Secondary Logo

Journal Logo

Ex-vivo antiretroviral potency of newer integrase strand transfer inhibitors cabotegravir and bictegravir in HIV type 1 non-B subtypes

Neogi, Ujjwala,*; Singh, Kamalendraa,b,c,*; Aralaguppe, Shambhu, G.a; Rogers, Leonard, C.b; Njenda, Duncan, T.a; Sarafianos, Stefan, G.b,c; Hejdeman, Bod; Sönnerborg, Andersa,e

doi: 10.1097/QAD.0000000000001726

Objective: To determine the antiretroviral activity of the integrase strand transfer inhibitors (INSTIs), raltegravir (RAL), elvitegravir (EVG), dolutegravir (DTG), cabotegravir (CAB) and bictegravir (BIC), against different subtypes as well as primary and acquired drug resistance mutations (DRMs) in a patient-cohort infected with diverse subtypes.

Design: Biochemical and virological drug sensitivity analyses using patient-derived HIV type 1 (HIV-1) genes and cross-sectional/longitudinal clinical study.

Methods: Assays for 50% inhibition of 3′-end processing (IC50-3EP), strand transfer (IC50-ST) and drug sensitivity for five INSTIs were done using patient-derived integrase or gag-pol genes from subtypes A1, B, C, 01_AE and 02_AG. Integrase from INSTI-naive (n = 270) and experienced (n = 96) patients were sequenced.

Results: RAL had higher IC50-ST than the other INSTIs for all subtypes. EVG had higher IC50-ST for HIV 1 subtype C (P < 0.05) and 02_AG (P < 0.05) than HIV 1 subtype B (HIV-1B). DTG showed lower IC50-ST in HIV 1 subtype C than HIV-1B (P = 0.003). In CAB , the non-B subtypes showed lower IC50-ST (P < 0.05) than HIV-1B. In BIC, lower IC50-ST in 01_AE (P = 0.017) and 02_AG (P = 0.045) than HIV-1B. In drug sensitivity assay, inhibiting virus replication by 50% for DTG [median (IQR) 2.14 (1.3–2.56)], CAB [1.68 (1.34–2.55)] and BIC [1.07 (0.22–2.53)] were lower than RAL and EVG. One patient had a primary DRMs (0.3%, 1/270), but 17 (6.3%) had one major accessory DRM, of which 12 were E157Q.

Conclusion: The equal or higher potency in non-B subtypes of DTG, CAB and BIC compared with RAL and EVG confirms their suitability for use in countries dominated by non-B subtypes. Any impact of the high prevalence of major accessory mutations, especially E157Q, requires long-term follow-up studies.

aDivision of Clinical Microbiology, Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden

bDepartment of Molecular Microbiology and Immunology

cBond Life Sciences Center, University of Missouri, Columbia, Missouri, USA

dDepartment of Infectious Diseases, Gay Men Health, South Hospital

eUnit of Infectious Diseases, Department of Medicine Huddinge, Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden.

*Ujjwal Neogi and Kamalendra Singh contributed equally in the article.

Correspondence to Ujjwal Neogi, Division of Clinical Microbiology, Department of Laboratory Medicine, Karolinska Institutet, Huddinge 14186, Stockholm, Sweden. Tel: +46 85 248 3680; e-mail:

Received 3 October, 2017

Revised 30 November, 2017

Accepted 4 December, 2017

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

Back to Top | Article Outline


Integrase strand transfer inhibitors (INSTIs) of the HIV type 1 (HIV-1) have high potency, durable efficacy and low toxicity [1]. Three approved INSTIs, raltegravir (RAL), elvitegravir (EVG) and dolutegravir (DTG) have been approved by FDA and European Medicines Agency. INSTIs, bictegravir (BIC) and cabotegravir (CAB), are currently in phase III studies [2,3]. The INSTIs DTG and BIC exhibit longer dissociation half-life in biochemical analyses of wild-type integrase/DNA complexes, resulting in a high genetic barrier to resistance [4,5]. The high efficacy in clinical trials [6,7] and in clinical care has prompted an increasing use of DTG in first-line antiretroviral therapy (ART).

