Cystic Fibrosis Newborn Screening: A Systematic Review-Driven Consensus Guideline from the United States Cystic Fibrosis Foundation

Abstract

Newborn screening for cystic fibrosis (CF) has been universal in the US since 2010; however, there is significant variation among newborn screening algorithms. Systematic reviews were used to develop seven recommendations for newborn screening program practices to improve timeliness, sensitivity, and equity in diagnosing infants with CF: (1) The CF Foundation recommends the use of a floating immunoreactive trypsinogen (IRT) cutoff over a fixed IRT cutoff; (2) The CF Foundation recommends using a very high IRT referral strategy in CF newborn screening programs whose variant panel does not include all CF-causing variants in CFTR2 or does not have a variant panel that achieves at least 95% sensitivity in all ancestral groups within the state; (3) The CF Foundation recommends that CF newborn screening algorithms should not limit CFTR variant detection to the F508del variant or variants included in the American College of Medical Genetics-23 panel; (4) The CF Foundation recommends that CF newborn screening programs screen for all CF-causing CFTR variants in CFTR2; (5) The CF Foundation recommends conducting CFTR variant screening twice weekly or more frequently as resources allow; (6) The CF Foundation recommends the inclusion of a CFTR sequencing tier following IRT and CFTR variant panel testing to improve the specificity and positive predictive value of CF newborn screening; (7) The CF Foundation recommends that both the primary care provider and the CF specialist be notified of abnormal newborn screening results. Through implementation, it is anticipated that these recommendations will result in improved sensitivity, equity, and timeliness of CF newborn screening, leading to improved health outcomes for all individuals diagnosed with CF following newborn screening and a decreased burden on families.

Keywords: F508del (p.Phe508del); cystic fibrosis; cystic fibrosis transmembrane conductance regulator; genetic testing; immunoreactive trypsinogen; newborn screening.

Figure 1

Example CF newborn screening algorithms for protocols using either IRT-DNA or IRT-DNA-SEQ. ^a State-determined; may be fixed or floating. ^b Variant panels may use sequencing methodology; however, if the list of variants to be released/reported is limited and pre-determined, then this is still considered a variant panel. ^c Not all states will utilize a very-high IRT (VHIRT) referral strategy. ^d Most states performing CFTR sequencing will only refer infants with two CF-causing variants for sweat testing.

Figure 2

Flowchart of the CF NBS algorithm and decision points for implementing consensus recommendations. All US programs initiate the CF NBS algorithm by measuring IRT; some perform two IRT measurements. Programs using a fixed IRT threshold should implement a floating IRT cutoff. Infants progressing to molecular analysis will undergo CFTR variant panel testing. This may be performed using sequencing technology if results are limited to a pre-determined list. Programs whose panel includes all CF-causing CFTR variants may choose whether to refer for sweat testing only infants with two CFTR variants OR infants with one or two CFTR variants. A program’s choice to refer only infants with two variants will likely result in decreased detection of CRMS/CFSPID, decreased sweat testing of carriers, but an increased risk of missing CF-affected infants with rare variants. The choice to refer infants with one or two variants will likely result in increased CRMS/CFSPID detection, increased sweat testing of carriers, but improved detection of CF-affected infants with rare variants. Programs whose variant panel does not include all CF- causing variants in CFTR2 and who are not performing CFTR sequencing should continue to refer infants with one or two variants detected. Programs performing CFTR sequencing following variant panel testing should refer only infants with two CF-causing variants. All programs should assess the sensitivity of their molecular analysis stage; if 95% detection within major ancestral sub-populations is not achieved and the variant panel does not include CF-causing variants from CFTR2, programs should maintain or implement a very-high IRT (VHIRT) referral strategy. ^a It is recommended that a variant panel achieve 95% detection rate in major ancestral sub-populations; ^b When making the determination of which infants should be referred for sweat testing, some programs will choose to include and report variants with interpretations other than CF-causing, such as variants of varying clinical consequences (VVCCs), variants of uncertain significance (VUS), and pathogenic or likely pathogenic variants.