Ananworanich, Jintanat MD*; Siangphoe, Umaporn MS*; Hill, Andrew PhD†; Cardiello, Peter MD*; Apateerapong, Wichitra MD*; Hirschel, Bernard MD‡; Mahanontharit, Apicha BS*; Ubolyam, Sasiwimol MS*; Cooper, David MD§; Phanuphak, Praphan MD, PhD*∥; Ruxrungtham, Kiat MD*∥
Whereas the introduction of highly active antiretroviral therapy (HAART) has significantly improved AIDS-free survival, lifelong treatment may lead to acute and progressive drug toxicities,1 and treatment costs may limit the long-term affordability of HAART.2,3 By decreasing the time patients receive medications, intermittent HAART could reduce the toxicity and cost of long-term treatment.4 Stopping HAART increases the risk of clinical disease progression if CD4 counts fall below 200 cells/μL,5,6 however, and there is the potential for the emergence of drug resistance from residual drug levels persisting after drug discontinuation.7,8 Treatment interruptions are likely to be more beneficial for patients with little or no prior HIV drug resistance than in the salvage setting, where clinical trials of treatment interruptions have led to mixed results.9,10
Intermittent treatment of patients with virologic suppression on HAART has been studied in prospective trials with 2 approaches. With CD4-guided treatment, HAART is stopped for those with high CD4 counts and restarted if CD4 counts fall below preset limits (eg, 350 cells/μL); these limits are set to ensure a continued low risk of HIV disease progression during CD4-guided treatment.6,11 With short-cycle treatment, HAART is stopped and restarted at regular intervals. A pilot study of week-on-week-off HAART with stavudine (d4T)/lamivudine (3TC) plus indinavir/ritonavir in 10 patients led to sustained viral suppression for 32 to 68 weeks.12 The Staccato study showed high rates of virologic failure using this same approach, however, leading to the premature discontinuation of the week-on-week-off arm.13 Regular cycles of treatment with longer treatment interruption intervals have led to viral rebound during the treatment discontinuation phase.14,15
The current pilot study, the HIV Netherlands Australia Thailand Research Collaboration (HIV-NAT) 001.4, was conducted to assess CD4-guided and week-on-week-off strategies in comparison to continuous HAART for patients who had been exposed to dual nucleoside reverse transcriptase inhibitors (NRTIs) before HAART. This report shows the results at the end of the study (week 108), which includes a HAART retreatment period between weeks 96 and 108 in all arms. To our knowledge, this is the first study to compare virologic suppression rates directly in patients treated with continuous HAART and those retreated with HAART after CD4-guided therapy.
This 3-arm, open-label, prospective 108-week trial was conducted at a single center in Thailand (HIV-NAT, Bangkok). The trial was approved by the Chulalongkorn University Institutional Review Board, and all patients signed a written informed consent form at screening.
The trial enrolled 74 HIV-1-infected adults with CD4 counts above 350 cells/μL and HIV RNA levels <50 copies/mL for at least 6 months before screening. The patients were randomized at a 1:1:1 ratio to receive continuous (n = 25), CD4-guided (n = 23), and week-on-week-off (n = 26) HAART. The primary end points were proportions of patients with CD4 counts >350 cells/μL and with HIV RNA levels <400 copies/mL at week 108.
All patients had received 66 weeks of dual NRTI treatment during which only 20% had HIV RNA levels below 400 copies/mL,16 followed by 160 weeks of HAART with saquinavir (SQV) in combination with 2 NRTIs (Fig. 1).17 The HAART regimen used for all patients during this structured treatment interruption (STI) study was 2 nucleoside analogues (NRTIs) plus 1600 mg/100 mg of SQV/ritonavir administered once daily. The study was conducted in 2 consecutive periods: weeks 0 through 48 and weeks 48 through 108. The results of the initial 48-week period have been reported.18 During the first 48 weeks, half of the patients received azidothymidine (AZT)/3TC and half received d4T/didanosine (ddI), whereas after week 48, all patients received d4T/ddI and the SQV-soft gel capsule (SGC) was switched to an SQV-hard gel capsule (HGC).
