Imiquimod treatment of anal intraepithelial neoplasia (patient 4). A, Complete circumferential demarcation of anal intraepithelial neoplasia (AIN) after 2 weeks of imiquimod therapy. Clinically, leukoplakic AIN was visible only at the lower part of the anus (arrow) before treatment. B, Complete clearance of AIN with no residual signs of imiquimod-related erythema 16 weeks later.
Imiquimod treatment of anal intraepithelial neoplasia (patient 7). A, Perianal indurated, erythematous plaque (2 cm in diameter) with superficial erosions (erythroplakic grade 3 anal intraepithelial neoplasia) before therapy. B, Clinical appearance after 16 weeks of imiquimod therapy.
Histologic features of anal intraepithelial neoplasia treated with imiquimod (patient 18). A, Keratinocytic polymorphism, hyperchromatic nuclei, and numerous mitoses consistent with grade 3 anal intraepithelial neoplasia before imiquimod therapy (hematoxylin-eosin, original magnification ×40). B, Almost-normal epidermis with a moderate residual inflammatory infiltrate at the end of therapy (hematoxylin-eosin, original magnification ×40).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Wieland U, Brockmeyer NH, Weissenborn SJ, et al. Imiquimod Treatment of Anal Intraepithelial Neoplasia in HIV-Positive Men. Arch Dermatol. 2006;142(11):1438–1444. doi:10.1001/archderm.142.11.1438
To evaluate the treatment of anal intraepithelial neoplasia (AIN) with the local immune response modifier imiquimod in human immunodeficiency virus (HIV)-positive men who have sex with men (MSM).
Prospective, nonrandomized, open-label pilot study, with a mean follow-up time of 9½ months.
Dermatology department of a university hospital.
Twenty-eight consecutive HIV-positive MSM with histologically confirmed perianal (n = 23) or intra-anal (n = 5) AIN.
Overnight treatment with self-applied imiquimod cream (perianal AIN) or suppositories (intra-anal AIN) 3 times a week for 16 weeks.
Main Outcome Measures
Response to treatment was documented using clinical, cytologic, and histologic criteria. Human papillomavirus (HPV) typing and HPV DNA load determination for the high-risk HPV types 16, 18, 31, and 33 were performed.
Seventeen (61%) of all 28 patients included in the study and 17 (77%) of the 22 patients with AIN, who applied imiquimod as instructed, showed clinical and histologic clearance at the end of therapy. Four patients had residual AIN and 1 patient did not improve. Clinical response was accompanied by a sharp decline in HPV DNA loads and by a reduction in the number of HPV types, but long-term HPV clearance was rarely achieved. In the follow-up period, AIN cleared in 3 patients with residual AIN. Fourteen (78%) of 18 imiquimod responders with at least 5 five months of follow-up had a normal cytologic and clinical picture at the end of the follow-up period. Three primary responders developed a recurrence. In 6 noncompliant patients, there was no clinical or morphological improvement and the HPV DNA loads remained high.
Imiquimod appears to be a safe and effective treatment option for AIN in HIV-positive MSM. Clinical response is accompanied by a significant decrease in high-risk HPV DNA load. These results should encourage controlled randomized studies of imiquimod treatment of AIN.
