Genetic Counseling and Genome Sequencing in Pediatric Rare Disease

Abstract

Both genome sequencing (GS) and exome sequencing (ES) have proven to be revolutionary in the diagnosis of pediatric rare disease. The diagnostic potential and increasing affordability make GS and ES more accessible as a routine clinical test in some centers. Herein, I review aspects of rare disease in pediatrics associated with the use of genomic technologies with an emphasis on the benefits and limitations of both ES and GS, complexities of variant classification, and the importance of genetic counseling. Indications for testing, the role of genetic counselors in genomic test selection, and the diagnostic potential of ES and GS in various pediatric multisystem disorders are discussed. The neonatal population represents an important cohort in pediatric rare disease. Rapid ES and GS in critically ill neonates can have an immediate impact on medical management and present unique genetic counseling challenges. This work includes reviews of recommendations for genetic counseling for families considering genome-wide sequencing, and issues of access to genetic counseling that affect clinical use and will necessitate implementation of innovative methods such as online decision aids. Finally, this work will also review the challenges of having a child with a rare disease, the impact of results from ES and GS on these families, and the role of various support agencies.

RARE DISEASE—DEFINITION AND FREQUENCY

In Europe, a rare disease is defined as one that affects less than one in 2000 individuals (European Organisation for Rare Diseases 2005). In the United States, the Rare Diseases Act defines a rare disease to be one that affects fewer than one in 200,000 individuals (H.R.4013 — 107th Congress 2002). Others consider a frequency of less than one in 2,000,000 to represent an “ultrarare” disease (Hennekam 2011). Although, individually, each disease is characterized as “rare,” collectively, rare diseases affect 4%–8% of the general population (Baird et al. 1988; Boycott et al. 2017). There are ∼7000 rare diseases, and ∼80% are thought to have a genetic basis (Amberger et al. 2009, 2015, 2019). The majority affect the pediatric population, and it is estimated that 30% of affected children do not survive beyond 5 years of age (European Organisation for Rare Diseases 2005). Rare diseases are generally multisystem and complex in nature. Because of their rarity, these disorders are generally not familiar to most clinicians. As a result, they represent significant diagnostic challenges.

Genetic heterogeneity (allelic and locus), variable clinical presentation (expressivity), incomplete penetrance, genetic modifiers, and environmental factors can further complicate the ability to obtain a specific diagnosis in individuals with rare disorders (as reviewed in Wright et al. 2018a). Furthermore, extremely rare or ultra-rare diagnoses have been even less well-characterized from a genetic and phenotypic standpoint.

Genetic conditions and malformations are the leading causes of infant deaths in the neonatal intensive care unit (NICU) (Weiner et al. 2011). Identification of rare disorders in this setting is uniquely challenging for a number of reasons: The complete clinical phenotype may not have evolved, clinical signs can be compounded by prematurity, and underlying genetic heterogeneity may delay diagnosis through a candidate gene approach or serial multiple gene panel testing. As a result, many patients with genetic disorders are discharged or die before the diagnosis has been established (Elliott et al. 2019).

THE CHANGING LANDSCAPE OF GENETIC TESTING

Serial investigations of single genes are time-consuming, expensive, and not always available. In the case of intellectual disability, more than 700 genes have been implicated (Grozeva et al. 2015; Kochinke et al. 2016), making serial single-gene investigations impractical. Consequently multigene panels are often used, but typically lack a complete comprehensive inclusion of all potentially relevant genes and variants. Chromosomal microarray analysis (CMA) allows for interrogation of unbalanced structural changes and copy number variants and, with greater resolution than cytogenetic analysis (50–100 kb vs. 5–7 Mb), it is instrumental in the detection of chromosomal microdeletion syndromes. However, it has been estimated that only ∼10% of patients with a rare pediatric disorder are diagnosed when using array-based techniques (Sagoo et al. 2009; Clark et al. 2018). Thus, technologies like genome sequencing/exome sequencing (GS/ES) are increasingly being used (as reviewed in Wright et al. 2018). These are more comprehensive evaluations than panel-based tests or CMA as a result of their unbiased approach to disease gene identification.

