Abstract
Introduction:
Recurrent respiratory papillomatosis (RRP) is a chronic disease of the upper respiratory tract caused by human papillomavirus types 6 and 11. The disease course is characteristically unpredictable, ranging from spontaneous remission to aggressive, recurrent disease. Thus, management is often challenging and requires unique approaches tailored to each individual patient. While recent literature has described risk factors for more aggressive disease, few sources have investigated the impact of smoking on RRP disease course and risk for malignant transformation.
Methods:
A retrospective chart review was conducted for adult RRP patients evaluated at an academic tertiary care center between 2005 and 2020. A total of 188 patients were identified. Demographic and clinical data were collected, including smoking and alcohol history, HPV subtype, history of dysplasia and/or carcinoma, voice handicap index scores, Derkay scores, debulkings (in office and operating room), and days to papilloma recurrence.
Results:
Malignant degeneration in RRP occurred in 16.3% of smokers and 3.6% of nonsmokers. Smokers who developed carcinoma had less debulkings per years of evaluation than those not developing carcinoma (0.21 vs 0.92, P = .004). Additionally, patients that either presented with or developed carcinoma during their course had a higher pack-year smoking history (18.0 vs 12.21, P = .0002). No difference in days to recurrence or inter-surgical interval was demonstrated between smokers and nonsmokers.
Conclusions:
The report demonstrates that smoking can increase the risk of malignant transformation in RRP patients.
Keywords
Introduction
Recurrent respiratory papillomatosis (RRP) is a chronic disease of the respiratory tract caused by human papillomavirus (HPV) types 6 and 11.1,2 Papillomas commonly appear as exophytic lesions arising from the larynx and more rarely, the nasopharynx, tracheobronchial tree, and pulmonary parenchyma. 1 Two types of RRP are recognized based on age of onset: juvenile- (JoRRP) and adult-onset (AoRRP). 2 The disease course for RRP is unpredictable, ranging from spontaneous remission to aggressive, recurrent disease.1,2 There is currently no cure for RRP, and management is often challenging and requires unique approaches tailored to each patient.1 -3
The primary goals of RRP management include (1) maintaining patency of airway, (2) improving voice quality, and (3) increasing time between surgical interventions. Due to variability of the disease, treatment options exist on a spectrum. Repeated surgical debulkings are typically required, and HPV vaccination is now often recommended.2 -5 For patients with high disease burden and frequent recurrence, adjuvant injections with intralesional cidofovir or bevacizumab may be used, and for the most advanced cases, IV therapies such as interferon or bevacizumab infusion may be required .2,3 In certain refractory and severe cases of RRP involving the airway, a tracheostomy may be required to avoid potentially fatal airway obstruction. 1
Risk factors for more aggressive disease have been studied widely but sometimes with disparate results, particularly in the adult population. Poor prognosis in JoRRP is fairly portended by younger age of onset, presence of HPV 11, firstborn vaginal delivery to a young mother, and lower socioeconomic status.6,7 In AoRRP, however, studies have yielded less consensus on investigated factors, including laryngeal irritants like tobacco use or laryngopharyngeal reflux, number of lifetime sexual partners or oral sex practices, oral HPV carrier status, male gender, presence of diabetes mellitus, or socioeconomic status.6 -12,14 -17
RRP poses a risk of malignant transformation, cited as <1% in children and ranging from 3% to 7% in adults in recent literature. 1 Identified risk factors have included infection with high-risk HPV subtypes (16 and 18), smoking, previous radiotherapy, cyto-toxic drug use, p53 gene mutation, and high severity score or high activity of 20-5′-oligoadenylate synthetase. 1 Additionally, low risk HPV subtypes 6 and 11, especially 11, have been shown to pose a risk for malignant transformation however the exact pathogenesis remains unclear. 1
While some studies have shown higher rates of dysplasia in smokers with RRP, they have not directly investigated smoking’s impact on morbidity, voice-related quality of life, recurrence rates, or transformation to carcinoma. 10 The primary outcome of this investigation was to determine correlation between malignant degeneration and smoking history. Additionally, we looked to describe the impact of current and former smoking on RRP clinical course and outcomes. Finally, our study focused on quantifying the point at which smoking history, quantified in pack-years, might continue to affect surgical outcomes and risk for malignant transformation.
Methods
A retrospective chart review was performed on all adult patients treated for RRP at The Ohio State University Wexner Medical Center from 2005 to 2020. Institutional Review Board (IRB) approval was obtained prior to data collection. International Classification of Disease (ICD) codes 9:212.1 and 10:D14.1 were used as the primary search criteria for RRP. Inclusion criteria included all patients >age 18 who were evaluated and treated for RRP at our institution: the majority with adult onset RRP. Patients with secondary pathologies such as vocal fold polyps or granulomas were excluded from the study. The primary outcome of this investigation was to determine correlation between malignant degeneration and smoking history. Malignant degeneration was defined as those with a history of recurrent respiratory papillomatosis who subsequently went on to develop malignancy of the airway as evidenced by biopsy. This included those who presented to our institution with a diagnosis of malignancy, including carcinoma in situ, with a history of previous RRP. It also included RRP patients who developed carcinoma while under our care.
