Incidence and clearance of penile human papillomavirus infection among circumcised Kenyan men

Abstract

Prospective data are limited on human papillomavirus (HPV) acquisition and clearance among circumcised men from resource-limited geographical regions, particularly Africa. The goal of this study was to estimate incidence and clearance of type-specific genital HPV infection in men. Penile exfoliated cell specimens were collected from the glans/coronal sulcus and shaft of 1,037 circumcised Kenyan men at baseline and 6-, 12- and 18-month follow-up visits between 2003–2007. Specimens were tested with GP5+/6+ PCR to detect 44 HPV types. The median age of participants at baseline was 21 years (range 18–28). The 12- and 18-month incidence rates (IRs) for any HPV were 34.9/100 person-years (95% confidence interval [CI]: 31.2–39.0) and 36.4/100 person-years (95% CI: 32.9–40.2), respectively. The 18-month cumulative risk for high-risk HPV was 30% compared to 16% for low-risk HPV. Cumulative risk was not associated with age or anatomical site. The estimated probability of any HPV infection clearing by 12 months was 0.92. Time until HPV clearance was not associated with age, anatomical site, or whether HPV infection type was high-risk or low-risk. HPV IRs among circumcised men in this study were comparable to other circumcised populations.

Keywords

Human papillomavirus circumcision men Africa

Introduction

Randomized controlled trials (RCT) in sub-Saharan Africa have shown that male circumcision (MC) reduces the acquisition of human immunodeficiency virus (HIV) infection among men.^1–3 The World Health Organization and the Joint United Nations Programme on HIV/AIDS recommended MC as a strategy for HIV prevention and identified several sub-Saharan African countries as priorities for MC scale-up programs.^4,5

Studies have found that MC reduces incidence, and increases clearance of penile human papillomavirus (HPV) infection among men^6,7 and incidence of HPV among their female sexual partners.⁸ Penile HPV infections are worth investigating, given that men are a key factor for increased transmission risk of HPV infections and HPV-associated cancers to their female sexual partners.⁹ Penile HPV infections are also associated with a higher risk of HIV acquisition.¹⁰ Furthermore, penile HPV infections, particularly HPV16, are associated with a higher risk of penile and anal cancers in men.^11,12

While HPV infections in women are well characterized, there are fewer longitudinal studies on HPV infection in men with considerable variation in incidence.^13–17 Reported incidence rates of penile high-risk HPV infection among circumcised men from population-based samples range from 7.5/1000 person-months among Chinese men and 28.7/1000 person-months among men participating in an international study in Brazil, Mexico and the US^13–16,18 The median duration of penile HPV infection has ranged from 3 to 7.5 months.^13,17,19 We previously reported a relatively high incidence rate (IR) of high-risk HPV infection (31.1/1000 person-months) among uncircumcised men participating in the control arm of an RCT in Kenya.²⁰ One other study⁶ has characterized the natural history of type-specific penile HPV infections among circumcised, HIV-negative men from Africa. As more men are circumcised through scale-up programs, these data will be increasingly important to guide cost-effectiveness models¹⁷ and implementation programs for male HPV vaccination.

This study aimed to characterize the 18-month type-specific incidence and clearance of HPV infection among 1,037 circumcised participants of an RCT of MC in Kisumu, Kenya.

Methods

Study population and enrollment

Uncircumcised men were screened between February 4, 2002 and September 6, 2005 in Kisumu, Kenya to participate in an RCT of male circumcision registered with ClinicalTrials.gov (NCT00059371),²¹ whose primary aim was to determine the effectiveness of MC in reducing HIV incidence. Inclusion criteria were: being uncircumcised, aged 18–24 years, HIV seronegative, sexually active, and blood hemoglobin ≥90 g/L. Participants were recruited from sexually transmitted infection (STI) clinics, workplaces, and community organizations. The study protocol was approved by the Institutional Review Boards of the Universities of Illinois at Chicago (#2005–0648), Manitoba (#B2001:026B), North Carolina (#04–0996), and Kenyatta National Hospital ethics and research committee (P163/9/2005).

Participants were randomized to treatment (i.e. circumcision; n = 1,391) or control (i.e. delayed circumcision; n = 1,393). This ancillary study of HPV incidence and clearance among circumcised men included participants who were circumcised before their 6-month visit in either the treatment or control arms. For circumcised men, HPV exfoliated penile samples were first collected 6 months after randomization; thus, for our analyses, “baseline” visit corresponds to the visit 6 months after the enrollment visit of the main RCT. HPV penile samples were then collected at 6-month intervals during the 18-month follow-up period: 12, 18, and 24 months after randomization.

Inclusion criteria for our analyses required participants’ consent to the collection of exfoliated penile cell samples for HPV testing, valid HPV DNA results from baseline and at least one follow-up visit, and circumcision within 6 months of randomization. Of 1,391 men enrolled in the intervention arm of the RCT, 1,083 consented to the ancillary study and had an HPV result at baseline. Of those, 1,050 (97%) had at least one additional follow-up HPV result and 1,027 (98%) were circumcised before their 6-month visit, and were included in our analyses. Additionally, 10 men randomized to the control arm crossed over within 6 months of randomization and met inclusion criteria, resulting in a sample size of 1,037 men.

