Penile Intraepithelial Neoplasia Is Frequent in HIV-Positive Men with Anal Dysplasia Alexander Kreuter1, Norbert H
View metadata, citation and similar papers at core.ac.uk brought to you by CORE
provided by Elsevier - Publisher Connector ORIGINAL ARTICLE
Penile Intraepithelial Neoplasia Is Frequent in HIV-Positive Men with Anal Dysplasia Alexander Kreuter1, Norbert H. Brockmeyer1, Soenke J. Weissenborn2, Thilo Gambichler1, Markus Stu¨cker1, Peter Altmeyer1, Herbert Pfister2 and Ulrike Wieland2, for the German Competence Network HIV/AIDS
Anogenital human papillomavirus (HPV)-infection is common in HIV-infected men who have sex with men (HIV þ MSM). These patients have a strongly increased risk of HPV-induced anal cancer and its precursor lesion, anal intraepithelial neoplasia (AIN), and a moderately increased risk for penile cancer. Only limited data exist on penile intraepithelial neoplasia (PIN) in HIV þ MSM. We determined the prevalence and evaluated the virologic characteristics of PIN and AIN in 263 HIV þ MSM. In case of histologically confirmed PIN (and AIN), HPV-typing, HPV-DNA load determination, and immunohistochemical staining for p16INK4a were performed. PIN was detected in 11 (4.2%) and AIN in 156 (59.3%) patients. Ten PIN patients also had AIN within the observation period. Four clinical types of PINs could be distinguished. High-risk-a-HPV-DNA was found in 10 PIN lesions, with HPV16 being the most frequent type. Infections with multiple HPV-types were common. All high-grade lesions had high- risk-HPV-DNA-loads X1 HPV-copy/b-globin-gene-copy. Cutaneous b-HPVs were found in PIN and AIN, but b-HPV-DNA loads were very low, irrespective of the histological grade. p16INK4a Expression was detectable in all PIN lesions and correlated both with the histological grade and with high-risk HPV-DNA loads. In view of the PIN prevalence found in our study, all HIV þ MSM should be screened for PIN in addition to AIN screening. Journal of Investigative Dermatology (2008) 128, 2316–2324; doi:10.1038/jid.2008.72; published online 3 April 2008
INTRODUCTION (CIN), anal (AIN), or penile (PIN) intraepithelial neoplasia. Human papillomaviruses (HPV) cause a wide spectrum of HPV-DNA has been found in 70–100% of PINs and in epithelial tumors ranging from benign warts to anogenital 29–81% of invasive penile cancers (depending on the cancers. Almost all cases of cervical and anal cancer, and the histological type), with HPV16 being the most prevalent type majority of basaloid and warty penile cancers are HPV- (Barrasso et al., 1987; Aynaud et al., 1994; Dillner et al., associated (Munoz et al., 2003; Gross and Pfister, 2004; 2000; Rubin et al., 2001; Ferreux et al., 2003; Lont et al., Palefsky, 2006). Phylogenetically, HPV are classified into 2006; Senba et al., 2006). HIV infected men who have sex genera a-, b-, g-, m-, and n-papillomavirus (de Villiers et al., with men (HIV þ MSM) have a strongly increased risk for 2004). b-, g-, m/n, and some a-HPV infect the skin, whereas development of AIN and anal cancer (Frisch et al., 2000; the majority of a-HPV is found on anogenital mucous Palefsky et al., 2005, 2006; Grulich et al., 2007). Highly membranes and the adjacent skin. a-HPV types regularly active antiretroviral therapy is not associated with reduction found in anogenital cancers as HPV16 or 18 are considered of AIN (Palefsky et al., 2005). Although up to 29% of younger high-risk (HR) HPV types, in contrast to low-risk HPV types men bear HR-HPV-DNA on the penis, PIN/penile cancer is a (for example, HPV6, 11), which are regularly found in genital relatively rare disease in immunocompetent patients in Europe warts and in low-grade dysplasias, but rarely in cancers and North America and mostly elderly men are affected (Munoz et al., 2003). The precursor lesions of anogenital (Dillner et al., 2000; van der Snoek et al., 2003; Nielson et al., cancers are called intraepithelial neoplasias, such as cervical 2007). Compared with HIV-negative men, those HIV positive have a somewhat higher penile HPV prevalence (van der Snoek et al., 2003; Sirera et al., 2006). Concerning PIN/penile 1Department of Dermatology, Ruhr-University Bochum, Bochum, Germany and 2Institute of Virology, University of Cologne, Koeln, Germany cancer in HIV-positive men, only few data exist (Frisch et al., 2000; Gomousa-Michael et al., 2000; Porter et al., 2002; ClinicalTrials.gov identifier: NCT00365729. Grulich et al., 2007). The goal of this study was to determine Correspondence: Dr Ulrike Wieland, Institute of Virology, University of Cologne, Fuerst-Pueckler-Strasse 56, Koeln 50935, Germany. the prevalence of PIN/penile cancer and to identify associated E-mail: [email protected] HPV types in a cohort of 263 HIV-positive MSM. Abbreviations: AIN, anal intraepithelial neoplasia; EQ, erythroplasia of Queyrat; HPV, human papillomavirus; HR, high-risk; HSIL, high-grade RESULTS squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial Prevalence of AIN and PIN lesion; MSM, men who have sex with men; PIN, penile intraepithelial neoplasia; SCC, squamous cell carcinoma A total of 263 HIV-positive MSM were prospectively Received 17 September 2007; revised 1 February 2008; accepted 9 February screened for the presence of anogenital lesions. A total of 2008; published online 3 April 2008 156 patients (59.3%) presented with anal dysplasia of any
2316 Journal of Investigative Dermatology (2008), Volume 128 & 2008 The Society for Investigative Dermatology A Kreuter et al. Penile Intraepithelial Neoplasia in HIV þ Men
