Validation of a Custom Next-Generation Sequencing Assay for Cystic Fibrosis Newborn Screening.

Newborn screening (NBS) for Cystic Fibrosis (CF) is associated with improved outcomes. All US states screen for CF; however, CF NBS algorithms have high false positive (FP) rates. In New York State (NYS), the positive predictive value of CF NBS improved from 3.7% to 25.2% following the implementation of a three-tier IRT-DNA-SEQ approach using commercially available tests. Here we describe a modification of the NYS CF NBS algorithm via transition to a new custom next-generation sequencing (NGS) platform for more comprehensive cystic fibrosis transmembrane conductance regulator (CFTR) gene analysis. After full gene sequencing, a tiered strategy is used to first analyze only a specific panel of 338 clinically relevant CFTR variants (second-tier), followed by unblinding of all sequence variants and bioinformatic assessment of deletions/duplications in a subset of samples requiring third-tier analysis. We demonstrate the analytical and clinical validity of the assay and the feasibility of use in the NBS setting. The custom assay has streamlined our molecular workflow, increased throughput, and allows for bioinformatic customization of second-tier variant panel content. NBS aims to identify those infants with the highest disease risk. Technological molecular improvements can be applied to NBS algorithms to reduce the burden of FP referrals without loss of sensitivity.

Low Psychosine in Krabbe Disease with Onset in Late Infancy: A Case Report.

Krabbe disease (KD) is a rare inherited neurodegenerative disorder caused by a deficiency in galactocerebrosidase enzyme activity, which can present in early infancy, requiring an urgent referral for hematopoietic stem cell transplantation, or later in life. Newborn screening (NBS) for KD requires identification and risk-stratification of patients based on laboratory values to predict disease onset in early infancy or later in life. The biomarker psychosine plays a key role in NBS algorithms to ascertain probability of early-onset disease. This report describes a patient who was screened positive for KD in New York State, had a likely pathogenic genotype, and showed markedly reduced enzyme activity but surprisingly low psychosine levels. The patient ultimately developed KD in late infancy, an outcome not clearly predicted by existing NBS algorithms. It remains critical that psychosine levels be evaluated alongside genotype, enzyme activity levels, and the patient's evolving clinical presentation, ideally in consultation with experts in KD, in order to guide diagnosis and plans for monitoring.

Newborn screening for Krabbe disease in New York State: the first eight years' experience.

PURPOSE: Krabbe disease (KD) results from galactocerebrosidase (GALC) deficiency. Infantile KD symptoms include irritability, progressive stiffness, developmental delay, and death. The only potential treatment is hematopoietic stem cell transplantation. New York State (NYS) implemented newborn screening for KD in 2006. METHODS: Dried blood spots from newborns were assayed for GALC enzyme activity using mass spectrometry, followed by molecular analysis for those with low activity (≤12% of the daily mean). Infants with low enzyme activity and one or more mutations were referred for follow-up diagnostic testing and neurological examination. RESULTS: Of >1.9 million screened, 620 infants were subjected to molecular analysis and 348 were referred for diagnostic testing. Five had enzyme activities and mutations consistent with infantile KD and manifested clinical/neurodiagnostic abnormalities. Four underwent transplantation, two are surviving with moderate to severe handicaps, and two died from transplant-related complications. The significance of many sequence variants identified is unknown. Forty-six asymptomatic infants were found to be at moderate to high risk for disease. CONCLUSIONS: The positive predictive value of KD screening in NYS is 1.4% (5/346) considering confirmed infantile cases. The incidence of infantile KD in NYS is approximately 1 in 394,000, but it may be higher for later-onset forms.

Screening for cystic fibrosis in New York State: considerations for algorithm improvements.

