Common aneuploidy testing

Aneuploidy describes additional or missing chromosomes, resulting in a chromosome complement other than 46,XX or 46,XY. Aneuploidy may occur in all cells of the body or only in certain cell lines (mosaicism). The most common aneuploid conditions are Down syndrome (trisomy 21), Edward syndrome (trisomy 18), Patau syndrome (trisomy 13) and sex chromosome aneuploidies such as Turner syndrome (45,X) as in clinical scenario 1. These conditions can be identified through QF-PCR: specific markers on chromosomes 13, 18, 21, X and Y are labelled, amplified and quantified, indicating the number of chromosomal copies in the patient’s sample. QF-PCR is also used in the prenatal setting to diagnose common aneuploidy via amniocentesis or chorionic villus sampling. QF-PCR is relatively cheap and results are usually available within a few days. The limited scope of the test means further testing may be needed.

Karyotyping

Karyotyping is the detailed assessment of chromosome number, size and structure under light microscopy by a cytogeneticist. A karyotype is part of the work-up for babies with ambiguous genitalia and is a follow-up investigation to common aneuploidy testing where low-level mosaicism is suspected or when structural information about the chromosomes is needed. For example, karyotyping identifies uncommon structural causes of aneuploidy such as Robertsonian translocations, which may have a high chance of recurrence in a future pregnancy. Karyotyping is relatively cheap, but for routine indications, it can take several weeks to receive a result.

Microarray

Copy number variations (CNVs) are deletions, duplications or insertions of DNA segments, which may cause a range of congenital anomalies. Microarray analysis screens for CNVs across the genome. Array comparative genomic hybridisation (array CGH) involves adding fluorescently labelled patient and control DNA samples to a slide with thousands of short DNA sequences (probes) from across the genome. The patient and control samples competitively bind to probes: an abnormal ratio of patient to control DNA across consecutive probe regions indicates a CNV at that genomic location. An alternative approach is single-nucleotide polymorphism (SNP) array. An SNP is a nucleotide, where ≥1% of the population has variation; there are millions of SNPs across the genome. Fluorescently labelled patient DNA binds to probes specific to different alleles at known SNPs. Analysis of SNP array data identifies CNVs and can also reveal loss of heterozygosity. This may uncover information about biological relationships, including misattributed parentage, consanguinity or incest.

Alongside common aneuploidy testing, microarray is a first-line test for newborns with multiple congenital anomalies or a suspected CNV, for example, Williams syndrome (7q11.23 microdeletion). Microarray can be performed prenatally if anomalies are detected on ultrasound. A microarray is a very accurate, high-resolution method to identify CNVs as small as 50 kilobases. Microarray analysis is relatively cheap, and when sent urgently, results are usually available within weeks. A microarray does not give structural information about chromosomes, so follow-up karyotyping or other tests such as fluorescence in situ hybridisation may be needed.

Targeted testing for specific indications

When there is clinical suspicion of a particular condition/group of conditions, a targeted testing strategy may be the best approach. While complex cases benefit from discussion with and assessment by a clinical geneticist, with the mainstreaming of genomic testing, treating teams can increasingly order testing directly. The goal is to improve access to testing and reduce time to diagnosis.8 In the UK, the National Genomic Test Directory details appropriate tests and eligibility criteria for a wide range of indications.9

For conditions with well-established single gene causes such as cystic fibrosis and achondroplasia, the gene in question can be sequenced or known pathogenic variants tested for using a range of approaches. The original sequencing technology is Sanger sequencing, a highly accurate way of determining the nucleotide sequence in a gene, which is then compared with a reference sample to identify pathogenic variants in the patient’s sample. For conditions or phenotypes which may have hundreds or thousands of known associated genes, such as infantile hypotonia or respiratory ciliopathies as in clinical scenario 2, panel testing can be carried out. Historically, all genes in a panel were individually sequenced but now, it is more common for a ‘virtual panel’ to be applied to data generated from whole genome sequencing (WGS) (see the Whole exome and whole genome sequencing section).

