AERD Genetics Protocol v6
AERD Genetics Protocol v6
AERD Genetics Protocol v6
Population
March 29, 2018
Principal investigator:
Doron Sommer, MD, FRCS(C)
Clinical Professor
Otolaryngology - Head and Neck Surgery, McMaster University, Hamilton Health Sciences
Phone: 905-521-2100 Ext. 73078
Email: [email protected]
Co-investigators:
Guillaume Paré, MD, MSc, FRCP(C)
Associate Professor
Pathology and Molecular Medicine, Clinical Epidemiology & Biostatistics, McMaster University,
Hamilton Health Sciences
Phone: 905-527-4322 Ext. 40356
Email: [email protected]
Yaanu Jeyakumar
Research Assistant
Otolaryngology - Head and Neck Surgery, McMaster University
Phone: 416-859-4579
Email: [email protected]
Background
Aspirin Exacerbated Respiratory Disease (AERD) – also known as Samter’s Triad – is a chronic medical
condition that manifests as the combination of chronic rhinosinusitis with nasal polyposis (CRSwP),
hypersensitivity to non-steroidal anti-inflammatory drugs (NSAIDs), and asthma. Additional features
include eosinophilic chronic rhinosinusitis, hypereosinophilia, anosmia, frequent absence atopy, and
intolerance to ingestion of red wine and other alcoholic beverages. Affected patients will present with
severe disease often refractory to maximal medical therapy and repeated surgeries (1).
The pathophysiology of this disease remains to be completely elucidated. Genetic associations have been
found with polymorphisms of genes involved in the synthesis of proteins related to arachidonic acid
metabolism (LTC4S, ALOX5), antigen presentation (HLA), inflammation (IL5, IL17), and aspirin
metabolism (CYP2C19) (2). Recent genome-wide association studies by Shin et al. (2012), Park et al.
(2013), and Wang et al. (2017) have identified further single nucleotide polymorphisms (SNPs) and
genetic interactions potentially contributing to this AERD (3-5). The identification of these SNPs can
further elucidate the pathogenesis of this disease, serve as diagnostic markers, and predict therapeutic
response, especially with the advent of biologic agents such as Omalizumab and Mepolizumab (6, 7).
Finally, such studies of AERD have only been done in South Korea thus far. The present study aims to
conduct a genome-wide association study (GWAS) of AERD in a Canadian population. The results could
highlight differences in associated SNPs between populations, lead to biomarkers used in a diagnostic
panel for AERD, and guide clinical decision-making regarding treatment.
Objectives
Inclusion Criteria:
Exclusion Criteria
Participants
Patients will be those who meet inclusion criteria treated at Hamilton Health Sciences, at St. Joseph’s
Health Care London, and at McGill University Health Centre from July 2019 to July 2019. We anticipate
there will be ~100 patients who will met inclusion criteria and have a diagnosis of AERD within this time
frame. A similar number of consecutive patients with CRSwP and asthma without AERD will be
recruited from this time frame and will be matched to age and sex of the recruited AERD patients.
Study Design
Case-control study
Methods
We will collect the following data from the patient and their chart: age, sex, ethnicity, symptoms (NSAID
reactions, food/alcohol reactions, 22-item Sino-Nasal Outcome Test, 7-item Asthma Control
Questionnaire), eosinophil level, method of diagnosis, clinical course (sinusitis: age of polyp onset, time
to polyp regrowth, if on daily oral glucocorticoids >6 mo per year, number of lifetime polypectomies;
asthma: age of diagnosis, number of exacerbations/hospitalizations; AERD treatment: if ASA
desensitized, if on low salicylate diet).
Patients will undergo a buccal cell swab to collect a DNA sample. Their samples will be divided into an
AERD group (cases) and an Aspirin Tolerant CRSwP group (control). All samples will be sent to Dr.
Paré’s lab at the David Bradley Institute (Hamilton Health Sciences). DNA will be extracted and purified
from buccal samples using standard methodologies. Each sample will be genotyped using the Axiom
Precision Medicine Array (Affymetrix) which interrogates > 800K genetic variants. Genetic data will be
processed using established quality control criteria including sex-check (i.e. concordance between genetic
and self-reported sex to exclude sample mix-up), variant allele frequency, missingness, Hardy-Weinberg
equilibrium, etc.
Genetic association analysis will be performed using logistic regression, with case-control status as
dependent variable and number of the alternate allele as the independent variable (i.e. one logistic
regression model for each SNP). Adjustment for population stratification will be done using variance
component models, which will enable us to include all participants irrespective of ancestry. As a
sensitivity analysis, we will also perform an analysis stratifying by ancestry. To adjust for multiple
hypothesis testing, we will use both a conservative Bonferroni correction as well as False Discovery Rate
analysis. To increase statistical power, we will also leverage publicly available genetic databases of
healthy participants (up to >500,000 participants) and include these individuals as controls in our
analyses.
Finally, to create a diagnostic panel, the identified SNPs will be combined in various permutations. Their
respective resulting receiver-operating characteristic (ROC) curves and area under the curve (AUC) will
be calculated. We will compare the SNPs to the various characteristics of the patients and the severity of
their AERD and look for correlations.
Benefits
No specific benefit will be afforded to the patient. However, a better understanding of the genetics and
pathophysiology of AERD may help in the future to better diagnose and treat this debilitating illness.
Patients will be reimbursed for their parking fees if an additional follow-up visit is required to collect a
buccal cell sample.
Risks
There is no perceivable risk in undertaking a buccal cell swab.
Data Identifiers
Documents will be kept on a password-protected computer accessible only to the principal investigator
and co-investigators. Genetic data will be doubly de-identified to ensure a maximum level of
confidentiality protection. In other words, genetic data will use a randomized key different from the one
use to identify patients in the study. No identifiable data will be transferred to anyone outside the research
team at any point in time. The data will not become part of repository or a database for future use. Once
the study has completed, patient identifiers will be removed and the data will be stored for 2 years, and
then destroyed.
Budget