Residual-Function and Splicing CFTR Variants Characterize False-Negative IRT Newborn Screening: A Ten-Year Single-Center Cohort From Türkiye.
Cystic fibrosis (CF) newborn screening in Türkiye uses a two-step immunoreactive trypsinogen (IRT/IRT) algorithm with fixed cutoffs. The infants missed by fixed-cutoff screening remain poorly characterized.
In this retrospective single-center cohort (2015-2025), 90 children with confirmed CF were classified as IRT-positive (n = 78; IRT-1 ≥ 90 ng/mL) or IRT-negative (n = 12; IRT-1 < 90 ng/mL). CFTR variants were re-evaluated (ACMG/AMP) and stratified by Welsh-Smith class and residual-function status. Clinical, perinatal, pre-analytical, microbiological and genotype data were compared with standardized effect sizes.
IRT-negative patients were diagnosed later (median 12 vs. 2 months; p < 0.001) and had higher early Pseudomonas aeruginosa colonization (33.3% vs. 9.0%; unadjusted p = 0.037). Height standard deviation score was higher in IRT-negative patients (p = 0.004; Cohen's d = -0.77; not retained after false-discovery-rate adjustment). Mean IRT-1 was four-fold lower (44.5 vs. 189.1 ng/mL; p < 0.001). Class IV-V variants accounted for 33.3% of classified IRT-negative versus 11.1% of IRT-positive alleles; IRT-negative patients were enriched for residual-function and splicing variants, including 2789 + 5 G > A (9.5% vs. 2.0%), R334W and Y1032C. Pre-analytical and perinatal factors did not differ.
In every missed case, the first-tier IRT value fell below the fixed national cutoff, without evidence of pre-analytical or perinatal contributions. Missed cases clustered among carriers of residual-function and splicing CFTR variants, whose attenuated pancreatic injury may keep neonatal IRT below the cutoff. Because the Turkish protocol forgoes second-tier testing in IRT-1-negative infants, this subgroup is invisible to the current algorithm; the fixed cutoff warrants optimization.
In this retrospective single-center cohort (2015-2025), 90 children with confirmed CF were classified as IRT-positive (n = 78; IRT-1 ≥ 90 ng/mL) or IRT-negative (n = 12; IRT-1 < 90 ng/mL). CFTR variants were re-evaluated (ACMG/AMP) and stratified by Welsh-Smith class and residual-function status. Clinical, perinatal, pre-analytical, microbiological and genotype data were compared with standardized effect sizes.
IRT-negative patients were diagnosed later (median 12 vs. 2 months; p < 0.001) and had higher early Pseudomonas aeruginosa colonization (33.3% vs. 9.0%; unadjusted p = 0.037). Height standard deviation score was higher in IRT-negative patients (p = 0.004; Cohen's d = -0.77; not retained after false-discovery-rate adjustment). Mean IRT-1 was four-fold lower (44.5 vs. 189.1 ng/mL; p < 0.001). Class IV-V variants accounted for 33.3% of classified IRT-negative versus 11.1% of IRT-positive alleles; IRT-negative patients were enriched for residual-function and splicing variants, including 2789 + 5 G > A (9.5% vs. 2.0%), R334W and Y1032C. Pre-analytical and perinatal factors did not differ.
In every missed case, the first-tier IRT value fell below the fixed national cutoff, without evidence of pre-analytical or perinatal contributions. Missed cases clustered among carriers of residual-function and splicing CFTR variants, whose attenuated pancreatic injury may keep neonatal IRT below the cutoff. Because the Turkish protocol forgoes second-tier testing in IRT-1-negative infants, this subgroup is invisible to the current algorithm; the fixed cutoff warrants optimization.
Authors
Korkmaz Korkmaz, Akpınar Akpınar, Kılıç Kılıç, Kartkaya Kartkaya, Güney Güney, Yılmaz Yılmaz, Öztosun Öztosun, Çollak Çollak, Sakallı Sakallı
View on Pubmed