Introduction
Recurrent respiratory papillomatosis is a rare disease in which benign squamous epithelial lesions develop within the airway, primarily involving the larynx. Affecting both children and adults, it is classified as juvenile-onset recurrent respiratory papillomatosis or adult-onset recurrent respiratory papillomatosis. This reflects the peak of onset at approximately 5 and 30 years of age, although an additional third peak at 64 years of age has also been reported.Reference San Giorgi, van den Heuvel, Tjon Pian Gi, Brunings, Chirila and Friedrich1
Recurrent respiratory papillomatosis is caused by ‘low risk’ human papillomavirus (HPV), primarily subtypes 6 and 11, and follows a different trajectory compared to high risk HPV 16 and 18.Reference Pim and Banks2 The high risk subtypes promote rapid proliferation in squamous cell carcinoma due to inhibition of tumour suppressor genes p53 and Rb by viral oncogenes E6 and E7.3, Reference Gluvajić, Hošnjak, Stegel, Novaković, Gale and Poljak4
The course of recurrent respiratory papillomatosis ranges from stable to rapid growth. Juvenile-onset recurrent respiratory papillomatosis generally is more aggressive and therefore more challenging to treat. Surgical management remains the mainstay of treatment and patients often require multiple operations to maintain voice and airway patency. As a chronic condition, recurrent respiratory papillomatosis can impede quality of life with avoidance of social activity and work absence.Reference So, McClellan and Best5 The HPV vaccine has decreased the incidence of recurrent respiratory papillomatosis and multiple adjuvant treatment options have been trialled to varying success.Reference Novakovic, Cheng, Zurynski, Booy, Walker and Berkowitz6–Reference Donne and Kinshuck8
Although the mechanism is not fully understood, HPV 6 and 11 in recurrent respiratory papillomatosis can promote cell proliferation and transformation to dysplasia and carcinoma. However, the rates of transformation are widely variable (13–78.4 per cent) for low-grade dysplasia.Reference Gluvajić, Hošnjak, Stegel, Novaković, Gale and Poljak4, Reference Carvalho, Huot, Charlois, Khalfallah, Chapuis and Froehlich9–Reference Blumin, Handler, Simpson, Osipov and Merati11 Potential influences contributing to the variability of dysplasia rates include patient factors (tobacco), disease and treatment factors (co-infection with high risk HPV, adjuvant cidofovir) and reporting factors (varying definitions of dysplasia between units, inter-pathologist variability, error secondary to small sample sizes).Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10
The low prevalence of recurrent respiratory papillomatosis, estimated to be 1.42 per 100 000 in the UK, makes gathering evidence and drawing conclusions challenging.Reference Donne, Keltie, Cole, Sims, Patrick and Powell12 A national UK registry has been developed, but at present there is no published evidence of recurrent respiratory papillomatosis patient outcomes from the United Kingdom in the last 30 years.Reference Sims, Keltie, Belilios, Burn, Brown and Jackson13, Reference Hartley, Hamilton, Birzgalis and Farrington14 We therefore aim to describe the demographics, evaluate the rate of dysplasia and carcinoma-ex-papillomatosis and assess risk factors in a large cohort of patients with recurrent respiratory papillomatosis at a tertiary adult otolaryngology unit.
Material and methods
A retrospective observational cohort study was performed. Patients, with histologically confirmed recurrent respiratory papillomatosis, who have been referred to or have been under the care of the Otolaryngology department at Liverpool University Hospital Foundation Trust, were included. Patients with recurrent respiratory papillomatosis were extracted from (1) the theatre logbook, (2) the voice clinic database held by Speech and Language Therapy department and (3) a prospective list of recurrent respiratory papillomatosis patients since January 2019. Patients with papillomas affecting only the palate, tongue, pharynx and/or other upper aerodigestive tract regions apart from the larynx and trachea were excluded.
Data were collected retrospectively from hospital clinic letters, operation notes and histology results. Data collected included date of birth, gender, smoking history, age of disease onset, number of procedures to date, anatomical distribution of papillomas, adjuvant cidofovir treatment, HPV vaccination status, dysplasia on at least one biopsy result and carcinoma-ex-papilloma. Number of procedures included the initial diagnostic endoscopy and biopsy and subsequent resections. In patients where a range of surgical procedures was provided (e.g. 5–10 surgeries), a number midpoint was chosen for analysis. Numbers of procedures documented as ‘several’ with no indication of numerical value were excluded. In patients with carcinoma-ex-papillomas, the number of procedures was recorded as the number of endoscopies until the diagnosis of malignancy. It was presumed that patients did not have the HPV vaccine if it was not documented in the clinic letters or visible on their general practitioner records.