clinicaltrials.gov Identifier: NCT00365729
Like cervical cancer, anal cancer is causally linked to persistent infection with high-risk human papillomavirus (HPV) types such as HPV-16 or HPV-18.1-3 In human immunodeficiency virus (HIV)-positive men who have sex with men (MSM), anal HPV prevalence is approximately 90%, and infections with multiple HPV types are common.4-6 Consequently, HIV-positive MSM have a strongly increased risk for anal cancer and its precursor, anal intraepithelial neoplasia (AIN). Despite highly active antiretroviral therapy (HAART), the incidence and prevalence of AIN and anal cancer have not decreased in HIV-positive MSM.6-8 In a recent study, 52% of HIV-positive men presented with advanced AIN.6
Imiquimod (1-[2-methylpropyl]-1H-imidazo-[4,5-c]-quinolin-4-amine) is a topical immune response modifier that was recently approved for the management of anogenital warts, superficial basal cell carcinoma, and actinic keratoses (in the United States). It has been used off-label for the treatment of recalcitrant warts, nodular basal cell carcinoma, Bowen disease, cutaneous squamous cell carcinoma, vulvar intraepithelial neoplasia, and molluscum contagiosum.9-12 Binding of imiquimod to the Toll-like receptor 7 on monocytes, macrophages, and dendritic cells activates both the innate and the cell-mediated (TH1) immune response by the induction of several proinflammatory cytokines and chemokines, such as interferon alfa, tumor necrosis factor α, and interleukin 1, 6, 8, and 12. Furthermore, imiquimod induces apoptosis and activates B lymphocytes.9,10,12-14 At present, its safety and efficacy in HPV-associated anogenital warts has been reported in about 40 clinical studies, including a randomized trial in HIV-infected patients.9,10,12
To date, there is still no universally accepted standard of treatment of AIN. Surgical excision or ablation is the primary form of treatment, although extensive anal surgery may lead to considerable morbidity, such as wound-healing problems, anal stenosis, or incontinence, especially in immunosuppressed patients.2,15 Therefore, topical therapies, including podophyllotoxin, 80% trichloroacetic acid or liquid nitrogen, carbon dioxide laser ablation, electrocautery, and infrared coagulation, have been considered for smaller lesions.2,3,15,16 Most ablative therapies, however, have considerable recurrence rates and frequently require retreatment.17,18 Reasons for recurrences could be the presence of multifocal AIN, HPV persistence in clinically healthy surrounding tissue, and HIV-associated immunosuppression leading to limited control of HPV. We recently reported on the treatment of AIN with imiquimod cream in 10 HIV-infected MSM and observed complete clearance or at least histologic regression in all patients.19 These results encouraged us to further evaluate the effects of imiquimod in an open-label prospective pilot study in HIV-positive MSM with histologically confirmed AIN.
Between November 2003 and December 2004, 124 HIV-positive MSM were screened for AIN as previously described.20 After giving informed consent, 28 consecutive patients with newly diagnosed perianal or intra-anal AIN were included in the study. The last follow-up sample was obtained in September 2005. Patients younger than 18 years and patients with concomitant malignancy or opportunistic infections were excluded. The patients ranged in age from 31 to 69 years (mean age, 43 years). The CD4 cell counts and HIV-1 RNA loads were between 39/μL and 1015/μL (mean, 422/μL) and less than 40 to 86 300 copies/mL (mean, 7369 copies/mL), respectively. The Centers for Disease Control and Prevention stages are listed in Table 1 and Table 2. Twenty-six patients received HAART. One patient (No. 11) did not receive HAART, and 1 patient (No. 13) started HAART during the follow-up period. The study, which was a substudy of the German Competence Network HIV/AIDS, was approved by the ethics review board of Ruhr University Bochum, Bochum, Germany. It was conducted according to Declaration of Helsinki principles.
All participants underwent a standardized anal screening program as previously described.16,20 Diagnosis of AIN was established according to clinical and histopathologic criteria (Table 1 and Table 2).20,21 Posttreatment biopsy specimens were obtained next to the site of the first specimen. Smears for cytology and HPV detection were collected from both the perianal area and the anal canal at each visit.20,22 Pathologists were blinded to the patients' diagnoses.
Patients with histologically confirmed AIN were treated overnight with self-applied imiquimod cream or suppositories 3 times a week for 16 weeks. Patients with perianal AIN received 5% imiquimod cream (Aldara; 3M Medica, Borken, Germany). In cases involving intra-anal AIN, suppositories containing 5% imiquimod cream were prepared as described.23 Imiquimod application was increased to 5 times a week in 1 patient (No. 1) because no signs of treatment-induced inflammation (erythema or mild erosion) were seen. In cases involving strong local adverse effects (eg, burning, marked erythema, or erosions), interruption of treatment for up to 7 days was allowed. All patients were instructed to refrain from receptive anal intercourse during imiquimod therapy. Follow-up visits were scheduled every 4 to 6 weeks.
Screening for HPV DNA was performed with nested polymerase chain reaction for anogenital HPV types.24 For HPV typing, polymerase chain reaction products were hybridized with oligonucleotide probes in an enzyme immunoassay.20,25,26 Determinations of HPV DNA loads were performed by real-time polymerase chain reaction with type-specific primers and probes for HPV-16, -18, -31, and -33.24,27 The HPV DNA load was expressed as HPV DNA copies per β-globin gene copy.27 If more than 1 of the 4 quantified HPV types were present in a sample, cumulative HPV DNA loads were calculated. The HPV-16, -18, -31, or -33 loads were determined for all samples from a patient if the respective HPV type had been found in any of the patient's samples (Table 3).
Statistical analyses (Wilcoxon test for paired samples) were performed with SPSS software (Version 11.0.1; SPSS Inc, Chicago, Ill). P<.05 was considered statistically significant.