Patients attended study visits at screening; baseline; weeks 4, 8, and 12; and every 8 weeks for the first 48 weeks and every 12 weeks thereafter to week 108. Patients were assessed for CD4 count, HIV RNA level (Roche Amplicor Ultrasensitive assay, Palo Alto, CA), fasting lipids, hematology, clinical chemistry, adverse events, and HIV disease progression. Clinical and laboratory adverse events were graded by severity.
For the CD4-guided arm, all patients started the trial by discontinuing HAART. They restarted at least 12 weeks of HAART if their CD4 count was below 350 cells/μL or dropped more than 30% from baseline according to patient preference. They stopped again if their CD4 count rose above 350 cells/μL or increased at least 70% from baseline. Virologic failure was classified as HIV RNA levels above 400 copies/mL on 2 consecutive visits or above 1000 copies/mL at a single visit in any patient in the continuous and week-on-week-off arms and in patients in the CD4-guided arm who had received at least 12 weeks of HAART. In the CD4-guided arm, all patients received continuous HAART from weeks 96 to 108. Those who did not achieve an HIV RNA level <50 copies/mL were treated for an additional 12 weeks (week 120).
An intention-to-treat (ITT) approach was used. In addition, an on-treatment (OT) approach was used to compare HIV RNA outcome at weeks 108 and 120 between the CD4-guided and continuous arms. Data were analyzed using SPSS for Windows, version 9.0, software (SPSS, Chicago, IL). Tests were matched to data type and distribution. Statistical analyses included the Kruskal-Wallis test, Mann-Whitney U test, Wilcoxon signed-rank test, χ2 test, Fisher exact test, McNemar test, and general descriptive statistics. Probability values less than 0.05 were considered statistically significant. Given the small sample size of this pilot study, there is no statistical power to detect differences in HIV RNA response or CD4 counts between treatment arms.
Seventy-four patients were enrolled into the trial. Baseline characteristics were well balanced across the treatment arms (Table 1). Overall, there were 36 men and 38 women, with a median age of 35 years and median body weight of 55 kg. Most patients were in Centers for Disease Control and Prevention (CDC) stage A (51%) or B (43%) at baseline. Patients had an overall duration of prior treatment of almost 5 years, with dual NRTIs and then HAART, including SQV. Before starting antiretrovirals (ARVs), the overall median CD4 count had been 358 cells/μL, which had improved to 644 cells/μL by the baseline visit for this trial. CD4 counts at baseline were above 500 cells/μL for 81% of patients overall; HIV RNA levels were below 400 copies/mL for all 74 patients at baseline, and below 50 copies/mL for 71 (96%) of 74 patients. The 3 patients who had HIV RNA levels above 50 copies/mL had HIV RNA levels below 50 copies/mL at the time of randomization.
All 25 patients in the continuous treatment arm completed the 108-week trial. In the CD4-guided arm, all 23 completed the trial but 1 had to stop ARVs because of severe mitochondrial toxicity (edema, weight loss, muscle weakness, and high lactate). In the week-on-week-off arm, 23 of 26 patients completed the trial and 1 interrupted ARV because of severe hepatotoxicity. The discontinuations were attributed to patient request (n = 1), lost to follow-up (n = 1), and death as the result of a brain tumor (n = 1).
Patients in the continuous arm maintained CD4 counts at baseline levels throughout the 108 weeks of the trial (Fig. 2), with HIV RNA levels remaining below 400 copies/mL for all patients and below 50 copies/mL for 24 (96%) of 25 patients by week 108 (Fig. 3).
The week-on-week-off arm was stopped after a median 15 months of follow-up after excessive virologic failure rates (46%) were observed. By week 24, 11 (42%) of 26 of these patients had a rebound in HIV RNA to above 50 copies/mL, but RNA levels were still below 400 copies/mL in 81% of patients overall (see Fig. 3). Interestingly, the CD4 counts remained close to baseline levels in this treatment arm, despite the rebound in viremia (see Fig. 2A). All 26 patients in this arm converted to continuous HAART with the same regimen (2 NRTIs with boosted SQV) by week 72 of the trial. At week 108, 21 (81%) of 26 patients had HIV RNA levels below 50 copies/mL. Of the 22 patients who were on treatment, 21 (95%) had HIV RNA levels <50 copies/mL.