Spécimen collection

After providing informed consent, a standardized questionnaire on sociodemographic characteristics and sexual behavior was administered by a trained male interviewer. At each visit, penile exfoliated cells were collected for HPV DNA detection from two anatomical sites: shaft (shaft specimen); and glans and coronal sulcus (glans specimen), using two individual prewetted Type 3 Dacron swabs (Roche Diagnostics, Basel, Switzerland).²²

Each cell sample was placed in an individual 15-mL conical centrifuge tube containing 2-mL 0.01 mol/L Tris-HCl, 7.4pH, buffer, and processed on the day of collection at the Universities of Nairobi, Illinois and Manitoba (UNIM) clinic laboratory by centrifugation at high speed (maximum, 3000 g) for 10 minutes. Excess Tris-HCl buffer was discarded using a Pasteur pipette, and the remaining cell pellet was resuspended in the same volume of 0.01 mol/L Tris-HCl buffer, and vortexed. Diluted cell pellets were then frozen at −75°C. Samples were sent using a dry shipper to the Department of Pathology, VU University Medical Center, Amsterdam, Netherlands, for HPV DNA testing.

HPV DNA and STI testing

DNA was isolated from penile exfoliated cell samples using the NucleoSpin 96 Tissue kit (Macherey-Nagel, Germany) and a Microlab Star robotic system (Hamilton, Germany), according to manufacturers’ instructions. Presence of human DNA was evaluated by β-globin specific PCR, followed by agarose gel electrophoresis. HPV positivity was assessed by GP5+/6+ PCR followed by hybridisation of PCR products using an enzyme immunoassay readout with two HPV oligoprobe cocktail probes that, together, detect 44 HPV genotypes. Subsequent HPV genotyping was performed by reverse line blot hybridisation of PCR products, as described previously.^23,24

At baseline, urine samples were tested for N. gonorrhoeae and C. trachomatis infections by PCR-based methods (Roche Diagnostics, Basel, Switzerland). Serum specimens were tested for herpes simplex virus type 2 (HSV-2) antibody (Kalon Biological Ltd, Aldershot, United Kingdom).

Statistical analyses

IRs for HPV detected over 12- and 18-month follow-up periods were estimated for individual HPV types and for HPV-type groupings (e.g. high-risk, low-risk) for both anatomical sites combined and stratified by site. An incident infection was defined as a type-specific HPV infection detected during follow-up that was not present at baseline. Time at risk for an incident infection was estimated by assuming the infection was acquired at the midpoint between the last HPV-negative result and first HPV-positive result. IRs for each HPV type or type grouping were estimated only among participants who were negative for the given type or grouping at baseline. Participants were right censored at the last study visit if they remained negative for the given HPV type or type grouping. Men with multiple HPV types were considered to have a high-risk HPV infection if ≥1 high-risk types were detected and to have a low-risk infection if only low-risk types were detected. Acquisition of multiple HPV types in a given type grouping at the same visit was considered a single incident acquisition event. Men with untyped HPV infections were excluded from both high-risk and low-risk categorizations unless a high-risk type was concurrently detected. Confidence intervals (CIs) for incidence rates were estimated by modeling the number of incident HPV infections as a Poisson variable.^13,25 Hazard ratios (HRs) comparing IRs for high-risk versus low-risk HPV infections, and for glans versus shaft infections, were estimated using Cox regression with robust (sandwich) variance estimators²⁶ used to adjust for possible correlation between multiple survival times per subject.²⁷ The Kaplan-Meier method and log-rank test were also used to examine the cumulative probability of acquiring a new HPV infection for each anatomical site, stratified by age group (≤19, 20–21, ≥22 years), and high-risk versus low-risk HPV status.

For clearance analyses, individual HPV infections were the units of analysis. Median durations of specific types of HPV infection and HPV type grouping were estimated for incident and prevalent infections using the Kaplan-Meier method. The start date of an incident HPV infection was calculated as the midpoint of the first HPV-positive result and the previous negative result. The end date was calculated as the midpoint of the last HPV-positive result and the first subsequent negative result. The start date of a prevalent infection was assumed to begin three months prior to baseline. The end date was calculated as the midpoint between the last HPV-positive result and the first subsequent negative result. Infections that did not clear by a participant’s last visit were right censored. The Kaplan-Meier method and log-rank test were also used to examine the probability of persistence for prevalent and incident HPV infections by anatomical site, age group (≤19, 20–21, ≥22 years), and high-risk versus low-risk HPV status. Cox models were fit to estimate HRs and 95%CIs for (i) the association between HPV types or type grouping and the time to HPV clearance, adjusted for age; and (ii) the association between age group and time to HPV clearance, using the robust variance estimator to adjust for multiple survival times per subject. Results are presented for prevalent and incident infections combined and incident infections alone, overall and stratified by anatomical site.

