Implementation of Routine HPV Vaccination in the Management of Recurrent Respiratory Papillomatosis

Introduction

Recurrent respiratory papillomatosis (RRP) is a disease characterized by papillomatous growths of the upper aerodigestive tract mucosa, predominantly affecting the larynx. The estimated incidence of RRP is approximately 4 per 100,000 in children and 2 per 100,000 in adults. ¹ It is caused by a DNA virus called the human papillomavirus (HPV), most commonly of subtypes 6 and 11, which infiltrates and proliferates within epithelial tissue. The disease can manifest in either childhood or adulthood, and the range of clinical severity is wide, with some experiencing malignant transformation to squamous cell carcinoma (SCC) and others seeing spontaneous remission. ² Many patients with RRP require frequent procedures to relieve airway obstruction and/or to rehabilitate voice.^3,4 Though surgical excision is the gold standard, there is no consensus on indications for surgery or optimal technique. ⁵ It is, however, widely accepted that achieving a consistent cure for RRP, especially in aggressive forms of the disease, will be nearly impossible without adjuvant therapy targeting the underlying cellular and immunologic processes. Some adjuvant treatments include interferon, intralesional cidofovir (Vistide), intralesional or systemic bevacizumab (Avastin), and more recently the prophylactic HPV vaccine (Gardasil).^2,3,5,6 Experts recommend consideration of adjuvant therapy in addition to surgical excision when patients require >4 procedures per year, have distal multisite spread, and/or demonstrate rapid regrowth with airway compromise.^3,7

The Centers for Disease Control and Prevention (CDC) has an Advisory Committee on Immunization Practices (ACIP) that recommends universal prophylactic HPV vaccination for all males and females at age 11 or 12 years. They also promote catch-up vaccination for females through age 26 and males through age 21 who missed vaccination at age 11 to 12.^8-10 The ACIP does not provide recommendations on vaccination over the age of 26 years or regarding therapeutic vaccination. Since most insurance companies base coverage of the HPV vaccine on CDC guidelines, they will not routinely cover immunization for adults affected by RRP if >26 years.

Several authors have described severe cases of refractory RRP, some with distal tracheobronchial or esophageal involvement, that responded dramatically to HPV vaccination.^11,12 Larger case-control and cohort studies have proposed that HPV vaccination can potentially modulate disease severity and even induce remission.^13-17 Due to these compelling findings, we elected to investigate this in our population of RRP patients. In addition, we sought to explore the effectiveness of patient education and financial assistance in encouraging HPV vaccination. The study objectives are to evaluate vaccine compliance rates and clinical outcomes after implementation of an office-based initiative to routinely provide the Gardasil HPV vaccine to patients with RRP. Vaccination rates and disease recurrence data were compared pre- and post-initiative and before and after vaccine series completion, respectively.

Materials and Methods

After obtaining institutional review board approval, we queried our electronic medical record for patients treated for diagnoses of RRP or laryngeal papillomatosis by the senior author (L.M.) from January 1, 2012, to December 31, 2017. We performed a retrospective chart review, collecting demographic, clinicopathologic, and treatment data, including age, sex, measures of disease burden, dates of office visits and surgical procedures, vaccine administration dates, and length of follow-up. We included surgeries performed in the operating room (microlaryngoscopy with potassium titanyl phosphate (KTP) laser ± microdebrider-assisted excision) and in-office procedures (awake flexible laryngoscopy with KTP laser ablation). We also noted use of adjuvant therapies such as cidofovir or bevacizumab.

We analyzed the entire cohort to compare vaccine rates before and after the start of the HPV vaccine initiative. The initiative began in December 2015 and involved a collaborative effort to streamline the process of providing the HPV vaccine to all amenable patients with RRP. First, our senior author began to consistently recommend the HPV vaccine for all unvaccinated RRP patients. This included actively counseling patients during office visits and providing handouts on the Gardasil vaccine. If patients agreed to vaccination, our specialty pharmacy appealed to their insurance companies for coverage of vaccine cost (Suppl. 1). If initial efforts were declined, we sent insurance companies a letter with literature supporting the therapeutic use of HPV vaccination in RRP (Suppl. 2). For uninsured patients and those who did not qualify for insurance coverage despite appeal, we obtained approval from hospital administration to provide the series of 3 HPV vaccines at cost with no additional facility or administration fees. Four patients received the quadrivalent Gardasil vaccine before 2016, while the remaining 10 patients vaccinated after 2016 received the Gardasil 9 vaccine.

