Understanding Results of CAPRISA 004: What do they mean for VOICE?
Z. Mike Chirenje MD FRCOG Protocol Co-Chair, VOICE Study UZ-UCSF Collaborative Research Program College of Health Science, Avondale Harare, Zimbabwe chirenje@uz_ucsf.co.zw
HIV/AIDS Burden
• Close to 3 decades after researchers identified HIV as cause of AIDS, 33.4 million living with HIV/AIDS (UNAIDS 2008) • Worldwide women carry disproportionate burden of HIV infection, 67% of people living with HIV/AIDS in SSA are women • WHO estimates that 2.5 million individuals become newly infected with HIV every year Women initiated HIV prevention methods • Women are unable to successfully negotiate mutual monogamy or condom use • New biological interventions are urgently need to curb the epidemic • Only 5 of 37 RCT which tested 39 HIV prevention strategies have demonstrated protection against sexual transmission of HIV infection HIV Prevention Strategies
• M circumcision is 50-60% effective in preventing F to M HIV acquisition (evidence from 3 trials, Auvert ‘05, Bailey ’07 Gray’07) • HIV vaccine trial several setbacks and modest 30% effectiveness (Perks-Ngarm S NEJM 2009) HIV Prevention Strategies
• STI treatment in Tanzania demonstrated 42% effectiveness in reducing HIV acquisition (Grosskurth H 1995) • Over the past 20 yrs, 11 effectiveness trials of six candidate microbicides • CAPRISA 004 is the first proof of concept microbicide and we now know that an effective microbicide has potential to alter the global HIV trajectory in next 3-5 yrs The Window of Opportunity for Prevention
Mucosal Epithelial Signaling Endocervix Transformation Zone
Adapted from Haase, Nature 2010, 464:217-23 Outside in signaling Recruits new target cells
Increased “Signal” Growth and spread from initial cluster by accretion of new infections influx of new target cells
Focus on critical early expansion Adapted from Haase, Necessary to establish Nature 2010, 464:217-23 systemic infection
How Could Microbicides Prevent Infection?
McGowan I, Biologicals, 2006 Progression in the product pipeline
C31G, N9 BufferGel
PRO2000, CS, Carraguard
Tenofovir, TMC 120, UC781, MIV150 CAPRISA 004 Effectiveness & safety of vaginal microbicide 1% tenofovir gel for prevention of HIV infection in women
1 CAPRISA 004 Participants, New York Times, July 20, 2010 HIV prevalence in pregnant women in rural Vulindlela,, South Africa (2005- -2008)
Age Group HIV Prevalence (Years) (N=1237)
≤16 10.6%
17-18 21.3%
19-20 33.0%
21-22 44.3%
23-24 51.1% 4 Why Tenofovir Gel?
Effective therapeutic agent
Very good safety profile Prevents mother-to-child transmission
Gel rapidly absorbed and long half-life
Very low systemic absorption – therefore expect fewer side effects Protects against SIV in monkey studies Methods
• Proof of concept double-blinded, randomized, placebo- controlled trial
• Enrolled high risk HIV uninfected women reporting two coital acts in past 30 days – known high risk populations from pre-trial feasibility studies
• Endpoint driven trial (92 HIV endpoints)
• HIV infection is primary safety & effectiveness endpoint: HIV negative: 2 negative rapid HIV tests HIV endpoint: PCR+ in 2 separate blood specimens Positive Western blot
• Intent-to-treat analysis except for adherence analysis CAPRISA 004 assessed the safety and effectiveness of 1% tenofovir gel
• BAT 24 coitally-related gel use Insert 1 gel up to 12 hours Before sex, insert 1 gel as soon as possible within 12 hours After sex, no more than Two doses in 24 hours
HIVNET 012 nevirapine regimen CAPRISA 004 tenofovir gel regimen asap asap 12 72 hrs hrs
Onset of Delivery labour
5 Screening and Enrollment 9