GS/ES has allowed for the discovery and further characterization of thousands of variants in Mendelian disorders—many of which represent rare disease in the pediatric population. Approximately 250 new gene-disease and 9200 variant-disease associations are reported each year (Wenger et al. 2017). This constantly evolving landscape not only contributes to rare disease discovery, but also underlies the importance of reanalyzing patients for whom a disorder was not identified on initial GS/ES analysis. Large-scale studies involving individuals with intellectual disability and suspected genetic disease have shown the diagnostic potential of this technology. The diagnostic rate of certain disorders of particular organ systems with ES is listed in Table 1 (Wright et al. 2018).

Table 1.

Diagnostic rate of certain disease classes with exome sequencing (ES)

Osteogenesis imperfecta	100%	Primary immunodeficiency	40%
Ciliary dyskinesia	76%	Limb girdle muscular dystrophy	37%
Epileptic encephalopathy	70%	Early-onset generalized dystonia	37%
Neurometabolic disorders	70%	Severe short stature	36%
Nonsyndromic hearing loss	56%	Inherited bone marrow failure	27%
Retinopathy	56%	Nephrolithiasis and/or nephrocalcinosis	17%
Suspected inborn errors of metabolism	50%	Congenital diaphragmatic hernia	<12%
Inherited thrombocytopenia	46%	Childhood solid tumors	10%
Intellectual disability	42%	Syndromic congenital heart disease	9.7%
Sporadic infantile spasms	40%	Autism spectrum disorder	8.4%

Adapted from Wright et al. 2018.

Many individuals enlisted in GS/ES studies can have more than one diagnosis. At least 4% of patients who enroll in GS/ES testing cohorts have at least two distinct genetic diagnoses (Balci et al. 2017). The disorders can be blended with overlapping phenotypes (e.g., variants resulting in both CHARGE and Kabuki syndromes) or distinct phenotypes (e.g., variants resulting in both neurofibromatosis and epileptic encephalopathy). The rate of dual diagnoses is even greater (14%) in cohorts of selected phenotypes (e.g., inclusion requirements of two or more conditions such as metabolic disease and intellectual disability) (Tarailo-Graovac et al. 2016).

From a genetic counseling perspective, the introduction of this technology is accompanied by a need for a shift from traditional single-gene counseling (e.g., for cystic fibrosis) to “genomic counseling” with its unique, inherent issues, which include variants of uncertain significance (VUSs) and incidental (secondary) findings (Ormond 2013; Patch and Middleton 2018).

EXOME VERSUS GENOME APPROACHES

Although GS and ES use overlapping advances in DNA-sequencing technology, there are distinct differences. ES only captures and reports on the exonic or protein-coding sequences, which represent <2% of the genome. GS attempts to capture and report on the entirety of the genome, but because of the challenges in sequencing certain regions, GS examines ∼90% of the genome.

By focusing on only the exonic, more well-characterized regions of the genome, ES offers pragmatic advantages over GS in overall cost savings in reagents and data storage; greater depth of coverage (100× vs. 30×) of reported sequences; and quicker, cheaper, and easier data interpretation (Warr et al. 2015).

However, as a result of the enrichment steps that are necessary for ES, the coverage of the final DNA sequences evaluated is not uniform. This results in reads with “hotspots” represented by an overabundance of coverage and other regions with too little coverage to provide reliable calls. Because GS does not require this up-front enrichment step, it generates much more uniform coverage of the genome, has the advantage of producing longer reads, and is better at capturing the high GC-rich content within exonic sequences. In addition to reporting on intronic sequences, the longer reads also allow for better detection and characterization of copy number variations, rearrangements, and other structural variations. Also, although ES has the ability to provide data for the mitochondrial genome, these results are not reported by many reference laboratories (Posey 2019).

LIMITATIONS OF EXOME AND GENOME SEQUENCING

Both ES and GS have limitations that affect their effectiveness as a diagnostic tool. Neither technology is capable of detecting changes in gene expression because of disorders of imprinting or providing reliable detection of variants in GC-rich regions such as centromeres and telomeres, trinucleotide repeat expansion, regions of highly homologous sequences, and low-level or tissue-specific mosaicism. It is also worth noting that with either ES or GS, the bioinformatic interpretation may rely on familiarity with the causative gene or variant to recognize its significance (Salgado et al. 2016).