Results

At baseline, the median age of participants was 21 years (range 18–28). Approximately one-third (33.8%) of men reported always using a condom during sex within the past 6 months. Most men did not live with their female sexual partner (98.2%) and bathed at least daily (98.3%). HSV-2 seroprevalence was 20.1%, with lower prevalence detected for HIV, C. trachomatis and N. gonorrhoeae (0.7%, 1.8%, 0.9%, respectively). Prevalence of any, high-risk and low-risk HPV infections were 50.4%, 25.1% and 24.2%, respectively. The most common individual HPV types were HPV16 (3.8%), HPVJC9710 (3.4%), and HPV35 (3.0%). Median time of study follow-up was 18.0 months (interquartile range [IQR] 0.6). Median time between study visits was 6.0 months (IQR 1.0).

Beta-globin positivity overall was 45.3% at baseline, 52.7% at 6 months, 41.3% at 12 months, and 51.8% at 18 months follow-up, respectively. Incidence and clearance results did not differ substantially when analyses were restricted to beta-globin-positive samples, thus, reported analyses utilized HPV DNA data from all penile exfoliated cell specimens regardless of beta-globin positivity, unless otherwise stated.

Incidence analyses

Incidence rates for any HPV were 34.9/100 person-years (95%CI:31.2–39.0) and 36.4/100 person-years (95%CI:32.9–40.2) at 12 and 18 months, respectively (Table 1). Twelve- and 18-month incidence rates were similar for high-risk HPV (25.9 and 24.1/100 person-years, respectively) and low-risk HPV (at 11.3/100 person-years). High-risk HPV infections were acquired more rapidly over 18 months of follow-up (HR = 1.94; 95%CI:1.45–2.58 versus low-risk HPV types).

Table 1.

Incidence of human papillomavirus (HPV) infection among 1,037 circumcised men in Kisumu, Kenya over 18 months of study follow-up.

	12-month incidence			18-month incidence
HPV type	Incident HPV infections, n	Person-years at risk	IR (95%CI)^a	Incident HPV infections, n	Person-years at risk	IR (95%CI)^a
Any HPV	295	845	34.9 (31.2–39.0)	374	1029	36.4 (32.9–40.2)
Multiple HPV	123	922	13.3 (11.2–15.9)	176	1314	13.4 (11.6–15.5)
High-risk HPV	224	867	25.9 (22.7-29.4)	283	1172	24.1 (21.5–27.1)
HPV16 and related
16	67	944	7.1 (5.6–9.0)	93	1406	6.6 (5.4–8.1)
31	17	971	1.8 (1.1–2.8)	26	1461	1.8 (1.2–2.6)
33	36	967	3.7 (2.7–5.2)	41	1449	2.8 (2.1–3.8)
35	28	967	2.9 (2.0–4.2)	33	1451	2.3 (1.6–3.2)
52	10	974	1.0 (0.6–1.9)	19	1466	1.3 (0.8–2.0)
58	9	975	0.9 (0.5–1.8)	20	1469	1.4 (0.9–2.1)
HPV18 and related
18	31	966	3.2 (2.3–4.6)	42	1449	2.9 (2.1–3.9)
39	11	973	1.1 (0.6–2.0)	13	1468	0.9 (0.5–1.5)
45	39	964	4.1 (3.0–5.5)	44	1446	3.0 (2.3–4.1)
59	13	971	1.3 (0.8–2.3)	17	1464	1.2 (0.7–1.9)
68	3	978	0.3 (0.1–1.0)	5	1476	0.3 (0.1–0.8)
Other high-risk
51	26	966	2.7 (1.8–4.0)	41	1449	2.8 (2.1–3.8)
56	41	959	4.3 (3.2–5.8)	59	1437	4.1 (3.2–5.3)
66	27	966	2.8 (1.9–4.1)	40	1451	2.8 (2.0–3.8)
Low-risk HPV	103	914	11.3 (9.3–13.7)	148	1310	11.3 (9.6–13.3)
6	27	968	2.8 (1.9–4.1)	42	1451	2.9 (2.1–3.9)
11	12	972	1.2 (0.7–2.2)	21	1464	1.4 (0.9–2.2)
26	3	978	0.3 (0.1–1.0)	5	1476	0.3 (0.1–0.8)
30	7	975	0.7 (0.3–1.5)	11	1470	0.8 (0.4–1.4)
32	5	976	0.5 (0.2–1.2)	5	1474	0.3 (0.1–0.8)
34	1	978	0.1 (0.0–0.7)	1	1478	0.1 (0.0–0.5)
40	18	968	1.9 (1.2–3.0)	33	1458	2.3 (1.6–3.2)
42	25	967	2.6 (1.8–3.8)	32	1455	2.2 (1.6–3.1)
43	28	963	2.9 (2.0–4.2)	34	1450	2.4 (1.7–3.3)
44	1	978	0.1 (0.0–0.7)	3	1478	0.2 (0.1–0.6)
53	1	978	0.1 (0.0–0.7)	3	1477	0.2 (0.1–0.6)
54	4	977	0.4 (0.2–1.1)	5	1475	0.3 (0.1–0.8)
55	8	974	0.8 (0.4–1.6)	12	1470	0.8 (0.5–1.4)
61	1	978	0.1 (0.0–0.7)	1	1479	0.1 (0.0–0.5)
67	17	970	1.8 (1.1–2.8)	33	1460	2.3 (1.6–3.2)
69	8	976	0.8 (0.4–1.6)	10	1472	0.7 (0.4–1.3)
70	17	971	1.8 (1.1–2.8)	24	1460	1.6 (1.1–2.5)
72	5	976	0.5 (0.2–1.2)	6	1474	0.4 (0.2–0.9)
73	11	973	1.1 (0.6–2.0)	16	1468	1.1 (0.7–1.8)
81	14	973	1.4 (0.9–2.4)	17	1466	1.2 (0.7–1.9)
82	4	977	0.4 (0.2–1.1)	7	1476	0.5 (0.2–1.0)
83	12	972	1.2 (0.7–2.2)	17	1466	1.2 (0.7–1.9)
84	4	976	0.4 (0.2–1.1)	6	1474	0.4 (0.2–0.9)
85	4	978	0.4 (0.2–1.1)	6	1476	0.4 (0.2–0.9)
86	12	974	1.2 (0.7–2.2)	14	1468	1.0 (0.6–1.6)
JC9710	27	964	2.8 (1.9–4.1)	35	1448	2.4 (1.7–3.4)
X^b	44	958	4.6 (3.4–6.2)	57	1434	4.0 (3.1–5.2)

