1. Failure 6th EACS Advanced HIV Course Montpellier, Sept. 3-5, 2008
Prof. Nathan CLUMECK
Head, Department of infectious diseases
Saint-Pierre University Hospital
Brussels, Belgium

2. Trends in HAART Failure
Baseline Characteristics
Patients With 2nd Virologic Failure
(n=5057)
Male (%) 85
HIV RNA (copies/mL)
15,723
Median CD4 (cells/mm3)
222
Antiretroviral naïve at HAART initiation (%) 31
Median time from HAART initiation (years)
2.9
Prior AIDS-defining
event (%) 37
Retrospective cohort study (1995 to 2005)
33,381 patients on HAART
46% had virologic failure
28% changed HAART regimen after failure
15% experienced a second virologic failure
3% died
Slide #37: NA-ACCORD: Trends in Second HAART Failure
Deeks and colleagues conducted a retrospective cohort study (1995 to 2005) to determine the frequency of second HAART failure and risk factors for resulting mortality.1
A cohort of 33,381 patients on HAART was studied. Of these, 46% had virologic failure, 28% changed HAART regimen after failure, 15% experienced a second virologic failure, and 3% died.1
Baseline characteristics of those with second virologic failure are listed in the table.1
Reference
Deeks S, North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) of the IeDEA. Trends in second virologic failure and predictors of subsequent mortality among ART-experienced patients: North American experience. Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Massachusetts. Abstract 41.
Deeks S, et al. 15th CROI. Boston, Abstract 41.

3. Trends in HAART Failure
Incidence of 2nd Virologic
Failure Over Time
Adjusted relative risk of second virologic failure has declined dramatically
Decreased from 1.46 (96-97) to 0.54 (04-05) per 100 patient-years
No improvement in mortality
Median survival: 7.1 years
Independent risk factors associated with increased risk of death
CD4 cell count and HIV RNA level at time of second virologic failure
No association
Prior treatment exposure
Pre-HAART
113.6
70.7
Incidence per 100 patient-years
Slide #38: NA-ACCORD: Trends in Second HAART Failure
The investigators found a 7.5-times decline in risk of second virologic failure from 1997 to 2005; from 113.6/100 person-years to 15.1/100 person-years. However, mortality following second HAART failure did not improve over time.1
HIV RNA levels and CD4 cell counts at the time of second virologic failure were independently associated with an increased risk of death, whereas prior treatment exposure and pre-HAART factors were not.1
These results do not take into account the potential impact of new drugs or classes of drugs that were available after 2005; a subsequent analysis is planned.1
Reference
Deeks S, North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) of the IeDEA. Trends in second virologic failure and predictors of subsequent mortality among ART-experienced patients: North American experience. Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Massachusetts. Abstract 41.
41.5
17.9
15.1
Calender Year
Deeks S, et al. 15th CROI. Boston, Abstract 41.

4. St-Pierre cohort : Viral load response to first line therapy

5. St-Pierre cohort :

6. Failure of initial HAART regimens
Rate of failure of HAART regimens have decreased in developed countries because
Better tolerance of new regimens
New regimens are easier to take
Less pill count
Once daily
No food/ water restrictions

7. EACS definition of failure
Confirmed Plasma HIV RNA > 50 copies/ml
6 months after starting therapy
But real failure or blip?

8. Incidence of low-level viremia more frequent than high-level viremia
HIV-1 RNA Level During Follow-up, %
Study Cohort
(N = 4447)
Persistently < 50 copies/mL
71.2
≥ 1 measurement > 50 copies/mL
28.8
≥ 1 measurement > 1000 copies/mL
6.7
van Sighem A, et al. J Acquir Immune Defic Syndr. 2008;48:

9. After initial virologic suppression with cART, approximately 30% of patients experienced transient viremia
7% classified as high level ( > 1000 copies/mL)
Low level viremia( c/mL) is not associated with clinical events, development of high-level viremia, or changes in CD4+ cell counts
High-level viremia is associated with occurrence of resistance and therapy changes
van Sighem A, et al. J Acquir Immune Defic Syndr. 2008;48:

10. Why do therapies fail? Virus Drug Patient Resistance Potency Toxicity
Non-adherence .

11. Adherence
Adherence may be the single most important factor affecting treatment outcomes to HAART
Many components contribute to non-adherence
- patient characteristics
- clinical care settings
- patient-provider relationships
- drug regimens characteristics

12. Checking adherence
Interview the patient to evaluate adherence and compliance
Re-explain the objectives and modalities of the treatment ant the potential risks of poor adherence
Exclude potential drug-drug or drug-food interactions
Use TDM if needed (and if available ...)

13. Improving adherence Reduce tablet load (fixed-dose)
Remove food restrictions  Adapt to P.
Reduce dose frequency (qd) way of life
Provide nurse counselling
Provide pill box
Provide psychological support

14. Increasing number of mutations
Which Barriers to resistance ?
Few change per mutation
BUT
Toxicity
NRTIs
High drug levels
Large change per mutation
NNRTIs
AND
Small change per mutation
Boosted-PI
High trough
High trough
Tox
Increasing EC50
EC50
EC50
Increasing number of mutations
Dale Kempf et al.

15. Which first-line ART ?
«  Boosted protease inhibitors have 60% lower risk of resistance compared with other classes «  (Lima et al J Infect Dis July 2008)
Factors associated with increased risk:=a higher baseline VL;an NNRTI-based combination;a high but still suboptimal adherence

16. Which options in case of failure ?

17. Management of failure
Decision to change treatment regimen must take into account :
the remaining treatment options
the level of failure based on kinetics of viral load and CD4 (decreased viral fitness)
the past treatment history, including resistance patterns, tolerability and adherence issues
Choice of treatment will be based on the number of active drugs within each class (genotypic testing)

18. ARV Failure; a historical perspective, before the 2006-08 new paradigm
First
Second-third
Multiple
Options
Many
Some
Few
Goals VL
VL/CD4
CD4
Action
Adapt or switch
Wait or switch
Do your best …

19. Before 2006 : Continue the same therapy
Stop therapy in order to revert to a wild type virus
Use a PI boosted regimen to try to overcome resistance
Re-cycle previous drugs
Use mega – or giga – HAART
Use new investigational agents, if available

20. Interruption of Therapy Continuation of Therapy?
Continuation of Therapy

21. Option 1 : Continuing therapy
Patients with CD4 < 50, failing all 3 classes, who continue therapy show better outcome, particularly those with more drugs (Miller et al., AIDS 2002)
This is probably in relationship to viral fitness (Deeks et al., 2002)
BUT this strategy may lead to development of further drug resistance, that could preclude the use of a new drug in an existing class

22. Option 2 : Stopping therapy
The goal is to restore the more sensitive wild type virus
One study (Gighaart) suggests a benefit of this strategy in patients with very low CD4 count, while studies in patients with higher CD4 suggest that this strategy is counter productive

23. Median changes in plasma viral load and CD4 cells in failing patients continuing or discontinuing therapy
From Deeks et al., NEJM, 2001

24. Median changes in plasma viral load and CD4 cells before and after switch of resistance in patients discontinuing therapy
From Deeks et al., NEJM, 2001

25. STI in patients with multiple failures
GIGHAART, Katlama et al., Boston 2003
70 patients randomized +/- STI 8 Weeks
VL 5.3 log CD4 27/mm3 Median ARV 6.6 years
Results No STI STI
At W 12/20 median VL SS
<400 cp 15% 38%
At W 24/32 median CD
>50 /mm3 25% 50% NS
HIV clinical events 3 5
At W48/56 median VL
median CD

26. Option 3 : Use of boosted regimens
The aim is to overcome resistance due to suboptimal plasma drug level
Ritonavir boosted regimens are widely used but the best results are usually obtained in the early stages of treatment (less mutations)

27. Option 4 : Recycle drugs
The goal is to maintain the selective pressure for some specific mutations such as M184V
The mutational effects on viral fitness may now be assessed by a replicative capacity assay
Patients with discordance failure (virological but not CD4) harbour viruses with lower replicative capacity (Deeks et al., 2001)

28. Option 5 : Use Mega/Giga – HAART
The goal is to keep patients, who have no options, alive and well, until new drugs become available

29. Option 6 : Use of new drugs
Enfuvirtide : Phase 3 Studies in Highly Experienced Patients
VL reduction
Enfuvirtide +OB OB p
TORO <0.0001
TORO <0.0001
Injection site reactions were the most frequent AE’s but with low discontinuation rate (3%)
Lalezari. NEJM 2003