Newborn screening for cystic fibrosis (CF), a chronic progressive disease affecting mucus viscosity, has been beneficial in both improving life expectancy and the quality of life for individuals with CF. In New York State from 2007 to 2012 screening for CF involved measuring immunoreactive trypsinogen (IRT) levels in dried blood spots from newborns using the IMMUCHEM(™) Blood Spot Trypsin-MW ELISA kit. Any specimen in the top 5% IRT level underwent DNA analysis using the InPlex(®) CF Molecular Test. Of the 1.48 million newborns screened during the 6-year time period, 7631 babies were referred for follow-up. CF was confirmed in 251 cases, and 94 cases were diagnosed with CF transmembrane conductance regulated-related metabolic syndrome or possible CF. Nine reports of false negatives were made to the program. Variation in daily average IRT was observed depending on the season (4-6 ng/ml) and kit lot (<3 ng/ml), supporting the use of a floating cutoff. The screening method had a sensitivity of 96.5%, specificity of 99.6%, positive predictive value of 4.5%, and negative predictive value of 99.5%. CONCLUSION: Considerations for CF screening algorithms should include IRT variations resulting from age at specimen collection, sex, race/ethnicity, season, and manufacturer kit lots. WHAT IS KNOWN: Measuring IRT level in dried blood spots is the first-tier screen for CF. Current algorithms for CF screening lead to substantial false-positive referral rates. WHAT IS NEW: IRT values were affected by age of infant when specimen is collected, race/ethnicity and sex of infant, and changes in seasons and manufacturer kit lots The prevalence of CF in NYS is 1 in 4200 with the highest prevalence in White infants (1 in 2600) and the lowest in Black infants (1 in 15,400).

Clinical Sensitivity of Cystic Fibrosis Mutation Panels in a Diverse Population.

Infants are screened for cystic fibrosis (CF) in New York State (NYS) using an IRT-DNA algorithm. The purpose of this study was to validate and assess clinical validity of the US FDA-cleared Illumina MiSeqDx CF 139-Variant Assay (139-VA) in the diverse NYS CF population. The study included 439 infants with CF identified via newborn screening (NBS) from 2002 to 2012. All had been screened using the Abbott Molecular CF Genotyping Assay or the Hologic InPlex CF Molecular Test. All with CF and zero or one mutation were tested using the 139-VA. DNA extracted from dried blood spots was reliably and accurately genotyped using the 139-VA. Sixty-three additional mutations were identified. Clinical sensitivity of three panels ranged from 76.2% (23 mutations recommended for screening by ACMG/ACOG) to 79.7% (current NYS 39-mutation InPlex panel), up to 86.0% for the 139-VA. For all, sensitivity was highest in Whites and lowest in the Black population. Although the sample size was small, there was a nearly 20% increase in sensitivity for the Black CF population using the 139-VA (68.2%) over the ACMG/ACOG and InPlex panels (both 50.0%). Overall, the 139-VA is more sensitive than other commercially available panels, and could be considered for NBS, clinical, or research laboratories conducting CF screening.

Utility of a very high IRT/No mutation referral category in cystic fibrosis newborn screening.

Newborn screening for Cystic Fibrosis (CF) began in New York in October, 2002 using immunoreactive trypsinogen (IRT)/DNA methodology. Infants with at least one CFTR mutation or very high IRT and no mutations (VHIRT) are referred for sweat testing. In a preliminary analysis, we noted a very low positive predictive value (PPV) and preponderance of Hispanic infants in the group of infants with CF referred for VHIRT, which led to a decision to revise, but not eliminate, the VHIRT category. Automatic referral for specimens with VHIRT collected on the day of birth was eliminated, and the VHIRT threshold was raised from 0.2% to 0.1%. In this report, we describe outcomes from VHIRT referrals among 2.4 million infants screened between March 2003 and February 2013. Following the algorithm change, referrals decreased by 37.8% overall (annual mean 1,485 vs. 923), and the VHIRT PPV improved (0.6-1.0%). The number of infants diagnosed has remained consistent at 1 in 4,400 births. The proportion of Black/Hispanic/Asian/Other infants with confirmed CF, CFTR-related metabolic syndrome (CRMS), or possible CF/CRMS was 21.3% in infants with 1-2 mutations, but 75.8% in the VHIRT group. In conclusion, although the PPV among VHIRT referrals remains low, had this category never been implemented, 24 infants with confirmed CF, and 9 infants with CRMS or possible CF/CRMS, most of whom were Hispanic, would have been missed over the 10 years. Information from this study may be helpful in assessing the need for the VHIRT category and algorithm changes in other screening programs.