IJC International Journal of Cancer

Transcription

1 IJC International Journal of Cancer Randomised trial on treatment of vaginal intraepithelial neoplasia Imiquimod, laser vaporisation and expectant Karoliina Tainio 1, Maija Jakobsson 1, Karolina Louvanto 1,2, Ilkka Kalliala 1,3, Jorma Paavonen 1, Pekka Nieminen 1 and Annika Riska 1 1 Obstetrics and Gynecology, University of Helsinki and Helsinki University Hospital 2 Obstetrics and Gynecology, Turku University Hospital 3 Imperial College London, Institute of Reproductive and Developmental Biology, Department of Surgery & Cancer Vaginal intraepithelial neoplasia (VAIN) is associated with human papillomavirus (HPV) infection. The most common treatment modality is laser vaporisation, but recurrences are common. Imiquimod is an immune response modulator which is used for the treatment of external condylomas and other HPV-related genital neoplasias. The aim of the study was to evaluate the efficacy and tolerability of vaginally administered imiquimod in comparison with laser vaporisation and expectant of high-grade VAIN. This proof of principle pilot study was a prospective 16-week randomised trial. We enrolled 30 patients with histologically confirmed VAIN 2 or 3 into three study arms: vaginally administered imiquimod, laser vaporisation and expectant. Follow-up colposcopy visits included high-risk human papillomavirus (hrhpv) testing, cytology and punch biopsies. At baseline 77% (n 5 20/26) of the patients were hrhpv positive. HPV clearance was significantly higher in the imiquimod arm (63%, n 5 5/8) than in the laser arm (11%, n 5 1/9) (p ) or in the expectant arm (17%, n 5 1/ 6) (p ). At baseline 25 patients (83%) had VAIN 2 and five (17%) had VAIN 3. None of the lesions progressed during the follow-up. Histological regression (VAIN 1) was observed in 80% (n 5 8/10) of patients in the imiquimod arm, 100% (n 5 10/10) of the laser arm (p ) and 67% (n 5 6/9) of the expectant arm (p ). Vaginal imiquimod appears to be as effective as laser treatment in high-grade VAIN. Introduction Vaginal intraepithelial neoplasia (VAIN) is a relatively rare premalignant lesion. It has an annual incidence of cases per 100,000 women and is strongly associated with human papillomavirus (HPV) infection. 1,2 VAIN is Key words: vaginal intraepithelial neoplasia, vaginal high-grade squamous intraepithelial lesion, imiquimod, HPV, human papillomavirus Abbreviations: CIN: cervical intraepithelial neoplasia; HPV: human papillomavirus; HrHPV: high-risk human papillomavirus; HRT: hormone replacement therapy; HSIL: high-grade squamous intraepithelial lesion; LEEP: loop electrosurgical excision procedure; LSIL: low-grade squamous intraepithelial lesion; NSAID: non-steroidal anti-inflammatory drug; VAIN: vaginal intraepithelial neoplasia; VIN: vulvar intraepithelial neoplasia Additional Supporting Information may be found in the online version of this article. DOI: /ijc History: Received 11 Apr 2016; Accepted 1 July 2016; Online 18 July 2016 Correspondence to: Dr. Karoliina Tainio, Department of Obstetrics and Gynaecology, University of Helsinki and Helsinki University Hospital, PO Box 140, FI Helsinki, Finland, Tel.: , Fax: , karoliina.tainio@hus.fi significantly less common than other HPV-related precancerous genital lesions such as cervical intraepithelial neoplasia (CIN) or vulvar intraepithelial neoplasia (VIN). 3 In the updated WHO 2014 classification VAIN lesions are graded as vaginal low-grade squamous intraepithelial lesion (LSIL) and vaginal high-grade squamous intraepithelial lesion (HSIL) with LSIL including VAIN 1 and HSIL VAIN 2 and 3. 4 VAIN 1 can also be considered a productive HPV infection and has a spontaneous regression rate of over 50% and therefore it can be expectantly managed. 5 7 VAIN 2 and 3 are considered precancerous and VAIN 3 has a relatively high risk of progression to carcinoma in approximately 11 13% of cases. 3,7 The incidence of vaginal cancer has been increasing in Finland and elsewhere. 8 VAIN is often multifocal and localised in the upper third of the vagina. 5,9 Patients with VAIN are predominantly perior postmenopausal and concomitant or previous CIN or VIN is common. 10,11 VAIN patients often have history of cervical cancer or have immunosuppression. 3 Recurrent VAIN lesions are common and require repeat treatments which may be mutilating and cause vaginal scarring. 12,13 The most common contemporary treatment for VAIN is laser vaporisation. Previous treatments for CIN or a previous hysterectomy make it more difficult to see the lesions, take biopsies and also to treat the patients with laser or surgical approaches.

