Terminal capillaries present in the lesion are a typical clinical sign for benign anogenital warts.
The lesion clinically appeared as an anogenital wart, but histopathologically revealed focal areas of high-grade dysplasia. p16INK4a is an indirect marker of high-risk human papillomavirus E7 oncogene expression. Strong nuclear and cytoplasmic p16INK4a expression is present in the areas of high-grade dysplasia, whereas p16INK4a staining is negative in the anogenital wart part of the lesion (hematoxylin-eosin, original magnification ×100).
eTable 1. Number of samples per patient, localization of AGWs, date of biopsy collection, and histologic diagnosis of 38 AGWs collected from 25 HIV-positive MSM.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Kreuter A, Siorokos C, Oellig F, Silling S, Pfister H, Wieland U. High-grade Dysplasia in Anogenital Warts of HIV-Positive Men. JAMA Dermatol. 2016;152(11):1225–1230. doi:10.1001/jamadermatol.2016.2503
Do anogenital lesions of HIV-positive men that clinically appear as benign warts contain areas of dysplasia, and if so, what are the virological characteristics of those lesions?
In this case series, a high proportion of anogenital warts contained areas of high-grade and low-grade dysplasia or even invasive cancer. Some of these lesions contained only low-risk-HPV types. Dysplasia was absent in all lesions of immunocompetent control patients.
All anogenital warts of men who are HIV-positive should be evaluated histopathologically to rule out dysplasia.
Human papillomavirus (HPV)-induced anogenital lesions are very frequent in men who have sex with men (MSM) who are HIV-positive (HIV+). Anogenital warts (AGWs) are considered benign lesions caused by low-risk HPV-types, whereas anogenital dysplasias are potential cancer precursors associated with high-risk HPV-types. Both types of lesions can usually be distinguished clinically.
To describe a case series of HIV+ MSM with typical AGW that harbored different grades of dysplasia.
Design, Setting, and Participants
For this retrospective virological analysis, we recruited 25 HIV+ MSM with AGWs (n = 38) harboring areas of dysplasia and 22 patients who were HIV-negative (HIV−) with AGWs seen between February 2013 and March 2015 at a tertiary dermatological referral center for anal cancer screening. Dysplasia-containing AGW tissue of HIV+ MSM were compared with randomly selected AGWs of patients who were HIV−.
Main Outcomes and Measures
Histopathological analysis, immunohistochemical staining for p16INK4a and Ki67, HPV-typing, and viral load determination in AGWs of HIV+ compared with patients who were HIV−.
Overall, 25 HIV+ MSM with AGWs (mean [SD] age, 47.3 [11.1] years) harboring areas of dysplasia and 22 patients who were HIV− (5 women, 17 men; mean [SD] age, 35.5 [12.8] years) with AGWs were included in this study. The 38 dysplasia-containing AGWs of HIV+ MSM harbored low-grade dysplasia in 6 cases (16%), high-grade dysplasia in 31 cases (81%), and areas of invasive anal carcinoma in 1 (3%) case. With the exception of 1 biopsy, all low-grade lesions were p16INK4a-negative, whereas 25 of 31 (81%) AGWs with high-grade lesions or an anal carcinoma were p16INK4a-positive. Only low-risk HPV-types were present in 11 samples (29%; 2 low-grade lesions and 9 high-grade lesions), low-risk and high-risk types were found in 19 samples (50%; 1 low-grade lesion and 18 high-grade lesions), and only high-risk HPV-types were present in 8 samples (21%; 3 low-grade lesions, 4 high-grade lesion, and 1 cancer-containing lesion). High low-risk HPV DNA loads were found in low-grade and high-grade lesions, while high high-risk HPV DNA loads were only found in AGWs harboring high-grade lesions. The 22 AGWs of patients who were HIV− showed no signs of dysplasia, and p16INK4a-staining was always negative. All of these samples carried low-risk HPV types, and in 2 cases high-risk HPV-types were detected additionally.
Conclusions and Relevance
In contrast to immunocompetent patients, AGWs of HIV+ MSM may harbor high-grade dysplasia or even invasive squamous cell carcinoma. A substantial proportion of theses lesions may only contain low-risk HPV-types. Anogenital warts in patients who are HIV+ should be evaluated histopathologically to exclude cancer precursors.
Human papillomavirus (HPV)-induced anogenital warts (AGWs) and anogenital dysplasias are very frequent in HIV-positive (HIV+) men who have sex with men (MSM).1-4 Anogenital warts (condylomas) are considered benign lesions caused by low-risk HPV-types such as HPV 6 or 11, whereas high-grade anal intraepithelial neoplasia (AIN) and penile intraepithelial neoplasia (PIN) are potential cancer precursors associated with high-risk HPV-types such as HPV 16 or 18.5 Based on morphologic criteria AGWs and AIN and/or PIN can usually be distinguished by clinical inspection or high-resolution anoscopy (HRA).3,6 This study reports a series of HIV+ MSM who clinically presented with typical AGWs that harbored areas of low-grade or high-grade dysplasia or, in 1 case, invasive carcinoma.
Patients were diagnosed and treated at the outpatient clinic for anogenital diseases of the Department of Dermatology, Venereology, and Allergology of HELIOS St. Elisabeth Hospital Oberhausen, Germany. Between February 2013 and March 2015, a total of 73 HIV+ MSM participated in an anal cancer screening program as described previously.2,4,7 Thirty-six (49%) of these 73 HIV+ MSM had AGWs at first presentation. Thirty-eight AGWs (35 perianal or intra-anal biopsy specimens and 3 penile biopsy specimens) harboring areas of dysplasia of 25 HIV+ MSM were included in this study for further virological and immunohistochemical analysis. Overall, 16 of 25 HIV+ MSM delivered 1 biopsy specimen, 5 patients delivered 2 biopsy specimens each, and 4 patients delivered 3 biopsy specimens each (eTable 1 in the Supplement). For comparison, randomly selected AGWs (10 anal biopsy specimens, 9 penile biopsy specimens, and 3 biopsy specimens from the mons pubis or groin) of 22 otherwise healthy patients who are HIV-negative (HIV−) seen in the same time period were analyzed. All patients included presented with AGWs showing typical morphological signs, both on inspection (dome- or comb-shaped verrucous papules) and on HRA (homogenous terminal capillaries, papillary structures) (Figure 1).3,6
The 25 HIV+ MSM were 23 to 70 years old (mean 47.3, standard deviation (SD) 11.1). The HIV-negative patients (5 females, 17 males) were 20 to 73 years old (mean 35.5, SD 12.8) and significantly younger than the HIV+ MSM (P = .001). All HIV+ MSM received antiretroviral therapy. Their median CD4 cell count was 551/µl (interquartile range, 402-690/µl) and, with the exception of 1 patient, they did not have detectable plasma HIV-1 RNA. Fourteen (56%) HIV+ MSM had received no prior therapy for their AGWs. The other 11 (44%) HIV+ MSM had received treatment for their AGWs before study inclusion: 5 had monotherapy with imiquimod 5% cream, 1 had monotherapy with topical 85% trichloroacetic acid, 1 had monotherapy with electrocautery, 2 patients had a combination of imiquimod 5% cream and topical 85% trichloroacetic acid, and 2 had a combination of imiquimod 5% cream and electrocautery. Six (27%) of the 22 HIV− controls were naïve to any kind of condyloma therapy, and 16 patients (73%) had received prior treatment for their AGWs (6 patients, electrocautery; 4 patients, imiquimod 5% cream; and 6 patients, therapy unknown). None of the HIV+ MSM and control patients who were HIV− had previously received prophylactic HPV vaccination with the bivalent or the quadrivalent HPV vaccine. The study was approved by the ethics review board of the Aerztekammer Nordrhein (Medical Council North-Rhine), and all participants provided written informed consent.
Human papillomavirus DNA detection and typing and HPV DNA load determination were performed.7,8 Briefly, DNA was isolated from formalin-fixed paraffin-embedded tissue (a total of 10 5-µm sections per sample) using the QIAamp DNA mini kit (Qiagen).2 The HPV-DNA detection and typing were performed by group-specific polymerase chain reaction assays (PCRs) (BSGP5+/6+ PCR and A6/A8 PCR) followed by hybridization with type-specific probes using a bead-based multiplex assay that covers 38 low-risk and high-risk HPV types (low-risk: HPVs 6, 11, 40, 42-44, 54, 55, 57, 61, 71, 72, 81, 83, 84, 89; probable high-risk types: HPVs 26, 30, 34, 53, 66-68, 70, 73, 82; and high-risk types: HPVs 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59).8 Samples that were low-risk HPV-positive and high-risk HPV-negative (n = 18) were additionally analyzed with a short-fragment group-specific HPV PCR (SPF10 PCR yielding 65 bp PCR products) followed by HPV typing with a reverse line-blot assay (INNO-LiPA; Innogenetics) that covers 28 HPV types (high-risk types: HPVs 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59; probable high-risk types: HPVs 26, 53, 66, 68, 69, 70, 73, 82; and low-risk types: HPVs 6, 11, 40, 43, 44, 54, 71, 74).8 Seven high-risk HPV-positive biopsy specimens were only found with the short-fragment group-specific HPV PCR. The HPV DNA loads were determined by real-time PCRs with type-specific primers and probes for low-risk HPV types 6, 11, 42 and for high-risk HPV types 16, 31, 45, 58 using a Light Cycler 480 (Roche).7 The HPV DNA loads were expressed as HPV DNA copies per betaglobin-gene copy.7 The Ki67 and p16INK4a immunohistochemical analysis were performed using a monoclonal anti-Ki67 antibody (Clone-MIB-1; Dako) and the CINtec histology kit (Roche) according to manufacturer instructions. The p16INK4a and Ki67immunohistochemical stainings were evaluated according the criteria defined by Reuschenbach et al.9 To be considered positive, the p16INK4a staining had to be diffuse (nuclear and/or cytoplasmatic), continuous, beginning in the basal and parabasal cells, and reaching intermediate to superficial cell layers. Focal or sporadic p16INK4a staining was considered negative, as was completely negative p16INK4a expression. Epithelial Ki67 expression was categorized into 4 groups: group 1, less than 10% of cells, only parabasal layers; group 2, 10% to less than 30% of cells, only the lower third of the epithelium; group 3, 30% to less than 70% of cells, reaching the upper third of the epithelium; and group 4, 70% or more cells, comprising the full thickness of the epithelium.9
Qualitative data were cross-tabulated and tested for association by Pearson χ2 tests (2-sided). Distributions of HPV DNA loads and patient age were analyzed using the Mann-Whitney U test for 2 independent samples. P values less than .05 were considered statistically significant. Statistical analyses were performed with SPSS Statistics 22.214.171.124 (IBM Corporation).
The 38 dysplasia-harboring biopsy specimens of HIV+ MSM comprised 35 anal and 3 penile AGWs. Histopathological evaluation revealed regions of low-grade AIN (AIN 1) in 5, high-grade AIN in 29 (12 AIN 2; 17 AIN 3), and areas of invasive squamous cell carcinoma in 1 of the 35 anal AGWs. The 3 penile AGWs harbored areas of PIN1 or PIN2. None of the 22 AGWs of the individuals who were HIV− comprised areas of dysplasia (Table 1).
Immunohistochemical analysis was performed in 59 of the 60 biopsy specimens. In HIV+ MSM, p16INK4a immunohistochemistry was positive in 25 of 31 (81%) of high-grade dysplasia or cancer-containing biopsies (Figure 2), compared with 1 of 6 (17%) in biopsy specimens with low-grade dysplasia (P = .005)). Most high-grade dysplasia-containing AGWs or cancer-containing AGWs (28 of 31 [90%]) showed grade 3 or 4 Ki67-staining (Table 2). Interestingly, 4 AIN 2 or AIN 3 biopsy specimens that only carried low-risk HPV-types were p16INK4a and Ki67 positive. In contrast to HIV+ MSM, all 22 AGWs of patients who were HIV− were p16INK4a negative (P < .001), and Ki67-staining reached grade 3 or 4 only in 6 (27%) (P < .001).
All analyzed AGWs were HPV DNA positive. Only low-risk HPV types were found in 11 of 38 (29%) of the dysplasia-harboring AGWs of HIV+ MSM. Low-risk and high-risk HPV types were found in 19 of 38 (50%) dysplasia-harboring AGWs of HIV+ MSM, and only high-risk HPV types were found in 8 (21%) (Table 1). Compared with HIV+ MSM, significantly more AGWs of individuals who were HIV− carried only low-risk HPV types (20 of 22 [91%]; P < .001). In contrast, high-risk HPV types were more prevalent in AGWs of HIV+ MSM (27 of 38 [71%] vs 2 of 22 [9%]; P < .001). The most frequent low-risk types found in both patients who were HIV+ and HIV− were HPV 6 and HPV 11, detected in 22 of 30 (73%) and 21 of 22 (96%) of low-risk HPV positive biopsy specimens, respectively. Low-risk HPV 42 was detected in 7 of 30 (23%) and 1 of 22 (5%) low-risk HPV-positive AGWs of patients who were HIV+ and HIV−, respectively. Human papillomavirus 16 was the most frequent high-risk HPV type in HIV+ MSM and occurred in 7 of 27 (26%) of high-risk HPV-positive biopsy specimens, followed by HPV 53 (5 of 27 [19%]) and HPV 31 and/or HPV 45 (4 of 27 [15%] each). Human papillomavirus 16 was only found in AGWs harboring AIN 2 or 3 or cancer. The 2 high-risk HPV-positive AGWs of patients who were HIV− carried HPV 66 and HPV 59 plus HPV 67 in addition to HPV 6. Infections with more than 1 HPV type were found in 20 of 38 (53%) AGWs of HIV+ MSM, compared with 3 of 22 (14%) in individuals who were HIV− (P = .003).
