Secondary Logo

Journal Logo

CLINICAL SCIENCE

Factors associated with virological rebound in HIV-infected patients receiving protease inhibitor monotherapy

Stöhr, Wolfgang; Dunn, David T.; Arenas-Pinto, Alejandro; Orkin, Chloe; Clarke, Amanda; Williams, Ian; Johnson, Margaret; Beeching, Nicholas J.; Wilkins, Edmund; Sanders, Karen; Paton, Nicholas I. for the Protease Inhibitor Monotherapy Versus Ongoing Triple Therapy (PIVOT) Trial Team

Author Information
doi: 10.1097/QAD.0000000000001206
  • Free

Abstract

Introduction

The Protease Inhibitor Monotherapy Versus Ongoing Triple Therapy (PIVOT) trial showed that protease inhibitor monotherapy, with regular HIV viral load monitoring and prompt reintroduction of combination therapy for viral load rebound, was noninferior to combination therapy in preserving future treatment options (i.e. drug resistance) and therefore is a well tolerated and effective alternative for long-term clinical management of HIV infection [1].

Like several other monotherapy trials, the PIVOT trial demonstrated a much higher rate of viral load rebound on protease inhibitor monotherapy than with standard combination therapy (32% higher in PIVOT, 10–13% difference found in a previous systematic review) [2]. The viral load rebound is generally of low level, easily and rapidly suppressed by re-introduction of triple therapy and without apparent long-term consequences [1]. Nevertheless, the ability to target patient selection to those at lower risk of viral load rebound might increase the appeal of the strategy.

Several previous trials and observational studies have aimed to find characteristics associated with a lower risk of viral load rebound. Lower nadir CD4+ cell counts [3–6], a shorter duration of viral load suppression [6–8] or treatment before monotherapy [9], detectable baseline HIV-viral load [9,10], hepatitis C virus (HCV) coinfection [10,11] and nonadherence to treatment [4,9] have all been identified as predictors, although not consistently. The impact of the type of viral load assay does not appear to have been explored.

The aim of this study was to analyse the risk of viral load rebound by duration of protease inhibitor monotherapy and identify clinical predictors of this risk.

Methods

Study design and population

PIVOT was a randomized controlled trial, performed in 43 centres in the United Kingdom. The main inclusion criteria were being on antiretroviral therapy (ART) comprising two nucleoside reverse transcriptase inhibitors (NRTIs) and one non-NRTI (NNRTI) or protease inhibitor for at least 24 weeks, having a viral load less than 50 copies/ml at screening and for at least 24 weeks beforehand (isolated blips <200 copies/ml allowed, if followed by two viral load <50 copies/ml) and having a CD4+ cell count more than 100 cells/μl at screening. The main exclusion criteria were the presence of known major protease inhibitor resistance mutation(s) on a resistance test before study entry, previous ART change for unsatisfactory virological response, undergoing or planning to start treatment for HCV coinfection, and coinfection with hepatitis B virus (HBV) unless the patient has had a documented HBV DNA measurement of less than 1000 copies/ml taken whilst off HBV active drugs. The full eligibility criteria are published elsewhere [1].

Participants were randomly assigned 1 : 1 to maintain ongoing triple therapy or switch to a strategy of physician-selected ritonavir-boosted protease inhibitor monotherapy with prompt return to combination therapy if viral load rebound (see below) occurred. This analysis is limited to participants assigned to the protease inhibitor monotherapy arm who commenced protease inhibitor monotherapy. Blood samples were taken at screening, baseline, weeks 4, 8, 12, and every 12 weeks thereafter. Viral load was measured in the clinical laboratory at each centre, using a variety of assays determined by local preference.

The protocol was approved by the Cambridgeshire 4 Research Ethics Committee and the Medicines and Healthcare products Regulatory Agency, UK. All participants provided written informed consent.

