The relationship between sweat chloride levels and mortality in cystic fibrosis varies by individual genotype

Julia C. Espel, Hannah L. Palac, Ankit Bharat, Joanne Cullina, Michelle Prickett, Marc A Sala, Susanna A McColley, Manu Jain*

*Corresponding author for this work

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Rationale The association between CFTR genotype, sweat chloride and mortality has been inconsistent, but no previous analyses have examined the association stratified by individual genotypes. Objectives To evaluate the genotype-specific association between sweat chloride and mortality. Methods The CFF Patient Registry was assessed and included all patients in the registry between 1996 and 2012 with at least one F508del allele. We excluded patients without a documented genotype or plausible sweat chloride level. The primary outcome was time to mortality during the observation period. We examined 15 genotypes using the three most prevalent alleles in each of 5 classes. We compared subgroups of sweat chloride using Kaplan-Meier curves, log-rank tests, and multivariable Cox PH models. The overall predictive value of sweat chloride on mortality was assessed using area under the receiver operating characteristic curves. Measurements and main results 18,893 subjects met inclusion criteria. Sweat chloride distribution was similar across genotypes in patients with class 1 mutations, but was significantly different across genotypes in mutation classes 2–5. The R117H/F508del genotype patients demonstrated an association between sweat chloride and mortality (HR: 1.32 for every 10 mmol/L increase in sweat chloride [95% CI 1.12–1.54]. There were also significant associations in patients with F508del/F508del, I507del/F508del, G551D/F508del and 2789 + 5G → A/F508del genotypes, though the clinical relevance for these genotypes is unclear. Conclusions There is significant variability in sweat chloride distribution across CFTR class 2–5 genotypes. The relationship between sweat chloride and mortality varies by genotype with a relatively strong relationship in R117H/F508del patients.

Original languageEnglish (US)
Pages (from-to)34-42
Number of pages9
JournalJournal of Cystic Fibrosis
Volume17
Issue number1
DOIs
StatePublished - Jan 1 2018

Fingerprint

Sweat
Cystic Fibrosis
Chlorides
Genotype
Mortality
Registries
Alleles
Mutation
Proportional Hazards Models
ROC Curve
Observation

Keywords

  • CFTR genotype
  • Cystic fibrosis
  • Mortality
  • Mutation
  • Sweat chloride

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health
  • Pulmonary and Respiratory Medicine

Cite this

@article{32affb240c3a46328b177e331b56ac3b,
title = "The relationship between sweat chloride levels and mortality in cystic fibrosis varies by individual genotype",
abstract = "Rationale The association between CFTR genotype, sweat chloride and mortality has been inconsistent, but no previous analyses have examined the association stratified by individual genotypes. Objectives To evaluate the genotype-specific association between sweat chloride and mortality. Methods The CFF Patient Registry was assessed and included all patients in the registry between 1996 and 2012 with at least one F508del allele. We excluded patients without a documented genotype or plausible sweat chloride level. The primary outcome was time to mortality during the observation period. We examined 15 genotypes using the three most prevalent alleles in each of 5 classes. We compared subgroups of sweat chloride using Kaplan-Meier curves, log-rank tests, and multivariable Cox PH models. The overall predictive value of sweat chloride on mortality was assessed using area under the receiver operating characteristic curves. Measurements and main results 18,893 subjects met inclusion criteria. Sweat chloride distribution was similar across genotypes in patients with class 1 mutations, but was significantly different across genotypes in mutation classes 2–5. The R117H/F508del genotype patients demonstrated an association between sweat chloride and mortality (HR: 1.32 for every 10 mmol/L increase in sweat chloride [95{\%} CI 1.12–1.54]. There were also significant associations in patients with F508del/F508del, I507del/F508del, G551D/F508del and 2789 + 5G → A/F508del genotypes, though the clinical relevance for these genotypes is unclear. Conclusions There is significant variability in sweat chloride distribution across CFTR class 2–5 genotypes. The relationship between sweat chloride and mortality varies by genotype with a relatively strong relationship in R117H/F508del patients.",
keywords = "CFTR genotype, Cystic fibrosis, Mortality, Mutation, Sweat chloride",
author = "Espel, {Julia C.} and Palac, {Hannah L.} and Ankit Bharat and Joanne Cullina and Michelle Prickett and Sala, {Marc A} and McColley, {Susanna A} and Manu Jain",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.jcf.2017.11.002",
language = "English (US)",
volume = "17",
pages = "34--42",
journal = "Journal of Cystic Fibrosis",
issn = "1569-1993",
publisher = "Elsevier",
number = "1",

}

The relationship between sweat chloride levels and mortality in cystic fibrosis varies by individual genotype. / Espel, Julia C.; Palac, Hannah L.; Bharat, Ankit; Cullina, Joanne; Prickett, Michelle; Sala, Marc A; McColley, Susanna A; Jain, Manu.

