HAART for the treatment experienced patient

HAART for the treatment experienced patient Prema Menezes PA-C Treatment Failure Definition and Reasons  Immunologic  No increase in CD4  Decrease in CD4  Virus  Resistant  Clinical  HIV related event at least 3 months after ART  Drug Therapy  Sub optimal  Sub therapeutic  Patient  Non adherence  Virologic  HIV RNA >400 (24wks), >50 (48 weeks) DHHS Treatment Guidelines Managing virologic failure: make a distinction between limited, intermediate, and extensive prior treatment exposure and resistance Goal of treatment: re-establish maximal virologic suppression  March 2004: Preservation of immune function and prevention of clinical progression  May 2006: Re-suppress HIV RNA levels maximally and prevent further selection of resistance mutations New Agents  Protease Inhibitors  Tipranavir  Darunavir  NNRTIs  Etravirine  Integrase Inhibitor  Raltegravir  CCR5 Inhibitor  Maraviroc Limited Treatment Failure Case 1: First Line Virologic Failure  31 yo woman s/p bilateral tubal ligation began     [fdc ZDV/3TC] + EFV 36 months ago Initial CD4 = 162; VL = 56,000 After starting HIV therapy, CD4 increased to 365 and VL fell to 340 and then to < 50 c/mL Difficulty taking medication over the past 2 months due to relapse of substance abuse Now returns with weight loss and thrush What is your next step? 1. 2. 3. 4. 5. Obtain HIV RNA, CD4, continue therapy and have her return within a month Obtain HIV RNA, CD4, stop therapy and have her return within a month Obtain a resistance test Choice 1 and 3 Choice 2 and 3 GS934: Resistance Development Through Week 96 TDF + FTC (n = 244) 14 4 10 10 2 ZDV/3TC (n = 243) 29 7 20 18 9* Patients genotyped, n Wild-type, n Any resistance, n EFV resistance mutations, n M184V/I, n TAMs, n K65R, n 0 0 1 0 Gallant J, et al IAC 2006. Abstract TUPE0064. Comparative trial of ZDV/3TC/EFV vs. TDF/FTC/EFV in treatment naïve patients KLEAN: Resistance FPV/r vs. LPV/r (with ABC/3TC FDC) FPV/r, n Confirmed virologic failures Unable to sequence 16 2 LPV/r, n 24 3 No treatment-emergent mutations Treatment-emergent mutations TAMs (M41M/L) 3TC-associated mutations (M184I, M184V, M184M/V) 9 0 3 14 1 4 NNRTI-associated mutations (V106V/A) PI-associated mutations: all minor (I54I/L, I93I/L, K20K/R, I62I/V) 0 3 2 2 Eron JJ Jr, Lancet 2006. Case 1: First Line Virologic Failure  CD4 now = 98 and VL = 29,000  Genotype reveals 184V and 103N mutations  ART is discontinued and PCP prophylaxis prescribed. The patient enters in-patient detox and 6 weeks later returns to restart HIV medications. She begins a series of adherence counseling sessions What ART to use?   Case 1: First Line Virologic Failure 1. Atazanavir + NRTIs 2. Atazanavir/ritonavir + NRTIs 3. Lopinavir/ritonavir + NRTIs 4. Fos-Amprenavir/ritonavir + NRTIs 5. [fdc ZDV/3TC/ABC] + tenofovir 6. Other If you chose NRTI + PI – which NRTIs 1. Abacavir + 3TC or FTC 2. Abacavir + tenofovir 3. Didanosine plus 3TC or FTC 4. Didanosine plus tenofovir 5. Tenofovir + 3TC or FTC 6. Tenofovir, ZDV + 3TC or FTC 7. Something else 3TC Alone vs Treatment Interruption Mean CD4+ Decrease (ITT) BL 4 8 Weeks 12 16 20 24 Mean VL Increase (ITT) 0 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 BL 4 8 12 Weeks 16 20 24 -50 -100 -150 -200 -250 3TC TI 3TC TI In contrast to treatment interruption arm, 3TC alone resulted in:  No recovery in RC  No increase in RT mutations  No reversion of PR mutations Castagna