INDICATIONS FOR AND APPROACHES TO EXOME AND GENOME SEQUENCING

Different groups have published indications for GS/ES, and a general summary is included here (adapted from Elliott et al. 2018).

The patient has a suspected genetic disorder plus one or more of the following:

• Previous genetic investigations, including CMA, appropriate single-gene or available panel testing, and first-tier biochemical testing for intellectual disability (Van Karnebeek et al. 2014) have not identified the genetic cause.

• The condition shows extensive genetic heterogeneity.

• The family history suggests a Mendelian single-gene disorder (e.g., affected parent and child, unaffected parents and affected child, parental consanguinity, multiple affected siblings).

As with older pediatric patients undergoing GS/ES, suspected neurologic disorders and multiple congenital anomalies are frequent indications for ES/GS studies in the neonatal intensive care setting (Willig et al. 2015; Petrikin et al. 2018; Stark et al. 2018; Elliott et al. 2019). Identifying appropriate patients for GS/ES and other genomic testing is an important clinical and economic consideration for providers. The integration of genetic counselors in the triage process of genomic tests has shown advantages that include a reduction in the misordering of tests, decreased time to diagnosis, cost efficiencies, education of ordering clinicians, and an increased diagnostic rate in ES/GS studies (Miller et al. 2014; Suarez et al. 2017; Dragojlovic et al. 2018; Elliott et al. 2018; Wakefield et al. 2018).

Trio-based (proband plus both biologic parents) approaches to GS/ES in rare pediatric disorders have diagnostic and efficiency advantages, particularly in neonates given the importance of rapid turnaround time for critically ill babies to immediately inform medical management. For example, de novo variants that present only in the child are easily identified, and the phase of variants in imprinted or recessive disorders can be detected. There is an ∼10-fold reduction in the number of candidate variants, as well as a 50% increase in diagnostic yield with the trio approach, and a result can be obtained earlier for families (Wright et al. 2015). For individuals with severe developmental disabilities, a diagnostic rate of 40% can be achieved using trios (Wright et al. 2018b).

GS/ES IN THE NEONATAL POPULATION AND RARE DISEASE

Rapid GS/ES have shown impressive diagnostic capability in the NICU setting, immediately influencing clinical management by enabling precision treatment, including therapeutic intervention or customized palliative care, as well as accurate genetic counseling (Saunders et al. 2012; Willig et al. 2015; Bowdin 2016; Berg et al. 2017; Meng et al. 2017; van Diemen et al. 2017; Petrikin et al. 2018; Stark et al. 2018; Elliott et al. 2019). Rapid GS/ES in the NICU setting can identify the genomic diagnosis in as little as 26 hours (Miller et al. 2015). A groundbreaking 2015 study showed the diagnostic effectiveness of rapid GS in 35 critically ill neonates: 57% diagnosed compared to 9% with conventional genetic testing, with results available in 5 days for some patients (Willig et al. 2015). Other studies have shown diagnostic rates of ∼30%–60% (Meng et al. 2017; Stark et al. 2018; Elliott et al. 2019). Combining ES results with multigene panels and CMA outcomes can result in diagnostic yields of >70% with significant impact on immediate medical management (e.g., customizing anticonvulsant medication) in 83% of patients given a diagnosis (Elliott et al. 2019).

VARIANT CLASSIFICATION AND RECLASSIFICATION

The diagnostic potential of GS/ES is realized only when interpretation of GS/ES data and subsequent classification of the variant(s) are appropriate. Each of us has 4,000,000–5,000,000 nucleotides in our genome that differ from those of the human reference genome, but in most cases only one or two of these variants are responsible for the disease in an individual with Mendelian disorder (Elliott and Friedman 2018). Identifying most of these nonreference variants as benign and classifying most of the remainder as unlikely to be causal with respect to disease is an important component of GS/ES interpretation.

Previously, terms such as “mutation” and “polymorphism” were used to describe nucleotide changes, with the latter describing a change in >1% of the population and, therefore, presumed to be benign. As we accrue data through GS/ES, we have found that this simplified form of classification is not optimally relevant or useful. Furthermore, genetic changes can have different implications and frequencies among different ethnic groups. Underrepresentation of certain populations in reference databases results in suboptimal variant classification and the generation of an increased number of variants of unknown significance.