1075 excluded: • 536 HIV positive Screened: 2160 •142 did not return •132 not sexually active • 51 pregnant or planning a pregnancy • 37 participating in other research • 33 refused participation • 26 not keen to use contraceptives • 24 allergic to latex • 23 planning to relocate Enrolled & • 23 medical condition • 19 unable to attend study visits randomized: • 14 unable to provide informed consent • 15 other reasons 1085
196 excluded: • 135 co-enrolled • 50 in other study <1 year ago • 1 <18 years (ineligible age) • 8 pre-existing HIV (PCR +) • 2 no follow up HIV test
Enrolled eligible: 889
Vulindlela: 611 eThekwini: 278 Enrollment & Retention
Enrolled Eligible: 889
Tenofovir: 445 Placebo: 444
• 15 lost to follow up • 10 lost to follow up • 8 terminated early Retention: • 12 terminated early • 1 died 94.8%
Completed study: 422 Completed study: 421
10 Comparability of the study arms at baseline: Selected sexual behavioral characteristic s
Tenofovir Placebo p- value N=445 N=444
Mean age at sexual debut 17.4 17.4 0.78
Mean # of sexual partners 3.0 3.6 0.78
Vaginal sex only (past month) 92.8% 92.3% 0.99
Any Anal sex (past month) 0.4% 0.5% 1.00
Coital frequency (past month) 8.6 8.2 0.49
Condom use always 28.8% 29.5% 0.99 Effectiveness of tenofovir gel in preventing HIV infection
Tenofovir Placebo
# HIV infections 38 60
Women-years (# women) 680.6 (445) 660.7 (444)
HIV incidence 5.6 9.1 (per 100 women-years)
Incidence rate ratio: 0.61 (CI: 0.4 to 0.94); p = 0.017
39% lower HIV incidence in tenofovir gel group
12 HIV infection rates in the tenofovir and placebo gel groups: Kaplan-Meier survival probability
0.20
0.18 Placebo 0.16
0.14 p=0.019p=0.017 0.12
0.10 Tenofovir 0.08 Probability of HIV infection 0.06 0.04 0.02 0.00
Monthsoffollow0.0-up 0.56 1.012 1.518 2.024 2.530 CumulativeHIVendpoints 37 65 Years 88 97 98 Cumulative women-years 432 833 1143 1305 1341
HIVincidencerates 6.0vs 11.2 5.2vs 10.5 5.3vs 10.2 5.6vs 9.4 5.6vs 9.1 (TenofovirvsPlacebo) Effectiveness 47% 50% 47% 40% 39% (p-value) (0.069) (0.007) (0.004) (0.013) (0.017)(0.019) Impact of adherence on effectiveness of tenofovir gel
HIV incidence # HIV N Effect TFV Placebo
High adherers 36 336 4.2 9.3 54% (>80% gel adherence)
Intermediate adherers 20 181 6.3 10.0 38% (50-80% adherence)
Low adherers (<50% gel adherence) 41 367 6.2 8.6 28%
16 21
Impact of tenofovir gel on Herpes Simplex Virus Type-2 infection Global epidemic of HSV-2 infection
High prevalence of HSV-2 infection • ~ 20% in sexually active adults globally • ~ 50 - 60% in South African sexually active adults • ~ 80% in HIV infected men and women globally
Commonest cause of genital ulcer disease
HSV-2 infection almost entirely asymptomatic up to 90% of HSV-2 positive have no prior GUD
Acyclovir & other antivirals effective HSV-2 treatment (viral suppression) but do not prevent or cure HSV-2 Purpose To assess the impact of coitally related tenofovir gel on HSV-2 acquisition in high risk women in South Africa
Methods • Samples for HSV-2 testing Stored sera (or plasma, if serum unavailable) tested Enrolment and study exit visit or last available visit • Kalon HSV-2 type specific EIA (Kalon Biological Ltd, United Kingdom) Sensitivity = 96.4% (CI 89.8% - 99.3%) Specificity = 99.1% (CI 96.8% - 99.5%) HSV-2 status at enrolment and study exit
Enrolled in HIV trial: 889 1 missing (CAPRISA 004) 454 HSV-2 positive
At risk for HSV-2: 434
Tenofovir: 208 Placebo: 226
3 missing 1 missing 3 equivocal 1 equivocal exit results exit result
Completed study: 202 Completed study: 224 Impact of tenofovir gel on HSV- -2 incidence
Tenofovir gel Placebo gel n=202* n=224*
# HSV-2 infections 29 58
Women-years of follow-up 292.3 287.3
HSV-2 incidence per 100wy 9.9 20.2 (95% CI) (6.6, 14.2) (15.3, 26.1)
*Note: Excludes equivocal HSV-2 results at study exit IRR = 0.49 (CI:0.30, 0.78); p = 0.003
51% protection against HSV-2 by tenofovir gel (CI: 22%-70%) How safe was tenofovir gel in CAPRISA 004?
