Clinical exome sequencing elucidates underlying cause of death in sudden unexpected death of infants: two case reports

Sudden unexpected death in infants (SUDI) is a traumatic event for families, and unfortunately its occurrence remains high in many parts of the world. Whilst cause of death is resolved for most cases, others remain undetermined following postmortem investigations. There has been a recognition of the role of genetic testing in unexplained cases, where previous studies have demonstrated the resolution of cases through DNA analyses. Here we present two case reports of SUDI cases admitted to Salt River Mortuary, South Africa, and show that underlying causes of death were determined for both infants using clinical exome sequencing. The first infant was heterozygous for a variant (rs148175795) in COL6A3, which suggested a bronchopulmonary dysplasia phenotype. This hypothesis led to finding of a second candidate variant in DMP1 (rs142880465), which may contribute towards a digenic/polygenic mechanism of a more severe phenotype. Histological analysis of retained tissue sections showed an asphyxial mechanism of death, where bronchiolar muscle weakness from an underlying bronchopulmonary dysplasia may have contributed to the asphyxia by affecting respiration. In the second infant, a homozygous variant (rs201340753) was identified in MASP1, which was heterozygous in each parent, highlighting the value of including parental DNA in genetic studies. Whilst mannose-binding lectin deficiency could not be assessed, it is plausible that this variant may have acted in combination with other risk factors within the triple-risk model to result in sudden death. These results may have genetic implications for family members, and represent possible new candidate variants for molecular autopsies.

Understanding the burden of unidentified bodies: a systematic review

While human identification is a crucial aspect of medico-legal investigations, many individuals remain unidentified each year across the world. The burden of unidentified bodies is often referred to when motivating for improved methods of identification, and anatomical teaching, yet the actual burden is somewhat unclear. A systematic literature review was undertaken to identify articles that empirically investigate the number of unidentified bodies experienced. Despite the large number of articles returned, an alarmingly low number (24 articles) provided specific and empirical details on the number of unidentified bodies, demographics and trends thereof. It is possible that this lack of data is due to the variable definition of 'unidentified' bodies and the use of alternative terminology such as 'homelessness' or 'unclaimed' bodies. Nevertheless, the 24 articles provided data for 15 forensic facilities across ten countries of both developed and developing statuses. On average, developing countries experienced more than double (9.56%) the number of unidentified bodies when compared to developed nations (4.40%). While facilities were mandated under different legislations and infrastructures available varied greatly, the most common issue faced is the lack of standardised procedures for forensic human identification. Further to this, the need for investigative databases was highlighted. Through addressing the standardisation of identification procedures and terminology, alongside the appropriate utilisation of existing infrastructure and database creation, the number of unidentified bodies could be significantly reduced globally.

Informing regional taphonomy research using retrospective forensic anthropology cases in the Western Cape, South Africa

The post-mortem interval (PMI) is important in death investigations as it can assist in reconstructing the context of the case and the identification of the unknown individual. However, in some instances the PMI is challenging to estimate due to the absence of regional taphonomy standards. To conduct accurate and locally relevant forensic taphonomy research, investigators require insight into the regional hotspot recovery areas. Thus, forensic cases examined by the Forensic Anthropology Cape Town (FACT) in the Western Cape (WC) province of South Africa between 2006 and 2018 (n = 172 cases; n = 174 individuals) were retrospectively reviewed. In our study, a considerable number of individuals did not have the PMI estimations (31 %; 54/174) and the ability to estimate a PMI was significantly associated with skeletal completeness, unburnt remains, absence of clothing and the absence of entomological evidence (p < 0.05 for each). Significantly fewer cases had the PMI estimated after FACT was formalised in 2014 (p < 0.0001). One third of cases with PMI estimations were made with large open-ended ranges, making them less informative. Factors significantly associated with these broad PMI ranges were fragmented remains, absence of clothing and absence of entomological evidence (p < 0.05 for each). Most decedents (51 %; 87/174) were discovered in police precincts from high crime areas, yet a considerable number (47 %; 81/174) were also found in low crime and sparsely populated areas regularly used for recreational activities. Common sites of body discovery were vegetated areas (23 %; 40/174), followed by the roadside (15 %; 29/174), aquatic environments (11 %; 20/174) and farms (11 %; 19/174). Decedents were discovered exposed (35 %; 62/174), covered with items such as bedding or shrubs (14 %; 25/174) or buried (10 %; 17/174). Our data highlight gaps in forensic taphonomy studies and clearly informs what regional research is needed. Overall, our study demonstrates how forensic case information can inform regional forensic taphonomy studies to locate common areas and contexts for the discovery of decomposed bodies and encourages similar studies in other areas of the world.