Data were collected using Microsoft Excel version 14,0.7268.5000 and analysed using computer software Jamovi (Version 2.3). Fisher’s exact test was used for nominal data and Welch’s t-test for ordinal data due to the unequal cohort sizes between adult-onset recurrent respiratory papillomatosis and juvenile-onset recurrent respiratory papillomatosis and between the number of patients with adult-onset recurrent respiratory papillomatosis who developed dysplasia compared to those who did not. A p-value ≤ 0.05 was considered significant.
Results and analysis
Adult versus juvenile-onset recurrent respiratory papillomatosis
Of the 123 patients identified with recurrent respiratory papillomatosis, 102 patients were male and 21 were female. Nine patients had juvenile-onset recurrent respiratory papillomatosis and 114 had adult-onset recurrent respiratory papillomatosis with an average age of disease onset of 5 and 41, respectively. There was a statistically significant male predominance in adult-onset recurrent respiratory papillomatosis (6:1, male:female) compared to a more equal distribution in juvenile-onset recurrent respiratory papillomatosis (1.25:1, male:female). The estimated number of surgical procedures per patient to date was 1–65. As expected, patients with juvenile-onset recurrent respiratory papillomatosis had more surgical procedures: the median number of procedures was 27 per person (range 3–165) in the juvenile-onset recurrent respiratory papillomatosis compared to 3 procedures per person (range 1–75) in the adult-onset recurrent respiratory papillomatosis cohort. Distal involvement of the subglottis and/or trachea was more common in patients with juvenile-onset recurrent respiratory papillomatosis (Table 1).
Comparing clinical information data between adult-onset recurrent respiratory papillomatosis (AORRP) and juvenile-onset recurrent respiratory papillomatosis (JORRP)
Dysplasia
Thirteen patients in the adult-onset recurrent respiratory papillomatosis cohort had at least one biopsy that showed dysplasia (11 per cent), of which one progressed to carcinoma-ex-papillomatosis of the larynx and was subsequently treated with a laryngectomy and adjuvant radiotherapy. There were no cases of dysplasia or malignancy in the juvenile-onset recurrent respiratory papillomatosis group although this difference was not statistically significant. Given the difference in clinical features between the groups (Table 1), juvenile-onset recurrent respiratory papillomatosis patients were excluded from analysis of risks factors for dysplasia or carcinoma-ex-papillomatosis.
The association between gender and dysplasia with a male predominance was similar in both dysplasia and benign groups (6:1, male:female). Estimated age of disease onset was missing in three patients. Patients with evidence of dysplasia had an average older age of disease onset compared to those without dysplasia (49 years vs 39 years, p = 0.029). Smoking history was missing in eight patients without dysplasia. Patients with dysplasia had a higher, but insignificant, smoking history compared to patients without dysplasia (46 per cent to 37 per cent, respectively).
A greater proportion of patients with dysplasia had distal involvement with subglottic and/or tracheal disease compared to patients without dysplasia (15 per cent vs 6 per cent) but did not reach statistical significance. Disease aggression was measured by the number of surgeries required and there was no difference between both groups. Data were missing on the number of procedures in two patients in the benign group and another was reported as ‘several’; all three were excluded from analysis. The rate of cidofovir use was low in our population overall and there was no statistical difference between groups (Table 2).
Comparing clinical information between patients with and without history of dysplasia; AORRP = adult-onset recurrent respiratory papillomatosis
HPV vaccination
Thirty-three patients out of 123 (27 per cent) had received at least one dose of the HPV vaccine, Gardasil quadrivalent or Gardasil 9. Vaccination uptake was significantly greater in the juvenile-onset recurrent respiratory papillomatosis population compared to the adult-onset recurrent respiratory papillomatosis (66 per cent vs 24 per cent, p = 0.01). This corresponds with a lower age of disease associated with a greater uptake of the vaccine. The number of procedures a patient had undergone on average was greater amongst those who had the vaccine but was not statistically significant (Table 3).
Comparison of clinical data between patients that had the and patients that did not have the HPV vaccine
Discussion
To our knowledge, this is the largest published series of recurrent respiratory papillomatosis patients from the UK. We evaluated the risk factors for dysplasia and carcinoma-ex-papillomatosis for patients at our tertiary otorhinolaryngology and head and neck unit. Our unit, which receives referrals from across the Liverpool City region and North Wales, serves a population of approximately 2.25 million.15, 16 Our rate of recurrent respiratory papillomatosis is approximately 5.2 per 100,000 population and is much higher than the predicted national prevalence rate of 1.42 per 100,000.Reference Donne, Keltie, Cole, Sims, Patrick and Powell12 Furthermore, it is likely patients with less severe disease are managed by local trusts without onward referral to our tertiary unit and therefore not captured within the data. The reason for this discordance is unknown.