• No major safety concerns – More diarrhea in the tenofovir gel group • No tenofovir resistance identified – Resistance tests checked in seroconverters • Safe in Hepatitis B virus infected women – Evidence from physical exams and laboratory tests of the liver • No evidence of increased risk behavior (also called “risk compensation”)
No resistance in seroconverters
• No tenofovir-related drug resistance found in women who acquired HIV infection during study • Standard drug resistance testing methods used • CAPRISA 004 team using new highly sensitive DNA technologies – Will be able to test participants for rare drug resistant types of HIV – These results are expected in a few months The VOICE Study
Z. Mike Chirenje MD FRCOG UZ-UCSF Collaborative Research Program Harare, Zimbabwe Jeanne Marrazzo MD MPH University of Washington Seattle, Washington, U.S. Protocol Co-Chairs, VOICE Study
16 VOICE Sites
• UGANDA – Makerere Univ./JHU, Kampala (1) • ZIMBABWE – UZ-UCSF, Harare (1) – UZ-UCSF, Chitungwiza (2) • MALAWI – College of Medicine Blantyre (1) • SOUTH AFRICA (DURBAN AREA) – Medical Research Council (7) – CAPRISA (1) • SOUTH AFRICA (JOHANNESBURG AREA) – RHRU Research and Training Centre (1) – PHRU (1) • SOUTH AFRICA (KLERKSDORP AREA) – Aurum Institute (1)
VOICE Design (Version 2.0)
Multi-site Total Double-blinded (5,000) Randomized
Two Routes of Vaginal Oral ARV Delivery (2,000) (3,000)
Placebo TFV Gel Placebo Gel Tenofovir Truvada Five Study Arms Tablet (1,000) (1,000) (1,000) (1,000) (1,000) The VOICE Study
Tenofovir Tenofovir Gel Truvada Tablets Tablets Which is effective? Is each safe? Which will women use?
VOICE Study Objectives
• Primary objectives – Effectiveness of daily tenofovir gel compared to placebo gel, and effectiveness of oral tenofovir & oral Truvada compared to oral placebo – Extended safety of daily tenofovir gel, oral tenofovir, and oral Truvada • Secondary objectives – Adherence, behavior, drug resistance, PK and delayed seroconversion
VOICE: Key Questions Addressed
• Effectiveness of each product relative to corresponding placebo • Substantial woman-years of f/u of safety data for oral and topical tenofovir – VOICE and CAPRISA 004 together ~2600 woman-years of safety data on tenofovir gel • Powered to compare adherence among different trial arms – Will be able to look at adherence by study group: gel vs. tablet
Key Questions Addressed
• Effectiveness of each product relative to corresponding placebo • Substantial woman-years of f/u of safety data for oral and topical tenofovir • Powered to compare adherence among different trial arms – Will be able to look at preferences in study population: gel vs. pill
Current Status (as of 13 AUG 2010)
3805 Screened
2015 Screened out
Uganda South Africa Zimbabwe Malawi 149 680 279 (pending)
Total Enrollment = 1211 Screen to Enroll Ratio = 2.7:1 Timeline
Concept Presented to Version Version Projected US NIH 1.0 2.0 End F/U Q2 2007 Q2 2008 Q4 2010 Q2 2012
Protocol First Site Projected Development Activation End Accrual Meeting Q3 2009 Q2 2011 Q4 2007
• Recent DSMB review in Q2 2010 • next in Q2 2011 • Study results anticipated available in Q1 2013 Opportunities in VOICE
• Chance to confirm effectiveness of TFV gel for prevention via examination of daily dosing • Investigate HSV-2 incidence by study arm via plasma archive • Additional adherence/safety data in multiple sub- Saharan African countries • Develop post-trial access protocols for non-pregnant, pregnant and breastfeeding women • Examine critical issues in male partner involvement/adherence - October 2010 VOICE & CAPRISA 004
VOICE CAPRISA 004
Location S. Africa, Zimbabwe, S. Africa (Durban area) Uganda, Malawi Participants ~5000 women 889 women 18 - 40 years 18 - 40 years Inclusions Intercourse last 3 months Intercourse >2 times, last 30 days Exclusions Similar Similar Study product DAILY BEFORE & AFTER SEX dosing strategy What about VOICE?
• VOICE more important than ever! • Basic VOICE design not expected to change – Will proceed with already planned increase in trial size and time on product – But, can we ask and answer other important questions about tenofovir gel, based on what we have learned from CAPRISA 004? • Regulators will be watching VOICE to see if CAPRISA 004 results are confirmed For the HIV prevention field
• The hope and encouragement we have been waiting for! • Highlights the potentially huge importance of adherence for any user-controlled biomedical intervention we study • Time is drawing sooner when we may no longer have placebo-controlled trials for HIV prevention What’s next? • WHO/UNAIDS consultation on results of CAPRISA and future of tenofovir gel – August 2010 – Representatives from CAPRISA 004, VOICE, other scientists, regulatory, community • Data needed for product licensure may vary by country – many eyes are on question across the world
Acknowledgements
MTN is funded by NIAID (5U01AI068633), NICHD and NIMH, all of the U.S. National Institutes of Health THANK YOU!!!