Reference No. 1

[X] MedicallyNecessary

[ ] Investigational

Population Reference No. 2

Whole Exome Sequencing for Children with a Suspected Genetic Disorder other than Multiple Congenital Anomalies or a Neurodevelopmental Disorder of Unknown Etiology Following Standard Workup; Individuals who are not Critically Ill

Clinical Context and Test Purpose

Most of the literature on WES is on neurodevelopmental disorders in children; however, other potential indications for WES have been reported (Table 2). These include limb-girdle muscular dystrophy, inherited retinal disease, and other disorders including mitochondrial, endocrine, and immunologic disorders.

The purpose of WES in patients who have a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following a standard workup is to establish a molecular diagnosis. The criteria under which diagnostic testing for a genetic or heritable disorder may be considered clinically useful are stated above.

The following PICO was used to select literature to inform this review.

Populations

The relevant population of interest is children presenting with a disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder that is suspected to have a genetic basis but is not explained by a standard clinical workup.

Intervention

The relevant intervention of interest is WES.

Comparators

The following practice is currently being used to diagnose a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder: a standard clinical workup without WES. A standard clinical workup for an individual with a suspected genetic condition varies by patient phenotype but generally involves a thorough history, physical exam (including dysmorphology and neurodevelopmental assessment, if applicable), routine laboratory testing, and imaging. If the results suggest a specific genetic syndrome, then established diagnostic methods relevant for that syndrome would be used.

Outcomes

There is no reference standard for the diagnosis of patients who have exhausted alternative testing strategies, therefore diagnostic yield will be the clinical validity outcome of interest.

Study Selection Criteria

For the evaluation of clinical validity of WES, studies that met the following eligibility criteria were considered:

Reported on the diagnostic yield or performance characteristics such as sensitivity and specificity of WES;
Patient/sample clinical characteristics were described;
Patient/sample selection criteria were described;
Included at least 20 patients.

Clinically Valid

A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Review of Evidence

Studies have assessed WES for a broad spectrum of disorders. The diagnostic yield in patient populations restricted to specific phenotypes ranges from 3% for colorectal cancer to 60% for unexplained limb-girdle muscular dystrophy (Table 2). Some studies used a virtual gene panel that is restricted to genes associated with the phenotype, while others have examined the whole exome, either initially or sequentially. An advantage of WES over individual gene or gene panel testing is that the stored data allows reanalysis as new genes are linked to the patient phenotype. Whole exome sequencing has also been reported to be beneficial in patients with atypical presentations.

Table 2. Diagnostic Yields of Whole Exome Sequencing for Conditions Other Than Multiple Congenital Anomalies or a Neurodevelopmental Disorder

Study	Patient Population	N	Design	Yield, n (%)	Additional Information
Kwong et al (2021)^21,	Patients with pediatric‑onset movement disorders and unrevealing etiologies	31	Cohort of patients who received WES	10 (32)	8 of 10 patients with a genetic diagnosis had alterations in management decisions
Gileles-Hillel et al (2020)^22,	Patients with symptoms highly suggestive of primary ciliary dyskinesia	48	Prospective WES in patients referred to a single-center	36 (75)	WES established an alternative diagnosis in 4 patients
Kim et al (2020)^23,	Patients with infantile-onset epilepsy who tested negative for epilepsy using a gene panel test	59	Cohort of patients who received WES	+9 (+8%)	WES provided an additional 8% diagnostic yield in addition to the original gene panel
Hauer et al (2018)^24,	Short stature in whom common nongenetic causes had been excluded	200 (mostly children)	Randomly selected from a consecutive series of patients referred for workup; trio testing performed	33 (17)	Standard diagnostic approach yield: 13.6% in the original cohort of 565 WES results had a possible impact on treatment or additional preventive measurements in 31 (16%) families
Rossi et al (2017)^25,	Patients with autism spectrum disorder diagnosis or autistic features referred for WES	163	Selected from 1200 consecutive retrospective samples from a commercial lab	42 (26)	66% of patients already had a clinician-reported autism diagnosis VUS in 12%
Walsh et al (2017)^26,	Peripheral neuropathy in patients ranging from 2 to 68 y	23 children 27 adults	Prospective research study at tertiary pediatric and adult centers	19 (38)	Initial targeted analysis with virtual gene panel, followed by WES
Miller et al (2017)^27,	Craniosynostosis in patients who tested negative on targeted genetic testing	40	Research study of referred patients^a	15 (38)	Altered management and reproductive decision making
Posey et al (2016)^28,	Adults (overlap of 272 patients reported by Yang et al [2014]),^15, includes neurodevelopmental and other phenotypes	486 (53% 18 to 30 y; 47% >30 y)	Review of lab findings in a consecutive retrospective series of adults	85 (18)	Yield in patients 18 to 30 y (24%) vs. those >30 y (10.4%)
Ghaoui et al (2015)^29,	Unexplained limb-girdle muscular dystrophy	60 families	Prospective study of patients identified from a specimen bank	27 (60)	Trio (60% yield) vs. proband only (40% yield)
Valencia et al (2015)^30,	Unexplained disorders: congenital anomalies (30%), neurologic (22%), mitochondrial (25%), endocrine (3%), immunodeficiencies (17%)	40 (<17 y)	Consecutive patients in a single-center	12 (30)	Altered management including genetic counseling and ending diagnostic odyssey VUS in 15 (38%) patients
Wortmann et al (2015)^31,	Suspected mitochondrial disorder	109	Patients referred to a single-center	42 (39)	57% yield in patients with a high suspicion of mitochondrial disorder
Neveling et al (2013)^32,	Unexplained disorders: blindness, deafness, movement disorders, mitochondrial disorders, hereditary cancer	186	Outpatient genetic clinic; post hoc comparison with Sanger sequencing	3% to 52%	WES increased yield vs. Sanger sequencing Highest yield for blindness and deafness

VUS: variant of uncertain significance; WES: whole exome sequencing. ^a Included both WES and whole genome sequencing.

Tables 3 and 4 display notable limitations identified in each study.

Table 3. Study Relevance Limitations

Study	Population^a	Intervention^b	Comparator^c	Outcomes^d	Duration of Follow-Up^e
Kwong et al (2021)^21,
Gileles-Hillel et al (2020)^22,	4. Most patients had high pre-test probability of disease
Kim et al (2020)^23,
Hauer et al (2018)^24,
Rossi et al (2017)^25,	4. Most patients had a clinical diagnosis; only 33% had testing for specific ASD genes before WES
Walsh et al (2017)^26,		3. Proband testing only
Miller et al (2017)^27,
Posey et al (2016)^28,	3. Included highly heterogeneous diseases	3. Proband testing only
Ghaoui et al (2015)^29,
Valencia et al (2015)^30,	3. Included highly heterogeneous diseases	2. Unclear whether WES performed on parents
Wortmann et al (2015)^31,		3. Proband testing only
Neveling et al (2013)^32,	3. Included highly heterogeneous diseases	3. Proband testing only

ASD: autism spectrum disorder; WES: whole exome sequencing.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aPopulation key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.^bIntervention key: 1. Classification thresholds not defined; 2. Version used unclear; 3. Not intervention of interest.^c Comparator key: 1. Classification thresholds not defined; 2. Not compared to credible reference standard; 3. Not compared to other tests in use for same purpose.^d Outcomes key: 1. Study does not directly assess a key health outcome; 2. Evidence chain or decision model not explicated; 3. Key clinical validity outcomes not reported (sensitivity, specificity, and predictive values); 4. Reclassification of diagnostic or risk categories not reported; 5. Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests).^e Follow-Up key: 1. Follow-up duration not sufficient with respect to natural history of disease (true-positives, true-negatives, false-positives, false-negatives cannot be determined).

Table 4. Study Design and Conduct Limitations

Study	Selection^a	Blinding^b	Delivery of Test^c	Selective Reporting^d	Data Completeness^e	Statistical^f
Kwong et al (2021)^21,
Gileles-Hillel et al (2020)^22,
Kim et al (2020)^23,
Hauer et al (2018)^24,
Rossi et al (2017)^25,
Walsh et al (2017)^26,
Miller et al (2017)^27,	2. Selection not random or consecutive
Posey et al (2016)^28,
Ghaoui et al (2015)^29,
Valencia et al (2015)^30,
Wortmann et al (2015)^31,	1,2. Unclear how patients were selected from those eligible
Neveling et al (2013)^32,	1,2. Unclear how patients were selected from those referred

The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aSelection key: 1. Selection not described; 2. Selection not random or consecutive (ie, convenience).^bBlinding key: 1. Not blinded to results of reference or other comparator tests.^c Test Delivery key: 1. Timing of delivery of index or reference test not described; 2. Timing of index and comparator tests not same; 3. Procedure for interpreting tests not described; 4. Expertise of evaluators not described.^d Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.^e Data Completeness key: 1. Inadequate description of indeterminate and missing samples; 2. High number of samples excluded; 3. High loss to follow-up or missing data.^f Statistical key: 1. Confidence intervals and/or p values not reported; 2. Comparison with other tests not reported.

Clinically Useful

Direct Evidence

No RCTs assessing the use of WES to diagnose a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder were identified.

Chain of Evidence

Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility.

A genetic diagnosis for an unexplained disorder can alter management in several ways: such a diagnosis may lead to genetic counseling and ending the diagnostic odyssey, and may affect reproductive decision making.

Because the clinical validity of WES for this indication has not been established, a chain of evidence cannot be constructed.