Imiquimod was applied according to instructions by 22 (79%) of the 28 patients with AIN. All 22 patients developed erythema at the application site (Figure 1A). At the end of therapy, a complete clinical and histologic response was observed in 17 (61%) of all 28 study patients and in 17 (77%) of the 22 compliant patients (Table 1 and Figures 1, 2, and 3). Residual AIN, either clinical and/or histologic, was seen in 4 patients (Nos. 1, 19, 20, and 23 [Table 1]). One compliant patient (No. 21) with monofocal AIN-1 did not respond to imiquimod therapy and progressed to AIN-2 within 16 weeks. This patient's disease was Centers for Disease Control and Prevention stage C3, but his CD4 cell counts were stable at approximately 1000/uL and his HIV replication was well controlled by HAART.
Neither the grade nor the localization of AIN seemed to influence the success of therapy. Six (75%) of the 8 compliant patients with AIN-1, 4 (100%) of 4 patients with AIN-2, and 7 (70%) of 10 patients with AIN-3 showed complete clinical and histologic clearance of AIN at the end of therapy. Three (75%) of the 4 patients with intra-anal disease and 14 (78%) of the 18 patients with perianal disease had a complete response.
Adverse effects of imiquimod therapy, in addition to the erythema that was seen in all patients at the application site, occurred in 11 of the 22 compliant patients. Seven patients (32%) (Nos. 1, 4, 6, 16, 18, 25, and 27) developed mild erosions that led to the interruption of therapy for a maximum of 7 days. One patient (No. 9) had severe erosions as a result of overdosage (imiquimod application twice daily for 2 weeks). Four patients (18%) (Nos. 6, 15, 20, and 21) reported influenzalike symptoms within the first 2 weeks of therapy.
Six (21%) of the 28 patients with AIN did not apply imiquimod at all or only occasionally (Table 2). None of these 6 patients had any erythema at follow-up visits or any adverse effects that were typical of imiquimod therapy. Also, they did not show any clinical response. The histologic AIN grade did not change in 4 of the noncompliant patients, but 2 patients (Nos. 5 and 22) progressed from AIN-1/2 to AIN-3 within 16 weeks. Unwillingness to refrain from sexual intercourse (patients 2, 22, 24, and 28), pain during and after imiquimod application (patients 2, 22, 24, and 28), or depression (patient 5) was given as the reason for noncompliance. One patient (No. 11) turned out to have an HIV encephalopathy and therefore was not able to apply imiquimod. Surgical therapy was offered to the noncompliant patients and was accepted by 2 (patients 5 and 28) of them. One patient (No. 22) asked for a second course of imiquimod; the other 3 patients were unavailable for follow-up.
Before therapy, all 28 patients were HPV DNA positive and carried high-risk HPV types (Table 3). Twenty-four patients (86%) were infected with more than 1 HPV type, and the mean number of types was 4.7. The most frequent high-risk types were HPV-16 (n=22 [79%]), HPV-18 (n=11 [39%]), HPV-52 (n=10 [36%]), HPV-58 (n = 9 [32%]), and HPV-33 (n=8 [29%]). Also, 22 patients carried low-risk HPV types (Table 3). Imiquimod therapy generally led to a rapid decrease in high-risk HPV DNA loads within the first 4 to 8 weeks of therapy. The mean and median HPV DNA loads in compliant patients with analyzable loads and normal or improved histologic findings (n = 17) were 2261 and 139 (range, 0.03-17 634) HPV DNA copies per β-globin gene copy before therapy compared with 12 and 0.3 (range, 0-111) HPV DNA copies per β-globin gene copy at the end of therapy (P<.001). During the follow-up period, HPV DNA loads again slightly increased in most patients (Table 3). The mean and median HPV DNA loads in the patients' last follow-up samples were 88 and 3 (range, 0-1340) HPV DNA copies per β-globin gene copy. In 1 compliant patient (No. 21) who did not respond to imiquimod therapy, the HPV-16 load first decreased from 177 to 2 HPV DNA copies per β-globin gene copy within the first 4 weeks of therapy, then increased gradually to 42 HPV DNA copies per β-globin gene copy at the end of therapy, and then further increased to 322 HPV DNA copies per β-globin gene copy during follow-up. In the 6 noncompliant patients, the HPV DNA loads persisted at a high level or increased within the observation period (Table 3). The mean and median high-risk HPV DNA loads in the noncompliant patients were 175 and 110 (range, 21-500) HPV DNA copies per β-globin gene copy before therapy and 799 and 229 (range, 91-3721) HPV DNA copies per β-globin gene copy 16 weeks later.