Beyond sequencing approaches, other targeted tests include methylation testing for imprinting disorders such as Angelman syndrome and short tandem repeat testing for myotonic dystrophy. In general, if the initial suspected diagnosis is correct, targeted approaches, where available, are an efficient way of reaching a molecular diagnosis. If there is clinical doubt, iterative targeted testing can prolong uncertainty for families, resulting in the so-called ‘diagnostic odyssey’.2

Whole exome and whole genome sequencing

Next-generation sequencing technology allows millions of DNA fragments to be sequenced in parallel, vastly reducing the time and cost of sequencing. Whole exome sequencing (WES) involves sequencing the 1–2% of the genome which is protein-coding (the exome), estimated to contain 85% of disease-causing variants.10 WGS sequences the entire genome, including the non-coding DNA, which has important regulatory functions. For both WES and WGS, variants in the patient sample are identified by comparison with a reference genome and analysed to determine their likelihood of being pathogenic, using the international American College of Medical Genetics and Genomics criteria.11 In the neonatal setting, a trio approach is often used where the baby’s DNA is sequenced alongside their biological parents. This enables the detection of de novo (new) variants in a baby. A de novo variant is more likely to be pathogenic than an autosomal inherited variant, in a baby whose parents do not share the phenotype.

WES or WGS may be performed if initial tests have not yielded a diagnosis or where targeted tests are not available. In the Neonatal Intensive Care Unite (NICU) setting, rapid trio WGS can be performed as a first-line test for critically ill babies in whom a genetic diagnosis could change immediate clinical management, with results available within just 2–3 weeks compared with current time spans of up to a year for non-urgent WGS. The landmark UK 100 000 Genomes Project resulted in a diagnosis for 25% of probands in the pilot analysis of 4660 participants.12 WGS in the NICU setting may lead to higher diagnostic rates of up to 50%.4

The downside of such broad analysis is the frequent identification of variants of uncertain significance without sufficient evidence to be classified as pathogenic or benign, leading to uncertainty for families. These approaches may also identify unexpected findings with health implications, for example, the identification of an adult-onset cancer-susceptibility variant when investigating congenital anomalies in a baby. As with SNP array, trio testing may reveal unknown or concealed information about biological relationships.

As WES generates less data than WGS, it is faster and cheaper.13 Although the non-coding DNA sequenced in WGS may contain pathogenic variants, it is currently difficult to interpret and, therefore, usually excluded from routine analysis.13 Although WES and WGS are highly accurate at identifying sequence changes, other test types may be needed to identify CNVs, structural changes, methylation and repeat expansion disorders.

At present, population newborn screening is limited to a relatively small number of treatable genetic conditions with international variation in screening approaches.14 In England, WGS is being explored as a universal screening approach for newborns to identify treatable, childhood-onset genetic conditions, however potential risks include false positives, leading to unnecessary testing and anxiety for families.15

Family perspectives and bioethical considerations

Pretest counselling facilitates informed consent for genomic testing, which acknowledges potential test outcomes, wider implications and potential limitations (see box 2). In the NICU setting, parents may be sleep-deprived or worried about an unwell infant, which can be barriers to meaningful, informed consent. Families can be supported through a sensitive approach, repeated conversations and measures to support equity in access to testing (see box 3). Genetic Alliance UK’s 2022 ‘good diagnosis’ report identified priorities such as the diagnosis being timely, accurate and collaborative, acknowledging the psychological needs of the whole family.16

Studies of family experiences with genomic testing have found positive overall perceptions of utility. In a US study of parents whose babies had received rapid genomic testing in NICU, 97% of parents felt it was ‘useful’ or ‘somewhat useful’ despite only 23% of infants receiving a diagnosis.17 Families reported that even a ‘negative’ result helped them feel more knowledgeable about their child’s health and better informed for future reproductive planning. It should be noted that for conditions which are life limiting or result in significant disability, receiving a diagnosis within the first days of the baby’s life may be very traumatic for the parents and disrupt early parent–child bonding.18 Alongside referral to Clinical Genetics and appropriate specialist teams, there are many sources of information and support for families following a genetic diagnosis.