Section Summary: Whole Exome Sequencing for a Suspected Genetic Disorder Other Than Multiple Congenital Anomalies or a Neurodevelopmental Disorder

There is an increasing number of reports assessing use of WES to identify a molecular basis for disorders other than multiple congenital anomalies or neurodevelopmental disorders. The diagnostic yields in these studies ranged from 3% for colorectal cancer to 60% for trio (parents and child) analysis of limb-girdle muscular dystrophy. Some studies have reported on the use of a virtual gene panel with restricted analysis of disease-associated genes, and the authors noted that WES data allow reanalysis as new genes are linked to the patient phenotype. Overall, a limited number of patients have been studied for any specific disorder, and study of WES in these disorders is at an early stage with uncertainty about changes in patient management.

Summary of Evidence

Reference No. 2

Population Reference No. 3

Repeat Whole Exome Sequencing

Clinical Context and Test Purpose

The purpose of repeat WES, including re-analysis of data from a previous test, in individuals who have previously received WES is to establish a molecular diagnosis.

The criteria under which diagnostic testing for a genetic or heritable disorder may be considered clinically useful are as follows:

A definitive diagnosis cannot be made based on history, physical examination, pedigree analysis, and/or standard diagnostic studies or tests;
The clinical utility of a diagnosis has been established (eg, by demonstrating that a definitive diagnosis will lead to changes in clinical management of the condition, changes in surveillance, or changes in reproductive decision making, and these changes will lead to improved health outcomes); and
Establishing the diagnosis by genetic testing will end the clinical workup for other disorders.

The following PICO was used to select literature to inform this review.

Populations

The relevant population of interest is individuals who have previously received WES.

Intervention

The relevant intervention of interest is repeat WES, including re-analysis of data from a previous test.

Repeat WES is intended to be used after a WES test has been performed without establishing a diagnosis. Repeat testing could lead to a diagnosis in a previously undiagnosed individual as new pathogenic genes or variants are identified or new diagnostic technologies are developed. Additionally, testing strategies might be revised based on the emergence of new clinical features as a child develops or the identification of congenital anomalies or developmental disorders in additional family members.

Comparators

The comparators of interest for this indication are no further molecular testing following an initial WES test, and WGS following an initial WES test.

Outcomes

There is no reference standard for the diagnosis of individuals who have exhausted alternative testing strategies; therefore, diagnostic yield will be the clinical validity outcome of interest.

Study Selection Criteria

For the evaluation of clinical validity of repeat WES, studies that met the following eligibility criteria were considered:

Reported on the diagnostic yield of repeat WES;
Patient/sample clinical characteristics were described; children with congenital anomalies or neurodevelopmental disorders were included;
Patient/sample selection criteria were described;
Included at least 20 patients.

Clinically Valid

A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Review of Evidence

Systematic Review

Dai et al (2022) conducted a systematic review to determine the diagnostic yield of sequencing reanalysis of data from cases with no diagnosis following an initial WES or WGS test (Table 5).^33,The primary measure of efficacy was the proportion of undiagnosed individuals reaching a positive diagnosis on reanalysis after first round sequencing and analysis. Results are summarized in Table 6. The overall diagnostic yield was 0.10 (95% confidence interval [CI], 0.06 to 0.13). Using the GRADE framework, the certainty of the evidence for this outcome was rated moderate certainty. Confidence in the estimate was downgraded due to significant heterogeneity across studies that could not be explained by subgroup analyses. The researchers performed subgroup analyses on the basis of time interval between the original analysis and reanalysis (<24 months compared with ≥24 months), sequencing methodology (WES vs. WGS), study sample size (<50, 50 to 100, >100 patients), sequencing of family members for segregation analysis, whether research validation of novel variants/genes were conducted, and whether any artificial intelligence-based tools were used in variant curation. These subgroup analyses did not identify any statistically significant differences in diagnostic yield estimates.

Table 5. Systematic Review of the Diagnostic Yield of Whole Exome Sequencing Re-analysis- Characteristics

Study	Objective	Literature Search Dates	Study Inclusion Criteria	Populations	Primary Outcome	Quality Assessment Method
Dai et al (2022)^33,	To determine the diagnostic yield, optimal timing, and methodology of next generation sequencing data reanalysis in suspected Mendelian disorders	2007 to 2021	Cohort study that included performed reanalysis of NGS data and reported the yield of new molecular diagnoses after reanalysis. Reanalysis defined as bioinformatic examination of the original sequencing data	Individuals with suspected Mendelian disorders who had previously undergone cES or cGS without a molecular diagnosis being reached	Proportion of cases without a molecular diagnosis after initial sequencing that subsequently reached a diagnosis upon reanalysis.	Checklist derived from the 2015 Standards for Reporting of Diagnostic Accuracy criteria; 19 items covering patient eligibility and selection, test protocols, reporting of results, and study limitations

cES: clinical exome sequencing; cGS: clinical genome sequencing; NGS: next-generation sequencing.

Table 6. Systematic Review of the Diagnostic Yield of Whole Exome Sequencing Re-analysis- Results

	N studies (n Individuals)	Pooled Result, (95% CI)	Heterogeneity
Dai et al (2022)^33,
Overall diagnostic yield	29 (9419)	0.10 (0.06 to 0.13)	I² = 95.33%; P <.01
Subgroup analyses
Re-analysis 24 months or more after initial testing	7 (2906)	0.13 (0.09 to 0.18)	I² = 84%; P =.000
Re-analysis < 24 months after initial testing	11 (1077)	0.09 (0.06 to 0.13)	I² = 66.45%; P =.00
Studies re-analyzing WES	25 (4664)	0.11 (0.08 to 0.14)	I² = 84.30%; P <.01
Studies re-analyzing WGS	5 (344)	0.04 (0.01 to 0.09)	I² = 62.59%; P <.01

CI: confidence interval; WES: whole exome sequencing; WGS: whole genome sequencing

Twenty-three of 29 studies (representing 429 individuals) provided the reasons for achieving a diagnosis with re-analysis. In 62% of these cases the reason was a new gene discovery, in 15% the reasons were unknown or unspecified, and in 11% the reason was validation of candidate variants through research or external collaboration. Other reasons included bioinformatic pipeline improvements (3.3%), laboratory errors/misinterpretations (2.8%), updated clinical phenotypes (2.1%), copy number variants (1.9%), and additional segregation studies in relatives (1.2%).

Only 7 of 29 studies provided individual clinical information of sequenced probands (e.g., diagnosed variant, or timing of reanalysis) but instead reported summary data of the overall population. There were 11 studies that reported the finding of variants of uncertain significance (VUS) and/or variants in novel genes but only 8 studies provided research evidence confirming their pathogenicity. Only 3 studies discussed whether a genetic diagnosis led to management changes, and the impact on management was only described in a subgroup of individuals. To address uncertainties in the evidence, the review authors recommended best practices for future research including detailed inclusion and exclusion criteria, detailed clinical information on each case, clear documentation of methodology used for initial and re-analysis, and reporting of the rationale for attribution of pathogenicity.

Nonrandomized Studies

Table 7 summarizes nonrandomized studies published after the Dai et al (2022) systematic review. The diagnostic yield in these studies was consistent with previous studies. Study limitations were similar to those identified in previous studies (Tables 8 and 9).

Table 7. Nonrandomized Studies of Diagnostic Yield with Whole Exome Sequencing Re-analysis

Study	Population	N	Design	Yield, n (%)
Ewans et al (2022)^34,	Individuals with undiagnosed suspected Mendelian disorders recruited from genetics units from 2013 to 2017	91 individuals from 64 families	Retrospective cohort	WGS: 13/38 WES-negative families (34%) WES re-analysis (average 2 years later): 7/38 families (18%)
Halfmeyer et al (2022)^35,	Individuals with disorders who had been analysed via WES between February 2017 and January 2022	1040 affected individuals from 983 families	Retrospective cohort	Initial WES: 155/1040 Re-analysis: 7/885 (0.8%) of all nondiagnostic cases (9 variants were identified; 7 were disease-causing)
Sun et al (2022)^36,	100 children with global developmental delay/intellectual disability who had undergone CMA and/or WES and remained undiagnosed	100 affected individuals; 62 had received nondiagnostic WES	Prospective cohort	Overall: 21/100 (21%) CMA only: (64.3%, 9/14) WES only families: 9.7%, 6/62 CMA + WES families: 6/24 (25.0%)

CMA: chromosomal microarray analysis; WES: whole exome sequencing.

Table 8. Study Relevance Limitations

Study	Population^a	Intervention^b	Comparator^c	Outcomes^d	Duration of Follow-Up^e
Ewans et al (2022)^34,	3. Included highly heterogeneous diseases 4. Only half were pediatric age
Halfmeyer et al (2022)^35,	1,2 Included diagnostic and non-diagnostic samples 3. Included highly heterogeneous diseases 4. Population was not limited to those with no diagnosis following WES; Only half were pediatric age
Sun et al (2022)^36,

WES: whole exome sequencing. The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aPopulation key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.^bIntervention key: 1. Classification thresholds not defined; 2. Version used unclear; 3. Not intervention of interest.^c Comparator key: 1. Classification thresholds not defined; 2. Not compared to credible reference standard; 3. Not compared to other tests in use for same purpose.^d Outcomes key: 1. Study does not directly assess a key health outcome; 2. Evidence chain or decision model not explicated; 3. Key clinical validity outcomes not reported (sensitivity, specificity, and predictive values); 4. Reclassification of diagnostic or risk categories not reported; 5. Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests).^e Follow-Up key: 1. Follow-up duration not sufficient with respect to natural history of disease (true-positives, true-negatives, false-positives, false-negatives cannot be determined).

