Structured treatment interruptions (STIs) have been investigated in several clinical trials,1 with the main objective being to reduce drug exposure.
Two main approaches have been explored: the first is based on alternate periods “on” and “off” therapy,2-6 whereas in the second (CD4-guided interruptions), patients who have successfully responded to highly active antiretroviral therapy (HAART) discontinue treatment and resume it when their CD4 cell count reaches predefined limits.7-11 So far, a clear benefit has not been demonstrated for either of these 2 strategies; moreover, concern has recently arisen about their safety. Indeed, among trials on fixed-length STIs, the 4-weeks-off-8-weeks-on study conducted at the National Institutes of Health (NIH),3 the 1-week-on-1-week-off (WOWO) arm of the Staccato trial,5 and the development of antiretroviral therapy (DART) study4 were prematurely stopped because of emergence of resistance, a higher rate of virologic failure, and increased clinical progression, respectively. Conversely, there were no safety concerns in other studies.2,6 Among CD4 cell count-guided STI trials, the strategies for management of antiretroviral therapy (SMART) and Trivacan studies10,11 have been stopped because of a higher rate of clinical progression in the treatment interruption arms (in the Trivacan study, the fixed-length STI arm is ongoing). Conversely, in the Staccato and BASTA studies, this STI modality was generally safe.7,8
We present here the final results of the ISS-PART, a large, open-label, randomized study based on the fixed-length STI approach, with interruptions of progressively longer duration. This design aimed at testing the noninferiority of STIs to continuous HAART in terms of CD4 cell count and the autovaccination hypothesis,12 which was still in vogue at the time the study was planned.
Between July 2001 and July 2002, 68 clinical centers in Italy enrolled HIV-infected patients meeting the following requirements: first-line antiretroviral therapy (3 or 4 drugs) for ≥9 months (1 previous therapy change for toxicity or noncompliance was allowed), HIV RNA level <400 copies/mL for at least 6 months, CD4 count >350 cells/mm3, pre-HAART CD4 count >100 cells/mm3, and no previous AIDS diagnosis. The study was approved by the ethics review committees of the coordinating center (the Istituto Superiore di Sanità) and of the participating sites. All the procedures followed were in accordance with ethical standards of the responsible committee on human experimentation and with the Helsinki Declaration of 1975, as revised in 2000. Written informed consent was obtained from all patients.
Randomization and Study Design
Patients were centrally randomized at a 1:1 ratio to continue their antiretroviral regimen (arm A) or to undergo STIs of increasing duration (1, 1, 2, 2, and 3 months), with each followed by a 3-month therapy period (arm B). Patients were stratified according to their current antiretroviral regimen (with or without protease inhibitors [PIs]) and their pre-HAART CD4 count (>200 or <200 cells/mm3). Patients were monitored every 3 months (arm A) or monthly (arm B).
During STIs, therapy was resumed in advance if the HIV RNA level exceeded 50,000 copies/mL or the CD4 T-cell decline was >25% of the baseline count (>35% for patients with a CD4 count >500 cells/mm3 at randomization). Early therapy resumption for 2 consecutive times mandated protocol discontinuation. During “on-HAART” periods, if HIV RNA levels <400 copies/mL were not attained within 2 months, patients did not undergo the following scheduled STI and received 3 additional months of treatment.
In the STI group, nonnucleoside reverse transcriptase inhibitors (NNRTIs) were interrupted for 3 days (in the case of nevirapine-based combinations) or 6 days (for efavirenz-based combinations) before continuing with the remaining agents in the regimen.
The primary endpoint was the proportion of patients with a CD4 count >500 cells/mm3 at the end of follow-up in the 2 treatment arms. Secondary endpoints included occurrence of grade 3 or 4 adverse events, proportion of patients with a CD4 count >350 cells/mm3 or with an HIV RNA level <400 copies/mL at the end of follow-up, rate of virologic failure (defined as any determination of HIV RNA >400 copies/mL at any time in arm A and at the end of “on-treatment” periods in arm B), emergence of resistance, and HIV-related events.
Viral Genotype in Plasma and Peripheral Blood Mononuclear Cells
In arm B, plasma samples for genotyping were obtained in all patients at all STIs 15 or 30 days after drug suspension and were sequenced by the HIV-1 TRUGENE assay (Bayer HealthCare LLC, Tarrytown, NY) whenever the HIV RNA viral load exceeded 400 copies/mL. “Resistance” was defined as the detection of ≥1 significant (according to the 2005 International AIDS Society [IAS] USA classification) mutation at any time.