Our findings, which are similar to other studies, demonstrate that juvenile-onset recurrent respiratory papillomatosis is associated with distal involvement of the subglottis/and or trachea and generally is a more aggressive disease requiring more surgeries.Reference Carvalho, Huot, Charlois, Khalfallah, Chapuis and Froehlich9 There were no cases of dysplasia within the juvenile-onset recurrent respiratory papillomatosis cohort, suggesting aggressive disease may not reflect dysplastic changes. It did not reach statistical significance which is likely a consequence of the low number of patients with juvenile-onset recurrent respiratory papillomatosis in our study, as most other studies, apart from Khan and Naidu,Reference Khan and Naidu20 that compared AORPP and juvenile-onset recurrent respiratory papillomatosis do report a lower rate of dysplasia amongst patients with juvenile-onset disease.Reference Carvalho, Huot, Charlois, Khalfallah, Chapuis and Froehlich9, Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10, Reference Hock, Kennedy, Howell, Friedman, de Alarcon and Khosla17–Reference Khan and Naidu20
The rate of dysplasia in our cohort was 11 per cent and rate of carcinoma-ex-papillomatosis was 0.9 per cent. Karatayli-Orzgursoy et al.’s review of the literature found a wide range of dysplasia rates (10–55 per cent) for moderate dysplasia and above.Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10 An updated review of the last 20 years shows our rate (11 per cent) sits at the lower end of the range overall but is lower than the two other studies with large cohorts of more than 100 patients with adult-onset recurrent respiratory papillomatosis, which both reported dysplasia rates of 20 per cent (Table 4).Reference Preuss, Klussmann, Jungehulsing, Eckel, Guntinas-Lichius and Damm18, Reference Omland, Lie, Akre, Sandlie, Jebsen and Sandvik19 Gluvajić’s large study of 163 patients from Slovenia including adult-onset recurrent respiratory papillomatosis and juvenile-onset recurrent respiratory papillomatosis is a significant outlier reporting a 100 per cent rate of dysplasia, of which 22 per cent were high-grade dysplasia or squamous cell carcinoma-ex-papillomatosis.Reference Gluvajić, Hošnjak, Stegel, Novaković, Gale and Poljak4 Potential contributing factors to the wide range include differing histological definitions of dysplasia, inter-pathologist variability and co-infection with high risk HPV subtypes.Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10 Whilst it may be possible that we have a particularly low rate of dysplasia, Karatayli-Orzgursoy et al. suggested that units which treat larger numbers of patients with recurrent respiratory papillomatosis may be willing to label some HPV-induced pathological changes as ‘atypia’ as opposed to ‘dysplasia’.Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10
Rates of dysplasia in recurrent respiratory papillomatosis in the literature between 2004 and 2024; AORRP = adult-onset recurrent respiratory papillomatosis; JORRP = juvenile-onset recurrent respiratory papillomatosis; pts = patients
An older age of disease onset was the only significant risk factor for dysplasia identified in our cohort. We report a difference of 10 years between the dysplasia and benign groups (39 vs 49 years of age). Three previous studies had similar findings, although one did not reach statistical significance, adding to the strength of evidence surrounding age as an important etiological factor.Reference Gluvajić, Hošnjak, Stegel, Novaković, Gale and Poljak4, Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10, Reference Blumin, Handler, Simpson, Osipov and Merati11 Patient tobacco use sex and have repeatedly been noted not to be associated with dysplastic malformation in the context of recurrent respiratory papillomatosis.Reference Gluvajić, Hošnjak, Stegel, Novaković, Gale and Poljak4, Reference Karatayli-Ozgursoy, Bishop, Hillel, Akst and Best10, Reference Blumin, Handler, Simpson, Osipov and Merati11 Freeman et al. reported smokers with recurrent respiratory papillomatosis had a high rate of malignant transformation compared to non-smokers, which is unsurprising because smoking is a known head and neck carcinogen in the general population.Reference Freeman, Erickson, Koch, Young, Allen and Kim28 Hall et al. reported patients with mild dysplasia did not progress to high-grade or carcinoma-ex-papillomatosis.Reference Hall, Chen, Yoo, Fletcher, Ossoff and Garrett25 The findings of these studies including ours question whether mild or moderate dysplasia in the context of recurrent respiratory papillomatosis is truly a pre-malignant condition or a separate non-concerning entity.