et al. AIDS 20:795 2006 Effects of M184V on other NRTI  Increases in susceptibility  Zidovudine  Stavudine  Tenofovir  Minimal Change or decrease in susceptibility  Abacavir  Didanosine − Both of these agents can select for M184V in vivo. First Line Virologic Failure  Good news  Lots of treatment options  Considerations  Keep 3TC/FTC in new regimen  Genotype  Resistance appears to be limited in first HAART failure ART Options for Extensively Treatment Experienced Patients Newer Agents Darunavir Tipranavir POWER 1 and 2 trials: design Randomization Investigator-selected CPI(s) + OBR • PI-, NRTI- and NNRTIexperienced • ≥1 PI mutation* • PI-based regimen • VL >1,000 copies/mL TMC114/r 400/100mg qd + OBR TMC114/r 800/100mg qd + OBR Investigatorselected CPI(s) + OBR (without NNRTIs) TMC114/r 400/100mg bid + OBR TMC114/r 600/100mg bid + OBR • The highest dose of TMC114/r (600/100mg bid) provided the greatest virologic response in the Week 24 analysis and is the selected dose for treatment-experienced patients The combined 48-week efficacy and safety interim analysis at this dose versus CPI(s) is reported here *Based on IAS-USA March 2003 at start of studies; updated to October 2004 list during studies VL = viral load, OBR = optimized background regimen (NRTIs ± enfuvirtide [ENF]) Lazzarin A, et al. XVI IAC 2006. Abstract TUAB0104 • POWER 1 and 2: BL characteristics TMC114/r 600/100mg bid n=131 Demographics Gender (% male) Mean age (years) Disease characteristics CDC class C (%) Mean duration of infection (years) Mean VL (log10 copies/mL; SD) Median CD4 count (cells/mm3; range) Previous ARV experience Mean duration (months; SD) NRTI NNRTI PI Fusion inhibitor (ENF) Genotypic and phenotypic information Median primary PI mutations* (n; range) Median PI resistance-associated mutations* (n; range) ≥1 sensitive† PI available (%) ≥1 sensitive† NRTI in OBR (%) *IAS-USA October 2004, CPI(s) n=124 88 44 43 12.9 4.49 (0.78) 163 (3–1,274) 89 44 36 12.0 4.61 (0.69) 153 (3–776) 100 (48) 28 (24) 65 (29) 14 (11) 3 (0–5) 8 (0–12) 36 72 †susceptibility 106 (45) 23 (15) 65 (28) 11 (9) 3 (0–5) 8 (1–13) 39 73 ARV = antiretroviral; SD = standard deviation was determined by Antivirogram® Lazzarin A, et al. XVI IAC 2006. Abstract TUAB0104 POWER 1 and 2: patients with VL <50 copies/mL over time to Week 48 (ITT-TLOVR) 100 80 Patients (%) TMC114/r 600/100mg bid CPI(s) 60 40 20 0 0 12 4 8 12 131 124 45%* (n=59/131) 46%* (n=50/110) 12% (n=15/124) 10% (n=12/120) 16 20 24 Weeks 131 124 28 32 130 124 36 40 120 121 44 48 110 120 TMC114/r n= 131 CPI(s) n= 124 *p<0.001 vs CPI(s) ITT = intent-to-treat, TLOVR = time to loss of virologic response Not all patients had reached Week 48 at the time of analysis; patients who had not reached Week 48 were censored at their last available visit Lazzarin A, et al. XVI IAC 2006. Abstract TUAB0104 POWER 1 and 2: virologic response defined as VL <50 copies/mL by BL subgroups at Week 48 (ITT-TLOVR) 70 60 21/36 TMC114/r 600/100mg bid CPI(s) Patients (%) 50 40 30 20 4/35 11/100 0/18 27/61 10 0 2/13 1/15 7/70 5/25 ENF used (naïve) ENF used (non-naïve) ENF not used 0 sensitive ARV in OBR ≥1 sensitive ARV in OBR ARV = antiretroviral drug; OBR = optimized background regimen; ENF = enfuvirtide. Use of ENF was not randomized in POWER 1 and 2. Lazzarin A, et al. XVI IAC 2006. Abstract TUAB0104 44/82 POWER 1 and 2: mean change from BL in CD4 count over time to Week 48 (LOCF) 140 Mean change in CD4 