We compared demographic and clinical variables between patients who had received the HPV vaccine and those who had not. We then performed a more detailed analysis of the 14 patients who had completed the HPV vaccine series. We used the date of final vaccine administration to separate disease courses into pre- and post-vaccine periods. We calculated the number of procedures performed per year and the interval of time between procedures (intersurgical interval or ISI) before and after completing the vaccine series. Remission was defined as no evidence of papilloma regrowth for >12 months during reliable clinical follow-up, though we recognize the persistent potential for disease reactivation and do not imply any patients were cured of disease.

The Fisher exact test was used to compare categorical variables, and the Mann-Whitney U test was used to compare continuous variables between the vaccinated and unvaccinated groups. We used the Fisher exact test to compare vaccination rates before and after the vaccine initiative. Paired t tests were used to compare the ISI and number of procedures per year before and after vaccine series completion. P values less than .05 were deemed clinically significant. Statistical analyses were performed using GraphPad Prism software.

Results

We identified 42 patients treated by the senior author (L.M.) for RRP between January 1, 2012, and December 31, 2017. The median age of the cohort was 48 years, comprising 33 males (78.6%) and 9 females (21.4%). Thirty-eight of the 42 patients (90.5%) were newly diagnosed with adult-onset RRP, while 4 patients (9.5%) had an existing diagnosis of juvenile-onset RRP without any intervention within the past 5 years. The median length of follow-up was 51 months (Table 1). Reliable long-term follow-up (>24 months) was available for 12/14 vaccinated patients and 14/28 unvaccinated patients. Three patients (7.14%), all nonvaccinated, developed malignant transformation of RRP into SCC. Two of these patients had other risk factors for SCC including alcohol and tobacco abuse.

OPEN IN VIEWER

Table 1.

Demographic and Clinicopathologic Characteristics of Patients with Recurrent Respiratory Papillomatosis. ^a

	All Patients (N = 42)		Vaccinated for HPV (n = 14)		Unvaccinated (n = 28)		P
Median age, y	48	(41-63)	46	(34-56)	51	(43-65)	.11
Male gender	33	(78.6%)	10	(71.4%)	23	(82.1%)	.70
Median length of follow-up, mo	51	(14-101.1)	81.3	(49.4-106.4)	30.1	(8.6-89.7)	.025
Measures of disease burden
Baseline anatomic median Derkay score	7	(3-9)	7	(2.75-7.75)	7	(3-9)	.65
Anterior commissure involvement	23	(54.8%)	7	(50%)	16	(57.1%)	.75
Vocal fold medial edge involvement	39	(92.9%)	14	(100%)	25	(89.3%)	.54
Mean total no. of procedures	3.14	(3.2)	4.43	(3.30)	2.5	(3)	.022
Mean baseline intersurgical interval, mo	3.2	(3.88)	3.49	(3.44)	3.06	(4.14)	0.75
Intralesional cidofovir	10	(22.7%)	6	(42.9%)	4	(14.3%)	.059 b
Intralesional bevacizumab	5	(11.4%)	3	(21.4%)	2	(7.14%)	.31
Remission	5	(20.7%)	5	(41.7%)	0	(0%)	.015

a

Results are expressed as frequencies (column percentage) for categorical variables and median (interquartile range) or mean (standard deviation) for continuous variables. P values are from Fisher exact tests for categorical variables and Mann-Whitney U tests for continuous variables. Remission rates were calculated from only 26 patients who had reliable follow-up data. Bolded P values achieved statistical significance.

b

P values approached significance.