Forensic DNA extraction methods for human hard tissue: A systematic literature review and meta-analysis of technologies and sample type

DNA identification of human remains has a valuable role in the field of forensic science and wider. Although DNA is vital in identification of unknown human remains, post-mortem environmental factors can lead to poor molecular preservation. In this respect, focus has been placed on DNA extraction methodologies for hard tissue samples, as these are the longest surviving. Despite decades of research being conducted on DNA extraction methods for bone and teeth, little consensus has been reached as to the best performing. Therefore, the aim of this study was to conduct a thorough systematic literature review to identify potential DNA extraction technique(s) which perform optimally for forensic DNA profiling from hard tissue samples. PRISMA guidelines were used, by which a search strategy was developed. This included identifying databases and discipline specific journals, keywords, and exclusion and inclusion criteria. In total, 175 articles were identified that detailed over 50 different DNA extraction methodologies. Results of the meta-analysis conducted on 41 articles - meeting further inclusion criteria - showed that statistically significant higher DNA profiling success was associated with solid-phase magnetic bead/resin methods. In addition, incorporating a demineralisation pre-step resulted in significantly higher profiling successes. For hard tissue type, bone outperformed teeth, and even though dense cortical femur samples were more frequently used across the studies, profiling success was comparable, and in some cases, higher in cancellous bone samples. Notably, incomplete data sharing resulted in many studies being excluded, thus an emphasis for minimum reporting standards is made. In conclusion, this study identifies strategies that may improve success rates of forensic DNA profiling from hard tissue samples. Finally, continued improvements to current methods can ensure faster times to resolution and restoring the identity of those who died in obscurity.

Forensic human identification: retrospective investigation of anthropological assessments in the Western Cape, South Africa

The identification of unknown persons, particularly those who are decomposed or burnt, is a global challenge. Forensic Anthropology Cape Town (FACT) is a service provider that assists the South African state with the identification of human remains. However, empirical data pertaining to anthropologically analysed forensic cases in the Western Cape Province of South Africa are lacking. Therefore, anthropological data pertaining to the forensic cases submitted to FACT between 2006 and 2018 from Forensic Pathology Services were retrospectively reviewed (n = 172). This study also sought to assess demographic, traumatic and pathological factors that impacted successful identification. Most decedents were male (67%) and older than 35 years (54%). While ante-mortem trauma was observed in 41% of decedents, the lack of medical records on ante-mortem injuries hindered the use of this information for identification. Positive identifications were reached for 37% of decedents, and of these, anthropological estimations were correct in 98% of cases for sex, 84% of cases for age at death and 100% of cases for stature. Considering globally accepted accuracies of 70-80%, these estimations were considered highly accurate, suggesting the anthropological methods used are suited to the population. However, 63% of cases remained unidentified, and this study showed that skeletal completeness and pathological conditions were the main factors that hindered demographic estimations. Lastly, not all unidentified bodies in the province were referred to FACT; given the highly accurate estimations, these data advocate for the routine, if not mandatory, use of forensic anthropology services for skeletonised remains in South Africa, with the overall purpose of positively impacting human identification. To improve identification rates globally, these data highlight the value of retrospective and region-specific studies to identify strengths and weaknesses in the system.

