Transcript The Bangkok Tenofovir Study

Bangkok Metropolitan
Administration
Thailand Ministry of
Public Health
The Bangkok Tenofovir Study
An HIV pre-exposure prophylaxis trial in
Thailand: participant adherence and
study results
A collaborative project involving: The Bangkok
Metropolitan Administration, the US Centers for
Disease Control and Prevention, and the Thailand
Ministry of Public Health
National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention
Division of HIV/AIDS Prevention
Background
• 2.5 million people infected with HIV in 2011
• One in ten of these new HIV infections caused by injecting
drug use
• Thailand, HIV spread rapidly among PWID in late 1980s and
prevalence remains high: 30% to 50%
Completed PrEP trials showing efficacy
1Grant
Trial
Risk Group
ARV used mITT
2° adherence analyses
iPrEx1
MSM
TDF/FTC
44%
92% (Drug level)
TDF22
Heterosexuals
TDF/FTC
62%
78% (As-treated)
Partners3
Discordant
TDF
67%
86% (Drug level)
Partners3
Discordant
TDF/FTC
75%
90% (Drug level)
RM, et al. Preexposure Chemoprophylaxis for HIV Prevention in Men Who Have Sex with Men. NEJM 2010;363: 2587-99.
MC, et al. Antiretroviral Preexposure Prophylaxis for Heterosexual HIV Transmission in Botswana. N Engl J Med 2012;367:423-34.
3Baeten JM, et al. Antiretroviral Prophylaxis for HIV Prevention in Heterosexual Men and Women. N Engl J Med 2012;367:399-410.
2Thigpen
Objectives and Design
Primary Objectives
• Determine if tenofovir reduces risk of HIV among PWID
• Determine if tenofovir safe among HIV-uninfected PWID
Design
• Phase 3, randomized, double-blind, placebo-controlled trial
• HIV-uninfected PWID
• Randomized to receive tenofovir or placebo
• BMA and MOPH ERC and CDC IRB approved protocol,
consent, and other trial materials
• DSMB annual safety reviews and interim efficacy analysis
Services offered to volunteers
• Individualized risk reduction education and
counseling
• Methadone maintenance
• Social services and primary medical care with
referrals
• Needles not provided: illegal, available in
pharmacies without prescription
Community Involvement
• Focus Group Discussions among PWID
– Assess interest, knowledge, concerns about PrEP
• Outreach to CBOs and NGOs working
with PWID and communities at risk of HIV
• Community Relations Committee
–Meet every 2 months with investigators
–PWID from each of 17 BMA clinics
Bangkok Metropolitan Administration
Drug Treatment Clinics
TUC Lab
BMA Lab
Bangkok
Thailand
Activities
Visitsss Activities
Monthly • Study drug diary
• DOT vs. nonDOT
• Adherence and risk reduction counseling
• Oral HIV test
3 Monthly • All monthly activities, plus:
• Assessed risk behavior
• Blood collected for safety testing
Daily visits (DOT) • Staff watch participant take study drug
Participant Flow
4094 Screened
1492 (36%) Ineligible
662 Abnormal lab results
447 HIV-positive
233 Hepatitis B surface antigen
101 Medical conditions
49 Other reasons
2413 (59%) Randomized
1204 assigned to Tenofovir
Included in ITT
189 (5%) Did not return to enroll
1209 assigned to Placebo
Included in ITT
2 HIV infected at enrollment
1204 Included in mITT analysis
4843 person-years
1207 Included in mITT analysis
4823 person-years
Baseline Characteristics
Tenofovir
Placebo
Total
Characteristic
(N = 1204)
(N = 1209)
(N = 2413)
Male — no. (%)
958 (79.6)
966 (79.9)
1924 (79.7)
Female — no. (%)
246 (20.4)
243 (20.1)
489 (20.3)
20-29 yr
516 (42.9)
517 (42.8)
1033 (42.8)
30-39 yr
458 (38.0)
450 (37.2)
908 (37.6)
40-49 yr
175 (14.5)
183 (15.1)
358 (14.8)
50-60 yr
55 (4.6)
59 (4.9)
114 (4.7)
Primary or less (<6 years)
561 (46.6)