The efficacy of the newer INSTIs (CAB and BIC) across various HIV-1 subtypes in vitro and ex vivo is however not well documented [8]. Similar to other antiretrovirals, the development of INSTIs was based to a large extent on clinical and virological data from HIV-1 subtype B (HIV-1B). Although HIV-1 subtype has not been considered as a predictor for ART failure, an AIDS Clinical Trial Group (ACTG) trial indicated HIV-1 subtype C (HIV-1C) as an independent predictor for viral treatment failure [9], similar to our study of the Swedish InfCare HIV cohort in which a higher risk of viral failure was found in HIV-1C patients than in HIV-1B, despite developed clinical care, modern laboratory monitoring and focused adherence support performed at highly HIV-specialized infectious disease clinics [10]. Also, subtype-specific differences in drug resistance to first generation INSTIs have been reported [8] as well as in DTG cross-resistance pattern in patients failing RAL [11]. Further sequence and structure-based analyses showed that the subtype-specific effects were caused by polymorphisms across the subtypes, which significantly affected native protein activity, structure, function of the drug-mediated inhibition of the enzyme [12].

The current study is aimed to analyze if subtype-specific differences exist between INSTIs (RAL, EVG, DTG, CAB and BIC) with regard to in-vitro and ex-vivo antiretroviral potency, using biochemical and virological assays. Furthermore, the presence of primary and major accessory drug resistance mutations (DRMs) as well as acquired DRMs was investigated in patients from Sweden to delineate any subtype-specific differences.

Back to Top | Article Outline

Material and methods

Cell lines, plasmids and antiretrovirals

TZM-bl cells, plasmid pNL43 and RAL were obtained from the NIH AIDS Reagent Program. 293T cells were purchased from ATCC (Manassas, Virginia, USA). The cells were maintained in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma, St. Louis, Missouri, USA), supplemented with 10% Fetal Bovine Serum and 2-mmol/l L-glutamine. EVG (Selleck-S2001) and DTG (Selleck-S2667) were purchased from Selleckchem (Selleck Chemicals, Munich, Germany). CAB and BIC were kindly provided by ViiV Healthcare/GSK (Research Triangle Park, North Carolina, USA) and Gilead Sciences (Foster City, California, USA), respectively.

Back to Top | Article Outline

Clinical specimens and data

Stored plasma samples, used in the ex-vivo studies, were randomly selected from 24 INSTI-naive adult patients from Stockholm, Sweden [10]. The viruses were pure subtypes (HIV-1B n = 6; HIV-1C n = 14; A1 n = 1) or circulating recombinant forms (CRFs) [01_AE (n = 1); 02_AG (n = 2)], as determined by near full-length sequencing using our earlier published protocol [13].

In addition, integrase sequences, which had been obtained within routine clinical care through direct population sequencing of INSTI-naive (n = 270) or INSTI-experienced (n = 96) patients were included for analysis of primary, major accessory or acquired DRM. Among the INSTI-experienced patients, 74% (71/96) were on RAL, 7% (7/96) on EVG and 17% (16/96) on DTG, whereas two were experienced with both RAL and DTG. The resistance interpretation for RAL, EVG and DTG were based on the Stanford HIV resistance database [14]. For CAB, the presence of E138A/Q148R, E138K/Q148K, E138K/Q148R, G140C/Q148R, G140S/Q148R or Q148R/N155H was interpreted as high-level resistance [2]. If a virus contained only one of these mutations, it was assigned as potential low-level resistance. For BIC, the presence of E138K/Q148R, G140S/Q148R or N155H/Q148R was interpreted as high-level resistance [3]. If a virus contained only one of these mutations, it was assigned as potential low-level resistance. In addition, if the M50I/R263K mutation was present, we termed it as intermediate resistance to BIC [3].

Ethical permissions were obtained from the Regional Ethics Committee Stockholm (Dnr: 2006/1367-31/4 and 2014/928-31/2).