Comparing the 14 patients who received the HPV vaccine to the 28 patients who did not, there were no significant differences in age (P = .11) or sex (P = .45). Measures of disease burden including baseline anatomic Derkay score and involvement of the anterior commissure and/or medial edge of the vocal folds were not significantly different between the vaccinated group (prior to vaccination) and the unvaccinated group. The mean number of procedures performed overall was higher in the vaccinated group (4.4 vs 2.5, P = .02). The mean baseline pre-vaccine ISI in the vaccinated group was 3.5 months compared to 3.06 months in the unvaccinated group (P = .75). There were no significant differences in rates of adjuvant bevacizumab administration between the vaccinated and unvaccinated groups (21.4% vs 7.14%, P = .31). The differences in intralesional injection rates of cidofovir approached clinical significance (42.9% vaccinated vs 14.3% unvaccinated, P = .059).

Of the 14 patients who were vaccinated, 11 patients received their vaccines after the initiative began in December 2015. The pre-initiative vaccination rate of all active established patients in the practice with RRP was 9.7% (3/31 patients); post-initiative rates improved to 43.8% (14/32 patients; P = .004, odds ratio, 7.26) (Figure 1). Two additional patients received the first Gardasil injection but had not completed the series. Appeals to insurance benefits for coverage of vaccine cost were successful in 6 of the 11 patients vaccinated after December 2015. Remaining patients paid an out-of-pocket cost of approximately $190 per injection. There were no reported side effects due to vaccination.

OPEN IN VIEWER

Figure 1.

Human papillomavirus (HPV) vaccination rates in active recurrent respiratory papillomatosis patients in L.M.’s practice pre- and post-initiative (December 2015).

Of the 14 vaccinated patients, there were significant differences between mean pre-vaccine ISI (3.5 months) and post-vaccine ISI (12.8 months) (P = .0021), such that patients averaged 9.3 months longer between procedures after completing their vaccine series (Figure 2). Similarly, there was a significant decrease in average number of surgical procedures per year before and after vaccination (2.7 vs 0.81; P = .014) (Figure 3). Of note, 6 vaccinated patients (42.9%) also received prior or concomitant adjuvant therapy with cidofovir, bevacizumab, or both (Table 2). Two of these patients (#2 and #7) received only a single injection of intralesional cidofovir. Two other patients (#3 and #13) had severe distal tracheobronchial involvement, present prior to vaccination, and were comanaged with a pulmonologist. Patient #13 was the only patient in our cohort to receive systemic bevacizumab therapy. Five of 11 patients (45.5%) who had an increase in ISI and 6 of 11 patients (54.5%) who had a decrease in number of procedures per year after vaccination had not received any other adjuvant therapy. Five of the vaccinated patients, 2 of whom had not received cidofovir or bevacizumab, attained remission after Gardasil vaccination and are still followed clinically. Two vaccinated patients were lost to follow-up after their last surgical procedure. The remaining 7 of 14 vaccinated patients continue to be followed closely for active disease.

OPEN IN VIEWER

Figure 2.

Change in intersurgical intervals for recurrent respiratory papillomatosis procedures in months pre- and postcompletion of 3-injection human papillomavirus vaccine series.

OPEN IN VIEWER

Figure 3.

Change in number of procedures performed per year for recurrent respiratory papillomatosis pre- and postcompletion of 3-injection human papillomavirus vaccine series with standard deviation.

OPEN IN VIEWER

Table 2.

Patients with RRP Vaccinated with Gardasil.