Table 9. Study Design and Conduct Limitations

Study	Selection^a	Blinding^b	Delivery of Test^c	Selective Reporting^d	Data Completeness^e	Statistical^f
Ewans et al (2022)^34,	1. selection not described
Halfmeyer et al (2022)^35,	1. selection not described
Sun et al (2022)^36,	1. selection not described				2. 5 cases were excluded due to the wrong samples (n = 2), poor sequencing data (n = 2), and (iii) variants were in the WES data but not detectable due to improper filtration

WES: whole exome sequencing.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment. WGS: whole genome sequencing.^aSelection key: 1. Selection not described; 2. Selection not random or consecutive (ie, convenience).^bBlinding key: 1. Not blinded to results of reference or other comparator tests.^cTest Delivery key: 1. Timing of delivery of index or reference test not described; 2. Timing of index and comparator tests not same; 3. Procedure for interpreting tests not described; 4. Expertise of evaluators not described.^d Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.^e Data Completeness key: 1. Inadequate description of indeterminate and missing samples; 2. High number of samples excluded; 3. High loss to follow-up or missing data.^f Statistical key: 1. Confidence intervals and/or p values not reported; 2. Comparison with other tests not reported.

Clinically Useful

Clinical utility of repeat WES testing would be demonstrated if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, more effective therapy, or avoid unnecessary therapy or testing.

Direct Evidence

No RCTs assessing the use of repeat WES to diagnose multiple unexplained congenital anomalies or a neurodevelopmental disorder following an initial WES test were identified.

Chain of Evidence

Due to heterogeneity and limitations in individuals studies, the evidence is insufficient to establish a chain of evidence for the clnical uitlity of repeat WES testing in individuals who are undiagnosed following an initial WES test.

Section Summary: Repeat Whole Exome Sequencing

In a systematic review of nonrandomized studies, re-analysis of WES data resulted in an 11% increase in diagnostic yield (95% CI 8% to 14%) in individuals who were previously undiagnosed via WES. However, the evidence is insufficient to establish the clinical utility of repeat testing. Individual studies lacked detail on the reasons for new diagnoses, changes in management based on new diagnoses, and the frequency of the identification of VUS. Additionally, the optimal timing of re-analysis has not been established, and there are no clear guidelines on what factors should prompt the decision to repeat testing.

Summary of Evidence

Reference No. 3

Population Reference No. 4

Whole Genome Sequencing for Children with Multiple Congenital Anomalies or a Neurodevelopmental Disorder of Unknown Etiology Following Standard Workup or Whole Exome Sequencing; Individuals who are not Critically Ill

Clinical Context and Test Purpose

The purpose of WGS in children who are not critically ill with multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following a standard workup is to establish a molecular diagnosis from either the coding or noncoding regions of the genome. The criteria under which diagnostic testing for a genetic or heritable disorder may be considered clinically useful are stated above.

The following PICO was used to select literature to inform this review.

Populations

The relevant population of interest is children who are not critically ill with multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following a standard workup.

Interventions

The relevant interventions being considered include: WGS with trio testing when possible. Several laboratories offer WGS as a clinical service. Medical centers may also offer rapid WGS (rWGS) as a clinical service. The median time for standard WGS is several weeks.

Note that this evidence review does not address the use of WGS for preimplantation genetic diagnosis or screening, prenatal (fetal) testing, or for testing of cancer cells.

Comparators

The following practices are currently being used to diagnose a suspected genetic disorder: A standard clinical workup without WES or WGS, and WES with trio testing when possible.

Outcomes

There is no reference standard for the diagnosis of patients who have exhausted alternative testing strategies; therefore, diagnostic yield will be the clinical validity outcome of interest.

Study Selection Criteria

For the evaluation of clinical validity of WGS, studies that met the following eligibility criteria were considered:

Reported on the diagnostic yield or performance characteristics such as sensitivity and specificity of rapid WGS or WGS;
Patient/sample clinical characteristics were described;
Patient/sample selection criteria were described;
Included at least 20 patients.

Whole Genome Sequencing Compared to Standard Clinical Workup

Clinically Valid

A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Review of Evidence

The use of WGS has been studied in children who are not critically ill with multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following standard workup in several observational studies, both prospective and retrospective. Studies are described in Table 10. The diagnostic yield of WGS has been between 20% and 40%. Additional indirect evidence is available from studies reporting diagnostic yield of WES in a similar population as summarized above, and it is reasonable to expect that WGS is likely to result in similar or better diagnostic yield for pathogenic or likely pathogenic variants as compared with WES.

Table 10. Diagnostic Yields with Whole Genome Sequencing in Children who are not Critically Ill with Multiple Unexplained Congenital Anomalies or a Neurodevelopmental Disorder of Unknown Etiology Following Standard Workup

Study	Patient Population	N	Design	Yield,n (%)	Additional Information
Lionel et al (2018)^37,	Well-characterized but genetically heterogeneous cohort of children <18 y that had undergone targeted gene sequencing Referral clinic: 44% metabolic, 23% ophthalmology, 15% joint laxity/hypermobility	103	Prospective Trio WGS testing for patients recruited from pediatric nongenetic subspecialists	42 (41)	Compared with a 24% yield with standard diagnostic testing and a 25% increase in yield from WES Limited information on change in management
Costain et al (2018), re-analysis^38, Stavropoulos et al (2016)^39,, original analysis	Children (<18 y) with undiagnosed congenital malformations and neurodevelopmental disorders Presentation: abnormalities of the nervous system (77%), skeletal system (68%), growth (44%), eye (34%), cardiovascular (32%), and musculature (27%)	64, re-analysis 100, original analysis	Prospective, consecutive Proband WGS was offered in parallel with clinical CMA testing	7 (11), re-analysis 34 (34), original analysis	Costain (2018) is a re-analysis of undiagnosed patients from Stavropoulos et al (2016) CMA plus targeted gene sequencing yield was 13% WGS yield highest for developmental delay 39% (22/57) and lowest (15%) for connective tissue disorders Change in management reported for some patients 7 incidental findings
Hiatt et al (2018) ^40, re-analysis Bowling et al (2017)^41, original analysis	Children with developmental and/or intellectual delays of unknown etiology 81% had genetic testing prior to enrollment	Original analysis included 244 Re-analysis included additional 123, for a total cohort of 494	Retrospective, selection method and criteria unclear Trio WGS in a referral center	54 (22)¹, original analysis	Re-analysis: Re-analysis yielded pathogenic or likely pathogenic variants that were not initially reported in 23 patients Downgraded 3 'likely pathogenic' and 6 VUS Original analysis: Compared to 30% yield for WES¹ Changes in management not reported 11% VUS in WGS
Gilissen et al (2014)^42,	Children with severe intellectual disability who did not have a diagnosis after extensive genetic testing that included whole exome sequencing	50	Trio WGS testing including unaffected parents	201 (42)	Of 21 with a positive diagnosis, 20 had de novo variants Changes in management not reported
Lindstrand et al (2022)^43,	Individuals with an intellectual disability diagnosis or a strong clinical suspicion of intellectual disability	229	Retrospective cohort; compared diagnostic yield from 3 genetic testing approaches: WGS 1st line, WGS 2nd line, and CMA with or without FMR1 analysis	WGS 1st line: 47 variants in 43 individuals (35%) WGS 2nd line: 48 variants in 46 individuals (26%) CMA/FMR1: 51 variants in 51 individuals (11%)	VUS: WGS 1st line: 12 of 47 variants were VUS WGS 2nd line: 14 of 34 variants were VUS CMA/FMR1:4 of 47 variants were VUS
van der Sanden et al (2022)^44,	Consecutive individuals with neurodevelopmental delay of suspected genetic origin; clinical geneticist had requested a genetic diagnostic test to identify the molecular defect underlying the individual's phenotype;	150	Prospective cohort; all had both SOC (including WES) and WGS with trio testing	SOC/WES: 43/150 (28.7%) WGS: 45/150 (30.0%)	VUS: WGS identified a possible diagnosis for 35 individuals of which 31 were also identified by the WES-based SOC pathway Management changes not addressed

CMA: chromosomal microarray analysis; SNV: single nucleotide variant; SOC: standard of care; VUS: variant of uncertain significance; WES: whole exome sequencing; WGS: whole genome sequencing. ¹ SNV/indel.

Tables 11 and 12 display notable limitations identified in each study.

Table 11. Study Relevance Limitations

Study	Population^a	Intervention^b	Comparator^c	Outcomes^d	Duration of Follow-Up^e
Lionel et al (2018)^37,	3. Included highly heterogeneous diseases	3. Proband testing only
Costain et al (2018), re-analysis^38,		3. Proband testing only
Bowling et al (2017)^41,	4. 19% had no prescreening performed
Gilissen et al (2014)^42,
Lindstrand et al (2022)^43,	3. Included highly heterogeneous diseases		3. No comparison to WES, 2nd line WGS cohort did not include individuals who had received WES
van der Sanden et al (2022)^44,	1. Individuals with a recognizable syndrome requiring confirmation were not excluded. 3. Included highly heterogeneous diseases			1. Management changes or health outcomes not addressed.