IR: incidence rate; CI: confidence interval; No incident HPV infections of type 57, 64 or 71 were detected during follow-up.

^aIncidence rate per 100 person-years for the first type detected in the glans or shaft. Infections of multiple HPV types were considered high-risk if one or more high-risk HPV types were detected; all other multiple infections were considered low-risk types unless they included HPVX.

^bUntyped HPV infection.

Individual HPV types with the highest 12-month incidence rates were HPV16 (7.1/100 person-years), HPV56 (4.3/100 person-years), HPV45 (4.1/100 person-years), HPV33 (3.7/100 person-years), and HPV18 (3.2/100 person-years). By 18 months, only three known HPV types had incidence rates ≥3.0/100 person-years (HPV16 = 6.6, HPV56 = 4.1, and HPV45 = 3.0). When restricting to beta-globin-positive samples, IRs were lower than IRs estimated among all samples (18-month IRs for any HPV = 24.6; 95%CI:22.2–27.3 and 36.4; 95%CI:32.9–40.2, respectively).

Overall, HPV incidence rates were similar in the glans (IR = 17.5/100 person-years) compared with the shaft (IR = 16.8/100 person-years) (HR = 1.05; 95%CI:0.92–1.19) (Table 2). Incidence rates in the glans versus shaft were similar for high-risk HPV (HR = 1.11; 95%CI:0.95–1.29), low-risk HPV (HR = 1.19; 95%CI:0.95–1.49) and HPV16 infections (HR = 1.09; 95%CI:0.80–1.49).

Table 2.

Incidence of human papillomavirus (HPV) infection over 18 months stratified by anatomical site among 1,037 circumcised men in Kisumu, Kenya.

HPV type	Incident HPV infections in glans, n	IR (95%CI)^a	Incident HPV infections in shaft, n	IR (95% CI)^a	Hazard ratio^b (95%CI)
Any HPV	305	17.5 (15.7–19.5)	295	16.8 (15.1–18.7)	1.1 (0.9–1.2)
High-risk HPV	217	12.4 (10.9–14.1)	200	11.3 (9.9–12.9)	1.1 (1.0–1.3)
HPV16 and related^c	159	1.4 (1.2–1.6)	140	1.2 (1.0–1.4)	1.1 (1.0–1.4)
HPV16	59	3.1 (2.4–3.9)	54	2.8 (2.2–3.7)	1.1 (0.8–1.5)
HPV16-related^c	100	1.0 (0.8–1.2)	86	0.9 (0.7–1.1)	1.2 (0.9–1.5)
HPV18 and related^d	90	0.9 (0.8–1.1)	64	0.7 (0.5–0.8)	1.4 (1.1–1.8)
HPV18	34	1.7 (1.3–2.4)	17	0.9 (0.5–1.4)	2.0 (1.3–3.2)
HPV18-related^d	56	0.7 (0.6–0.9)	47	0.6 (0.5–1.8)	1.2 (0.9–1.6)
Other high-risk^e	111	1.9 (1.6–2.3)	82	1.4 (1.1–1.7)	1.4 (1.1–1.7)
Low-risk HPV	119	6.5 (5.5–7.8)	102	5.5 (4.6–6.7)	1.2 (1.0–1.5)

IR: incidence rate; CI: confidence interval.

^aIR=incidence per 100 person-years.

^bHazard ratio for glans versus shaft infections.

^cHPV types more closely related phylogenetically to HPV16 than to other types: HPV31, 33, 35, 52, 58.

