Cholesteatoma and family history: An international survey

To explore the relative frequency of a family history of cholesteatoma in patients with known cholesteatoma, and whether bilateral disease or earlier diagnosis is more likely in those with a family history. Associations between cleft lip or palate and bilateral disease and age of diagnosis were also explored.

A systematic review of the genetics of cholesteatoma 5 was conducted and identified a range of studies reporting familial clustering.
Whilst supportive of a genetic predisposition to cholesteatoma, there was not enough evidence to describe cholesteatoma as a heritable trait. The review hypothesised that subtypes of cholesteatoma may exist with different aetiological pathways and suggested there may be rare genetic variants that underlie the disease in some families.
Subsequently, the Genetics of Cholesteatoma (GoC) project (https://www.uea.ac.uk/rhino logy-group /resea rch/activ e-projects) established a database and sample bank with the aim of identifying candidate genetic variants of interest that co-segregate with a cholesteatoma diagnosis in families with several affected individuals. Whole exome sequencing of DNA collected from participants in one family revealed variants of interest in two genes. 6 However, these are preliminary studies, and the variants are of unknown significance to the disease pathology. Any inherited risk for a complex trait like cholesteatoma is likely to be polygenic in origin; but a rare variant with a major functional effect may be a significant risk factor in some families.
There have been numerous reports of associations between cholesteatoma and congenital conditions, namely cleft palate disorders and Turner syndrome. A Danish study found a 20-fold increase in the risk of cholesteatoma for those with cleft palate 7 and a Canadian study reported an even higher 200-fold increase in the rate of acquired cholesteatoma when comparing children with cleft lip and palate to the baseline rate. 8 There have also been reports that individuals who have a sibling with palate problems are more likely to have cholesteatoma. 7 Similarly, strong associations between cholesteatoma and Turner syndrome have been reported. 9,10 One study found that 7 out of 179 (3.9%) individuals with Turner syndrome had cholesteatoma 10 and another reported 26 out of 173 (15%) individuals with Turner syndrome had cholesteatoma. 9

| Objectives
This study aims to explore the relative frequency of a family history of cholesteatoma in patients with established cholesteatoma, and whether bilateral disease or earlier diagnosis is more likely in those with a positive family history. Associations between cleft lip or palate and bilateral disease and age of diagnosis were also explored.
Other hypothesis-generating associations are described in supplementary tables, namely whether a history of grommet insertion or tonsillectomy is associated with a younger age of cholesteatoma diagnosis.

| Study design
An online survey (http://smart survey.co.uk/s/chole steatoma) was created to gather information from those who have been diagnosed with cholesteatoma and gathered retrospective data on demographics, genetic factors and condition-specific factors.

| Setting
Initially data collection focused on patients presenting in Norfolk to ENT clinics at two sites. Patients with a diagnosis of cholesteatoma were personally invited to complete the questionnaire.
Subsequently, the project was expanded globally, whereby anyone in the world could complete the survey if they had access to an online computer. The survey continues to collect responses from participants.

| Participants
Participants were included if they had a diagnosis of cholesteatoma, had the capacity to consent to participation and were English speaking. There was no age restriction with some parents completing the survey on their child's behalf. Participants were excluded if they did not complete the majority of the questionnaire (at least 11 questions).

| Variables
Demographic questions encompassed; age, sex, ethnic background, geographic distribution, occupation and level of qualification.
Questions regarding the genetic nature of the condition focused on whether there was any family history of cholesteatoma, palate problems or inherited medical conditions. Condition-specific questions were also included in the survey and explored side of disease, age of diagnosis, history of other ear, nose or sinus conditions (including grommet insertion and tonsillectomy/adenoidectomy) and handedness (Tables S1 and S2). Survey questions are summarised in Table S4.

Key points
• 89 of 857 (10.4%) of respondents with cholesteatoma reported an affected family member.
• This study found a positive association between family history of cholesteatoma and bilateral cholesteatoma.
• Further population-based research is suggested to determine the heritability of cholesteatoma.

| Data sources
Data were collected via a web-based questionnaire over a 19-month period between October 2017 and April 2019. The survey consisted of 24 closed questions each with a selection of available answers.
Twelve questions provided open text boxes with the option to provide further details. Respondents could provide their postcode or country of origin in a free field text box. Participants were either invited following a hospital clinic visit or self-selected through invitations posted in patient support groups on social media. Social media groups were identified by searching social media sites for groups with the word "cholesteatoma" in their title. These groups were then contacted and asked whether they would share the invitation to participate, and the survey link, with their members.

| Bias
To ensure respondents had a cholesteatoma diagnosis, they were asked at the beginning of the questionnaire to confirm whether they had been invited to participate following a clinic visit or that they had a diagnosis of cholesteatoma and came across the survey online.

| Study size
Data collection was based on an online survey and no pre-set target was set.

| Quantitative variables
Most of the questionnaire consisted of questions with categorical answers. Quantitative variables collected included age at completion of survey and age at diagnosis of cholesteatoma.

| Statistical methods
The software package SPSS was used to generate descriptive statistics and conduct t tests, chi-squared tests and logistic regression. Independent samples t test was used to compare the average age, and chi-squared tests were conducted for each demographic category to compare differences between participants recruited online and from clinic. Subsequent analysis was conducted on the sample as a whole-combining participants from clinic and online.
Logistic regression was performed to compare unilateral and bilateral disease to any family history of cholesteatoma, having a first-degree relative with cholesteatoma and personal or family history of cleft lip/palate. T tests were conducted to determine any difference in mean age of diagnosis for those with a family history of cholesteatoma and other past medical history (Table S3).