In 2015, the American College of Medical Genetics and Genomics (ACMG) recommended the replacement of the terms mutation and polymorphism with the term “variant” (Richards et al. 2015) and published guidelines regarding the classifications of the spectrum of pathogenicity. There are five resulting categories: “pathogenic,” “likely pathogenic,” “uncertain significance,” “likely benign,” and “benign.” Corresponding details relevant to scoring, classification, and prediction algorithms (e.g., criteria including segregation with disease, evidence of the variant being de novo [i.e., not inherited], and its effect or predicted effect on the protein) were included.

The ACMG guidelines serve to standardize nomenclature for variant interpretation across laboratories, and outline criteria for classification of variants based on various lines of evidence. Classifying variants according to the guidelines takes into account different lines of evidence of pathogenicity. One of the supporting lines of evidence used is the agreement of multiple in silico tools (representing computational evidence), which predict a deleterious effect of the specific variant on the gene or gene product. An example of a common and preferred annotation tool that assists in the interpretation of a sequence variant is the Combined Annotation Dependent Depletion Score (aka “CADD score”). The CADD score integrates multiple annotations into a single metric and generates a score that is a measure of deleteriousness (pathogenicity of the variant) (Kircher et al. 2014). Typically, a CADD score of 20 means that a variant is among the top 1% of deleterious variants in the human genome, and a CADD score of 30 indicates the variant is in the top 0.1%. In spite of these attempts at standardization, interpretations of pathogenicity can vary among laboratories.

A number of databases are available to assess the pathogenicity of a variant, and include those associated with (1) sequence variation, (2) disease, and (3) population. Population databases may include presumably unaffected individuals (e.g., gnomAD) or genomic data from affected and unaffected individuals (e.g., dbSNP). Examples are shown in Tables 2–4 (Richards et al. 2015).

Table 2.

Sequence databases

NCBI Genome, www.ncbi.nlm.nih.gov/genome	Contains full human genome reference sequences
RefSeqGene, www.ncbi.nlm.nih.gov/refseq/rsg	Contains medically relevant gene reference sequences
MitoMap, www.mitomap.org/MITOMAP/HumanMitoSeq	Revised Cambridge Reference Sequence (rCRS) for human mitochondrial DNA

Adapted from Richards et al. 2015.

Table 3.

Disease databases

The Human Gene Mutation Database (HGMD), www.hgmd.org	Contains genetic variants associated with disease in the literature, not including somatic variants and variants in the mitochondrial genome
The Online Mendelian Inheritance in Man (OMIM), www.ncbi.nlm.nih.gov/omim	Contains content from the published scientific literature. Many of the genes and associated variants in OMIM are considered pathogenic and associated with human disease (the “Clinome”), but some genes not currently associated with human disease are also included (the “NonClinome”), and can serve as potential research candidates
ClinVar, www.ncbi.nlm.nih.gov/clinvar	Unlike HGMD and OMIM, variants do not have to be reported in the literature to be entered into ClinVar
The Database of Genomic Variation and Phenotype in Humans Using Ensembl Resources (DECIPHER/DDD), decipher.sanger.ac.uk/ddd#research-variants	Contains sequence variants and copy number variants with corresponding patient data and phenotypic information

Adapted from Richards et al. 2015.

Table 4.

Population databases

gnomAD, gnomad.broadinstitute.org	Contains 125,748 exomes and 15,708 genomes from unrelated individuals sequenced as part of various disease-specific and population genetic studies, totaling 141,456 individuals. Although individuals with severe pediatric disease have been reported to be removed, some remain
Exome Variant Server, evs.gs.washington.edu/EVS	Database of variants (>6000 samples) found during exome sequencing (ES) of several large cohorts of individuals of European and African American ancestry
1000 Genomes, browser.1000genomes.org; grch37.ensembl.org/index.html	The 1000 Genomes Project ran between 2008 and 2015, creating the largest public catalog of human variation and genotype data. As the project ended, the Data Coordination Centre at the European Molecular Biology Laboratory-European Bioinformatics Institute (EMBL-EBI) received continued funding from the Wellcome Trust to maintain and expand the resource
dbSNP, www.ncbi.nlm.nih.gov/snp	Database of short genetic variations (typically 50 base pairs or less) submitted from many sources. May lack details of originating study and may contain pathogenic variants
dbVar, www.ncbi.nlm.nih.gov/dbvar	Database of structural variation (typically >50 base pairs) submitted from many sources

Adapted from Richards et al. 2015.