30. Since 2006 :
Use a genotype - driven salvage therapy plus
Use at least 2 new active agents
New PI: Tipranavir, Darunavir
New NNRTI: Etravirine
CCR5 inhibitors
Fusion inhibitors: enfuvirtide
Integrase inhibitors
The new Paradigme for multi-experienced patients is to reach undetectable viral load

31. The « New Paradigme »
HIV-1 RNA suppression to < 50 copies/mL should be the therapeutic goal for treatment-experienced HIV-infected patients as defined by :
2006 US DHHS
2006 International AIDS Society-USA guidelines
2007 EACS guidelines

32. HIV-1 RNA < 50 copies/mL, %
Week 48 Virologic Efficacy of New Drugs Defined as HIV-1 RNA < 50 c/mL
Study
Drug Regimen
HIV-1 RNA < 50 copies/mL, %
TORO [1]
Enfuvirtide + OBR OBR alone
RESIST [2]
Tipranavir + OBR Comparator PI + OBR
POWE [3]
Darunavir/ritonavir + OBR Comparator PI + OBR
DUET
[4,5]
Etravirine + darunavir/ritonavir-containing OBR Placebo + darunavir/ritonavir-containing OBR
MOTIVATE [6]
Maraviroc QD + OBR Maraviroc BID + OBR Placebo + OBR
BENCHMRK
[7,8]
Raltegravir + OBR Placebo + OBR
1. Nelson M, et al. J Acquir Immune Defic Syndr. 2005;40: Hicks CB, et al. Lancet. 2006;368: Clotet B, et al. Lancet. 2007;369: Haubrich R, et al. CROI Abstract Johnson M, et al. CROI Abstract Lalezari J, et al. ICAAC Abstract H-718a. 7. Cooper DA, et al. N Engl J Med In press. 8. Steigbigel R, et al. N Engl J Med In press.

33. New PI
Tipranavir
Darunavir

34. Virologic response with respect to baseline number of LPV mutation
0–5 mutations
6–7 mutations
Percentage of responders
8–10 mutations
Calvez et al. Scotsdale resistance meeting, Kempf et al. Scotsdale resistance meeting, 2001

35. Tipranavir
nonpeptidic PI with potent activity against PI-resistant HIV-1 both in vitro and in vivo
Approved in United States and Europe for use in PI-experienced patients in combination with ritonavir

36. Saquinavir/ritonavir Amprenavir/ritonavir
Study Design RESIST 1 and 2 in which tipranavir plus OBR compared with r-boosted comparator PI (CPI) plus optimized background regimen (OBR)
Week 48
Tipranavir/ritonavir n = 746
Patients failing PI-containing HAART
(N = 1483)
Baseline genotypic resistance testing
Preselection of CPI plus OBR by investigator
CPI Arm
Lopinavir/ritonavir
Indinavir/ritonavir
Saquinavir/ritonavir
Amprenavir/ritonavir
n = 737
NRTI, NNRTI, and PI experience for ≥ 3 consecutive months
≥ 2 PI-based regimens for ≥ 3 months
On PI-based regimen at enrollment
≥ 1 documented primary PI mutation (30N, 46I/L, 48V, 50V, 82A/F/L/T, 84V, 90M)
≤ 2 mutations at codons 33, 82, 84, and 90

37. Tipranavir/Ritonavir (n = 746)
Outcome at Week 48
Tipranavir/Ritonavir (n = 746)
CPI/Ritonavir (n = 737)
P Value
Treatment response, %
33.6
15.3
< .0001
In patients using ENF
48.5
20.0
< .0001*
In patients using first-time ENF
58.5
21.6
HIV-1 RNA < 400 copies/mL, %
30.4
13.8
43.2
18.5
HIV-1 RNA < 50 copies/mL, %
22.8
10.2
28.4
14.1
Mean HIV-1 RNA reduction, log10 copies/mL (SD)
1.14 (1.30)
0.54 (1.02)
Mean CD4+ cell count increase, cells/mm3 (SD)
45 (104)
21 (89)
ENF, enfuvirtide; IQR, interquartile range.
*Comparison between use and nonuse of enfuvirtide within treatment arm.

38. Patient Outcomes
Several factors significantly and independently associated with treatment response to tipranavir/ritonavir:
ENF use: Odds Ratio 4.07 ; P < .0001
Higher tipranavir trough concentration ; Odds Ratio , 2.16 ; P < .05
Fewer baseline tipranavir mutations (0-2 vs 5-6): Odds Ratio , 0.14; P < .0001
Other factors associated with treatment response to tipranavir/r
≤ 2 primary PI mutations at baseline: 40.8% vs 31.5% (P = .03)
Prior treatment with ≤ 3 PIs: 40.9% vs 30.5% (P = .007)
CI, confidence interval.

39. Tipranavir/Ritonavir (n = 749)
Other Outcomes
Incidence of exposure-adjusted adverse events similar between arms
Significantly higher triglycerides, ALT/AST, and cholesterol in the tipranavir/ritonavir arm
Adverse Events and Grade 3/4 Laboratory Abnormalities, n (Rate per 100 Patient-Years)
Tipranavir/Ritonavir (n = 749)
CPI/Ritonavir (n = 737)
P Value
Any adverse event leading to study discontinuation
90 (12.4)
48 (10.6) --
Triglycerides
184 (30.8)
94 (23.1)
< .0001
ALT
71 (10.1)
15 (3.3)
AST
45 (6.3)
13 (2.9)
.002
Cholesterol
31 (4.3)
3 (0.7)
ALT, alanine aminotransferase; AST, aspartate aminotransferase.

40. Tipranavir-r : Conclusions in highly experienced P.
Associated with significantly superior treatment outcomes vs r-boosted CPI plus OBR through 48 weeks of treatment
Treatment response rate significantly higher with tipranavir/r
Significantly longer time to treatment failure with tipranavir/r
Inclusion of enfuvirtide in OBR increased likelihood of effective treatment outcomes
Safety profile similar to that of other r-boosted Pis

41. Disadvantages of Tipranavir-r
4 pills (2 Tipra+2 RTV 100mg) BID
Liver toxicity
Lipid profile is worsened
Drug-drug interaction:
no other PI allowed
Etravirine not allowed

42. Darunavir
Darunavir/ritonavir a potent boosted PI active against wild-type and many PI-resistant viruses[1]

43. POWER 1 and 2 : Study Design
Investigator-selected CPI(s) + OBR
DRV/RTV 400/100 mg QD
+ OBR
PI-, NRTI- and NNRTI-experienced
 1 PI mutation
PI-based regimen
VL > 1000 copies/mL
Investigator-selected CPI(s)
+ OBR (without NNRTIs)
DRV/RTV 800/100 mg QD
+ OBR
DRV/RTV 400/100 mg BID
+ OBR
DRV/RTV 600/100 mg BID
+ OBR
DRV/RTV 600/100 mg BID provided greatest virologic response in Wk 24 analysis; Now ,FDA-approved dose for treatment-experienced pts
VL, viral load; OBR, optimized background regimen (NRTIs ± enfuvirtide)
Lazzarin A, et al. IAC Abstract TUAB0104.

44. POWER 1 and 2 : VL < 50 c/mL at week 48 (ITT-TLOVR)
100
DRV/RTV 600/100 mg BID 80
*P < .001 vs comparator PI/RTV.
Control
45*
46* 60
Patients With VL< 50 c/mL (%) 40 12 10 20 1 2 4 8 12 16 20 24 28 32 36 40 44 48
Weeks
Lazzarin A, et al. IAC Abstract TUAB0104.

45. POWER 1 and 2: VL < 50 c/mL at Week 48 by Baseline Subgroups46
Overall
n = 120 10
ENF Used (Naive)
n = 35
n = 36 58 11
ENF Not Used
n = 61 44
n = 70 10
DRV/RTV 600/100 BID
≥ 3 Primary PI Mut
n = 55 44
n = 74 5
CPI/RTV
No Sensitive ARVs in OBR
n = 25 20
n = 18
≥ 1 Active Agent in OBR
n = 44 54
n = 11 11 20 40 60 80
100
Patients With VL < 50 copies/mL at Wk 48 (ITT, NC = F) (%)
Lazzarin A, et al. IAC Abstract TUAB0104.