Patient No.	Pre-ISI, mo	Post-ISI, mo	Δ in ISI, mo	Pre-, No./Year	Post-, No./Year	Δ in No./Year	Adjuvant Therapy, No.	Remission	Length of Follow-up, mo
1	3.26	7.76	4.5	2.74	1.96	−0.78	None	Yes	74.08
2	12.47	20.23	7.77	0.77	0.83	0.06	C(1)	No	110.33
3	1.29	6.03	4.74	3.87	2.17	−1.7	C(5)+B(3)	No	44.42
4	-	-	-	1.88	0	−1.88	None	No	103.5
5	1.25	9.89	8.64	10.74	1.23	−9.51	C(3)+B(4)	Yes	119.83
6		10.68	-	0	0.73	0.73	None	Unknown	105.08
7	5.69	11.87	6.18	1.93	0	−1.93	C(1)	Yes	88.42
8	2.45	14.17	11.72	1.53	1.55	0.02	None	Unknown	94.5
9	2.48	19.5	17.02	3.5	0	−3.5	C(2)	Yes	51
10	5.47	27.57	22.1	1.63	0	−1.63	None	Yes	19.83
11	1.97	16.03	14.06	3.08	0	−3.08	None	No	33.25
12	1.97	7.37	5.4	2.32	2.18	−0.14	None	No	183.17
13	0.12	24.4	24.28	2.32	0.74	−1.58	C(2) +systemic B	No	65.92
14	6.97	0.87	−6.1	1.53	0	−1.53	None	No	61.92

Abbreviations: B, bevacizumab; C, cidofovir; ISI, intersurgical interval; No./year, number of procedures per year; Post, after receiving the third vaccine in the series; Pre, before completing the vaccine series; RRP, recurrent respiratory papillomatosis.

Discussion

Though RRP is a benign disease with low risk of malignant degeneration, its sometimes unrelenting clinical course and heavy treatment burden negatively impact patients’ quality of life and have major financial implications for individuals and the health care system.^3,18 High rates of disease recurrence despite aggressive intervention have inspired extensive research efforts to identify adjuvant treatments to surgical excision that can synergistically improve outcomes.

The HPV vaccine is one of the newer therapies to be explored in the prevention and treatment of RRP. Newer research on the upper aerodigestive manifestations of HPV has benefited from the framework established by studies on prophylactic and therapeutic uses of vaccination in the genitourinary manifestations of HPV. Multicenter randomized controlled trials have been instrumental in establishing the prophylactic efficacy of the HPV vaccine in preventing genitourinary disease. The international FUTURE I/II trials showed that quadrivalent HPV vaccination provided sustained protection of 96% to 100% against anogenital condylomas and low-grade cervical and vulvovaginal neoplasias caused by HPV types 6, 11, 16, and 18. ¹⁹ Complementing the established research on genital immunity, Ahn et al ²⁰ were the first to show that the vaccine was also effective in inducing oral immunity against HPV16 in a murine model.

The therapeutic potential of the current vaccine in genitourinary disease was first explored in a retrospective review from Korea published in 2013 that evaluated 737 patients with high-grade cervical intraepithelial neoplasia (CIN 2-3) treated with loop electrosurgical excision procedure (LEEP). ²¹ Approximately half of their patients received the HPV vaccine after LEEP and then had significantly lower rates of recurrence compared to the unvaccinated group (2.5% vs 7.2%). In the RRP literature, several studies on the therapeutic capacity of the prophylactic vaccine in preventing recurrence also exist. A European group followed 11 adult patients with aggressive forms of laryngeal papillomatosis who were vaccinated with Gardasil. ¹⁶ They noted an increase in ISI from 271.2 to 537.4 days and a decrease in mean number of surgeries per year from 2.16 to 0.93. Complete remission was seen in 1 patient, and 7 patients had partial remission. Young and colleagues ¹⁵ performed a review of 17 patients with RRP and found a significant increase in ISI of 3.1 months after HPV vaccination. In their cohort, 8 patients experienced complete remission and 5 patients experienced partial remission. A prospective cohort study from Romania offered the HPV vaccine to 13 patients whose RRP recurred despite serial excisions with intralesional cidofovir injection. They achieved complete remission in 11 patients (85%) at 1-year follow-up. ¹⁴ A systematic review by Dion et al ²² established that 9 out of 12 studies (including the 3 cited above) examining therapeutic use of Gardasil in patients with active clinical disease demonstrated decreased disease recurrence, decreased disease burden, or increased ISI. Our analysis parallels the above studies as our patients had increased ISI of 9.3 months and required 1.89 fewer average procedures per year after completing the HPV vaccination series. In our group, 5 of 14 vaccinated patients achieved remission.