LIMITATIONS OF VARIANT CLASSIFICATION

GS/ES provide a higher diagnostic rate if performed in a hospital setting rather than a reference laboratory remote from the patient (Clark et al. 2018). This is likely because the hospital setting provides access to patient health records and consultation with clinicians who have deep knowledge of the patient's family history, clinical presentation, and laboratory and imaging results. This is distinct from the clinical laboratories that classify genomic variants according to the guidelines mentioned, through an algorithmic assessment and limited phenotypic correlation (Elliott and Friedman 2018). Given their access to additional information, clinicians may interpret a variant as causal for the patient's phenotype in spite of the variant being reported as a VUS (Shashi et al. 2016). These discrepancies can result in challenges for the patient and health-care team. Standardization of phenotypic information provided to laboratories can help to reduce these discrepancies (Bowdin et al. 2016).

One of the challenges to interpretation of variants is that certain groups—such as Indigenous populations—are underrepresented in reference databases. These shortcomings (genomic inequities) of reference databases have been acknowledged as critical areas to address in reducing challenges associated with variant interpretation (Morgan et al. 2019).

Another challenge to variant interpretation is that it is a dynamic process and, as new data emerge, variants can be reclassified (e.g., VUS to benign). This has significant implications for medical management and family planning. But who is responsible for informing the family? Recent European guidelines indicate that the duty to recontact family exists for findings with clinical or personal utility, and the best interest of the family must be kept in mind. The process should be a joint effort involving the clinical team, the family, and the laboratory and needs to be sustainable for the health-care system (Carrieri et al. 2019). The shared responsibility indicated in the European guidelines is echoed in the recently published American College of Genetics and Genomics position statement on recontacting patients after revision of genomic results and should be discussed as part of pretest genetic counseling (David et al. 2019). Many individuals undergo GS and ES through research protocols, in part, because genome-wide sequencing is not routinely available clinically. Bombard et al. (2019) recently published a position statement on behalf of the American Society of Human Genetics, which includes 12 recommendations for recontacting research participants after reinterpretation of genetic and genomic testing.

GENETIC COUNSELING AND GS/ES

Genetic counseling is a communication process that is both educational and supportive; it involves helping people understand and adapt to the medical, psychological, and familial aspects of the genetic contribution to disease (Resta et al. 2006).

Informed consent is not the same as genetic counseling. Issues specific to GS/ES and pre- and posttest genetic counseling have been recently addressed (Elliott and Friedman 2018). The complex issues associated with genetic counseling and GS/ES include the challenges connected with a rare disease diagnosis, and the possibility that some patients will be diagnosed with more than one genetic disorder. Pretest genetic counseling should include taking a detailed family history, explanation of the method of testing used, the associated risks and benefits, and the possibility that uncertain, undefined, and difficult to interpret results can be generated. In addition, it is necessary to communicate that results may only explain a portion of the child's phenotype. Incidental findings and potential implications for relatives need to be discussed, including the chance that other individuals may need to be tested to determine whether or not a particular variant is carried by all family members with a disease or condition. Privacy issues and concerns regarding insurance need to be addressed. The pretest counseling process should include emotional support for families and help to guide them to make informed decisions that are consistent with their values. Families need to be prepared for the uncertainties that can accompany GS/ES results and that a diagnosis may not be reached. Importantly, families should understand that they can decline GS/ES before providing “informed consent.”

Canadian clinical practice guidelines state that before diagnostic GS/ES, genetic counseling should be provided for the patient/family and documented in the medical record by a qualified individual with a thorough understanding of clinical GS/ES (Boycott et al. 2015). An explanation of what will happen with data, including how long they will be stored and if and when additional analysis or reanalysis will be performed in the future, should be provided. Patients/families should be given the option of having coded or anonymized genome-wide and phenotypic data deposited and stored in an international database to assist in interpretation of genome-wide studies of themselves and other patients and the opportunity to enroll in current or future research studies (Boycott et al. 2015).

