Male Circumcision The Road from Evidence to Practice

Male Circumcision: The Road from Evidence to Practice Prof. Robert C. Bailey Division of Epidemiology School of Public Health University of Illinois at Chicago rcbailey@uic. edu

Current Prevention Tools Against Heterosexual HIV Transmission · Risk Reduction – Abstinence – Faithfulness – Condoms · STI Treatment · Voluntary Counseling and Testing · Pr. EP? · Anti-retroviral therapy? · Microbicides? · Diaphragm · HSV-2 suppressive therapy? · Vaccines?

Male circumcision is probably the oldest and most common surgery performed in humans

Male Circumcision Background Information · Globally, approximately 30% of men are circumcised. In Africa, approximately 68%. · It is a practice observed mostly for cultural and religious reasons, less often for health reasons. · It is a simple procedure that may confer health benefits, but being a surgical procedure it entails risks. · The benefits of MC must be weighed against the potential harm.

Protective Effect of MC against Other Conditions Outcome RR (95% CI) Urinary tract infection 0. 13 (0. 08, 0. 20) Chlamydia 0. 18 (0. 06, 0. 58) Chancroid 0. 40 (0. 24, 0. 66) Invasive penile cancer 0. 43 (0. 24, 0. 77) Cervical cancer 0. 54 (0. 39, 0. 74) Syphilis 0. 67 (0. 54, 0. 83) HPV 0. 80 (0. 61, 1. 05) HSV-2 0. 88 (0. 77, 1. 01) . 1 . 5. 75 1 2 Risk ratio 3

Observational Studies of the Association Between Circumcision and HIV Infection · · 4 ecological studies 37 cross-sectional studies 15 prospective studies The adjusted relative risk of HIV infection for circumcised men found in the prospective studies is 0. 52 – 0. 14

Meta-analysis (Weiss et al. , AIDS 2000; 14: 2261 -70)

Schema of Uncircumcised Penis Flaccid (A) and Erect with Foreskin Retracted (B) A B Mc. Coombe & Short, AIDS 2006 20: 1491 -1495

Outer surface of foreskin Inner surface of foreskin Patterson et al. , 2002

Outer surface of foreskin Inner surface of foreskin Mc. Coombe and Short, 2006

Inner (A) and Outer (B) Surfaces of Foreskin Infected by HIVBal in Explant Culture A Red = uninfected cells (Patterson et al. , 2002) B Yellow = infected Langerhans’ cells Green = infected CD 4+ T cells

Need for Clinical Trials · “Randomized clinical trials are needed to determine the utility of circumcision as an HIV preventive measure. ” · Reasons: – All epidemiological studies had been observational. Confounding could exist. – Risk of too early resumption of sexual activity after circumcision or subsequent behavioural disinhibition (risk compensation) could counteract any protective effect. – Risk of post-surgical complications must be balanced against any protective effect.

Randomised controlled trials of male circumcision to reduce HIV infections Rakai, Uganda Gray et. al. (2007) Lancet; 369: 657 – 66 Kisumu, Kenya Bailey et. al. (2007) Lancet; 369: 643 – 56 Orange Farm, South Africa Auvert et. al. (2005) Source: 2006 Report on the global AIDS Epidemic (UNAIDS, May 2006) PLo. S Med; 2 (11): e 298

Effect of MC on HIV Incidence: Observational Studies and 3 RCTs Risk ratio (95% CI) Observational 0. 42 (0. 34, 0. 52) South Africa 0. 40 (0. 24, 0. 68) Kenya 0. 41 (0. 24, 0. 70) Uganda 0. 49 (0. 28, 0 , 0. 84) Summary RCTs (95% CI) 0. 43 (0. 32, 0. 58) . 15. 2 . 3 1. 4. 5 Risk ratio 1. 5

WHO/UNAIDS Consultation Montreux, Switzerland March 6, 2007

WHO/UNAIDS Statement March 28, 2007 · “The efficacy of male circumcision in reducing female to male HIV transmission has now been proven beyond reasonable doubt. This is an important landmark in the history of HIV prevention. ” · “Scaling up male circumcision in (certain) countries will result in immediate benefits to individuals. ” · “Male circumcision should be considered as part of a comprehensive HIV prevention package. ”

What is the Impact of Circumcision on Incidence of other STI? Circumcision STI HIV

MC Impact on STIs: Results from RCTs Outcome Orange Farm S Africa Kisumu, Kenya HSV-2 incidence 0. 84 (0. 56 -1. 26) 0. 99 (0. 67 -1. 46) Gonnorhea 0. 68 (0. 29 -1. 62) 0. 97 (0. 69 -1. 35) Chlamydia 1. 10 (0. 64 -1. 78) 0. 93 (0. 69 -1. 26) Syphilis 1. 02 (0. 50 -2. 10) 1. 68 (0. 40 -7. 03) Trichomonas - 0. 78 (0. 46 -1. 34)

Incidence of Signs and Symptoms of Selected STI: Kisumu Trial

Will Sexual Disinhibition erode the efficacy of MC?