Secondary outcomes included assessments of disease morbidity related to smoking, defined by the average days to recurrence, inter-surgical intervals, abbreviated Voice Handicap Index (VHI-10) scores, and Derkay scores. Recurrence was defined as increased or new growth of papilloma compared to previous exam, as evaluated with flexible laryngoscopy or laryngostroboscopy. “Surgery” was defined as either intra-operative debulking of the papillomatous lesions or in-office treatment using the potassium titanyl phosphate (KTP) laser.
The VHI-10 is a validated patient-centered scale that quantifies degree of voice impairment, and the Derkay score is a system for quantifying the amount of papillomatous disease burden. 13 VHI-10 scores were routinely obtained during clinic visits, and VHI-10 and Derkay scores were included if collected within 4 weeks before and or after an operation.
Demographic and clinical data were collected, including gender, age, history of dysplasia and/or carcinoma, ethnicity, smoking and alcohol history, type of RRP (JoRRP or AoRRP), surgery details and site (office or OR), and adjuvant utilization (Cidofovir or Bevacizumab), in addition to the data noted in the above paragraph. We separated groups into current smokers (CS, n = 29), former smokers (FS, n = 75), and never-smokers (NS, n = 84). We also analyzed groups by number of pack-years. Heavy alcohol use was defined as >7 drinks per week while moderate alcohol use was defined as 1 to 7 drinks per week.
Data Analysis
All analyses were performed through Microsoft Excel and R Studio. Comparison of means between groups was performed using Welch’s t-tests or Wilcoxon tests for 2-sample comparisons and ANOVA or the Kruskal-Wallis test for 3-sample comparisons. Fisher’s Exact Test was used for qualitative analyses of categorical values. Figures and tables were made in Microsoft Excel and RStudio. We used an alpha level of .05 for hypothesis testing. The data for this study is available upon reasonable request.
Results
The cohort consisted of 188 patients diagnosed with RRP prior to or upon presentation to an academic tertiary care center. Table 1 shows demographic characteristics of the cohort.
Demographic Information from the Analyzed Cohort Separated by Smoking Status.
Malignant degeneration in RRP occurred in 16.3% of current or former smokers and in 3.6% of never smokers (see Table 2). The group of former smokers averaged more pack-years (26.9 years) than the group of current smokers (24.5 years). Median observation/evaluation time significantly varied amongst current smokers, former, and never smokers (1.0 vs 2.8 vs 3.0 years, respectively, P = .03). For years under evaluation, smokers who developed carcinoma averaged less debulkings per year than those that did not develop carcinoma (0.21 vs 0.92, P = .004).
Recurrent Respiratory Papillomatosis (RRP) Data Separated by Smoking Status with Associated Comparisons.
Note. Categorical values were analyzed using the Fisher’s Exact Test while group means were compared using ANOVA or Welch’s t-test for 3- and 2-group comparisons, respectively. Kruskal-Wallis rank sum test and 2-sample Wilcoxon test were used to analyze pack history, average days to recurrence, and length of follow up as these distributions were right-skewed.
Abbreviations: AoRRP, adult onset RRP; JoRRP, juvenile onset RRP; RRP, recurrent respiratory papillomatosis; VHI, voice handicap index. p1, never smokers versus current smokers; p2, current smokers versus former smokers; p3, former smokers versus never smokers.
Current vs former smokers.
Never vs former smokers.
Only one value is available for this specific cohort.
The combination of former and never smokers together had more of an average VHI decrease per operation (−11.6) than current smokers (−5.25, P-value = .007).
Regarding the secondary objective of morbidity associated with smoking in RRP patients, current smokers had less improvement in VHI-10 score per operation (−5.25) than the combination of FS and NS (−11.6, P = .007). Patients with a pack-year history greater than or equal to 15, whether current or former smokers, had less improvement in VHI-10 score per operation than patients with less than 15 pack-years (−4.98 vs −15.9, P = .01); there were no other significant differences between those with <15 versus ≥15 pack-years (see Table 3). There was not enough available data to compare the change in Derkay score per operation among current smokers, former smokers, or nonsmokers; note the limitations described in Table 2. No significant differences among current smokers, former smokers, and nonsmokers were demonstrated in median days to recurrence (303.5 vs 186.0 vs 205.7, respectively, P = .96)
Recurrent Respiratory Papillomatosis (RRP) Data Separated by Average Pack Years (PY) with Associated P-Values.
Note. Categorical values were analyzed using the Fisher’s Exact Test while group means were compared using Welch’s t-test or 2-sample Wilcoxon test.
Abbreviations: RRP, recurrent respiratory papillomatosis; VHI, voice handicap index; AoRRP, adult onset RRP; JoRRP, juvenile onset RRP.
We controlled for factors such as alcohol use, age of onset, baseline VHI-10 or Derkay scores, and demographic factors. There were no significant differences among the 3 groups for any of these factors.