grade. In 82 patients (31.2%) high-grade AIN (AIN2/AIN3/ imiquimod as previously described (Palefsky, 2006; HSIL) and in 86 patients (32.7%) low-grade AIN (AIN1/LSIL) Wieland et al., 2006). was diagnosed within the observation period. Twelve patients with low-grade AIN developed high-grade AIN in the course HPV spectrum and HPV-DNA load in patients with PIN of the study. A total of 69 (44.2%) of the 156 patients with and AIN AIN had more than one episode of AIN. Nine (5.8%) of the HPV-DNA was found in 10 of the 11 PIN patients. In one 156 AIN lesions were subclinical (6 AIN1, 3 AIN2). Six patient with focal PIN1 (patient 8), no HPV-DNA was patients (2.3%) had invasive anal cancer (5 squamous cell detectable in the penile lesion. All other PIN patients carried carcinomas (SCCs), 1 adenocarcinoma). PIN was diagnosed 1–6 different HR-a-HPV types, with HPV16 being the most in 4.2% of all patients (11/263), in 6.4% of patients with AIN frequent type occurring in eight (73%) PIN lesions. Two of any grade (10/156), in 4.1% of patients with low-grade AIN HPV16-negative high-grade PIN lesions carried HR types 26, (3/74), and in 8.5% of patients with high-grade AIN (7/82). 31, 39 (patient 1) or 35 (patient 6). Low-risk a-HPV types Six of the 11 PIN patients had PIN grade 3 (PIN3), two PIN2 were found in five PIN lesions (Table 1). Histological signs of and three had PIN1 (Table 1). No patient had invasive penile HPV infection (koilocytes) were detectable in 8 of the 11 PIN cancer. Five patients presented with PIN at the baseline visit lesions (patients 2–5, 8–11), including in the HPV-DNA- (patients 1, 3, 4, 5, 9) and six patients (patients 2, 6, 7, 8, 10, negative PIN1. HR-a-HPV types were detected in all AIN 11) developed PIN within the course of the study (incident lesions of the patients with PIN and in 96% (150/156) of all cases). Ten of the 11 PIN patients also had AIN within the anal dysplasias, with HPV16 again being the most frequent observation period: five patients had concomitant AIN and type (109/156, 70%). A total of 80% of the AIN patients with PIN, four patients had AIN 2–18 months before developing PIN (Table 1) and 87% (136/156) of all AIN patients carried PIN, and one patient first had PIN and 16 months later had additional anal low-risk a-HPV types. b-HPV-DNA was AIN (Table 1). We could observe four clinical types of found in four PIN and five AIN lesions of 7 of the 11 patients PIN, bowenoid (36%, n ¼ 4), verrucous (27%, n ¼ 3), shown in Table 1. The maximum number of b-HPV types erythroplakic (18%, n ¼ 2), and leukoplakic (18%, n ¼ 2) found in a lesion was seven (patient 10). Eight of the 11 PIN (Table 1; Figure 1). Each patient exhibited only one type patients had one or more identical a-orb-HPV types in their of PIN. Subclinical PIN lesions were not observed. Patients PIN and AIN lesions (Table 1). In patients with AIN and PIN with verrucous or erythroplakic PIN had the same clinical (n ¼ 10), the mean/median numbers of HPV-types were type of AIN, whereas all patients with bowenoid and one slightly higher in AIN than in PIN lesions, but the differences patient with leukoplakic PIN had different clinical types of were not significant (for example, HR-a-HPV types: 2.8/3 vs AIN (Table 1). Patient’s age at diagnosis of PIN ranged 1.8/1, P ¼ 0.105; a þ b HPV types: 5.7/5 vs 4.4/2, P ¼ 0.441, between 33 and 69 (mean 49) years, and the time between Wilcoxon test). The total number of HPV types and the diagnosis of HIV-infection and PIN diagnosis ranged between number of HR-a-HPV types were significantly higher in 1 and 16 (mean 10) years. With the exception of two patients verrucous PIN/AIN (n ¼ 8) than in non-verrucous (bowenoid, (no. 8,11), all PIN patients received highly active antiretro- erythroplakic, leukoplakic, n ¼ 13) PIN/AIN (mean number viral therapy. A total of 64% of the PIN patients were in CDC a þ b HPV types: 8.3 (verrucous) vs 3.1 (non-verrucous), stage C3, compared with 32% in the whole cohort and their P ¼ 0.010; mean number HR-a-HPV types: 3.4 (verrucous) vs mean CD4 counts were lower (396 ml 1 compared with 476 1.7 (non-verrucous), P ¼ 0.037, Mann–Whitney test). All in all patients). Five PIN patients were circumcised before patients with 10 or more HPV types per lesion had verrucous study entry: one patient since his first year of life (patient 9) PIN (patients 1, 5) or AIN (patients 5, 11) (Table 1). and four patients due to prior removal of penile condylomas HPV-DNA loads of HR-a-HPV types 16, 18, 31, and 33 (patient 1, 5, 7, 10). The risk for anal dysplasia of any grade were X1 HPV copy per b-globin gene copy for at least one was enhanced in smokers (93/143, 65%) compared with non- HR type in all 12 cases of PIN2/3 and AIN2/3 that carried the smokers (63/120, 52.5%) (odds ratio ¼ 1.68 (95% confidence respective types, and around or above 100 in 4 of 7 PIN2/3 interval 1.02–2.77)). For PIN, the risk increase in smokers and in 4 of 5 AIN2/3 cases analyzed (Table 1). HR-a-HPV- (8/143, 5.6%) compared with non-smokers (3/120, 2.5%) was DNA loads below 1 were found in the PIN1 cases. In contrast not significant (odds ratio ¼ 2.31 (95% confidence interval to elevated HR-a-HPV loads in high-grade AIN/PIN, b-HPV- 0.60–8.91)). DNA loads were generally very low, irrespective of the grade In contrast to anal cytology and histology, penile cytology of AIN or PIN (oo1 HPV copy per b-globin gene copy). correlated with penile histology only in 7 of the 11 cases. In Only one patient with PIN1 (patient 10) had higher b-HPV three patients the grade of PIN was underestimated by loads (0.9 and 0.5) than HR-a-HPV loads (0.1 and 0.03) cytology (patients 2, 5, 6) and in one patient (patient 11), (Table 1). Mean and median cumulative HR-a-HPV load of cytology was not evaluable due to scant cellularity (Table 1). PIN and AIN lesions were 3 orders of magnitude higher than Eight of the 11 PIN lesions were treated with electro- cumulative b-HPV loads (632 vs 0.25 (mean) and 54 vs 0.02 cautery and/or circumcision, depending on size and localiza- (median), P ¼ 0.025, Wilcoxon test). When comparing HR-a- tion. In two patients, PIN lesions were surgically removed. HPV loads of the few patients that had penile and anal lesions One patient (patient 3) refused treatment and was lost to of the same grade and carried the same quantifiable HR types follow-up. Depending on the site end extension, AIN was in these lesions, penile loads were higher than anal loads in treated with electrocautery, 85% trichloroacetic acid, or these three patients (patients 1–3).
