Modelling HIV Transmission and Treatment for US MSM to Estimate the Impact of HPTN 078 on HIV Incidence

Transcription

1 Modelling HIV Transmission and Treatment for US MSM to Estimate the Impact of HPTN 078 on HIV Incidence Kate Mitchell, PhD Imperial College London London, UK 14 th June 2016

2 Introduction: HPTN 078 Enhancing Recruitment, Linkage to Care and Treatment for HIV-Infected Men Who Have Sex with Men (MSM) in the United States Protocol chair: Chris Beyrer Protocol co-chair: Robert Remien Trial sites: Atlanta GA, Baltimore ML, Birmingham AL, Boston MA

3 Introduction: HPTN 078 Enhancing Recruitment, Linkage to Care and Treatment for HIV-Infected Men Who Have Sex with Men (MSM) in the United States Intervention (n = 178) Case Manager Intervention Package Deep-Chain Respondent Driven Sampling MSM Not virally suppressed Individual Randomization Control (n = 178) SOC for Linkage and Treatment

4 Introduction: HPTN 078 Enhancing Recruitment, Linkage to Care and Treatment for HIV-Infected Men Who Have Sex with Men (MSM) in the United States Intervention (n = 178) Case Manager Intervention Package MSM Not virally suppressed Control (n = 178) SOC for Linkage and Treatment Deep-Chain Respondent Driven Sampling Individual Randomization Modelling: predict population-level HIV incidence reduction

5 HPTN 078: Modelling Centre involvement Before During After Protocol design Questionnaires Model development Data analysis Model calibration Initial analyses Final impact analysis

6 HPTN 078: Modelling Centre involvement Before During After Protocol design Questionnaires Model development Data analysis Model calibration Initial analyses Final impact analysis

7 Introduction: Baltimore HIV prevalence among MSM: 41% 2011, 31% 2014 (NHBS) Higher HIV prevalence among black MSM than white MSM Levels of viral suppression among MSM: 24% 2011, 31% 2013 (Maryland Health Dept) HIV prevalence by race Baltimore MSM 60% 40% 20% black MSM white MSM 0% Year NHBS data

8 Model development and calibration Stages: Model design, coding and development Finding model inputs (parameters) and calibration data Review of data Data analysis Calibration of model to data

9 Model structure HIV disease progression: CD4 decline by viral load Risk groups: age (<25, 25+) x race (black, white)

10 Model structure HIV disease progression: CD4 decline by viral load Risk groups: age (<25, 25+) x race (black, white) Care cascade: Never testing, undiagnosed Testing, undiagnosed Diagnosed In care On ART Dropped out

11 Model inputs (parameters) Domain Examples Data source Disease progression Infection probabilities Intervention efficacy Sexual risk behaviour Intervention behaviour Initial CD4 and viral load distribution HIV-related mortality CD4 progression rates Per-sex-act transmission probability Relative infectiousness different disease stages Reduction in HIV transmission risk: condoms, ART Number and type of partners Condom use Age and race of partners HIV testing Linkage/dropout from HIV care ART linkage and dropout Published studies: cohorts in North America and Europe Published studies: meta analyses, study of Australian MSM Published studies: clinical trials, metaanalyses NHBS surveillance data, 078 trial NHBS surveillance data, clinical cohorts, state health department data, 078 trial

12 Model inputs (parameters) Domain Examples Data source Disease progression Infection probabilities Intervention efficacy Sexual risk behaviour Intervention behaviour Initial CD4 and viral load distribution HIV-related mortality CD4 progression rates Per-sex-act transmission probability Relative infectiousness different disease stages Reduction in HIV transmission risk: condoms, ART Number and type of partners Condom use Age and race of partners HIV testing Linkage/dropout from HIV care ART linkage and dropout Published studies: cohorts in North America and Europe Published studies: meta analyses, study of Australian MSM Published studies: clinical trials, metaanalyses NHBS surveillance data, 078 trial NHBS surveillance data, clinical cohorts, state health department data, 078 trial

13 Model inputs (parameters) Domain Examples Data source Disease progression Infection probabilities Intervention efficacy Sexual risk behaviour Intervention behaviour Initial CD4 and viral load distribution HIV-related mortality CD4 progression rates Per-sex-act transmission probability Relative infectiousness different disease stages Reduction in HIV transmission risk: condoms, ART Number and type of partners Condom use Age and race of partners HIV testing Linkage/dropout from HIV care ART linkage and dropout Published studies: cohorts in North America and Europe Published studies: meta analyses, study of Australian MSM Published studies: clinical trials, metaanalyses NHBS surveillance data, 078 trial NHBS surveillance data, clinical cohorts, state health department data, 078 trial

