Article Text

PDF

Can manifesting heterozygotes have cystic fibrosis?
  1. MAURICE SUPER
  1. Cystic Fibrosis Clinic and
  2. Department of Clinical Genetics
  3. Royal Manchester Children’s Hospital
  4. Pendlebury
  5. Manchester M27 4HA
  6. UK

Scientific advances and the passage of time are causing us to review what constitutes a diagnosis of cystic fibrosis (CF). Over and above those with a typical presentation and laboratory findings, there have always been patients who fall in the grey area—“does she or doesn’t she have CF?” Since the discovery of the CFTR gene in 1989 and the ongoing discovery of mutations associated with CF (the mutation count now stands at over 800), genetic analysis has brought a new sophistication to the area but has not solved all the problems. A consensus conference on diagnostic criteria arranged by the North American Cystic Fibrosis Foundation has published the agreed criteria.1 In addition to those with typical features (sweat chloride levels above 60 mmol/l and the finding of two mutations of CFTR known to be associated with the disease), the consensus statement admits that approximately 2% have an atypical phenotype with chronic sinopulmonary disease, pancreatic sufficiency, and either borderline (40–60 mmol/l) or normal (<40 mmol/l) sweat chloride concentrations. The predominance of single features may still allow a diagnosis of CF if two CF mutations, or abnormal sweat electrolytes, or abnormal ion transport across the nasal epithelium are found.2 3 A truly borderline diagnostic situation may apply when congenital bilateral absence of the vas deferens (CBAVD) or another form of obstructive azoospermia is the sole clinical feature with one or two CFTR mutations or the incompletely penetrant mutation (5T) in intron 8.4 In borderline patients the persistent finding of mucoid Pseudomonas aeruginosa in the sputum is highly suggestive of CF, while persistent colonisation withStaphylococcus aureus,Haemophilus influenzae, andBurkholderia cepacia may support a diagnosis of CF.

The interesting case described on page 278 of this issue ofThorax raises a number of points.5 The authors claim that their patient, born to consanguineous parents and with a history of having lost two infant siblings with CF, is a manifesting heterozygote—that is, is someone with only one CF mutation but showing the phenotype of the disease. While single system heterozygote effects are accepted as occurring in cases of CBAVD,6 disseminated bronchiectasis,7 allergic bronchopulmonary aspergillosis,8 and chronic pancreatitis,9 in these patients other features or abnormal tests that would allow a diagnosis of CF are absent. One could consider them as having CFTR disease but not CF. The degree of chest disease likely in the heterozygotes in whom the lung is involved would generally not approach that found in adults with CF. In this patient with a history of recurrent nasal polyps and severe sinopulmonary disease with clubbing and chronic Staphylococcus aureus infection a diagnostically abnormal nasal potential difference test allows a diagnosis of CF, even with a normal sweat test and normal pancreatic function.1 The authors’ claim is that, on the basis of parental consanguinity, adverse homozygosity at a number of alleles has resulted in a reduction in CFTR production as coded for by the patient’s “normal” CFTR gene. Thus, for one set of intragenic polymorphisms TG12T7 from intron 8 they quote from the literature that 30% of exon 9 transcript would not mature. One would have been more impressed had the finding been homozygosity of 5T rather than 7T, for with the smaller polyT tract there is abundant evidence of underproduction of CFTR due to inefficient splicing of exon 9.4 No CFTR mRNA data from the patient are presented. The authors have gone to considerable trouble to search for CFTR mutations by single stranded conformational polymorphism (SSCP) which included the promoter region and exon flanking intron regions and they also searched for deletions. One still cannot entirely exclude a functionally active intron mutation away from exon flanking region; the accepted CF mutation 3849+10 kb C>T9 is an example of such a mutation (incidentally giving CF with a normal sweat test, as in this patient). Similarly, denaturing gradient gel electrophoresis (DGGE) may reveal some mutations missed by SSCP. Thus, the claim by the authors of a manifesting heterozygote with CF can only remain theoretical.

Is there anything more which the authors could do to strengthen our belief in their hypothesis? What of the family history, of two infant siblings dying of severe early onset disease and necroscopic examinations having been carried out in at least one? At a clinical level the occurrence of meconium ileus and thus, by definition, pancreatic insufficiency had the child survived provides some support with the patient pancreatic sufficient. If tissue blocks are still available from the necroscopy then DNA extraction could permit a mutation search, starting first by testing for homozygosity for the single rare mutation, 1898+3 AG found in the study patient. However, it might be simpler to trace the consanguineous parents to check their mutation status. The occurrence of both mutations in the necroscopic tissue or of a CF mutation in each parent would provide strong supportive evidence for the hypothesis. On the other hand, if the fetal tissues revealed CFTR genetic identity to the case reported or a CF mutation could only be shown in one parent, this would support the existence of a second as yet undiscovered mutation in this patient. Even if only one parent were available for testing, the exercise would be worthwhile; if that person proved negative on a mutation search again an undiscovered mutation would become the more likely explanation, despite the rigour of the search by the authors for a mutation.

The hypothesis proposed by the authors could provide a possible explanation for the poor outlook of Pakistani children with CF in Britain11 whose parents are often consanguineous. Homozygosity of adverse polymorphisms, either within or outside the CF gene, such as the loci described on chromosome 19 could be playing a role.12

The finding of recurrent nasal polyps together with significant chest disease would have caused many physicians, especially those from CF centres, to have regarded this patient as having atypical CF worthy of an active therapeutic approach. Regardless of the test results, the clinical picture remains the most important guide in this respect. An extra responsibility is devolving onto paediatricians to try to detect those patients with an atypical picture in childhood who are going to run an adverse course, as in the case described. Mutation analysis will help in this respect but full measurement of nasal potential difference2 3 13 should be developed to the standards of a reliable test in CF centres so that the full diagnostic armamentarium is available.