| Participants
A total of 859 participants completed the survey. Two respondents were excluded as they failed to complete the survey, answering at most 5 of 22 questions. The sample consisted of 857 participants, 581 (68%) of which were female. The majority, 796 (93%), were recruited online as opposed to clinic. The mean age of diagnosis was 24.6 years of age, with a range of 0-80 years (Table 1). There was a peak in the age of diagnosis in childhood in the age range 3-10 years old.

| Descriptive data
Participants recruited online had a different gender distribution compared with the clinic participants; with a higher proportion of females (69.2%) than males (30.7%) compared with the almost even distribution of the clinic sample ( There was a range in the level of qualifications for the participants, with most having achieved GCSE or above. It must be noted that the sample included children, whom naturally will not have gained qualifications. The online recruits had on average higher levels of qualifications compared with the clinic recruits (Table 1, P < .001). There was an almost equal distribution in terms of right or left ear affected (40% and 37% respectively), 194/857 (23%) had both ears affected and there was no difference between online or clinic participants (Table 1; P = .46).

| Outcome data
When participants were asked whether they had a family member with cholesteatoma 89 of 857 (10.4%) answered "yes," 47 (5.5%) of these stating it was a first-degree relative. In terms of cleft palate or lip condition, 39 (4.6%) said they had a family history of cleft lip/palate and 16 (1.9%) stated they had this condition.

| Main results
There was a positive association between participants reporting an affected family member (any relation) and bilateral cholesteatoma, those with an affected family member were more likely to have bilateral cholesteatoma than those with no such family history ( There was no difference in mean age of diagnosis between those with a family history (any relative and first-degree relative) and those without such history (P = .23 (CI −1.39 to 6.08) and P = .52 (−6.62 to 3.36), respectively). Similarly, there was no difference in mean age of diagnosis for those with a cleft lip/palate or a family history of this condition (P = .11 (CI −1.61 to 15.12) and P = .78 (CI −4.68 to 6.23), respectively).

| Key results
Given that cholesteatoma is present in approximately 0.01% of the population it is interesting that 89 (10.4%) of 857 of respondents report an affected family member, 47 (5.5%) of whom are first-degree relatives. The association between family history of cholesteatoma and bilateral disease is a novel finding of this study.
There was no evidence in the sample of a positive association between a personal history of cleft lip/palate and bilateral ear disease, though the small number of participants with such history suggests these results should be considered with caution. There was also no evidence in the sample that a family history of cholesteatoma or a history of cleft lip/palate resulted in a younger mean age of cholesteatoma diagnosis.

| Interpretation
The number of respondents whom reported a family history of cholesteatoma is consistent with studies that report familial clustering and the suspicion of cholesteatoma as a heritable trait. 5 The association between family history of cholesteatoma and bilateral disease may represent a genetic liability, whereby those with a genetic predisposition are more likely to have more severe disease. It might be expected that those with a genetic predisposition for cholesteatoma would have a younger mean age of diagnosis, though these data do not support this hypothesis.
Various studies have reported an association between cleft lip/ palate and cholesteatoma. 7,8 The sample of this study is consistent with this; with 16/857 (1.9%) reporting to have had a cleft lip/palate problem themselves. Given that the incidence of cleft lip/palate in newborn babies in the UK is 1.7 per 1000, 12 or 0.17%, the proportion of people in the sample who report to have cleft lip/palate is higher than would be expected.
This sample includes a large number of participants with a wide age range of 1-80 years. The 2.1-1 female to male ratio differs from the widely reported higher incidence in men. 1,3,11 The gender ratio may represent a bias in the sampling technique whereby there is likely to be gender differences in health-seeking behaviour and social media engagement. The distribution of the age of diagnosis is consistent with other studies, with a peak in childhood. 2,3

| Limitations
The opportunistic sampling method presents challenges in how widely the results can be generalised. Participants were self-selected and, for the majority, their diagnosis of cholesteatoma was self-reported.
However, it can be argued that, given the rare nature of cholesteatoma, participants are unlikely to have self-diagnosed or have come across the survey by chance. Nevertheless, the survey relied on participants recalling details of their own diagnosis and past medical history.
In addition, there were limitations in the survey design including that questions regarding qualifications were originally formatted for UK respondents, subsequent international distribution of the survey resulted in respondents having to convert their education level into UK answers. Ethnicity data had to be excluded due to ambiguity in the available categorical answers.
Despite the limitations, this study is a relatively large survey of people with cholesteatoma. The sample captured people throughout the world affected by the condition from the very young to older generation and included a wide range of data on demographic, genetic, and condition-specific factors.

| CON CLUS ION
This study aimed to explore genetic factors in cholesteatoma. A larger number than would be expected reported a positive family history of cholesteatoma and a personal or family history of cleft lip or palate. There was a positive association between having a family history of cholesteatoma and bilateral cholesteatoma.

| Implications of research
This study offers further support for a genetic component to cholesteatoma and possible association with more severe disease.
Preliminary findings of genetic-based research have identified two gene variants of interest in cholesteatoma. 6 Further population-based research is suggested to determine the heritability of cholesteatoma.

CO N FLI C T S O F I NTE R E S T
None to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.