Statistical methods

The endpoint of this analysis was viral load rebound defined as three consecutive viral load tests at least 50 copies/ml, with at least 4 weeks between the first and last test; one of these three tests could be a repeat of the same sample. Time from start of protease inhibitor monotherapy to first viral load rebound (date of third consecutive viral load at least 50 copies/ml) was analysed using time-to-event methods, censoring at the earlier of switch from protease inhibitor monotherapy for any reason (stop of all antiretroviral drugs, or restart of combination ART) or last clinical follow-up if the outcome had not occurred. We ignored subsequent viral load rebounds in patients who had more than one episode.

We used flexible parametric survival models to estimate the probability of not having viral load rebound by duration of protease inhibitor monotherapy, including extrapolation beyond the actual follow-up. In brief, these models extend standard parametric models using restricted cubic splines rather than linear functions for the underlying log cumulative hazard and are implemented in Stata stpm2 [12]. We fitted models on the log cumulative hazard scale to parallel the hazard ratio calculated from the standard Cox model. Akaike Information Criteria was used to identify the best-fitting model, testing 1–6 degrees of freedom of the underlying spline for the log cumulative hazard; the best fitting model had degrees of freedom = 5.

Univariable and multivariable flexible parametric survival models were used to find predictors of viral load rebound. We evaluated the following baseline factors: age, sex, ART at randomization (NRTIs + NNRTI versus NRTIs + protease inhibitor), time since first consistently suppressed viral load below 50 copies/ml, CD4+ cell count nadir, baseline CD4+ cell count, viral load at least 50 copies/ml at baseline or in the previous 6 months, ethnicity (white, nonwhite), BMI and HCV status (antibody positive). In addition, we evaluated several time-varying factors: ART nonadherence (missed at least one dose since the last visit, self-report), current protease inhibitor (some patients changed protease inhibitor during follow-up because of adverse events or other clinical decision) and HIV-viral load assay (some centres changed their local viral load assay during the trial period). For illustration of the predictive utility in clinical practice, we predicted time without viral load rebound for hypothetical patients with selected values of duration of viral load suppression, CD4+ cell count nadir and viral load assay adjusted for all factors with P less than 0.1 in the multivariable model; stpm2 achieves this by estimating a survival curve for each individual and then averaging these.

Stata software, version 14.0 (StataCorp, College Station, Texas, USA), was used for all analyses.

Results

Of the 587 participants enrolled in PIVOT, 296 were randomized to protease inhibitor monotherapy. Of these, six patients never started monotherapy and were excluded from this analysis. Baseline characteristics are summarized in Table 1. Initial ritonavir-boosted protease inhibitor was darunavir in 233 (80%), lopinavir in 40 (14%), atazanavir in 16 (6%) and saquinavir in one (<1%). At study entry, the median time since first consistently suppressed viral load less than 50 copies/ml was 3 years.

T1-7
Table 1:
Baseline characteristics.

Rate and risk of viral load rebound over time

The 290 participants included had a median trial follow-up time of 44 (range 3–59) months. Of these, 93 (Kaplan–Meier estimate 38.5%) had viral load rebound whilst treated with protease inhibitor monotherapy with a median peak viral load of 526 copies/ml [peak was <400 copies/ml in 39 (42%)]. The follow-up was censored because of death (n = 1), withdrawal/lost to follow-up (n = 2), discontinuation of monotherapy for other reasons than viral load rebound (n = 40 – 15 for toxicity, nine detectable viral load not meeting protocol criteria for viral load rebound, nine patient decision, seven other or unknown reasons) and end of trial (n = 154).

The cumulative risk of viral load rebound estimated from the parametric model is shown in Fig. 1 and was very close to the curve obtained using the Kaplan–Meier approach. The rate of viral load rebound increased initially, peaking at around 9 months after the start of protease inhibitor monotherapy. It declined subsequently until around 18 months and then remained comparatively stable (approximately 5 per 100 person-years).

F1-7
Fig. 1:
Overall probability and hazard of viral load rebound.