46. Effect of Baseline DRV Fold Change on Response to DRV
Baseline fold-change to DRV (by Antivirogram) was strong predictor of Week 24 response in POWER 1, 2, and 3
100
Distribution of pts by baseline DRV FC 80 60 50
Patients With VL < 50 copies/mL at Wk 24 (%) 40 25 20 13
n =
255 65 48
FC ≤ 10
FC 11-40
FC > 40
Baseline DRV FC
DeMeyer S, et al. Resistance Workshop Abstract 73.

47. Effect of Baseline DRV-associated Mutations on Response to DRV
11 PI resistance mutations associated with reduced response
V11I, V32I, L33F, I47V, I50V, I54L, I54M, G73S, L76V, I84V and L89V
100
Distribution of pts by baseline # of DRV mutations 80 64 60
Patients With VL < 50 copies/mL at Wk 24 (%) 50 42 40 22 20 10
n = 67 94
113 58 41 1 2 3
≥ 4
Baseline No. of DRV Mutations
DeMeyer S, et al. Resistance Workshop Abstract 73.

48. Relationship Between Activity of OBR and Response to DRV/RTV 600/100
Highest rates of VL < 50 copies/mL in pts with ≥ 2 active agents in OBR[1]
No incremental benefit of active ENF if ≥ 1 active NRTI in OBR[1]
Phenotypic susceptibility score (PSS) of OBR also predicted VL < 50 at Wk 24[2]
PSS ≤ 0.5: 34%
PSS : 49%
PSS > 1.5: 52%
Characteristic of OBR
VL < 50 at Wk 24, %
# of active agents 26 1 46
≥ 2 49
Type of active agent
Active ENF only 43
≥ 1 active NRTI, no ENF 51
≥ 1 active NRTI + active ENF 53
1. Pozniak A, et al. BHIVA, Abstract P3.
2. Vangeneugden T, et al. Resistance Workshop, Abstract 1138.

49. Other Outcomes Most adverse events mild to moderate in severity
25% of patients reported ≥ 1 grade 3 or 4 adverse event
Grade 3-4 Adverse Events With Incidence ≥ 2, %
Darunavir/Ritonavir
(n = 298)
Diarrhea 14
Nausea 10
Elevated cholesterol 4
Elevated triglycerides 6
Elevated pancreatic amylase 7
Elevated ALT 2
Elevated AST
Molina JM, et al. J Acquir Immune Defic Syndr. 2007;46:24-31.

50. Summary of Key Conclusions
Darunavir/ritonavir 600/100 mg twice daily safe and effective in treatment-experienced patients with drug-resistant HIV
Rates of virologic suppression is the highest in pts with > 1 active agent in OBR
Full susceptibility to darunavir strongest predictor of response
Safety profile of darunavir/r is good
Molina JM, et al. J Acquir Immune Defic Syndr. 2007;46:24-31.

51. NNRTI
Etravirine
active against wild-type HIV and strains resistant to currently available NNRTIs in phase IIb trials

52. DUET-1[1] and 2[2] :assess long-term efficacy, safety, and tolerability of etravirine in treatment-experienced patients; 24-week results shown
Week 24
Week 48
Etravirine 200 mg BID +
Darunavir/Ritonavir-containing OBR*
(DUET-1: n = 304;
DUET-2: n = 295)
HIV-infected patients with virologic failure on current HAART regimen, history of ≥ 1 NNRTI RAM, ≥ 3 primary PI mutations, and
HIV-1 RNA > 5000 copies/mL
(DUET-1: N = 612;
DUET-2: N = 591)
Placebo +
Darunavir/Ritonavir-containing OBR*
(DUET-1: n = 308;
DUET-2: n = 296)
*Investigator-selected OBR included darunavir/ritonavir 600/100 mg twice daily + ≥ 2 NRTIs ± enfuvirtide.
1. Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

53. Baseline Characteristics
DUET-1
DUET-2
Etravirine
(n = 304)
Placebo
(n = 308)
(n = 295)
(n = 296)
Median HIV-1 RNA, log10copies/mL (range)
4.8 ( )
4.9 ( )
4.8 ( )
4.8 ( )
HIV-1 RNA ≥ 100,000 copies/mL, % 39 41 37 31
Median CD4+ count, cells/mm3 (range)
99 (1-789)
109 (1-694)
100 (1-708)
108 (0-912)
Median previous NRTIs, n 6 5
Median previous NNRTIs, n 1
Median previous PIs, n 4
≥ 4 NNRTI RAMs 21 20 17
≥ 4 NRTI RAMs 93 90
≥ 4 primary PI mutation 60 59 65 66
≤ 2 darunavir RAMs 58 56
Enfuvirtide use 40 52 53
De novo 24 26 27
PSS 0 or 1* 50 46 51
*Lower clinical cutoff (10-fold) used to define susceptibility to darunavir.
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

54. Description of Analysis
Primary endpoint: HIV-1 RNA < 50 copies/mL at Week 24
First trial to use HIV-1 RNA < 50 copies/mL as primary endpoint in treatment-experienced patients
Secondary endpoints: HIV-1 RNA < 400 copies/mL at Week 24, change in HIV-1 RNA from baseline, change in CD4+ cell count from baseline, toxicity
Analysis
Intent to treat, time to loss of virologic response
95% power to detect significance
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

55. Significantly more patients achieved HIV-1 RNA < 50 copies/mL with etravirine vs placebo
Etravirine treatment resulted in greater CD4+ cell count increases from baseline compared with placebo (statistical significance reached in DUET-1 only)
Outcome at Week 24
DUET-1
DUET-2
Etravirine
(n = 304)
Placebo
(n = 308) P
(n = 295)
(n = 296)
HIV-1 RNA < 50 copies/ mL, % 56 39
.005 62 44
.0003
HIV-1 RNA < 400 copies/mL,% 74 51
.0001 75 54
HIV-1 RNA reduction,
log10copies/mL
2.4
1.7
< .0001
2.3
Mean increase in CD4+ cell count from baseline, cells/mm3 89 64
.0002 78 66
.3692
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

56. HIV-1 RNA < 50 copies/mL at Week 24 (%)
Etravirine + OBR
Placebo + OBR
100
DUET-1
DUET-2
DUET-1
DUET-2 82 80 80 73
DUET-1
DUET-2 70 68 66 65
DUET-1
DUET-2 62 61 59 60
HIV-1 RNA < 50 copies/mL at Week 24 (%) 47 44 40 34 24 20 9 7
n = 45 88 46 43 45
105 95 64
116 66 93 82 64 61 49 49 51 1 2
≥ 3
Number of Fully Active Agents in OBR (Assessed by PSS)
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

57. Enfuvirtide Reused or Not Used Enfuvirtide Used de Novo
Patients reusing or not using enfuvirtide achieved greater response with etravirine vs placebo
Findings unchanged when further stratified by number of baseline NNRTI RAMs and fold change in darunavir susceptibility
Among patients who used enfuvirtide de novo, no significant difference in response to etravirine compared with placebo
Outcome at Week 24, %
Enfuvirtide Reused or Not Used
Enfuvirtide Used de Novo
DUET-1
DUET-2
ETR
(n = 230)
PBO
(n = 229)
(n = 216)
(n = 215)
(n = 74)
(n = 79)
(n = 81)
HIV-1 RNA < 50 copies/ mL
55* 33
58* 34 60 56 73 68
HIV-1 RNA < 400 copies/mL
70* 44
71* 45 84 86 78
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

58. Other Outcomes
13 baseline mutations associated with diminished response to etravirine: V90I, L100I, V106I, Y181C/I/V, A98G, K101E/P, V179D/F, G190A/S
Response diminished by ~ 20% in presence of 1 or 2 mutations
≥ 3 mutations present at baseline in only 14% of study population
Rash most common AE; most mild/moderate (grade 3: 1%; grade 4: 0%)
AEs Through Week 24, %
DUET-1
DUET-2
Etravirine
(n = 304)
Placebo
(n = 308)
(n = 295)
(n = 296)
Rash 20 10 14 9
Central nervous system symptoms† 15 17
Nausea 12
Diarrhea 18
Psychiatric event‡ 16
Headache 13 11
Combined grade 3 or 4 AEs 21 28 27
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

59. Results at week 48 (CROI 2008) DUET-1 DUET-2
ETR
PBO
HIV-1 RNA < 50 copies/ mL 60 39 61 41 p
<.0001
HIV-1 RNA < 400 copies/mL 71 47 72 48
Mean CD4 Change cells/mm3
103 74 94
.0025
.0160
Duet 1:Haubrich R Abstract 790
Duet 2: Johnson M Abstract 791

60. Summary of Key Conclusions
Etravirine (TMC125) in combination with an OBR of darunavir/r and optimized NRTIs (with optional enfuvirtide) associated with significantly greater rates of HIV-1 RNA < 50 copies/mL at w48 in highly treatment– experienced (≥ 3 primary PI mutations) patients
Etravirine also superior to placebo in increasing CD4+ cell counts from baseline
Toxicity comparable in etravirine and placebo arms except rash
Madruga JV, et al. Lancet. 2007;370: Lazzarin A, et al. Lancet. 2007;370:39-48.