2 2354 Randomised trial on treatment of vaginal intraepithelial neoplasia What s new? When precancerous vaginal intraepithelial neoplasias develop, the usual treatment is laser vaporisation but they tend to return. These authors wondered whether imiquimod might do a better job eradicating these HPV-related lesions. Thirty patients received either imiquimod, laser treatment or no intervention. None of the lesions progressed; with no intervention, 6 of 9 patients saw their neoplasias spontaneously recede. Laser treatment and imiquimod achieved comparable results: 100% regression with laser, and 80% with the drug. Imiquimod performed considerably better against HPV more than half of imiquimod patients were rid of HPV, while all but one of the laser patients kept their HPV infection after treatment. Imiquimod is an immune response modulator, which has been effective and safe when applied topically as 5% cream in the treatment of external condylomas. 14 Imiquimod is a tolllike receptor 7 agonist which activates local immune response through dendritic cell activation and cytokine release resulting in activation of innate and T-helper cell immunity. 15 Previous small non-randomised studies of the treatment of VAIN with imiquimod have been promising. 16,17 There are also ongoing imiquimod studies on other HPV-related lesions CIN and VIN The aim of this pilot study was to assess the efficacy and tolerability of self-administered vaginal imiquimod compared with conventional laser vaporisation and expectant. Our hypothesis was that a treatment targeting on HPV would be more effective and decrease the risk for recurrent disease and repeat treatment episodes. Material and Methods This study was a prospective 16-week randomised controlled three-arm proof of principle pilot trial in a single tertiary centre at the Women s Hospital s Outpatient Colposcopy Clinic, Helsinki University Hospital, Finland. Eligible patients were enrolled in the study between December 2012 and May 2015 and followed until September The main outcome measure was to evaluate the rate of histological regression (VAIN 1 or less). Secondary outcome measures were complete histological regression (no dysplasia), HPV clearance and tolerability. Eligible patients for the study were women over 18 years of age diagnosed with histologically confirmed VAIN 2 3 or persistent VAIN 1, which had been expectantly managed for at least 2 years previously. However, no patients with VAIN 1 were ultimately enrolled in the study. Exclusion criteria were concomitant diagnosis of CIN 21 requiring loop electrosurgical excision procedure (LEEP), lack of reliable contraception in premenopausal patients, pregnancy or lactation, allergy to imiquimod, vaginal cancer and known HIV infection. Study visits were at 4, 8 and 16 weeks after the enrolment visit (Table 1). All clinical investigators were experienced consultants and colposcopists or gynaecological oncologists. The study was approved by Helsinki University Hospital s Ethical Committee (385/13/03/03/2012) and the Finnish Medicines Agency Fimea (EudraCT ). The study is registered in the ISRCTN Registry (ISRCTN ). Patients who gave written informed consent were randomised 1:1:1 by computer-assisted permuted-block randomisation with random block sizes of four to six. The investigators did not participate in the randomisation process. Allocation concealment was achieved by sequentially numbered sealed opaque envelopes. The three arms were vaginally administered imiquimod, laser vaporisation and expectant. Because only small, non-randomised, studies on the effect of imiquimod in VAIN lesions exist, we were not able to perform power calculations to reach statistically representative sample size for primary or secondary outcomes. We enrolled ten women in each arm in this proof of principle pilot study. Imiquimod was self-administered by the patients as vaginal suppositories for a period of 8 weeks with one dose of 12.5 mg imiquimod per week during the first 2 