count (cells/mm3) 120 92* TMC114/r 600/100mg bid CPI(s) 102* 100 80 60 40 20 0 0 2 4 8 12 131 124 17 19 16 20 24 Weeks 131 124 28 32 130 124 36 40 120 121 44 48 110 120 TMC114/r n= 131 CPI(s) n= 124 Lazzarin A, et al. XVI IAC 2006. Abstract TUAB0104 *p<0.001 vs CPI(s) LOCF = last observation carried forward Baseline Resistance and Response to DRV/r  Baseline fold-change was strongest  Δ VL at Wk 24, (NC = F) predictor of Week 24 response 11 mutations associated with reduced response  V11I, V32I, L33F, I47V, I50V, I54L, I54M, G73S, L76V, I84V and L89V Baseline DRV FC FC  10 FC 11-40 (n = 65) (17% of pts) FC > 40 (n = 48) (13% of pts) 0 -0.5 -1.0 (n = 255) (70% of pts)  73% of pts had  2 of these mutations -0.78 -1.08 No. of BL mutations % with VL < 50, Wk 24 0 64 1 50 2 42 3 22 4 10 -1.5 -2.0 -2.04 -2.5 DeMeyer S, et al. Resistance Workshop 2006. Abstract 73. VL < 50 c/mL at Wk 24 50% 25% 13% POWER 1 and 2: observed incidence of the most common treatment-emergent AEs* during treatment, regardless of severity and 80 causality TMC114/r 600/100mg bid 70 60 Patients (%) CPI(s) 50 40 30 20 10 0 Diarrhea Nausea Headache Nasopharyngitis Fatigue Upper respiratory tract infection Herpes simplex Pyrexia *AEs = adverse events reported in ≥10% of patients and excluding ENF-associated injection site reaction (TMC114/r: 28%; CPI: 22%) Lazzarin A, et al. XVI IAC 2006. Abstract TUAB0104 TITAN DRV/RTV vs LPV/RTV in Tx-Experienced, LPV-Naive Patients  Intermediate treatment experience  Stratified by treatment site, NNRTI in OBR, HIV RNA > or < 50,000 c/mL  Darunavir/ritonavir met criteria for superiority to lopinavir/ritonavir in proportions with HIV-1 RNA < 400 copies/mL and < 50 copies/mL at week 48  Safety comparable between darunavir/ritonavir and lopinavir/ritonavir The Randomized Evaluation of Strategic Intervention in Multidrug Resistant Patients With Tipranavir (RESIST) Resist 1 and 2 – 96 weeks Proportion of patients with viral load <400 and <50 copies/ml at week 96 Resist 1 and 2 – 96 weeks Proportion of patients who took Enfuvirtide as a new drug and achieved virologic suppression. Safety - laboratory abnormalities TPV/r (%) Grade 3/4 AST Grade 3/4 ALT Grade 3/4 total cholesterol Grade 3/4 triglycerides 6.1 9.7 2.1 24.9 CPI/r (%) 1.8 4.2 0.4 13.0  Patients with Grade 3/4 elevations in liver enzymes or lipids were able to continue TPV/r therapy without developing clinical AEs Highly Treatment Experienced Patients Highly Treatment Experienced  48 yo white man HIV + since 1991 (CD4 180)  Received ZDV from 1994 to 1996, added 3TC    then indinavir VL on this regimen was initially BDL and CD4 rose to 300 cells over two years. VL rose to 5,600 then 10,100, CD4 was 390. In 2000 treatment was interrupted VL peaked at 386,000 and CD4 fell to 220 Case: Multi-drug Resistance  Over several years he was on a series of regimens:     EFV, SQV/RTV, d4T Abacavir, ddI, LPV/r TDF, FTC, LPV/r plus SQV HIV RNA rebound was eventually observed on each regimen  He feels well and has had no AIDS defining illness. His current CD4 is 310 and his VL repeated several times is between 10,000 and 15,000 on therapy Case: Multi-drug Resistance  Given long treatment history with only intermittent suppression of HIV RNA to BDL you order a genotype  This is the test available to you  RT mutations include:  41L, 74V, 118I, 184V, 215Y and 219Q  Protease mutations include:  10V, 20R, 33F, 46L,54V, 82A, 84V and 90M Case: Multi-drug Resistance What will you do at this point? 