Emerging patterns and implications of HIV-1 integrase inhibitor resistance

Transcription

1 REVIEW C URRENT OPINION Emerging patterns and implications of HIV-1 integrase inhibitor resistance Anna Maria Geretti a, Daniele Armenia b, and Francesca Ceccherini-Silberstein b Purpose of review This review highlights recent data on the pathways of resistance that impact the clinical activity of firstgeneration and second-generation integrase inhibitors. Recent findings Raltegravir (RAL) and elvitegravir (EVG) are highly efficacious in first-line antiretroviral therapy, with small numbers of virological failures observed in clinical trials. Durable activity in treatment-experienced patients requires a fully supportive background regimen. RAL and EVG show a low-to-moderate genetic barrier to resistance and extensive cross-resistance, which preclude their sequential use. Resistance to dolutegravir (DTG) is not selected as readily in vitro and has not emerged in studies of treatment-naïve patients to date. Both in vitro and in vivo, DTG retains activity against several RAL and EVG resistant strains, but susceptibility is variably impaired by multiple mutations within the G148 pathway, which are common after RAL or EVG failure. Cross-resistance can be partially overcome by doubling DTG dosing to twice daily, but durability of responses remains dependent on a supportive background regimen. There is variability in the integrase gene of circulating HIV strains, which does not appear to reduce drug activity, although it may influence the emergence and evolution of integrase resistance. Transmission of integrase resistance remains rare but surveillance is required. Summary Integrase inhibitors provide a potent option for the treatment of HIV infection. Drug resistance remains a challenge, which may be partially overcome by the introduction of second-generation compounds. Prompt management of RAL and EVG failure is required to prevent the accumulation of multiple resistance mutations that reduce DTG susceptibility. Keywords genotype, HIV-1, integrase, phenotype, resistance INTRODUCTION The integrase enzyme is a 288 amino acid protein encoded by the 3 0 end of the HIV pol gene, which functions as a tetramer and is essential for viral replication. In 2007, raltegravir (RAL; MK-0518) became the first integrase inhibitor to receive approval for the treatment of HIV-1 infection. RAL potently inhibits HIV-1 and HIV-2 replication by blocking the strand-transfer reaction that catalyses integration of HIV DNA into the host genome. Two other integrase strand-transfer inhibitors (INSTIs), elvitagravir (EVG; GS-9137) and dolutegravir (DTG; S/GSK ), are expected to be soon approved for clinical use. Several potential second-generation INSTIs are also under development. RAL, EVG and DTG, although structurally diverse, bind to a common D 64 D 116 E 152 moiety in the integrase catalytic domain (amino acids ), causing it to disengage from the viral DNA (reviewed by Hare et al. [1] and Blanco et al. [2]). RAL and EVG offer a low-to-moderate genetic barrier to resistance and continued activity is dependent upon the presence of a supportive background regimen. Resistant strains emerge relatively rapidly both during in-vitro passage experiments and in patients a Institute of Infection and Global Health, University of Liverpool, Liverpool, UK and b Department of Experimental Medicine and Surgery, University of Rome Tor Vergata, Italy Correspondence to Professor Anna Maria Geretti, MD, PhD, FRCPath, Institute of Infection and Global Health, University of Liverpool, Liverpool L69 3BX, UK. Tel: ; geretti@liverpool.ac.uk Curr Opin Infect Dis 2012, 25: DOI: /QCO.0b013e32835a1de ß 2012 Wolters Kluwer Health Lippincott Williams & Wilkins

2 Antimicrobial agents KEY POINTS The integrase inhibitors raltegravir (RAL) and elvitegravir (EVG) potently suppress HIV replication but are vulnerable to the emergence of drug resistance, whereas dolutegravir (DTG) offers the promise of an improved genetic barrier to resistance. With both RAL and EVG, durable activity in treatmentexperienced patients requires support from other active drugs in the antiretroviral regimen. RAL and EVG show extensive cross-resistance precluding their sequential use. RAL and EVG resistant strains show variable levels of cross-resistance to DTG, which can be partially overcome by doubling the dose of DTG to twice daily and ensuring the presence of other active drugs in the antiretroviral regimen. Transmission of integrase resistance remains rare, but surveillance is warranted. experiencing viraemia during treatment. Single mutations in integrase are sufficient to confer high-level phenotypic resistance to the two drugs. Furthermore, compensatory mutations emerge commonly to restore the fitness of drug resistant strains. There is extensive cross-resistance between RAL and EVG, which precludes their sequential use. DTG is a second-generation INSTI that offers the promise of an improved genetic barrier and preserved activity against some RAL and EVG resistant strains [3 &,4,5], although further clinical validation from the ongoing phase 3 studies is awaited. RALTEGRAVIR AND ELVITEGRAVIR RESISTANCE IN CLINICAL TRIALS RAL, dosed twice daily, has shown potent and durable antiretroviral activity and a favourable safety profile in large phase 3 trials of treatment-naïve (STARTMRK study) and treatment-experienced (BENCHMRK study) patients [2,6,7]. In contrast, reduced virological efficacy was observed in treatment-naïve patients receiving once daily RAL (and tenofovir/ emtricitabine) compared with twice daily dosing, particularly when baseline plasma HIV-1 RNA levels exceeded copies/ml (QDMRK study) [8 ]. EVG is dosed once daily but requires boosting by either ritonavir or cobicistat. In phase 3 clinical trials, the coformulation of tenofovir, emtricitabine, EVG and cobicistat (QUAD) demonstrated noninferiority as first-line antiretroviral therapy (ART) when compared with either efavirenz (Study ) [9 ] or ritonavir-boosted atazanavir (Study ) [10 ], each in combination with tenofovir/emtricitabine. RAL has also been tested in phase 3 switch studies of patients with suppressed plasma HIV-1 RNA as a substitute for enfuvirtide (EASIER study) [11,12 ] or ritonavir-boosted lopinavir (LPV/r) (SWITCHMRK study; reviewed by Blanco et al. [2]. In contrast with the good to excellent responses observed in both BENCHMRK, which typically included a ritonavirboosted protease inhibitor (PI/r) (with or without enfuvirtide) in the background regimen, and EASIER, which included a PI/r in the background regimen, reduced virological responses were observed in ARTexperienced patients randomised to the RAL arm of SWITCHMRK, which included only nucleoside/ nucleotide reverse transcriptase inhibitors (NRTIs) in the background regimen. The suboptimal efficacy of RAL in SWITCHMRK was mostly confined to patients who had previously experienced treatment failure and were likely to harbour resistance to the NRTIs in the regimen. RAL and EVG resistant strains were readily detected in patients experiencing virological failure within these studies (Tables 1 3) [2,6,7,8 10, 12,13 & ]. Furthermore, in first-line ART studies with RAL or EVG, although failures were uncommon, patients with viraemia typically acquired resistance to both the INSTI and the NRTI backbone. Interestingly, the rates of NRTI resistance (to both emtricitabine and tenofovir) were increased in patients receiving EVG compared with those receiving efavirenz in Study [9 ]. Rapid loss of response with high resistance rates was reported in a phase 2 trial of ART-experienced patients receiving ritonavir-boosted EVG with a poorly active background regimen (reviewed by Blanco et al. [2]). However, in a phase 3 comparative trial, RAL and EVG, each in combination with a PI/r-based background regimen, showed similar efficacy in ART-experienced patients (Study 145) [13 & ]. A prematurely discontinued, open-label trial investigating first-line dual therapy with RAL and atazanavir (SPARTAN study) suggested a high incidence of RAL resistance at the time of virological failure (5/6 patients) [14]. Similarly poor outcomes were reported from a small phase 2 study of dual therapy with RAL and darunavir/ritonavir (800/100 mg once daily) (ACTG A5262 Study) [15]. These data have not, however, been consistent across studies. A small phase 3 trial exploring the dual combination of RAL and LPV/r (PROGRESS study) gave more promising results, however, with one of four viraemic patients developing RAL resistance (reviewed by Blanco et al. [2]). A substantial proportion of RAL failures nonetheless occur in the absence of detectable resistance Volume 25 Number 6 December 2012

3 Patterns of HIV-1 integrase inhibitor resistance Geretti et al. Table 1. Detection of integrase resistance mutations within selected clinical trials of raltegravir and elvitegravir Study Population Drug Resistance in tested patients n (%) Study week STARTMRK First-line RAL 4/16 (25) 192 QDMRK o.d. First-line RAL 9/27 (33) 48 QDMRK b.i.d. First-line RAL 2/12 (17) 48 BENCHMRK Experienced RAL 95/148 (64) 260 Study 145 Experienced RAL 15/72 (21) 48 Study 145 Experienced EVG 16/60 (27) 48 EASIER Switch RAL 3/39 (8) 48 SWITCHMRK Switch RAL 8/11 ((73) First-line EVG 7/14 (50) First-line EVG 4/12 (33) 48 Adapted from Rockstroh et al. [6 ]; DeJesus et al. [7]; Blanco et al. [2]; Eron et al. [8 ]; Molina et al. [13 & ]; Gallien et al. [12 ]; Sax et al. [9 ]; DeJesus et al. [10 ]. In a Spanish cohort treated with RAL, 50/89 (56%) patients lacked emerging integrase mutations at failure [16]. Plasma concentrations of RAL were significantly greater in patients with integrase mutations than in those without mutations, suggesting a correlation between drug exposure and selection of integrase resistance. In line with these observations, emergence of major integrase resistance mutations has been documented in patients experiencing low-level viraemia ( copies/ml) while on RAL-based ART [12 ]. PATHWAYS OF RESISTANCE TO RALTEGRAVIR AND ELVITEGRAVIR Integrase mutations that confer a RAL or EVG resistant phenotype have been identified both in vitro and in vivo. Several publicly available algorithms, such as those maintained by ANRS Table 2. Resistance analysis of patients enrolled into the STARTMRK trial week 192 a RAL EFV Randomised Genotypic resistance test b RAL-resistance 4/16 c EFV-resistance 7/14 FTC-resistance 6/18 5/14 TDF-resistance 0/18 1/14 EFV, efavirenz; FTC, emtricitabine; RAL, raltegravir; TDF, tenofovir. Adapted from Rockstroh et al. [6 ]; DeJesus et al. [7]. a Treatment-naïve patients were randomised to RAL or EFV each in combination with TDF/FTC. b Sequencing results not obtained for all patients. c The four patients with RAL resistance showed Q148HþG140S; Q148RþG140S; Y143HþL74L/MþE92QþT97A; and Y143R, respectively. ( REGA ( rega.kuleuven.be/cev/) and STANFORD ( hivdb.stanford.edu/dr/) aid the interpretation of integrase sequencing data in clinical practice, and continue to evolve as data emerge. HIV-1 resistance to RAL develops along three primary genetic pathways that involve