weeks, and 12.5 mg doses twice weekly 3 4 days apart in the remaining 6 weeks (14 doses in total). Imiquimod was manufactured as suppositories with inactive binding materials (macrogol g and macrogol g) by a collaborating pharmacy and the patients received all suppositories from the investigators with written and oral instructions. Patients were instructed to apply the suppositories in the evening before bedtime and to store the suppositories at room temperature. A diary for recording any side effects was given. Patients were instructed to halve the suppositories longitudinally and continue the treatment with half a dose (6.25 mg) if side effects were not tolerable after paracetamol or NSAID (nonsteroidal anti-inflammatory drug) medication. In the laser arm carbon dioxide laser vaporisation (6 12 W, continuous beam, by the discretion of the colposcopist) was applied to all visible lesions with 2 mm margins and to a depth of 2 mm. Patients in the expectant arm were followed without any intervention except punch biopsies. Colposcopy was performed with 5% acetic acid with or without Lugol s iodine solution depending on the preference of the colposcopist. Punch biopsies for histology were obtained at 8-week-visit if any new lesions were suspected (Table 1). Patients with persisting VAIN 2 3 at the 16-weekvisit were treated with laser vaporisation or surgical excision regardless of study arm. The institution s pathologists,

3 Tainio et al Table 1. Study visits in all study arms in the randomised trial on treatment of vaginal intraepithelial neoplasia hrhpv Colposcopy Cytology Histology test 0 weeks, enrolment 4 weeks 8 weeks () 16 weeks, exit Abbreviation: hrhpv test, high-risk human papillomavirus test. Table 2. Characteristics of 30 women participating in the randomised trial on treatment of vaginal intraepithelial neoplasia Imiquimod Laser Expectant Age (years) Median Range Previous VAIN (n) Previous VIN (n) Previous CIN (n) Previous cervical cancer (n) Hysterectomy (n) Total Supravaginal Immunosuppressive medication (n) Current tobacco use (n) Current systemic HRT (n) Current local HRT (n) Differences between any two groups did not reach statistical significance (p < 0.05) for any of the baseline explanatory variables studied and listed above. Abbreviation: HRT, hormone replacement therapy. Flow diagram 1. Progress of the randomised trial on treatment of vaginal intraepithelial neoplasia. specialised in gynaecological histopathology, reviewed both the cytological and the histological samples. During the study, VAIN lesions were still classified according to WHO 2003 classification. 21 High-risk human papillomavirus (hrhpv) testing at enrolment and at exit was performed with Hybrid Capture II (Digene Corporation, Gaithersburg, MD), which tests for HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68. The key baseline characteristics and baseline histological and hrhpv results of the three patient groups were compared. In addition, comparisons between two groups together were done (imiquimod vs. laser, imiquimod vs. expectant and laser vs. expectant ). The histological, cytological and HPV results at 16 weeks follow-up were compared. Regression of the disease was defined as follows: complete histological regression as regression to no dysplasia Table 3. Baseline histological and HPV results of the randomised trial on treatment of vaginal intraepithelial neoplasia Imiquimod Laser Expectant HPV, n/total, (%) Positive 8/8 (100) 7/9 (78) 5/9 (56) Negative 0/8 (0) 2/9 (22) 4/9 (44) Histology, n VAIN VAIN Cytology, n Normal ASC-US LSIL ASC-H/HSIL Focality of VAIN, n Unifocal Multifocal Differences between any two groups did not reach statistical significance (p < 0.05) for any of the variables studied and listed above. Abbreviations: ASC-US, atypical squamous cells of undetermined significance; LSIL, low-grade squamous intraepithelial lesion; ASC-H, atypical squamous cells, cannot exclude HSIL; HSIL, high-grade squamous intraepithelial lesion.