1. 2. 3. 4. 5. Change to > 1 NRTI, tipranavir/r and enfuvirtide Change to > 1 NRTI, darunavir/r and enfuvirtide Interrupt therapy and after 6 months begin > 1 NRTI, darunavir/r and enfuvirtide Continue current therapy Holding regimen Need to Know Likelihood of Success CD4 310, clinically stable Let’s Look at Genotype RT mutations include:  41L, 74V, 118I, 184V, 215Y and 219  Resistance predicted to 3TC, FTC, ddI, ABC, ZDV  Intermediate activity to TDF and d4T Protease mutations include:  10V, 20R, 33F, 46L,54L, 82A, 84V and 90M TPV Score and Treatment Response 10V, 20R, 33F, 46L,54L, 82A, 84V, and 90M TPV Score 0-1 Median FC: 0.7-0.9 Median Change in VL at Wk 24* (log10 copies/mL) 0 2-3 1.1-1.4 4-5 2.0-3.1 6-7 3.3-3.9 8-9 14.7-52.5 -0.08 (n = 4) -1 -0.89 (n = 242) -0.45 (n = 260) -0.49 (n = 68) -2 -2.10 (n = 144) -3 TPV Score Mutations 10V, 13V, 20M/R/V, 33F, 35G, 36I, 43T, 46L, 47V, 54A/M/V, 58E, 69K, 74P, 82L/T, 83D, 84V *24-week data from patients in RESIST-1 and -2 given TPV/r Valdez H, et al. Resistance Workshop 2005. Abstract 27. Baseline Resistance and Response to DRV/r 10V, 20R, 33F, 46L,54L, 82A, 84V, and 90M  Baseline fold-change was strongest  predictor of Week 24 response 11 mutations associated with reduced response  V11I, V32I, L33F, I47V, I50V, I54L, I54M, G73S, L76V, I84V and L89V  73% of pts had  2 of these mutations 0 Δ VL at Wk 24, (NC = F) -0.5 -1.0 Baseline DRV FC FC  10 (n = 255) (70% of pts) FC 11-40 (n = 65) (17% of pts) FC > 40 (n = 48) (13% of pts) -0.78 -1.08 No. of BL mutations % with VL < 50, Wk 24 0 64 1 50 2 42 3 22 4 10 -1.5 -2.0 -2.04 -2.5 DeMeyer S, et al. Resistance Workshop 2006. Abstract 73. VL < 50 c/mL at Wk 24 50% 25% 13% Multi-Drug Resistance  This case illustrates several points to consider when making the difficult decision of when to switch  This patient is in no clinical danger.  His CD4 cell count is above 300  His viral load is over 1 log10 lower than a peak off therapy.  He has had no AIDS defining illness and feels well  On the other hand his current therapy is not optimal with sustained viral replication  Continued current therapy may lead to CD4 cell decline and further accumulation of resistance Case: Multi-drug Resistance  There is no clinical urgency; what are treatment options?  He is naïve to T-20 therefore likely very active initially − T-20 low barrier to resistance; incomplete suppression leads to rapid resistance evolution  He has no NNRTI mutations − By history he had previous viral rebound while on EFV − NNRTI may add modest activity, − NRTI will likely have modest ADDITIONAL activity  There are substantial PI mutations limiting PI options − He has 5 TPV-associated mutations decreasing the likelihood of a sustained response − He has 3 DRV mutations which will impact activity Should He Wait for New Agents? How quickly will new mutations evolve? Risk of Delayed Switch on Stable HAART  SCOPE cohort of ART-experienced subjects (n = 106)[1] Proportion Without New Mutation     Stable HAART for  120 days HIV-1 RNA > 1000 c/mL  1 resistance mutation Resistance testing every 4 mos until HAART modification 1.00 0.75 0.50 0.25 0 0 Proportion Without Loss of 1 Drug 1.00 0.75 0.50 0.25 0 0 4 1 new major PI mutation 1 new NRTI mutation* Any new mutation  Emergence of new mutns at 1 yr  Any: 44% (95% CI: 33%-56%)  NAMs: 23% (95% CI: 15%-34%)  PI: 18% (95% CI: 9%-34%) *PI-treated subjects (n = 71) 4 8 12 16 20 24  Those with persistent viremia on HAART run risk of limiting future treatment options Time to loss of 1 drug equivalent Number of available antiretrovirals from the following: ZDV, 3TC, ddI, ABC,TDF, EFV, IDV, NFV, SQV, RTV, APV, LPV  Other studies show similar results[24] 1. Hatano H, et al. CROI 2006. Abstract 615. 