For the CD4-guided arm, stopping HAART at baseline led to a wide range of outcomes during the 96-week randomized phase of the trial. Overall, patients in this arm were taking HAART for a median of 44 weeks (46% of the 96-week trial duration).
The median HIV RNA level had risen from <50 copies/mL at baseline to 63,095 copies/mL at week 4, with the CD4 count falling from 766 cells/μL at baseline to 598 cells/μL during the same interval.
Four patients (17%) stopped HAART for the entire 96-week duration of the randomized trial without restarting. For the 19 patients who restarted HAART, 5 restarted within 12 weeks of baseline, 4 restarted from 12 to 24 weeks after baseline, and 10 restarted HAART from 24 to 96 weeks after baseline. Eight of the 19 patients who restarted HAART remained on treatment to week 96 because they chose to, whereas 11 subsequently stopped HAART again before week 96. The median CD4 count at the time of first restarting HAART was 442 cells/μL (interquartile [IQ] range: 340-498 cells/μL); at this time, the CD4 count had fallen by a median 41% from its baseline level. During the 96-week randomized phase, more than 70% of patients maintained CD4 counts above the threshold of 350 cells/μL on or off HAART.
After the 96-week randomized phase, all patients in the CD4-guided arm restarted HAART for 12 weeks. The proportion of patients with HIV RNA levels less than 400 copies/mL rose from 30% at week 96 to 91% (21 of 23 patients) at week 108, with the proportion less than 50 copies/mL rising from 26% to 57% (13 of 23 patients) during the same interval (Fig. 4A). The 10 patients whose HIV RNA level was not below 50 copies/mL at week 108 had a median HIV RNA level of 4.7 log10 at week 96 before retreatment, whereas the 13 with undetectable viremia at week 108 had a lower median HIV RNA of 2.8 log at week 96. The patients who did not have an HIV RNA level <50 copies/mL were treated with continuous HAART for another 12 weeks (week 120). Five patients were not on HAART at week 120 for various reasons, however, including toxicity (n = 1), stopped by themselves (n = 2), physician agreed for patient to stop because his HIV RNA level was 52 copies/mL and he wanted to stop (n = 1), and lost to follow-up (n = 1). Of the remaining 5 patients on HAART, 4 had HIV RNA levels <50 copies/mL. Therefore, at week 120, of the 18 evaluable patients, 17 (94%) of 18 had HIV RNA levels below 50 copies/mL (see Fig. 4B). The only patient who had virologic failure (HIV RNA level above 400 copies/mL twice) was a patient who did not comply with the CD4-guided start and stop procedures. This patient had no mutations on genotyping. At week 108, CD4 counts rose from a median of 439 cells/μL at week 96 to 489 cells/μL, with all patients above 350 cells/μL. The final median CD4 value was still 278 cells below the baseline level, however (see Fig. 2A).
There were no differences in rates of clinical disease progression between the arms at week 108, with 1 CDC stage B event in the continuous treatment (oral candidiasis) and CD4-guided (oral hairy leukoplakia) arms and 3 CDC stage B events (2 oral hairy leukoplakia and 1 pruritic papular eruption) and 1 CDC stage C event (esophageal candidiasis) in the week-on-week-off arm.
The incidence of grade I through IV adverse events at least possibly related to study medication was 64% for continuous treatment, 60% for CD4-guided treatment, and 89% for week-on-week-off treatment. The overall incidence of lipodystrophy measured subjectively using a questionnaire was 53%, with no significant differences between treatment groups. Liver enzyme levels showed no significant changes within or between groups during the trial.
Surprisingly, there was a 25% rise in median high-density lipoprotein (HDL) cholesterol and a 10% reduction in low-density lipoprotein (LDL) cholesterol in all 3 treatment arms, which improved the overall ratio of total cholesterol to HDL cholesterol (P < 0.001). Triglyceride levels showed small rises in the 3 treatment groups of borderline statistical significance. Glucose levels rose by 7% overall, with statistically significant rises in all 3 treatment groups but no overall difference between the treatment groups. There were no differences in lipodystrophy and quality of life between the treatment arms.