In most compliant patients with complete or partial histologic response (n = 21), not only was there a decrease in HPV DNA loads, but there was also a reduction in the number of infecting HPV types. The mean and median numbers of HPV types were 4.6 and 4.0 (range, 1-13) before therapy compared with 2.6 and 1.0 (range, 0-11) at the end of therapy (P = .01). In the last follow-up samples, the mean and median numbers of HPV types were 3.7 and 3.0 (range, 0-10). At the end of therapy, only 4 patients (Nos. 6, 9, 14, and 19) were HPV DNA negative, and only 1 (patient 14) of them remained HPV DNA negative during the follow-up period (Table 3). Patient 14 and another patient (No. 18) who became HPV DNA negative during the follow-up period were in the earlier stages of HIV infection (Table 1).
Follow-up cytology results were available in all cases with normal histologic findings and clinical healing (n = 17) or with residual AIN at the end of therapy (n = 4). The duration of follow-up after the end of therapy ranged from 1 to 20 months (mean, 9.5 months; median, 7 months) (Table 1). Fourteen of the 17 patients with complete response at the end of therapy had normal cytologic results and a normal clinical picture at the last follow-up visit. Two patients (Nos. 3 and 7) had low-grade squamous intraepithelial lesions (LSIL) 5 and 6 months after the end of therapy, but the lesions appeared clinically inconspicuous. One year after the end of therapy, 1 patient with AIDS (No. 16) developed a clinical relapse and high-grade squamous intraepithelial lesions (HSIL)/AIN-2. One patient (No. 23) with residual clinical disease and normal histologic findings at the end of therapy still had LSIL on cytologic examination but had a normal clinical picture 6 months later. Interestingly, 3 patients (Nos. 1, 19, and 20) who had residual AIN-1 at the end of therapy demonstrated normal cytologic findings during follow-up, without further intervention, and had a normal clinical picture 5, 13, and 18 months after the end of therapy. In summary, of the 21 patients with normal or improved histologic findings at the end of therapy, 17 (81%) had normal cytologic findings and a normal clinical picture at their last visit. When only patients with a follow-up of at least 5 months are considered, 78% (14/18) had normal cytologic findings and a normal clinical picture at their last visit.
Human immunodeficiency virus–positive MSM develop AIN and anal cancer more frequently than HIV-negative MSM, and progression from normal findings or LSIL to HSIL occurs fast.1-3,5-8,15,29,30 Therefore, regular anal screening has been recommended for MSM to detect and treat anal dysplasias as early as possible.16,31 Compared with AIN in HIV-negative individuals, AIN in HIV-positive patients is more recalcitrant to therapeutic interventions.3,15,17 Many different approaches, including topical therapies, electrocautery, laser ablation, infrared coagulation, and surgical excision, have been used. Most approaches, however, have not been validated in controlled randomized trials.2,3,15,16 The successful use of imiquimod cream for the treatment of AIN has been described in an HIV-positive man with carcinoma in situ, in a nonimmunocompromised woman with Bowen disease, and, recently by us, in 10 HIV-positive MSM.19,32,33 In the present study, we used imiquimod to treat 28 HIV-positive MSM with histologically confirmed AIN. After 16 weeks of therapy, 17 (77%) of 22 compliant patients and 17 (61%) of all study patients were free of disease by clinical and histologic criteria. Four (22%) of 18 primary imiquimod responders with a follow-up period of at least 5 months had cytologic abnormalities at the end of the observation period, but only 1 patient (6%) developed HSIL. The relapse rate during follow-up compares favorably with the relapse rates in studies in which ablative techniques were used.17,18 Chang et al17 found persistent or recurrent HSIL in 23 (79%) of 29 HIV-positive patients with high-grade AIN who were treated with excision or cauterization, but their duration of follow-up was longer than ours and the mean time to recurrence was 12 months. We had 9 patients with a follow-up period of 12 months or longer; 8 were clinically and cytologically normal at the end of follow-up, and only 1 had developed HSIL/AIN-2. Goldstone et al18 used infrared coagulation to treat 165 high-grade AIN lesions in 68 HIV-positive MSM. Only 28% of the lesions persisted, which is similar to our results. However, 65% of the patients developed new or persistent HSIL within a median follow-up time (217 days) that was similar to the follow-up period in our study. One advantage of imiquimod therapy over ablative therapies could be the induction of an HPV-specific T-cell immunity, as has been shown with imiquimod treatment of condylomas and vulvar intraepithelial neoplasia.11,34 The decrease in HPV DNA load and infecting HPV types during therapy and the persistent low HPV loads in the follow-up period that were seen in our patients could be the result of an imiquimod-induced TH1 response. The TH1 response has also been assumed to be the reason for the decreases in HPV DNA and messenger RNA that have been seen after imiquimod treatment of condylomas.13,34,35 In our study, 3 patients with residual AIN-1 at the end of therapy had complete clearance of their AIN in the follow-up period, a result that further supports this assumption. It is possible that the “end-of-treatment” biopsy specimen should not be obtained in the first week after the last application of imiquimod, as in this study, but, instead, should be obtained a few weeks later to give time for final clearance of dysplasias.
