Delivering health care for patients with primary ciliary dyskinesia: diagnosis and life-long care

Rubbo, Bruna (2020) Delivering health care for patients with primary ciliary dyskinesia: diagnosis and life-long care. Doctoral Thesis, 361pp.

Record type: Thesis (Doctoral)

Abstract

Primary ciliary dyskinesia (PCD) is a rare genetic disease that affects the function of cilia lining the upper and lower respiratory airways. High-speed video microscopy analysis (HSVA) is the only clinically available diagnostic test to assess ciliary function; however, European and North American diagnostic guidelines could not recommend its use as a confirmatory test due to insufficient evidence. The inability of cilia to beat in a coordinated manner can lead to varying degrees of progressive lung disease, recurrent otitis media with hearing loss and chronic rhinosinusitis. Disease expression is likely linked to genotype, with some genes responsible for a more severe phenotype. There is a lack of disease-specific validated clinical outcome measures to monitor disease progression and for use in clinical trials and prospective cohorts. Furthermore, service delivery models vary from country to country, yet there have been no comparisons on the effect of these differences in delivery of care for patients with PCD.

Objectives: This work sets out to examine different aspects of diagnosis and management of PCD patients by determining: a) the accuracy and reliability of HSVA to diagnose PCD, b) genotype-phenotype associations, c) appropriate clinical outcome measures to monitor disease progression and for use in research, and d) differences in service delivery that could affect disease outcomes.
Methods: We calculated the sensitivity, specificity, inter- and intra-observer reliability of HSVA to diagnose PCD compared to both European diagnostic guidelines and the multidisciplinary team clinical decisions (a). In order to investigate genotype-phenotype associations, we collected diagnostic, clinical and genetic data from three countries. We then applied a novel data-driven clustering method (i.e. topological data analysis) to generate models that would highlight underlying patterns in the phenotypic data and guide the selection of two genes for standard hypothesis testing (b). We conducted a scoping review in order to identify clinical outcome measures that have been used in PCD research and to recommend the use of a set of outcomes in future longitudinal studies (c). Finally, we performed interviews with PCD experts from various countries to develop a survey that would quantify similarities and differences between service delivery models. These were compared to results from a survey conducted ten years ago (d).

Results: We found that HSVA was both sensitive and specific to diagnose PCD, with high inter- and intra-observer agreement when performed by experienced scientists (a). This was the first powered and blinded study to assess HSVA in isolation compared to the current evidence-based European diagnostic guidelines.

Our topological models revealed a cluster of patients with mutations on the CCDC39 gene that had more severe respiratory phenotype, with statistically significantly lower FEV1 z-scores at diagnosis and higher proportion of history of neonatal respiratory distress (b). We also found a cluster of patients with DNAH11 mutations that had statistically significant milder respiratory phenotype. This was the first large scale multi-centred European study to investigate phenotype-genotype associations in PCD across multiple diagnostic and clinical parameters, using a novel data-driven approach.

The scoping review identified a variety of clinical outcome measures that have been used in PCD research, with significant differences in measurement and reporting of outcomes between studies (c). It also highlighted the lack of standardisation in selection, definition and application of outcomes across studies. We recommended using disease-specific validated outcome measures such as QOL-PCD in future randomised controlled trials in order to generate comparable results.

Comparisons between service delivery models revealed that diagnostic testing for PCD has become more standardised and centralised compared to ten years ago (d). However, management still varies considerably depending on geographical region and size of the PCD centre, with larger centres prioritising therapies that have been shown to be effective in PCD such as airway clearance and nasal rinsing.
Discussion: The findings from these studies provide evidence to improve the delivery of health care for patients with PCD. We showed that HSVA can play an important part in the diagnostic pathway. Our genotype-phenotype association study provided evidence for early and more aggressive intervention in patients with specific genetic mutations. The adoption of a core set of clinical outcome measures will generate the evidence needed for the development of management guidelines, the importance of which were highlighted when comparing service delivery models between countries.