593 (49.0)
1154 (47.8)
Secondary (7 - 12 years)
545 (45.3)
500 (41.4)
1045 (43.3)
98 (8.1)
116 (9.6)
214 (8.9)
Age group — no. (%)
Education level — no. (%)
Post-secondary
Baseline Characteristics
Characteristic
Tenofovir
Placebo
Total
Risk data available on 2405 participants
(N = 1201)
(N = 1204)
(N = 2405)
In police holding cell (jail) in the past 3 months
272 (22.6)
280 (23.3)
552 (23.0)
In prison in the past 3 months
200 (16.6)
189 (15.7)
389 (16.2)
Currently in methadone program
257 (21.4)
267 (22.2)
524 (21.8)
Injected drugs in past 3 months
739 (61.5)
768 (63.8)
1507 (62.7)
Injected heroin
268 (22.3)
259 (21.5)
527 (21.9)
Injected methamphetamine
416 (34.6)
385 (32.0)
801 (33.3)
Injected midazolam
270 (22.5)
289 (24.0)
559 (23.2)
Injected other drugs
76 (6.3)
97 (8.1)
173 (7.2)
Daily
101 (8.4)
103 (8.6)
204 (8.5)
Weekly
267 (22.2)
274 (22.8)
541 (22.5)
Less than weekly
371 (30.9)
391 (32.5)
762 (31.7)
222 (18.5)
213 (17.7)
435 (18.1)
Incarceration
Drug use
Injection frequency in the past 3 months — no. (%)
Shared needles in past 3 months — no. (%)
Baseline Characteristics
Characteristic
Tenofovir
Placebo
Total
(N = 1201)
(N = 1204)
(N = 2405)
Sexual behaviours
Number of opposite sex sexual partners in past 3 months
0
365 (30·4)
334 (27·7)
699 (29·1)
1
585 (48·7)
599 (49·8)
1184 (49·2)
>1
251 (20·9)
271 (22·5)
552 (21·7)
Reported sex with live-in partner in past 3 months
526 (43·8)
518 (43·0)
1044 (43·4)
Reported sex with casual partner in past 3 months
433 (36·0)
481 (40·0)
914 (38·0)
35 (3·7)
56 (5·8)
91 (4·8)
Male participants (n=1913) reporting sex with
male partner in past 3 months
Retention
Reason
Tenofovir
(n=1204)
Placebo
(n=1209)
All
(n=2413)
Log-rank
p-value
Lost to follow-up
179 (14.9)
176 (14.6)
355 (14.7)
0.83
Withdrew consent
112 (9.3)
95 (7.9)
207 (8.6)
0.22
Death
49 (4.1)
58 (4.8)
107 (4.4)
0.34
Ineligible
3 (0.2)
6 (0.5)
9 (0.4)
0.32
Withdrawn for medical reason
4 (0.3)
4 (0.3)
8 (0.3)
0.99
Withdrawn because refused
study activities or threat to staff
3 (0.2)
3 (0.2)
6 (0.2)
0.99
Pregnant
33 (2.7)
25 (2.1)
58 (2.4)
0.37
False reactive HIV test
9 (0.8)
10 (0.8)
19 (0.8)
0.63
Efficacy
Infections/
person-years
Incidence/100 pys
(95% CI)
Efficacy (95% CI)
P-value
Tenofovir
(n=1204)
17/4843
0.35/100 (0.21 – 0.56)
48.9 (9.6 – 72.2)
0.01
Placebo
(n=1207)
33/4823
0.68/100 (0.47 – 0.96)
Tenofovir
(n=1204)
17/4843
0.35 (0.21 – 0.56)
Placebo
(n=1209)
35/4823
0.73 (0.51 – 1.01)
mITT
ITT
All (n=2411) 50/9665
0.52/100 (0.38 – 0.68)
51.8 (15.3 – 73.7) 0.01
0.08
0.10
Modified Intention-To-Treat Analysis
Kaplan-Meier estimates of time to HIV infection
0.00
0.02
0.04
0.06
mITT 48.9% (9.6 – 72.2); P=0.01
0
Number at risk
TDF 1147
Placebo 1158
12
24
1007
1029
933
948
36
48
Months on study
857
844
TDF
736
722
60
72
84
521
500
241
234
0
0
Placebo
0.08
0.10
Modified Intention-To-Treat Analysis
Kaplan-Meier estimates of time to HIV infection
0.00
0.02
0.04
0.06
mITT 48.9% (9.6 – 72.2); P=0.01
0
Number at risk
TDF 1147
Placebo 1158
12
24
1007
1029
933
948
36
48
Months on study
857
844
TDF
736
722
60
72
84
521
500
241
234
0
0
Placebo
Pre-specified adherence-defined analysis
• Adherent defined: DOT participant taking
study drug >71% of days with no more than 2
consecutive day off study drug
• Efficacy = 56% (-25.1 - 84.5; p=0.12)
• Limiting analysis to participants with
detectable tenofovir, efficacy = 74% (16.6 –
94.0; p=0.03)
Unmatched case-control study
limited to tenofovir recipients
• The odds of HIV infection were 3-times lower
(OR 0.30; 95% CI, 0.09 to 0.98; p=0.04) among
participants with detectable tenofovir levels,