^dHPV types more closely related phylogenetically to HPV18 than to other types: HPV39, 45, 59, 68.

^eHPV51, 56, 66.

Cumulative incidence probabilities were estimated by HPV type groupings, age group, and anatomical site. The 18-month cumulative probability for high-risk HPV (30%; 95%CI:28–33%) was higher than for low-risk HPV (16%; 95%CI:14–18%; p < 0.001) (Figure 1(a)). Cumulative HPV incidence probabilities did not differ when stratified by age group (log rank p = 0.38) or anatomical site (log rank p = 0.22) (Figure 1(b) and (c)).

Figure 1.

Kaplan Meier estimates of the cumulative incidence of human papillomavirus infection. HPV: human papillomavirus; CI confidence interval. (a) stratified by high-risk and low-risk HPV status: 18-month incidence proportions= 0.30 (95%CI:0.28–0.33) for high-risk and 0.16 (95%CI:0.14–0.18) for low-risk HPV; (b) stratified by age group: 18-month incidence proportions =0.30 (95%CI:0.24–0.36) for ≤19 years, 0.29 (95%CI:0.25–0.34) for 20–21 years and 0.31 (95%CI:0.27–0.37 for ≥22 years; and (c) stratified by anatomical site: 18-month incidence proportion = 0.33 (95%CI:0.30–0.36) for glans and 0.30 (95%CI:0.27–0.33) for shaft.

Median duration and clearance analyses

The probability of any HPV infection clearing by 6 months was 0.38 and by 12 months was 0.92 (Figure 2). Median clearance time for HPV infections was 6.0 months (95%CI:5.4–11.6) for high-risk HPV infections and 6.0 months (95%CI:5.3–12.2) for low-risk HPV infections. Median clearance time for individual HPV types ranged from 5.9 months for HPV31 to 11.9 months for HPV32. Median clearance times were similar when comparing high-risk versus low-risk HPV (p = 0.54), by age groups (p = 0.76), and by anatomical site (p = 0.55) (Figure 2(a) to (c)).

Figure 2.

Kaplan Meier estimates for the time to HPV clearance (in months). HPV: human papillomavirus; CI: confidence interval. (a) Stratified by high-risk and low-risk HPV status: median clearance time for high-risk = 6.0 (95%CI:5.4–11.6) and low-risk =6.0 (95%CI:5.3–12.2); (b) stratified by age: median clearance time (months)= 6.0 (95%CI:5.5–11.0) for ≤19 years, 6.0 (95%CI:5.5–11.1) for 20–21 years and 6.0 (95%CI:5.5–11.4) for ≥22 years; and (c) stratified by anatomical site: median clearance time for glans = 6.0, (95%CI:5.7–11.2) and shaft = 6.0 (95%CI:5.6–11.0).

HPV infections among men aged ≤19 years (HR = 1.08; 95%CI:0.77–1.51 for any HPV) and 20–21 years (HR = 1.14; 95%CI:0.85–1.55) did not clear at a different rate than HPV infections among men ≥22 years of age (Table 3). No differences in clearance rates for prevalent and incident HPV infections across age groups were observed when stratified by anatomical site. Further, there were no significant differences in clearance rates for high-risk HPV types compared to low-risk types.

Table 3.

Hazard ratios and 95% confidence intervals for clearance of human papillomavirus (HPV) infection among 1,037 Kenyan men, stratified by HPV type and age group.