The effect of NucleoSpin® Forensic Filters on DNA recovery from trace DNA swabs

Forensic DNA profiling is a globally accepted method for human identification, however, obtaining full DNA profiles from trace DNA can be challenging. The optimal recovery of DNA from trace DNA swabs is therefore crucial. Methods for extracting DNA from swabs often make use of a spin basket combined with a centrifugation step, to enhance the release of cells from the swab prior to DNA extraction. The NucleoSpin® Forensic Filter (Macherey-Nagel, Düren) is a type of spin basket, but it has not been thoroughly assessed on trace DNA samples. This study aimed to assess if the inclusion of the NucleoSpin® Forensic Filter significantly improved DNA recovery and DNA profiling success from cotton and flocked swabs used to collect trace DNA and buccal cells (control). Buccal cells and trace DNA samples were collected from 25 volunteers using each swab type (cotton and flocked) in duplicate. DNA was extracted from the samples using the NucleoSpin® DNA Forensic kit, one set with, and the other set without, NucleoSpin® Forensic Filters. DNA concentration was assessed using real time PCR, and DNA profiling was done using the PowerPlex® ESX 16 system. The inclusion of the NucleoSpin® Forensic Filters significantly improved DNA concentration for buccal cells that were collected using flocked swabs (p = 0.035). However, no significant differences were noted for trace DNA samples for either swab type. There was also no significant difference in DNA profiling success when NucleoSpin® Forensic Filters were used, regardless of swab and sample type. These results may be helpful for laboratories that are considering the NucleoSpin® Forensic Filters in the DNA extraction workflow, particularly for trace DNA samples.

A review of the causes and risk factors for sudden unexpected death in the young

Sudden unexpected death in the young (SUDY) is a tragic event resulting in the fatality of seemingly healthy individuals between the ages of one and 40 years. Whilst studies have been performed on sudden unexpected death in infants, children, and adults respectively, little is known about trends in risk factors and causes of death of SUDY cases. Understanding the factors surrounding these deaths could lead to targeted interventions for at-risk individuals. Hence, a systematic approach to investigate the reported possible causes of SUDY was employed using three major databases and Primo, wherein 67 relevant articles were identified and 2 additional guidelines were read. Sudden unexpected death in epilepsy and sudden cardiac events were well-established causes of death with risk factors such as male predominance, substance use and a familial history identified. It was acknowledged that while the cause of death is established following post-mortem examination in many cases, some remain non-specific or undetermined. Considering the genetic etiology, these cases would be ideal candidates for molecular autopsies in the future. Thus, this review emphasized the significance of acquiring the relevant information to aid in resolving cause of death of these SUDY cases and subsequently highlighted the potential for further studies on risk factors and the value of molecular autopsies.

Massively Parallel Sequencing of 43 Arrhythmia Genes in a Selected SUDI Cohort from Cape Town

Sudden unexpected death in infants (SUDI) is a devastating event, and unfortunately occurs frequently in developing countries. The emerging molecular autopsy has added value to post-mortem investigations, where genetic variants were able to explain the unexpected demise. Many of these variants have been found in genes involved in arrythmia pathways. The aim of this study was to sequence 43 genes previously associated with cardiac arrhythmia in a selected cohort of SUDI cases (n = 19) in South Africa. A total of 335 variants were found among the 19 infants, of which four were novel. The variants were classified as “likely pathogenic” (n = 1), “variant of unknown significance” (n = 54), “likely benign” (n = 56) or “benign” (n = 224). The likely pathogenic variant was LMNA NM_170707.2:c.1279C > T (p.Arg427Cys) and was found in a 3-week-old male infant of African ancestry. Variants in LMNA have previously been associated with dilated cardiomyopathy, with a typical age of onset in adulthood; therefore, this may be the first report in an infant. The yield of pathogenic or likely pathogenic variants in the classic genes typically associated with channelopathies and sudden death, was less in this study compared with other settings. This finding highlights the importance of population-specific research to develop a molecular autopsy which is locally relevant.