The same assay was used to sequence proviral DNA from baseline samples after extraction and amplification from peripheral blood mononuclear cells (PBMCs; QIAamp DNA Blood Mini Kit, Qiagen, Hilden, Germany).
The study was designed as a noninferiority trial. A total of 250 patients provided a power of (1-β) = 80% to demonstrate the noninferiority of STI-based HAART to continuous treatment in terms of the proportion of subjects with a CD4 count >500 cells/mm3 at 24 months under the following assumptions: 10% margin of noninferiority, global proportion of success expectancy of 90%, significance level of α = 0.05, and 15% loss to follow-up.
The primary endpoint was evaluated according to per-protocol (PP) and intention-to-treat (ITT) analyses. PP analysis included all patients completing follow-up and fully compliant with the protocol; ITT analysis included all enrolled patients. ITT was used for CD4 cell count and HIV RNA level data. The analysis of resistance, adverse events, and toxicities included enrolled patients on treatment. In arm B, HIV RNA level and CD4 cell count analysis was based on values observed at the end of each STI or treatment period.
Descriptive statistics and parametric and nonparametric tests were used to summarize and compare baseline characteristics in the 2 arms. The difference in the proportion of patients with a CD4 count >500 cells/mm3 in the continuous and STI arms was assessed using a 1-sided 95% confidence interval (CI). Time to virologic failure and time to emergence of resistance were analyzed using the Kaplan-Meier method and the log-rank test. The Cox proportional hazards model was used to estimate relative risks of events. Logistic regression models were used to identify factors associated with the probability of achieving the primary endpoint and predictors of resistance in arm B. Predictors of CD4 cell count decline during first STI were explored with a multiple linear regression model. The weighted mean of repeated HIV RNA measures during STIs (with the weight being the length of each STI period) was used to investigate the relation between HIV RNA and CD4 cell values at the end of the study in individual patients. This association was estimated by the Pearson correlation coefficient.
Statistical calculations were performed using SAS statistical package, version 8.2 (SAS Institute, Cary, NC).
A total of 284 patients were randomized, and 273 were enrolled in the study. The trial profile is described in Figure 1. Baseline characteristics of the enrolled patients are reported in Table 1. There were no statistically significant differences between the 2 arms.
Median follow-up was 24 months in both arms (range: 4 to 24 months in arm A and 1 to 24 months in arm B). The cumulative risk of protocol discontinuation at the end of follow-up was 19.4% (95% CI: 12.3 to 26.5) in arm A and 66.5% (95% CI: 50.4 to 82.6) in arm B (P < 0.0001, log-rank test; hazard ratio = 4.6, 95% CI: 3.0 to 7.3).
Early HAART resumption during STI for protocol-mandated criteria occurred in 52 (38.2%) arm B subjects (HIV RNA level >50,000 copies/mL in 21 subjects and CD4 cell count decline in 31 subjects). In 9 cases, this occurred 2 times consecutively and resulted in permanent protocol discontinuation.
Median time off therapy was 256 days (35.3% of the total 725 days of median follow-up).
Noninferiority of arm B was not demonstrated: the proportion of subjects with a CD4 count >500 cells/mm3 at the end of follow-up was 86.5% in arm A and 69.1% in arm B in the PP analysis (observed difference = 17.4, lower limit of the 1-sided 95% CI = 5.8) and 84.8% in arm A and 75.7% in arm B in the ITT analysis (difference = 9.1, lower limit of the 1-sided 95% CI = 0.2). In the PP analysis, arm A was superior to arm B (2-sided 95% CI: 3.6 to 31.2; P = 0.0075). Nevertheless, proportions of patients with a CD4 count >350 cells/mm3 were not different in the 2 arms in the ITT approach (98.3% in arm A and 96.1% in arm B, difference = 2.2, 95% CI: −2.2 to 6.6). In addition, no significant change was seen in CD4 cell count at the end of follow-up in either arm (median change: +6 in arm A and −26 in arm B; Fig. 2). Higher pre-HAART CD4 cell count and male gender were independently associated (P = 0.003 and P = 0.007, respectively) with “success” in arm B, defined as completing the study according to protocol and achieving the primary endpoint.
The magnitude of CD4 cell decline during first STI (median value = 72 cells/mm3) progressively increased with lower pre-HAART CD4 count and higher CD4 cell gain during HAART (pre-STI) (P = 0.03 and P < 0.001, respectively).