Overall, the estimated proportion of participants without viral load rebound was 77% (95% CI 73–82%), 68% (95% CI 62–73%), and 62% (95% CI 56–69%) by 1, 3, and 5 years after start of protease inhibitor monotherapy. Extrapolated probabilities of not having viral load rebound after 7 and 10 years on protease inhibitor monotherapy, that is beyond the actual follow-up in PIVOT, were 58% (49–66%) and 53% (42–63%).

Predictors of viral load rebound

Across all viral load measurements, Abbott RealTime (Abbott RealTime HIV-1; Abbott Molecular Inc., Des Plaines, Illinois, USA) (40% of all measurements) and Roche TaqMan-2 (COBAS AmpliPrep/COBAS TaqMan HIV-1 Test, v2.0; Roche Molecular Systems, Inc., Pleasanton, California, USA) (35%) were the most commonly used viral load assays; all other assays, including in-house assays (11%), Siemens Versant 3.0 bDNA (VERSANT HIV-1 RNA 3.0 Assay (bDNA); Siemens Healthcare Diagnostics, Frimley, Camberley, Surrey, UK) (10%), Nuclisens 2.0 (NucliSENS EasyQ HIV-1 v2.0; bioMérieux SA, Marcy L’Etoile, France) (2%) and various other assays [≤1% each, including Roche Amplicor (Roche COBAS Amplicor HIV-1 Monitor Test, v1.5.; Roche Molecular Systems, Inc.)] were combined in a third category for analysis. Viral load assay was changed during follow-up in 41 of 290 (14%) participants.

Results of univariable and multivariable models are shown in Table 2. Independent predictors of viral load rebound were shorter time since first viral load suppression (hazard ratio 0.81 per year longer suppressed; P < 0.001), lower CD4+ cell count nadir (hazard ratio 0.73 per additional 100 cells/μl; P = 0.008), a higher baseline CD4+ cell count (hazard ratio 1.13 per 100 cells/μl increase; P = 0.023), nonwhite ethnicity (hazard ratio 1.87 versus white, P = 0.025), and testing with the TaqMan-2 assay (hazard ratio 1.87 compared with Abbott RealTime assay, P = 0.012). There was also a suggestion of a higher rate of viral load rebound associated with lower adherence (P = 0.069). There was no significant difference between lopinavir and darunavir monotherapy (P = 0.13), and no significant association with any other factor tested.

T2-7
Table 2:
Predictors of viral load rebound during protease inhibitor monotherapy.

The impact of the type of assay used is illustrated in Fig. 2. The overall probability of no viral load rebound by year 5 on protease inhibitor monotherapy was 68% for Abbott RealTime, and 50% for the TaqMan-2 assay.

F2-7
Fig. 2:
Predicted time to viral load rebound, by viral load assay and time since viral load suppression.(a) Abbott RealTime assay and (b) TaqMan-2 assay.

For a hypothetical patient with a CD4+ nadir of 350 cells/μl, 6 years’ prior viral load suppression with triple therapy, and follow-up testing using the Abbott RealTime assay, the model predicted that the probability of no viral load rebound by 5 years would be 86% (88% at 3 years). For a patient with a CD4+ nadir of 100 cells/μl, 1.5 years prior viral load suppression on triple therapy and follow-up testing with the Abbott RealTime assay, the model predicted a probability of no viral load rebound by 5 years of 51% (58% at 3 years).

Discussion

In this analysis, we have provided further information on the timing of viral load rebounds in patients taking protease inhibitor monotherapy in the PIVOT trial. We have shown a high rate of early viral load rebound which peaked around 9 months from switch. This early rate of confirmed viral load rebound appears to be greater than that observed in previous trials of protease inhibitor monotherapy [2] for reasons that are unclear (although differences in trial populations on some of the factors below, especially the method of viral load testing, are likely to have played a role). Significantly, this high rate was short-lived: in the period after 18 months from the time of switch only about 5% of patients experienced viral load rebound each year. Thus a subgroup of patients are likely to be able to remain on protease inhibitor monotherapy in the longer term, highlighting the importance of identifying factors that predict a sustained response. The size of the PIVOT trial, the duration of follow-up, and the larger number of viral load rebound episodes observed means that this trial population is able to make a definitive contribution to the understanding of these factors.