4 2356 Randomised trial on treatment of vaginal intraepithelial neoplasia Table 4. Histological and HPV results at 16 weeks of the randomised trial on treatment of vaginal intraepithelial neoplasia Imiquimod Laser Expectant (n 5 9) p Values 1 HPV, n/total, (%) Positive 4/10 (40) 6/10 (60) 4/9 (44) Negative 6/10 (60) 4/10 (40) 3/9 (33) Cleared HPV 5/8 (63) 1/9 (11) 1/6 (17) 0.05 Persistent HPV 3/8 (38) 6/9 (67) 3/6 (50) Histology, n Normal VAIN VAIN Complete regression Partial regression Regression of disease (any) Persistent disease Cytology, n Normal ASC-US LSIL ASC-H/HSIL Cytological regression Statistical analyses between the imiquimod and laser groups. Analyses between the imiquimod and expectant and laser and expectant did not reach statistical significance (p < 0.05) in any of the categories studied here. 2 Defined as regression to <VAIN 1. 3 Defined as regression to VAIN 1. 4 Defined as regression to normal cytology. Abbreviations: ASC-US, atypical squamous cells of undetermined significance; LSIL, low-grade squamous intraepithelial lesion; ASC-H, atypical squamous cells, cannot exclude HSIL; HSIL, high-grade squamous intraepithelial lesion. (<VAIN 1), and partial histological regression to VAIN 1. Cytological regression was defined as regression to normal cytology. Frequency tables were analysed using the v 2 -test or Fisher s exact test for categorical variables. Differences in the means of continuous variables were analysed using nonparametric (Kruskal Wallis) tests for two and multiple independent samples, respectively. Statistical analyses were done with STATA/SE 13.1 (StataCorp, College Station, TX). Results Progress of the study is presented in Flow diagram 1. The results are based on an intention-to-treat analyses. The median age of the 30 women was 54 years (range 31 82). Patient characteristics are presented in Table 2. There were no differences between the study arms by any of the characteristics considered. Half of the patients had previously been treated for one or multiple HPV-related conditions: 37% (n 5 11/30) had been treated for VAIN, 10% (n 5 3/30) for VIN and 23% (n 5 7/30) for CIN. Four patients (13%) had a history of cervical cancer treated more than 5 years ago. Four patients had concomitant CIN 1 and one patient had concomitant VIN 3. The VAIN lesions were multifocal (two or more separate vaginal lesions) in 63% (n 5 19/30) (Table 3). An additional more detailed description of each individual s characteristics and outcomes are shown in Supplementary material Table 1. At baseline 77% of the patients (n 5 20/26) were hrhpv positive. HPV clearance at 16 weeks was significantly higher in the imiquimod group (63%) compared with 11% in the laser group (p ) and 17% in the expectant group (p ). HPV status was missing in six patients at 16 weeks (Table 4). At baseline 25 patients (83%) had VAIN 2 and five (17%) had VAIN 3 (Table 3). None of the lesions progressed during the follow-up period. Histological regression was common in all of the groups with no differences in the regression rates between the study arms (Table 4). In the expectant group untreated patients (n 5 5) had all VAIN 2 at baseline. Two patients had persistent VAIN 2 at 16 weeks and were treated with laser vaporisation. Of the untreated patients three out of five had complete histological regression after the followup period. Baseline cytological results varied in all study groups

5 Tainio et al with cytology ranging from normal to high-grade squamous intraepithelial lesion (HSIL) (Table 3). At 16 weeks cytological regression to normal was seen in nearly half of the patients, ranging between 33% and 55% in the three arms (Table 4). All patients in the imiquimod group reported side effects from the treatment, but none discontinued treatment. Only one patient halved the imiquimod dose during the treatment period. Most common side effects were fever and flu-like symptoms (n 5 9), local irritation in the vagina and vulva (n 5 6) and lower abdominal pain (n 5 3). The side effects and fever up to 398C occurred within 12 hr of application. None needed further evaluation for the fever as it rapidly subsided or was alleviated with paracetamol or NSAID medication. Other side effects similarly did not require additional evaluation. Local estrogen gel was recommended to two patients because of vulvar irritation. One patient had a vulvar ulceration at the 4- and 8-week-visit, but this healed by the exit visit at 16 weeks. Discussion This is the first randomised prospective study of VAIN treatment with self-administered vaginal imiquimod compared with laser vaporisation or expectant. Imiquimod appears to be superior in achieving hrhpv clearance: 63% in the imiquimod group and 11% in the laser group. In the laser vaporisation group hrhpv persisted among 67% of the patients. Persistent hrhpv infection is a risk factor for recurrent disease. 