References

Statistics from Altmetric.com

Scientific advances and the passage of time are causing us to review what constitutes a diagnosis of cystic fibrosis (CF). Over and above those with a typical presentation and laboratory findings, there have always been patients who fall in the grey area—“does she or doesn’t she have CF?” Since the discovery of the CFTR gene in 1989 and the ongoing discovery of mutations associated with CF (the mutation count now stands at over 800), genetic analysis has brought a new sophistication to the area but has not solved all the problems. A consensus conference on diagnostic criteria arranged by the North American Cystic Fibrosis Foundation has published the agreed criteria.1 In addition to those with typical features (sweat chloride levels above 60 mmol/l and the finding of two mutations of CFTR known to be associated with the disease), the consensus statement admits that approximately 2% have an atypical phenotype with chronic sinopulmonary disease, pancreatic sufficiency, and either borderline (40–60 mmol/l) or normal (<40 mmol/l) sweat chloride concentrations. The predominance of single features may still allow a diagnosis of CF if two CF mutations, or abnormal sweat electrolytes, or abnormal ion transport across the nasal epithelium are found.2 3 A truly borderline diagnostic situation may apply when congenital bilateral absence of the vas deferens (CBAVD) or another form of obstructive azoospermia is the sole clinical feature with one or two CFTR mutations or the incompletely penetrant mutation (5T) in intron 8.4 In borderline patients the persistent finding of mucoid Pseudomonas aeruginosa in the sputum is highly suggestive of CF, while persistent colonisation withStaphylococcus aureus,Haemophilus influenzae, andBurkholderia cepacia may support a diagnosis of CF.

The interesting case described on page 278 of this issue ofThorax raises a number of points.5 The authors claim that their patient, born to consanguineous parents and with a history of having lost two infant siblings with CF, is a manifesting heterozygote—that is, is someone with only one CF mutation but showing the phenotype of the disease. While single system heterozygote effects are accepted as occurring in cases of CBAVD,6 disseminated bronchiectasis,7 allergic bronchopulmonary aspergillosis,8 and chronic pancreatitis,9 in these patients other features or abnormal tests that would allow a diagnosis of CF are absent. One could consider them as having CFTR disease but not CF. The degree of chest disease likely in the heterozygotes in whom the lung is involved would generally not approach that found in adults with CF. In this patient with a history of recurrent nasal polyps and severe sinopulmonary disease with clubbing and chronic Staphylococcus aureus infection a diagnostically abnormal nasal potential difference test allows a diagnosis of CF, even with a normal sweat test and normal pancreatic function.1 The authors’ claim is that, on the basis of parental consanguinity, adverse homozygosity at a number of alleles has resulted in a reduction in CFTR production as coded for by the patient’s “normal” CFTR gene. Thus, for one set of intragenic polymorphisms TG12T7 from intron 8 they quote from the literature that 30% of exon 9 transcript would not mature. One would have been more impressed had the finding been homozygosity of 5T rather than 7T, for with the smaller polyT tract there is abundant evidence of underproduction of CFTR due to inefficient splicing of exon 9.4 No CFTR mRNA data from the patient are presented. The authors have gone to considerable trouble to search for CFTR mutations by single stranded conformational polymorphism (SSCP) which included the promoter region and exon flanking intron regions and they also searched for deletions. One still cannot entirely exclude a functionally active intron mutation away from exon flanking region; the accepted CF mutation 3849+10 kb C>T9 is an example of such a mutation (incidentally giving CF with a normal sweat test, as in this patient). Similarly, denaturing gradient gel electrophoresis (DGGE) may reveal some mutations missed by SSCP. Thus, the claim by the authors of a manifesting heterozygote with CF can only remain theoretical.

Is there anything more which the authors could do to strengthen our belief in their hypothesis? What of the family history, of two infant siblings dying of severe early onset disease and necroscopic examinations having been carried out in at least one? At a clinical level the occurrence of meconium ileus and thus, by definition, pancreatic insufficiency had the child survived provides some support with the patient pancreatic sufficient. If tissue blocks are still available from the necroscopy then DNA extraction could permit a mutation search, starting first by testing for homozygosity for the single rare mutation, 1898+3 AG found in the study patient. However, it might be simpler to trace the consanguineous parents to check their mutation status. The occurrence of both mutations in the necroscopic tissue or of a CF mutation in each parent would provide strong supportive evidence for the hypothesis. On the other hand, if the fetal tissues revealed CFTR genetic identity to the case reported or a CF mutation could only be shown in one parent, this would support the existence of a second as yet undiscovered mutation in this patient. Even if only one parent were available for testing, the exercise would be worthwhile; if that person proved negative on a mutation search again an undiscovered mutation would become the more likely explanation, despite the rigour of the search by the authors for a mutation.

The hypothesis proposed by the authors could provide a possible explanation for the poor outlook of Pakistani children with CF in Britain11 whose parents are often consanguineous. Homozygosity of adverse polymorphisms, either within or outside the CF gene, such as the loci described on chromosome 19 could be playing a role.12

The finding of recurrent nasal polyps together with significant chest disease would have caused many physicians, especially those from CF centres, to have regarded this patient as having atypical CF worthy of an active therapeutic approach. Regardless of the test results, the clinical picture remains the most important guide in this respect. An extra responsibility is devolving onto paediatricians to try to detect those patients with an atypical picture in childhood who are going to run an adverse course, as in the case described. Mutation analysis will help in this respect but full measurement of nasal potential difference2 3 13 should be developed to the standards of a reliable test in CF centres so that the full diagnostic armamentarium is available.

References

View Abstract

Request permissions

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.