In arm B, all patients but 1 experienced viral rebound during ≥1 STI. The HIV RNA median log change was +2.3, +2.1, +2.3, +2.1, and +2.0 during the 5 STIs, respectively. Patients with pre-HAART HIV RNA >50,000 copies/mL were more likely to experience viral rebound to >50,000 copies/mL during ≥1 STI (68.8% vs. 31.3% among patients with <50,000 copies/mL; P < 0.0001). At the end of follow-up, the median HIV RNA level was not different from baseline in either arm; similarly, the proportion of patients with an HIV RNA level <400 copies/mL was 92.3% in arm A and 91.1% in arm B (P = 0.7), and the proportion with an HIV RNA level <50 copies/mL was 72.6% in arm A and 75.2% in arm B (P = 0.7). No correlation was found between the HIV RNA weighted mean over the whole “off-therapy” period and the CD4 cell value at the end of follow-up. Virologic failure was not different in arms A and B, occurring in 27 of 137 patients in arm A and in 27 of 136 patients in arm B, corresponding to a cumulative risk of 24% (95% CI: 15.3 to 32.3) in arm A and 26% (95% CI: 15.4 to 36.2) in arm B (P = 0.8, log-rank test; Fig. 3).
Two patients in arm A and 2 in arm B developed an HIV-related event (cervical dysplasia and oral leukoplakia with oral candidiasis in arm A and in situ cervical carcinoma and herpes zoster in arm B). There were 14 serious adverse events in arm A and 14 in arm B, none of which were related to treatment except for a case of acute retroviral syndrome at the first STI. There were 27 grade 3 or greater laboratory toxicities in arm A and 12 in arm B.
Emergence of Resistance
Over 5 STIs, 303 plasma samples with an HIV RNA level >400 copies/mL were amplified and sequenced. In 38 (27.9%) of 136 subjects, 1 or more mutations were detected on at least a single occasion, corresponding to a cumulative risk of 30% (95% CI: 21.8 to 38.2) by the end of follow-up. Fifty percent of subjects had 1 mutation, and 28.9% had 2 mutations. Ninety percent of all mutations was found in the reverse transcriptase gene, with the most frequent being M184V or M184I (21 [32.9%] of a total of 70 mutations), K103N (8 mutations [11.4%]), and M41L (5 mutations [7.1%]). In 6 patients, thymidine analog mutations (TAMs) clustered according to a TAM1 (M41L, L210W, or T215Y; 3 patients) or TAM2 (D67N, K70R, or K219Q/E; 3 patients) pathway. Forty-two percent of patients developed their first mutation during STI 1, 18.4% during STI 2, 21% during STI 3, and 18.4% during STI 4. Mutations occurred in 20 (20.2%) of 99 subjects treated with NNRTI-based HAART, 14 (46.7%) of 30 treated with PI-based HAART, and 4 (57.1%) of 7 treated with nucleoside reverse transcriptase inhibitor (NRTI)-based HAART (P = 0.004). The M184V/M184I mutation was found in 36.6% of PI-treated, 6% of NNRTI-treated, and 57.1% of NRTI-treated patients.
No differences in baseline characteristics, including previous changes of therapy and HIV plasma viremia <50 copies/mL, were observed in patients with or without mutations during STIs. HIV RNA and CD4 cell patterns over 24 months did not differ in the presence of mutations; however, 34.2% (13 of 38) of patients with mutations versus 14.6% (14 of 96) of patients without mutations experienced virologic failure. A time-dependent Cox regression model yielded a higher risk of virologic failure in the presence of mutations (hazard ratio = 2.6, 95% CI: 1.2 to 5.9).
In a subset of 82 patients with available samples (27 with and 55 without plasma mutations), we sequenced proviral DNA from baseline PBMCs. Mutations were found in 9 (33.3%) of 27 patients developing plasma mutations during STIs and in 1 (1.8%) of 55 of those retaining a wild-type virus during STIs (P < 0.001). M41L was the most frequent mutation (3 [30%] of 10 subjects), followed by M184V, D67N, and T215Y, with each detected in 2 (20%) of 10 subjects. In general, mutations found in PBMCs re-emerged during STIs, although intermittently. In 3 cases, additional “new” mutations were found in plasma (Table 2). In a logistic regression model, the presence of mutations in proviral DNA at baseline and the use of PI-based HAART were associated with the emergence of plasma mutations during STIs (P = 0.002 for DNA mutations and P = 0.048 for PI-based HAART).