11,22 We did not detect differences in histological regression between imiquimod treatment or laser vaporisation. However, at baseline we detected four cases of VAIN 3 in the imiquimod group, but none in the laser group, which might have biased the results. Patients adhered well to imiquimod treatment and the tolerability was good despite side effects. Only one ulceration of the vulva was observed and this healed after the cessation of imiquimod. Strengths of our study are the prospective randomised design, repeated hrhpv testing and good compliance and follow-up. Study limitations included small sample size and short follow-up period. This was a pilot study and therefore the effects of for example cigarette smoking, multifocality of the disease, immunosuppression and hormone replacement therapy could not be thoroughly assessed. Some HPV samples were missing. Some patients in the expectant group wanted to have laser vaporisation making it difficult to analyse the biopsy effect on disease regression. Disease regression can certainly be induced by biopsies. This is in accordance with our results as three of five patients had histological regression with no other treatment. Our cohort is representative of VAIN patients in general. The median age of our patients with VAIN 2 3 was 54, in comparison with the mean ages of in previous studies. 3,9,10,23 In our study, 77% of patients were hrhpv positive at baseline in comparison with 48 85% reported by others. 10,11,23 Of our patients 67% had been previously treated for or had concomitant HPV-related genital neoplasias while others report rates between 58% and 68%. 3,9,10 Multifocality of VAIN has been previously reported in 29 51% compared with 63% in our study. 3,11 Our rate of hysterectomised patients (37%) was lower than in other studies (54 88%). 3,9 11,23 Imiquimod has been previously used in three studies among VAIN patients. Our study differs in design and dosage of imiquimod. Haidopoulos et al. treatedsevenpatientswith VAIN 2 3 with imiquimod that was applied under colposcopic control for 8 weeks. 17 All patients in this study were hrhpv positive at baseline, but were not re-tested. A 57% complete response rate was reported during median follow-up of 18.4 months, which is in line with our results of complete histological regression rate of 70% in a shorter follow-up period. Two retrospective studies with self-administered imiquimod cream comprised mainly of VAIN 1 patients, with regression rates of 86% (n 5 42) and 54% (n 5 26), respectively. 16,24 The study by Lin et al. included a combined group of CIN/VAIN patients without a clear distinction between the two. 24 In the study by Buck et al. the patients were younger with a mean age of 20 years, and partially the diagnosis of VAIN was made on colposcopic impression without histological confirmation. 16 Considering the spontaneous regression rate of VAIN 1 the results are difficult to interpret. Contemporary treatment of VAIN is laser vaporisation. Partial or total vaginectomy, 5-fluorourasil, excisional procedures and radiotherapy have also been used. Efficacy rates have been similar to that of laser vaporisation also with regard to recurrence rates, but adverse effects including chronic ulcerations after 5-fluorourasil have been more common. 25 Previous studies have reported 46 86% cure rates after one laser vaporisation treatment round ,22,26 Our result of 70% complete histological regression with imiquimod and 90% with laser treated patients can be considered equally effective as conventional laser vaporisation. Risk factors for relapse of VAIN have been shown to be persistent hrhpv infection and the severity of VAIN. 11,13,22,26 HPV clearance could decrease the risk of recurrent disease. HPV persistence after laser vaporisation has been in previous studies 23 62%. 13,22 In our study, HPV persistence after laser vaporisation was 67%. In summary, imiquimod seems to be a promising and safe treatment modality for high-grade VAIN and is likely to have less adverse effects. If future studies prove HPV clearance by imiquimod is long-lasting repeat treatment episodes could be avoided. We are now conducting a large randomised trial of imiquimod and laser treatment for VAIN, and assess lesion regression in relation to different HPV genotypes (ISRCTN , EudraCT ). We also continue to monitor the patients of this study for hrhpv clearance and histological regression over time. Acknowledgements We want to thank the staff of the Colposcopy Clinic of Helsinki University Hospital Women s Hospital. The study was funded by the Research Council for Health (Finland).