2. Lafeuillade A, et al. IAC 2004. Abstract WeOrB1293. 3. Margot NA, et al. JAIDS. 2003;33:15-21. 4. Napravnik S, et al. JAIDS. 2005;40:34-40. 8 12 16 Time (Mos) 20 24 Case 1  Patient now has CD4 cell count of 120  VL is 15,000  Repeat Genotype – virtual phenotype show no new mutations     DRV fold change = 45 (cut-offs 3.4 – 96.8) TPV fold change = 4.0 (cut-offs 1.2 – 5.4) Intermediate susceptibility to TDF resistant to all NRTI No NNRTI mutations (failed EFV in the past) Case 1  If you were to start the patient on new drugs which would you use? A. Maraviroc B. Raltegravir C. Etravirine D. Raltegravir/Etravirine E. Maraviroc/Etravirine F. Maraviroc/Raltegravir DUET-1 and -2: Etravirine + DRV/RTVContaining OBR Phase III Trials Week 24† Etravirine 200 mg BID + DRV/RTV-containing OBR* (n = 599) Placebo + DRV/RTV-containing OBR* (n = 604) Week 48 HIV-infected patients with VF on current HAART regimen, history of ≥ 1 NNRTI resistance mutations, ≥ 3 primary PI mutations, HIV-1 RNA > 5000 copies/mL (DUET-1: N = 612; DUET-2: N = 591) *Investigator-selected OBR to consist of DRV/RTV (600/100 mg/mL) + ≥ 2 NRTIs ± enfuvirtide. †Planned Week 24 analysis: primary endpoint HIV-1 RNA < 50 copies/mL (TLOVR). Madruga JV, et al. Lancet. 2007;370:29-38. Lazzarin A, et al. Lancet. 2007;370:39-48. Mills A, et al. IAS 2007. Abstract WESS204.1. Katlama C, et al. IAS 2007. Abstract WESS204.2. Cahn P, et al. ICAAC 2007. Abstract H-717. DUET-1 and -2: Pooled Virologic and Immunologic Responses Outcome at Week 24 HIV-1 RNA < 50 copies/mL, % Etravirine (n = 599) 59 -2.4 Placebo (n = 604) 41 -1.7 P Value < .0001 < .0001 Mean change in HIV-1 RNA from baseline, log10 copies/mL Mean change in CD4+ cell count from baseline, cells/mm3 +86 +67 < .0001  In patients using enfuvirtide for the first time (n = 201), the difference between treatment arms (67% and 62% for etravirine vs placebo, respectively) was not significant (P = .427) Cahn P, et al. ICAAC 2007. Abstract H-717. DUET-1 and -2: Response Based on Active Agents in OBR 100 Etravirine + OBR Placebo + OBR 74 67 60 HIV-1 RNA < 50 copies/mL at Week 24 (%) 80 60 45 40 30 20 n= 88 8 91 199 211 258 257 0 0 1 ≥2 No. of Fully Active Agents in OBR (assessed by PSS) Cahn P, et al. ICAAC 2007. Abstract H-717. DUET-1 and -2: BL ETR Mutations and Virologic Response at Week 24  13 mutations associated with ETR resistance L100I V106I Y181C/I/V K101E/P V179D/F G190A/S Patients With HIV-1 RNA < 50 copies/mL (%) V90I A98G 100 90 80 70 60 50 40 30 20 10 0 0 1 2 3  Presence of ≥ 3 ETR mutations associated with response similar to placebo + OBR – 70% of patients had 0 or 1 ETR resistance mutations at BL – 14% of patients had ≥ 3 ETR resistance mutations at BL – Response diminished by ~ 20% in presence of 1 or 2 mutations Katlama C, et al. IAS 2007. Abstract WESS204.2. 