Evaluation of direct PCR for routine DNA profiling of non-decomposed deceased individuals

Forensic DNA profiling is a standard method used in the attempt to identify deceased individuals. In routine investigations, and if available, the preferred sample type is usually blood. However, this requires the invasive re-opening of the body, days or weeks after the autopsy, which is undesirable in resource-constrained mortuary settings. Motivated by the ease of sampling as well as reduced health and safety risks, this study aimed to establish the success rate of generating a full DNA profile on first attempt from buccal swab lysates using a direct PCR approach. Buccal swab samples were collected from 100 unidentified deceased males, and were subjected to direct DNA profiling with use of the Promega PowerPlex® Y23 Kit. At the time of sample collection, these individuals had been stored for between 1 and 887 days. This study shows that full DNA profiles were initially obtained from 73% of samples, which constitutes the first empirical data pertaining to first time success rates of direct PCR from post-mortem buccal lysates. Further investigation of partial and failed DNA profiles using real-time PCR showed that samples did not contain PCR inhibitors, DNA was not degraded, but DNA concentration was particularly low. Repeating DNA profiling with increased lysate input and extra PCR cycles yielded an additional six full DNA profiles, resulting in an overall success rate of 79%. Overall, DNA profile success rate was not associated with the duration of storage (p = 0.387). Lastly, massively parallel sequencing with the ForenSeq™ Signature DNA Prep kit provided more informative profiles for three additional samples. These results indicate that blood should therefore remain the sample of choice in a post-mortem setting, yet buccal lysates hold potential to be optimised further, which may ease the human identification workflow.

A first for forensic genetics in Africa: successful identification of skeletal remains from the marine environment using massively parallel sequencing

In unrelated circumstances, two young adult males allegedly went missing off the coast of Cape Town, South Africa, within two months of each other. Weeks after the second disappearance, a decomposed human lower limb was recovered from a beach in Cape Town, followed by a washed-up decomposed hand three days later. An item of female clothing was found with the remains, and preliminary analysis of the skeleton indicated a female, leading to confusion regarding the possible identity of the decedent. Consequently, DNA analyses were requested to determine the biological sex of the remains, and whether the two sets of remains originated from the same individual. Various samples were collected, including bone, nails and swabs of soft tissue. DNA quantity and quality varied between sample types, with better results obtained from metacarpal bone and swab lysates. DNA profiling revealed a male sex, which suggests cognitive bias may have played a role in initial sex estimations. In addition, massively parallel sequencing confidently matched the two sets of remains (random match probability: 1 in 2.70 x 10³¹). These results were a first for Africa where massively parallel sequencing was successfully used and assisted in the identification of human remains, thus, affording closure to the next-of-kin. Moreover, this constitutes the first global report where soft tissue lysates from a marine decomposition case yielded full DNA profiles with a massively parallel sequencing approach.

Assessment of candidate variants causative of inborn metabolic diseases in SUDI cases in South Africa, and a case report

Sudden unexpected death in infants (SUDI) is a devastating event, and unfortunately is still a burden in many parts of the world, including in South Africa. Due to the absence of routine testing for inborn metabolic diseases in newborns and in a post-mortem context, little is known about the presence of metabolic diseases in local SUDI cases. The aim of this study was to genotype five candidate variants previously associated with metabolic disorders in a cohort of SUDI cases (n = 169) from Salt River Mortuary, Cape Town. DNA was isolated from blood, and SNaPshot® PCR and Sanger sequencing were used to genotype the following variants: ACADM: c.583G > A, ACADM: c.985A > G, GCDH: c.877G > A/T, GALT: c.404C > G/T and GALT: c.563A > G. Four carriers of GCDH: c.877G > A/T were identified, while one infant was homozygous for the founder mutation GALT: c.404C > G/T; the latter which is causative of galactosaemia and was previously undiagnosed. During the follow-up with the family, it emerged that the affected infant's identical twin had subsequently demised. The findings in this study highlight possible new candidate variants to assess in South African SUDI cases, and these results directly contribute to the development of a molecular autopsy which is locally relevant. It is evident that until newborn screening becomes routine and accessible in South Africa, molecular autopsies should include testing for inherited metabolic disorders, as it holds potential to save lives.