After nearly a decade of clinical trials, it is not clear whether HAART can be safely interrupted in subjects who have achieved sustained viral suppression. In contrast to the positive results of some studies,6-8 since January 2006, 3 large randomized trials have been halted because of a higher rate of clinical events in the treatment interruption arms.4,10,11 Two of these (SMART and Trivacan) were based on CD4 cell-guided interruptions, whereas the DART trial consisted of time-fixed cycles “on-off” therapy. SMART study results were particularly relevant for the huge size of this trial and because the interruption strategy performed worse than continuous HAART irrespective of baseline CD4 cell count. Similarly, the STI arm had a worse outcome in the Trivacan and DART trials, which were conducted in resource-limited settings. The reasons for these controversial results have not been fully elucidated. In this scenario, although its design was based on an outdated hypothesis, the ISS-PART study may provide some insight into factors influencing the outcome of STI. Among these, our finding of a 58.8% rate of discontinuation in the STI arm, mainly attributable to low protocol compliance, indicates the poor feasibility of this kind of time-fixed STI, requiring intensive clinical and laboratory monitoring. In addition, we cannot exclude the possibility that a shift in the perception in the field with regard to STI may have contributed to the high discontinuation rate.
In the ISS-PART study, noninferiority of intermittent HAART was not demonstrated in terms of the proportion of patients with a CD4 count >500 cells/mm3 at 24 months. Two factors may have had an impact on this outcome, however, with the first being the high CD4 cell count threshold adopted as the primary endpoint. Indeed, with a threshold of 350 cells/mm3, the ITT approach did not show differences between arms; similar results were found in the Window study,6 in which STI strategy was not inferior to continuous HAART in terms of the proportion of subjects with a CD4 count ≥300 cells/mm3. The second factor comprised the lower (although not statistically significant) baseline proportion of patients with a CD4 count >500 cells/mm3 in arm B (79.9% vs. 84.7% in arm A) and the lower baseline median CD4 count in the same arm (659 vs. 731 cells/mm3 in arm A). In addition, CD4 cell median values in the 2 arms did not change over 24 months (median change: +6 in arm A and −26 in arm B). In arm A, this observation is of interest because it suggests that in a population of subjects who started HAART with a high CD4 cell count and were treated for a median of 2 years, no further increase in CD4 cell count occurs, at least with the regimens used in our study. In the ISS-PART study, success in the STI arm was defined as the likelihood of completing the study protocol and achieving a CD4 count >500 cells/mm3, and it was independently predicted by pre-HAART CD4 cell count and male gender. The role of male gender is difficult to explain, because no gender-related significant differences were observed for other variables. Our results on pre-HAART CD4 cell counts, which were strongly associated with the decrease in CD4 cell counts during STIs, are consistent with those previously reported in the literature,8,9,11 which had generated the belief that, irrespective of modalities, STIs could be safely proposed to patients with high CD4 cell levels at HAART initiation. The results of the SMART trial, in which the increased rate of progression and death in the STI arm occurred irrespective of CD4 cell nadir,10 have partially reversed this view, suggesting that other factors may influence the outcome of STIs. For instance, the magnitude of CD4 cell gain during HAART seemed to affect the CD4 cell decline after treatment interruption significantly in our study as well as in those of others,13,14 and immunologic factors may also play a role in CD4 cell-guided therapy interruptions, as recently shown.15
In the ISS-PART study, 24 months of intermittent therapy in subjects who had been on HAART for a median of 2 years did not increase the rate of virologic failure (24% in arm A and 26% in arm B), defined as a single determination of HIV RNA level >400 copies/mL; this rate is consistent with the data in the literature, such as those of the UK cohort, in which the failure (defined as HIV RNA level >500 copies/mL) rate was 33% by 4 years of HAART.16 In our study, the risk of failing was not associated with any baseline factor; however, in arm B, it was significantly higher in patients with resistance-associated mutations in the virus rebounding during STIs. Resistance has been reported during STI trials,17,18 with an incidence varying according to the setting, the extent of monitoring, and the sensitivity of the assay; its clinical implications are not clear.18,19 In the ISS-PART study, we decided to characterize the virus rebounding after drug removal genotypically and assess potential links between the presence of mutations and the response to therapy reinstitution. To this purpose, plasma samples were obtained in all patients at all STIs and sequenced whenever HIV viremia exceeded 400 copies/mL. Despite the fact that all subjects were on first-line HAART and had no previous virologic failure, in a relatively high proportion (27.9%) of them, the rebounding virus carried major mutations; the majority (approximately 66%) were associated with NRTI resistance, with the M184V mutation accounting for 50% of these and being found in 20 subjects, with all but 1 receiving lamivudine in their current HAART. This finding was not unexpected, because viruses with this mutation have been described in several trials of STIs as minority variants or as more commonly represented species.20,21 The low genetic barrier of lamivudine, together with its long intracellular half-life (12 hours), allowing for the persistence of suboptimal drug levels in the initial period of STI, may have played a role. In agreement with data found in the literature,17 the M184V mutation tended to emerge early in our study, during the first or second STI. Because we did not test samples taken before day 15 of STI and the assay we used does not identify minority variants, we cannot exclude the presence of the M184V mutation at even earlier time points or as mutants representing <20% of the whole viral population. TAM1 and TAM2 clusters were equally represented (each in 3 patients), and 1 case of failure per cluster was observed.