Discussion
This study is a first step in addressing the gap in knowledge about the role of tobacco use in RRP disease course in adults.6,7,11,14 -17 Our results show that smokers experience higher rates of malignant degeneration and have less improvement in VHI scores following intervention. These cancers were largely located in the larynx, as that’s where the majority of papilloma burden was. Each of these patients had biopsy proven RRP with later pathology showing malignancy. While these were biopsy proven papilloma, HPV testing was not routinely performed. This limited our ability to compare HPV subtypes prior to malignant degeneration. The length of time between benign disease and development of malignancy widely varied. The majority of patients had RRP for 5 to 10 years before malignant degeneration, though some were diagnosed within a few years of being at our facility. The exact amount of time was difficult to discern because many people were not diagnosed at this tertiary care center but transferred after an unknown amount of time.
This cohort of RRP patients was largely treated with excision by direct microlaryngoscopy with a potassium titanyl phosphate (KTP) crystal laser. Some patients also received adjuvant therapy with cidofovir or bevacizumab, though of the patients that developed malignancy, only 3 had previously been on adjuvant therapy for RRP. Of those with malignancy, many underwent total laryngectomy or chemoradiation with anti-neoplastic agents such as cisplatin.
Cigarette smoking and tobacco dependance remains the leading cause of preventable disease, death, and disability in the United States.18,19 As of 2019, 14.0% of adults over the age of 18 smoke cigarettes, translating to 34.1 million Americans. 18 Prevalence of former smokers in the United States is between 20% and 30%. 16 Our patient population consisted of 15.3% current smokers and 39.9% former smokers: rates higher than the national averages.
Similar to the negative effects tobacco can have on the heart and lungs, smoking also affects the vocal cords in a variety of well-defined ways, including malignancy, dysplastic changes, and polypoid corditis.20,21 In addition to these entities, smoking increases generalized inflammation of the laryngeal mucosa, affecting the voice in less discrete ways and providing an explanation for the worse VHI-10 outcomes after clearance of papilloma (Figure 1). Smoking also negatively impacts wound healing, thought to be due to the effect of toxins such as hydrogen cyanide, nicotine, and carbon monoxide on cellular function.22,23 Delayed wound healing could be another reason for worse VHI-10 outcomes after surgery and diminished improvement.
Boxplot detailing the differences in Average VHI change per operation comparing current smokers (CS) to former smokers (FS) and never smokers (NS).
Among existing literature examining the impact of smoking on RRP outcomes, few have examined the specific relationship between smoking and the risk of malignant degeneration.10,24,25 A study by Blumin et al investigated dysplasia rates in smokers with RRP, finding dysplasia present in 20% of RRP patients; however smoking was not determined to be an identifiable risk factor. 10 In a multicenter prospective study, 42 patients with RRP were followed to evaluate RRP disease course and potential reasons for malignant transformation. Of 5 patients who presented with malignant transformation during the study, 4 were smokers. However, the correlation between incidence of malignant transformation and smoking was not found to be statistically significant. 16 Our study is unique in demonstrating a significant correlation between smoking history (whether current or former) in patients with RRP and malignant degeneration to carcinoma. Additionally, our study demonstrated that smokers had worse voice outcomes following intervention. However, no statically significant difference in RRP recurrence rate was found between smokers, nonsmokers, or former smokers.
The failure to show a difference in recurrence rates could have occurred for other reasons as well. First, inter-surgical interval may be primarily driven by patient discretion and symptomology as opposed to environmental factors such as smoking. In a study by Derkay et al 26 a staging system was analyzed for prediction of surgical interval in RRP disease course. The study concluded that elements of the staging system, including Derkay score and stridor with activity, were effective in prediction of surgical interval; voice quality was not predictive. 26 Additionally, other factors like surgical technique or use of adjuvants may affect inter-surgical interval. Multiple studies have reported an increased inter-surgical interval among patients who receive adjuvant therapy compared who do not receive these treatments. 24 While we did control for adjuvant use, more subtle variations in surgical approach would not have been controlled.
Weaknesses of this study should be noted. While no increase in recurrence rate or difference in inter-surgical interval was found in smokers, it should be noted that smokers may accept greater degrees of dysphonia than non-smokers. This may keep them from presenting to the clinic as early when recurrence does occur. Routine intervals of observation were recommended for most patients reviewed in this study, which could prevent this error, but not all patients kept appointments when they were not sufficiently bothered by their voice quality. Another weakness was incomplete data on Derkay scores within an acceptable time interval before or after surgical intervention. We also note that our cohort did not include a sufficient number of patients with dyspnea or obstructive symptoms to examine the role of smoking in relation to these factors.
Conclusion
The report demonstrates that smoking can increase the chance of malignant transformation in RRP patients and lead to a more severe clinical course, requiring additional surgical interventions.
Footnotes
Authors’ Note
All authors listed significantly contributed to multiple aspects of the manuscript submitted including design, data collection, data analysis, and manuscript preparation and review
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