We found an increasing risk of virological rebound with shorter duration of viral load suppression prior to switch to monotherapy, as has been found in other studies [6–8]. A related measure, shorter time of prior ART exposure, was found to be a risk factor in the MONOI trial on darunavir monotherapy [9]. Prolonged viral suppression could be a marker for better adherence to antiretroviral therapy or favourable genetic factors but might also be a surrogate marker of lower viral reservoir size [13,14]. In the MONOI trial, a baseline ultrasensitive viral load more than 1 copy/ml and higher baseline HIV-1 DNA were predictive of virological rebound [9]. Similar to a study on combination ART [13], we did not find any ceiling on this effect – the longer that patients were suppressed before switching (even beyond 2–3 years) the better they appeared to do on monotherapy. This suggests that protease inhibitor monotherapy may be best positioned as a clinical management option for patients who have already demonstrated long-standing viral load suppression rather than being considered as part of a planned stepwise reduction strategy in patients starting ART for the first time [15].

A lower CD4+ cell count nadir was also associated with a higher risk for viral load rebound, independently of time since HIV-RNA suppression, which was also seen in other monotherapy studies [3–6,16]. Patients with low nadir CD4+ appear to have deficits in immune function that recover only slowly and have a larger HIV reservoir despite prolonged viral load suppression on ART [17–19]. Although effective ART is essential for control of viral load replication in most patients, this does not negate a potentially important contribution of the immune system to maintaining viral load suppression, and this may vary between individuals [20,21]. A higher baseline CD4+ cell count had no effect in the univariable model but was associated with a higher risk of rebound in the multivariable analysis. This effect appeared only after adjusting for CD4+ nadir and time since RNA suppression and is therefore likely a statistical artefact introduced by over-adjustment for these factors, or possibly because of residual confounding.

We found a strong association between the type of viral load assay and viral load rebound, with TaqMan-2 showing a nearly two-fold higher rate of rebound than the Abbott RealTime assay. This result is consistent with several studies that tested samples in parallel with these two assays and found that TaqMan-2 has greater sensitivity for detecting low-level HIV RNA [22,23]. Although not based on a direct comparison of assay methods in each patient, this finding has some important implications in the context of protease inhibitor monotherapy. First, for clinical management of patients on protease inhibitor monotherapy, it may be more appropriate to define higher thresholds of viral load rebound for more sensitive assays [24]. The PIVOT protocol specified a conservative approach for re-initiating combination therapy, and patients tested using the TaqMan-2 assay may have reintroduced triple therapy prematurely (especially given that resistance resulting from the strategy was very rare). Second, it may explain in part the higher rate of rebound seen in PIVOT compared with some other protease inhibitor monotherapy trials. The impact is hard to assess as a variety of viral load assays have been used in previous trials [2], and in many cases this was not reported. Third, it will be important that this factor is assessed in any future studies or analyses of protease inhibitor monotherapy data.

PIVOT was a strategic trial intended to resemble the use of protease inhibitor monotherapy in clinical practice and hence we allowed clinicians to select the protease inhibitor for use as monotherapy (although we recommended the use of darunavir or lopinavir). This allows us to compare several protease inhibitors within the same trial, whereas previous protease inhibitor monotherapy trials mandated the use of a single protease inhibitor. Most clinicians/patients selected darunavir although a sizeable minority (14%) used lopinavir. We did not see a difference between darunavir and lopinavir in virological rebound which is consistent with observational studies of protease inhibitor monotherapy [6,8,25]. Two small trials have randomized patients to either lopinavir or darunavir monotherapy and reported no difference in virological failure [26,27]. In the absence of clear evidence for differences in viral load suppression, convenience, tolerability, and the potential for longer term toxicity may be the most important factors in the selection between darunavir and lopinavir for use as monotherapy. Of note, only one participant in PIVOT developed clinically significant resistance to a protease inhibitor taken as monotherapy, and this was on atazanavir [1].