The study of baseline factors associated with resistance provided interesting results. The presence of archived mutations in proviral DNA before STI was the strongest predictor of plasma mutations at viral rebound, followed by the use of unboosted PI-based HAART (in comparison to NNRTI-based HAART). Both findings were quite surprising. Archived mutations have been detected in subjects previously exposed to monotherapy or dual therapy22; however, this was not the case in our patients. Thus, the presence of archived mutations in 10 of the 82 individuals tested may indicate infection by a resistant strain or the selection of mutations during an early phase of treatment that persist in the cellular reservoir23 and re-emerge when drugs are removed. The preexistence of mutations in these subjects is confirmed by the similar patterns of mutations in plasma and PBMCs, except for the M184V mutation (found in only 12.5% of the subjects subsequently developing this mutation) and the K103N mutation (absent in all the baseline proviral DNA samples), suggesting their de novo selection during STIs. The other factor associated with resistance during STIs was the use of PI-based HAART (mainly consisting of unboosted PIs in our population) versus NNRTI-based HAART. This finding is in apparent disagreement with data found in the literature reporting a higher rate of resistance in subjects interrupting an NNRTI-based regimen.24 A factor that may have played a role is the sequential modality of drug interruption adopted in our protocol to account for the long half-life of NNRTIs.25,26 Furthermore, less residual replication in subjects treated with NNRTIs may have reduced viral evolution and emergence of mutations conferring resistance to the NRTI backbone. In fact, in a previous report, we found that NNRTI-based regimens were associated with lower values of residual HIV viremia when compared with regimens that included mainly unboosted PIs.27
The ISS-PART trial has some limitations and some strengths. The high proportion of protocol discontinuations, although not preventing statistical evaluation of the primary endpoint, represented the most important drawback of our strategy. The major strength of the study was its focus on safety, which led us to adopt strict criteria for defining immunologic and virologic failure and to search actively for resistance in the rebounding virus.
In conclusion, in the ISS-PART study, which was conducted in a population with a relatively high CD4 cell count before HAART and at enrollment, the STI strategy did not fulfill the immunologic criteria for noninferiority to continuous HAART but did not increase the overall rate of virologic failure at the end of follow-up. Patients with archived mutations and receiving unboosted PI-based HAART were more likely to have mutated virus at rebound and to exhibit a poorer response to HAART reintroduction.
The STI strategy has not been completely abandoned, but it has been restricted to research settings.28 In the ISS-PART study, we identified factors that should be considered as exclusion criteria from any clinical trial evaluating treatment interruptions.
The authors thank the independent Data and Safety Monitoring Board (R. Gulick, Cornell University, NY; J. Montaner, The University of British Columbia, Vancouver, British Columbia, Canada; and P. Yeni, Hôpital Bichat-Claude Bernard, Paris, France), the technical staff at Istituto Superiore di Sanità (M. Mirra, F. Innocenti, and A. Mattei), and the patients and their families.
1. Pai NP, Tulsky JP, Lawrence J, et al. Structured treatment interruptions (STI) in chronic suppressed HIV infection in adults. Cochrane Database Syst Rev
2. Dybul M, Nies-Kraske E, Dewar R, et al. A proof-of-concept study of short-cycle intermittent antiretroviral therapy with a once-daily regimen of didanosine, lamivudine, and efavirenz for the treatment of chronic HIV infection. J Infect Dis
3. 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
4. Hakim J, on behalf of the DART Trial Team. A structured treatment interruptions (STI) strategy of 12 weeks cycles on and off ART is clinically inferior to continuous treatment in patients with low CD4 counts before ART: a randomisation within the DART Trial [abstract THLB0207]. Presented at: XVI International AIDS Conference; 2006; Toronto.