Behavioural Disinhibition/Risk Compensation • Agot et al. , 2007 • prospective study: no difference in risk behaviours • 3 RCTs • Orange Farm, South African • 5 of 5 risk behaviors were greater in the circumcision group • Mean # sexual encounters increased (p<0. 05) • Adjusting for differences in sexual behaviour had no effect on the RR of HIV infection • Rakai, Uganda • Difference in inconsistent condom use at 6 months (p<0. 05) • Control group more likely to drink alcohol • Kisumu, Kenya • Risk behaviours were reduced significantly in both groups • Circumcised men had more unprotected sex (p<0. 03) and less consistent condom use at 24 months (p<0. 03)

Sexual Risk Scores Baseline, 6 Months and 12 Months: Kisumu Trial Mattson et al. , submitted

Results III Gonorrhea, Chlamydia, or Trichomonas Infections *Prevalent infections included since circumcised men had more infections at baseline

What will be the Impact of Circumcision Interventions?

Correlation of male circumcision and HIV prevalence in adults (%), 2005 10 20 30 40 Sub-Saharan Africa Swaziland Botswana Lesotho Zimbabwe South Africa Zambia Mozambique Malawi Tanzania Uganda 0 Madagscar 20 40 60 Male circumcision prevalence (%) 80 Kenya Cameroon Angola Ghana Benin 100

Correlation of male circumcision and HIV prevalence 2 Asia HIV prevalence in adults (%), 2005. 5 1 1. 5 Papua New Guinea Cambodia Thailand India Republic of Korea 0 China Japan 0 Indonesia Pakistan Philippines 20 40 60 Male circumcision prevalence (%) 80 100

Modeling the Impact of MC on HIV Prevalence/Incidence · Williams et al. , 2006 – 100% uptake of MC could avert 2. 0 million new infections and 0. 3 million deaths over ten years in sub-Saharan Africa – Could avert 5. 7 million new infections over 20 years · Mesesan et al. , 2006 – 50% uptake of MC could avert 32, 000 – 53, 000 new infections in Soweto, SA over 20 yrs. Prevalence would decline from 23% to 14%. · Nagelkerke et al. , 2007 – Prevalence in Nyanza Province, Kenya would decline from 18% to 8% with 50% uptake of circumcision over 10 years. · Gray et al. , 2007 – Assuming 50% uptake in Rakai, incidence would decline from 1. 4% to. 81%, and Ro would decline to 0. 89.

Cost-effectiveness Models · Kahn et al. , 2006 – Cost is $181 per HIV infection averted over 20 years in Guateng, SA. With 25. 6% prevalence. – Cost-effectiveness is sensitive to HIV prevalence, cost of MC. – If HIV prevalence is 8%, cost per infection averted is $550.

What is the impact of the pace of scale-up?

The Impact of the Pace of Scale up Linear scale- Slower scale Faster scaleup -up up 2008 -2020 % difference with linear scale-up Avg annual # MCs 27, 473 -1. 6% +1. 5% Avg annual # IAs 4, 533 -16. 4% +15. 9% Cost per IA $292 +14% -10%

Cost per HIV Infection Averted Across Prevention Interventions · · · Adult MC PMTCT Condoms Treatment of STI School-based $ 176 - $2, 631 $ 20 – 21, 000 $ 11 – 2, 198 $ 271 - $514 $ 7, 288 - $13, 326

Location of Studies of Acceptability Thirteen studies from nine sub-Saharan African countries were identified through a comprehensive search of electronic databases (MEDLINE) and contact with authors.

Main Barriers and Facilitators for Acceptability of MC Barriers: · Cost · Fear of pain · Concern for safety Facilitators: · Hygiene · Reduced STIs · Attractiveness Uncircumcised men for themselves: Women (for their partners): Men for their son: Women for their son: 65% (29 -87%) 69% (47 -79%) 71% (50 -90%) 81% (70 -90%) Westercamp and Bailey, 2006

The Road to Practice Minimum package: § HIV counseling and testing § STI diagnosis, treatment and partner referral • Behavioral counseling, including safe sex after MC Additionally: Comprehensive RH Services § Couples counseling § Family planning § Gender equality § Access to female partners

The Road to Practice · Male circumcision services should be made available as soon as possible in regions where HIV prevalence is high and most infections occur through heterosexual transmission. · MC should be integrated with other HIV/STI prevention strategies – not a stand alone procedure. · Install systems for monitoring and evaluation of adverse events and behavioural risk compensation. · Consider innovative means to reach the maximum number of males; e. g. , mobile services; medical missions, “circumcision weekends” other?

Research Questions · Is there a protective effect for females? · Is there a protective effect for MSM? · How long after the procedure does it take for full healing and keratinization to occur? · How much of the protective effect of MC is attributable to its effect against GUD? · What is the impact of MC programs on HIV at the population level? (Phase 4) · Will behavioral disinhibition/risk compensation occur? Will risk behaviours decline? · Operational research

Summary of Advantages and Challenges Advantages · · Clearly protective One-time intervention Cost-effective Opportunity to reach men · Opportunity to reach sexually active women · Opportunity to improve infrastructure and related services Challenges · · Only partially protective Safety Disinhibition Excuse not to use condoms · Benefits take 5+ years · Must have high coverage · Potential stigma

Many Thanks Helen Weiss Bertran Auvert Stephen Moses Christine Mattson Nelli Westercamp Corette Parker Norma Pugh Jack Ndinya-Achola Kawango Agot Ian Mac. Lean Carolyn Williams Ron Gray Maria Wawer Alan Landay Bruce Patterson Lori Bollinger James Kahn Daniel Halperin Brian Williams Richard Hayes Nico Nagelkerke Gayle Martin John Stover Scott Mc. Coombe Roger Short Brian Morris Melanie Bacon Jeffrey Klausner Richard Campbell Cate Hankins George Schmidt