The HPV DNA load determination revealed high HPV 6, HPV 11, or HPV 42loads (>1 HPV DNA copy per betaglobin-gene copy) in all but 1 AGW of HIV+ MSM, and in all AGWs of patients who were HIV− carrying these low-risk HPV-types (Table 1). Cumulative low-risk HPV DNA loads (n = 50) were higher in HIV+ MSM compared with patients who were HIV− (median, 181.2 vs 76.0), but the difference was not significant (P = .10). High HPV 16 loads were detected in 7 HPV 16-positive AGWs with AIN 2 or 3 or cancer. High-risk HPV DNA loads found in 2 HPV 31-positive or HPV 45-positive AIN 1-containing AGWs were below 1, and high-risk HPV DNA loads found in 7 HPV 31-positive, HPV 45-positive, or HPV 58-positive AGWs with AIN 2 or 3 were above 1 in 4 samples.
Recent evidence indicates that anogenital lesions of patients who are HIV+ morphologically appearing as benign AGWs may harbor focal areas of high-grade dysplasia.3,10-12 We have performed detailed HPV analyses including genotyping, viral load determination, and immunohistochemical analysis of such lesions in comparison to similar-looking AGWs of immunocompetent individuals. In contrast to AGWs of HIV+ MSM, none of the evaluated lesions of immunocompetent patients harbored focal areas of low-grade or high-grade dysplasia. In 4 high-grade AIN-containing AGWs of HIV+ MSM with strong p16INK4a and Ki67 staining, only low-risk HPV types with high DNA loads were detected. Thus, low-risk HPV types may be capable of inducing high-grade dysplasia in immunosuppressed individuals.12 This has also been documented in giant Buschke-Löwenstein AGWs.13 We cannot completely rule out that we have missed 1 or more high-risk HPV types in the low-risk HPV-positive AIN 2or 3–containing AGWs mentioned above. However, we have analyzed each of these samples with 3 different highly sensitive PCR protocols8 which makes it rather unlikely that high-risk HPV-types were missed, even if they were only present in small quantities. Moreover, during the review process of this article, a Dutch study14 that included HPV genotyping of 42 lesions of HIV+ MSM that clinically appeared as intra-anal warts revealed very similar results: in 4 anal condylomas that harbored regions of high-grade dysplasia (AIN2) only low-risk HPV types (HPV 6 or HPV 11) were detected.14
In our study, 5 of the AGWs with AIN 2 or 3 or cancer harbored only high-risk HPV types with high HPV DNA loads combined with a strong p16INK4a and Ki67 expression. Accordingly, high-risk HPV types seem to be capable of inducing AGW-like lesions that lack typical HRA signs of high-grade dysplasia as punctation, mosaicism, or neovascularization. Our results are in line with recent case reports of (giant) condylomas associated with HPV 16.15,16 Interestingly, Sturegard et al17 have found HPV 16 in 12.9% of 621 cytobrush samples collected from the surface of clinically diagnosed genital condylomas of mostly heterosexual patients from Malmö, Sweden. In 31 of 621 (5%) of the samples, HPV 16 was the only HPV type detected.17 However, histopathological analysis was not performed in the study of Sturegard et al, and contamination with HPV from other anogenital sites cannot be ruled out because surface swabs instead of lesional biopsies were analyzed.17 On the other hand, HPV 16 monoinfections have also been detected in studies on HPV type distribution of AGWs that analyzed lesional biopsy samples.18,19 Wikström et al18 have found HPV 16 monoinfections in 6 of 256 (2.3%) of benign anogenital lesions of Swedish men, and Hernandez-Suarez et al19 have detected HPV 16 as the single genotype in 5 of 155 (3.2%) of AGWs of women and in 3 of 106 (2.8%) of AGWs of men in Bogota, Colombia.