This was a randomized, 108-week, pilot STI study of patients with full viral suppression and high CD4 counts before ARVs and before STI who were randomized to continuous, CD4-guided, or week-on-week-off treatment. The week-on-week-off arm had a 46% virologic failure rate, and all patients resumed continuous HAART with the same regimen by week 72. Patients in the continuous and week-on-week-off arms had CD4 counts >350 cells/μL, but the patients in the CD4-guided arm did not recover their CD4 count up to what it was before STI. Between weeks 96 and 108, all patients in the CD4-guided arm were retreated with HAART to assess whether viral suppression after STI was possible. At week 108, the HIV RNA suppression rate below 400 copies/mL was similar between the 3 arms. Fewer patients in the CD4-guided arm had HIV RNA levels <50 copies/mL compared with the other 2 arms, however. We found that the patients in the CD-guided arm who were able to achieve viral suppression after 12 weeks of HAART retreatment had a lower HIV RNA level before retreatment. After knowing these results, we tried to assess whether the unsuppressed patients would also be able to achieve HIV RNA levels <50 copies/mL if they were treated for an additional 12 weeks after the study ended. At the follow-up 12 weeks later (week 120), however, half of the patients were already off ARVs and were not able to be evaluated. Of the 18 evaluable patients, 94% had HIV RNA levels less than 50 copies/mL. Therefore, the rate of HIV RNA suppression for patients who received adequate retreatment time was similar to that of patients receiving continuous HAART. There was no progression to CDC stage C during the trial.
Results from this small randomized trial are consistent with those of cohort studies of CD4-guided treatment. For those with high CD4 counts at discontinuation of HAART and no prior failure on ARVs, a range of slopes of CD4 decline have been seen in other studies, with high interpatient variability, allowing some patients to remain off treatment for long periods.19,20 For those with CD4 counts remaining above 200 cells/μL during treatment interruptions, the risk of clinical disease progression was low in this trial and previous studies.6,11,14 Considering that 80% of our patients had virologic failure on dual NRTI before they were treated with HAART, the potential risk of drug resistance emerging after discontinuation does not seem to have affected the response to reinitiation of boosted SQV-based HAART in this trial. A similar conclusion emerged from the Staccato trial, where 8 patients with rebounding HIV RNA levels during week-on-week-off treatment with boosted SQV-based HAART had no evidence of PI resistance and could be resuppressed on the same treatment.13 Similar observations have been made for patients stopping treatment with other ritonavir-boosted PIs, where no evidence of PI resistance is found.21 This may be different from the situation with nonnucleoside reverse transcriptase inhibitor (NNRTI)-based HAART, however, where resistance generated from single amino-acid substitutions has been detected in a significant proportion of patients discontinuing NNRTI-based HAART.8
There were no rises in CD4 count in the continuous treatment arm. The baseline CD4 count in this arm (653 cells/μL) was close to the lower limits of the normal range of CD4 counts in healthy volunteers, however. An apparent plateau in CD4 count increases has also been seen in analysis of long-term HAART treatment of patients with high pre-HAART CD4 counts.22 It is unclear whether the results of this trial could be extrapolated to other populations with lower pretherapy CD4 counts, for whom the risk of opportunistic infections may be higher during treatment interruptions; however, regular monitoring for CD4 counts in this situation should minimize this risk.
Failure of the week-on-week-off treatment is consistent with the results of the Staccato trial,13 which also evaluated mainly boosted SQV-based HAART in a randomized comparison with continuous HAART. A single-arm pilot study using 2 NRTIs plus boosted indinavir did show continued suppression of HIV RNA for more than 1 year in 10 patients,12 however, with undetectable HIV RNA at baseline. In addition, a recent study of 20 patients with HIV RNA suppression at baseline given NNRTI-based HAART for 5 days with an interruption of 2 days per week also showed durable suppression of HIV RNA levels.23 In a large treatment interruption study of a variety of HAART regimens with intensive sampling, HIV RNA levels started to rebound within 4 days in some patients.24 The reasons for these apparent mismatches in results are unclear.