the major mutations Y143R, Q148H/R/K, and N155H (Fig. 1), typically in combination with one or more additional mutations (Table 4) [2,3 &,17 &,18 20,21 ]. Although several mutations have been implicated, more or less convincingly, in integrase resistance and cross-resistance, strains from RAL-experienced patients show predominantly N155HþE92Q, Q148þG140, and Y143RþT97A mutations (Fig. 1). The combination of N155H þ Y143R is rare, but associated with very high phenotypic resistance to both RAL and EVG. Mutations seen in patients failing EVG include typically T66I/A/K, E92Q/G, T97A, S147G, Q148H/ R/K and N155H. The ranking order of the associated resistance effects is Q148R > E92Q ¼ N155H > T66I > S147G > T97A (Table 4). Importantly, the resistance profiles selected in vitro by the EVG metabolites M1 (EVG hydroxide) and M4 (EVG glucuronide) overlap with those of EVG [22]. There is extensive cross-resistance between RAL and EVG [23]. However, mutations at codon 143 affect RAL susceptibility to a much greater extent than susceptibility to other INSTIs, because only RAL makes direct interactions with this residue [1,17 &,24]. Conversely, the EVG resistance mutations T66I/A E92G, and S147G do not cause RAL resistance in isolation [17 & ]. Patients that show drug-specific mutants as dominant quasispecies may harbour low-frequency crossresistant strains that escape detection in routine genotypic testing but may emerge rapidly under selective drug pressure. Sequencing of RAL and EVG is not recommended regardless of the prevalent ß 2012 Wolters Kluwer Health Lippincott Williams & Wilkins 679

4 Antimicrobial agents Table 3. Resistance analysis of patients enrolled into the and trials week 48 a EVG EFV EVG ATV Randomised Genotypic resistance test EVG-resistance b 7/14 4/12 EFV-resistance 8/17 ATV-resistance 0 FTC-resistance 8/14 2/17 4/12 0 TDF resistance 3/14 2/17 1/12 0 ATV, atazanavir; EFV, efavirenz; EVG, elvitegravir; FTC, emtricitabine; TDF, tenofovir. Adapted from Sax et al. [9 ]; DeJesus et al. [10 ]. a Treatment-naïve patients were randomised to EVG versus ERV or EVG versus ATV each in combination with TDF/FTC. b EVG resistance mutations included E92Q, T66I, Q148R and N155H. resistance pattern detected at virological failure. Of note, despite a 40% difference in integrase genes between HIV-1 and HIV-2, the genetic pathways leading to RAL resistance are similar [25 & ]. The factors determining the evolution of integrase resistance pathways are being elucidated. Q148 mutations markedly decrease integrase function, but the defect is largely reversed by compensatory mutations at position G140 and, to a lesser extent, at position E138. The N155H mutation also decreases viral fitness, although less than Q148 mutations; the addition of E92Q to N155H further decreases drug susceptibility but does not improve viral fitness. Thus, although N155H may emerge first during virological failure, the G140/Q148 double mutant tends to replace N155H as the dominant species with prolonged drug selective pressure [26 & ]. Similarly, the addition of T97A to Y143C/R further reduces susceptibility to RAL and rescues the integrase catalytic function [27]. Presumably due to the effect of compensatory mutations, there appears to be no residual therapeutic benefit of maintaining RAL (and by extension EVG) in patients with major drug resistance mutations (reviewed by Vandamme et al. [28]). Interestingly, Q148H/R and N155H/S as single mutations appear to confer higher levels of RAL resistance in macrophages compared with those observed in CD4 T cells, suggesting that HIV-1 replication in macrophages facilitates the development of RAL resistance [29]. RESISTANCE TO DOLUTEGRAVIR Crystal structural studies suggest that DTG makes more intimate contacts with viral DNA than RAL and EVG, and has the ability to subtly readjust its Percentage Y143 Q148 N155 Others a None b FIGURE 1. Prevalence of the three major pathways of raltegravir (RAL) resistance detected at the time of virological failure within the BENCHMRK study. The trial recruited highly treatment-experienced patients with documented triple class drug resistance associated with mutations in reverse transcriptase and protease. Patients were randomised to RAL or placebo in combination with optimised background therapy. At the time of virological failure in the RAL arm most patients showed at least two integrase resistance mutations. a Other mutations recognised to play a role in raltegravir resistance (included L74M, E92Q, T97A, E138A, E138K, G140A, G140S, G163R, and S230R). b None indicates no recognised RAL resistance mutations were detected [2] Volume 25 Number 6 December 2012

5 Patterns of HIV-1 integrase inhibitor resistance Geretti et al. Table 4. Mutations in HIV-1 integrase associated with resistance to raltegravir and elvitegravir a Mutations by associated fold-change in phenotypic susceptibility 2 to 3 4 to 10 >10 to <100 >100 Other b RAL H51Y T66K G118R c E138A/Q148R V54I E92I E92Q/V Y143R E138K/Q148K T66A/I Y143C/H F121Y d Q148H/R/K E138K/Q148R L68I/V E157Q V151A/L N155H/S/T G140C/Q148R L74M/R T66I/L74M G140S/Q148K T66I/E92Q G140S/Q148H Q95K T66K/L74M G140S/Q148R T97A F121Y/T125K E92Q/N155H E138K/A E138K/Q148H Q148R/N155H G140S/A/C L74M/N155H V151I T97A/N155H M154I/L Y143H/N155H G163R/K N155H/G163K I203M N155H/G163R T206S N155H/D232N D232N EVG H51Y T66A/I T66K P145S e L68I/V T97A E92I/G E92Q/V T66I/E92Q L74M/R H114Y G118S/R c Q148R/K T66K/L74M Q95K Y143R S147G V151L E138A/Q148R A128T S153F/Y Q148H N155H/S/T E138K/Q148K E138K/A E157Q F121Y d T66I/L74M