61. Entry Inhibitors
Let’s start with the entry inhibitors.

62. HIV-1 Entry Inhibitors CD4 binding Coreceptor binding Virus-cell
fusion
Enfuvirtide
gp41
CCR5 antagonists
Maraviroc
Vicriviroc
TNX-355
gp120
V3 loop
CD4
This slide displays the different steps involved in viral entry. This process begins with the binding of HIV gp120 to a CD4+ cell followed by a conformational change that allows the V3 loop of gp120 to interact with the chemokine coreceptors, either CCR5 or CXCR4. A conformational change then occurs that facilitates fusion. The first entry inhibitor to be approved, enfuvirtide, is a fusion inhibitor. In this presentation, I will talk about some of the drugs that have been developed to target both CD4 and CCR5.
Cell
membrane
CCR5/CXCR4
(R5/X4)
CXCR4 antagonists
Figure adapted from Doms R, et al. Genes Dev. 2000;14:

63. Coreceptor Usage of HIV-1 VariantsX4
X4/R5
Dual R5
CD4
CXCR4
CCR5
This schematic demonstrates the mode of action for HIV via certain chemokine coreceptors. On the far left is X4 virus, which primarily uses the CXCR4 coreceptor. CXCR4 chemokine coreceptors are primarily located on T-cell tumor lines as well as primary lymphocytes in peripheral blood.The R5 virus is located on the far right. These viruses are defined by the use of the CCR5 chemokine coreceptor, which is present on monocytes and macrophages as well as primary lymphocytes. In the middle of the schematic are those viruses designated as X4/R5 dual‑tropic viruses that are actually able to use either one of these chemokine coreceptors. The terms syncytium‑inducing (SI) and nonsyncytium-inducing (NSI) viruses were used before the identification of chemokines as secondary receptors. Syncytium‑inducing viruses were isolated from the blood of patients that demonstrated the ability to induce syncytia in T-cell tumor lines. We now know that this ability correlates highly with the use of the CXCR4 coreceptor. The NSI viruses, on the other hand, are able to infect cells but do not induce syncytia. These viruses are now known to use CCR5.
T-cell lines
Primary lymphocytes
Monocyte/macrophages

64. Association Between Emergence of SI Virus and CD4+ Cell Count
NSI virus predominates early in disease
Dual/mixed virus detected in approximately 50% of patients over time
CD4+ cell count decline accelerated following detection of SI in patients in whom NSI-only virus was previously detected
800
600
Mean (SE) CD4+ Cell Count (cells/mm3)
400
NSI
200
NSI, nonsyncytium inducing; SE, standard error; SI, syncytium inducing.
More than a decade ago, data were presented that suggested there may be a relationship between SI virus—or what we now know are X4 viruses—and the natural history of the disease. Koot and colleagues from a Dutch cohort followed a group of patients over time and performed assays to determine whether there was detectable SI virus. The investigators found that there was a gradual decline in CD4+ cell counts, but with the emergence of the SI viruses a more precipitous decline was seen than that which was seen among patients who did not have this viral population. This observation has been confirmed in many data sets during the last 10‑15 years, and there are now data that suggest very similar findings in people who develop X4 viruses.
What is still not clearly defined is whether there exists a cause-and-effect relationship. More specifically, is it the development of these X4 viruses that is actually causing CD4+ cell decline? Although the answer to this question remains unknown, it is important to bear in mind the implications of the association between X4 viruses and CD4+ cell decline when considering the use of CCR5 antagonists. There is the possibility that we might select for X4 viruses with the use of these agents, which could influence the natural history of the disease. This possibility must be taken into account as these drugs move forward in clinical trials and in drug development.
NSI → SI SI
-48
-36
-24
-12 12 24 36
Time (Months)
Koot M, et al. Ann Intern Med. 1993;118:

65. MOTIVATE : Maraviroc in Treatment-Experienced Patients With R5 Virus
Randomized, double-blind, placebo-controlled, parallel phase IIb/III studies
Primary endpoint : mean change in VL at Week 24
2:2:1 randomization;
stratified by ENF use and VL
Planned interim analysis Week 24
Week 48
Maraviroc 150 mg or 300 mg twice daily + OBR
R5 virus; ≥ 5000 copies/mL;
Triple-class resistant or triple-class experienced patients
MOTIVATE 1 (N = 601)
(Canada, US)
MOTIVATE 2 (N = 475)
(Europe, Australia, US)
Maraviroc 150 mg or 300 mg once daily + OBR
ENF, enfuvirtide; MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; OBR, optimized background regimen; R5, CCR5 tropic; TPV, tipranavir; X4, CXCR4 tropic; VL, viral load.
The next slide shows recently presented randomized control data from 2 large trials called MOTIVATE 1 and MOTIVATE 2. Combined, these studies evaluated more than 1000 patients, and they used an essentially identical study design. MOTIVATE 1 was performed in Canada and the United States, and MOTIVATE 2 was performed in Europe, Australia, and the United States. These trials randomized highly treatment–experienced patients 2 to 2 to 1 to receive an optimized background regimen with 1 of 2 doses of maraviroc or placebo. The 2 investigational arms were once-daily or twice-daily maraviroc in 2 different doses. Maraviroc is a substrate for cytochrome P450 and is boosted by ritonavir and the NNRTI delavirdine so the patients on those 2 drugs received 150 mg; all others received 300 mg. Patients on tipranavir/ritonavir also received the 300-mg dose because of a drug-drug interaction between tipranavir and maraviroc. The primary planned interim analysis at Week 24 has been presented, and the study is designed to continue until Week 48. The primary endpoint is mean change in HIV-1 RNA at Week 24.
For more information about this study, see the Capsule Summary at:
Placebo + OBR
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.

66. MOTIVATE 1 and 2 : VL < 400 copies/mL (ITT, NC = F)
Placebo + OBR (n = 209)
MVC QD + OBR (n = 414)
MVC BID + OBR (n = 426)
100
100
MOTIVATE 1
MOTIVATE 2 90 90 80 80
P < .0001*
P < .0001* 70 70
60.4%
61.3% 60 60
54.7%
55.5% 50 50
Patients (%)
Patients (%)
P < .0001*
P < .0001* 40 40
31.4% 30 30
23.1% 20 20
BID, twice daily; ITT, intent to treat; MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; MVC, maraviroc; NC = F, noncompleter equals failure; OBR, optimized background regimen; QD, once daily; VL, viral load.
The data on this slide are of individuals who achieved an HIV-1 RNA of < 400 copies/mL in an intent-to-treat, noncompleter-equals-failure, analysis. MOTIVATE 1 is on the left and MOTIVATE 2 on the right. The data are nearly superimposable, with better viral suppression among those who received once- or twice-daily maraviroc vs those who received placebo.
For more information about this study, see the Capsule Summary at: 10 10 2 4 6 8 10 12 14 16 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24
Time (Weeks)
Time (Weeks)
*P values vs placebo at Week 24.
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.

67. MOTIVATE 1 and 2: VL < 50 copies/mL (ITT, NC = F)
Placebo + OBR (n = 209)
MVC QD + OBR (n = 414)
MVC BID + OBR (n = 426)
100
100
MOTIVATE 1
MOTIVATE 2 90 90 80 80 70 70 60 60
P < .0001*
P < .0001* 50 50
Patients (%)
Patients (%)
48.5%
45.6% 40
42.2% 40
40.8%
P = .0006*
P = .0005* 30 30
24.6% 20 20
20.9%
BID, twice daily; ITT, intent to treat; MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; MVC, maraviroc; NC = F, noncompleter equals failure; OBR, optimized background regimen; QD, once daily; VL, viral load.
The next slide shows the same kind of data for those achieving HIV-1 RNA < 50 copies/mL. Again, the results are essentially superimposable with significantly better viral suppression among those who received the investigational drug with an optimized background regimen vs those who received an optimized background regimen with placebo.
For more information about this study, see the Capsule Summary at: 10 10 2 4 6 8 10 12 14 16 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24
Time (Weeks)
Time (Weeks)
*P values vs placebo at Week 24.
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.