WES: whole exome sequencing; WGS: whole genome sequencing.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aPopulation key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.^bIntervention key: 1. Classification thresholds not defined; 2. Version used unclear; 3. Not intervention of interest.^c Comparator key: 1. Classification thresholds not defined; 2. Not compared to credible reference standard; 3. Not compared to other tests in use for same purpose.^d Outcomes key: 1. Study does not directly assess a key health outcome; 2. Evidence chain or decision model not explicated; 3. Key clinical validity outcomes not reported (sensitivity, specificity, and predictive values); 4. Reclassification of diagnostic or risk categories not reported; 5. Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests).^e Follow-Up key: 1. Follow-up duration not sufficient with respect to natural history of disease (true-positives, true-negatives, false-positives, false-negatives cannot be determined).

Table 12. Study Design and Conduct Limitations

Study	Selection^a	Blinding^b	Delivery of Test^c	Selective Reporting^d	Data Completeness^e	Statistical^f
Lionel et al (2018)^37,	1,2. Unclear how patients were selected from those eligible
Costain et al (2018), re-analysis^38,
Bowling et al (2017)^41,	1,2. Unclear how patients were selected from those eligible
Gilissen et al (2014)^42,
Lindstrand et al (2022)^43,	1. selection not described
van der Sanden et al (2022)^44,

The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment. WGS: whole genome sequencing.^aSelection key: 1. Selection not described; 2. Selection not random or consecutive (ie, convenience).^bBlinding key: 1. Not blinded to results of reference or other comparator tests.^cTest Delivery key: 1. Timing of delivery of index or reference test not described; 2. Timing of index and comparator tests not same; 3. Procedure for interpreting tests not described; 4. Expertise of evaluators not described.^d Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.^e Data Completeness key: 1. Inadequate description of indeterminate and missing samples; 2. High number of samples excluded; 3. High loss to follow-up or missing data.^f Statistical key: 1. Confidence intervals and/or p values not reported; 2. Comparison with other tests not reported.

Whole Genome Sequencing Compared to Whole Exome Sequencing

A 2020 Health Technology Assessment conducted by Ontario Health, with literature searches conducted in January 2019, included a comparative review of the diagnostic yield of WES and WGS in children with unexplained developmental disabilities or multiple congenital anomalies.^45, The diagnostic yield across all studies was 37% (95% CI, 34% to 40%). More studies, with an overall larger sample size, were included in the examination on WES (34 studies, N=9142) than on WGS (9 studies, N=648). Confidence intervals for studies using WES versus WGS overlapped (37%; 95% CI, 34% to 40%, vs. 40%; 95% CI, 32% to 49%). Diagnostic yield ranged between 16% and 73%, with variation attributed largely to technology used and participant selection. The overall quality of the evidence was rated as very low, downgraded for risk of bias, inconsistency, indirectness, and imprecision.

In some studies of WGS, the genes examined were those previously associated with the phenotype, while other studies were research-based and conducted more exploratory analysis. It has been noted that genomes sequenced with WGS are available for future review when new variants associated with clinical diseases are discovered.^37,

Studies have shown that WGS can detect more pathogenic variants than WES, due to an improvement in detecting copy number variants, insertions and deletions, intronic single-nucleotide variants, and exonic single-nucleotide variants in regions with poor coverage on WES. A majority of studies have described methods for interpretation of WGS indicating that only pathogenic or likely pathogenic variants were included in the diagnostic yield and that VUS were not reported. Five studies included in the Ontario Health Technology Assessment review provided data on the yield of VUS, with an overall yield of 17%. Only 1 of the 5 studies used WGS, however. The review authors noted, "Whole genome sequencing always results in substantially longer lists of variants of unknown significance than whole exome sequencing does. Interpreting and acting upon variants of unknown clinical significance is the single greatest challenge identified by clinicians….”^45,

Clinically Useful

A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence

No RCTs assessing the use of WGS to diagnose multiple unexplained congenital anomalies or a neurodevelopmental disorder outside of critical care were identified.

Chain of Evidence

Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility.

Clinical validity is established based on the meaningful diagnostic yield associated with WGS when a genetic etiology is uncertain after standard workup. Studies on WGS report changes in management that would improve health outcomes. The effect of WGS results on health outcomes are the same as those with WES, including avoidance of invasive procedures, medication changes to reduce morbidity, discontinuation of or additional testing, and initiation of palliative care or reproductive planning.

Section Summary: Whole Genome Sequencing for Children with Multiple Congenital Anomalies or a Neurodevelopmental Disorder of Unknown Etiology Following Standard Workup; Individuals who are not Critically Ill

Whole genome sequencing has been studied in non-critically ill children with congenital anomalies and developmental delays of unknown etiology following a standard workup. The diagnostic yield for WGS has been reported between 20% and 40%. A majority of studies described methods for interpretation of WGS indicating that only pathogenic or likely pathogenic variants were included in the diagnostic yield and that VUS were frequently not reported. Although the diagnostic yield of WGS is at least as high as WES in individuals without a diagnosis following standard clinical workup, it is unclear if the additional yield results in actionable clinical management changes that improve health outcomes. Further, while reporting practices of VUS found on exome and genome sequencing vary across laboratories, WGS results in the identification of more VUS than WES. The clinical implications of this difference are uncertain as more VUS findings can be seen as potential for future VUS reclassification allowing a diagnosis. However, most VUS do not relate to the patient phenotype, the occurrence of medical mismanagement and patient stress based on misinterpretation of VUS is not well defined, and provider reluctance to interpret VUS information lessen the value of additional VUS identification by WGS. As such, higher yield and higher VUS from WGS currently have limited clinical utility.

Summary of Evidence

For individuals who are children who are not critically ill with multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following a standard workup or WES who receive whole genome sequencing (WGS) with trio testing when possible, the evidence includes nonrandomized studies and a systematic review. Relevant outcomes are test validity, functional outcomes, changes in reproductive decision making, and resource utilization. In studies of children with congenital anomalies and developmental delays of unknown etiology following standard clinical workup, the yield of WGS has ranged between 20% and 40%. A majority of studies described methods for interpretation of WGS indicating that only pathogenic or likely pathogenic variants were included in the diagnostic yield and that VUS were frequently not reported. In a systematic review, the pooled (9 studies, N=648) diagnostic yield of WGS was 40% (95% CI, 32% to 49%). Although the diagnostic yield of WGS is at least as high as WES in individuals without a diagnosis following standard clinical workup, it is unclear if the additional yield results in actionable clinical management changes that improve health outcomes. Further, while reporting practices of VUS found on exome and genome sequencing vary across laboratories, WGS results in the identification of more VUS than WES. The clinical implications of this difference are uncertain as more VUS findings can be seen as potential for future VUS reclassification allowing a diagnosis. However, most VUS do not relate to the patient phenotype, the occurrence of medical mismanagement and patient stress based on misinterpretation of VUS is not well defined, and provider reluctance to interpret VUS information lessen the value of additional VUS identification by WGS. As such, higher yield and higher VUS from WGS currently have limited clinical utility. The evidence is insufficient to determine that the technology results in an improvement in the net health outcome.

Reference No. 4

Population Reference No. 5

Whole Genome Sequencing for a Suspected Genetic Disorder Other Than Multiple Congenital Anomalies or a Neurodevelopmental Disorder; Individuals who are not Critically Ill

Clinical Context and Test Purpose

The purpose of WGS in patients with a suspected genetic disorder of unknown etiology following a standard workup is to establish a molecular diagnosis from either the coding or noncoding regions of the genome. The criteria under which diagnostic testing for a genetic or heritable disorder may be considered clinically useful are stated above.

The following PICO was used to select literature to inform this review.

Populations

The relevant population of interest is children with a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following a standard workup.

Interventions

The relevant interventions being considered include: WGS with trio testing when possible. Several laboratories offer WGS as a clinical service. Medical centers may also offer WGS as a clinical service. The median time for standard WGS is several weeks.

Note that this evidence review does not address the use of WGS for preimplantation genetic diagnosis or screening, prenatal (fetal) testing, or for testing of cancer cells.

Comparators

The following practice is currently being used to diagnose a suspected genetic disorder: standard clinical workup without WES or WGS. A standard clinical workup for an individual with a suspected genetic condition varies by patient phenotype but generally involves a thorough history, physical exam (including dysmorphology and neurodevelopmental assessment, if applicable), routine laboratory testing, and imaging. If the results suggest a specific genetic syndrome, then established diagnostic methods relevant for that syndrome would be used.

Outcomes

There is no reference standard for the diagnosis of patients who have exhausted alternative testing strategies; therefore, diagnostic yield will be the clinical validity outcome of interest.

Study Selection Criteria

For the evaluation of clinical validity of WGS, studies that met the following eligibility criteria were considered:

Reported on the diagnostic yield or performance characteristics such as sensitivity and specificity of rapid WGS or WGS;
Patient/sample clinical characteristics were described;
Patient/sample selection criteria were described;
Included at least 20 patients.

Clinically Valid

A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Review of Evidence

The use of WGS has been studied in children with a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder in several observational studies, both prospective and retrospective. Studies are described in Table 13. The diagnostic yield of WGS has been between 9% and 55%. However, these studies include mixed indications with heterogeneous populations and include little information about associated changes in management following genetic diagnosis.