There was no difference in risk for viral load rebound between participants on protease inhibitor and those on NNRTI at enrolment. In contrast to other studies, we did not see an association between HCV coinfection and virological rebound [10,11], although the number of coinfected patients in PIVOT was relatively small (5%) and limited to those not on HCV treatment or with stable HCV disease. We found nonwhite patients had a greater risk for viral load rebound. Ethnicity has not been examined in other protease inhibitor monotherapy trials, but there are similar findings in trials with combination ART. For example, a combined analysis of several HIV trials in the United States found a 40% higher virological failure risk in blacks than in whites that was not fully explained by demographic, clinical, socioeconomic, or adherence factors and was consistent over a wide range of regimens. It was suggested that it may be driven by unmeasured social factors, although biological factors could not be ruled out [28].

Our analysis has some limitations. First, patients who participate in randomized trials may not be representative of the general clinic population. In particular, the very low rate of virological rebound observed in the triple therapy arm suggests that PIVOT enrolled patients who are more adherent than average [1]. Second, as the predictive factors that we analysed (including protease inhibitor used as monotherapy) were not randomly allocated, the observed associations with viral load rebound are potentially (partly) explained by unmeasured confounders. Third, we have speculated that the effects of nadir CD4+ cell count and duration of viral load suppression are because of the size of the HIV reservoir. It would have been informative to have examined this directly (e.g. quantification of HIV DNA in peripheral blood mononuclear cells), but the appropriate samples were not collected in PIVOT.

In spite of the apparent lack of harm associated with short-term viral load rebound that was shown by the main analysis of the PIVOT trial, the strategy is nevertheless likely to have greater appeal to clinicians and patients if selection criteria can identify those with lower probability of rebound. We showed that patients who had suppressed viral load for several years prior to switch to monotherapy and have a higher CD4+ cell count nadir would be most suitable. Our findings may help to reassure both patients and clinicians that a strategy of protease inhibitor monotherapy (with prompt reintroduction of triple therapy in the event of viral load rebound) is an acceptable alternative to the management of chronic HIV infection. For patients who have remained without rebound on protease inhibitor monotherapy for about 18 months, our findings also provide reassurance that the good response is likely to be sustained in the longer term.

Acknowledgements

N.I.P. and D.T.D. designed the PIVOT trial. A.A.-P., C.O., A.C., I.W., M.J., N.J.B., and E.W. enrolled participants into the study. N.I.P., W.S., A.A.-P., K.S., and D.T.D. coordinated and oversaw the trial. W.S. and D.T.D. did the statistical analysis. All authors interpreted data. W.S., D.T.D., and N.I.P. drafted the report. All authors provided input into the report and approved the final version of the report.

We thank all the patients and staff from all the centres participating in the PIVOT trial, and the UK Community Advisory Board and African Eye for community support.

Controlled-Trials.com number: ISRCTN 04857074.

Funding: This work was supported by the National Institute for Health Research Health Technology Assessment programme (project number 06/403/90). The views and opinions expressed herein are those of the authors and do not necessarily reflect those of the HTA programme, NIHR, NHS, or the Department of Health.

The PIVOT Trial Team are:

Participating UK Sites:Elton John Centre, Brighton: Martin Fisher, Amanda Clarke, Wendy Hadley, and David Stacey. Royal Free Hospital, London: Margaret Johnson, and Pat Byrne. Mortimer Market Centre, London: Ian Williams, Nahum De Esteban, Pierre Pellegrino, Lewis Haddow, and Alejandro Arenas-Pinto. Barts & The London Hospital: Chloe Orkin, James Hand, Carl De Souza, Lisa Murthen, and Andrew Crawford-Jones. Royal Berkshire Hospital, Reading: Fabian Chen, Ruth Wilson, Elizabeth Green, and John Masterson. Manchester Royal Infirmary: Vincent Lee, Kamlesh Patel, and Rebecca Howe. St Mary's Hospital, London: Alan Winston and Scott Mullaney. Southmead Hospital, Bristol: Mark Gompels and Louise Jennings. Royal Liverpool University Hospital: Nicholas Beeching, Rebecca Tamaklo, and Paula Davies. Guys and St Thomas’ Hospital, London: Julie Fox, Alistair Teague, Isabelle Jendrulek, and Juan Manuel Tiraboschi. North Manchester General Hospital: Ed Wilkins, Yvonne Clowes, and Andrew Thompson. Central Middlesex Hospital: Gary Brook and Manoj Trivedi. Avenue House Clinic, Eastbourne: Kazeem Aderogba and Martin Jones. Gloucester Royal Hospital: Andrew DeBurgh-Thomas and Liz Jones. Homerton University Hospital, London: Iain Reeves and Sifiso Mguni. James Cook University Hospital, Middlesbrough: David Chadwick, Pauline Spence, and Nellie Nkhoma. Derriford Hospital, Plymouth: Zoe Warwick, Suzanne Price, and Sally Read. Royal Bournemouth Hospital: Elbushra Herieka, James Walker, and Ruth Woodward. Southend University Hospital: John Day, and Laura Hilton. St Mary's Hospital, Portsmouth: Veerakathy Harinda, and Helen Blackman. St George's Hospital, London: Phillip Hay, Wendy Mejewska, and Olanike Okolo. Royal Victoria Infirmary, Newcastle: Edmund Ong, Karen Martin, and Lee Munro. Royal Hallamshire Hospital, Sheffield: David Dockrell and Lynne Smart. North Middlesex University Hospital: Jonathan Ainsworth and Anele Waters. Queen Elizabeth Hospital, Woolwich: Stephen Kegg and Sara McNamara. Birmingham Heartlands Hospital: Steve Taylor and Gerry Gilleran. Chelsea & Westminster Hospital, London: Brian Gazzard and Jane Rowlands. University Hospital of Coventry: Sris Allan and Rumun Sandhu. Ealing Hospital, London: Nigel O’Farrell and Sheena Quaid. Harrogate District Hospital: Fabiola Martin and Caroline Bennett. Northwick Park Hospital: Moses Kapembwa. St James’ Hospital, Leeds: Jane Minton and James Calderwood. King's College Hospital, London: Frank Post, Lucy Campbell, and Emily Wandolo. Leicester Royal Infirmary: Adrian Palfreeman and Linda Mashonganyika. Luton & Dunstable Hospital: Thambiah Balachandran and Memory Kakowa. Newham University Hospital, London: Rebecca O’Connell and Cheryl Tanawa. Edith Cavell Hospital, Peterborough: Sinna Jebakumar and Lesley Hagger. Royal Victoria Hospital, Belfast: Say Quah and Sinead McKernan. York Teaching Hospital: Charles Lacey, Sarah Douglas, Sarah Russell-Sharpe, and Christine Brewer. Western General Hospital, Edinburgh: Clifford Leen and Sheila Morris. Barking Hospital, London: Sharmin Obeyesekera and Shirley Williams. Norfolk and Norwich University Hospital: Nelson David. Worcester Royal Hospital: Mark Roberts and Julie Wollaston.

MRC Clinical Trials Unit at UCL: Nicholas Paton, Wolfgang Stöhr, Alejandro Arenas-Pinto, Karen Scott, David Dunn, Emma Beaumont, Sue Fleck, Mark Hall, Susie Hennings, Ischa Kummeling, Sara Martins, Ellen Owen-Powell, Karen Sanders, Fionna van Hooff, Livia Vivas, and Ellen White.

Independent event reviewer: Brian Angus.

Trial Steering Committee: Andrew Freedman (Chair), Ben Cromarty, Danielle Mercey, Sarah Fidler, Estee Torok, Abdel Babiker, Brian Gazzard, Chloe Orkin, and Nicholas Paton.

Data Monitoring Committee: Tim Peto (Chair), David Lalloo, Andrew Phillips, and Robert James.

Conflicts of interest

There are no conflicts of interest.

References

1. Paton NI, Stohr W, Arenas-Pinto A, Fisher M, Williams I, Johnson M, et al Protease inhibitor monotherapy for long-term management of HIV infection: a randomised, controlled, open-label, noninferiority trial. Lancet HIV 2015; 2:e417–e426.
2. Mathis S, Khanlari B, Pulido F, Schechter M, Negredo E, Nelson M, et al Effectiveness of protease inhibitor monotherapy versus combination antiretroviral maintenance therapy: a meta-analysis. PLoS One 2011; 6:e22003.
3. Ripamonti D, Hill A, DeMasi R, Bicer C, Moecklinghoff C. Predictors of HIV RNA suppression on darunavir/ritonavir monotherapy or triple therapy in the MONET and PROTEA trials. Conference on Retroviruses and Opportunistic Infections, Seattle, USA, 2015.
4. Pulido F, Perez-Valero I, Delgado R, Arranz A, Pasquau J, Portilla J, et al Risk factors for loss of virological suppression in patients receiving lopinavir/ritonavir monotherapy for maintenance of HIV suppression. Antivir Ther 2009; 14:195–201.
5. Gutmann C, Cusini A, Gunthard HF, Fux C, Hirschel B, Decosterd LA, et al Randomized controlled study demonstrating failure of LPV/r monotherapy in HIV: the role of compartment and CD4-nadir. AIDS 2010; 24:2347–2354.
6. Curran A, Monteiro P, Domingo P, Villar J, Imaz A, Martinez E, et al Effectiveness of ritonavir-boosted protease inhibitor monotherapy in the clinical setting: same results as in clinical trials? The PIMOCS Study Group. J Antimicrob Chemother 2014; 69:1390–1396.
7. d’Arminio Monforte A, Gianotti N, Cozzi-Lepri A, Pinnetti C, Andreoni M, di Perri G, et al Durability of lopinavir/ritonavir monotherapy in individuals with viral load ≤50 copies/ml in an observational setting. Antivir Ther 2014; 19:319–324.
8. Guiguet M, Ghosn J, Duvivier C, Meynard JL, Gras G, Partisani M, et al Boosted protease inhibitor monotherapy as a maintenance strategy: an observational study. AIDS 2012; 26:2345–2350.
9. Lambert-Niclot S, Flandre P, Valantin MA, Peytavin G, Duvivier C, Haim-Boukobza S, et al Factors associated with virological failure in HIV-1-infected patients receiving darunavir/ritonavir monotherapy. J Infect Dis 2011; 204:1211–1216.
10. Arribas J, Pulido F, Hill A, Delft Y, Moecklinghoff C. Predictors of long-term HIV RNA suppression on darunavir/ritonavir monotherapy in the MONET trial. Int J STD AIDS 2013; 24:679–681.
11. Castagna A, Spagnuolo V, Galli L, Vinci C, Nozza S, Carini E, et al Simplification to atazanavir/ritonavir monotherapy for HIV-1 treated individuals on virological suppression: 48-week efficacy and safety results. AIDS 2014; 28:2269–2279.
12. Royston P, Parmar MK. Flexible parametric proportional-hazards and proportional-odds models for censored survival data, with application to prognostic modelling and estimation of treatment effects. Stat Med 2002; 21:2175–2197.
13. Lima VD, Bangsberg DR, Harrigan PR, Deeks SG, Yip B, Hogg RS, et al Risk of viral failure declines with duration of suppression on highly active antiretroviral therapy irrespective of adherence level. J Acquir Immune Defic Syndr 2010; 55:460–465.
14. Rosenblum M, Deeks SG, van der Laan M, Bangsberg DR. The risk of virologic failure decreases with duration of HIV suppression, at greater than 50% adherence to antiretroviral therapy. PLoS One 2009; 4:e7196.