Despite these promising conclusions, the HPV vaccine has not been accepted widely as a therapeutic option. However, research to establish an immunologic mechanism for its therapeutic implications is ongoing. One study of 6 patients with RRP demonstrated that in vivo seroreactivity to the virus remained unexpectedly low at baseline despite chronic HPV infection and rose significantly after vaccination. ²³ Makiyama and colleagues ²⁴ also found that men with active RRP from HPV infection saw a substantial increase in serum antibody titers after Gardasil injection. The hypothesis we share is that despite being a prophylactic vaccine, Gardasil triggers a stronger cell-mediated and/or antibody-mediated immune response to HPV that was not activated by HPV infection itself and therefore can be used in a therapeutic fashion in patients with previous or active HPV infection. Interestingly, Ahn et al ²⁵ have started investigating immunologic responses to a new therapeutic vaccine that encodes the HPV11 E6 and E7 genes rather than the L1 capsid protein contained within the existing Gardasil vaccines. Their novel vaccine generated a vigorous CD8+ T-cell response against the HPV11 E6 peptide that translated to lower tumor growth rate and tumor volumes in vaccinated mice. We anticipate that interest in therapeutic vaccines for RRP will heighten as the medical community continues to seek a cure for this devastating disease.

Overall HPV vaccination rates in the United States have slowly increased since FDA approval of the Gardasil vaccine in 2006, with current national rates of approximately 60% in male and female adolescents. ⁸ One of the most remarkable HPV vaccine efforts was carried out in Australia, where the government initiated a free school-based vaccination program in 2007 that resulted immediately in high vaccination rates of 73.8% in females by 2007, with sustained rates of 78.6% in females and 72.9% in males by 2016. National surveillance data in Australia have depicted substantial declines in the incidence of anogenital warts in females and males after the 2007 initiative. ²⁶ Wangu and Hsu ²⁷ performed a systematic review of all such population-level surveillance studies after introduction of the HPV vaccine and found a 31% decrease in prevalence of anogenital warts across multiple countries. Their literature search identified no population-level studies on the impact of HPV vaccination on RRP, but they suggest that these studies are warranted and that results may mirror the findings of anogenital warts.

Given the minimal risks and potential benefits associated with the HPV vaccine, we began to consistently offer vaccination to all RRP patients after December 2015. These efforts led to a significant increase in vaccination compliance, demonstrating that patients are receptive to vaccination after receiving adequate counseling. As the CDC ACIP does not endorse vaccination for adults >26 years, citing a lack of population-level impact, insurance plans will not routinely cover the Gardasil vaccine despite evidence suggesting it may reduce disease burden of RRP on an individual level. We successfully implemented an algorithm through our specialty pharmacy to petition insurance companies to consistently cover vaccine costs for RRP patients (Suppl. 1). These efforts can be easily replicated by other providers who wish to offer the HPV vaccine to their RRP patients at subsidized or no cost. Yet we must continue to bolster the existing literature such that the CDC ACIP may eventually endorse the therapeutic use of HPV vaccines for RRP. This would standardize insurance benefits to cover HPV vaccination in this special population and relieve patients of a prohibitive financial burden.

Conclusion

Our study shows that coordinated efforts to provide counseling and financial support for the HPV vaccine can increase vaccination rates in patients with RRP. We present our findings to assist other practices looking to increase their vaccination rates and to strengthen the existing literature on the potential benefits of therapeutic use of the HPV vaccine. As more of the overall population is vaccinated in adolescence, we are optimistic that incidence of RRP may decline in a manner similar to population-based studies of other HPV-mediated diseases. For now, we believe that the current prophylactic HPV vaccine should be endorsed as a therapeutic measure by the CDC to standardize insurance coverage of vaccine costs for patients being treated for RRP as research continues toward the formulation of a true therapeutic vaccine.

Supplemental Material

Supplemental Material