E138K/Q148R G140S/A/C R263K Q146R/P L74M/N155H G140C/Q148R V151I V151A E138K/Q148H T97A/N155H G140S/Q148H M154I/L F121Y/T125K G140S/Q148R G163R/K G140S/Q148K E92Q/N155H D232N Y143H/N155H Q148R/N155H N155H/G163K N155H/G163R N155H/D232N EVG, elvitegravir; RAL, raltegravir. Adapted from Blanco et al. [2]; Van Wesenbeeck et al. [17 & ]; Kobayashi et al. [3 & ]; Malet et al. [18]; Souza et al. [19]; Lampiris [20]; Quashie et al. [21 ]. a Data derived from the characterisation of clinical isolates, passaged virus and site-directed mutants. Data are evolving rapidly and the list may not be conclusive. Fold-changes may show small discrepancies in different studies. b Includes common accessory mutations that in isolation do not confer drug resistance but enhance resistance or viral fitness in combination with major resistance mutations, most commonly T97A (which often occurs with Y143 mutations), E138 and G140 mutations (which typically occur with G148 mutations), L74M/R, Q95K, V151I, G163R/K, S230R, and R263K. Some mutations are common polymorphisms. c G118R is selected in vitro by MK-2048 and occasionally by dolutegravir (in HIV-1 subtype C and CRF02_AG) [21 ] and has also been detected in a RALtreated patient infected with CRF02_AG and showing phenotypic resistance to RAL (fold change ¼ 25) and EVG (fold change ¼ 9) [18]. The site-directed mutant G118S confers approximately four-fold resistance to EVG (but not to RAL) [3 & ]. d Detected transiently in a RAL-treated patient prior to the emergence of Y143R [19]. e Markedly reduces susceptibility to EVG (but not RAL) [3 & ]. position and conformation in response to structural changes in the active site of RAL-resistant integrase [1]. DTG also shows slower dissociation than RAL and EVG from both wild-type and mutated integrase DNA complexes [30]. Selection of DTG resistance in vitro does not occur as readily as with RAL or EVG, and accumulation of multiple mutations is required to cause phenotypic resistance with foldchanges (FC) above 10. In-vitro passage experiments with DTG initially detected the emergence of T124A (day 14), S153F (day 28), and L101I (day 70) [3 & ]. The significance of these mutations is uncertain. Although integrase position 153 is conserved, positions 101 and 124 are polymorphic. The mutations, even in combination, do not confer significant phenotypic resistance to DTG (FC 1.9), RAL (FC 1.4) or EVG (FC 2) [3 &,4 ]. Other changes, similarly associated with small shifts in phenotypic ß 2012 Wolters Kluwer Health Lippincott Williams & Wilkins 681

6 Antimicrobial agents susceptibility, occur in vitro at codons E92 and G193 (reviewed by Quashie et al. [21 ]). More recently, in-vitro selection experiments with DTG using subtype B, subtype C, and CRF02_AG HIV-1 strains showed that the most common mutation to emerge after 20 weeks was R263K, which over time was replaced by S153Y in some cultures. The R263K site-directed mutant (SDM) conferred reduced susceptibility to DTG (FC ¼ 11) with a small effect on EVG but not RAL, and also reduced viral fitness [21 ]. G118R was the second most common substitution observed during in-vitro selection with DTG. G118R confers low-level resistance to both DTG and MK-2048 in vitro. One potential pathway of DTG resistance may include R263K (possibly with E138K as a compensatory mutation) in subtype B strains and G118R in subtype C strains [21 ]. DTG retains full activity in vitro against viruses with N155H (E92Q or Y143R) or Y143CR (T97A). However, susceptibility is reduced by about 10-fold to 20-fold by Q148 in combination with mutations at codons L74, E138 and G140 (Fig. 2) [3 &,4 ]. SDMs containing Y143R, Q148K, N155H, and G140S/ Q148H show fold-changes consistently less than 3 [3 & ]. With clinical isolates, the combination Q148H/G140S confers variable levels of DTG susceptibility due to the associated secondary mutations (Table 5) [4 ]. A similar impact of the Q148 mutational pathway has also been reported with other investigational INSTIs such as MK-2048 [17 & ]. A recent study analysed the genotypic and phenotypic susceptibility of 39 clinical isolates obtained from 18 patients that had exhibited incomplete viral suppression on a RAL-based regimen. The 30 samples with genotypic and phenotypic resistance to RAL (median FC >81; range, 3.7 to >87) showed levels of DTG resistance often close to those of wild-type variants (median FC 1.5; range, ). The median FC for isolates containing G140S þ Q148H; G140S þ Q148R; T97A þ Y143R; and N155H were 3.75, 13.3, 1.05, and 1.37, respectively [31 & ]. In clinical studies of ART-naïve patients, 10 days of DTG monotherapy did not result in the selection of resistant mutants (reviewed by Lenz and Rockstroh [32]; Katlama and Murphy [5]). A recent phase 2b dose-ranging trial explored the activity of once daily DTG (and tenofovir/emtricitabine or abacavir/lamivudine) compared with efavirenz in 205 ART-naïve adults (SPRING-1 study) [33 & ]. At week 16 (primary analysis) and week 48, proportions of patients with a plasma HIV-1 RNA load below 50 copies/ml were 93 and 97%, respectively (with little difference between DTG dose groups), T66K T66A/I E92Q/I/V G118S F121Y Y143R P145R Q146R Q148H/R/K V151L N155H N155S N155T T66K/L74M T66I/E92Q F121Y/T125K L74M/N155H T97A/N155H Y143H/N155H N155H/G163K N155H/G163R N155H/D232N E138K/Q148H E138A/Q148R E138K/Q148K E138K/Q148R G140C/Q148R G140S/Q148H G140S/Q148K G140S/Q148R E92Q/N155H Q148R/N155H Fold-change FIGURE 2. Phenotypic resistance (mean fold-change in EC 50 ) to dolutegravir assessed in vitro with molecular clones. Adapted with permission from Kobayashi et al. [3 & ] Volume 25 Number 6 December 2012

7 Patterns of HIV-1 integrase inhibitor resistance Geretti et al. Table 5. Phenotypic profiles of clinical isolates from raltegravir-experienced patients a Average fold-change Patient Weeks of RAL therapy Mutation RAL DTG 1 8 Y143K T97A E138A Y143K T97A Y143C Y143G L74M T97A Y143S T97A Y143S Y143R T97A Y143R G140S, Q148H G140S, Q148R G140S, Q148R G163R T112A G140S Q148H G163R V72I, N155H V54I, Y143R, N155H DTG, dolutegravir; RAL, raltegravir. Adapted from Canducci et al. [4 ]. a In patients 1, 2, 3, 4, and 6 clinical isolates were obtained from sequential time points. compared with 60 and 82% for the efavirenz arm. There were three virological failures in the DTG arm, but none showed integrase mutations. The small phase 2b VIKING trial studied the efficacy of DTG (50 mg once daily or twice daily) in 51 RAL-experienced patients with genotypic integrase resistance (all three pathways) (reviewed by Katlama and Murphy [5]). DTG either replaced RAL or was added to the existing failing regimen (in patients who had already discontinued RAL) for 10 days, followed by optimisation of the background regimen and follow-up to week 24. The primary endpoint was achieving a plasma HIV-1 RNA load below 400 copies/ml or at least 0.7log 10 reduction in viral load at day 11. Overall, at day 11 21/27 patients on once daily DTG and 23/24 patients on twice daily DTG met the primary endpoint. The mean plasma HIV-1 RNA load decline from baseline was 0.72 and 1.76 log 10 copies/ml in the two groups, respectively. At week 24, 41% of patients in the once daily arm and 75% of patients in the twice daily arm achieved a viral load below 50 copies/ml. The viral load declined significantly in patients harbouring the N155 or Y143 mutational pathways; however responses were reduced in patients with Q148 pathway mutations (Fig. 3), especially in the presence of Q148 mutations with mutations at positions 138 and 140 and E92Q or T97A. Detailed resistance data are yet to be presented, however, including a refined genotypic score and a possible clinical cut-off for the interpretation of DTG geotypic and phenotypic resistance respectively. Importantly, several RAL and EVG-associated mutations (including N155H) were seen to emerge in patients experiencing viraemia in VIKING, highlighting the potential for broad class crossresistance. No data have been presented on the sensitive assessment of preexisting and emergent drug resistant strains to assist with the interpretation of these findings. Based upon these considerations, prolonged virological failure during treatment with RAL or EVG should be avoided to prevent the emergence of strains with multiple HIV-1 RNA 0 decline (log10 copies/ml) All 1.45 Q N155 or Y = 27 = 9 = 18 FIGURE 3. Plasma HIV-1 RNA decline from baseline after 10 days of once daily dolutegravir in patients with preexisting raltegravir resistance (n ¼ 27). Adapted with permission from Lenz and Rockstroh [32] ß 2012 Wolters Kluwer Health Lippincott Williams & Wilkins 683

8 Antimicrobial agents integrase mutations as these show increasing resistance and cross-resistance [4 ]. There is promising evidence that DTG (dosed twice daily) will retain at least partial activity in patients with RAL or EVG experience. However, long-term responses will depend on the supporting activity of the background regimen. INTEGRASE MUTATIONS IN TREATMENT- NAIVE PATIENTS To date, the detection of major integrase resistance mutations in INSTI-naïve patients remains a very rare event when assessed by routine genotyping methods [34 38]. This is the case for both HIV-1 and HIV-2 infected patients [39,40]. A few cases of transmitted integrase resistance have been reported in ART-naïve HIV-1 infected patients, including the detection of Q148H and N155H in two patients, who subsequently failed their first-line RAL-containing regimen [41,42]. As use of INSTIs expands in clinical practice, it will be important to monitor the rates of transmitted drug resistance and adapt testing guidelines accordingly [43]. The occurrence of the key accessory mutations E138 and G140 is similarly rare in INSTI-naïve patients. Some accessory mutations (Table 4), some of the mutations seen during in-vitro passage with DTG (T124A and L101I), and several other integrase polymorphisms occur in INSTI-naïve patients infected with HIV-1 or HIV-2 [34 38,40]. By ultrasensitive testing, low-frequency integrase mutations are detected at variable prevalence; although major drug-resistance mutations are overall uncommon, accessory mutations are often detected [44 &,45 & ]. Importantly, even when multiple mutations are detected, they do not tend to coexist on the same viral genome [44 & ]. Although it remains important to monitor the activity of INSTIs against diverse HIV strains, there is no evidence to date that natural genetic variation in integrase affects clinical responses. In phase 3 studies in treatment-naive and treatment-experienced patients, RAL showed comparable efficacy against B and non-b HIV-1 subtypes [46]. In one study, patients experiencing virological failure on RAL were more likely to show low-frequency integrase mutations (both major and accessory) at baseline, but the difference did not achieve statistical significance, possibly due to the small study size [44 & ]. The preexistence of certain integrase variations, including accessory mutations such as T97A, V151I and G163R/K does not consistently relate to the mutational pathways detected at virological failure, which appear mostly dictated by considerations of overall resistance and viral fitness [44 &,45 & ]. Nonetheless, further studies are warranted to elucidate how such genetic variability may impact on drug susceptibility and genetic barriers. For instance, although baseline T124A and T124A/L101I do not appear to impair responses to DTG, the mutations are enriched in RAL-treated patients infected with HIV-1 subtype B (prevalence of 39% and 20%, respectively) compared with INSTInaïve patients (24% and 10%) and may thus influence the evolution of DTG resistance [37]. Furthermore, variations at integrase codon positions 140, 148, 151, 157, and 160 predict higher genetic barriers to G140S/C in HIV-1 subtypes C, CRF02_AG, and A/CRF01_AE, as well as higher genetic barriers toward acquisition of V151I in subtypes CRF02_AG and A/CRF01_AE [34]. Other subtype-specific differences have also been suggested [47]. CONCLUSION The first-generation INSTIs RAL and EVG are highly efficacious in ART-naïve patients and show durable activity in ART-experienced patients receiving a fully supportive background regimen. Whereas treatment failure is uncommon when RAL and EVG are used within a fully active regimen, patients who experience viraemia readily develop resistance mutations in integrase, which continue to accumulate with ongoing selective pressure causing high levels of phenotypic resistance and cross-resistance. DTG is a promising second-generation INSTI that shows an improved genetic barrier to resistance and both in vitro and in vivo retains activity against certain RAL and EVG resistant strains containing the Y143 and N155 pathway mutations. Clinical isolates with Q148 and additional integrase mutations (which are commonly observed in clinical practice) possess a broader range of susceptibility to DTG. Cross-resistance can be partially overcome by increasing the DTG dose, but long-term responses remain dependent on the presence of a supportive background regimen. Detailed resistance data, including the sensitive assessment of the spectrum of resistance mutations occurring at RAL or EVG failure and a refined genotypic score and clinical cut-off for the interpretation of DTG resistance data are awaited. Further clinical data are also required to confirm DTG robustness in INSTI-naïve patients. Meanwhile, patients experiencing viraemia with RAL or EVG should be managed promptly to avoid the accumulation of DTG cross-resistant mutants. Acknowledgements None Volume 25 Number 6 December 2012

9 Patterns of HIV-1 integrase inhibitor resistance Geretti et al. Conflicts of interest AMG: consultancy and speakers bureau for Merck, Gilead, GSK, ViiV, and research support from Merck and ViiV. FCS: consultancy and speakers bureau for Merck, Gilead, Abbott, and research support from Merck. D.A. has no conflicts of interest to declare in relation to this review. REFERENCES AND RECOMMENDED READING Papers of particular interest, published within the annual period of review, have been highlighted as: & of special interest of outstanding interest Additional references related to this topic can also be found in the Current World Literature section in this issue (pp ). 1. Hare S, Smith SJ, Metifiot M, et al. Structural and functional analyses of the second-generation integrase strand transfer inhibitor dolutegravir (S/GSK ). Mol Pharmacol 2011; 80: Blanco JL, Varghese V, Rhee SY, Gatell JM. HIV-1 integrase inhibitor resistance and its clinical implications. J Infect Dis 2011; 203: Kobayashi M, Yoshinaga T, Seki T, et al. In vitro antiretroviral properties of & S/GSK , a next-generation HIV integrase inhibitor. Antimicrob Agents Chemother 2011; 55: The study presents a detailed analysis of the resistance profile of DTG. 4. Canducci F, Ceresola ER, Boeri E, et al. Cross-resistance profile of the novel integrase inhibitor dolutegravir (S/GSK ) using clonal viral variants selected in patients failing raltegravir. J Infect Dis 2011; 204: The study details the evolution of integrase resistance and cross-resistance at multiple time points during RAL therapy, demonstrating increasing levels of phenotypic resistance to RAL and cross-resistance to DTG. The findings ihighlight the risk of maintaining RAL (and by extension EVG) therapy in treated patients who experience viraemia. 5. Katlama C, Murphy R. Dolutegravir for the treatment of HIV. Expert Opin Investig Drugs 2012; 21: Rockstroh JK, Lennox JL, Dejesus E, et al. Long-term treatment with raltegravir or efavirenz combined with tenofovir/emtricitabine for treatment-naive human immunodeficiency virus-1-infected patients: 156-week results from STARTMRK. Clin Infect Dis 2011; 53: Important study describing the durability of RAL-based first-line antiretroviral therapy. The investigators did not identify new cases of virological failure with integrase resistance after study week 96, although it should be noted that only patients with plasma HIV-1 RNA levels above 400 copies/ml underwent resistance testing. 7. DeJesus E, Rockstroh JK, Lennox JL, et al. Efficacy of raltegravir versus efavirenz when combined with tenofovir/emtricitabine in treatment-naïve HIV-1-infected patients: week-192 overall and subgroup analyses from STARTMRK. HIV Clin Trials 2012; 13: Eron JJ Jr, Rockstroh JK, Reynes J, et al. Raltegravir once daily or twice daily in previously untreated patients with HIV-1: a randomised, active-controlled, phase 3 noninferiority trial. Lancet Infect Dis 2011; 11: The large phase 3 clinical trial demonstrated that in first-line antiretroviral therapy, once daily RAL is inferior to twice daily dosing, especially in patients with high plasma HIV-1 RNA load at baseline. Most patients who experienced virological failure showed resistance to both RAL and emtricitabine. 