also see C Dyer, ‘Parents convicted of killing to have their cases reviewed’, British Medical Journal, Vol. 328, No. 7443, 2004, 183.
[29] Cannings, above note 25.
[30] Ibid, [178].
[31] S Golru, ‘The extrapolation dilemma: toxicological evidence and toxic torts’, Torts Law Journal, Vol. 27, No. 3, 2022, 210–233.
[32] AM McIntosh et al, ‘Effects of vaccination on onset and outcome of Dravet syndrome: a retrospective study’. Lancet Neurolology, Vol. 9, No. 6, 2010, 592–598.
[33] A Rudolph, J Chang-Claude and Schmidt MK, ‘Gene-environment interaction and risk of breast cancer’, British Journal of Cancer, Vol. 114, No. 2, 2016, 125–33.
[34] CM Lewis and E Vassos, ‘Polygenic risk scores: from research tools to clinical instruments’, Genome Medicine, Vol. 12, No. 44, 2020.
[35] M Sabatello and PS Appelbaum, ‘Behavioral genetics in criminal and civil courts’, Harvard Reviews in Psychiatry, Vol. 25, No. 6, 2017, 289–301.
[36] R Ptácek, H Kuzelová and Stefano GB, ‘Dopamine D4 receptor gene DRD4 and its association with psychiatric disorders’, Medical Science Monitor, Vol. 17, No. 9, 215–220.
[37] SM Phelps, M Okhovat and A Berrio, ‘Individual differences in social behavior and cortical vasopressin receptor: Genetics, epigenetics, and evolution’, Frontiers in Neuroscience, Vol. 4, No. 11, 2017, 537–549; DA Baribeau and E Anagnostou, ‘Oxytocin and vasopressin: Linking pituitary neuropeptides and their receptors to social neurocircuits’, Frontiers in Neuroscience, Vol. 24, No. 9, 2015, 335.
[39] HL Peay, ‘Genetic Risk Assessment in Psychiatry’, Cold Spring Harbour Perspectives in Medicine, Vol. 10, No. 12, 2020, a036616.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/PrecedentAULA/2023/20.html