Table 13. Diagnostic Yields with Whole Genome Sequencing in Children with a Suspected Genetic Disorder other than Multiple Unexplained Congenital Anomalies or a Neurodevelopmental Disorder of Unexplained Etiology Following Standard Workup

Study	Patient Population	N	Design	Yield, n (%)	Additional Information
Costain et al (2020)^46,	Children with medical complexity (children with at least 1 feature from each of the following: technology-dependent or use of high-intensity care, fragility, chronicity, and complexity)	138 (49 probands)	Prospective WGS in patients referred to a single-center	15 (30.6)	Management decisions beyond genetic and reproductive counseling were influenced in at least 11 families
Thiffault et al (2019)^47,	Patients with suspected genetic disorders referred for genetic testing between 2015 and 2017. The majority had previous genetic testing without a diagnosis. The mean age was 7 yrs.	80	Prospective. The majority underwent trio sequencing; WGS was performed for the proband and WES was done for both parents	19 (24)	2 partial gene deletions detected with WGS that would not be detectable with WES
Alfares et al (2018)^48,	Undiagnosed patients (91% pediatric) who had a history of negative WES testing 70% Consanguinity	154 recruited; 108 included in analysis	Retrospective, selection method and criteria unclear	10 (9)	Reported incremental yield of WGS in patients with negative CGH and WES
Carss et al (2017)^49,	Unexplained inherited retinal disease; ages not specified	605	Retrospective NIHR-BioResource Rare Diseases Consortium	331 (55)	Compared with a detection rate of 50% with WES (n=117)
Ellingford et al (2016)^50,	Unexplained inherited retinal disease; ages not specified	46	Prospective WGS in patients referred to a single-center	24 (52)	Estimated 29% increase in yield vs. targeted NGS
Taylor et al (2015)^51,	Broad spectrum of suspected genetic disorders (Mendelian and immunological disorders)	217	Prospective, multicenter series Clinicians and researchers submitted potential candidates for WGS and selections were made by a scientific Steering Committee. Patients were eligible if known candidate genes and large chromosomal copy number changes had been excluded. Trio testing for a subset of 15 families.	46 (21)	34% yield in Mendelian disorders; 57% yield in trios
Yuen et al (2015)^52,	Individuals with diagnosed ASD	50	Prospective; unclear how patients were selected; quartet testing of extensively phenotyped families (parents and 2 ASD-affected siblings)	21 (42)	12/20 had change in management; 1/20 had change in reproductive counseling

ASD: autism spectrum disorder; CGH: comparative genomic hybridization; NGS: next-generation sequencing; NIHR: National Institute for Health Research; WES: whole exome sequencing; WGS: whole genome sequencing.

Tables 14 and 15 display notable limitations identified in each study.

Table 14. Study Relevance Limitations

Study	Population^a	Intervention^b	Comparator^c	Outcomes^d	Duration of Follow-Up^e
Costain et al (2020)^46,	3. Included heterogeneous diseases
Thiffault et al (2019)^47,	3. Included heterogeneous diseases
Alfares et al (2018)^48,	3: Clinical characteristics not described 4: 70% consanguinity	3. Appears to be proband testing only but not clear
Carss et al (2017)^49,	4. 25% had no prescreening performed
Ellingford et al (2016)^50,		3. Proband testing only
Taylor et al (2015)^51,	3. Included highly heterogeneous diseases
Yuen et al (2015)^52,	4: All patients had a clinical diagnosis		3: Results of standard diagnostic methods not discussed

The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aPopulation key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.^bIntervention key: 1. Classification thresholds not defined; 2. Version used unclear; 3. Not intervention of interest.^c Comparator key: 1. Classification thresholds not defined; 2. Not compared to credible reference standard; 3. Not compared to other tests in use for same purpose.^d Outcomes key: 1. Study does not directly assess a key health outcome; 2. Evidence chain or decision model not explicated; 3. Key clinical validity outcomes not reported (sensitivity, specificity, and predictive values); 4. Reclassification of diagnostic or risk categories not reported; 5. Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests).^e Follow-Up key: 1. Follow-up duration not sufficient with respect to natural history of disease (true-positives, true-negatives, false-positives, false-negatives cannot be determined).

Table 15. Study Design and Conduct Limitations

Study	Selection^a	Blinding^b	Delivery of Test^c	Selective Reporting^d	Data Completeness^e	Statistical^f
Costain et al (2020)^46,
Thiffault et al (2019)^47,	1,2: Unclear how patients were selected from those eligible
Alfares et al (2018)^48,	1,2: Unclear how patients were selected from those eligible
Carss et al (2017)^49,
Ellingford et al (2016)^50,
Taylor et al (2015)^51,
Yuen et al (2015)^52,	1,2. Unclear how patients were selected from those eligible

The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment. ^aSelection key: 1. Selection not described; 2. Selection not random or consecutive (ie, convenience).^bBlinding key: 1. Not blinded to results of reference or other comparator tests.^cTest Delivery key: 1. Timing of delivery of index or reference test not described; 2. Timing of index and comparator tests not same; 3. Procedure for interpreting tests not described; 4. Expertise of evaluators not described.^d Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.^e Data Completeness key: 1. Inadequate description of indeterminate and missing samples; 2. High number of samples excluded; 3. High loss to follow-up or missing data.^f Statistical key: 1. Confidence intervals and/or p values not reported; 2. Comparison with other tests not reported.

Clinically Useful

Direct Evidence

No RCTs assessing the use of WGS to diagnose a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder were identified.

Chain of Evidence

Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility.

Because the clinical validity of WGS for this indication has not been established, a chain of evidence cannot be constructed.

Section Summary: Whole Genome Sequencing for a Suspected Genetic Disorder Other Than Multiple Congenital Anomalies or a Neurodevelopmental Disorder; Individuals who are not Critically Ill

Whole genome sequencing has also been studied in children with a suspected genetic disorder other than multiple unexplained congenital anomalies or a neurodevelopmental disorder of unknown etiology following standard workup. The diagnostic yield of WGS has been between 9% and 55%. However, these studies include mixed indications with heterogeneous populations and include little information about associated changes in management following genetic diagnosis.

Summary of Evidence

Reference No. 5

Population Reference No. 6

Rapid Whole Exome or Genome Sequencing in Critically Ill Infants or Children

Clinical Context and Test Purpose

The purpose of rapid WES (rWES) or rWGS in critically ill patients with a suspected genetic disorder of unknown etiology is to establish a molecular diagnosis from either the coding or noncoding regions of the genome. The criteria under which diagnostic testing for a genetic or heritable disorder may be considered clinically useful are stated above.

The most common cause of death in neonates in the United States is genetic disorders. Currently, critically ill neonates with suspected genetic diseases are frequently discharged or deceased without a diagnosis. There are thousands of rare genetic disorders. The presentation of many of these disorders in neonates may be nonspecific or differ from the presentation in older patients and the disorder may produce secondary involvement of other systems due to the fragility of the neonate that obscures the primary pathology. The neonatal intensive care unit (NICU) treatment of suspected genetic diseases is often empirical. Rapid diagnosis is critical for delivery of interventions that reduce morbidity and mortality in genetic diseases for which treatments exist. For many genetic diseases there is no effective treatment and timely diagnosis limits futile intensive care.

The following PICO was used to select literature to inform this review.

Populations

The relevant population of interest is critically ill infants presenting with any of a variety of disorders and anomalies suspected to have a genetic basis but not explained by a standard workup. For example, patients may have a phenotype that does not correspond with a specific disorder for which a genetic test targeting a specific gene is available. Specifically for critically ill infants, the population would also include patients for whom specific diagnostic tests available for that phenotype are not accessible within a reasonable timeframe. Petrikin (2018) identified critically ill infants that are appropriate for rapid testing as meeting the following inclusion criteria: multiple congenital anomalies; an abnormal laboratory test suggests a genetic disease or complex metabolic phenotype; an abnormal response to standard therapy for a major underlying condition; significant hypotonia; or persistent seizures. Exclusion criteria included: an infection with normal response to therapy; isolated prematurity; isolated unconjugated hyperbilirubinemia; Hypoxic Ischemic Encephalopathy; confirmed genetic diagnosis explains illness; Isolated Transient Neonatal Tachypnea; or nonviable neonates.^53,

Interventions

The relevant interventions being considered include:

rapid WES with trio testing when possible
rapid WGS with trio testing when possible

Several laboratories offer WES or WGS as a clinical service. Medical centers may also offer rWES or rWGS or standard WES or WGS as a clinical service. The median time for standard WGS is several weeks. In its 2021 guideline, the American College of Medical Genetics and Genomics defines rapid and ultrarapid testing as 6 to 15 days and 1 to 3 days, respectively.^54,

Note that this evidence review does not address the use of WES or WGS for preimplantation genetic diagnosis or screening, prenatal (fetal) testing, or for testing of cancer cells.

Comparators

The following practice is currently being used to diagnose a suspected genetic disorder: a standard clinical workup without WES or WGS. A standard clinical workup for an individual with a suspected genetic condition varies by patient phenotype but generally involves a thorough history, physical exam (including dysmorphology and neurodevelopmental assessment, if applicable), routine laboratory testing, and imaging. If the results suggest a specific genetic syndrome, then established diagnostic methods relevant for that syndrome would be used.

Outcomes

Outcomes of interest are as described above for use of WES in patients with multiple congenital anomalies or a neurodevelopmental disorder. For critically ill infants, rapid diagnosis is important; therefore, in addition to the outcomes described in the previous section, time to diagnosis and time to discharge are also outcomes of interest.

Of course, mortality is a compelling outcome. However, many of the conditions are untreatable and diagnosis of an untreatable condition may lead to earlier transition to palliative care but may not prolong survival.

Study Selection Criteria

For the evaluation of clinical validity of rWES or rWGS, studies that met the following eligibility criteria were considered:

Reported on the diagnostic yield or performance characteristics such as sensitivity and specificity of rWES or rWGS;
Patient/sample clinical characteristics were described;
Patient/sample selection criteria were described;
Included at least 20 patients.