6 2358 Randomised trial on treatment of vaginal intraepithelial neoplasia References 1. Henson D, Tarone R. An epidemiologic study of cancer of the cervix, vagina, and vulva based on the Third National Cancer Survey in the United States. Am J Obstet Gynecol 1977;129: De Vuyst H, Clifford GM, Nascimento MC, et al. Prevalence and type distribution of human papillomavirus in carcinoma and intraepithelial neoplasia of the vulva, vagina and anus: a meta-analysis. Int J Cancer 2009;124: Sillman FH, Fruchter RG, Chen YS, et al. Vaginal intraepithelial neoplasia: risk factors for persistence, recurrence, and invasion and its. Am J Obstet Gynecol 1997;176: WHO Classification of Tumours of Female Reproductive Organs. 4th edn., vol. 6, Lyon: International Agency for Research on Cancer. 5. Aho M, Vesterinen E, Meyer B, et al. Natural history of vaginal intraepithelial neoplasia. Cancer 1991;68: Massad LS. Outcomes after diagnosis of vaginal intraepithelial neoplasia. J Low Genit Tract Dis 2008;12: Rome RM, England PG. Management of vaginal intraepithelial neoplasia: a series of 132 cases with long-term follow-up. Int J Gynecol Cancer 2000;10: Finnish Cancer Registry, Institute for Statistical and Epidemiological Cancer Research. Retrieved 4 January 2016, from 9. Boonlikit S, Noinual N. Vaginal intraepithelial neoplasia: a retrospective analysis of clinical features and colpohistology. J Obstet Gynaecol Res 2010;36: Gunderson CC, Nugent EK, Elfrink SH, et al. A contemporary analysis of epidemiology and of vaginal intraepithelial neoplasia. Am J Obstet Gynecol 2013;208:410.e1.e Jentschke M, Hoffmeister V, Soergel P, et al. Clinical presentation, treatment and outcome of vaginal intraepithelial neoplasia. Arch Gynecol Obstet 2016;293: Perrotta M, Marchitelli CE, Velazco AF, et al. Use of CO 2 laser vaporization for the treatment of high-grade vaginal intraepithelial neoplasia. J Low Genit Tract Dis 2013;17: Wang Y, Kong WM, Wu YM, et al. Therapeutic effect of laser vaporization for vaginal intraepithelial neoplasia following hysterectomy due to premalignant and malignant lesions. J Obstet Gynaecol Res 2014;40: Beutner KR, Spruance SL, Hougham AJ, et al. Treatment of genital warts with an immuneresponse modifier (imiquimod). J Am Acad Dermatol 1998;38: Schon MP, Schon M. Imiquimod: mode of action. Br J Dermatol 2007;157: Buck HW, Guth KJ. Treatment of vaginal intraepithelial neoplasia (primarily low grade) with imiquimod 5% cream. J Low Genit Tract Dis 2003;7: Haidopoulos D, Diakomanolis E, Rodolakis A, et al. Can local application of imiquimod cream be an alternative mode of therapy for patients with high-grade intraepithelial lesions of the vagina? Int J Gynecol Cancer 2005;15: Grimm C, Polterauer S, Natter C, et al. Treatment of cervical intraepithelial neoplasia with topical imiquimod: a randomized controlled trial. Obstet Gynecol 2012;120: Koeneman MM, Kruse AJ, Kooreman LF, et al. TOPical imiquimod treatment of high-grade cervical intraepithelial neoplasia (TOPIC trial): study protocol for a randomized controlled trial. BMC Cancer 2016;16:132. doi: /s van Seters M, van Beurden M, ten Kate FJ, et al. Treatment of vulvar intraepithelial neoplasia with topical imiquimod. NEnglJMed2008;358: WHO Classification of Tumours. Pathology and genetics of tumours of the breast and female genital organs. Lyon: IARC Press, Frega A, French D, Piazze J, et al. Prediction of persistent vaginal intraepithelial neoplasia in previously hysterectomized women by high-risk HPV DNA detection. Cancer Lett 2007;249: Rhodes HE, Chenevert L, Munsell M. Vaginal intraepithelial neoplasia (VaIN 2/3): comparing clinical outcomes of treatment with intravaginal estrogen. J Low Genit Tract Dis 2014;18: Lin CT, Qiu JT, Wang CJ, et al. Topical imiquimod treatment for human papillomavirus infection in patients with and without cervical/vaginal intraepithelial neoplasia. Taiwan J Obstet Gynecol 2012;51: Krebs HB, Helmkamp BF. Chronic ulcerations following topical therapy with 5-fluorouracil for vaginal human papillomavirus-associated lesions. Obstet Gynecol 1991;78: Kim HS, Park NH, Park IA, et al. Risk factors for recurrence of vaginal intraepithelial neoplasia in the vaginal vault after laser vaporization. Lasers Surg Med 2009;41: