Dual Therapy Treatment Strategies for the Management of Patients Infected with HIV: A Systematic Review of Current Evidence in ARV-Naive or ARV-Experienced, Virologically Suppressed Patients

Objective We reviewed the current literature regarding antiretroviral (ARV)-sparing therapy strategies to determine whether these novel regimens can be considered appropriate alternatives to standard regimens for the initial treatment of ARV-naive patients or as switch therapy for those patients with virologically suppressed HIV infection. Methods A search for studies related to HIV dual therapy published from January 2000 through April 2014 was performed using Biosis, Derwent Drug File, Embase, International Pharmaceutical Abstracts, Medline, Pascal, SciSearch, and TOXNET databases; seven major trial registries, and the abstracts of major conferences. Using predetermined criteria for inclusion, an expert review committee critically reviewed and qualitatively evaluated all identified trials for efficacy and safety results and potential limitations. Results Sixteen studies of dual therapy regimens were critiqued for the ARV-naive population. Studies of a protease inhibitor/ritonavir in combination with the integrase inhibitor raltegravir or the nucleoside reverse transcriptase inhibitor lamivudine provided the most definitive evidence supporting a role for dual therapy. In particular, lopinavir/ritonavir or darunavir/ritonavir combined with raltegravir and lopinavir/ritonavir combined with lamivudine demonstrated noninferiority to standard of care triple therapy after 48 weeks of treatment. Thirteen trials were critiqued in ARV-experienced, virologically suppressed patients. The virologic efficacy outcomes were mixed. Although overall data regarding toxicity are limited, when compared with standard triple therapy, certain dual therapy regimens may offer advantages in renal function, bone mineral density, and limb fat changes; however, some dual combinations may elevate lipid or bilirubin levels. Conclusions The potential benefits of dual therapy regimens include reduced toxicity, improved tolerability and adherence, and reduced cost. Although the data reviewed here provide valuable insights into the effectiveness and tolerability of dual therapy regimens, it remains unclear whether these potential benefits can be maintained long-term. Appropriately powered studies with longer follow-up periods are needed to more definitively assess potential toxicity reduction advantages with dual therapy.


Introduction
In the late 1980s/early 1990s, the sequential use of nucleoside reverse transcriptase inhibitor (NRTI) monotherapy and dual therapies in patients with HIV infection rapidly led to treatment failure because of the emergence of resistance-associated mutations [1]. The use of combination antiretroviral therapy (cART) began in the mid-1990s, in which 2 NRTIs were combined with a third agent from a different therapeutic class. Current treatment guidelines continue the convention of preferred cART based on combining a dual NRTI backbone with a third "anchor" agent, such as a ritonavir (r)-boosted protease inhibitor (PI; PI/r), a non-nucleoside reverse transcriptase inhibitor (NNRTI), or an integrase inhibitor [2][3][4]. The toxicities associated with long-term use of NRTIs have led to the assessment of dual therapy approaches that do not include an NRTI component. A higher risk of treatment failure was observed in early NRTI-sparing studies compared with current standard triple therapy regimens [5][6][7]. Cohort studies suggest that patients with HIV are now living longer and are encountering an increased prevalence of comorbidities associated with natural aging, including renal, cardiovascular, or liver diseases; cognitive decline; metabolic disorders (diabetes and dyslipidaemia); and osteoporosis [8,9]. Drug-related adverse events (AEs) associated with the long-term use of antiretroviral therapy (ARV) may contribute to these comorbidities [10][11][12][13].
With the improved potency, tolerability, and durability of newer drugs and the higher barrier to the development of resistance, interest has re-emerged for ARV-sparing strategies, including monotherapy and dual therapies. These strategies have been applied as initial therapy in ARV-naive patients or as a switch strategy in those patients who have become virologically suppressed on standard regimens. Ideally, these regimens should achieve and maintain viral suppression and immunologic control while minimizing short-and long-term AEs, improve adherence and convenience, and reduce costs.
One well-studied therapeutic approach is the use of PI/r monotherapy following suppression with standard triple therapy. Although successful for a majority of patients, PI monotherapy was found to be associated with a statistically significant increased risk of virologic failure and an increased incidence of PI-associated resistance [14]. Although most failures were resuppressed by reinitiating NRTI therapies, this strategy is reserved for special circumstances.
Current guidelines do not include dual therapy regimens as a standard treatment strategy unless specific clinical characteristics (eg, comorbidities, pre-treatment viral load, and CD4 cell counts) of the individual patient warrant their use [2][3][4].
The objective of this report was to summarise data in the published literature regarding dual therapy approaches for treating ARV-naive patients and as a switch strategy for virologically suppressed patients on ARV therapy. We reviewed the literature from January 2000, with the approval of the first PI/r, until April 2014, in order to evaluate the efficacy of dual therapy regimens and the on long-term safety, AEs, and comorbidities associated with these regimens.

Methods
PRISMA guidelines for reporting systematic reviews were followed. The checklist is available as dreorting information (S1 PRISMA Checklist).

Search Strategy
ProQuest Dialog, Biosis, Derwent Drug File, Embase, International Pharmaceutical Abstracts, Medline, Pascal, and SciSearch databases were searched from January 2000 through April 2014 for studies related to HIV dual therapy. TOXNET was searched for AEs. Subject headings and keywords were tailored for each electronic resource using the following concepts: (atazanavir OR darunavir OR dolutegravir OR fosamprenavir OR indinavir OR lopinavir OR saquinavir) AND (efavirenz OR enfuvirtide OR etravirine OR lamivudine OR maraviroc OR nevirapine OR raltegravir OR rilpivirine OR saquinavir OR tenofovir OR tipranavir). The term "HIV dual therapy" was searched separately to capture potential combinations not explicitly stated above. The CRD/Cochrane Highly Sensitive Search Strategy [15] was used to restrict the research to randomised controlled trials in PubMed. For conference proceedings, we searched NLM Gateway

Eligibility Criteria
Eligible studies included randomised controlled or prospectively designed trials evaluating dual therapy combinations with a protease inhibitor (with or without r boosting [PI/r and PI, respectively]), an integrase inhibitor, an NNRTI, a CCR5 inhibitor, or lamivudine. A minimum 24-week duration of treatment in adults with HIV-1 infection who were ARV-naive or were switched after being virologically suppressed was required. Pilot/proof-of-concept studies and studies presented only as abstracts, which may not have been adequately powered, were also considered for inclusion. Case reports, reviews, correspondence, and research letters were excluded, as were phase 1, laboratory, pharmacokinetic/pharmacodynamic, prevention of vertical transmission, and retrospective studies or those including patients who were ARV-experienced but not suppressed, pediatric, or pregnant or patients who had a coinfection.
A primary outcome of suppression of viral load, change in viral load, or virologic failure (VF) was assessed; other outcomes were acceptable as a primary endpoint if they were supplemented by secondary endpoints that included the aforementioned criteria. Toxicity and/or comorbidity-related secondary outcomes were also evaluated.

Data Abstraction and Qualitative Data Synthesis
The primary endpoint was efficacy (achieved or maintained virologic suppression, usually defined as <50 copies mL of HIV-1 RNA). Intent-to-treat (ITT) analyses were preferentially reported; however, per-protocol or observed analyses were also permitted. Changes in CD4 cell counts, discontinuation rates and tolerability or toxicity, lipid levels, renal function, bone mineral density (BMD), and body fat redistribution were examined, whether these parameters were predefined or reviewed post hoc. Subanalyses of included studies were also examined. In the trials, patients were classified as ARV-naive or virologically suppressed, and the results were examined qualitatively based on efficacy and safety results, as well as power and other study design limitations. The study entry criteria for inclusion, endpoints (actual rates and definitions), and comparators are presented in tabular format due to the large number of studies summarised in this review (n = 29).

Results
Twenty-nine studies examining novel dual therapy regimens were included in the analysis ( Fig  1); 16 trials contained ARV-naive patients and 13 had ARV-experienced, virologically suppressed patients.

Trials in ARV-Naive Patients
Sixteen trials of novel dual therapy regimens in ARV-naive patients were included; of these trials (Fig 1), many were underpowered to confirm noninferiority of the strategy relative to standard of care. Trial designs and key findings from these studies are summarised in Tables 1,  2 and 3 and are discussed briefly by strategy

PIs in Combination With Raltegravir
Atazanavir + raltegravir therapy. The SPARTAN study [16] (a noncomparative pilot study; N = 93) compared atazanavir (ATV) + raltegravir (RAL) with ATV/r + tenofovir/emtricitabine (TDF/FTC) and was terminated early. Although the efficacy of ATV + RAL was similar to ATV/r + TDF/FTC, combination therapy with ATV + RAL was associated with a high rate of severe hyperbilirubinaemia. Despite a high dose of ATV (300 mg twice daily [BID]), the development of RAL resistance was seen in regimen failures.
Darunavir/r + RAL therapy. Results from studies evaluating darunavir (DRV)/r plus RAL are mixed. The RADAR study [17,18] (N = 80) provided some initial evidence for the effectiveness of DRV/r + RAL therapy; however, the 48-week results did not confirm the noninferiority of the DRV/r + RAL arm because of a higher level of discontinuations compared with the DRV/r + TDF/FTC arm. Likewise, in ACTG A5262 [19], DRV/r + RAL therapy demonstrated poorer than expected results in this single-arm phase 2 study. Among 112 patients, HIV-RNA levels were not suppressed in 11 patients and 6 and 11 patients rebounded by weeks 24 and 48, respectively, resulting in a 26% failure rate overall [19]. Although high baseline viral loads and poor adherence could have contributed to the poor results, the lack of a comparator arm limits the assessment of these data.
The NEAT 001 study [20], a large (N = 805) noninferiority randomised open-label study comparing the efficacy and safety of DRV/r in combination with either TDF/FTC or RAL, utilized time to treatment failure (virologic or clinical) as the primary endpoint. Per Kaplan-Meier methodology, therapy failure occurred in an estimated 17.4% of patients in the RAL arm and 13.7% in the TDF/FTC arm after 96 weeks (adjusted difference, 3.7% [95% CI, −1.1 to 8.6]), falling within a pre-specified noninferiority margin of 9%. In subgroup analyses, patients with CD4 counts <200 cells/μL had substantially higher rate of treatment failures using RAL therapy compared with TDF/FTC therapy. The reasons for the less than optimal treatment response in the non-NRTI arm in this subset of patients are being studied further. Although the number of virologic failures was low, 5 patients in the RAL arm developed integrase resistance, whereas no patients in the TDF/FTC arm had with PI resistance.

PIs in Combination With Maraviroc
Four studies assessed a PI/r in combination with maraviroc (MVC). The VEMAN (N = 50) and A4001078 (N = 121) studies demonstrated virologic suppression with LPV/r + MVC and ATV/r + MVC, respectively, in ARV-naive patients [26][27][28][29][30]. In the A4001078 study, the grade 3 and 4 elevations in bilirubin levels in patients treated with ATV/r + MVC (36.7%) versus   ATV/r + TDF/FTC (19.7%) were of concern. Ten patients in this study switched from ATV/r to DRV/r or LPV/r due to toxicity issues [26,30]. Whether the ATV/r + MVC regimen is associated with improvement with other toxicity issues will require further examination of between-group differences in bone and immune activation markers [28]. In the MIDAS study (DRV/r + MVC; N = 24), the rate of VF was high, especially in patients with higher baseline viral loads; the virus was not suppressed at 48 weeks in 16.7% patients (4/24), despite reported perfect adherence to therapy [31]. A fourth study, A4001095 [32] (MODERN; N = 812), was designed to assess therapy with DRV/r + TDF/FTC and DRV/r + once-daily MVC, and utilized 150 mg MVC with DRV/r 800/100 mg once daily. This study was terminated early because of inferior efficacy in the MVC arm [32]. A switch to this combination in patients with virologic suppression remains under study in the MARCH trial, which has been fully recruited; however, results have not yet been reported.

PIs in Combination With Lamivudine
Two studies examined a PI/r in combination with lamivudine (3TC). In a single-arm pilot study (LOREDA, N = 39) [33], the LPV/r + 3TC combination demonstrated moderate virologic efficacy (66.7% of patients with viral load <48 copies/mL, ITT; 81.2%, as treated); however, VF was high (13%) [33]. The GARDEL study (N = 426) was a randomised, controlled, and powered study that compared LPV/r + 3TC with LPV/r + 2 NRTIs [34]. LPV/r + 3TC was noninferior to standard triple therapy at 48 weeks regardless of baseline viral loads (<50 copies/mL: dual therapy, 88.3%; triple therapy, 83.7%; P = 0.171). There were fewer discontinuations in the LPV/r + 3TC arm largely because of safety and toxicity reasons. VF occurred at low levels in both treatment arms and did not result in any PI resistance in either arm. The M184V mutation was identified in 2 of 5 patients in the LPV/r + 3TC arm. Because the second NRTI in the triple therapy arm was most commonly zidovudine (ZDV), the generalizability of these results to all NRTIs may be limited. However, when the comparison was limited to non-ZDVcontaining regimens, noninferiority was confirmed.

Trials of ARV-Experienced, Virologically Suppressed Patients
Another approach to an ARV-sparing strategy is to suppress patient HIV-RNA levels using standard triple ARV regimens and then switch or simplify to a dual therapy regimen. Thirteen trials were identified that examined novel dual therapy regimens in ARV-experienced, virologically suppressed patients. Trial designs and key findings from these studies are summarised in Tables 4, 5

PIs in Combination With 3TC
Two studies of a PI/r in combination with 3TC were identified. In the single-arm ATLAS trial [45][46][47] (N = 40), simplification to therapy with ATV/r + 3TC demonstrated maintenance of virologic efficacy (<50 copies/mL) and no grade 4 laboratory toxicities or treatment interruptions due to the development of new laboratory toxicities at 96 weeks. In the SALT study [48] (N = 131), 87.5% of virologically suppressed patients who received ATV/r + 3TC maintained virologic suppression (<50 copies/mL) compared with 92.5% of patients who received ATV/r + 2 NRTIs (95% CI, −26.3% to 15.5%) at 24 weeks. The OLE study is an open-label prospective study evaluating LPV/r + 3TC or FTC compared with LPV/r + 2 NRTIs, which is ongoing through 48 weeks; results have not yet been reported [49].

PIs in Combination With NNRTIs
In A5116 [50] (N = 236), the combination of LPV/r + EFV was associated with increased toxicity-related discontinuations and a trend towards increased rates of VF (2 consecutive plasma HIV-1 RNA >200 copies/mL) compared with EFV + 2 NRTIs. In the NEKA study [51] (N = 31), in which virologically suppressed patients were switched to LPV/r + NVP or continued with LPV/r + 2 NRTIs, the proportion of patients who maintained virologic suppression were comparable, and LPV/r + NVP was generally well-tolerated over 48 weeks. Data regarding simplification of a 2-ARV regimen combining PI/r with rilpivirine, which could decrease pill burden and lessen toxicity, are not currently available.

Integrase Inhibitors in Combination With MVC
Two studies examined simplification to RAL in combination with MVC from a suppressive ARV [54] or RAL + MVC + TDF/FTC [55]. The ROCnRAL ANRS 157 study [54] (N = 44) was discontinued because of a high rate (n = 5) of VF (2 consecutive plasma viral load >50 copies/mL) after a median duration of 20 weeks. In the No Nuc No Boost study [55] (N = 10),  HIV-RNA levels remained <50 copies/mL in 9 of 10 patients following 44 weeks of treatment with RAL + MVC.

Secondary Endpoint Findings From Trials of ARV-Naive and ARV-Experienced, Virologically Suppressed Patients
The main rationale for dual therapy is to optimize health status and quality of life without compromising control of HIV infection. Toxicities and comorbidities may interfere with patient quality of life and negatively affect adherence to ARV regimens and can lead to additional costs for diagnosis and management. Thus, an ideal regimen would provide potent virologic suppression while conferring a lower risk of long-term, ARV-related AEs and would minimally affect any comorbidities that may be present.  Dual Therapy HIV Infection Treatment Strategies No significant differences between LPV/r + RAL (n = 37) and HAART (n = 18) in DEXA scans for BMD No significant differences between LPV/r + RAL (n = 37) and HAART (n = 18) in DEXA scans for total body fat composition HAART (n = 20) (Continued) No study thus far has been able to examine the impact of dual therapy on these outcomes in the long-term. Key secondary endpoint findings from studies identified in this report are summarised in Tables 3 and 6.

Renal Markers
Renal impairment, which is occasionally observed among individuals with HIV infection, may be related to chronic HIV infection, underlying comorbidities such as diabetes or hypertension, the use of medications, or a combination of factors [9,[56][57][58]. Cohort studies have demonstrated that cumulative exposure to some ARVs, for example, TDF, is associated with increased rates of chronic kidney disease or decreases in the estimated glomerular filtration rate (eGFR) [9,56,57]. Sparing such agents may help preserve renal function.
Few studies included in this report reported renal function as an outcome. In the PROG-RESS trial [24], a significantly greater decrease in eGFR from baseline was noted with LPV/r + TDF/FTC therapy than with LPV/r + RAL (−7.33 mL/min vs −1.43 mL/min, respectively; P = 0.035), potentially due to the presence of TDF. A similar result was observed in the NEAT 001 study [20], where a reduction in eGFR was significantly greater in the DRV/r + TDF/FTC arm compared with the DRV/r + RAL arm at week 96 (−3.8 vs +0.9 mL/min, respectively; Dual Therapy HIV Infection Treatment Strategies P = 0.02). Creatinine clearance at 48 weeks in the A4001078 study [26,28,30] was stable with ATV/r + MVC therapy, but decreased with ATV/r + TDF/FTC therapy; at week 96, creatinine clearance decreased by 5.5 mL/min and 18 mL/min with ATV/r + MVC and ATV/r + TDF/ FTC, respectively. In the SPARE study [42], there was no statistically significant difference between treatment groups in the number of patients who achieved a >10% improvement from baseline in eGFR with either DRV/r + RAL therapy or LPV/r + TDF/FTC. In ATLAS [45][46][47], eGFR was statistically increased versus baseline with ATV/r + 3TC (Cockroft-Gault equation, +6 mL/min, P<0.001 at 48 weeks; Modified Diet in Renal Disease equation, +15 mL/min, P<0.001 at 96 weeks). In the Reliquet 2014 study [52], serum creatinine improved in patients who switched to RAL + NVP therapy. The long-term clinical relevance of these statistical differences in renal parameters remains unexplored.

Changes in Lipid Parameters
Changes in lipid parameters could serve as markers for future cardiovascular disease (CVD), an important issue in the aging populations of patients with HIV [9]. Along with traditional risk factors, such as smoking, some ARV agents have been associated with CVD [13,[59][60][61].
Although NRTI-sparing dual therapy regimens may potentially improve lipid profiles and hence CVD risk, this question must be evaluated prospectively in the context of Framingham risk. Although the third agent included in triple therapy regimens may contribute to this risk, triple therapy regimens containing TDF tend to have more favourable lipid profiles compared with non-TDF regimens [18,19,21,23,24,34,36,37,44,50,62]. With the exception of the RADAR study (DRV/r + RAL), across the studies identified here, lipid abnormalities, especially elevations in triglyceride levels, were more frequent with dual therapy in CCTG 589, PROGRESS, KITE (all LPV/r + RAL), ACTG A5262 (DRV/r + RAL), ACTG 5142 and A5116 (both LPV/r + EFV) [18,19,21,23,24,36,37,44,50]. However, in the VEMAN (LPV/r + MVC) and SPAR-TAN (ATV + RAL) studies, the metabolic profile was found to be fairly stable between dual and triple therapy regimens [16,27,29]. In BMS-121 [35], the higher ATV dose was associated with a greater change in fasting triglycerides. The implications of these changes for CVD risk are uncertain. Interestingly, the positive impact of NVP on lipid profiles was maintained when used in combination with LPV/r [51]. Investigators postulated that the improved atherogenic profile of LPV/r + NVP therapy among ARV-experienced, virologically suppressed patients could help lower the risk of cardiovascular events [51]. In the Reliquet 2014 (RAL + NVP) and ROCnRAL ANRS157 (RAL + MVC) studies, ARV-experienced, virologically-suppressed patients experienced decreases in lipid parameters after the switch to dual therapy [52,54].

Hyperbilirubinaemia
Hyperbilirubinemia is frequently reported with ATV therapy. In some NRTI-sparing regimens, the rate of hyperbilirubinemia is increased, potentially limiting the use of ATV in dual therapy regimens [16,26,30]. However, total bilirubin levels were not elevated by ATV used in combination with 3TC in the ATLAS trial [45,47].

Bone Health
Osteoporosis is common in aging populations, especially in postmenopausal women. Thus, it is disconcerting that increased rates of BMD loss are observed in patients with HIV [63]. The relative contributions of HIV or its therapy to these observations remains uncertain. As has been noted in a number of recent studies, there is a decline in BMD of 2% to 6% with the initiation of ARV therapy [64]. This has been found to be more marked in those regimens in which TDF is included [65,66]. Because HIV treatment is continued over the long-term, the implications for bone health are worrisome. Few studies identified in this analysis examined the effect of dual therapy regimens on BMD. PROGRESS [24] evaluated the change in BMD through 96 weeks and markers of bone turnover. The mean change from baseline in BMD was −2.48% for the LPV/r + TDF/FTC arm compared with an increase of 0.68% for the LPV/r + RAL arm (P<0.001). Bone turnover markers C-terminal telopeptide (CTx) and osteocalcin increased in both groups to week 96, with greater increases in the LPV/r + TDF/FTC arm. Early changes in bone turnover markers were associated with a >5% BMD decrease at week 96, and increases at 4 weeks in CTx were associated with clinically significant bone loss. TDF had a greater impact on bone turnover and was associated with a higher incidence of clinically significant bone loss compared with RAL in combination with LPV/r. Similar positive findings on bone markers were observed in ATLAS, where there was a trend toward increased lumbar spine vertebrae 2 and 4 BMD in the ATV/r + 3TC arm [47]. The ROCnRAL study [67] noted a mean increase in BMD during a 26-week evaluation period, corresponding to an estimated increase of 2% per year. Although these data provide evidence that measurement of bone markers may be predictive of changes in BMD and may be useful for biotherapeutic agent evaluation, caution is warranted because of the small number of study subjects, the less specific imaging used, and the wide variation in bone marker measures.

Body Fat Redistribution
Lipoatrophy has frequently been associated with NRTIs; previous studies have shown that patients switched from zidovudine (AZT) and stavudine (d4T) experience improvements in peripheral fat. Therefore, a regimen that spares one or more NRTIs may improve or prevent lipoatrophy. However, compared with earlier NRTIs, the rate of lipoatrophy with newer NRTIs (eg, TDF and abacavir) appears to be low, making it unclear whether NRTI-sparing regimens will have much clinical impact. In the studies examined here, an increase in limb fat was observed in NRTI-sparing regimens compared with triple therapy regimens, both in ARVnaive and virologically suppressed patients [25,36,45,50].

Discussion
A total of 29 studies that evaluated the effectiveness of novel dual therapy regimens were summarised in ARV-naive (16 studies) or ARV-experienced, virologically suppressed patients (13 studies). They provide preliminary, short-term insights into the potential of dual therapy regimens for achieving and maintaining adequate virologic suppression in these populations. Whether long-term viral suppression and reduced toxicity will be achieved and maintained remains unclear. Given the small patient numbers and short durations of most of the existing studies, it remains unknown to what extent resistance will develop upon failure of a particular regimen, or if re-suppression will be possible with a switch or intensification of therapy. Certain dual therapy combinations seem to have insufficient efficacy or safety profiles to be recommended as alternative options to current standard of care, specifically in ARV-naive patients: ATV 300 mg BID + RAL 400 mg BID due to high rates of severe bilirubinaemia [16], DRV/r 800/100 mg QD + RAL 400 mg BID in patients with CD4 counts <200 cells/μL [19], DRV/r 800/100 mg + MVC 150 mg QD due to inferior efficacy [31], and LPV/r 533/133 mg BID + EFV 600 mg QD due to poor tolerability [36,37]. Long-term studies evaluating efficacy, safety, adherence, pill burden, and cost-effectiveness are required to more completely understand the clinical value of other dual therapy approaches.
Because they were randomised and sufficiently powered to demonstrate noninferiority of the viral response at 48 weeks in the dual therapy arm compared with triple drug therapy comprising a PI combined with 2 NRTIs, the NEAT 001 study, which compared DRV/r + RAL with DRV/r + TDF/FTC, the PROGRESS study, which compared LPV/r + RAL with LPV/r + TDF/FTC, and the GARDEL study, which compared an ARV-sparing regimen of LPV/r +3TC with triple therapy, provide the most definitive evidence to date. NEAT 001 and PROG-RESS assessed virologic efficacy up to 96 weeks [20,23,24], and all three studies found potential signals of reduced toxicity [23,24,34]. The GARDEL trial is further strengthened by the larger sample size, including a reasonable percentage of patients with a higher baseline viral load. To date, however, the findings of these studies have not been confirmed in second independent, adequately powered studies. In addition, no dual therapy regimens have been directly compared with each other in prospective, randomised controlled trials. As such, no definitive statement can be made regarding the dual therapy treatment strategy that would achieve the best virologic efficacy and safety outcomes in HIV-infected patients. In rare situations where NRTIs cannot be used owing to transmitted resistance or toxicity, the combinations of LPV/r + 3TC, LPV/r + RAL, or DRV + RAL could be considered as alternative options in treatmentnaive patients.
Because some studies have suggested that TDF use is linked to renal injury [68,69] and an increased cardiovascular risk may be associated with ABC (although this remains controversial) [70][71][72][73][74], avoiding their use in certain patients may be prudent. According to current treatment guidelines, dual therapy with LPV/r + 3TC, LPV/r + RAL, or DRV + RAL can be considered when regimens containing ABC or TDF are not recommended or are contraindicated because of patient comorbidities, such as the presence of cardiovascular risk factors or pre-existing renal disease and a positive HLA-B Ã 5701 test [2][3][4].
Carr et al [75] have cautioned that, although virologic noninferiority is an essential endpoint for simplification studies, it should not be the only endpoint, as virologic noninferiority alone is not a benefit. The disadvantages of existing ARV regimens, with respect to AEs, quality of life, cost, or other effects can only be addressed in a simplification study if the disadvantage is important, if the entry criteria are well-defined, and if there is adequate recruitment of at-risk participants to provide the statistical power that is required to yield clinically meaningful results. Furthermore, a clinically relevant endpoint is exponentially more meaningful than, for example, a statistically significant change in a laboratory parameter.
Unfortunately, most of the studies cited in this report have not adequately addressed the potential short-and long-term AEs with dual therapy regimens or demonstrated that the newer strategies provide important clinical benefits or are cost-effective in the long term. Taken as a whole, the potential benefits of dual therapy regimens for some organ systems (eg, kidney and bone) must be balanced against potential detriments in others (eg, CVD). Further, any negative effects of treatment may necessitate additional monitoring or treatment (eg, lipidlowering agents). Finally, medication-related changes in lipids and other markers must be considered in the context of traditional risk factors on disease outcomes (ie, smoking) in aging HIV-positive populations [76,77]. This information will be critical before dual therapy regimens gain widespread use.
Implementation of a dual therapy regimen requires consideration of potential drug-drug interactions with other agents used as part of the ARV regimen, as well as other prescription and over-the-counter drugs. For example, because of effects mediated through cytochrome P450 metabolism, coadministration of NVP or MVC with certain PIs or other classes of drugs (eg, some NNRTIs, antifungals, antivirals, and antibiotics) can affect the plasma concentration of both agents; thus, coadministration may not be recommended or dose adjustments may be required [78,79]. Before initiation of an ARV regimen in the individual patient, evaluation of concomitant medication use and pertinent prescribing information, as well as other available resources [80], should be consulted to assess for any potential drug-drug interactions.
Limitations of this review include the restriction of eligible trials to those that evaluated dual therapy regimens in either ARV-naive or ARV-experienced, virologically suppressed patients. An examination of studies that assess the ability of dual therapy regimens to provide virologic suppression in ARV-experienced patients with VF (salvage therapy) [81,82] were beyond the scope of this report. Furthermore, patient subgroups, such as individuals with hepatitis C virus coinfection, were not examined, nor were study results stratified by age or gender. The effectiveness of dual therapy regimens in ARV-naive patients with baseline HIV-1 RNA levels >100,000 copies/mL is inadequately investigated in most of the studies reported. Therefore, careful interpretation of the findings is warranted. An evaluation of the effectiveness of dual therapy regimens in diverse patient population is also required. A potential concern with the use of dual therapy is the persistence of viral replication in reservoir sites such as the central nervous system (CNS) [83,84]. The evaluation of dual therapy regimens regarding penetration and effectiveness in potential reservoir sites, and specifically, CNS viral escape and the predictive value of CNS penetration effectiveness scores, requires further study.
Preliminary data from some studies are encouraging; however, it is not currently possible to recommend widespread adoption of novel dual therapy regimens. Future trials demonstrating adequate long-term efficacy and safety are required before dual therapy regimens can be incorporated into routine clinical practice for ARV-naive and ARV-experienced, virologically suppressed patients. Other dual therapy regimens that do not include a boosted PI are currently being investigated, and initial results show promise. For example, the exploratory proof-ofconcept study using dolutegravir and lamivudine in ARV-naive subjects (PADDLE) showed no virologic failure in 20 patients after 24 weeks of therapy [85]. Another recent study conducted in ARV-experienced, virologically suppressed patients (LATTE) reported positive results using a dual regimen of an integrase inhibitor, cabotegravir, with rilpivirine [86]. Results from additional studies could lead to a shift in the treatment paradigm for HIV.