compared to those with undetectable
tenofovir
• Represents reduction in risk 70% (95% CI, 2.3
to 90.6)
Adherence by treatment group
100%
Overall, participants in DOT
follow-up 87% of days on study
80%
P=0.16
60%
nonDOT
DOT
40%
20%
0%
Tenofovir
Placebo
Adherence among HIV infected and
not-infected participants
Tenofovir
Placebo
100%
100%
80%
80%
P=0.45
60%
60%
40%
20%
P=0.01
Quartiles
HIV-infected
adherence
Not infected
adherence
75%
98.8%
98.8%
Median
94.0%
94.3%
25%
65.0%
81.2%
0%
40%
20%
Quartiles
HIV-infected
adherence
Not infected
adherence
75%
94.1%
98.6%
Median
76.5%
94.0%
25%
65.3%
77.8%
0%
HIV-infected
Not infected
HIV-infected
Not infected
Efficacy by level of adherence
Efficacy
100%
84%
80%
68%
60%
54%
72%
58%
49%
40%
20%
mITT
>67%
>75%
>90%
Adherence
>95%
>97.5%
Safety
Tenofovir (n=1204)
Placebo (n=1209)
no. of participants (%) no. events no. of participants (%) no. events
Adverse Event (AE)
†p
value
Any AE
1098 (91.2)
10965
1083 (89.6)
11550
0.455
Any serious AE
227 (18.9)
340
246 (20.3)
375
0.352
49 (4.1)
49
58 (4.8)
58
0.369
156 (13.0)
414
160 (13.2)
389
0.886
Nausea/vomit
96 (8.0)
113
59 (4.9)
71
0.002
Renal disease
13 (1.1)
18
11 (0.9)
15
0.687
Creatinine: grade 1
37 (3.1)
114
28 (2.3)
33
0.268
Creatinine: grade 2
2 (0.2)
3
0 (0)
0
0.249
Creatinine: grade 3 or 4
3 (0.2)
4
3 (0.2)
3
0.996
Death
Any grade 3 or 4 event
4
Nausea
3
2
Tenofovir
1
Placebo
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Months since randomization
Safety
Tenofovir (n=1204)
Placebo (n=1209)
no. of participants (%) no. events no. of participants (%) no. events
Adverse Event (AE)
†p
value
Any AE
1098 (91.2)
10965
1083 (89.6)
11550
0.455
Any serious AE
227 (18.9)
340
246 (20.3)
375
0.352
49 (4.1)
49
58 (4.8)
58
0.369
156 (13.0)
414
160 (13.2)
389
0.886
Nausea/vomit
96 (8.0)
113
59 (4.9)
71
0.002
Renal disease
13 (1.1)
18
11 (0.9)
15
0.687
Creatinine: grade 1
37 (3.1)
114
28 (2.3)
33
0.268
Creatinine: grade 2
2 (0.2)
3
0 (0)
0
0.249
Creatinine: grade 3 or 4
3 (0.2)
4
3 (0.2)
3
0.996
Death
Any grade 3 or 4 event
4
Nausea
3
2
Tenofovir
1
Placebo
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Months since randomization
Genotyping and Resistance
• HIV viruses from 49 of the 52 HIV-infected
participants were successfully amplified:
– 43 (88%) were CRF01_AE
– 5 (10%) were subtype B'
– 1 (2%) was CRF01_AE/subtype B' strain
• No resistance mutations associated with
tenofovir (K65R, K70E) were detected
Limitations
• We used self reports of risk behaviors
• HIV infections may have occurred due to
sexual activity [MOLBPE27]
Oral pre-exposure prophylaxis trials
mITT results
75% (55-87)
67% (44-81)
Partners (serodiscordants): TDF/FTC
TDF
TDF2 (heterosexuals): TDF/FTC
62% (22-83)
Bangkok Tenofovir Study (PWID): TDF
49% (10-72)
iPrex (MSM): TDF/FTC
44% (15-63)
FemPrEP (women): TDF/FTC
VOICE (women): TDF
TDF/FTC
0%
10
20
30
40
50
Efficacy
60
70
80
90
100%
Oral pre-exposure prophylaxis trials
2° adherence-based results
90% (56-98)
86% (57-95)
Partners (serodiscordants): TDF/FTC
TDF
TDF2 (heterosexuals): TDF/FTC
78% (41-94)
84%
Bangkok Tenofovir Study (PWID): TDF
74% (17-94)
iPrex (MSM): TDF/FTC
92% (40-99)
FemPrEP (women): TDF/FTC
VOICE (women): TDF
TDF/FTC
0%
10
20
30
40
50
Efficacy
60
70
80
90
100%
Conclusions
• Daily oral tenofovir, in combination with other
HIV prevention strategies, reduced the risk of
HIV infection among people who inject drugs
• The protective efficacy increased with
improved adherence
• We did not identify significant safety concerns,
and no tenofovir resistance detected
Next Steps