Due to the limited numbers of patients, we were unable to calculate number of surgeries within a predetermined time frame and report the total number of surgeries per patient to date. Whilst this limits accurate assessment of disease aggression, our findings support others that aggressive disease does not predispose dysplasia.Reference Blumin, Handler, Simpson, Osipov and Merati11, Reference Hall, Chen, Yoo, Fletcher, Ossoff and Garrett25
Our study is limited by lack of dysplasia grading and HPV typing. As suggested previously, low-grade dysplasia may not progress to high-grade dysplasia, therefore further studies may benefit from examining these groups separately. This was not possible in our study as biopsies were taken at multiple hospitals including peripheral units prior to referral to the tertiary centre. The importance of HPV typing is debated in the literature with some studies reporting a higher number of endoscopies with HPV 11 whilst others argue age of onset is a more important prognostic factor.Reference Carvalho, Huot, Charlois, Khalfallah, Chapuis and Froehlich9, Reference Buchinsky, Valentino, Ruszkay, Powell, Derkay and Seedat29 Lele et al. suggested HPV 11 was more likely to predispose malignancy, but that finding is challenged by Carvalho et al. who reported no increased risk of cancer, pulmonary extension and equal remission length between HPV 6 and 11.Reference Carvalho, Huot, Charlois, Khalfallah, Chapuis and Froehlich9, Reference Lele, Pou, Ventura, Gatalica and Payne30
Current research in recurrent respiratory papillomatosis is centred on the preventative and adjuvant therapies to eradicate HPV. A national prospective study in Australia by Novakovic et al. Reference Novakovic, Cheng, Zurynski, Booy, Walker and Berkowitz6 reported a decline in the incidence on juvenile-onset recurrent respiratory papillomatosis following the implementation of the quadrivalent HPV vaccination programme for females aged 12–26 years. They concluded the vaccine (targeting HPV 6, 11, 16 and 18) reduced the risk of vertical transmission and subsequent development of juvenile-onset recurrent respiratory papillomatosis in children.Reference Novakovic, Cheng, Zurynski, Booy, Walker and Berkowitz6 In the UK, Gardasil 9 is offered to all children aged 12 and 13, and we therefore expect similar reduction in incidence in years to come. There is evidence that the HPV vaccination used as adjuvant treatment increases the inter-surgical interval, and the British Laryngological Association released a position statement in 2019 supporting the Gardasil HPV vaccination in patients with recurrent respiratory papillomatosis above the age of 9 years.Reference Donne and Kinshuck8, Reference Rosenberg, Philipsen, Mehlum, Dyrvig, Wehberg and Chirilǎ31, 32 Surprisingly, only 27 per cent of our patients have had the Gardasil HPV vaccine. Reasons for poor uptake have not been studied but contributing factors are likely to include lack of awareness and difficulty accessing or funding the treatment.
Novel therapeutic vaccines are in development with the aim to inactivate and cure recurrent respiratory papillomatosis. INO-3107, a HPV-specific DNA immunotherapy agent, elicits targeted T-cell response against HPV 6 and 11. A recent publication from a phase 1/2 trial of 21 patients reported over 75 per cent had a reduction in number of surgeries per year with a tolerable safety profile following INO-3107 administration.Reference Mau, Amin, Belafsky, Best, Friedman and Klein33
We conclude age of recurrent respiratory papillomatosis onset to be the only risk factor for dysplasia. In our large series of adult patients with recurrent respiratory papillomatosis, we found that gender, tobacco use, subglottic or tracheal disease involvement, number of surgeries, and cidofovir were not prognostic factors. Further studies should (1) examine the differences between low- and high-grade dysplasia to understand the significance, and (2) explore and address the poor HPV vaccination uptake as an adjuvant treatment, whilst awaiting trial results from novel DNA immunotherapy agents that may offer a cure to recurrent respiratory papillomatosis.
• Recurrent respiratory papillomatosis, caused by HPV 6 and 11, is deemed ‘low risk’ but can promote cell proliferation and transformation to dysplasia and carcinoma
• Although recurrent respiratory papillomatosis is a rare condition and data on outcomes and risk factors for dysplasia in the UK are limited, the rate of dysplasia in 123 patients was 11 per cent and one patient developed carcinoma-ex-papillomatosis
• An older age of recurrent respiratory papillomatosis onset was the only significant risk factor for dysplasia
• Gender, tobacco use, subglottic and/or tracheal involvement, number of surgeries and cidofovir were not risk factors
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
The authors declare none.