DNA barcoding of forensically important flies in the Western Cape, South Africa

Forensic entomology aids the determination of post mortem interval based on arthropods associated with a deceased body. This relies on the accurate identification of insects that visit the body, particularly first colonisers such as Calliphoridae (Diptera). Traditional species identification though morphological keys can be challenging as immature or closely related specimens can look similar. Some of these challenges have been overcome through “DNA barcoding”, which involves the sequencing of informative regions within a species’ DNA and comparison to a database of reference sequences. However, reference DNA sequences of blow fly species in South Africa is currently limited. In this study, adult blow flies representing four species common to the Western Cape, South Africa (Chrysomya chloropyga, Chrysomya albiceps, Chrysomya marginalis, Lucilia sericata) were examined using morphological keys and DNA barcoding of two regions: COI and ITS2. These DNA sequences were then used as references for the successful identification of seven unknown immature specimens. Intraspecific divergence showed a maximum of 0.36% and 2.25% for COI and ITS2, respectively; interspecific divergence showed a minimum of 6.14% and 64.6% for COI and ITS2, respectively. According to these results, COI and ITS2 have sufficient discriminatory power for species-level identification for the four species studied.

A Systematic Review of Molecular Autopsy Studies in Sudden Infant Death Cases

Sudden unexpected death is an upsetting event, which can remain unexplained even after post-mortem investigation. Internationally, molecular autopsies have shown to resolve up to 44% of unexplained cases; however, it is currently unclear how many of these were infants. This systematic literature review showed that significantly fewer infant cases were resolved (median: 4%) compared with cohorts of 1 to 45 years old (median: 32%). Further, no study involving indigenous African participants has yet been published. Overall, molecular autopsies hold immense value to living family members and is motivation to explore new avenues in infant cohorts.

Ethical, legal and social implications of forensic molecular phenotyping in South Africa

Conventional forensic DNA analysis involves a matching principle, which compares DNA profiles from evidential samples to those from reference samples of known origin. In casework, however, the accessibility to a reference sample is not guaranteed which limits the use of DNA as an investigative tool. This has led to the development of phenotype prediction, which uses SNP analysis to estimate the physical appearance of the sample donor. Physical traits, such as eye, hair and skin colour, have been associated with certain alleles within specific genes involved in the melanogenesis pathways. These genetic markers are also associated with ancestry and their trait prediction ability has mainly been assessed in European and North American populations. This has prompted research investigating the discriminatory power of these markers in other populations, especially those exhibiting admixture. South Africa is well known for its diversity, and the viability of these particular SNPs still needs to be assessed within this population. South African law currently restricts the use of DNA for molecular phenotyping, and there are also numerous ethical and social considerations, all of which are discussed.

A review of the optimisation of the use of formalin fixed paraffin embedded tissue for molecular analysis in a forensic post-mortem setting

Molecular analyses in a post-mortem setting are becoming increasingly common, particularly in cases of sudden unexplained death, with the aim of identifying genetic mutations which may be responsible for causing death. In retrospective investigations, the access to suitable autopsy biological samples may be limited, and often formalin fixed paraffin embedded (FFPE) tissue is the only sample available. The preservation of tissue in formalin is known to damage DNA through crosslinking activity. This results in the extraction of severely fragmented DNA of variable yields, which subsequently reduces the ability to perform downstream molecular analyses. Numerous studies have investigated possible improvements to various aspects of the DNA extraction and amplification procedures from FFPE tissue and this review aims to collate these optimization steps in a cohesive manner. A systematic review was performed of three major databases, which identified 111 articles meeting the inclusion criteria. Five main areas for optimization and improvements were identified in the workflow: (1) tissue type, (2) fixation process, (3) post-fixation, (4) DNA extraction procedure and (5) amplification. It was found that some factors identified, for example tissue type and fixation process, could not be controlled by the researcher when conducting retrospective analyses. For this reason, optimization should be performed in other areas, within the financial means of the laboratories, and in accordance with the purposes of the investigation. Implementation of one or more of the optimization measures described here is anticipated to assist in the extraction of higher quality DNA. Despite the challenges posed by FFPE tissue, it remains a valuable source of DNA in retrospective molecular forensic investigations.