Richel et al3 identified high-grade AIN by HRA in 18% of intra-anal condylomatous lesions of HIV+ MSM, and Pimenoff et al12 detected AIN 2 or 3 histopathologically in 38% of 37 HIV+ MSM with intra-anal AGWs.3,12 Siegenbeek van Heukelom et al14 found focal areas of high-grade dysplasia in 23.8% of anal AGWs in HIV+ MSM that had participated in a clinical study on different treatment options for AIN.14 Within the observation period of the present study, 36 of 73 (49%) HIV+ MSM seen by us had AGWs, including the 25 men presented here with dysplasia-containing AGWs. Hence, a substantial number of benign-looking AGWs of HIV+ MSM may harbor cancer precursors. Based on the reported previous findings3,12,14 and on the results of the present study, we would like to suggest practical recommendations that might be helpful for the clinical and histopathological examination and treatment of AGWs in HIV+ MSM (Box). According to our recommendations, p16INK4a and/or Ki67 immunostaining and (optional) HPV genotyping should be performed in AGWs of patients who are HIV+ that reveal areas of dysplasia in histopathological analysis.
Clinical inspection of the entire anogenital area
Documentation of size and location of anogenital warts (AGWs) and search for clinical signs of dysplasia (eg, punctuation, mosaicism, neovascularization/abnormal vessels)
Perform (or send patient to) high-resolution anoscopy to exclude anal intraepithelial neoplasia and/or anal carcinomaa
Obtain biopsy specimens of representative lesions
Perform routine histopathological examination (standard fixation with hematoxylin-eosin staining)
In case of histopathological signs for dysplasia, add immunohistochemical staining for Ki67 and p16INK4a
Optional: add HPV genotyping in case of signs of dysplasia in histopathology
Perform ablative treatment of AGWs (eg, electrocautery, infrared coagulation, surgical excision)
Consider additional treatment with imiquimod 5% cream
a Anal AGWs represent a risk factor for AIN and anal carcinoma
Concerning potential coverage by the quadrivalent or nonavalent HPV vaccine,21 82% of the AGWs of the patients who were HIV− analyzed here would have been completely covered by either vaccine. Only 37% and 53% of the dysplasia-harboring AGWs of HIV+ MSM carried HPV types completely covered by the quadrivalent and the nonavalent HPV vaccine, respectively. In contrast to individuals who were HIV−, broad gender-neutral HPV vaccination might not suffice to prevent the vast majority of HPV-induced lesions in HIV+ MSM.
The results presented here should be interpreted in light of the limitations of our study. This was a retrospective study including a relatively small number of samples. Moreover, although different highly sensitive PCR protocols were used, HPV analyses comprised all lesions. Further studies on a large population of HIV+ MSM including microdissection and in-situ hybridization would corroborate our findings.22 Finally, the control patients were significantly younger than the HIV+ MSM, and this could have influenced our results. On the other hand, in contrast to women, it has been shown that anogenital HPV incidence and clearance are fairly stable in men throughout life.23
This study demonstrates that, in contrast to patients who are HIV−, AGWs of HIV+ MSM may harbor high-grade dysplasia or even cancer, although the clinical appearance is that of classical AGWs. Accordingly, all HPV-induced anogenital lesions of high-risk patients such as HIV+ MSM should be evaluated histopathologically and by p16INK4a and/or Ki67 immunohistochemical analysis and HPV genotyping should be considered if areas of dysplasia are detected.
Corresponding Author: Alexander Kreuter, MD, Department of Dermatology, Venereology and Allergology, HELIOS St. Elisabeth Hospital Oberhausen, University of Witten-Herdecke, Josefstrasse 4, 46045 Oberhausen, Germany (email@example.com).
Published Online: July 27, 2016. doi:10.1001/jamadermatol.2016.2503
Author Contributions: Drs Kreuter and Wieland had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analyses.
Acquisition, analysis, or interpretation of data: All Authors.
Drafting of the manuscript: Kreuter, Wieland.
Critical revision of the manuscript for important intellectual content: Siorokos, Oellig, Silling, Pfister.
Statistical analysis: Wieland.
Obtaining funding: Pfister, Wieland.
Administrative, technical, or material support: All Authors.
Conflict of Interest Disclosures: Dr Kreuter received lecture fees and fees for serving on advisory boards from Sanofi Pasteur MSD.
Funding/Support: This study was supported by the German Federal Ministry of Health (German National Reference Center for Papilloma- and Polyomaviruses, grant no. 1369-401).
Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: Monika Junk and Nabila Ristow, both medical technical assistants employed by the University Hospital of Cologne, provided excellent technical assistance. They were not compensated for their contributions.
Create a personal account or sign in to: