Fax: +31-20-444 2964
Cancer Cell Biology
HPV-associated flat penile lesions in men of a non-STD hospital population: Less frequent and smaller in size than in male sexual partners of women with CIN
Article first published online: 25 AUG 2004
Copyright © 2004 Wiley-Liss, Inc.
International Journal of Cancer
Volume 113, Issue 1, pages 36–41, 1 January 2005
How to Cite
Bleeker, M. C.G., Hogewoning, C. J.A., Voorhorst, F. J., van den Brule, A. J.C., Berkhof, J., Hesselink, A. T., Lettink, M., Starink, T. M., Stoof, T. J., Snijders, P. J.F. and Meijer, C. J.L.M. (2005), HPV-associated flat penile lesions in men of a non-STD hospital population: Less frequent and smaller in size than in male sexual partners of women with CIN. Int. J. Cancer, 113: 36–41. doi: 10.1002/ijc.20502
- Issue published online: 25 OCT 2004
- Article first published online: 25 AUG 2004
- Manuscript Accepted: 7 JUN 2004
- Manuscript Received: 9 MAY 2004
- ZON-MW. Grant Number: 28-2725
- penile HPV;
- viral load;
- flat penile lesions
Human papillomavirus (HPV) infections and HPV-associated penile lesions are frequently found in male sexual partners of women with cervical intraepithelial neoplasia (CIN). To determine the significance of these findings, we studied the prevalence of HPV and HPV associated penile lesions in a male hospital population with non-STD complaints. Penoscopy was performed after application of acetic acid to identify flat lesions, papular lesions, condylomata acuminata and pearly penile papules (PPPs). Presence of HPV DNA in penile scrapes was tested by GP5+6+ PCR. In case of HPV 16 positivity, viral loads were quantified using a LightCycler based real-time PCR method. Comparing the non-STD male hospital population (n = 118) with the male sexual partners of women with CIN (n = 238), flat penile lesions were found in 14% vs. 60% and penile HPV in 25% vs. 59% of the men, respectively. We found that the presence of penile HPV and, in case of HPV 16 positivity, higher viral loads were associated with the presence of flat penile lesions. Amongst the HPV-positive men, flat penile lesions were more common and larger in size in male sexual partners of women with CIN than in the non-STD hospital population. HPV infections and HPV-associated flat penile lesions are commonly found in the non-STD male population. However, these lesions are less frequently present and smaller in size than in male sexual partners of women with CIN. Higher viral loads in penile scrapes of male sexual partners of women with CIN are reflected by a higher prevalence of flat penile lesions and a larger size of these lesions.
Extensive research in women during the past decades has led to the conclusion that sexually transmitted human papillomavirus (HPV) infection is etiologically linked to precancerous and invasive neoplastic lesions of the cervix. In fact, high-risk types of HPV have been detected in 99.7% of the cervical carcinomas.1
Multiple studies have reported a role for the male sexual partner in the epidemiology of carcinoma of the cervix.2, 3, 4 However, studies on penile lesions and HPV infection in men are scarce and have been mainly restricted to “high risk” populations such as those visiting clinics for sexually transmitted diseases (STD) or male partners of women with HPV-associated diseases.5, 6, 7, 8, 9, 10 These studies, mainly reported from the late 1980s to the early 1990s, showed a wide spectrum of HPV-associated penile lesions varying from classical condylomata acuminata to lesions consistent with severe penile intraepithelial neoplasia (PIN). More recently, we described penile lesions in 68% of the male sexual partners of women with cervical intraepithelial neoplasia (CIN).11 Most of these lesions were flat and could be clearly visualised only after application of acetic acid. By polymerase chain reaction (PCR), HPV DNA was detected in 59% of the corresponding penile scrapes.
Interpretation of these findings, however, was hampered by lack of data about HPV infections and HPV-associated penile lesions in men from the general population. These data are necessary to obtain sufficient insight in the epidemiology and clinical manifestations of HPV infection in men.
Here we present data about penile lesions and presence of HPV in a male non-STD hospital population. The findings were compared to data in male sexual partners of women with CIN. In case of HPV 16 positivity, viral DNA loads were quantified and related to the presence of flat penile lesions.
Material and methods
From April 2002 until November 2002, men between 18 and 75 years, visiting the outpatient non-STD clinic of the Department of Dermatology of the VU Medical Center in Amsterdam were asked to participate in the study. Not eligible were men who consulted the clinic because of a sexually transmitted disease or those having anogenital cancer. Of 181 eligible men, 156 were willing to participate. These men visited the Department of Dermatology because of nongenital skin cancer screening (n = 21), pigmented naevi (n = 21), eczema (n = 28), allergy (n = 20), psoriasis (n = 17), nongenital warts (n = 8), syndrome related skin disorders (n = 6), itching (n = 4), acne (n = 4) or other nongenital dermatological problems (n = 27). Of the 156 men of the non-STD hospital population, 118 (76%) did have a female partner and 38 (24%) did not. In the latter population 4 men reported a male partner and 34 no steady relationship. In the present study, only findings of men with a female partner were presented (population A). The presence of CIN in the female partners of these men was not evaluated and therefore unknown. However, men were asked for the presence of abnormal cervical cytology and other cervical problems in their partner.
Male sexual partners of women with CIN (n = 238) were recruited at the colposcopy clinic of the Albert Schweitzer hospital in Dordrecht (the Netherlands) from December 1995 to June 2002. Details of the latter group of men have been described elsewhere.11, 12 Penoscopic findings and HPV detection in penile scrapes from 175 of the 238 men with a female partner with CIN were previously published.11 The study protocols were approved by the ethics review board of the hospitals and all participants signed informed consent.
Specimen collection and penoscopic examination
For HPV testing, a Cervex brush® was used to collect cells from the top of penis (including the glans, corona, sulcus, frenulum and the inner part of the foreskin). The brush was collected in a phosphate-buffered saline solution, sent to the laboratory and stored at 4°C until processing. After penile scraping and several minutes after the application of 3% acetic acid solution, penoscopy was performed to identify lesions. For penoscopy, a colposcope was used with a magnification factor ranging from 6 to 16 times (Olympus OCS-AW4) and findings were documented by photographs. Penile lesions were classified as flat lesions, papular lesions and condylomata acuminata as described and illustrated earlier.11 When penile lesions were present, it was estimated whether the total area of penile epithelium affected was smaller or larger than 5 mm2 in size. In addition, the presence of pearly penile papules, a normal benign variation, was administrated.
After vortexing, suspensions of penile scrapes were centrifuged, resuspended in 100 to 200 μL TE (10 mmol/L TRIS, 1 mmol/L EDTA, pH 7.5) and stored at −80°C until use. Initially, aliquots of 10 μL of crude extracts were used for GP5+/6+ enzyme immunoassay (EIA) PCR, as described previously.13, 14 This method enables group detection of 14 high-risk types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68) and 23 low-risk types (6, 11, 26, 34, 40, 42, 43, 44, 53, 54, 55, 57, 61, 70, 71, 72, 73, 81, 82/MM4, 82/MM4, 83, 84 and CP6108) by using a high-risk and low-risk oligoprobe cocktail, respectively. On HPV-positive cases, individual typing was performed for high-risk HPV 16, 18, 31 and 33, and for low-risk HPV 6 and 11, using type specific probes in an EIA format. All HPV PCR-negative samples were subjected to a β-globin PCR, as DNA quality control of the sample. Specimens that were negative in both HPV and β-globin PCR assays were excluded from the HPV analyses because of inadequate sample quality.
In order to select samples to be used for subsequent HPV 16 viral load assessment, DNA was extracted from half of the penile scrape suspension left using the HPPTP kit (Roche, Mannheim, Germany). These samples were subjected to another GP5+6+ PCR EIA round followed by typing specific for HPV 16.
Quantification of HPV 16 and β-globin by a LightCycler based real-time PCR
Viral load testing of HPV 16 positive purified DNA extracts was done using a LightCycler based real-time HPV 16 type specific PCR method, as described by Hesselink et al. (manuscript in preparation). For this assay, primers and probes had been selected within the E7 region. Sequences of HPV 16 oligonucleotides were the following: forward PCR primer 5′-gaggaggaggatgaaatagatggt-3′(nt position 658–681), reverse PCR primer 5′-gcccattaacaggtcttccaa-3′(nt position 796–817), donor probe 5′-acaaaaggttacaatattgtaatgggctct-3′fluorescence (nt position 735–706) and acceptor probe red 640 5′-ccggttctgcttgtccagctgg-3′ (nt position 703–682). Amplification of the β-globin gene was used to quantify the amount of cellular DNA per sample. Sequences of β-globin oligonucleotides were the following: forward PCR primer 5′-gagccatctattgcttacatttgc-3′(nt position 62122–62145), reverse PCR primer 5′-ttggtctccttaaacctgtcttgt-3′(nt position 62318–62295), donor probe 5′-ccagggcctcaccaccaacttc-3′fluorescence (nt position 62275–62254) and acceptor probe red 640 5′-ccacgttcaccttgccccacag-3′ (nt position 62251–62230). For quantification of HPV 16, standard curves were generated using 10-fold dilutions of HPV 16 plasmid (ranging from 10 to 100,000 fg per reaction) spiked in 100 ng human placenta DNA (hpDNA). The β-globin gene was quantified with the aid of standard curves of 10-fold dilutions of hpDNA ranging from 3 pg to 30 ng per reaction. LightCycler tests were run in duplicate and data were averaged for ultimate calculation of viral load.
Between the populations (population A vs. B), potential differences in age, circumcision rate and ethnicity were examined using t-tests, chi-square tests and Fisher exact tests, where appropriate. Odds ratios (OR) and 95% confidence intervals (CI) were computed of the prevalence of penile lesions and HPV. Stratified analyses were performed to adjust for age differences between 2 populations.15 Each population was divided in homogeneous age-strata (<35, 35–45 and >45years) and the OR was computed for each stratum separately. As a following step, Breslow and Day's test of homogeneity was applied to check whether the stratum-specific ORs could be pooled into the Mantel Haenszel's common odds ratio (ORMH).16 An analogous approach was used restricting the stratified analyses to men who were not circumcised. Determinants for flat penile lesions and penile HPV were estimated within each cohort using logistic regression analyses. Viral load values were analysed both as continuous and categorised variables, using Mann-Whitney or χ2 tests, respectively. HPV 16 positive samples which were negative in LightCycler PCR tests, thus containing number of HPV copies or cells under the detection level, were set at 10 copies or 5 cells input, respectively13 (Hesselink et al., manuscript in preparation). Ten copies input per PCR reaction corresponds to approximately 250 copies per penile scrape. Analysing HPV copies per cell, samples were excluded when both number of HPV 16 copies and cells were under the detection level. Level of statistical significance was set at 0.05 (2-sided).
Characteristics of the study populations
Of the 118 men of the non-STD hospital population with a female partner, the mean age was 46.1 years (range 22.8–73.2). Circumcised were 21 (17.8%) men and non-Caucasian were 13 (11%) men.
Male sexual partners of women with CIN (n = 238) averaged 37.6 years (range 22.5–57.7). Eleven (5%) of them were circumcised and 13 (5%) men were non-Caucasian. In population B, the mean age was lower and less widely ranged (p ≤ 0.001) and less men were circumcised compared to population A (p ≤ 0.001). Number of non-Caucasian men in population B was not significantly different from population A (p = 0.058).
Penile lesions, HPV detection, and HPV 16 viral load
Flat penile lesions were found in 16 (14.3%) men of population A (Table I). In 5 (31.3%) of these men more than 5 mm2 of the penile epithelium was affected. Condylomata acuminata were not found in this population. Papular lesions were found in 23 (20.5%) and PPPs in 27 (24.1%) men. Flat penile lesions were found in 139 (60.4%) men of population B which was significantly more prevalent compared to population A (ORMH 9.6, 95% CI 5.2–17.9). Flat penile lesions in population B sized more than 5 mm2 in 75% of the men indicating a higher number of men with larger sized lesions than in population A. Condylomata acuminata were found in 11 (4.8%) men, which was significantly more prevalent than in population A (p = 0.019). Papular lesions and PPPs were found in 49 (21.3%) and in 79 (34.3%) men, respectively. The prevalences of papular lesions and PPPs were not significantly different in both populations. Not suitable for penoscopic evaluation due to previous surgery or phimosis were 6 (5.1%) men of population A and 8 (3.4%) men of population B.
|Population||A (N %)||B (N %)||A vs. B ORMH (95% CI)1||A vs. B uncircumcised ORMH (95% CI)1|
|Not suitable for evaluation2||6||8|
|Flat penile lesions|
|No||96||85.7||91||39.6||1.0 reference||1.0 reference|
|Yes||16||14.3||139||60.4||9.6 (5.2–17.9)||7.8 (4.1–14.6)|
|No||89||79.5||181||78.7||1.0 reference||1.0 reference|
|Yes||23||20.5||49||21.3||1.0 (0.6–1.8)||1.1 (0.6–2.1)|
|No||85||75.9||151||65.7||1.0 reference||1.0 reference|
|Yes||27||24.1||79||34.3||1.3 (0.8–2.2)||1.3 (0.7–2.4)|
|Penile HPV detection|
|GP5+/6+ HPV PCR EIA|
|HPV negative||62||74.7||69||40.6||1.0 reference||1.0 reference|
|HPV positive||21||25.3||101||59.4||4.2 (2.3–7.6)||3.7 (2.0–7.0)|
HPV DNA was detected in 21 (25.3%) penile scrapes of population A (Table I) vs. 101 (59.4%) in population B (ORMH 4.2, 95% CI 2.3–7.6). Proportions of penile scrapes that were negative in both HPV and β-globin PCR assays were similar in the populations, i.e., 29.7% in population A and 28.6% in population B.
In HPV-positive individuals, infection with a single HPV type was most common, i.e., 16 (76.2%) men of population A and 81 (80.2%) men of population B (Table II). In both populations, infection with high-risk types was more prevalent than infection with low-risk types. However, HPV 16 was less frequently detected in population A than in population B, i.e., in 4 (19.0%) vs. 47 (46.5%) men, respectively (p = 0.028). Low-risk types were more frequently found in population A than in population B, i.e., in 10 (47.6%) and in 32 (31.7%) men but this difference was not statistically significant (p = 0.257).
|Population||A (N %)||B (N %)||A vs. B p|
|HPV positive subjects||21||101|
HPV 16 viral load
Compared to HPV testing of crude extracts, higher numbers of HPV 16-positive penile scrapes were found after testing purified DNA extracts indicating the higher analytical sensitivity of the GP5+6+ PCR assay using purified DNA.17 HPV 16 was detected in purified DNA extracts of 20 penile scrapes of population A and in 58 scrapes derived from population B. HPV 16 DNA quantification of these samples showed that less HPV 16 copy numbers were found in scrapes derived from population A compared to those from population B (Table III). One out of the 58 HPV 16-positive men of population B could not be tested for viral load due to insufficient sample material. Eighteen of 20 (90.0%) tested samples of population A yielded <10 HPV copies input, corresponding with less than 250 HPV 16 copies per scrape (range <250–2,712). This proportion was significantly lower (p = 0.024) in population B, i.e., 35/57 (61.4%) samples counted less than 250 copies (range <250–389,026; p = 0.015). When only scrapes with >250 copies were considered, a median of 1,572 copies (range 432–2,712; n = 2) was found in penile scrapes of population A vs. 2,834 copies (range 267–389,026; n = 22) in scrapes of population B (p = 0.35). A median of 1.2 copies per cell (range 0.1–11.4; n = 8) was found in scrapes of population A vs. 5.0 copies per cell (range 0.0–210.0; n = 28) in population B (p = 0.033). Twelve of 20 (60%) samples from population A and 29/57 (51%) samples from population B yielded both <10 copies and <5 cells input and were therefore excluded from the analyses on HPV 16 copy numbers per cell.
|Population||A (N %)||B (N %)||A vs. B p|
|HPV 16 positive subjects||20||57|
|HPV 16 ≈ <250 copies/scrape||18||90.0||35||61.4||0.0241|
|HPV 16 ≈ >250 copies/scrape||2||10.0||22||38.6|
Determinants of flat penile lesions and penile HPV
Determinants of flat penile lesions are presented in Table IV. The positive association between the presence of penile HPV and the presence of flat penile lesions was found both in population A (OR 6.4, 95% CI 1.9–22.2) and in population B (OR 3.2, 1.6–6.2). Both in HPV-positive and in HPV-negative men, rates of flat penile lesions were higher in population B than in population A. Considering HPV-positive subjects, flat penile lesions were found in 77.1% in population B vs. 42.1% in population A (ORMH 4.7, 95% 1.6–13.3). In HPV-negative men, these figures were 51.5% vs. 10.2%, respectively (ORMH 8.3, 95% 3.1–22.7). Age was not related to the presence of flat penile lesions. In population A, all flat penile lesions were found among uncircumcised men and none among those circumcised (p = 0.069).
|Total N||Flat+||OR (95% CI)||Total N||Flat+||OR (95% CI)|
|Not in analyses3||34||62||31|
|Negative||59||6||10.2||1.0 Reference||66||34||51.5||1.0 Reference|
|Positive||19||8||42.1||6.4 (1.9–22.2)||96||74||77.1||3.2 (1.6–6.2)|
|< 35 years||29||5||17.2||1.0 Reference||94||58||61.7||1.0 Reference|
|35–45 years||29||4||13.8||0.8 (0.2–3.2)||88||50||56.8||0.8 (0.5–1.5)|
|> 45 years||54||7||13.0||0.7 (0.2–2.5)||42||31||73.8||1.7 (0.8–3.9)|
Neither age nor circumcision was associated with the presence of penile HPV. In population A, HPV positive were 6/19 (31.1%) men <35 years, 7/24 (29.2%) men between 35 and 45 years (OR 0.9, 95% CI 0.3–3.3) and 8/40 (20.0%) men >45 years (OR 0.5, 95% CI 0.2–1.9). In population B, these figures were 41/71 (57.7%), 38/65 (58.5%; OR 1.0, 95% CI 0.5–2.0) and 22/34 (64.7%; OR 1.3, 95% CI 0.6–3.1), respectively. HPV positive were 18/70 (25.7%) uncircumcised and 3/13 (23.1%) circumcised men (OR 0.9, 95% CI 0.2–3.5) from population A. In population B, 93/160 (58.1%) uncircumcised and 8/10 (80.0%; OR 2.9, 95% CI 0.6–14.0) circumcised men were HPV positive.
The association between HPV 16 viral load and flat penile lesions
When HPV 16-positive subjects were considered, more HPV 16 copies were present in scrapes derived from men with flat penile lesions compared to those without (Table V). Per scrape, all HPV 16-positive men without flat penile lesions (n = 28) yielded less than 250 HPV 16 copies vs. 267 (range <250–389,026; n = 45) when flat penile lesions were present (p <0.001). Per cell, the median copy number was 0.7 (range 0.3–1.7; n = 4) when lesions were absent vs. 4.3 (range 0.0–210.0; n = 31) when lesions were present (p = 0.070). In population A, 100% of the penile scrapes of men without flat penile lesions yielded less than 250 HPV 16 copies (n = 15) vs. 60% when lesions were present (range <250–2,712; n = 5; p = 0.012). Median copies per cell were 1.1 (range 0.3–1.7; n = 3) and 1.4 (range 0.1–11.4; n = 5), respectively (p = 0.9). In population B, these figures were <250 copies per scrape when lesions were absent (n = 13) vs. 342 copies per scrape (range <250–389,026; n = 40) when lesions were present (p = 0.001). Copies per cell were 0.3 (n = 1) vs. 5.0 (range 0.0–210.0; n = 26), respectively (p = 0.2). Besides the presence of higher viral loads when flat penile lesions are present these data suggest that, in case of prevalent flat penile lesions, higher viral loads are present in penile scrapes of population B than in those of population A. However, these differences were not statistically significant (p = 0.431 for copies per scrape and p = 0.107 for copies per cell).
|Population||A + B||A||B|
|Flat lesions at penoscopy||Flat −||Flat +||p||Flat −||Flat +||p||Flat −||Flat +||p|
|Number of HPV 16 copies|
|Under detection level (<250)||28||100||22||49||<0.001||15||100||3||60||0.053||13||100||19||48||0.001|
|Above detection level (>250)||—||0||23||51||—||0||2||40||—||0||21||52|
|Per scrape (median/range)||<250||267/<250–389,026.4||<0.001|
|Per cell (median/range)3||0.7/0.3–1.7||4.3/0.0–210.0||0.070|
In order to determine the significance of the high prevalence of penile HPV DNA and HPV-associated penile lesions in male sexual partners of women with CIN, we evaluated a male non-STD population, visiting the outpatient clinic of the Department of Dermatology as a representative of the general male population in the Netherlands. Besides the lower prevalence of penile HPV and HPV-associated penile lesions, lower HPV 16 viral loads were found in penile scrapes of this non-STD hospital population compared to the group of male sexual partners of women with CIN. Consistent with the presence of lower viral loads, we found that flat penile lesions were not only less frequently observed but also smaller in size in the non-STD hospital population than in male sexual partners of women with CIN.
Our findings on the prevalence of flat penile lesions in both populations are consistent with data presented in literature. Based on small populations, published rates of flat penile lesions found in male partners of women without HPV-associated lesions range from 2 to 21%.6, 10, 18, 19 The highest rate was found in a hospital-based population of nearly a hundred men attending an infertility clinic.19 The lower rates were found in male partners of women with non-HPV-related gynaecologic problems,18 chlamydial cervicitis10 or normal cytology in their cervical smears.6 For male partners of women with abnormal cervical cytology, rates between 50% and 63% have been reported.6, 8, 10, 18 Besides the lower prevalence of flat penile lesions, we found that lesions sized smaller in the non-STD hospital population than in the male sexual partners of women with CIN. The prevalence of condylomata acuminata in the populations investigated was low though our findings indicate that they were more prevalent in male sexual partners of women with CIN. This difference however might be biased as men of the non-STD population were considered not eligible to participate in this study when the clinic was visited because of a sexual transmitted disease, for example, genital condylomata. On the other hand, one would expect an elevated rate of genital condylomata in male sexual partners of women with CIN because infections with both high- and low-risk HPV types share identical sex-related risk factors.20 In both populations, we found similar rates of PPPs and papular lesions. These findings support earlier observations that there exists no association between HPV and these penile features.12, 21, 22
As earlier described by us for male sexual partners of women with CIN,11 the association between the presence of flat penile lesions and HPV detection in penile scrapes was also found in men of the non-STD hospital population who had a female partner. Consistent with the lower prevalence of flat penile lesions, penile HPV was less frequently detected in men of the non-STD population compared to those having a female partner with CIN. Case-control studies in populations at high (Colombia) and low (Spain) risk for cervical cancer have also found higher rates of penile HPV in husbands of women with cervical neoplasia compared to those without.3 However, in comparison to our study, lower rates of penile HPV were reported in these studies. Using PCR techniques, HPV was detected in 17% of the case patients and in 5% of the control subjects in Spain and these rates were 26% and 19%, respectively, in Colombia. In part this could be due to the fact that a different PCR method was used for these studies. Additionally, in these studies, penile sampling included the distal urethra as well as the external surface of the penis. A low prevalence of HPV and HPV-associated lesions in the urethra23, 24, 25 in combination with a more easy collection of cells sampled from this nonkeratinised site of the penis may be responsible for this apparently lower HPV detection rate.
Considering HPV-positive subjects, the majority was positive for a high-risk type. However, HPV 16 was less frequently detected in men of the non-STD population than in male sexual partners of women with CIN. Interestingly, this difference was less pronounced using purified extracts. In our experience, the use of purified DNA enables detection of lower viral loads, i.e., the method has a higher analytical sensitivity compared to crude extracts.17 Apparently, higher viral loads are present in male sexual partners of women with CIN. Indeed, HPV 16 viral load quantification on purified DNA showed that higher copy numbers were present in penile scrapes of male sexual partners of women with CIN than in those of the non-STD male population. The presence of higher viral loads is a likely explanation for the finding that flat penile lesions were not only more common but also larger in size in male partners of women with CIN. This is substantiated by the observed association between viral load and flat penile lesions. All HPV 16-positive subjects without flat penile lesions yielded low levels of HPV copy numbers whereas higher numbers were found in scrapes of men with flat penile lesions. These findings indicate that a considerable part of penile HPV infections do not result in development of penile lesions.
The non-STD population was composed of men who visited a non-STD Department of Dermatology, who were generally in good health and who were not selected for age or other specific characteristics that might be HPV related. The advantage of this approach was that these men were likely to have a similar risk for HPV infections and related lesions as a general male population in the Netherlands. However, the consequence is that this study group did not match with the population of men selected on the basis of a prevalent CIN lesion in the female partner. Indeed, substantial differences between both populations were present for age and history of circumcision. Possible bias due to this selection however is unlikely because our analyses showed very similar risk estimates for the presence of penile lesions and HPV after adjustment for these demographic characteristics. Moreover, both cohorts showed no particular relationship between age and the presence of flat penile lesions or HPV. Therefore, it is more likely that the observed difference in HPV prevalence is owing to the presence vs. absence of a female partner with CIN and consequently a differential risk of being exposed to HPV of the sexual partner.
Lack of association between age and penile HPV has also been described by others.3, 26, 27 On the contrary, in women HPV prevalence is age related.28, 29, 30 In Dutch women with normal cytology, the HPV prevalence is decreasing from 20% in women aged between 20 and 24 years to 6% in women over 30 years.28 An HPV prevalence of 25% in non-STD men with a mean age of 46 years suggests that HPV infection is more common among men than among women, especially when taken into account that the number of HPV copies collected in penile scrapes is much lower compared to those collected from the cervix and might result in false negativity when testing penile scrapes.
Our data indicate that the presence of flat penile lesions is related to the presence of higher viral loads of HPV, whereas HPV detection without the presence of flat penile lesions indicates very low levels. Therefore, identification of flat penile lesions is highly informative in the evaluation of HPV infection in men.
In conclusion, penile HPV detection and HPV-associated flat penile lesions are not uncommon in the male non-STD hospital population but were significantly less prevalent than in male sexual partners of women with CIN. Detection of penile HPV without the presence of HPV-associated lesions indicates very low levels of HPV DNA. The presence of higher viral loads in penile scrapes of male sexual partners of women with CIN explains not only the higher prevalence of flat penile lesions but also the larger size of these lesions.
We thank P.M. van Diemen for her excellent technical assistance on DNA purification and HPV testing. The medical doctors of the outpatient clinic of the Department of Dermatology are gratefully thanked for the introduction of their male patients.
- 15Epidemiology: an introduction. New York: Oxford University Press, 2002. 144–67..
- 29POBASCAM, a population-based randomized controlled trial for implementation of high-risk HPV testing in cervical screening: design, methods and baseline data of 44,102 women. Int J Cancer 2004; 110: 94–101, , , , , , , , , , , , et al.