www.jidonline.org 2317 2318 eieItapteilNolsai HIV in Neoplasia Intraepithelial Penile Kreuter A ora fIvsiaieDraooy(08,Vlm 128 Volume (2008), Dermatology Investigative of Journal Table 1. Patients’ characteristics, p16 expression, lesional HPV spectrum, and lesional HPV-DNA load in HIV+ men with PIN and AIN Stage of HPV-DNA Prevalence Histo- HPV-DNA Year of HIV HIV-RNA Clinical Histologic load5 of PIN Clinical logical Anal load5
Patient Age HIV (CDC/ (copies type of grade of Penile P16INK4a Penile [HPV and AIN type of grade Anal HPV [HPV al. et no. (yr) diagnosis WHO) CD4 ll 1 ml 1) PIN1 PIN cytology2 (IHC)3 % HPV types4 type] (m6 later) AIN1 of AIN cytology2 types4 type]
1 61 1995 C3 460 120 V III HSIL 82 *26, 31, 943 [31] Concomitant V III HSIL *16, 26, 143 [31] 39* 31, 68* 111, 831 0.02 [8] 16, 831 0.6 [16] 8, 15, odl [15] 8, 15, 0.001 [8] 20, 21, 38 38 0.01 [20] odl [15]
odl [38] odl [38] þ Men
5 57 1991 C3 329 567 V III LSIL 64 *16, 26, 0.5 [16] Concomitant V I LSIL *16, 26, 5,470 [16] 33, 39, 59, 73* 59, 66* 140, 44, 4 [33] 111, 811 odl [5] 70, 72, 81, 84, 891 5 odl [5] 5, 23, 0.03 [23] 36, 38, RTRX5 odl [36] 0.01 [38]
4 69 1993 C3 419 70 E III HSIL 82 *16* 2 [16] PIN4AIN (16 E I LSIL *16, 18, 6 [16] m) 56, 82* 1421 odl [18]
9 47 2000 B2 419 o40 E III HSIL 97 *16, 45, 677 [16] Only PIN no AIN ND nl *16* 1 [16] 73* 161, 721 1611
3 36 1999 B3 269 53 B III HSIL 46 *16, 18* 256 [16] Concomitant V III HSIL *16, 18, 111 [16] 31* 0.1 [18] 1831 17 [18] 23 0.8 [31] odl [23]
2 33 1993 A3 813 o40 L III LSIL 92 *16* 3,348 [16] Concomitant V III HSIL *16* 99 [16] 1611 Table 1. Continued Stage of HPV-DNA Prevalence Histo- HPV-DNA Year of HIV HIV-RNA Clinical Histologic load5 of PIN Clinical logical Anal load5 Patient Age HIV (CDC/ (copies type of grade of Penile P16INK4a Penile [HPV and AIN type of grade Anal HPV [HPV no. (yr) diagnosis WHO) CD4 ll 1 ml 1) PIN1 PIN cytology2 (IHC)3 % HPV types4 type] (m6 later) AIN1 of AIN cytology2 types4 type]
11 48 1988 C3 39 86,300 V II us 45 *16* 1 [16] AIN4PIN (5 V III HSIL *16, 18, 221 [16] m) 52, 73, 82* 1721 111, 721 62 [18] 8, 9, 0.5 [8] 14, 15, 36, 93 0.003 [15] 0.003 [36]
6 56 1996 C3 343 543 B II LSIL 17 *35* — AIN4PIN (18 L II HSIL *56* — m) 25
7 61 1989 C3 84 30,200 B I LSIL 28 *16* 0.4 [16] AIN4PIN (8 L III HSIL *26* — m) 111, 541 odl [5] 5
8 35 2003 A1 678 23,400 B I LSIL 7 Neg AIN4PIN (2 L I LSIL *16, 73* 3 [16] m) 16, 541
10 41 1992 C3 500 475,000 L I LSIL 9 *16, 31* 0.1 [16] Concomitant L III HSIL *16, 52, 58* 9 [16] HIV in Neoplasia Intraepithelial Penile 5, 12, 14, 15, 0.03 [31] 140, 831 21, 23, 80 0.9 [5] odl [15] 0.5 [23] AIN, anal intraepithelial neoplasia; ND, not determined; PIN, penile intraepithelial neoplasia; yr, years. 1v, verrucous; b, bowenoid; e, erythroplakic; l, leukoplakic. 2HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; us, unsatisfactory sample (scant cellularity); nl, normal. 3IHC, immunohistochemistry; %, percentage of p16-positive cells per section. 4flanked by *: mucosal (HPV-group a) high-risk types as defined by Munoz et al. (2003); flanked by 1: mucosal (HPV-group a) low-risk types as defined by Munoz et al.; in italic: HPV group beta types; neg,
no HPV-DNA detectable; the underlined HPV types were found in both the penile and the anal lesion of the respective patient. Kreuter A www.jidonline.org 5HPV load was defined as HPV-DNA copies per b-globin gene copy; odl, below detection limit of real-time PCR; , no quantifiable HPV type. 6m, months. þ tal. et Men 2319 A Kreuter et al. Penile Intraepithelial Neoplasia in HIV þ Men
a b
c d
Figure 1. Four clinical types of PIN in HIV-positive men. (a) Slightly scaling, INK4a verrucous plaque affecting the complete glans penis (verrucous PIN, patient Figure 2. Immunohistochemical staining of a PIN lesion for p16 . 5). (b) Sharply defined, dark red, glistening, velvety, non-tender, eroded Penile intraepithelial neoplasia grade 3 (patient 4). Immunohistochemical plaque on the glans penis, coronal sulcus, and prepuce (erythroplakic PIN, analysis (detail) using anti-p16 monoclonal antibody demonstrates a strong patient 4). (c) Multiple, slightly elevated, red papules on the glans penis immunoreactivity for p16 both in the nuclei and the cytoplasm, reaching the (bowenoid PIN, patient 3). (d) Slightly infiltrated, white to reddish plaque on upper third of the epidermis (82% of cells were p16-positive in the complete the prepuce (leukoplakic PIN, patient 2). biopsy specimen; original magnification 40). Focal areas of p16-negative cells were observed. Bar ¼ 0.1 mm.
p16INK4a expression in PIN p16INK4a expression, which is an indirect marker of HPV-E7 high-grade PINs detected in 30 HIV-positive patients with oncogene expression, was detectable in all PIN lesions by penile lesions, compared with 10 low- and two high-grade immunohistochemistry (Table 1; Figure 2). The percentage of PINs found in 40 HIV-negative men with penile lesions. To p16-positive cells correlated both with the histological grade our knowledge, prevalence data for PIN in HIV-positive men of PIN (n ¼ 11, Spearman coefficient of rank correlation: have not yet been reported. In this study, we have screened 0.881, Po0.001) and with cumulative HR-HPV-DNA loads 263 HIV-positive MSM and have found AIN in 59.3% and (n ¼ 9, Spearman coefficient of rank correlation: 0.854, PIN in 4.2% of all patients, with about half of the AIN and P ¼ 0.003). 73% of the PIN lesions being high-grade lesions. The AIN prevalence found in our study is comparable to previous DISCUSSION studies reporting high rates of AIN in HIV-positive men HPV-associated PIN is considered a precursor of some forms (Palefsky et al., 2005, 2006). The fact that PIN is less frequent of penile SCC. In immunocompetent patients, only 5–30% of than AIN is in line with the above mentioned numbers for the PIN cases will progress to invasive SCC and even high-grade respective invasive cancers in HIV-positive men. Further- PIN lesions may regress spontaneously (Gross and Pfister, more, in studies comparing penile and anal HPV-DNA 2004). In western countries, penile cancer is a rare disease prevalence of MSM, penile HPV prevalence has been lower that occurs mainly in elderly patients (Dillner et al., 2000). In than anal HPV prevalence (van der Snoek et al., 2003; Sirera HIV-positive men, an increased risk for penile cancer et al., 2006). This suggests both a lower susceptibility of the compared with HIV-negative men has been shown, but the penile epithelium for HPV infection and a lower risk of figures are much lower than those reported for anal cancer dysplasia/cancer development of penile HPV infections (Frisch et al., 2000). In a recent meta-analysis comprising 21 compared with the anal transformation zone. cases of penile cancer and 303 cases of anal cancer, Grulich Two-thirds of the patients with PIN were in CDC stage C3, et al. (2007) calculated standardized incidence ratios of 4.42 compared with one-third in the whole cohort. In a study by for penile and of 28.75 for anal cancer of HIV/AIDS patients. Porter et al. (2002) all HIV-infected patients with PIN had Concerning PIN in HIV-positive men, only few reports exist. CD4 counts below 200 mm 3. In our study, only two of the Bernard et al. (1992) found that HIV-positive men more 11 PIN patients had CD4 counts below 200 ml 1 and eight frequently had high-grade AIN/PIN than those HIV-negative. had below 500 ml 1. However, mean and median CD4 Gomousa-Michael et al. (2000) describe five low- and 10 counts were lower in PIN patients compared with the whole
2320 Journal of Investigative Dermatology (2008), Volume 128 A Kreuter et al. Penile Intraepithelial Neoplasia in HIV þ Men
cohort. Three clinical variants of PIN3 have been described, of the very low b-HPV loads, a commensal colonization of bowenoid papulosis, Bowen’s disease, and erythroplasia of the anogenital skin also seems possible. Furthermore, no b- Queyrat (EQ) (Gross and Pfister, 2004). We have recently HPV type was detectable in the two EQ-like PIN cases of this tried to categorize perianal AIN in four clinical types, study. On the other hand, b-HPV loads were 10-fold higher bowenoid, erythroplakic, leukoplakic, and verrucous AIN than HR-a-HPV loads in PIN patient 10, which could indicate (Kreuter et al., 2005). This classification might also be that b-HPV play a role in some cases of PIN development. suitable for all grades of PIN, with bowenoid PIN corre- Interestingly, all four PIN patients with penile b-HPV had sponding to bowenoid papulosis and erythroplakic PIN to been circumcised before study entry. Perhaps this facilitated EQ. Verrucous PIN partially resembles Bowen’s disease the spread of b-HPV to the distal part of the penis. The finding (well-demarcated scaly plaques); however, hyperkeratotic that eight patients had 1–6 identical anal and penile HPV forms of this PIN type have condylomatous aspects. types supports the assumption that dissemination of HPV Leukoplakic PIN does not correspond to bowenoid papulosis, types between different anogenital regions occurs within a Bowen’s disease, or EQ. Porter et al. (2002) have recently given patient. described the clinical spectrum of 35 PIN cases. All of their Several studies have shown that elevated HR-a-HPV-DNA eight HIV-positive patients presented with bowenoid loads in cervical scrapes are predictive of the severity of the papulosis, one additionally with Bowen’s disease and