Clinically Valid

A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Review of Evidence

The use of rWES and rWGS has been studied in critically ill children in multiple observational studies, both prospective and retrospective, and in 3 RCTs. Studies are described in Table 16. The RCTs are discussed in more detail in the following ‘clinically useful’ section. One study included only infants with cardiac defects and had a diagnostic yield of 6% with WGS. The remaining studies included phenotypically diverse but critically ill infants and had yields of between 30% and 60%.

Table 16. Diagnostic Yields With Rapid Whole Exome or Whole Genome Sequencing in Critically Ill Infants or Children With a Suspected Genetic Disorder of Unknown Etiology

Study	Patient Population	N	Design	Yield, n (%)	Additional Information
rapid WES
Wu et al (2019)^55,	Pediatric patients (<18 yr old) who were critically ill (PICU; 68%) and suspected of having a genetic disease or newborns who were suspected of having a serious genetic disease after newborn screening. The primary phenotypes were neurologic (35%), cardiac (22.5%), metabolic (15%), and immunological (15%). Ages ranged from 0.2 mos to 13 yrs.	40	Eligibility and selection from eligible patients were unclear. Trio testing was performed.	21 (52.5)	Clinical management was changed for 81%: medications were recommended for 10 patients , transplantation was advised for 5 , and hospice care was suggested for 2
Elliott et al (2019) ^56,RAPIDOMICS	NICU neonates with unexplained seizures, metabolic disturbances (4%), neurological disorders (28%), multiple congenital anomalies (56%), or significant physiological disturbance for which diagnosis would likely change clinical management	25	Patients were evaluated for enrollment by a clinical geneticist and a neonatologist and approved by the research team. Trio analysis was performed. All patients with suspected definitely, possibly, or partially causal variants generated by rWES underwent Sanger validation	15 (60)	3 additional patients diagnosed with multi-gene panel testing or CMA 34 discrete and immediate medical decisions were identified for 15 of the 18 diagnosed patients
Gubbels et al (2019) ^57,	Infants age <6 mos admitted to ICU with recent presentation of seizures (20%), hypotonia (40%), multiple congenital anomalies (72%), complex metabolic phenotype (32%), or other.	50	New ICU admissions were triaged daily by a patient selection algorithm developed by a multidisciplinary medical team (neonatology, genetics, and neurology); whole-blood samples were collected from probands and parents for trio-based exome sequencing.	29 (58)	Results informed medical management changes in 24 of 29 patients. For 21 patients, there was an acute impact on care: switch to comfort care, specialist referral, decision not to pursue further diagnostic testing
Stark et al (2018)^12,	Acutely unwell pediatric patients with suspected monogenic disorders; 22% congenital abnormalities and dysmorphic features; 43% neurometabolic disorder; 35% other.	40	Recruited during clinical care by the clinical genetics services at the 2 tertiary pediatric hospitals; panel of study investigators reviewed eligibility; used singleton rWES.	21 (53)	Clinical management changed in 12 of the 21 diagnosed patients (57%) Median time to report of 16 days (range, 9 to 109)
Meng et al (2017)^58,	Critically ill infants within the first 100 days of life who were admitted to a tertiary care center between 2011 and 2017 and who were suspected to have genetic disorders. 208 infants were in the NICU or PICU at time of sample.	278 overall; 208 in NICU or PICU; 63 received rWES	Referred to tertiary care; proband WES in 63%, trio WES in 14%; critical trio rWES in 23%.	102 (37) overall; 32 (51) for rWES	Molecular diagnoses directly affected medical management in 53 of 102 patients (52%) overall and in 23 of 32, 72% who received rWES
rapid WGS
French et al (2019)^59,	Infants and children in the NICU and PICU admitted between 2016 and 2018 with a possible single gene disorder. Exclusion criteria for infants included: admitted for short stay post-delivery surveillance, prematurity without additional features, babies with a clear antenatal or delivery history suggestive of a non-genetic cause and those babies where a genetic diagnosis was already made. Median age, NICU: 12 days, PICU: 24 mos	Overall: 195 NICU: 106 PICU: 61 Pediatric neurology or clinical genetics department: 28	Trio WGS testing (when available) for the prospective cohort of families recruited in the NICU and PICU at a single site in the U.K.	Overall: 40 (21) NICU: 13 PICU: 25	Diagnosis affected clinical management in more than 65% of cases (83% in neonates) including modification of treatments (13%) and care pathways (35% in PICU, 48% in NICU) and/or informing palliative care decisions. For at least 7 cases, distinguishing between inherited and de novo variants informed reproductive decisions. VUS in 2 (1%)
Sanford et al (2019)^60,	Children 4 mos to 18 yrs admitted to a single-center PICU between 2016 and 2018 with suspicion for an underlying monogenic disease. Median age: 3 yrs Primary reasons for admission: respiratory failure (18%), shock (16%), altered mental status (13%), and cardiac arrest (13%)	38	Trio rWGS testing (when available) in a retrospective cohort study of consecutive children who had rWGS after admission to a single-center tertiary hospital in the U.S.	17 (45)	VUS identified in all cases but were not reported to patients. Changes in ICU management in 4 diagnosed children (24%), 3 patients had medication changes, 14 children had a subacute (non-ICU) change in clinical management that had implications for family screening
Hauser et al (2018)^61,	Neonatal and pediatric patients born with a cardiac defect in whom the suspected genetic disorder had not been found using conventional genetic methods	34	Trio rWGS testing for patients recruited from the NICU, PICU, or general inpatient pediatric ward of a single-center	2 (6)	VUS in 10 (26%)
Farnaes et al (2018)^62,	Critically ill infants with undiagnosed, highly diverse phenotypes. Median age 62 days (range 1 to 301 days). Multiple congenital anomalies, 29%; neurological, 21%; hepatic, 19%	42	Retrospective; comparative (received rWGS) and standard testing (mostly commonly CMA) Trio testing (when available) using rWGS	18 (43)	10% were diagnosed by a standard test Change in management after WGS in 13 of 18 (72%) patients with a new genetic diagnosis Estimated that rWGS reduced length of stay by 124 days
Mestek-Boukhibar et al (2018)^63,	Acutely ill infants with a suspected underlying monogenetic disease. Median age 2.5 mos. Referred from clinical genetics, 42%; immunology 21%; intensive care, 13%	24	Prospective; rWGS trio testing in a tertiary children's hospital PICU and pediatric cardiac intensive care unit.	10 (42)	Change in management in 3 patients
Van Diemen (2018)^64,	Critically ill infants with an undiagnosed illness excluding those with a clear clinical diagnosis for which a single targeted test or gene panel was available; median age 28 days. Presentation: cardiomyopathy, 17%, severe seizure disorder, 22%, abnormal muscle tone, 26%, 13% liver failure	23	Prospective rWGS Trio testing of patients from NICU/PICU; decision to include a patient was made by a multidisciplinary team; regular genetic and other investigations were performed in parallel	7 (30)	2 patients required additional sequencing data 1 incidental finding from WGS led to the withdrawal of unsuccessful intensive care treatment in 5 of the 7 children diagnosed
Willig (2015)^65,	Acutely ill infants with an undiagnosed illness, suspected genetic etiology; 26% congenital anomalies; 20% neurological; 14% cardiac; 11% metabolic Median age 26 days	35	Retrospective; enrolled in a research biorepository (nominated by treated physician, reviewed by panel of experts); had rWGS and standard diagnostic tests to diagnose monogenic disorders of unknown cause; trio testing	20 (57)	Had diagnoses with ‘strongly favorable effects on management’; Palliative care initiated in 6 infants of 20; WGS diagnoses were diseases that were not part of the differential at time of enrollment

CMA: chromosomal microarray analysis; ICU: intensive care unit; NICU: neonatal intensive care unit; PICU: pediatric intensive care unit; RAPIDOMICS: rapid genome-wide sequencing in a neonatal intensive care unit-successes and challenges; rWES: rapid whole exome sequencing; rWGS: rapid whole genome sequencing; VUS: variant of uncertain significance; WGS: whole genome sequencing; WES: whole exome sequencing.

Tables 17 and 18 display notable limitations identified in each study.

Table 17. Study Relevance Limitations

Study	Population^a	Intervention^b	Comparator^c	Outcomes^d	Duration of Follow-Up^e
Wu et al (2019)^55,			3: Results of standard diagnostic methods not discussed
Elliott et al (2019) ^56,
Gubbels et al (2019) ^57,			3: Results of standard diagnostic methods not discussed
Stark et al (2018)^12,	3. Included highly heterogeneous diseases	3. Proband testing only	3: Results of standard diagnostic methods not discussed
Meng et al (2017)^58,		3: Not all patients received rapid testing	3: Chromosomal microarray analysis was completed for 85% but results not discussed
French et al (2019)^59,			3: No comparator
Sanford et al (2019)^60,			3: No comparator
Hauser et al (2018)^61,			3: No comparator
Farnaes et al (2018)^62,	3. Included highly heterogeneous diseases
Mestek-Boukhibar et al (2018)^63,	3. Included highly heterogeneous diseases		3: No comparator
Van Diemen (2018)^64,	3. Included highly heterogeneous diseases		3: Results of standard diagnostic methods not discussed; were available after rWGS
Willig et al (2015)^65,	3. Included highly heterogeneous diseases		3: Results of standard diagnostic methods not discussed
Gilissen et al (2014)^42,

 rWGS: rapid whole genome sequencing.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aPopulation key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.^bIntervention key: 1. Classification thresholds not defined; 2. Version used unclear; 3. Not intervention of interest.^c Comparator key: 1. Classification thresholds not defined; 2. Not compared to credible reference standard; 3. Not compared to other tests in use for same purpose.^d Outcomes key: 1. Study does not directly assess a key health outcome; 2. Evidence chain or decision model not explicated; 3. Key clinical validity outcomes not reported (sensitivity, specificity and predictive values); 4. Reclassification of diagnostic or risk categories not reported; 5. Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests).^e Follow-Up key: 1. Follow-up duration not sufficient with respect to natural history of disease (true-positives, true-negatives, false-positives, false-negatives cannot be determined).

