Statistics from Altmetric.com
Editor—Cystic fibrosis (CF) is the most common, lethal, autosomal recessive childhood disorder in the white population, occurring in about 1 in 2500 live births.1 The incidence (1975-1995) is 1 in 936 live births, with a carrier rate of 1 in 15 in inhabitants of Saguenay-Lac-Saint-Jean (SLSJ), a geographically isolated region in north eastern Quebec.2
One hundred and sixty three patients in 143 families are known at the “Clinique de la Fibrose Kystique” in Chicoutimi (Saguenay), which is the referral centre for the whole region (285 000 inhabitants) and has followed all the CF patients but one since 1973. Molecular characterisation has been performed on all identified living CF patients.3 Three mutations account for 94.4% of the CF chromosomes; these are the ΔF508 (64.1%), 621+1G→T (22.6%), and A455E (7.7%) mutations.
SLSJ offers a unique opportunity to investigate whether there is a relationship between CFTR genotype, mutation, and survival. Indeed, all the CF patients are followed by the same multidisciplinary team, one of the paediatricians having been affiliated with the clinic since its opening. Although therapy has improved over this period of time, there is continuity and homogeneity in the follow up and treatment. Furthermore, the presence of three mutations, two severe and one mild, at high frequencies allowed us to compare different classes ofCFTR mutations.
Data collected on each patient were extracted from the files kept at the CF clinic in Chicoutimi where the patients come on a regular basis (approximately every two months) as outpatients for advice, follow up, and treatment. This medical visit consists of a physical examination, a growth and nutritional status evaluation, pulmonary function tests by spirometry, and a sputum culture. This phase of data gathering has been described in detail elsewhere.4
We only considered the genotyped patients, alive or dead after 1973, diagnosed before they reached 5 years old, in order to have a more homogeneous group consisting of patients having sufficient clinical signs and symptoms for the diagnosis to be made early in life. The Kaplan-Meier survival analysis was used to test whether there were significant differences (p<0.05) in the cumulative survival of the CF patients carrying one of the three main mutations.
Table lA shows the distribution of the 137 CF patients, including 15 dead (10.9%), by age groups and genotypes. It does not include 10 CF patients who died between 1973 and 1989 and were not genotyped. The proportion of patients homozygous for ΔF508 remains quite constant between the three age groups, whereas there is a decline in that of the patients carrying the 621+1G→T/ΔF508 genotype and an increase of those having a ΔF508/A455E or 621+1G→T/A455E genotype.
Because of the small numbers involved, the 15 genotypes were grouped according to the three main CFTR mutations (table 1B). The A455E mutation group contains all the patients with that mutation, independently of the second mutation composing their genotype. The CF patients carrying the 621+1G→T mutation, but not A455E, were included in a unique group. Patients homozygous or compound heterozygous for the ΔF508 mutation, without the 621+1G→T or the A455E mutation, were grouped together. We observed an increase of A455E and a depletion of 621+1G→T at older age groups, whereas the frequency of the ΔF508 mutation stayed constant (table 1B).
Fig 1 shows the cumulative survival by mutation. The Kaplan-Meier survival analysis could not be performed considering the three mutations because there were no dead patients in the A455E group. No significant difference was found between the ΔF508 and the 621+1G→T groups (p>0.05).
Evaluating survival as a function of theCFTR genotype or mutation requires that patients have the same access to medical care and had been diagnosed very early (before 5 years old). Our approach is different from those who determined the genotypes of adult patients or patients diagnosed at a later age5-7 and found rare mutations. Indeed, those patients should have a less severe phenotype or disease evolution, which does, in part, correlate with milder mutations.
The ΔF508 and 621+1G→T alleles are known to be severe mutations conferring pancreatic insufficiency when they are combined with another severe mutation.4 8 9 The A455E allele is a mild mutation associated with pancreatic sufficiency and exerts a dominant effect on the severe mutations. Compound heterozygotes for the A455E mutation have a milder pulmonary disease, no meconium ileus, and no late complications, such as diabetes and liver cirrhosis.2Therefore, since pulmonary insufficiency is the major cause of mortality in cystic fibrosis, it is not unexpected that no CF patients carrying the A455E mutation have died unlike the 13% of those with two severe mutations.
Since our series is small, the results should be considered as preliminary. However, it appears that survival is not only related to good care and treatment, but is also, at least in part, genetically determined.
The authors thank Mrs Simone Aubin, Claudette Larochelle, and Suzanne Migneault from the Clinique de Fibrose Kystique in Chicoutimi for their invaluable help. This study was supported in part by Organon Canada.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.