5. Ananworanich J, Nuesch R, Le Braz M, et al. Failures of 1 week on, 1 week off antiretroviral therapies in a randomized trial. AIDS
6. Marchou B, Tangre P, Charreau I, et al. Intermittent antiretroviral therapy in patients with controlled HIV infection. AIDS
7. Ananworanich J, Gawet-Ageron A, Le Braz M, et al. CD4-guided scheduled treatment interruptions compared with continuous therapy for patients infected with HIV-1: results of the Staccato randomised trial. Lancet
8. Maggiolo F, Ripamonti D, Gregis G, et al. Effect of prolonged discontinuation of successful antiretroviral therapy on CD4 T cells: a controlled, prospective trial. AIDS
9. Mussini C, Bedini A, Borghi V, et al. CD4 cell-monitored treatment interruption in patients with a CD4 cell count >500 × 106
10. El-Sadr WM, Lundgren JD, Neaton JD, et al. CD4+ count guided interruption of antiretroviral treatment. N Engl J Med
11. Danel C, Moh R, Minga A, et al. CD4-guided structured antiretroviral treatment interruption strategy in HIV-infected adults in West Africa (Trivacan ANRS 1269 trial): a randomised trial. Lancet
12. Lori F, Macerati R, Foli A, et al. Structured treatment interruptions to control HIV-1 infection. Lancet
13. 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
14. Fagard C, Bandelier CY, Ananworanich J, et al. Biphasic decline of CD4 cell count during scheduled treatment interruptions. AIDS
15. Nemes E, Lugli E, Nasi M, et al. Immunophenotype of HIV+ patients during CD4 cell-monitored treatment interruption: role of the IL-7/IL 7 receptor system. AIDS
16. The UK Collaborative Group on HIV Drug Resistance and UK CHIC Study Group. Long term probability of detection of HIV-1 drug resistance after starting antiretroviral therapy in routine clinical practice. AIDS
17. Arnedo-Valero M, Garcia F, Gil C, et al. Risk of selecting de novo drug-resistance mutations during structured treatment interruptions in patients with chronic HIV infection. Clin Infect Dis
18. Yerly S, Fagard C, Günthard HF, et al. Swiss HIV Cohort Study. Drug resistance mutations during structured treatment interruptions. Antivir Ther
19. Papasavvas E, Grant RM, Sun J, et al. Lack of persistent drug-resistant mutations evaluated within and between treatment interruptions in chronically HIV-1 infected patients. AIDS
20. Metzner KJ, Bonhoeffer S, Fischer M, et al. Emergence of minor populations of human immunodeficiency virus type 1 carrying the M184V and L90M mutations in subjects undergoing structured treatment interruptions. J Infect Dis
21. Martinez-Picado J, Morales-Lopetegi K, Wrin T, et al. Selection of drug-resistant HIV-1 mutants in response to repeated structured treatment interruptions. AIDS
22. Opravil M, Hirschel B, Lazzarin A, et al. A randomized trial of simplified maintenance therapy with abacavir, lamivudine, and zidovudine in human immunodeficiency virus infection. J Infect Dis
23. Ghosn J, Pellegrin I, Goujard C, et al. HIV-1 resistant strains acquired at the time of primary infection massively fuel the cellular reservoir and persist for lengthy periods of time. AIDS
24. Barreiro P, de Mendoza C, Gonzalez-Lahoz J, et al. Superiority of protease inhibitors over nonnucleoside reverse-transcriptase inhibitors when highly active antiretroviral therapy is resumed after treatment interruption. Clin Infect Dis
25. 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: 11th Conference on Retroviruses and Opportunistic Infections; 2004; San Francisco.