Table 18. Study Design and Conduct Limitations

Study	Selection^a	Blinding^b	Delivery of Test^c	Selective Reporting^d	Data Completeness^e	Statistical^f
Wu et al (2019)^55,	1: Criteria for selection unclear
Elliott et al (2019)^56,	2: Potential enrollees selected by a panel
Gubbels et al (2019)^57,	2: New ICU admissions were triaged by 1 team and enrollment criteria were applied by a panel
Stark et al (2018)^12,	2: Eligibility determined by panel; a minimum of 2 clinical geneticists had to agree rWES was appropriate for a patient to be enrolled
Meng et al (2017)^58,	1,2 Unclear if the patients were randomly or consecutively chosen from those who were eligible
French et al (2019)^59,	1,2. Unclear how patients were selected from those eligible
Sanford et al (2019)^60,
Hauser et al (2018)^61,
Farnaes et al (2018)^62,	2: Patients nominated by clinicians
Mestek-Boukhibar et al (2018)^63,	2: Eligibility criteria established after first 10 enrolled.
Van Diemen (2018)^64,	2: Decision to include a patient was made by a multidisciplinary team
Willig et al (2015)^65,	2: Nominated by treated physician, reviewed by panel of experts for inclusion
Gilissen et al (2014)^42,

ICU: intensive care unit; rWES: rapid whole exome sequencing.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment.^aSelection key: 1. Selection not described; 2. Selection not random or consecutive (ie, convenience).^bBlinding key: 1. Not blinded to results of reference or other comparator tests.^cTest Delivery key: 1. Timing of delivery of index or reference test not described; 2. Timing of index and comparator tests not same; 3. Procedure for interpreting tests not described; 4. Expertise of evaluators not described.^d Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.^e Data Completeness key: 1. Inadequate description of indeterminate and missing samples; 2. High number of samples excluded; 3. High loss to follow-up or missing data.^f Statistical key: 1. Confidence intervals and/or p values not reported; 2. Comparison with other tests not reported.

Clinically Useful

Direct Evidence

Randomized Controlled Trials

Three RCTs have evaluated rWGS or rWES in critically ill infants or children.

Kingsmore et al (2019) reported early results of A Randomized, Blinded, Prospective Study of the Clinical Utility of Rapid Genomic Sequencing for Infants in the Acute-care Setting (NSIGHT2) trial^66,. NSIGHT2 was a randomized, controlled, blinded trial of the effectiveness of rapid whole-genome or -exome sequencing (rWGS or rWES, respectively) in seriously ill infants with diseases of unknown etiology primarily from the NICU, pediatric intensive care unit (PICU), and cardiovascular intensive care unit (CVICU) at a single hospital in San Diego. Details of the study are provided in Table 19 and results are shown in Table 20. Ninety-five infants were randomized to rWES and 94 to rWGS. In addition 24 infants who were gravely ill received ultrarapid WGS (urWGS). The initial Kingsmore et al (2019) publication included only the diagnostic outcomes. The diagnostic yield of rWGS and rWES was similar (19% vs. 20%, respectively), as was time to result (median, 11 vs. 11 days). Although the urWGS was not part of the randomized portion of the study, the proportion diagnosed by urWGS was 11 of 24 (46%) and time to result was a median of 4.6 days. The incremental diagnostic yield of reflexing to trio testing after inconclusive proband analysis was 0.7% (1 of 147). In 2020, Dimmock et al reported results of the primary endpoint of NSIGHT2: clinician perception that rWGS was useful and clinican-reported changes in management.^67, Clinicians provided perceptions of the clinical utility of diagnostic genomic sequencing for 201 of 213 infants randomized (94%). In 154 (77%) infants, diagnostic genomic sequencing was perceived to be useful or very useful; perceptions of usefulness did not differ between infants who received rWES and rWGS, nor between urWGS and rWES/rWGS. Thirty-two (15%) of 207 clinician responses indicated that diagnostic genomic sequencing changed infant outcomes (by targeted treatments in 21 [10%] infants, avoidance of complications in 16 [8%], and institution of palliative care in 2 [1%] infants). Changes in outcome did not differ significantly between infants randomized to rWES and rWGS, although urWGS was associated with a signficantly higher rate of change in managment than rWES/rWGS (63% vs. 23%; p=.0001).

Petrikin et al (2018) reported on the Prospective Randomized Trial of the Clinical Utility of Rapid Next Generation Sequencing in Acutely Ill Neonates (NSIGHT1; NCT02225522) RCT of rWGS to diagnose suspected genetic disorders in critically ill infants.^53, In brief, NSIGHT1 was an investigator-initiated (funded by the National Human Genome Research Institute and Eunice Kennedy Shriver National Institute of Child Health and Human Development), blinded, and pragmatic trial comparing trio rWGS with standard genetic tests to standard genetic tests alone with a primary outcome of the proportion of NICU/PICU infants receiving a genetic diagnosis within 28 days. Parents of patients and clinicians were unblinded after 10 days and compassionate cross-over to rWGS occurred in 5 control patients. The study was designed to enroll 500 patients in each group but was terminated early due to loss of equipoise on the part of study clinicians who began to regard standard tests alone as inferior to standard tests plus trio rWGS. Intention-to-treat analyses were reported, ie, crossovers were included in the group to which they were randomized. The trial required confirmatory testing of WGS results, which lengthened the time to rWGS diagnosis by 7 to 10 days. Study characteristics are shown in Table 19 and results are shown in Table 20.

In the NICUSeq RCT, Krantz et al (2021) compared rWGS (test results returned in 15 days) to a delayed reporting group (WGS with test results returned in 60 days) in 354 infants admitted to an intensive care unit (ICU) with a suspected genetic disease at 5 sites in the US.^68, In 76% of cases, both parents were available for trio testing. Overall, 82 of 354 infants received a diagnosis (23%), with a higher yield in the 15-day group (Table 19). The primary outcome was change in management, measured at day 60. Significantly more infants in the rWGS group had a change in management compared with the delayed arm (21.1% vs 10.3%; p=.009; odds ratio, 2.3; 95% CI, 1.22 to 4.32). Changes in management included subspecialty referral (21 of 354, 6.0%), changes to medication (5 of 354, 1.4%), therapeutics specific to the primary genetic etiology (7 of 354; 2.0%) and surgical interventions (12 of 354; 3.4%). Survival and length of stay did not differ between the groups.

Table 19. Characteristics of RCTs of Rapid Whole Genome Sequencing in Critically Ill Infants

Study; Trial	Countries	Sites	Dates	Participants	Interventions
					Active	Comparator
Krantz et al (2021)^68, NICUSeq (NCT03290469)	U.S.	5	2017 to 2019	Infants aged 0 to 120 days who were admitted to an ICU (83% NICU, 7% PICU, 10% CVICU ) with a suspected genetic disease based on objective clinical findings for which genetic testing would be considered. At least 1 biological parent was required for participation. Exclusions: established genetic diagnosis, high clinical suspicion for trisomy 13, 18, 21, or monosomy X, or full explanation of the patient's phenotype by complications of prematurity.	N=176 WGS testing results returned 15 days after enrollment	N=178 WGS testing results 60 days after enrollment
Kingsmore et al (2019) ^66, Dimmock et al (2020)^67, NSIGHT2 (NCT03211039)	U.S.	1	2017 to 2018	Acutely ill infants, primarily from the NICU, PICU, and CVICU; age <4 mos; time from admission or time from development of a feature suggestive of a genetic condition of <96 h; excluding infants in whom there was a very low likelihood that a genetic disease diagnosis would change management.	N=94, rWGS initially performed with proband sequences alone; if diagnosis was not made, analysis was performed again, with parental samples	N=95, rWES initially performed with proband sequences alone; if diagnosis was not made, analysis was performed again, with parental samples
Petrikin (2018)^53,; NSIGHT1 (NCT02225522)	U.S.	1	2014 to 2016	Infants (<4m) in the NICU/PICU with illnesses of unknown etiology and: 1. genetic test order or genetic consult; 2. major structural congenital anomaly or at least 3 minor anomalies; 3. abnormal laboratory test suggesting genetic disease; or 4. abnormal response to standard therapy for a major underlying condition. Primary system involved: CA/musculoskeletal, 35%; Neurological, 25%; Cardiovascular, 17%; Respiratory, 6%	N=32 rWGS on specimens from both biological parents and affected infants simultaneously	N=33 Standard clinical testing for genetic disease etiologies was performed in infants based on physician clinical judgment, assisted by subspecialist recommendations

CA: congenital anomalies; CVICU: cardiovascular intensive care unit; ICU: intensive care unit; NICU: neonatal intensive care unit ; NSIGHT1: Prospective Randomized Trial of the Clinical Utility of Rapid Next Generation Sequencing in Acutely Ill Neonates; NSIGHT2; A Randomized, Blinded, Prospective Study of the Clinical Utility of Rapid Genomic Sequencing for Infants in the Acute-care Setting; PICU: pediatric intensive care unit; RCT: randomized controlled trial; rWES: rapid whole exome sequencing; rWGS: rapid whole genome sequencing; WGS: whole genome sequencing.