The main limitations of this trial are its small sample size and the selection of the patients; the eligible patients were those who had successfully responded to HAART for several years and had high CD4 counts before ARVs and before STI and may not be representative of the wider population. It is also uncertain whether similar results would be seen for ARV combinations with different pharmacokinetics or safety profiles. The potential saving in treatment cost for CD4-guided treatment could be partially offset by a need for more intensive monitoring for CD4 counts. Using new technologies, there is the potential for low-cost CD4 testing, however, on the order of $5 US per test.25 The interval of retreatment at week 96 may have been too short to ensure HIV RNA suppression to less than 50 copies/mL for all patients in the CD4-guided arm, and the OT analysis of data to week 108 is to some extent exploratory; longer treatment durations were necessary in another study of HAART based on SQV-ritonavir in naive patients.26 The lack of difference in adverse events between the continuous and CD4-guided arms may be a function of the small sample size in this pilot study as well as the selection of this patient population with a history of long prior exposure to NRTI/PI-based HAART before the trial.
The median pre-ARV CD4 count in this trial was 358 cells/μL (IQ range: 309-436 cells/μL); this is relatively high for initiation of treatment in many countries. In other cohort studies, patients with lower pre-HAART CD4 counts had a higher risk of the CD4 count falling to low levels during treatment interruptions.11 This may provide support in the future for earlier initiation of HAART: starting HAART earlier, when the CD4 count is above 350 cells/μL, could offer the potential for extended treatment interruptions, which may not be feasible for those starting treatment later but with lower CD4 counts.
In summary, in this small randomized study of patients pretreated with dual NRTI who had high CD4 counts and undetectable HIV RNA levels before STI, CD4-guided treatment had a virologic outcome comparable to continuous HAART with a 54% ARV cost savings. The week-on-week-off strategy had a high rate of virologic failure and was prematurely stopped. CD4-guided treatment can be recommended for use in clinical practice only if the benefits are confirmed by larger trials (eg, SMART, Windows, ISS-Part, Staccato, DART) that are also evaluating these strategies in larger sample sizes and with longer follow-up time.
Hoffman La Roche and Abbott provided saquinavir and ritonavir respectively for use in this study. Hoffman La Roche also provided funding for the conduct of the study during the first 48 weeks.
1. Fellay J, Boubaker K, Ledergerber B, et al. Prevalence of adverse events associated with potent antiretroviral treatment: Swiss HIV Cohort Study. Lancet. 2001;358:1322-1327.
2. Creese A, Floyd K, Alban A, et al. Cost-effectiveness of HIV/AIDS interventions in Africa: a systematic review of the evidence. Lancet. 2002;359:1635-1643.
3. Hellinger FJ, Fleishman JA. Estimating the national cost of treating people with HIV disease: patient, payer, and provider data. J Acquir Immune Defic Syndr. 2000;24:182-188.
4. Walmsley S, Loutfy M. Can structured treatment interruptions (STIs) be used as a strategy to decrease total drug requirements and toxicity in HIV infection? J Int Assoc Physicians AIDS Care. 2002;1:95-103.
5. Hogg R, Heath K, Bangsberg D, et al. Intermittent use of triple-combination therapy is predictive of mortality at baseline and after 1 year of follow up. AIDS. 2002;16:1051-1058.
6. Lundgren J, Vella S, Paddam L, et al. Interruption/stopping antiretroviral therapy and the risk of clinical disease: results from the EuroSIDA Study [abstract 48]. Presented at: Eighth Conference on Retroviruses and Opportunistic Infections; 2002; Boston.
7. Taylor S, Allen S, Fidler S, et al. Stop study: after discontinuation of efavirenz, plasma concentrations may persist for 2 weeks or longer [abstract 131]. Presented at: 10th Conference on Retroviruses and Opportunistic Infections; 2002; San Francisco.
8. Palmisano L, Giulano M, Pirillo M, et al. Baseline predictors and virological outcome in subjects developing mutations during intermittent HAART [abstract 143]. Presented at: XIII International HIV Drug Resistance workshop; 2004, Canary Islands. Also published in Antivir Ther. 2004;9(Suppl):S159.
9. Lawrence J, Mayers D, Hullsiek K, et al. Structured treatment interruption in patients with multidrug-resistant human immunodeficiency virus. N Engl J Med. 2003;349:837-846.
10. Katlama C, Dominguez S, Gourlain K, et al. Benefit of treatment interruption in HIV-infected patients with multiple therapeutic failures: a randomized controlled trial (ANRS 097). AIDS. 2004;18:217-226.