EQ. underlying cervical lesion and reflect an increased risk for In contrast to Porter et al., only 4 of our 11 PIN patients had cervical cancer. It has been difficult, however, to define a bowenoid PIN. PIN2/3 occurred with all clinical types, but all clear cut-off for HR-a-HPV-DNA load indicating the presence PIN1 lesions were bowenoid or leukoplakic. of advanced CIN (Ylitalo et al., 2000; Snijders et al., 2006). In HR-a-HPV-DNA was detectable in all PIN cases, with the HIV-positive women, cervical HPV16-DNA loads above 1 exception of a case of focal PIN1. This is in line with previous HPV copy per b-globin gene copy were associated with HSIL publications that have reported HPV-DNA prevalence rates (Weissenborn et al., 2003). In this study, we determined HR- around or above 90% in penile dysplasias of immunocom- a-HPV-DNA loads only for 4 of the 16 HR-a-HPV-types petent patients (Aynaud et al., 1994; Cupp et al., 1995; Rubin found in the PIN and AIN lesions (HPV16, 18, 31, and 33). et al., 2001). In two-thirds of our PIN lesions, more than one All patients with PIN2/3 or AIN2/3 had HR-a-HPV loads at or a-HPV type was found, which is more than that reported for above this level, but some patients with AIN1 also had high penile dysplasias of HIV-negative patients (16–22%) (Aynaud viral loads. However, HR-a-HPV-DNA loads below 1 were et al., 1994; Rubin et al., 2001). Higher rates of infections consistently found in PIN1 lesions, one of which may have with multiple HPV types have already been reported for AIN been caused by low-risk types 11 and/or 54. a-HPV-DNA of HIV-positive compared with HIV-negative patients (Pa- loads were also determined for low-risk HPV types 6 and 11 lefsky et al., 1998). HPV16 occurred in 73% of our PIN (Kreuter et al. 2006). HPV6/11 loads did not correlate with lesions, again more than that reported for HIV-negative the grade of dysplasia (data not shown). All HPV6/11 positive patients (41–43%) (Aynaud et al., 1994; Rubin et al., 2001). patients shown in Table 1 had a history of penile or anal Since HPV16 is considered more carcinogenic than other condylomata acuminata. HR-a-HPV types (Smith et al., 2007), the high HPV16 p16INK4a Expression is an indirect marker of HPV-E7 prevalence might pose a higher risk for progression to penile oncogene expression and p16INK4a immunohistochemistry is cancer. The total number of HPV types, as well as the number a suitable marker for detection and grading of CIN, of HR-a-HPV types, was significantly higher in verrucous independent of the infecting HR-a-HPV type (Sano et al., PIN/AIN than in non-verrucous lesions, possibly indicating an 1998; Vinokurova et al., 2005). Ferreux et al. (2003) found influence of the number of HPV types on the clinical picture. p16INK4a staining in 65% of HPV-positive, in 6% of HPV- However, the etiologic significance of multiple HPV types is negative, in 81% of HR-a-HPV-DNA-positive, and in 92% of largely unclear. Synergistic effects of HPV types infecting the HPV16-positive penile carcinomas. In our study, p16 same cell or lesion seem possible, but commensalism or expression was detectable in all PIN lesions, also in an surface contamination cannot be excluded. HPV types with HPV-negative focal PIN1 with few p16-positive cells. low viral loads may rather be passengers than drivers. Probably this PIN lesion also carried low amounts of HR-a- b-HPV-DNA was found in 4 PIN and 5 AIN lesions of HPV-DNA that were missed by PCR. The percentage of p16- seven patients. b-HPV were originally discovered in benign positive cells correlated both with the histological grade of skin lesions and SCC of patients with the genodermatosis PIN and with cumulative HR-a-HPV-DNA loads as pre- epidermodysplasia verruciformis, where they occur in high viously described for CIN and AIN (Vinokurova et al., 2005; copy numbers. These patients usually do not develop Kreuter et al., 2007). This makes p16 a useful adjunct for the anogenital lesions (Orth, 2006). b-HPV frequently persist in diagnosis and grading of PIN. the skin and hair bulbs of healthy individuals (de Koning A drawback of our study is the inconsistent observation et al., 2007; Hazard et al., 2007). Only few reports on the periods of the individual patients ranging from a single visit to detection of b-HPV in mucosal/genital lesions have been 38 months. PIN and AIN rates might have been higher if all published (Favre et al., 1998; Ekeowa-Anderson et al., 2007; patients could have been followed for a longer period of time. Mermet et al., 2007). We speculated before that PIN A further limitation is that penile swabs/biopsies were only presenting as EQ could result from co-infection with HR-a- obtained in case of clinically visible lesions and/or high- HPV types and b-HPV type 8 (Wieland et al., 2000). In view resolution anoscopic signs indicating HPV-related disease.