68. Combined Analysis: MOTIVATE 1 and 2
MOTIVATE 1 and 2 : VL < 50 c/mL at Wk 24 by Number of Active Drugs in OBR
Placebo + OBR
MVC QD + OBR
MVC BID + OBR
100 90
Combined Analysis: MOTIVATE 1 and 2 80 70 61 58 60 55 52 53
Patients (%) 50 43 43 40 29 30 18 19 20 9
BID, twice daily; c/mL, copies/mL; ITT, intent to treat; MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; MVC, maraviroc; OBR, optimized background regimen; QD, once daily; VL, viral load; Wk, week.
This is a combined analysis of MOTIVATE 1 and 2 looking at the proportion of people achieving HIV-1 RNA < 50 copies/mL at Week 24 based on the number of active drugs in the optimized background regimen. These individuals were analyzed by genotypic and phenotypic assays at baseline. Their response based on the number of active drugs in the regimen0, 1, 2, or 3 or moreis shown. All of the groups that received 3 or more active drugs did well because of the high number of active drugs in their regimens. Among those who received 0, 1, or 2 active drugs, better virologic responses were seen in those patients who received maraviroc, either once or twice daily, vs placebo. In those participants who had no other active drugs within their regimen except maraviroc, there is the suggestion that those who received twice‑daily therapy did better than those who received once-daily therapy. However, both maraviroc treatment groups did far better than those who received placebo.
For more information about this study, see the Capsule Summary at: 10 3
n = 35 51 56 44
130
134 59 88
104 64
132
121
Number of Active Drugs in OBR 1 2
≥ 3
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.

69. MOTIVATE 1 and 2 : Mean Change in VL at Wk 24 by ENF Use in OBR
Placebo + OBR
MVC QD + OBR
MVC BID + OBR
MOTIVATE 1
MOTIVATE 2
No ENF
ENF*
No ENF
ENF*
n = 67
130
127 49 98
107
n = 50
113
112 41 67 73
-0.5
-0.5
-1.0
-1.0
-0.97
Change in VL, log10 copies/mL
-1.12
-1.20
-1.31
-1.5
-1.5
BID, twice daily; ENF, enfuvirtide; MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; MVC, maraviroc; OBR, optimized background regimen; QD, once daily; VL, viral load; Wk, week.
The next slide looks the mean change in HIV-1 RNA based on enfuvirtide use. Studies often use this drug with the idea that enfuvirtide may be active in many treatment-experienced patients. Looking at MOTIVATE 1 on the left and MOTIVATE 2 on the right, it is clear that there is better virologic suppression in the maraviroc groups (orange and blue) than the placebo groups (green). However, when comparing the group that used enfuvirtide with the group that did not, it is evident that there is not much of a difference.
The one thing that was not done in this particular trial is an analysis within the enfuvirtide‑treated group of those who were enfuvirtide experienced and possibly resistant to this drug vs those who were treatment naive. The previous slide suggested that the number of active drugs, which would include enfuvirtide in an enfuvirtide-naive patient, is associated with virologic response, and in general, this may true even though this analysis is not able formally to show that.
For more information about this study, see the Capsule Summary at:
-2.0
-1.97
-2.0
-2.08
-2.08
-2.02
-2.18
-2.17
-2.22
-2.5
-2.5
-2.45
*Includes all those who received ENF as part of OBR, whether ENF naive or experienced.
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.

70. MOTIVATE 1 and 2: Change in CD4+ Count at Time of Failure
Mean Change in CD4+ Cell Count in Patients With Treatment Failure
Placebo + OBR
MVC QD + OBR
MVC BID + OBR
All treatment failures
+14 (n = 97)
+49 (n = 68)
+71 (n = 77)
In Patients With Tropism Results at Baseline and Failure
R5  R5
+15 (n = 80)
+61 (n = 18)
+138 (n = 17)
R5  D/M or X4
+67 (n = 4)
+37 (n = 31)
+56 (n = 32)
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.
~ 8% of patients experienced shift in detected tropism between screening and baseline
Among patients with treatment failure, shift in detected tropism more common among maraviroc vs placebo recipients
Among maraviroc recipients with tropism results at time of failure, approximately 2/3 had dual/mixed or X4 virus detected
BID, twice daily; D/M, dual/mixed tropic; MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; MVC, maraviroc; OBR, optimized background regimen; R5, CCR5 tropic; QD, once daily; X4, CXCR4 tropic.
These data look at CD4+ cell count at the time of virologic failure. There are a few issues that need to be considered with any new drug and some unique issues associated with CCR5 antagonists in particular. For any drug, it is important to think about the possibility of true drug resistance. In the case of CCR5 antagonists, one must also consider the possibility of a shift in the viral population from R5 virus only at baseline to a population of dual/mixed-tropic virus.
As mentioned earlier, it is important to be sure that if dual/mixed-tropic viruses emerge, this event is not associated with adverse outcomes. If we look at all treatment failures, CD4+ cell counts were maintained in both maraviroc groups, even among those who experienced virologic failure. The tropism data are even more interesting if one looks at the group of people who had R5 virus only at baseline and at failure. Although the numbers are fairly small, CD4+ cell counts remained high. The most important groups to analyze, however, are those who experienced virologic failure with dual/mixed-tropic or X4 virus. CD4+ cell counts were maintained in this group as well suggesting that, at least in the short term, this shift in tropism was not associated with CD4+ cell count decline.
The other thing that is important to note is that if you look at the proportion of people in the different groups that failed with dual/mixed-tropic or X4 virus, only approximately 5% of those who received optimized background regimen with placebo are included in this group. However, approximately 65% or two thirds of those who were treated with CCR5 antagonists ultimately failed with dual/mixed-tropic or X4 virus. Whereas true resistance data were not formally described (there was some suggestion that there may be mutations within the V3 loop), it appears that those who are failing on a CCR5 antagonist are failing with viruses that use the other receptor. The good news is that this does not appear to be associated with a decline in CD4+ cell count.
The other interesting observation that came out of this study is that there was minimal variability in the results seen with the tropism assay. All of the patients who wanted to enroll in the study were screened and were included only if they did not have any evidence of X4 or dual/mixed virus (ie, their virus was R5 only). A tropism assay was performed again at study entry, several weeks after the initial screening test but before treatment with a CCR5 antagonist. Approximately 8% of the patents who had R5 virus only at screening were found to have dual/mixed-tropic virus at the time of entry. This has been seen in other studies as well. In fact, there is approximately 5% to 10% of patients who, over time, will have viral populations that switch from R5 virus only to dual/mixed or X4 virus only and then perhaps back and forth again. It is also believed that low levels of dual/mixed-tropic virus are below the sensitivity of the assay may result in a false negative at baseline.
For more information about this study, see the Capsule Summary at:

71. MOTIVATE 1 and 2 : Adverse Events and Resistance
Similar incidence of adverse events in maraviroc and placebo arms
Similar low incidence of hepatotoxicity in maraviroc and placebo arms
Lymphoma diagnosed in 3 patients in maraviroc arms and 2 patients in placebo arms
Resistance
Mutations seen in V3 loop among patients who failed on the maraviroc arms with R5 virus
No signature R5 mutations have been defined yet
MOTIVATE, Maraviroc plus Optimized Background Therapy in Viremic, ART-Experienced Patients; R5, CCR5 tropic.
Adverse events and the development of resistance were similar across the treatment groups. Although it is important to remember that these patients all received optimized background regimens, there did not appear to be any difference in adverse events between those who received once-daily or twice-daily maraviroc vs placebo. Similar low levels of hepatotoxicity were found in all of the groups.
There was an issue of malignancy raised in an earlier trial. It was suggested that there may have been more malignancies among those who received a CCR5 antagonist when compared with those who did not. That was analyzed very carefully in these 2 studies. In this case, the investigators were unable to see any relationship between malignancies and the use of CCR5 antagonists. In fact, the overall number of malignancies was small and similar results were found in both groups.
Regarding resistance, no real data have been formally presented yet. There are in vitro data, however, showing that resistance may develop. This needs to be studied carefully in these trials. Lastly, no signature R5 mutations have been defined at this time.
For more information about this study, see the Capsule Summary at:
Nelson M, et al. CROI Abstract 104aLB. Lalezari J, et al. CROI Abstract 104bLB.