9. Sax PE, DeJesus E, Mills A, et al. Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir versus co-formulated efavirenz, emtricitabine, and tenofovir for initial treatment of HIV-1 infection: a randomised, doubleblind, phase 3 trial, analysis of results after 48 weeks. Lancet 2012; 379: Large clinical trial demonstrating the efficacy and safety of EVG/colcibistat in firstline antiretroviral therapy. The resistance analysis performed at study week 48 among patients experiencing virological failure indicated a relatively high prevalence of dual class-resistance affecting EVG, emtricitabine and in a few cases tenofovir. 10. DeJesus E, Rockstroh JK, Henry K, et al. Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir disoproxil fumarate versus ritonavir-boosted atazanavir plus co-formulated emtricitabine and tenofovir disoproxil fumarate for initial treatment of HIV-1 infection: a randomised, double-blind, phase 3, noninferiority trial. Lancet 2012; 379: The large phase 3 trial demonstrates the noninferiority of EVG/cobicistat when compared with atazanavir/ritonavir as first-line antiretroviral therapy, with high rates of virological success and good tolerability in both arms. The resistance analysis at study week 48 showed higher overall rates of drug resistance in the EVG arm compared with the atazanavir arm. 11. Gallien S, Braun J, Delaugerre C, et al. Efficacy and safety of raltegravir in treatment-experienced HIV-1-infected patients switching from enfuvirtidebased regimens: 48 week results of the randomized EASIER ANRS 138 trial. J Antimicrob Chemother 2011; 66: Gallien S, Delaugerre C, Charreau I, et al. Emerging integrase inhibitor resistance mutations in raltegravir-treated HIV-1-infected patients with lowlevel viremia. AIDS 2011; 25: The study demonstrated that RAL resistance can be detected in patients experiencing low-level viraemia (up to 500 copies/ml) during therapy. This indicates that the exclusion of patients with low-level viraemia from the analysis can underestimate the prevalence of emerging drug resistance within clinical trials. 13. & Molina JM, Lamarca A, Andrade-Villanueva J, et al. Efficacy and safety of once daily elvitegravir versus twice daily raltegravir in treatment-experienced patients with HIV-1 receiving a ritonavir-boosted protease inhibitor: randomised, double-blind, phase 3, noninferiority study. Lancet Infect Dis 2012; 12: The study demonstrated that RAL and EVG are equally efficiacious in treatmentexperienced patients when combined with a fully supportive background regimen. 14. Kozal MJ, Lupo S, DeJesus E, et al. A nucleoside- and ritonavir-sparing regimen containing atazanavir plus raltegravir in antiretroviral treatment-naïve HIV-infected patients: SPARTAN study results. HIV Clin Trials 2012; 13: Taiwo B, Zheng L, Gallien S, et al. Efficacy of a nucleoside-sparing regimen of darunavir/ritonavir plus raltegravir in treatment-naive HIV-1-infected patients (ACTG A5262). AIDS 2011; 25: Garrido C, de Mendoza C, Alvarez E, et al. Plasma raltegravir exposure influences the antiviral activity and selection of resistance mutations. AIDS Res Hum Retroviruses 2012; 28: & Van Wesenbeeck L, Rondelez E, Feyaerts M, et al. Cross-resistance profile determination of two second-generation HIV-1 integrase inhibitors using a panel of recombinant viruses derived from raltegravir-treated clinical isolates. Antimicrob Agents Chemother 2011; 55: Elegant characterisation of the resistance profile of viral strains emerged during failure of RAL-based therapy. 18. Malet I, Fourati S, Charpentier C, et al. The HIV-1 integrase G118R mutation confers raltegravir resistance to the CRF02_AG HIV-1 subtype. J Antimicrob Chemother 2011; 66: Souza Cavalcanti J, Minhoto Lança A, de Paula Ferreira JL, et al. In-vivo selection of the mutation F121Y in a patient failing raltegravir containing salvage regimen. Antiviral Res 2012; 95: Lampiris HW. Elvitegravir: a once-daily, boosted, HIV-1 integrase inhibitor. Expert Rev Anti Infect Ther 2012; 10: Quashie PK, Mesplède T, Han YS, et al. Characterization of the R263K mutation in HIV-1 integrase that confers low-level resistance to the secondgeneration integrase strand transfer inhibitor dolutegravir. J Virol 2012; 86: Elegant article that demonstrates the selection of drug resistance during in-vitro passage with DTG and identifies R263K and G118R as signature DTG mutations. Detailed functional studies are presented. 22. Margot NA, Hluhanich RM, Jones GS, et al. In vitro resistance selections using elvitegravir, raltegravir, and two metabolites of elvitegravir M1 and M4. Antiviral Res 2012; 93: Garrido C, Villacian J, Zahonero N, et al. Broad phenotypic cross-resistance to elvitegravir in HIV-infected patients failing on raltegravir-containing regimens. Antimicrob Agents Chemother 2012; 56: Métifiot M, Vandegraaff N, Maddali K, et al. Elvitegravir overcomes resistance to raltegravir induced by integrase mutation Y143. AIDS 2011; 25: Charpentier C, Roquebert B, Delelis O, et al. Hot spots of integrase genotypic & changes leading to HIV-2 resistance to raltegravir. Antimicrob Agents Chemother 2011; 55: The study demonstrates that despite significant integrase sequence variability, HIV-1 and HIV-2 share similar genotypic pathways of RAL resistance. 26. Mukherjee R, Jensen ST, Male F, et al. Switching between raltegravir & resistance pathways analyzed by deep sequencing. AIDS 2011; 25: The study supports the concept that the evolution of RAL resistance during therapy is influenced by the overall phenotypic impact of resistance mutations on drug susceptibility and considerations of viral fitness. 27. Reigadas S, Masquelier B, Calmels C, et al. Structure-analysis of the HIV-1 integrase Y143C/R raltegravir resistance mutation in association with the secondary mutation T97A. Antimicrob Agents Chemother 2011; 55: Vandamme AM, Camacho RJ, Ceccherini-Silberstein F, et al. European recommendations for the clinical use of HIV drug resistance testing: 2011 update. AIDS Rev 2011; 13: Marsden MD, Avancena P, Kitchen CM, et al. Single mutations in HIV integrase confer high-level resistance to raltegravir in primary human macrophages. Antimicrob Agents Chemother 2011; 55: Hightower KE, Wang R, Deanda F, et al. Dolutegravir (S/GSK ) exhibits significantly slower dissociation than raltegravir and elvitegravir from wild-type and integrase inhibitor-resistant HIV-1 integrase-dna complexes. Antimicrob Agents Chemother 2011; 55: ß 2012 Wolters Kluwer Health Lippincott Williams & Wilkins 685

10 Antimicrobial agents 31. Underwood MR, Johns BA, Sato A, et al. The activity of the integrase inhibitor & dolutegravir against HIV-1 variants isolated from raltegravir-treated adults. J Acquir Immune Defic Syndr [Epub ahead of print] Elegant studies detailing the phenotypic profile of clinical isolates recovered from RAL-experienced patients. 32. Lenz JC, Rockstroh JK. S/GSK , a new integrase inhibitor for the treatment of HIV: promises and challenges. Expert Opin Investig Drugs 2011; 20: & van Lunzen J, Maggiolo F, Arribas JR, et al. Once daily dolutegravir (S/GSK ) in combination therapy in antiretroviral-naive adults with HIV: planned interim 48 week results from SPRING-1, a dose-ranging, randomised, phase 2b trial. Lancet Infect Dis 2012; 12: Important initial data on the activity of DTG in first-line combination antiretroviral therapy, providing the evidence base for phase 3 clinical trials. 34. Brenner BG, Lowe M, Moisi D, et al. Subtype diversity associated with the development of HIV-1 resistance to integrase inhibitors. J Med Virol 2011; 83: Cossarini F, Boeri E, Canducci F, et al. Integrase and fusion inhibitors transmitted drug resistance in naive patients with recent diagnosis of HIV- 1 infection. J Acquir Immune Defic Syndr 2011; 56:e51 e Garrido C, Soriano V, Geretti AM, et al. Resistance associated mutations to dolutegravir (S/GSK ) in HIV-infected patients impact of HIV subtypes and prior raltegravir experience. Antiviral Res 2011; 90: Malet I, Wirden M, Fourati S, et al. Prevalence of resistance mutations related to integrase inhibitor S/GSK in HIV-1 subtype B raltegravir-naive and -treated patients. J Antimicrob Chemother 2011; 66: Iamarino A, de Melo FL, Braconi CT, Zanotto PM. BF integrase genes of HIV-1 circulating in São Paulo, Brazil, with a recurrent recombination region. PLoS One 2012; 7:e Gottlieb GS, Smith RA, Dia B, Ba, et al. HIV-2 integrase variation in integrase inhibitor naïve adults in Senegal, West Africa. PLoS One 2011; 6:e Treviño A, de Mendoza C, Caballero E, et al. Drug resistance mutations in patients infected with HIV-2 living in Spain. J Antimicrob Chemother 2011; 66: Boyd SD, Maldarelli F, Sereti I, et al. Transmitted raltegravir resistance in an HIV-1 CRF_AG-infected patient. Antivir Ther 2011; 16: Young B, Fransen S, Greenberg KS, et al. Transmission of integrase strandtransfer inhibitor multidrug-resistant HIV-1: case report and response to raltegravir-containing antiretroviral therapy. Antivir Ther 2011; 16: Hurt CB. Transmitted resistance to HIV integrase strand-transfer inhibitors: right on schedule. Antivir Ther 2011; 16: & Liu J, Miller MD, Danovich RM, et al. Analysis of low-frequency mutations associated with drug resistance to raltegravir before antiretroviral treatment. Antimicrob Agents Chemother 2011; 55: Similarly to the study from Armenia et al. [45 & ] the investigators employed sensitive testing for the detection of integrase mutations and demonstrated that major and, to a greater extent, accessory drug resistance mutations can be detected at low frequency prior to the introduction of drug pressure. In this study, the prevalence of integrase mutations was higher in patients who experienced virological failure on subsequent RAL-based therapy compared with patients who achieved virological suppression, although the evidence for a significant impact of pre-existing lowfrequency mutants was not conclusive. 45. & Armenia D, Vandenbroucke I, Fabeni L, et al. Study of genotypic and phenotypic HIV-1 dynamics of integrase mutations during raltegravir treatment: a refined analysis by ultra-deep 454 pyrosequencing. J Infect Dis 2012; 205: The study employed an ultrasensitive assay with a lower limit of detection of 1% to detect integrase mutations. Results showed that the evolution of integrase drug resistance under selective pressure is not strictly correlated to the composition of the pre-treatment quasispecies, highlighting the impact of other virological factors (e.g., level of resistance and fitness impact of mutational pathways) and the possible role of stochastic events. 46. Rockstroh JK, Teppler H, Zhao J, et al. Clinical efficacy of raltegravir against B and non-b subtype HIV-1 in phase III clinical studies. AIDS 2011; 25: Nguyen HL, Ruxrungtham K, Delaugerre C. Genetic barrier to the development of resistance to integrase inhibitors in HIV-1 subtypes CRF01_AE and B. Intervirology 2012; 55: Volume 25 Number 6 December 2012