15. Cameron DW, da Silva BA, Arribas JR, Myers RA, Bellos NC, Gilmore N, et al A 96-week comparison of lopinavir–ritonavir combination therapy followed by lopinavir–ritonavir monotherapy versus efavirenz combination therapy. J Infect Dis 2008; 198:234–240.
16. Antinori A, Clarke A, Svedhem-Johansson V, Arribas JR, Arenas-Pinto A, Fehr J, et al Week 48 efficacy and central nervous system analysis of darunavir/ritonavir monotherapy versus darunavir/ritonavir with two nucleoside analogues. AIDS 2015; 29:1811–1820.
17. Boulassel MR, Chomont N, Pai NP, Gilmore N, Sekaly RP, Routy JP. CD4 T cell nadir independently predicts the magnitude of the HIV reservoir after prolonged suppressive antiretroviral therapy. J Clin Virol 2012; 53:29–32.
18. Besson GJ, Lalama CM, Bosch RJ, Gandhi RT, Bedison MA, Aga E, et al HIV-1 DNA decay dynamics in blood during more than a decade of suppressive antiretroviral therapy. Clin Infect Dis 2014; 59:1312–1321.
19. Hong FF, Mellors JW. Changes in HIV reservoirs during long-term antiretroviral therapy. Curr Opin HIV AIDS 2015; 10:43–48.
20. Corbeau P, Reynes J. Immune reconstitution under antiretroviral therapy: the new challenge in HIV-1 infection. Blood 2011; 117:5582–5590.
21. Klatt NR, Chomont N, Douek DC, Deeks SG. Immune activation and HIV persistence: implications for curative approaches to HIV infection. Immunol Rev 2013; 254:326–342.
22. Amendola A, Marsella P, Bloisi M, Forbici F, Angeletti C, Capobianchi MR. Ability of two commercially available assays (Abbott RealTime HIV-1 and Roche Cobas AmpliPrep/Cobas TaqMan HIV-1 Version 2. 0) to quantify low HIV-1 RNA Levels (<1,000 copies/milliliter): comparison with clinical samples and NIBSC working reagent for nucleic acid testing assays. J Clin Microbiol 2014; 52:2019–2026.
23. Sollis KA, Smit PW, Fiscus S, Ford N, Vitoria M, Essajee S, et al Systematic review of the performance of HIV viral load technologies on plasma samples. PLoS One 2014; 9:e85869.
24. Brumme CJ, Swenson LC, Wynhoven B, Yip B, Skinner S, Lima VD, et al Technical and regulatory shortcomings of the TaqMan version 1 HIV viral load assay. PLoS One 2012; 7:e43882.
25. Santos JR, Llibre JM, Berrio-Galan D, Bravo I, Miranda C, Perez-Alvarez S, et al Monotherapy with boosted PIs as an ART simplification strategy in clinical practice. J Antimicrob Chemother 2015; 70:1124–1129.
26. Santos JR, Molto J, Llibre JM, Rosado DG, Bravo I, Muñoz-Moreno JA, et al Randomized study to evaluate lopinavir/ritonavir vs. darunavir/ritonavir monotherapies as simplification of triple therapy regimens. Kuala-Lumpur, Malaysia: IAS; 2013.
27. Pinnetti C, Lorenzini P, Cozzi-Lepri A, Sandrine O, Tommasi C, Zaccarelli M, et al Randomized trial of DRV/r or LPV/r QD monotherapy vs maintaining a PI/r-based antiretroviral regimen in persons with suppressed HIV replication. Glasgow, UK: HIV Drug Therapy Glasgow Congress; 2014.
28. Ribaudo HJ, Smith KY, Robbins GK, Flexner C, Haubrich R, Chen Y, et al Racial differences in response to antiretroviral therapy for HIV infection: an AIDS clinical trials group (ACTG) study analysis. Clin Infect Dis 2013; 57:1607–1617.
Keywords:

clinical trial; HIV; monotherapy; protease inhibitor; virological rebound

Copyright © 2016 Wolters Kluwer Health, Inc.