11. Mussini C, Cozzi-Lepri A, Borghi V, et al. CD4-guided treatment interruptions: a new possible treatment strategy [abstract F11/1]. Presented at: Ninth European AIDS Conference; 2003; Warsaw.
12. Dybul M, Chun T, Yoder C, et al. Short-cycle structured intermittent treatment of chronic HIV infection with highly active antiretroviral therapy: effects on virologic, immunologic, and toxicity parameters. Proc Natl Acad Sci USA. 2001;98:15161-15166.
13. Ananworanich J, Nuesch R, Le Braz M, et al. Failures of 1 week on, 1 week off antiretroviral therapies in a randomised trial. AIDS. 2003;17(Suppl):F33-F37.
14. Fagard C, Oxenius A, Gunthard H, et al. A prospective trial of structured treatment interruptions in human immunodeficiency virus infection. Arch Intern Med. 2003;163:1220-1226.
15. Dybul M, Nies-Kraske E, Daucher M, et al. Long-cycle structured intermittent versus continuous highly active antiretroviral therapy for the treatment of chronic infection with human immunodeficiency virus: effects on drug toxicity and on immunologic and virologic parameters. J Infect Dis. 2003;188:388-396.
16. Kroon ED, Ungsedhapand C, Ruxrungtham K, et al. A randomized, double-blind trial of half versus standard dose of zidovudine plus zalcitabine in Thai HIV-1-infected patients (study HIV-NAT 001). HIV Netherlands Australia Thailand Research Collaboration. AIDS. 2000;14:1349-1356.
17. Cardiello P, van Heeswijk R, Hassink E, et al. Simplifying protease inhibitor therapy with once-daily dosing of saquinavir soft-gelatin capsules/ritonavir (1600/100 mg): HIVNAT 001.3 study. J Acquir Immune Defic Syndr. 2002;29:464-470.
18. Cardiello P, Hassink E, Ananworanich J, et al. Prospective randomized trial of structured treatment interruption (STI) in patients with chronic HIV infection. Clin Infect Dis. 2005;40:594-600.
19. Tebas P, Henry K, Mondy K, et al. Effect of prolonged discontinuation of successful antiretroviral therapy on CD4+ T cell decline in human immunodeficiency virus-infected patients: implications for intermittent therapeutic strategies. J Infect Dis. 2002;186:851-854.
20. Tarwater P, Parish M, Gallant J. Prolonged treatment interruption after immunologic response to highly active antiretroviral therapy. Clin Infect Dis. 2003;37:1541-1548.
21. Kempf DJ, King MS, Bernstein B. Incidence of resistance in a double-blind study comparing lopinavir/ritonavir plus stavudine and lamivudine to nelfinavir plus stavudine and lamivudine. J Infect Dis. 2004;189:51-60.
22. Kaufmann G, Perrin L, Pantaleo G, et al. CD4 T-lymphocyte recovery in individuals with advanced HIV-1 infection receiving potent antiretroviral therapy for 4 years. Arch Intern Med. 2003;163:2187-2195.
23. Cohen C, Morris A, Bazner S, et al. The FOTO Study: a pilot study of short-cycle treatment interruption, taking antiretroviral medications for five days on, two days off (FOTO), for those with viral load suppression on either PI or EFV-based regimens [abstract TuPeB4575]. Presented at: XV International AIDS Conference; 2004; Bangkok.
24. Fischer M, Hafner R, Schneider C, et al. HIV RNA in plasma rebounds within days during structured treatment interruption. AIDS. 2003;17:195-199.
25. Pattanapanyasat K, Lerdawa S, Shain H, et al. Low-cost enumeration in HIV-infected patients in Thailand. Asian Pac J Allergy Immunol. 2003;21:104-113.
26. Staszewski S, Stark T, Knecht G, et al. The Quad study: a pilot study to assess the efficacy and safety of Trizivir + RTV boosted saquinavir (TZV+SQV/r) compared to Combivir + RTV-boosted saquinavir (CBV-SQV/r) in ART naive patients with high viral load (VL) and low CD4 count. 24 week interim results [abstract F1/1]. Presented at: Ninth European AIDS Conference (EACS); 2003; Warsaw.
© 2005 Lippincott Williams & Wilkins, Inc.