• Additional analyses underway: adherence, risk
behavior, drug level
• Participants will be offered tenofovir for one
year
• CDC published initial guidance in MMWR
• CDC is currently working to finalize full Public
Health Service clinical guidelines on PrEP use
Bangkok Tenfovir Study Group
Principal Investigator Kachit Choopanya
Advisory Group
Sompob Snidvongs Na Ayudhya, Sithisat
Chiamwongpaet, Kraichack Kaewnil, Praphan
Kitisin, Malinee Kukavejworakit, Manoj
Leethochawalit, Pitinan Natrujirote, Saengchai
Simakajorn, Wonchat Subhachaturas
Study Clinic Coordination Team
Lead: Suphak Vanichseni; Members: Boonrawd
Prasittipol, Udomsak Sangkum, Pravan
Suntharasamai
Bangkok Metropolitan Administration
Rapeepan Anekvorapong, Chanchai
Khoomphong, Surin Koocharoenprasit,
Parnrudee Manomaipiboon, Siriwat Manotham,
Pirapong Saicheua, Piyathida Smutraprapoot,
Sravudthi Sonthikaew, La-Ong Srisuwanvilai,
Samart Tanariyakul, Montira Thongsari,
Wantanee Wattana, Kovit Yongvanitjit
Thailand Ministry of Public Health
Sumet Angwandee, Somyot Kittimunkong
Thailand MOPH - U.S. CDC Collaboration
Wichuda Aueaksorn, Benjamaporn Chaipung,
Nartlada Chantharojwong, Thanyanan
Chaowanachan, Thitima Cherdtrakulkiat,
Wannee Chonwattana, Rutt Chuachoowong,
Marcel Curlin, Pitthaya Disprayoon, Kanjana
Kamkong, Chonticha Kittinunvorakoon, Wanna
Leelawiwat, Robert Linkins, Michael Martin,
Janet McNicholl, Philip Mock, Supawadee NaPompet, Tanarak Plipat, Anchala Sa-nguansat,
Panurassamee Sittidech, Pairote Tararut,
Rungtiva Thongtew, Dararat Worrajittanon,
Chariya Utenpitak, Anchalee
Warapornmongkholkul, Punneeporn
Wasinrapee
U.S. CDC
Jennifer Brannon, Monique Brown, Roman
Gvetadze, Lisa Harper, Lynn Paxton, Charles
Rose
Johns Hopkins University
Craig Hendrix, Mark Marzinke
Community Involvement
• Conducted Focus Group Discussions among PWID in 2004
– PREP trial important, why tenofovir, side-effects, access to care
• Outreach to organizations working with PWID and
communities at risk of HIV/AIDS
– Met with 16 CBOs and NGOs: distributed draft protocols, consents,
education tools
• Met with community representatives at 3 drop-in sites
• Meetings informed development of protocol, consent form
and consent process, and education materials
• Community Relations Committee
– PWID from each of the 17 BMA clinics
– Committee meets every 2 months with investigators
– Raised concerns about study procedures, confidentiality,
police harassment, incarceration, side-effects, medical
care, compensation, and post-trial access to tenofovir
Injecting and sharing by treatment group
80%
% Reporting
Tenofovir -------Placebo - - - - - 60%
No difference in TDF/placebo
groups (p=0.71)
40%
Injecting declined (p<0.001)
20%
No difference in TDF/placebo
groups (p=0.84)
Sharing declined (p<0.001)
0%
0
12
24
36
48
60
72
Study visit (month)
TDF 1201
919
840
771
662
521
341
Placebo 1204
960
874
770
667
512
326
Sex by treatment group
50%
No difference in TDF/placebo
groups (p=0.28)
40%
% Reporting
Tenofovir -----------Placebo - - - - - -
30%
Sex with casual partners
declined (p<0.001)
20%
10%
No difference in TDF/placebo
groups (p=0.24)
Sex with more than one
partner declined (p<0.001)
0%
0
12
24
36
48
60
72
Study visit (month)
TDF 1201
919
840
771
662
521
341
Placebo 1204
960
874
770
667
512
326
Risk Behavior
Results
• Risk behavior (injecting, sharing, sex) declined
during follow-up
• Participants randomized to tenofovir and placebo
reported similar HIV-associated risk behaviors
• Multivariable analysis: age (20-29 years) (p=0.02),
sharing needles (p<0.001), and prison (p=0.003)