European guidelines also indicate that pretest counseling is necessary for families considering GS/ES and should include a discussion on both expected results and the potential for unsolicited and secondary findings (Matthijs et al. 2016). The ACMG recommends GS/ES be accompanied by consultation with a genetics professional and adequate genetic counseling (Green et al. 2013). International recommendations require adequate pretest counseling, interpretation results, and provision of posttest counseling when GS/ES is being considered clinically (Bowdin et al. 2016).

GUIDELINES REGARDING INCIDENTAL AND SECONDARY FINDINGS

Determining the patient's preference for receiving incidental and secondary findings (IFs/SFs) is an important component of pretest genetic counseling for GS/ES and a key recommendation of multiple guidelines (Knoppers et al. 2015; Williams et al. 2015). American diagnostic laboratories provide patients undergoing GS/ES with the choice to opt in or out of the reporting of SFs. For individuals who opt in, the patient's sample is interrogated independently for the 59 genes identified by the working group (Green et al. 2013). Furthermore, it is recommended that the seeking and reporting of SFs to ordering clinicians not be limited by the age of the person being sequenced.

Unlike diagnostic laboratories in the United States, the handling of SFs in the Canadian context follows the Canadian College of Medical Genetics guidelines for research studies “…until the benefits of reporting incidental findings are established, we do not endorse the intentional clinical analysis of disease-associated genes other than those linked to the primary indication; clinicians should provide genetic counseling and obtain informed consent prior to undertaking clinical genome-wide sequencing. Counseling should include discussion of the limitations of testing, likelihood and implications of diagnosis and incidental findings…” (Boycott et al. 2015). The Canadian approach is similar to European guidelines (Matthijs et al. 2016). For adults who opt in, pathogenic, medically actionable adult-onset variants are disclosed. For children, pathogenic, pediatric-onset medically actionable variants are automatically disclosed and adult-onset variants are not disclosed (Zawati et al. 2014; Boycott et al. 2015). SFs/IFs are an inevitable consequence of GS/ES and an important genetic counseling issue to discuss with families considering GS/ES.

Access to genetic counseling is a necessity for families considering GS/ES, and this requirement puts increased pressure on an already strained resource. This will be compounded once GS/ES becomes routinely available clinically. Consequently, innovative methods (e.g., videoconferencing, telehealth, group counseling, telephone counseling, and online decision aids) for the provision of genetic counseling services for GS/ES must be considered.

Given their inherent scalability, online decision aids are a logical solution to support the increasing demand for GS/ES. Decision aids have been shown to be effective when users are faced with complex and difficult treatment or screening decisions (Sheehan and Sherman 2012), including genetic screening and single-gene testing (Wakefield et al. 2008; Kuppermann et al. 2009; Yee et al. 2014). Decision aids have been proposed, but not yet thoroughly evaluated, in applications as complex as prenatal GS/ES and newborn screening (Lewis et al. 2016; Chen and Wasserman 2017). Compared to usual care, decision aid users are better informed, and have shown improved knowledge and more accurate understanding of the pros and cons of their options. Several studies suggest that users typically participate to a greater extent in the decision, feel clearer about what matters to them, and make choices that are more congruent with their values and respectful of patient autonomy (Johnston et al. 2017; Stacey et al. 2017).

The effectiveness and cost efficiency of decision aids compared to in-person genetic counseling for GS/ES have yet to be established from the perspective of either the patient-user or the health-care system. Decision aids for GS/ES have been introduced in some clinical settings. “DECIDE” (decision aid and e-counseling for inherited disorder evaluation) is designed to streamline, enhance, and improve the accessibility of genetic counseling for clinical GS/ES (Birch et al. 2016). DECIDE provides educational material in a variety of formats, tailored to users’ needs, and presents pros and cons of GS/ES and IFs for users to evaluate in the context of their own values. In a recent study of families undergoing GS/ES for pediatric intractable epilepsy, a noninferiority trial comparing DECIDE to conventional in-person genetic counseling revealed a significant increase in parental knowledge in both groups, with no difference between the two groups (Adam et al. 2019).

The genomics “ADvISER” is a patient-centered support tool that is specific to the selection of incidental sequencing results. Usability testing showed this tool to be effective for delivering genomic information, was acceptable to patients, and was sufficient for them to make an informed hypothetical decision (Bombard et al. 2018).