26. Muro E, Droste JA, Hofstede HT, et al. Nevirapine plasma concentrations are still detectable after more than 2 weeks in the majority of women receiving single-dose nevirapine: implications for intervention studies. J Acquir Immune Defic Syndr
27. Palmisano L, Giuliano M, Nicastri E, et al. Residual viremia in subjects with chronic HIV infection and viral load <50 copies/ml: the impact of highly active antiretroviral therapy. AIDS
The participating clinical sites are as follows: N. Abrescia, M. Figoni, and R. Viglietti (Azienda Ospedaliera D. Cotugno, IV Divisione Malattie Infettive, Napoli); G. Angarano and A. Saracino (Università degli Studi di Foggia, Clinica Malattie Infettive, Foggia); M. Anselmo (Ospedale San Paolo, U. O. Malattie Infettive, Savona); A. Antinori, P. Sette, M. Zaccarelli, and G. Liuzzi (IRCCS “Lazzaro Spallanzani,” III Divisione Malattie Infettive, Roma); M. Arlotti, L.T. Martelli, and P. Ortolani (Ospedali “Infermi” di Rimini, Divisione Malattie Infettive, Rimini); D. Bassetti, A. Di Biagio, and F. Bisio (Università di Genova, Clinica Malattie Infettive, Genova); P. Bellissima (Azienda Ospedaliera Gravina, Divisione Malattie Infettive, Caltagirone); F. Branz and N. Dorigoni (Presidio Ospedaliero “Villa Igea,” Divisione Malattie Infettive, Trento); G. Cadeo, D. Vangi, D. Bertelli, and A. Bergamasco (Spedali Civili di Brescia, I Divisione Malattie Infettive, Brescia); L. Caggese and A. Volonterio (Ospedale Niguarda Cà Granda, Divisione Malattie Infettive, Milano); P. Caramello, G. C. Orofino, S. Carosella, and L. Gennero (Ospedale Amedeo di Savoia, U. O. Malattie Infettive, Divisione A, Torino); M. Caremani and D. Tacconi (Ospedale S. Donato, Divisione Malattie Infettive, Arezzo); G. Carosi, F. Castelli, L. Tomasoni, and A. Patroni (Università degli Studi di Brescia, Clinica di Malattie Infettive e Tropicali, Spedali Civili, Brescia); F. Chiodo, M. Borderi, and L. Calza (Policlinico S. Orsola, Malpighi, Sezione di Malattie Infettive, Bologna); F. Gritti and G. Fasulo (Ospedale Maggiore, U. O. Malattie Infettive, Bologna); A. Chirianni and M. Gargiulo (Azienda Ospedaliera D. Cotugno, III Divisione Malattie Infettive, Napoli); A. Colomba, E. R. Dalle Nogare, F. Di Lorenzo, and T. Prestileo (P. O. Casa del Sole, Divisione Malattie Infettive, Palermo); A. d'Arminio Monforte, T. Bini, and P. Cicconi (Università degli Studi di Milano, Istituto Malattie Infettive, Milano); F. De Lalla and M. T. Giordani (Ospedale S. Bortolo USL no. 8 di Vicenza, Divisione Malattie Infettive, Vicenza); C. De Stefano and G. De Stefano (Ospedale S. Carlo, U.O. Malattie Infettive, Potenza); S. Delia and M. Ciardi (Policlinico Umberto I, III Divisione Malattie Infettive, Roma); G. Di Perri, A. Sinicco, and P. Sales (Università di Torino, Clinica Malattie Infettive, Torino); M. Dini and M. Simeone (Azienda Ospedaliera, Umberto I, Divisione Malattie Infettive, Ancona); R. Esposito, G. Guaraldi, and B. Beghetto (Università degli Studi di Modena, Clinica Malattie Infettive e Tropicali, Modena); F. Fatuzzo and R. La Rosa (Azienda Ospedaliera Vittorio Emanuele, Ferrarotto, S. Bambino, Divisione Malattie Infettive, Catania); C. Ferrari and C. Calzetti (Azienda Ospedaliera di Parma, Divisione Malattie Infettive, Parma); T. Ferraro and L. Cosco (Az. Ospedaliera A. Pugliese Ciaccio, Divisione Malattie Infettive, Catanzaro); F. Ghinelli and L. Sighinolfi (Arcispedale Sant'Anna, Divisione Malattie Infettive, Ferrara); V. Guadagnino, B. Caroleo, and A. Izzi (Policlinico “Mater Domini,” U.O. Malattie Infettive, Catanzaro); C. Izzo and A. Franco (Azienda Ospedaliera D. Cotugno, VIII Divisione Malattie Infettive, Napoli); A. Lazzarin, A. Castagna, and G. Fusetti (Ospedale San Raffaele, Centro Ricerca e Cura per le Patologie HIV Correlate, Milano); F. Leoncini, M. Pozzi, S. Sbaragli, and M. Marzetti (Ospedale Careggi, Divisione Malattie Infettive, Firenze); G. Magnani, L. Bonazzi, E. Barchi, and G. Zoboli (ASMN Reggio Emilia, Divisione Malattie Infettive, Reggio Emilia); A. Pintus, A. Mandas, M.L. Soddu, and F. Zucca (Università degli Studi di Cagliari, Centro Operativo SIDA, Cagliari); P. M. Mannucci and A. Gringeri (IRCCS Ospedale Maggiore di Milano, Centro Emofilia e Trombosi, Divisione Medicina Interna, Milano); G. Marani Toro, R. V. Graziani, and A. Consorti (Ospedale Civile, Divisione Malattie Infettive, Pescara); F. Mazzotta, M. Di Pietro, and C. Blè (Ospedale S. M. Annunziata, U. O. Malattie Infettive, Firenze); F. Meneghetti, L. Sasset, and A. M. Cattelan (Azienda Ospedaliera di Padova, Divisione Malattie Infettive e Tropicali, Padova); F. Menichetti and E. Savalli (Azienda Ospedaliera Pisana, Spedali Riuniti di S. Chiara, U. O. Malattie Infettive, Pisa); P. Mian and R. Pristerà (Ospedale Generale Regionale di Bolzano, Divisione Malattie Infettive, Bolzano); E. Mignani and S. Artioli (Azienda Usl no. 5 Spezzino Regione Liguria Ospedale Felettino, Malattie Infettive, La Spezia); M. S. Mura and M. Mannazzu (Università degli Studi di Sassari, Istituto di Malattie Infettive, Sassari); P. Narciso and R. Bellagamba (IRCCS “Lazzaro Spallanzani,” IV Divisione Malattie Infettive, Roma); A. Orani and P. Perini (Ospedale Azienda Alessandro Manzoni, Divisione di Malattie Infettive, Lecco); L. Ortona, A. De Luca, and R. Murri (Policlinico “A. Gemelli” Università Cattolica del Sacro Cuore, Clinica Malattie Infettive, Roma); G. Pagano and A. Alessandrini (Ospedale San Martino, U. O. Malattie Infettive, Genova); A. Paladini, M.A. Vinattieri, and S. Carbonai (Ospedale di Prato Misericordia e Dolce, U. O. Malattie Infettive, Prato); G. Pastore, N. Ladina, and M. Tateo, (Università degli Studi di Bari, Clinica Malattie Infettive, Bari); N. Piersantelli and G. Penco (E. O. Ospedali Galliera, U. O. Malattie Infettive, Genova); E. Petrelli and M. Balducci (Ospedale Civile di Pesaro, Divisione Malattie Infettive, Pesaro); L. Pippi and A. Gonnelli (Policlinico Le Scotte, U. O. Malattie Infettive 1, Siena); F. Puppo and G. Murdaca (Università degli Studi di Genova, Centro per le Immunodeficienze, Dipartimento di Medicina Interna, Genova); E. Raise and A. Pasquirucci (Ospedale Civile SS. Giovanni e Paolo, Divisione Malattie Infettive, Venezia); G. Riccio, V. Bartolacci, and G. Carrega (Ospedale Santa Corona, Sezione di Malattie Infettive, Pietra Ligure); G. Rizzardini and G. Migliorino (Ospedale Civile, Divisione Malattie Infettive, Busto Arsizio); R. Russo, S. Casentino, and M. Celesia (Università degli Studi di Catania, Istituto Malattie Infettive, Catania); M. L. Soranzo, A. Macor, and B. Salassa (Ospedale Amedeo di Savoia, U. O. Malattie Infettive B, Torino); F. Soscia, L. Roberti, and M. T. Di Toro (Ospedale S. Maria Goretti, Centro Riferimento AIDS, Latina); A. Stagno and C. Beltrami (Azienda Usl di Cesena, Ospedale Bufalini, U. O. di Malattie Infettive, Cesena); F. Suter, F. Maggiolo, and D. Ripamonti (Ospedali Riuniti di Bergamo, Reparto Malattie Infettive, Bergamo); G. Tantimonaco and B. Grisorio (Azienda Ospedali Riuniti, Divisione Malattie Infettive, Foggia); A. Tassara and P. Rossi (Ospedale Regionale di Aosta, Divisione Malattie Infettive, Aosta); M. Tinelli and A. Regazzetti (Ospedale Delmati, U. O. Malattie Infettive, S. Angelo Lodigiano); U. Tirelli, G. Voltaggio, and R. Cinelli (Centro di Riferimento Oncologico, Divisione di Oncologia Medica A, Aviano); M. Toti, M. Baldari, T. Carli, B. Ricciardi, and M. Trezzi (Ospedale della Misericordia, Divisione Malattie Infettive, Grosseto); G. M. Vigevani, A. Capetti, and S. Landonio (Ospedale L. Sacco, I Divisione Malattie Infettive e Servizio di Allergologia, Milano); V. Vullo and P. Massetti (Policlinico Umberto I, Università di Roma La Sapienza, Istituto Malattie Infettive e Tropicali, Roma); and T. Zauli and S. Casolari (Servizio Aziendale di Malattie Infettive c/o Presidio Ospedaliero di Ravenna, Ravenna).