www.jidonline.org 2321 A Kreuter et al. Penile Intraepithelial Neoplasia in HIV þ Men
Therefore some early or discrete PIN lesions could have measuring 3 mm were obtained after local anesthesia in conditions been missed. 1, 2, and 3. All 28 biopsies taken from condylomata acuminata were In summary, we have detected a surprisingly high number histologically confirmed. The same was true for 11 biopsies of HIV-positive AIN patients who additionally developed suspicious of PIN (see Results). All biopsies (n ¼ 22) taken from PIN. Since HIV-positive patients have an increased risk for subclinical lesions revealed no signs of dysplasia. penile cancer (Frisch et al., 2000; Grulich et al., 2007), and Swabs for cytology and HPV analyses were obtained as described since explosive courses of penile SCC in AIDS patients have previously (Kreuter et al., 2005). Cytological results were reported been described (Aboulafia and Gibbons, 2001; Theodore according to the Bethesda system as normal, ASCUS (atypical et al., 2002), all HIV-positive MSM should regularly be squamous cells of undetermined significance), LSIL (low-grade screened for PIN in addition to AIN screening. squamous intraepithelial lesion), HSIL (high-grade squamous in- traepithelial lesion), or cancer. AIN and PIN were histologically MATERIALS AND METHODS classified as grades 1–3 depending on the degree of dysplasia. Patients Histological results were evaluated independently by two histo- Between October 2003 and December 2006, 263 White, pathologists without knowledge of the clinical status. In case of HIV-positive MSM were screened every 6 months for HPV-related discrepancies, all specimens were assessed by a third observer. diseases in the context of a sub-study of the German ‘‘Kompetenz- netzwerk HIV/AIDS’’ basically as previously described (Kreuter HPV-DNA analyses et al., 2005). The medical ethical committee of the Ruhr University DNA isolation and HPV-DNA analyses were performed as described of Bochum approved the study protocol. The study was conducted before (Kreuter et al., 2005; de Koning et al., 2006). Briefly, DNA according to Declaration of Helsinki Principles. Patients gave written was isolated using the QIAamp DNA Mini kit (Qiagen, Hilden, informed consent to participate in the study. At study entry, patients’ Germany) and b-globin gene PCR was carried out to demonstrate ages ranged between 19 and 69 years (mean 41, median 40). CD4 that the samples contained adequate DNA and were free of counts were between 5 and 1,476 ml 1 (mean 476, median 456), and substances inhibitory to PCR (Kreuter et al., 2005). The samples HIV-1 RNA loads were between o40 and 2,710,000 copies ml 1 shown in Table 1 were analyzed by different PCRs for the detection (mean 47,905, median 79). A total of 87 patients were in CDC stage of a- and b-HPV-DNA as previously described (Kreuter et al., 2005; A, 89 in B, 83 in C, and CDC stage was unknown for 4 patients. A de Koning et al., 2006). Briefly, A5-A10 nested PCR was used for the total of 186 (71%) of the 263 patients received highly active detection of a-HPV-DNA (Wieland et al., 2000). For HPV typing, antiretroviral therapy at study entry. A total of 143 patients were internal biotinylated PCR products were hybridized with 37 type- smokers (54%) and 120 non-smokers (46%). The mean follow-up specific digoxigenin-labelled oligonucleotide probes in an enzyme time was 11 months (range 0 (single visit)—38 months). Eight immunoassay (Jacobs et al., 1997; Kreuter et al., 2005). Pluck me patients were circumcised before study entry: one patient (patient 9; (PM)-PCR followed by a reverse hybridization assay (DDL, Table 1) in his first year of life, one patient 25 years before study Voorburg, the Netherlands) was used for the detection and typing entry (phimosis surgery), and six patients between 1 and 5 years of 25 different b-HPV types (de Koning et al., 2006). HPV-DNA loads before study entry (four due to surgical removal of penile were determined by real-time PCRs, with type-specific primers and condylomas (patients 1, 5, 7, 10), one due to Kaposi sarcoma probes for a-HR-HPV types 16, 18, 31, and 33 (Weissenborn et al., surgery, one due to cosmetic (piercing) surgery). 2003), and for b-HPV types 5, 8, 15, 20, 23, 36, and 38, as described earlier (Wieland et al., 2000; Weissenborn et al., 2005). HPV-DNA Clinical examination, cytology, and histology loads were expressed as HPV-DNA copies per b-globin gene copy All participants underwent standardized HPV-screening program at (Weissenborn et al., 2003). the Department of Dermatology of the Ruhr University Bochum, as previously described (Berry et al., 2004; Kreuter et al., 2005). Immunohistochemical staining for p16INK4a Clinical examination included inspection of the oral cavity, glans p16INK4a Staining of 3-mm paraffin-embedded sections of penile penis, corona, sulcus, frenulum, the inner part of the foreskin, biopsies was performed using the CINtec p16INK4a Histology kit scrotum, inguinal folds, perianal skin, anal squamous-columnar (DakoCytomation, Hamburg, Germany) according to the manufac- transitional zone (dentate line), and anal canal. Application of 3% turer’s instructions, as previously described (Kreuter et al., 2007). acetic acid was used to detect subclinical anal and penile lesions. Nuclear staining with or without cytoplasmatic staining was Mucous membranes were 30-fold magnified using high-resolution considered positive for p16INK4a. Quantitative results were expressed anoscopy (Kolposkop 150 FC; Zeiss, Oberkochen, Germany). Penile as the percentage of positive cells per field on total cell count lesions were categorized as follows: (1) condylomata acuminata (Table 1). Only cells within the epidermis were counted. All sections (exophytic lesions with irregular surface and terminal capillaries); (2) were separately evaluated by two histopathologists without knowl- lesions suspicious of PIN (clinical features as reported by Porter et al. edge of the histological diagnosis. If there was a difference of more 2002; Kreuter et al. 2005); (3) subclinical lesions (lesions only visible than 5%, sections had to be re-evaluated by both the observers. after application of acetic acid with signs of HPV-related disease such as punctuation or mosaic as reported by Gross and Pfister, Statistical analysis 2004); and (4) unspecific lesions (acetowhitening seen after Statistical analyses were performed with SPSS 11.0. 1 (SPSS Inc., application of acetic acid, mainly caused by traumatic abrasions, Chicago, IL), using Wilcoxon test (paired samples) and Mann–Whit- superficial erosions, or inflammation such as balanitis without any ney test (independent samples) for assessment of non-normally signs of HPV-infection; (Gross and Pfister, 2004)). Punch biopsies distributed data. Correlation studies were conducted using Spearman
2322 Journal of Investigative Dermatology (2008), Volume 128 A Kreuter et al. Penile Intraepithelial Neoplasia in HIV þ Men
coefficient of rank correlation. The risk of anal dysplasia was Gomousa-Michael M, Gialama E, Gomousas N, Gialama G (2000) Genital estimated by means of odds ratios, including its 95% confidence human papillomavirus infection and associated penile intraepithelial neoplasia in males infected with the human immunodeficiency virus. interval. Since this was an explorative study, we did not constrain Acta Cytol 44:305–9 experiment-wise error rates due to multiple comparisons. P-values Gross G, Pfister H (2004) Role of human papillomavirus in penile cancer, less than 0.05 were considered statistically significant. penile intraepithelial squamous cell neoplasias and in genital warts. Med Microbiol Immunol 193:35–44 CONFLICT OF INTEREST Grulich AE, van Leeuwen MT, Falster MO, Vajdic CM (2007) Incidence of The authors state no conflict of interest. cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: a meta-analysis. Lancet 370:59–67 ACKNOWLEDGMENTS Hazard K, Karlsson A, Andersson K, Ekberg H, Dillner J, Forslund O (2007) We thank Monika Junk, Nabila Ristow, Tanja Blome, Nicole Girrulat, and Cutaneous human papillomaviruses persist on healthy skin. J Invest Barbara Panz for excellent technical assistance. This work was supported by Dermatol 127:116–9 the Federal Ministry of Education and Research (BMBF), German Competence Jacobs MV, Snijders PJ, van den Brule AJ, Helmerhorst TJ, Meijer CJ, Network HIV/AIDS, Grant No. 01 KI 0501, and by Koeln Fortune, Faculty of Walboomers JM (1997) A general primer GP5+/GP6(+)-mediated PCR- Medicine of University of Cologne, Grant No. 167/2006. enzyme immunoassay method for rapid detection of 14 high-risk and 6 low-risk human papillomavirus genotypes in cervical scrapings. J Clin Microbiol 35:791–5 REFERENCES Kreuter A, Brockmeyer NH, Hochdorfer B, Weissenborn SJ, Stucker M, Swoboda J et al. (2005) Clinical spectrum and virologic characteristics of Aboulafia DM, Gibbons R (2001) Penile cancer and human papilloma virus anal intraepithelial neoplasia in HIV infection. J Am Acad Dermatol (HPV) in a human immunodeficiency virus (HIV)-infected patient. 52:603–8 Cancer Invest 19:266–72 Kreuter A, Brockmeyer NH, Weissenborn SJ, Wafaisade A, Pfister H, Altmeyer Aynaud O, Ionesco M, Barrasso R (1994) Penile intraepithelial neoplasia. P et al. (2006) 5% Imiquimod suppositories decrease the DNA load of Specific clinical features correlate with histologic and virologic findings. Cancer 74:1762–7 intra-anal HPV types 6 and 11 in HIV-infected men after surgical ablation of condylomata acuminata. Arch Dermatol 142:243–4 Barrasso R, De Brux J, Croissant O, Orth G (1987) High prevalence of papillomavirus-associated penile intraepithelial neoplasia in sexual Kreuter A, Wieland U, Gambichler T, Altmeyer P, Pfister H, Tenner-Racz K partners of women with cervical intraepithelial neoplasia. N Engl J et al. (2007) p16(ink4a) Expression decreases during imiquimod Med 317:916–23 treatment of anal intraepithelial neoplasia in human immunodeficiency virus-infected men and correlates with the decline of lesional high-risk Bernard C, Mougin C, Madoz L, Drobacheff C, Van Landuyt H, Laurent R human papillomavirus DNA load. Br J Dermatol 157:523–30 et al. (1992) Viral co-infections in human papillomavirus-associated anogenital lesions according to the serostatus for the human immuno- Lont AP, Kroon BK, Horenblas S, Gallee MP, Berkhof J, Meijer CJ et al. (2006) deficiency virus. Int J Cancer 52:731–7 Presence of high-risk human papillomavirus DNA in penile carcinoma predicts favorable outcome in survival. Int J Cancer 119:1078–81 Berry JM, Palefsky JM, Welton ML (2004) Anal cancer and its precursors in HIV-positive patients: perspectives and management. Surg Oncol Clin N Mermet I, Guerrini JS, Cairey-Remonnay S, Drobacheff C, Faivre B, Gaillard Am 13:355–73 M et al. (2007) Cervical intraepithelial neoplasia associated with epidermodysplasia verruciformis HPV in an HIV-infected patient: a Cupp MR, Malek RS, Goellner JR, Smith TF, Espy MJ (1995) The detection of manifestation of immune restoration syndrome. Eur J Dermatol human papillomavirus deoxyribonucleic acid in intraepithelial, in situ, 17:149–52 verrucous and invasive carcinoma of the penis. J Urol 154:1024–9 Munoz N, Bosch FX, de Sanjose S, Herrero R, Castellsague X, Shah KV et al. de Koning M, Quint W, Struijk L, Kleter B, Wanningen P, van Doorn LJ et al. (2003) Epidemiologic classification of human papillomavirus types (2006) Evaluation of a novel highly sensitive, broad-spectrum PCR- associated with cervical cancer. N Engl J Med 348:518–27 reverse hybridization assay for detection and identification of beta- papillomavirus DNA. J Clin Microbiol 44:1792–800 Nielson CM, Flores R, Harris RB, Abrahamsen M, Papenfuss MR, Dunne EF et al. (2007) Human papillomavirus prevalence and type distribution in de Koning MN, Struijk L, Bavinck JN, Kleter B, ter Schegget J, Quint WG et al. male anogenital sites and semen. Cancer Epidemiol Biomarkers Prev (2007) Betapapillomaviruses frequently persist in the skin of healthy 16:1107–14 individuals. J Gen Virol 88:1489–95 Orth G (2006) Genetics of epidermodysplasia verruciformis: Insights into host de Villiers EM, Fauquet C, Broker TR, Bernard HU, zur Hausen H (2004) defense against papillomaviruses. Semin Immunol 18:362–74 Classification of papillomaviruses. Virology 324:17–27 Palefsky J (2006) Human papillomavirus-related tumors in HIV. Curr Opin Dillner J, von Krogh G, Horenblas S, Meijer CJ (2000) Etiology of squamous Oncol 18:463–8 cell carcinoma of the penis. Scand J Urol Nephrol Suppl 205:189–93 Palefsky JM, Holly EA, Efirdc JT, Da Costa M, Jay N, Berry JM et al. (2005) Ekeowa-Anderson AL, Harwood CA, Perrett CM, Sahota A, Annan H, Ran H Anal intraepithelial neoplasia in the highly active antiretroviral therapy et al. (2007) Vulval intraepithelial neoplasia and periungual Bowen0s era among HIV-positive men who have sex with men. AIDS 19:1407–14 disease concordant for mucosal (HPV-34) and epidermodysplasia verruciformis (HPV-21) human papillomavirus types. Clin Exp Dermatol Palefsky JM, Holly EA, Ralston ML, Jay N (1998) Prevalence and risk factors 32:304–7 for human papillomavirus infection of the anal canal in human immunodeficiency virus (HIV)-positive and HIV-negative homosexual Favre M, Majewski S, De Jesus N, Malejczyk M, Orth G, Jablonska S (1998) A possible vertical transmission of human papillomavirus genotypes men. J Infect Dis 177:361–7 associated with epidermodysplasia verruciformis. J Invest Dermatol Porter WM, Francis N, Hawkins D, Dinneen M, Bunker CB (2002) Penile 111:333–6 intraepithelial neoplasia: clinical spectrum and treatment of 35 cases. Br Ferreux E, Lont AP, Horenblas S, Gallee MP, Raaphorst FM, von Knebel J Dermatol 147:1159–65 Doeberitz M et al. (2003) Evidence for at least three alternative Rubin MA, Kleter B, Zhou M, Ayala G, Cubilla AL, Quint WG et al. (2001) mechanisms targeting the p16INK4A/cyclin D/Rb pathway in penile Detection and typing of human papillomavirus DNA in penile carcinoma, one of which is mediated by high-risk human papilloma- carcinoma: evidence for multiple independent pathways of penile virus. J Pathol 201:109–18 carcinogenesis. Am J Pathol 159:1211–8 Frisch M, Biggar RJ, Goedert JJ (2000) Human papillomavirus-associated Sano T, Oyama T, Kashiwabara K, Fukuda T, Nakajima T (1998) Expression cancers in patients with human immunodeficiency virus infection and status of p16 protein is associated with human papillomavirus oncogenic acquired immunodeficiency syndrome. J Natl Cancer Inst 92:1500–10 potential in cervical and genital lesions. Am J Pathol 153:1741–8
www.jidonline.org 2323 A Kreuter et al. Penile Intraepithelial Neoplasia in HIV þ Men
Senba M, Kumatori A, Fujita S, Jutavijittum P, Yousukh A, Moriuchi T et al. Vinokurova S, Wentzensen N, von Knebel Doeberitz M (2005) Analysis of (2006) The prevalence of human papillomavirus genotypes in penile p16INK4a and integrated HPV genomes as progression markers. cancers from northern Thailand. J Med Virol 78:1341–6 Methods Mol Med 119:73–83 Sirera G, Videla S, Pinol M, Canadas MP, Llatjos M, Ballesteros AL et al. Weissenborn SJ, Funke AM, Hellmich M, Mallmann P, Fuchs PG, Pfister HJ (2006) High prevalence of human papillomavirus infection in the anus, et al. (2003) Oncogenic human papillomavirus DNA loads in human penis and mouth in HIV-positive men. AIDS 20:1201–4 immunodeficiency virus-positive women with high-grade cervical Smith JS, Lindsay L, Hoots B, Keys J, Franceschi S, Winer R et al. (2007) lesions are strongly elevated. J Clin Microbiol 41:2763–7 Human papillomavirus type distribution in invasive cervical cancer and Weissenborn SJ, Nindl I, Purdie K, Harwood C, Proby C, Breuer J et al. (2005) high-grade cervical lesions: a meta-analysis update. Int J Cancer Human papillomavirus-DNA loads in actinic keratoses exceed those in 121:621–32 non-melanoma skin cancers. J Invest Dermatol 125:93–7 Snijders PJ, Hogewoning CJ, Hesselink AT, Berkhof J, Voorhorst FJ, Bleeker Wieland U, Brockmeyer NH, Weissenborn SJ, Hochdorfer B, Stucker M, MC et al. (2006) Determination of viral load thresholds in cervical Swoboda J et al. (2006) Imiquimod treatment of anal intraepithelial scrapings to rule out CIN 3 in HPV 16, 18, 31 and 33-positive women neoplasia in HIV-positive men. Arch Dermatol 142:1438–44 with normal cytology. Int J Cancer 119:1102–7 Wieland U, Jurk S, Weissenborn S, Krieg T, Pfister H, Ritzkowsky A (2000) Theodore C, Androulakis N, Spatz A, Goujard C, Blanchet P, Wibault P Erythroplasia of queyrat: coinfection with cutaneous carcinogenic (2002) An explosive course of squamous cell penile cancer in an AIDS human papillomavirus type 8 and genital papillomaviruses in a patient. Ann Oncol 13:475–9 carcinoma in situ. J Invest Dermatol 115:396–401 Van der Snoek EM, Niesters HG, Mulder PG, van Doornum GJ, Osterhaus Ylitalo N, Sorensen P, Josefsson AM, Magnusson PK, Andersen PK, Ponten J AD, van der Meijden WI (2003) Human papillomavirus infection in men et al. (2000) Consistent high viral load of human papillomavirus 16 and who have sex with men participating in a Dutch gay-cohort study. Sex risk of cervical carcinoma in situ: a nested case-control study. Lancet Transm Dis 30:639–44 355:2194–8
2324 Journal of Investigative Dermatology (2008), Volume 128