GS/ES AND THE DIAGNOSTIC ODYSSEY

Genetic counselors represent the front line of genomic medicine and are instrumental in guiding and supporting patients and their families throughout their clinical journey, often well beyond the delivery of diagnostic results.

Pediatric patients with rare diseases often have complex, multisystem involvement and require the care of numerous health-care providers from diverse subspecialties. Adherence to medical management recommendations is an important component of the care trajectory for these families. The inclusion of a genetic counselor in initial pediatric visits in genetics has been shown to significantly increase patient adherence and can be considered a metric by which genetic counseling is assessed. A chart review of 198 pediatric patients seen for their initial appointment in clinical genetics revealed that appointments including a genetic counselor were associated with significantly increased adherence across three domains (follow-up with genetics, referral to specialist, and testing) as compared to appointments in which there was no genetic counselor involved in the initial visit (Rutherford et al. 2014).

The increased uncertainty experienced by parents of a child with an undiagnosed disorder has been associated with lower levels of optimism and feelings of control, in addition to increased perception of severity (O'Daniel et al. 2010). Isolation, frustration, and hopelessness have been reported in families with rare disorders (Zurynski et al. 2008; Helm 2015; Baumbusch et al. 2019). Uncertainty, fear, and loss of control were shown in another study of parents of children with undiagnosed disorders (Spillmann et al. 2017). Families dealing with rare diseases experience challenges that affect their psychological well-being and include financial burden, lack of information, lack of access to appropriate care, delays in diagnosis, and increased social isolation and uncertainty (Zurynski et al. 2008; Kole and Faurisson 2009; Baumbusch et al. 2019), whereas, for some parents of undiagnosed children, the uncertainty associated with their child's disorder can be associated with parental social integration, a component of adaptation (Yanes et al. 2017).

ES and GS have helped to end the “diagnostic odyssey” for thousands of pediatric patients with suspected genetic disease. A diagnosis has clinical, social, and economic benefits. These include optimization of patient management (and in some cases, customizing treatment), an improved understanding of prognosis and surveillance, and a reduction in unnecessary testing. For families, an answer can provide access to social programs, allied health-care services, specialized educational programs, and relevant support groups (Strande and Berg 2016; Boycott et al. 2017). From a genetic counseling perspective, a diagnosis allows for informed family planning, more accurate recurrence risks, and the option of prenatal diagnosis or preimplantation diagnosis for interested parents. Additionally, it allows genetic counselors to characterize genetic risk for other family members.

Diagnostic sessions have been shown to be correlated with a positive parental experience when a genetic counselor is present, likely attributable to the specialized training, emotional support, and resources provided (Waxler et al. 2013). Using semistructured in-person interviews with families whose children received a diagnosis, emotional support, counseling, providing hope, and perspective and explaining follow-up were associated with more positive experiences (Ashtiani et al. 2014). The importance of families being “prepared” was emphasized, which is particularly relevant in the space of genetic counseling for ES/GS. Genetic counselors taking an active and defined role in diagnosis sessions can result in a more positive experience for families (Waxler et al. 2013; Ashtiani et al. 2014).

Education is an important component of pre- and posttest genetic counseling, and ES/GS is a complex topic to navigate. Parents of pediatric patients who underwent clinical ES were surveyed to assess perceived and actual understanding (Tolusso et al. 2017). Parents, in general, had a good understanding (actual and perceived), but areas for improved understanding included how genes are analyzed and the lack of protection with respect to life insurance discrimination. There was also low actual understanding related to certain aspects of SFs. The importance of providers to explain to families at the pretest counseling session that ES may not find a diagnosis was a common theme emphasized by multiple respondents independently.

In a study surveying 192 parents whose children had diagnostic ES, the parents’ interpretation of the child's result agreed with the clinicians’ interpretation in 79% of cases. There was more frequent discordance when the clinician's interpretation was uncertain. Most parents (79%) reported no regret with respect to their decision to pursue ES, and most (65%) reported complete satisfaction with the genetic counseling encounter. Satisfaction was positively correlated with their genetic counselor's number of years of clinical experience (Wynn et al. 2018).