	Overall HPV infections			Glans HPV infections			Shaft HPV infections
	Events^a	Person years	HR (95%CI)	Events^a	Person-years	HR (95%CI)	Events^a	Person-years	HR (95%CI)
Prevalent and incident infections combined
Age (years)
≤19	226	197	1.1 (0.8–1.5)	169	148	1.2 (0.9–1.8)	114	103	0.9 (0.6–1.5)
20–21	329	287	1.1 (0.9–1.6)	235	207	1.2 (0.9–1.7)	186	157	1.2 (0.8–1.7)
≥22	291	265	1.0	211	188	1.0	157	138	1.0
HPV type
High-risk	444	386	1.0 (0.8–1.2)	321	278	1.0 (0.8–1.2)	241	209	1.1 (0.8–1.3)
HPV16 and related^b	216	184	0.9 (0.7–1.2)	159	134	0.9 (0.7–1.2)	114	94	1.0 (0.8–1.4)
HPV16	68	64	1.0 (0.7–1.4)	46	43	0.9 (0.6–1.4)	37	33	0.9 (0.6–1.5)
HPV16-related^b	148	120	0.9 (0.7–1.2)	113	91	0.9 (0.6–1.3)	77	62	1.1 (0.8–1.5)
HPV18 and related^c	115	88	1.0 (0.7–1.4)	80	63	1.1 (0.8–1.6)	66	52	1.0 (0.7–1.4)
HPV18	32	29	1.2 (0.8–1.9)	25	22	1.3 (0.7–2.1)	17	14	1.3 (0.7–2.3)
HPV18-related^c	83	60	0.9 (0.6–1.3)	55	40	1.0 (0.6–1.6)	49	38	0.9 (0.6–1.4)
Other high-risk^d	113	114	1.1 (0.9–1.4)	82	82	1.1 (0.9–1.5)	61	63	1.2 (0.9–1.6)
Low-risk	371	335	1.0	275	246	1.0	202	177	1.0
Incident Infections
Age (years)
≤19	100	122	1.1 (0.8–1.5)	69	89	1.2 (0.8–1.7)	49	66	1.0 (0.6–1.5)
20–21	122	164	1.2 (0.9–1.6)	79	113	1.2 (0.9–1.8)	74	97	1.3 (0.9–1.9)
≥22	117	150	1.0	79	99	1.0	68	88	1.0
HPV type
High-risk	191	234	0.8 (0.7–1.0)	132	163	0.9 (0.7–1.1)	103	135	0.9 (0.7–1.2)
HPV16 and related^b	92	108	0.8 (0.7–1.1)	60	74	0.9 (0.7–1.2)	54	62	0.9 (0.7–1.2)
HPV16	30	39	0.9 (0.7–1.3)	19	25	1.0 (0.7–1.4)	17	21	0.8 (0.5–1.3)
HPV16-related^b	62	69	0.8 (0.6–1.0)	41	49	0.8 (0.6–1.2)	37	41	1.0 (0.7–1.4)
HPV18 and related^c	53	53	0.8 (0.5–1.1)	35	36	0.8 (0.5–1.3)	27	31	0.9 (0.6–1.4)
HPV18	16	20	0.8 (0.5–1.4)	13	15	0.8 (0.4–1.5)	7	9	1.1 (0.6–2.0)
HPV18-related^c	37	33	0.7 (0.5–1.2)	22	21	0.9 (0.5–1.5)	20	22	0.8 (0.5–1.4)
Other high-risk^d	46	72	0.9 (0.7–1.2)	37	53	0.9 (0.6–1.2)	22	42	1.0 (0.7–1.4)
Low-risk	129	180	1.0	81	124	1.0	82	108	1.0

HR: hazard ratio; CI: confidence interval.

^aEvents = # of infections cleared.

^bHPV types more closely related phylogenetically to HPV16 than to other types: HPV31, 33, 35, 52, 58.

^cHPV types more closely related phylogenetically to HPV18 than to other types: HPV39, 45, 59, 68.

^dHPV51, 56, 66.

Discussion

The acquisition of high-risk HPV infections was common in this large follow-up study of circumcised Kenyan men. Similar HPV incidence rates were found in the glans and shaft. Approximately one third of penile HPV infections among men cleared by 6 months, and the vast majority cleared by 12 months. HPV clearance did not vary by age group nor by anatomical site.

The incidence rate of 24.1/100 person-years (equivalent to 20.1/1000 person-months) for high-risk HPV found among circumcised men aged 18–28 years in this study is consistent with most^6,13,18 though not all¹⁶ prior studies. Our estimated incidence is similar to that observed among circumcised men in Uganda (27.1/100 person-years),⁶ Arizona (20.8/1000 person-months),¹³ and Brazil, Mexico and the US (28.7/100 person-years),¹⁸ yet higher than among circumcised men in China (7.5/1000 person-months).¹⁶ Our findings are consistent with RCTs indicating that HPV incidence is lower among circumcised versus uncircumcised men^6,7 given that the incidence rate for high-risk HPV infection among circumcised men in this study was much lower than the 12-month incidence rate reported among uncircumcised men from the same Kenyan population (37.3/100 person-years).²⁰

The 18-month cumulative incidence estimated in our study were similar in the glans and shaft specimens (33% and 30%, respectively). Our findings are consistent with those of a US study of 240 men, the majority (77%) of whom were circumcised,¹⁴ wherein cumulative HPV incidence was similar across anatomical sites (44% in glans versus 45% in shaft). These results are in contrast to a previous study of 966 uncircumcised men from Kenya where the incidence rate of HPV infection in the glans was higher than that on the shaft (44.4 glans versus 21.6 shaft, HR = 2.1; 95% CI 1.7 to 2.4).²⁰ Similar rates of HPV acquisition across penile sites among circumcised men may be due to greater tissue keratinization and lower retention of infectious secretions below the foreskin.²⁸ We also found similar rates of HPV acquisition across different age groups, which is consistent with findings of a study of 1,159 men aged 18–70 from Brazil, Mexico and the US¹⁷ although the age range in our study was relatively narrow.

The median duration of incident and prevalent HPV infections (overall) in our study (6.0 months) was similar to the findings of a previous US study reporting a median duration of 5.9 months,¹³ but higher than the median duration reported among circumcised men from Hawaii (3.5 months), possibly due in part to the relatively shorter interval between study visits in that study.¹⁹ Longer median duration of penile HPV infection was reported among circumcised men in an international study in Mexico, Brazil and the US (7.5 months)¹⁸ and among Chinese men (7.3 months).¹⁶ It should be noted, however, that median duration estimates should be interpreted with caution when estimating HPV infection duration, given that they are dependent upon our definition of infection start date – as beginning three months prior to baseline (prevalent) or as the midpoint of the first HPV-positive result and the previous negative result (incident), and the end date as the midpoint between last HPV-positive result and first subsequent negative result.