associated with HIV infection
• Reporting sex with live-in, casual, or men
reporting sex with male partner not associated
with incident HIV
Needles
• Thai law prohibits distribution of needles to
inject illicit drugs
• Needles and syringes are available in
pharmacies at low cost without a prescription
• Participants and community representatives
reported that access to clean needles and
syringes is not a problem
Measures to improve adherence
• Offer daily directly observed therapy
• Monthly adherence counseling: participant centered and sharing
discussed
• Pill counts with participant at each monthly study visit
– Participants reminded that we understand taking all doses is difficult
and that we need them to tell us about missed doses
• Medication diary
– At monthly visits diaries reviewed, compared to pill count, and
discrepancies resolved
• Tenofovir plasma and intracellular levels determined
– HIV seroconverters – blood collected visit that HIV test reactive and
study drug stopped
– HIV uninfected – blood collected from volunteers at exit visit at 4
clinics (n=284 specimens matched PBMCs/plasma)
Pre-specified adherence-defined analysis
• Adherent defined: DOT participant taking
study drug >71% of days with no more than 2
consecutive day off study drug
• 17 met the adherent definition
• 5 tenofovir/12 placebo = 56% (-25.1 - 84.5;
p=0.12)
• 2 of 5 undetectable tenofovir; removing
these 74% (16.6 – 94.0; p=0.03)
Unmatched case-control study
limited to tenofovir recipients
HIV
infected
Tenofovir detected 5 (39%)
HIV
uninfected
Total
93 (67%)
98
Tenofovir not
detected
8
45
53
Total
13
138
151
• The odds of HIV infection were 3-times lower (OR 0.30;
95% CI, 0.09 to 0.98; p=0.04) among participants with
detectable tenofovir levels, compared to those with
undetectable tenofovir
• Represents reduction in risk 70% (95% CI, 2.3 to 90.6)
Plasma tenofovir levels
5/13 (38.5%) Detectable
Tenofovir (ng/ml)
400
93/138 (67.4%) Detectable
300
200
100
Median
0
0
HIV-positive
1
HIV-negative
2
Participant Flow
1681 (41%) ineligible
4094 Screened
2413 (59%) Enrolled
1204 Tenofovir
662 abnormal lab results
447 HIV-positive
233 HBsAg
189 did not return
101 medical conditions
49 other reasons
1209 Placebo
2 were HIV-infected at enrollment
1204 followed for HIV infection (mITT)
4843 person-years
Annual retention
12 months, 1059/1204 (88%)
24 months, 987/1030 (96%)
36 months, 933/956 (98%)
48 months, 860/893 (96%)
60 months, 769/788 (98%)
72 months, 596/615 (97%)
84 months, 390/399 (98%)
179 (15%) Were lost to follow-up
112 (9%) Withdrew from study
49 (4%) Died
33 (3%) Became pregnant
17 (1%) Became HIV-infected
9 (1%) Had a falsely reactive HIV test
4 (0·3%) Withdrawn for medical reason
3 (0·3%) Did not meet eligibility criteria
3 (0·3%) Had other reason
1207 followed for HIV infection (mITT)
4823 person-years
176 (15%) Were lost to follow-up
95 (8%) Withdrew from study
58 (5%) Died
25 (2%) Became pregnant
33 (3%) Became HIV-infected
10 (0·8%) Had a falsely reactive HIV test
4 (0·3%) Withdrawn for medical reason
6 (0·5%) Did not meet eligibility criteria
3 (0·3%) Had other reason
Annual retention
12 months, 1079/1209 (89%)
24 months, 1006/1046 (96%)
36 months, 944/978 (97%)
48 months, 849/886 (96%)
60 months, 758/775 (98%)
72 months, 584/595 (98%)
84 months, 375/377 (99%)
Annual loss to follow-up by study group
Tenofovir
Placebo
Total
Month 12
98
92
190/2413 (7.9%)
Month 24
24
24
48/2072 (2.3%)
Month 36
10
23
33/1943 (1.7%)
Month 48
16
20
36/1856 (1.9%)
Month 60
14
10
24/1749 (1.4%)
Month 72
10
6
16/1685 (1.0%)
Month 84
7
1
8/1623 (0.5%)