72. Advantages and Disadvantages of CCR5 Antagonists
Novel antiretroviral class
Effective against NRTI-, NNRTI-, PI-, and ENF-resistant virus
Synergistic with ENF in vitro
Short-term tolerability data promising
Orally administered
Disadvantages
May not be effective for significant portion of patient population—those with X4 or D/M virus
Uncertain risk/implications of emerging D/M or X4 virus
Long-term safety and resistance not well defined
Cost and availability of tropism assay
Drug interactions: complexity of dosing with PI
D/M, dual/mixed tropic; ENF, enfuvirtide.
The advantages of CCR5 antagonists include that it is a novel antiretroviral class and it is effective against those with resistance to NRTIs, NNRTIs, enfuvirtide, and PIs. There is some suggestion in vitro that there may be synergy with enfuvirtide. The short‑term tolerability data appear to be quite promising, and it is an orally available drug.
Unfortunately, on the other hand, this class of drugs may not be effective for a significant proportion of patients, particularly those in the more advanced stage of disease who may have dual/mixed-tropic virus. There are uncertain risk implications of the emergence of dual/mixed-tropic or X4 virus, although the preliminary data suggest that this class appears to be safe. Long‑term safety and resistance has yet to be defined. There are also issues involving cost and the availability of a tropism assay, which will be a vital tool in determining for which patients these drugs are appropriate.

73. Clinical Trials of Integrase Inhibitors

74. Integrase Enzyme Viral enzyme essential to replication of both
HIV-1 and HIV-2
Integration
Follows reverse transcription, where DNA copy of HIV-1 RNA synthesized after infection
Essential step before viral DNA can be transcribed back into viral RNA
Incorporates or “integrates” viral DNA into host cell’s DNA

75. BENCHMRK 1 & 2 Phase III : RAL in Treatment-Experienced Pts
Week 48 current analysis
Week 156 planned follow-up
HIV-infected patients with triple-class resistance and HIV-1 RNA > 1000 copies/mL
BENCHMRK 1 (N = 350;
Europe, Asia/Pacific, Peru)
BENCHMRK 2 (N = 349;
North, South America)
Raltegravir 400 mg BID + OBR*
(BENCHMRK 1: n = 232;
BENCHMRK 2: n = 230)
Placebo + OBR*
(BENCHMRK 1: n = 118;
BENCHMRK 2: n = 119)
*Investigator-selected OBR based on baseline resistance data and history; inclusion of DRV and TPV permitted.
Cooper DA, et al. CROI Abstract 788.
Steigbigel R, et al. NEJM359:339, July 24, 2008

76. BENCHMRK 1: BL Characteristics
Raltegravir + OBR
(n = 234)
Placebo + OBR
(n = 118)
Mean HIV-1 RNA, log10 copies/mL
4.6
4.5
Median CD4+ cell count, cells/mm3
140
105
AIDS diagnosis, % 94 89
Median duration of ARV exposure, yrs 11 10
GSS = 0, % 30 29
GSS = 1, % 33 41
PSS = 0, % 19 18
PSS = 1, %
First-time use of DRV in OBR, % 27 25
First-time use of ENF in OBR, % 21 20
GSS, genotypic sensitivity score; HBV, hepatitis B virus; HCV, hepatitis C virus; OBR, optimized background regimen; PSS, phenotypic sensitivity score; SD, standard deviation.
Steigbigel RT, et al. N Engl J Med 76 76

77. BENCHMRK 2 : BL Characteristics
Raltegravir + OBR
(n = 230)
Placebo+ OBR
(n = 119)
Mean HIV-1 RNA, log10 copies/mL
4.7
Median CD4+ cell count, cells/mm3
102
132
AIDS diagnosis, % 91 92
Median duration of ARV exposure, yrs 10
GSS = 0, % 20 27
GSS = 1, % 44 40
PSS = 0, % 19
PSS = 1, % 34 28
First-time use of DRV in OBR, % 45 50
First-time use of ENF in OBR, %
GSS, genotypic sensitivity score; HBV, hepatitis B virus; HCV, hepatitis C virus; PSS, phenotypic sensitivity score; SD, standard deviation.
Steigbigel RT, et al. N Engl J Med 77 77

78. BENCHMRK 1 & 2: Efficacy by BL HIV-1 RNA and CD4+ Cell Count
Patient Group, %
HIV-1 RNA < 50 copies/mL at Week 48*
Raltegravir + OBR
(n = 443)
Placebo + OBR
(n = 228)
All patients 64 34
HIV-1 RNA at BL, copies/mL
> 100,000
48 (n = 156)
16 (n = 76)
≤ 100,000
73 (n = 287)
43 (n = 152)
CD4+ cell count at BL, cells/mm3
≤ 50
50 (n = 139)
20 (n = 75)
> 50 to ≤ 200
67 ( n = 167)
39 (n = 82)
> 200
76 (n = 136)
44 (n = 71)
*Virologic failures carried forward.
Steigbigel RT, et al. N Engl J Med

79. BENCHMRK 1 & 2: HIV-1 RNA < 50 c/mL at Week 48, Overall and by GSS*
*The genotypic sensitivity score is the total number of antiretroviral drugs used as part of the optimized background therapy to which a patient's HIV was fully susceptible, as determined with the use of genotypic resistance testing.
Raltegravir
Placebo
Subgroup n
Patients (%)
443 64
Total
228 34
GSS:
112 45 65 3
166 67
For patients with GSS=1, 4 ART agents represented at least 80% of the active agents in OBT: darunavir (52%, 52% in raltegravir and placebo groups, respectively), enfuvirtide (8%, 16%), tenofovir (12%, 6%), and tipranavir (11%, 11%).
Rates of virologic suppression also greater with RAL vs placebo when analyzed by baseline PSS
Similar results when assessing PSS by number of fully active drugs and by number of fully or partially active drugs 1 92 37
158 75
≥ 2 68 59 20 40 60 80
100
David A. Cooper et al. N Engl J Med 79 79 79

80. BENCHMRK 1 & 2: HIV-1 RNA < 50 c/mL at Week 48 by BL PSS*
* The phenotypic sensitivity score
PSS n
Patients (%)
Raltegravir + OBR
PSS = 0 (based on lower cutoff) 65 51 44 2
Placebo + OBR
PSS = 0 (based on upper cutoff) 33 52 12 8
PSS = 1 (based on lower cutoff)
PSS = 1 (based on lower cutoff)
PSS = 1 (based on lower cutoff)
PSS = 1 (based on lower cutoff)
137 61 69 29
PSS = 1 (based on upper cutoff) 71 48 54 13
PSS ≥ 2 (based on lower cutoff)
221 71
108 48
PSS ≥ 2 (based on upper cutoff)
PSS ≥ 2 (based on upper cutoff)
313 70
153 43 20 40 60 80
100
David A. Cooper et al. N Engl J Med

81. BENCHMRK 1 & 2: Virologic Failure, Resistance Through Week 48
Virologic failure generally associated with mutations at Q148 or N155, in combination with at least 1 other mutation
Virologic failure*: BENCHMRK 1 (n = 50); BENCHMRK 2 (n = 48)
Virologic failure defined as
< 1 log10 ↓ HIV-1 RNA from BL and > 400 c/mL at Week 16 or
> 1 log10 ↑ HIV-1 RNA above nadir or > 400 c/mL from nadir after response of < 400 c/mL (on 2 consecutive measurements ≥ 1 week apart)
*In patients for whom integrase genotypic data were available.
David A. Cooper et al. N Engl J Med

82. BENCHMRK 1 & 2: Adverse Events Through Week 48
Clinical adverse events
Raltegravir groups: 89%
Placebo groups: 87%
Considered treatment related in each group: 54%
Laboratory adverse events
Raltegravir groups: 23%
Placebo groups: 22%
Considered treatment related: 14% and 13%, respectively
Most common drug-related clinical adverse events in both treatment groups
Diarrhea, nausea, headache
Most common drug-related laboratory adverse events
Increased serum lipid, aminotransferase, creatinine levels
Steigbigel RT, et al. N Engl J Med