In a recent review article, imiquimod was not recommended for intra-anal AIN or high-grade perianal AIN.2 In our study, neither the localization nor the grade of AIN seemed to have an influence on the response to treatment. However, larger and controlled studies are needed to confirm this finding, especially since imiquimod is not licensed for the treatment of AIN or for intra-anal application.
One of our cases progressed to AIN-2 during imiquimod therapy. Reasons for the failure of imiquimod therapy could be (1) a reduction of dermal dendritic cells, as has been shown in cases involving imiquimod-resistant condylomas; (2) the lack of an HPV-specific TH1 immunity, as has been seen in patients with vulvar intraepithelial neoplasia that did not respond to imiquimod, or (3) Toll-like receptor 7 polymorphisms.11,36,37
A considerable number of patients (21%) were noncompliant. The main reason patients gave for not applying imiquimod was incompatibility with their sexual lifestyle. This issue should be discussed with patients before they begin therapy. Ablative therapies are probably the better choice for patients who are not ready to change their sexual activities during imiquimod application. An advantage of ablative techniques is that they do not require long-term compliance or an ability to administer the therapy. Therefore, ablative therapies presumably should be the first choice for patients with mental disorders. In our study, noncompliant patients had verrucous and multifocal AIN more often than compliant patients (Table 1 and Table 2). This outcome possibly reflects a more reluctant attitude to seek and adhere to medical advice, especially since mean HIV RNA loads and CD4 cell counts were also less favorable in noncompliant patients than in compliant patients (HIV RNA, 26 642 copies/mL vs 2113 copies/mL; CD4 cell count, 359/μL vs 484/μL).
The limitations of our study are the lack of a control group receiving an ablative therapy, the unequal length of the follow-up period, and the relatively small number of patients. The high primary response rates, the low recurrence rates, the significant decrease in HPV loads, and the tolerable side effects shown in this pilot study should encourage the initiation of larger studies to compare imiquimod therapy for AIN with ablative techniques. It has been suggested that a combination of vaccination (inducing a HPV-specific TH1 response) and local imiquimod application could increase the likelihood of a strong clinical response in patients with vulvar intraepithelial neoplasia.11 This assumption could be incorporated in the planning of future studies of imiquimod therapy for AIN.
Correspondence: Alexander Kreuter, MD, Department of Dermatology and Allergology, Ruhr University Bochum, Gudrunstrasse 56, D-44791 Bochum, Germany (firstname.lastname@example.org).
Financial Disclosure: None reported.
Accepted for Publication: April 29, 2006.
Author Contributions: Dr Kreuter had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Wieland, Brockmeyer, Pfister, and Kreuter. Acquisition of data: Wieland, Hochdorfer, Stücker, Swoboda, and Kreuter. Analysis and interpretation of data: Wieland, Brockmeyer, Weissenborn, Altmeyer, and Kreuter. Drafting of the manuscript: Wieland and Kreuter. Critical revision of the manuscript for important intellectual content: Wieland, Brockmeyer, Weissenborn, Hochdorfer, Stücker, Swoboda, Altmeyer, Pfister, and Kreuter. Statistical analysis: Weissenborn. Obtained funding: Wieland. Administrative, technical, and material support: Wieland, Hochdorfer, Stücker, and Swoboda. Study supervision: Wieland, Brockmeyer, Hochdorfer, Stücker, Altmeyer, Pfister, and Kreuter.
Funding/Support: This study was supported by grant 01 KI 0211 from the Federal Ministry of Education and Research, German Network of Competence HIV/AIDS.
Role of the Sponsor: The funding organization (German Federal Ministry of Education and Research) was not involved in the design and conduct of the study; the collection, management, analysis, and interpretation of data; or in the preparation, review, or approval of the manuscript.
Acknowledgment: We thank Monika Junk, Nicole Girrulat, and Tanja Blome for excellent technical assistance.