HPV clearance during follow-up was similar for high-risk and low-risk HPV infections and did not appear to differ by age. The relatively small age range of our study participants limited the assessment of HPV clearance among men in relatively older age groups. Our findings are similar, however, to those of a previous study among 290 men aged 18–44 years from the US in which HPV clearance was associated with neither high-risk or low-risk HPV status, nor age.¹³ In contrast, natural history studies among women have generally found lower clearance rate of high-risk compared to low-risk HPV infections,²⁹ and a corresponding higher duration of HPV persistence.³⁰ In one international study among men, median time to clearance was longer in younger (18–30 years old) men compared to those aged 31–70 years.¹⁷

A sensitive GP5+/6+ PCR assay was used to detect a wide range of HPV types in a central laboratory, allowing for greater quality control of study results. Separate HPV laboratory testing for glans and shaft specimens allowed for stratified analyses by anatomical site. To our knowledge, this study is one of the first to characterize the natural history of HPV infection among HIV-negative circumcised men from Africa and one of the largest follow-up studies of HPV infection among circumcised men to date.

The 6-month interval between each visit was a limitation of this study as participants could have acquired a new HPV infection and cleared it before their subsequent follow-up visit, potentially underestimating HPV incidence in our study. Additionally, participants could have cleared and re-acquired an HPV infection of the same type within the testing interval. Beta-globin positivity was also relatively low; however, our results did not change substantially when restricting analyses to beta-globin-positive samples, except that IRs were lower among beta-globin positive samples. It has also been previously found that beta-globin-PCR-negative samples often contain detectable HPV, as HPV copies may often exceed those of the beta-globin gene²² thus restricting the analysis to beta-globin-positive samples may underestimate the true IRs. Given that participants were young, recently circumcised men at high-risk of STIs, the generalizability of our findings may be somewhat limited.

In conclusion, we found an incidence rate among circumcised Kenyan men comparable to other circumcised populations, and that the vast majority of penile HPV infections among men cleared by 1 year. While MC can reduce risk for HPV infections, incident HPV infections were still quite common among circumcised young men participating in this study; thus, additional interventions to reduce HPV acquisition and decrease HPV clearance times are needed. Our findings can help inform future HPV vaccine implementation programs in young men prior to sexual initiation to help reduce the risk of oral, anal, and penile cancers in men, as well as HPV transmission to their female sexual partners to reduce the risk of cervical, oral, vaginal, vulvar and anal cancers. They may also assist policy makers and programmers in assessing cost and program implementation challenges that are currently limiting vaccine availability in many countries.

Footnotes

Acknowledgements

The authors would like to thank the young men from Kisumu, Kenya who participated in this study and the UNIM staff for all of their hard work on this project. The authors would also like to thank the late Professor J.O. Ndinya-Achola for all of his effort and commitment to this project.

Declaration of conflicting interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: PJFS had received speaker’s fees from Hologic-Gen-Probe, Roche, Abbott, and Qiagen, and has limited stock of Self-Screen, a spin-off company of VU University Medical Center (no payments made); CJLM is an occasional advisory member to QIAGEN, Roche, GSK, Merck and Hologic-Gen-probe, a consultant to QIAGEN (finished 12/31/10), and has small amount of shares of Self-Screen BV (no payments made). JSS has reviewed research grants or served as a consultant for GSK, Merck, Hologic-Gen-probe and QIAGEN. For the remaining authors, none were declared.

Funding

The authors disclose receipt of the following financial support for the research, authorship and/or publication of this article: This study was supported by the National Cancer Institute, National Institutes of Health (NIH) (grant R01 CA114773-04). The RCT was supported by grant number AI50440 from the Division of AIDS, NIAID, NIH and by grant number HCT 44180 from the Canadian Institutes of Health Research (CIHR). RCB was supported by the Chicago Developmental Center for AIDS Research, an NIH funded program P30 AI 082151. CB was supported by the Cancer Health Disparities Program, a National Cancer Institute funded program (grant T32-CA128582).

ORCID iD

Claire Bosire

References

Bailey

Egesah

Rosenberg

Male circumcision for HIV prevention: a prospective study of complications in clinical and traditional settings in Bungoma, Kenya. Bull World Health Organ 2008; 86: 669–677.

Auvert

Taljaard

Lagarde

, et al. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 trial. PLoS Med 2005; 2: e298.

Gray

Kigozi

Serwadda

, et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007; 369: 657–666.

World Health Organization. New data on male circumcision and HIV prevention: policy and programme implications. Geneva: WHO and UNAIDS, www.who.int/hiv/pub/malecircumcision/research_implications/en/ (2007, accessed 23 July 2020).

World Health Organization. Progress in scale-up of male circumcision for HIV prevention in Eastern and Southern Africa: focus on service delivery. Geneva: WHO and UNAIDS, www.who.int/hiv/pub/malecircumcision/mc_country_progress2011/en/ (2011, accessed 23 July 2020).

Gray

Serwadda

Kong

, et al. Male circumcision decreases acquisition and increases clearance of high-risk human papillomavirus in HIV-negative men: a randomized trial in Rakai, Uganda. J Infect Dis 2010; 201: 1455–1462.

Smith

Backes

Bailey

, et al. The effect of male circumcision on incident and persistent penile HPV infections in men from Kenya. In: 26th international papillomavirus conference and clinical and public health workshops, Montréal, Canada, 3–8 July, 2010.

Wawer

Tobian

Kigozi

, et al. Effect of circumcision of HIV-negative men on transmission of human papillomavirus to HIV-negative women: a randomised trial in Rakai, Uganda. Lancet 2011; 377: 209–218.

Castellsagué

Bosch

Muñoz

The male role in cervical cancer. Salud Publica Mex 2003; 45: S345–S353.

10.

Smith

Moses

Hudgens

, et al. Increased risk of HIV acquisition among Kenyan men with human papillomavirus infection. J Infect Dis 2010; 201: 1677–1685.

11.

Backes

Kurman

Pimenta

, et al. Systematic review of human papillomavirus prevalence in invasive penile cancer. Cancer Causes Control 2009; 20: 449–457.

12.

Hoots

Palefsky

Pimenta

, et al. Human papillomavirus type distribution in anal cancer and anal intraepithelial lesions. Int J Cancer 2009; 124: 2375–2383.

13.

Giuliano

Nielson

, et al. Age-specific prevalence, incidence, and duration of human papillomavirus infections in a cohort of 290 US men. J Infect Dis 2008; 198: 827–835.

14.

Partridge

Hughes

Feng

, et al. Genital human papillomavirus infection in men: incidence and risk factors in a cohort of university students. J Infect Dis 2007; 196: 1128–1136.

15.

Lajous

Mueller

Cruz-Valdéz

, et al. Determinants of prevalence, acquisition, and persistence of human papillomavirus in healthy Mexican military men. Cancer Epidemiol Biomarkers Prev 2005; 14: 1710–1716.

16.

Wei

Guo

, et al. The impact of male circumcision on the natural history of genital HPV infection: a prospective cohort study. Zhonghua Yu Fang Yi Xue Za Zhi 2018; 52: 486–492.

17.

Giuliano

Lee

Fulp

, et al. Incidence and clearance of genital human papillomavirus infection in men (HIM): a cohort study. Lancet 2011; 377: 932–940.

18.

Albero

Castellsagué

Lin

, et al. Male circumcision and the incidence and clearance of genital human papillomavirus (HPV) infection in men: the HPV infection in men (HIM) cohort study. BMC Infect Dis 2014; 14: 75.

19.

Hernandez

Shvetsov

Goodman

, et al. Reduced clearance of penile human papillomavirus infection in uncircumcised men. J Infect Dis 2010; 201: 1340–1343.

20.

Backes

Snijders

Hudgens

, et al. Sexual behaviour and less frequent bathing are associated with higher human papillomavirus incidence in a cohort study of uncircumcised Kenyan men. Sex Transm Infect 2013; 89: 148–155.

21.

Bailey

Moses

Parker

, et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007; 369: 643–656.

22.

Smith

Backes

Hudgens

, et al. Prevalence and risk factors of human papillomavirus infection by penile site in uncircumcised Kenyan men. Int J Cancer 2010; 126: 572–577.

23.

van den Brule

Pol

Fransen-Daalmeijer

, et al. GP5+/6+ PCR followed by reverse line blot analysis enables rapid and high-throughput identification of human papillomavirus genotypes. J Clin Microbiol 2002; 40: 779–787.

24.

Snijders

van den Brule

Jacobs

, et.al. HPV DNA detection and typing in cervical scrapes by general primer GP5+/6+ PCR. In: CEDJE

Davy

(ed.) Methods in molecular medicine: human papillomaviruses. Methods and protocols. Totowa: Humana Press, 2005. pp.101–114.

25.

Ulm

A simple method to calculate the confidence interval of a standardized mortality ratio (SMR). Am J Epidemiol 1990; 131: 373–375.

26.

Lin

Wei

LJ.

The robust inference for the proportional hazards model. J Am Stat Assoc 1989; 84: 1074–1078.

27.

Xue

Gange

Zhong

, et al. Marginal and mixed-effects models in the analysis of human papillomavirus natural history data. Cancer Epidemiol Biomarkers Prev 2010; 19: 159–169.

28.

Templeton

DJ.

Male circumcision to reduce sexual transmission of HIV. Curr Opin HIV AIDS 2010; 5: 344–349.

29.

Muñoz

Méndez

Posso

, et al. Incidence, duration, and determinants of cervical human papillomavirus infection in a cohort of Colombian women with normal cytological results. J Infect Dis 2004; 190: 2077–2087.

30.

Rositch

Koshiol

Hudgens

, et al. Patterns of persistent genital human papillomavirus infection among women worldwide: a literature review and meta-analysis. Int J Cancer 2013; 133: 1271–1285.