83. EACS Management of virologic failure
If Plasma HIV RNA confirmed > 500/1000 copies/ml, change regimen
as soon as possible: what to change will depend on the resistance
testing results:
No Resistance mutations found: re-check for adherence, perform TDM
Resistance mutations found: switch to a suppressive regimen based on drug history; multidisciplinary experts discussion advised
Goal of new regimen:
Plasma HIV RNA < 400 c/ml after 3 months
Plasma HIV RNA < 50 c/ml after 6 months

84. EACS In case of resistance mutations demonstrated
General recommendations :
Use 2 or preferably 3 active drugs in the new regimen (including active drugs from previously used classes)
Any regimen should use at least 1 drug from a class not used previously e.g. fusion, integrase or CCR inhibitor
Defer change if < 2 active drugs available, based on resistance data, except in patients with low CD4 count (<100/mm3) or with high risk of clinical deterioration for whom the goal is the preservation of immune function through partial reduction of Plasma HIV RNA (> 1 log reduction) by recycling.
If limited options, consider experimental and new mechanistic drugs, favouring clinical trials (but avoid functional monotherapy)
Treatment interruption is not recommended

85. EACS In case of resistance mutations demonstrated
Optimisation of new regimen :
Avoid NNRTI in NNRTI-experienced patients; Etravirine potentially active in selected NNRTI-resistance profiles
Consider continuation of 3TC or FTC even if documented resistance mutation (M184V/I)
Select other potentially active NRTI(s), on treatment history and full resistance (past and present) evaluation
Select 1 active ritonavir-boosted PI. If at all possible avoid double boosted PIs
Always check for drug-drug-interactions, and when necessary perform TDM of drugs of new regimen if available
If many options are available, criteria of preferred choice include : simplicity of the regimen, toxicity risks evaluation, drug-drug-interactions, future salvage therapy

86. Estimated Timeline for Availability of New Antiretrovirals
CXCR4 inhibitors
Entry inhibitors
(anti-gp120, CCR5)
GS-9137
Maturation inhibitors
Vicriviroc
Integrase inhibitors
Maraviroc
TNX-355
MK-0518
Bevirimat
2006
2007
2008
2009
2010
Etravirine
TMC278
PIs
Brecanavir
NNRTI
Apricitabine
NRTI

87. Conclusions (1)
Failure of therapy is multifactorial.The virologic failure is a progressive increase of HIV RNA that further leads to a decrease in CD4 cells.
Prevention of therapeutic failure starts as soon as first-line therapy detect defect in adherence due to any reasons.
Identification of therapeutic failure should mobilize treating HIV physicians.

88. Conclusions (2)
An optimal analysis of the  failing situation must be performed with ARV history and resistance assays results .
an « expert group » decision is the ideal situation.
Do not jeopardize any chance for success therapy by using a single new potent drug
Combining new drugs is the only solution to multi salvage situations

89. The near future of Antiretroviral therapy ?
…at least and unfortunately in developped countries only …

90. TRIO Study : Combining Raltegravir, Darunavir and Etravirine
24 Week Phase II, non-comparative, Multicenter Trial
All pts viremic on current regimen (n=103)
HIV RNA > 1000 /mL, any CD4 count
Documented multidrug-resistant virus
≥ 3 NRTI mutations
≥ 3 major PI mutations
Susceptible to DRV : ≤ 3 DRV mutations*
Previous virologic failure on NNRTIs
Susceptible to ETR : < 3 ETR NNRTI mutations
All initiate Raltegravir, Darunavir and Etravirine (naïve to all)
Additional ARVs allowed : NRTIs and ENF (based on clinical judgment)
Yazdanpannah, et al. 17th IAC; Mexico City, Aug 3-8, 2008; Abst. THAB0406. 90 90

91. TRIO Study : Baseline Characteristics
HIV RNA log10, copies/ml, median (IQR)
4.0
(3.6 – 4.6)
CD4 cells/mm3, median (IQR)
255
(132 – 350)
# mutations at screening, median (IQR)
Major PI 4
(3 – 5)
NRTIs 5
(4 – 6)
NNRTIs 1
(0 – 2)
% with 0 / 1 / 2 / 3 mutations
DRV
4% / 31% / 30% / 35%
ETR
34% / 31% / 31% / 3%
Additional ARVs in Optimized Background Regimen:
None 14%
NRTIs 83%
Enfuvirtide (most – 10/12 - ENF naive) 12%
Yazdanpannah, et al. 17th IAC; Mexico City, Aug 3-8, 2008; Abst. THAB0406. 91 91

92. TRIO Study : Primary Outcome at 24 Weeks (ITT, M=F)
90%
(95% CI:
85% to 96%)
% Pts with HIV RNA <50 cp/ml (95% CI)
Weeks 00 02 04 08 12 16 20 24 10 30 40 50 60 70 80 90
100
Yazdanpannah, et al. 17th IAC; Mexico City, Aug 3-8, 2008; Abst. ThAB0406. 92 92

93. How to choose a salvage therapy ?

94. Week 48: results in press: NEJM

95. Episodes of low-level viremia less likely associated with clinical event or change in therapy than episodes of high-level viremia
Low-level viremia
Without clinical event or therapy change : 79.6%
Change in therapy occurred : 13.9%
High-level viremia
Without clinical event : 41.7%
Change in therapy occurred : 52.3%
CD4+ cell counts increased during periods of virologic suppression but decreased during episodes of high level viremia
van Sighem A, et al. J Acquir Immune Defic Syndr. 2008;48:

96. Viremia : Low level 50-1000 cp/ml High level >1000 cp/ml
After achieving virologic suppression, many patients experience transient, measurable viremia while on antiretroviral therapy [1,2]
Transient viremia is associated with
Low-level viral replication
Activation of latently infected cells and subsequent viral production
Rise in target cell availability
1. Easterbrook PJ, et al. AIDS. 2002;16: Havlir DV, et al. JAMA. 2001;286:

97. Nucleosides Analogues (NRTI) Resistance
very common
mutations archived  indefinitely
cross-resistance between NRTIs ++
AZT-D4T
- high cross-resistance
> 3 TAMs including T215 F D4T efficacy
DDI
- antiviral efficacy if  3TAMS including T215 F
Abacavir
- antiviral efficacy if  4 mut. among 41,67,74,184,210,215

98. Nucleosides Analogues (NRTI) resistance
3TC induces M184V. high level of resistance for intrinsic antiviral activity
M184V reduces viral fitness
M184V prevents accumulation of other mutations(K65R)
Data suggest a benefit to maintain M184V in a regimen

99. Management of failure on first line therapy
Failure is defined as the detection of a viral load greater than 50 copies > 6 months after the initiation of a first treatment regimen
Interview the patient to evaluate adherence and compliance
Re-explanation of the objectives and modalities of the treatment and the potential risks of poor adherence
Exclusion of potential drug-drug or drug-food interactions
Re-Test viral load : If detectable :
- Resistance testing
- Therapeutic drug monitoring

100. Adding new drugs / drugs with a remaining sensitivity is a key issue in the succes of a salvage regimen

101. Management of virologic failure General measures (1)
If 50< Plasma HIV RNA < copies/ml
Check for adherence
Check Plasma HIV RNA 1 to 2 months later
Improve boosted PI's PK (if applicable)

102. Management of virologic failure General measures (2)
Perform resistance testing (if plasma HIV RNA levels > copies/ml) and obtain historical resistance testing for archived mutations
Review antiretroviral history
Identify treatment options, active, potentially active drugs/combinations

103. Failure in (multi)experienced patients
Patients who have been treated suboptimally in the past, have a long treatment history and have developed sequential resistance complex pattern of resistance which makes viral suppression difficult
Patients who have been unable to comply to and/or to tolerate their previous regimens less complex pattern of resistance but compliance and tolerability issues for the long term

104. Resistance testing : impact on treatment
Decision to change treatment regimens must take into account :
the remaining treatment options
the level of failure based on kinetics of viral load and CD4 (decreased viral fitness)
the past treatment history, including resistance patterns, tolerability and adherence issues
Co-infections and comorbidities

105. Treatment interruption in salvage therapy : a case for caution
May be associated with rapid decline of CD4
Need to optimalize OI prophylaxis
Optimal time for re-initiation of therapy is not established