4 5 No. of BL ETR Mutations Patients (%) 40 30 16 8 5 1 DUET-1 and -2: Study Conclusions  Proportion of patients achieving HIV-1 RNA < 50 copies/mL significantly greater in etravirine arms  CD4+ cell count increase from baseline significantly greater in ETR arm in DUET-1  DUET-1: 89 vs 64 cells/mm3 (P = .0002)  DUET-2: 78 vs 66 cells/mm3 (P = .3692)  Presence of ≥ 3 mutations from list of 13 ETR RT mutations associated with decreased response  K103N not associated with ETR resistance  Incidence of adverse events and laboratory abnormalities similar to placebo arm BENCHMRK-1 and -2: Raltegravir in Treatment-Experienced Pts  Randomized, double-blind, placebo-controlled, parallel phase III studies Primary endpoints: Week 16 HIV infected; triple-class resistant; VL > 1000 copies/mL BENCHMRK-1 (N = 350) (Europe, Asia/Pacific, Peru) BENCHMRK-2 (N = 349) (North, South America) Planned duration: Week 48 Raltegravir 400 mg twice daily + OBR BENCHMRK-1 (n = 232) BENCHMRK-2 (n = 230) Placebo + OBR BENCHMRK-1 (n = 118) BENCHMRK-2 (n = 119) Cooper D, et al. CROI 2007. Abstract 105aLB. Steigbigel R, et al. CROI 2007. Abstract 105bLB. Benchmark 1 and 2 Patient Disposition at study entry BENCHMRK 1 and 2: HIV-1 RNA < 50 copies/mL (ITT, NC = F) Raltegravir + OBR 100 Percent of Patients with HIV RNA <50 Copies/mL BENCHMRK-1 100 Placebo + OBR BENCHMRK-2 80 80 61% 60 40 20 0 0 2 4 8 12 Weeks 16 24 P < .001 at Week 16 60 62% P < .001 at Week 16 40 20 0 0 2 4 8 12 Weeks 16 24 33% 36% Cooper D, et al. CROI 2007. Abstract 105aLB. Steigbigel R, et al. CROI 2007. Abstract 105bLB. BENCHMRK 1 and 2: HIV-1 RNA < 400 c/mL at Wk 16 by Agents in OBR Raltegravir + OBR Placebo + OBR % of Patients 43 79 n Overall Efficacy Data Efficacy by Agents in OBR Enfuvirtide Darunavir 447 230 + + – – + : First use in OBR – : No use in OBR + – + – 44 23 42 24 80 47 191 90 0 20 29 40 60 55 74 80 63 98 87 90 90 100 Cooper D, et al. CROI 2007. Abstract 105aLB. Steigbigel R, et al. CROI 2007. Abstract 105bLB. BENCHMRK 1 and 2: HIV-1 RNA < 400 c/mL at Wk 16 by PSS/GSS of OBR Raltegravir + OBR Placebo + OBR % of Patients 43 5 41 87 57 57 10 43 85 89 61 76 79 n Overall Efficacy Data 447 230 62 44 141 68 222 110 111 63 170 93 159 70 0 (PSS) 0 1 2 or more (GSS) 0 1 2 or more 71 20 40 60 80 100 Cooper D, et al. CROI 2007. Abstract 105aLB. Steigbigel R, et al. CROI 2007. Abstract 105bLB. Case Summary CD4 now 200 cell/mm3, VL is 65,000; patient is symptomatic RT mutations include:  41L, 74V, 118I, 184V, 215Y and 219  Resistance to 3TC, FTC, ddI, ABC, ZDV  Intermediate activity to TDF and d4T Protease mutations include:  10V, 20R, 33F, 46L,54L, 82A, 73S, 84V and 90M  Predicted phenotype TPV = 4.0 (1.2, 5.4) DRV 64 (3.4, 96.9)  above upper cut-off for all other PI. Previous NNRTI experience Raltegravir, etravirine and maraviroc are now available.  Entry phenotype is dual-mixed Summary Treatment experienced patients  Re-suppress HIV RNA levels maximally and prevent further selection of resistance mutations  Need to know likelihood of success  Must have at least two active drugs World AIDS Day 2006 What Is the “Resistance Penalty” of Continued Nonsuppressive Therapy?  Studies of resistance accumulation in states of “incomplete viral suppression”  68% with new mutations after median of 22 mos[1]  33% with new TAMs, 2% K65R during 96 wks of FU[2]  60% with new mutations after median of 9.3 mos, but no shift on virtual phenotype[3]  Studies lack results of subsequent switches  No fully powered randomized studies of early vs deferred switching 1. Lafeuillade A, et al. IAC 2004. Abstract WeOrB1293. 2. Margot NA, et al. J Acquir Immune Defic Syndr. 2003;33:15-21. 3. Napravnik S, et al. J Acquir Immune Defic Syndr. 2005;40:34-40.