Table 20. Results of RCTs of Rapid Whole Genome Sequencing in Critically Ill Infants

Study	Diagnostic yield	Time to diagnosis	Age at at discharge/length of stay	Changes in management	Mortality
Krantz et al (2021)^68, NICUSeq NCT03290469	Diagnosis at day 60
WGS results at 15 days	55/176 31.0% (95% CI, 25.5% to 38.7%)	Data in graph only; "overall time to diagnosis was broadly associated with time to return of WGS testing."	No differences between groups in length of stay	34/161 21.1% (95% CI, 15.1% to 28.2%)	No differences between groups in survival observed
WGS results at 60 days	27/178 15.0% (95% CI, 10.2% to 21.3%)			17/165 10.3% (95% CI, 6.1% to 16.0%)
Treatment effect (95% CI)				Odds ratio, 2.3 (1.22 to 4.32)
Kingsmore et al (2019) ^66, Dimmock et al (2020)^67, NSIGHT2 (NCT03211039)	Genetic diagnosis, timing unspecified (%)	Proportion of results reported within 7 days (%)			Mortality at 28 days (%)
N	189	189	NR		189
rWGS	20%	11%		19/90 (21%)	3%
rWES	19%	4%		23/93 (25%)	0%
Treatment effect (95% CI)	p=.88	p=.10		p=.60	p=.25
Petrikin et al (2018)^53,; NSIGHT1	Genetic diagnosis within 28 days of enrollment (%)	Time (days) to diagnosis from enrollment, median	Age (days) at hospital discharge, mean	Change in management related to test results (%)	Mortality at 180 days (%)
N	65	65	65	65	65
rWGS	31%	13	66.3	41%¹	13%
Standard testing	3%	107	68.5	24%¹	12%
Treatment effect (95% CI)	p=.003	p=.002	p=.91	p=.11	NR

CI: confidence interval; NR: not reported; NSIGHT1: Prospective Randomized Trial of the Clinical Utility of Rapid Next Generation Sequencing in Acutely Ill Neonates; NSIGHT2; A Randomized, Blinded, Prospective Study of the Clinical Utility of Rapid Genomic Sequencing for Infants in the Acute-care Setting; RCT: randomized controlled trial; rWES: rapid whole exome sequencing; rWGS: rapid whole genome sequencing; WGS: whole genome sequencing.¹ Includes changes related to positive result (diagnosis); does not include impact of negative test results on management.

Tables 21 and 22 display notable limitations identified in each study.

Table 21. Study Relevance Limitations

Study	Population^a	Intervention^b	Comparator^c	Outcomes^d	Follow-Up^e
Krantz et al (2021)^68, NICUSeq NCT03290469			2. usual care testing varied	Patient and family-reported outcome measures not validated	1,2. 90 days might not have been long enough to assess outcomes
Kingsmore et al (2019) ^66, Dimmock et al (2020)^67, NSIGHT2 (NCT03211039)			2. no non-WGS/WES comparator	4: Outcomes based on clinician surveys 5: No discussion of clinically significant differences
Petrikin et al (2018)^53, NSIGHT1

NSIGHT1: Prospective Randomized Trial of the Clinical Utility of Rapid Next Generation Sequencing in Acutely Ill Neonates; NSIGHT2; A Randomized, Blinded, Prospective Study of the Clinical Utility of Rapid Genomic Sequencing for Infants in the Acute-care Setting; WES: whole exome sequencing; WGS: whole genome sequencing.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment. ^aPopulation key: 1. Intended use population unclear; 2. Clinical context is unclear; 3. Study population is unclear; 4. Study population not representative of intended use.^bIntervention key: 1. Not clearly defined; 2. Version used unclear; 3. Delivery not similar intensity as comparator; 4. Not the intervention of interest.^c Comparator key: 1. Not clearly defined; 2. Not standard or optimal; 3. Delivery not similar intensity as intervention; 4. Not delivered effectively.^d Outcomes key: 1. Key health outcomes not addressed; 2. Physiologic measures, not validated surrogates; 3. No CONSORT reporting of harms; 4. Not establish and validated measurements; 5. Clinical significant difference not prespecified; 6. Clinical significant difference not supported.^e Follow-Up key: 1. Not sufficient duration for benefit; 2. Not sufficient duration for harms.

Table 22. Study Design and Conduct Limitations

Study	Allocation^a	Blinding^b	Selective Reporting^d	Data Completeness^e	Power^d	Statistical^f
Krantz et al (2021)^68, NICUSeq NCT03290469	3: Allocation concealment not described
Kingsmore et al (2019) ^66, Dimmock et al (2020)^67, NSIGHT2 (NCT03211039)	3: Allocation concealment not described					4 :Only p-values reported; no treatment effects
Petrikin et al (2018)^53, NSIGHT1		1: Parents/clinicians unblinded at day 10 but analyses were intention-to-treat so crossovers would bias toward null			4: Trial stopped early, power for secondary outcomes will be very low	3, 4: Only p-values reported with no treatment effects or CIs

CI: confidence interval; NSIGHT1: Prospective Randomized Trial of the Clinical Utility of Rapid Next Generation Sequencing in Acutely Ill Neonates; NSIGHT2; A Randomized, Blinded, Prospective Study of the Clinical Utility of Rapid Genomic Sequencing for Infants in the Acute-care Setting.The study limitations stated in this table are those notable in the current review; this is not a comprehensive gaps assessment. ^aAllocation key: 1. Participants not randomly allocated; 2. Allocation not concealed; 3. Allocation concealment unclear; 4. Inadequate control for selection bias.^bBlinding key: 1. Not blinded to treatment assignment; 2. Not blinded outcome assessment; 3. Outcome assessed by treating physician.^c Selective Reporting key: 1. Not registered; 2. Evidence of selective reporting; 3. Evidence of selective publication.^d Data Completeness key: 1. High loss to follow-up or missing data; 2. Inadequate handling of missing data; 3. High number of crossovers; 4. Inadequate handling of crossovers; 5. Inappropriate exclusions; 6. Not intent to treat analysis (per protocol for noninferiority trials).^e Power key: 1. Power calculations not reported; 2. Power not calculated for primary outcome; 3. Power not based on clinically important difference; 4: Target sample size not achieved.^f Statistical key: 1. Analysis is not appropriate for outcome type: (a) continuous; (b) binary; (c) time to event; 2. Analysis is not appropriate for multiple observations per patient; 3. Confidence intervals and/or p values not reported; 4. Comparative treatment effects not calculated.

Chain of Evidence

Nonrandomized studies with over 200 infants are available to estimate performance characteristics of rWES in the NICU setting. Studies on rWGS report changes in management that would improve health outcomes. The effect of WGS results on health outcomes are the same as those with WES, including avoidance of invasive procedures, medication changes to reduce morbidity, discontinuation of or additional testing, and initiation of palliative care or reproductive planning. A chain of evidence linking meaningful improvements in diagnostic yield and changes in management expected to improve health outcomes supports the clinical value of WES and WGS for critically ill infants.

Section Summary: Rapid Whole Exome or Genome Sequencing in Critically Ill Infants or Children

For critically ill infants, disease may progress rapidly and genetic diagnoses must be made quickly. Several retrospective and prospective observational studies with sample sizes ranging from about 20 to more than 275 (in total including more than 450 critically ill infants or children) reported on diagnostic yield for rWGS or rWES. These studies included phenotypically diverse, but critically ill, infants and had yields between 30% and 60% and reports of changes in management such as avoidance of invasive procedures, medication changes, discontinuation of or additional testing, and initiation of palliative care.

Three RCTs have evaluated rWGS in critcially ill infants or children. An RCT comparing trio rWGS with standard genetic tests to diagnose suspected genetic disorders in critically ill infants funded by the National Institutes of Health was terminated early due to loss of equipoise on the part of study clinicians who began to regard standard tests alone as inferior to standard tests plus trio rWGS. The rate of genetic diagnosis within 28 days of enrollment was higher for rWGS versus standard tests (31% vs. 3%; p=.003) and the time to diagnosis was shorter (13 days vs. 107 days; p=.002). The age at hospital discharge and mortality rates were similar in the 2 groups. However, many of the conditions are untreatable and diagnosis of an untreatable condition may lead to earlier transition to palliative care, but may not prolong survival. A second RCT compared rWGS to rWES in seriously ill infants with diseases of unknown etiology from the NICU, PICU, and CVICU. The diagnostic yield of rWGS and rWES was similar (19% vs. 20%, respectively), as was time to result (median, 11 vs. 11 days). The NICUSeq RCT compared rWGS (test results returned in 15 days) to a delayed reporting group (WGS with test results returned in 60 days) in 354 infants admitted to an ICU with a suspected genetic disease. Diagnostic yield was higher in the rWGS group (31.0%; 95% CI, 25.5% to 38.7% vs. 15.0%; 95% CI, 10.2% to 21.3%). Additionally, significantly more infants in the rWGS group had a change in management compared with the delayed arm (21.1% vs. 10.3%; p=.009; odds ratio, 2.3; 95% CI, 1.22 to 4.32).

Summary of Evidence

Reference No. 6