Rosell and colleagues examined parental perceptions of ES in pediatric disorders that were previously diagnosed. Some families experienced frustration and isolation because of the limited information available about the rare disorder. Parents wanted more information and hoped to identify with other families with the disorder (Rosell et al. 2016). A recent meta-analysis of psychological outcomes for individuals who underwent GS/ES for a variety of indications (including a pediatric cohort) found that there were no significant psychological harms from the return of GS/ES results across multiple clinical settings (Robinson et al. 2019).

GS/ES has the capacity to identify ultra-rare disorders and is a constant source of gene discovery in the pediatric rare disease space. A recent study of parents whose children had been diagnosed with “new” genetic conditions revealed that, in spite of limited information about the child's disorder, most parents experienced relief and perceived value in the diagnosis (having an explanation for the cause of the condition). Some families reported a fear that their child would develop cancer after parental internet searches turned up literature mentioning changes in the gene in cancer cell lines (e.g., somatic, not germline) (Inglese et al. 2019). Other investigators have indicated that a parent's desire to obtain an answer is a strong motivator for pursuing GS/ES (Sapp et al. 2014; Rosell et al. 2016; Smith et al. 2019).

Informational and emotional support are essential benefits of peer support for children with a rare disease (Baumbusch et al. 2019). Various investigators have described the benefits to families of connecting with other families who have children with the same disorder (Krabbenborg et al. 2016; Rosell et al. 2016; Inglese et al. 2019). Social networking sites are helpful when a support group specific to a child's disorder has not been established. Some resources for families who have a child with a rare genetic disorder are listed in Table 5.

Table 5.

Resources for families with rare disorders

Organization name	Website
Rare Diseases Europe (EURODIS)	www.eurordis.org
Orphanet	www.orpha.net
National Organization for Rare Disorders (NORD)	rarediseases.org
Canadian Organization for Rare Disorders (CORD)	www.raredisorders.ca
Rare Disease Foundation (RDF)	rarediseasefoundation.org
Unique	www.rarechromo.org
RareShare	rareshare.org
MyGene2	mygene2.org/MyGene2
Facebook	www.facebook.com

Parents who have children with a suspected genetic disease considering GS/ES should be informed during pretest genetic counseling that, even if a diagnosis is found, there may not be many resources or information specific to the disorder. It is important that genetic counselors and other health-care providers identify relevant support groups for families once a diagnosis is established. If a relevant support group does not exist, encouraging linkages through portals such as MyGene2, or establishing an independent group through Facebook are options for interested families.

GENETIC COUNSELING ISSUES AND GS/ES IN THE NEONATAL INTENSIVE CARE UNIT

The importance of availability of genetic counseling for rapid GS/ES has previously been addressed (Stark et al. 2018; Smith et al. 2019). There are challenges to enrollment of parents in genomic sequencing studies of newborns. Parents declining to participate have cited such reasons as feeling overwhelmed, logistical issues, lack of interest in research involving genetic testing, and unavailability of both parents for trio-based testing (Willig et al. 2015; Petrikin et al. 2018; Genetti et al. 2019).

Maximizing communication between the clinical and study teams, in addition to flexibility in the genetic counselor's availability, can ensure appropriate pretest genetic counseling occurs. This may require counseling parents separately to accommodate family schedules (Elliott et al. 2019).

Parents of patients in the NICU experience increased stress and depression (Chourasia et al. 2013; Alkozei et al. 2014; Turner et al. 2015). Lack of enrollment in neonatal GS/ES studies indicates different concerns and issues in this group than in families of older children with suspected genetic disorders. A recent study comparing parents of neonates and parents of older pediatric children (average age of 10 years) undergoing GS/ES revealed that after pretest genetic counseling, parents of the NICU patients were significantly less likely to opt in to find out incidental findings for themselves, significantly more likely to identify “diagnosis” as their primary motivation for pursuing GS/ES, and significantly less likely to identify “no concerns” than the parents of older children. Parents in both cohorts had increased anxiety and depression relative to the general population (Smith et al. 2019).

CONCLUSION

Families of children with rare disease can benefit from the tremendous diagnostic potential of GS/ES. However, GS/ES should be delivered as a holistic service that includes pre- and posttest genetic counseling. The eventual clinical implementation of GS/ES results in the need for innovative methods to deliver genetic counseling services. Genetic counselors represent the front line of genomic medicine and are integral to ensuring families receive appropriate support as new genetic disorders continue to be discovered and characterized.