106. HIV Replication Cycle and Drug Targets
Entry inhibitors
Reverse transcriptase inhibitors
Protease inhibitors
3′-processing inhibitors
Strand transfer inhibitors
Current antiretrovirals target (a) entry of HIV into the host cell and the viral enzymes (b) reverse transcriptase and (c) protease. Integrase, the third viral enzyme, catalyzes 2 steps in the viral replication cycle. First, integrase catalyses the processing of the 3′-ends of the viral DNA (3′-processing) (d); integrase then remains bound in a complex with the viral DNA ends in the preintegration complex (PIC). Second, after the PIC enters the host nucleus, integrase catalyses the insertion (strand-transfer) of the viral DNA ends into host chromosomes (e).
Pommier Y, et al. Nat Rev Drug Discov. 2005;4:
106
106

107. Percentage of Patients with Plasma HIV-1 RNA Levels of Less Than 400 or Less Than 50 Copies per Milliliter during the BENCHMRK Studies, According to Study Group
Figure 1. Percentage of Patients with Plasma HIV-1 RNA Levels of Less Than 400 or Less Than 50 Copies per Milliliter during the BENCHMRK Studies, According to Study Group. Percentages are shown for patients in the BENCHMRK-1 study (Panel A), patients in the BENCHMRK-2 study (Panel B), and patients in the combined studies (Panel C). The data, based on standard and ultrasensitive assays, are for the patients included in the analysis counting noncompletion as treatment failure. I bars represent 95% confidence intervals.
Steigbigel RT et al. N Engl J Med 2008;359:
107

108. BENCHMRK 1 : Patients With HIV-1 RNA < 50 c/mL at Week 48
Raltegravir + OBR (n = 232)
Placebo + OBR (n = 118)
P Value
HIV-1 RNA < 50 c/mL (NC = F), % 65 31
< .001
HIV-1 RNA < 400 c/mL (NC = F), % 74 36
Mean change in HIV-1 RNA vs BL, log10 c/mL
-1.7
-0.7
Mean change in CD4+ cell count vs BL, cells/mm3
120 49
BL, baseline; DRV, darunavir; ENF, enfuvirtide; NC = F, noncompleter = failure; OBR, optimized background regimen.
*P value derived from a logistic regression model adjusted for BL HIV-1 RNA level (log10), first ENF use in OBR, first DRV use in OBR, active PI in OBR.
Steigbigel RT, et al. N Engl J Med
108
108

109. BENCHMRK 2 : Patients With HIV-1 RNA < 50 c/mL at Week 48
Raltegravir + OBR (n = 232)
Placebo + OBR (n = 119)
P Value
HIV-1 RNA < 50 c/mL (NC = F), % 60 34
< .001
HIV-1 RNA < 400 c/mL NC = F), % 71 38
Mean change in HIV-1 RNA vs BL, log10 c/mL
-1.8
-0.9
Mean change in CD4+ cell count vs BL, cells/mm3 98 40
BL, baseline; DRV, darunavir; ENF, enfuvirtide; NC = F, noncompleter = failure; OBR, optimized background regimen.
*P value derived from a logistic regression model adjusted for BL HIV-1 RNA level (log10), first ENF use in OBR, first DRV use in OBR, active PI in OBR.
Steigbigel RT, et al. N Engl J Med
109
109

110. TRIO Study : Combining Raltegravir, Darunavir and Etravirine
24 Week Phase II, non-comparative, Multicenter Trial
All pts viremic on current regimen (n=103)
HIV RNA > 1000 /mL, any CD4 count
Documented multidrug-resistant virus
≥ 3 NRTI mutations (2006 IAS list)
≥ 3 major PI mutations (2006 IAS list)
Susceptible to DRV (using 1st Power algorithm): ≤ 3 DRV mutations*
Previous virologic failure on NNRTIs
Susceptible to ETR (using 1st Tibotec analysis of ETR RAMs): < 3 ETR NNRTI mutations
All initiate Raltegravir, Darunavir and Etravirine (naïve to all)
Additional ARVs allowed : NRTIs and ENF (based on clinical judgment)
* V11I, V32I, L33F, I47V, I50V, I54L/M, G73S, L76V, I84V and L89V
Yazdanpannah, et al. 17th IAC; Mexico City, Aug 3-8, 2008; Abst. THAB0406.
110
110

111. Option 6 : «Continuous» genotypic - driven salvage therapy
The goal is to achieve a selective pressure by using a genotypic-driven salvage therapy that is changed as soon as emergence of new resistant variants is documented Maggiolo et al. (Barcelona 2000) multi class experienced patients - VL every 2 months - Therapy changed if VL > , based on genotype - Over 24 months, VL contained below copies and CD4 increase of 84 cells (with max. 4 drugs)

112. POWER 3 : VL < 50 copies/mL at Week 24 by ITT-TLOVR
POWER 3: ongoing phase III open-label study, DRV/RTV 600/100 mg[1]
Safety analysis similar to POWER 1 and 2[2] 70
POWER 1 (n = 65) 60
POWER 2 (n = 66) 53 50
POWER 3 (n = 327) 40 40
Patients With VL < 50 copies/mL (%) 39 30 20 10
B/L 2 4 8 12 16 20 24
Weeks
1. Molina JM, et al. IAC Abstract TUPE Madruga V, et al. IAC Abstract TUPE0062.
112

113. J Acquir Immune Defic Syndr. 2007;46:24-31.
POWER 3 Study Confirms Safety and Efficacy of Darunavir/Ritonavir in Treatment-Experienced Patients
J Acquir Immune Defic Syndr. 2007;46:24-31.
POWER 1 and 2 studies demonstrated efficacy and safety of darunavir/ritonavir in treatment-experienced individuals
Parallel dose-ranging trials in treatment-experienced patients
Current POWER 3 study designed to provide additional data on efficacy and safety of darunavir/ritonavir 600/100 mg in treatment-experienced, HIV-infected patients

114. Description of Current Analysis
Data for RESIST 1 and 2 pooled in current analysis, given similar study design and patient demographics
Patients assessed at Weeks 2, 4, 8, 16, 24, 32, 40, and 48 for clinical and laboratory evaluations
Primary endpoints
Treatment response, defined as confirmed reduction in HIV-1 RNA ≥ 1 log10 copies/mL at Week 48
Time to treatment failure
Safety assessed via adverse-event monitoring
Adverse events and laboratory abnormalities graded according to Division of AIDS grading scale
Total cholesterol abnormalities graded according to Common Toxicity Criteria Scale
Intent-to-treat analyses using noncompletion-equals-failure and last-observation-carried-forward methods
114

115. Main Findings
Darunavir/ritonavir plus OBR associated with substantial virologic responses and immunologic improvement at Week 24
Outcome at Week 24
Patients
> 1 log10 copies/mL reduction in HIV-1 RNA, % 65
HIV-1 RNA < 400 copies/mL, % 57
HIV-1 RNA < 50 copies/mL, % 40
Mean change in HIV-1 RNA, log10 copies/mL
-1.65
Mean change in CD4+ cell count, cells/mm3
+80
Molina JM, et al. J Acquir Immune Defic Syndr. 2007;46:24-31.

116. BLQ Study : DRV/RTV + ENF in Triple-Class Experienced Patients
142 triple-class-experienced, DRV/RTV-naive and ENF-naive patients with HIV-1 RNA > 2000 copies/mL
Switched from failing regimen to DRV/RTV (600/100 mg BID), ENF (90 mg SC BID), and other investigator-selected antiretrovirals
Single arm, nonrandomized design
Overall, 60% achieved HIV-1 RNA < 50 copies/mL at Week 24
No difference in response according to baseline DRV susceptibility
Baseline Phenotypic Susceptibility to Darunavir
HIV-1 RNA < 50 copies/mL at Week 24, %
Categorical cutoffs
FC < 10 (n = 87)
64.4
FC (n = 19)
57.9
FC > 40 (n = 8)
62.5
BID, twice daily; BLQ, below the level of quantitation; DRV, darunavir; ENF, enfuvirtide; FC, fold change; RTV, ritonavir; VL, viral load
De Jesus E, et al. ICAAC Abstract 367.
116

117. Various Causes of failure
- Non-adherence : Side effects
Complex regimens
Lifestyle conflicts
- Toxicity
- Pharmacologic variations : drug-drug interaction pregnancy
- Infection with resistant HIV-1 strains
- Selection of resistant HIV-1 strains