14 Model calibration (fitting) Ensure that model outputs can reproduce observed data: HIV prevalence trends - by age and race (NHBS) MSM demography (NHBS, census) Age and race Viral suppression level (NHBS, Maryland Health Dept, CDC national estimates)

15 Model calibration: challenges Uncertainty around MSM population demography 60% % aged overall NHBS data census data 40% 20% 0% Year

16 Model calibration: challenges Uncertainty around MSM population demography 60% % aged overall NHBS data census data 40% 20% 0% Year Dealt with: fitted to 2 different scenarios

17 Model calibration: challenges Uncertainty around MSM population demography 60% 40% 20% % aged overall NHBS data census data 0% Year Dealt with: fitted to 2 different scenarios Uncertainty in current level of viral suppression 100% 75% 50% 25% 0% 2011 estimates MSM NHBS Baltimore Health dept Baltimore CDC National diagnosed on ART virally suppressed

18 Model calibration: challenges Uncertainty around MSM population demography 60% 40% 20% % aged overall NHBS data census data 0% Year Dealt with: fitted to 2 different scenarios Uncertainty in current level of viral suppression 100% 75% 50% 25% 0% 2011 estimates MSM NHBS Baltimore Health dept Baltimore CDC National diagnosed on ART virally suppressed Dealt with: wide uncertainty range

19 Model calibration (fitting) Select combinations of inputs giving model outputs reproducing demography, HIV prevalence and viral suppression data Age HIV prevalence 60% 40% 20% % aged overall census data census fits NHBS fits NHBS data 0% Year 80% 60% 40% 20% black MSM 0% white Year MSM 80% 60% 40% 20% 0% Year 2015

20 Initial analysis What level of viral suppression must be reached to reduce HIV incidence among Baltimore MSM by 20% or 50% after 5 and 10 years?

21 Initial analysis meeting incidence reduction targets 0.03 HIV incidence Baseline 100% % of HIV-infected MSM virally suppressed % 60% % Baseline 20% %

22 Initial analysis meeting incidence reduction targets HIV incidence Baseline 20% 100% 80% 60% 40% % of HIV-infected MSM virally suppressed 5 years 10 years 20% 5 years Baseline 0%

23 Initial analysis meeting incidence reduction targets HIV incidence Baseline 100% 80% 60% 40% % of HIV-infected MSM virally suppressed 5 years 10 years 20% 5 years % Baseline 0%

24 Preliminary results: required increase in viral suppression level To reduce incidence by 20% absolute increase in VS after 5 yrs overall 17% (12-25%) 11% (8-23%) 5 years 10 years

25 Preliminary results: required increase in viral suppression level absolute increase in VS after 5 yrs To reduce incidence by 20% overall 17% (12-25%) 11% (8-23%) 5 years 10 years absolute increase in VS after 5 yrs To reduce incidence by 50% Could not be reached for 10% of fits overall 42% (32-59%) 37% (25-58%) 5 years 10 years

26 Preliminary results: required increase in viral suppression level To reduce incidence 20% by demography census NHBS absolute increase in VS after 5 yrs years 10 years

27 Preliminary results: required increase in viral suppression level To reduce incidence 20% by demography To reduce incidence 20% by initial viral suppression census NHBS absolute increase in VS after 5 yrs years 10 years absolute increase in VS after 5 yrs % 25% 35% 45% % HIV-positive MSM virally suppressed 2015

28 Summary Large increases in viral suppression are needed to achieve moderate reductions in HIV incidence among Baltimore MSM, especially short-term Results are robust to uncertainty in MSM demography Results are influenced by uncertainty in current levels of viral suppression

29 Next steps Estimate how much individual intervention components - testing, linkage, ART initiation and retention - need to be increased to achieve this reduction in HIV incidence Estimate predicted impact upon HIV incidence of meeting UNAIDS and CDC targets for diagnosis, treatment and viral suppression Extend analysis to other HPTN 078 sites Atlanta, Boston, Birmingham

30 ACKNOWLEDGEMENTS The HIV Prevention Trials Network is sponsored by the National Institute of Allergy and Infectious Diseases, the National Institute of Mental Health, and the National Institute on Drug Abuse, all components of the U.S. National Institutes of Health. HPTN Modelling Centre: Marie-Claude Boily, Dobromir Dimitrov, Tim Hallett, Christophe Fraser NHBS data: Gabriela Paz-Bailey, Brooke Hoots, Danielle German, Colin Flynn HPTN 078: Chris Beyrer, Robert Remien, Protocol and site teams Imperial College London: Anne Cori, Mike Pickles

31