In: Journal of Cystic Fibrosis, Vol. 17, No. 1, 01.01.2018, p. 34-42.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The relationship between sweat chloride levels and mortality in cystic fibrosis varies by individual genotype

AU - Espel, Julia C.

AU - Palac, Hannah L.

AU - Bharat, Ankit

AU - Cullina, Joanne

AU - Prickett, Michelle

AU - Sala, Marc A

AU - McColley, Susanna A

AU - Jain, Manu

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Rationale The association between CFTR genotype, sweat chloride and mortality has been inconsistent, but no previous analyses have examined the association stratified by individual genotypes. Objectives To evaluate the genotype-specific association between sweat chloride and mortality. Methods The CFF Patient Registry was assessed and included all patients in the registry between 1996 and 2012 with at least one F508del allele. We excluded patients without a documented genotype or plausible sweat chloride level. The primary outcome was time to mortality during the observation period. We examined 15 genotypes using the three most prevalent alleles in each of 5 classes. We compared subgroups of sweat chloride using Kaplan-Meier curves, log-rank tests, and multivariable Cox PH models. The overall predictive value of sweat chloride on mortality was assessed using area under the receiver operating characteristic curves. Measurements and main results 18,893 subjects met inclusion criteria. Sweat chloride distribution was similar across genotypes in patients with class 1 mutations, but was significantly different across genotypes in mutation classes 2–5. The R117H/F508del genotype patients demonstrated an association between sweat chloride and mortality (HR: 1.32 for every 10 mmol/L increase in sweat chloride [95% CI 1.12–1.54]. There were also significant associations in patients with F508del/F508del, I507del/F508del, G551D/F508del and 2789 + 5G → A/F508del genotypes, though the clinical relevance for these genotypes is unclear. Conclusions There is significant variability in sweat chloride distribution across CFTR class 2–5 genotypes. The relationship between sweat chloride and mortality varies by genotype with a relatively strong relationship in R117H/F508del patients.

AB - Rationale The association between CFTR genotype, sweat chloride and mortality has been inconsistent, but no previous analyses have examined the association stratified by individual genotypes. Objectives To evaluate the genotype-specific association between sweat chloride and mortality. Methods The CFF Patient Registry was assessed and included all patients in the registry between 1996 and 2012 with at least one F508del allele. We excluded patients without a documented genotype or plausible sweat chloride level. The primary outcome was time to mortality during the observation period. We examined 15 genotypes using the three most prevalent alleles in each of 5 classes. We compared subgroups of sweat chloride using Kaplan-Meier curves, log-rank tests, and multivariable Cox PH models. The overall predictive value of sweat chloride on mortality was assessed using area under the receiver operating characteristic curves. Measurements and main results 18,893 subjects met inclusion criteria. Sweat chloride distribution was similar across genotypes in patients with class 1 mutations, but was significantly different across genotypes in mutation classes 2–5. The R117H/F508del genotype patients demonstrated an association between sweat chloride and mortality (HR: 1.32 for every 10 mmol/L increase in sweat chloride [95% CI 1.12–1.54]. There were also significant associations in patients with F508del/F508del, I507del/F508del, G551D/F508del and 2789 + 5G → A/F508del genotypes, though the clinical relevance for these genotypes is unclear. Conclusions There is significant variability in sweat chloride distribution across CFTR class 2–5 genotypes. The relationship between sweat chloride and mortality varies by genotype with a relatively strong relationship in R117H/F508del patients.

KW - CFTR genotype

KW - Cystic fibrosis

KW - Mortality

KW - Mutation

KW - Sweat chloride

UR - http://www.scopus.com/inward/record.url?scp=85039942761&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039942761&partnerID=8YFLogxK

U2 - 10.1016/j.jcf.2017.11.002

DO - 10.1016/j.jcf.2017.11.002

M3 - Article

VL - 17

SP - 34

EP - 42

JO - Journal of Cystic Fibrosis

JF - Journal of Cystic Fibrosis

SN - 1569-1993

IS - 1

ER -