Back to Top | Article Outline

Cloning and purification of integrase

The complete coding region of the integrase genes (HXB2: 4230-5096) of four randomly selected INSTI-naive patients infected by HIV-1B, C, CRF01_AE and CRF02_AG, respectively, were cloned into the pRSFDuet-integrase vector with N-terminal H6-tag. Recombinant integrase proteins were expressed in BL21-DE3-RIL cells. The expression of the IN protein was induced by addition of 1-mmol/l IPTG in incubator-shaker for 3 h at 37 °C. The cells were harvested and resuspended in 20-mmol/l Tris–HCl pH 8.0, 1-mol/l NaCl, 4-mmol/l 2-Mercaptoethanol (βME), 5-mmol/l imidazole, 10-mmol/l 3-[(3-cholamidopropyl) dimethylammonio]-1-propanesulfonate hydrate, 1-mmol/l phenylmethylsulfonyl fluoride and 0.15-mg/ml lysozyme. The homogenized and sonicated cells were centrifuged, and the IN proteins were purified by Ni-affinity chromatography.

Back to Top | Article Outline

Inhibition of 3′-end processing assay

To assess the 3′-end processing activity of integrase (IN) in the presence or absence of INSTIs, a 21-mer oligonucleotide was annealed to a complementary 5-32P-labeled 21-mer oligonucleotide that contained a 3′-end-CAGT sequence. The endonucleolytic cleavage of 3′-end GT, resulting in an exposed CA-dinucleotide at the 3′-end, was conducted by incubating 300-nmol/l IN, 2-nmol/l DNA substrate, in a buffer containing 20-mmol/l MOPS (3-(N-morpholino)propanesulfonic acid), pH 7.2, 25-mmol/l NaCL, 1-mmol/l MnCl2 and 3-mmol/l βME. The cleavage products were resolved on a 16% polyacrylamide 8-mol/l urea gel. The cleavage of the GT-dinucleotide resulted in a 19-mer product. The reactions were allowed to proceed for different times and stopped by adding of Sanger's gel loading solution (95% formamide, 0.01% bromophenol blue, 0.01% xylene cyanol and 25 mmol/l ethylenediaminetetraacetic acid). The fraction of cleaved GT-dinucleotide was calculated by measuring the 19-mer products, which were then plotted as a function of time. A time point representing the linear range of 3′-end processing reaction (90 min) was selected, and the 3′-endonuclease activity of INs was monitored with increasing concentrations of INSTIs (0–100 nmol/l). The gels were exposed to PhosphoImager (GE Healthcare, North Richland Hills, Texas, USA), scanned on Typhoon FLA 700 (GE Healthcare), and the radioactive bands were quantitated by ImageQuant 5.2 (GE Healthcare). The amount of 3′-end excision products (19-mer) was plotted as the function of INSTI concentration, and the data were fit to a dose–response curve to estimate inhibition of 3′-end processing assay (IC50–3P) (three replicates).

Back to Top | Article Outline

Inhibition of strand-transfer activity assay

The inhibition of strand transfer activity in the presence INSTIs was assessed in gel-based assay. This assay separates the products of the strand-transfer reaction, which are longer than the 5′-32P-labeled 19-mer substrate. Briefly, a 32P-5′-end-labeled 19-mer single-stranded DNA (ssDNA) was annealed to a complementary 21-mer ssDNA. The 19-mer substrate has a CA-dinucleotide exposed for the INST reaction. Three hundred nmol/l IN was first incubated with 2 nmol/l 32P-5′-end-labeled 19-mer for 15 min followed by addition of 500 nmol/l unlabeled 31 base-pair long double-stranded DNA and 1 mmol/l MnCl2. The reaction was allowed to proceed for 90 min in the presence of increasing concentrations of INSTIs (0–25 nmol/l) followed by stopping by Sanger's gel loading solution. The products were resolved on a 16% polyacrylamide 8 mol/l urea gel. The gels were exposed to PhosphoImager (GE Healthcare), scanned and quantitated as described above. The radioactive bands corresponding to products of greater than 19-mer were results of the IN strand-transfer reaction. These bands were quantitated and plotted as the function of INSTI concentration. The data were fitted to a dose–response curve to determine inhibition of strand-transfer activity assay (IC50-ST) of IN proteins for strand-transfer reaction (three replicates).

Back to Top | Article Outline

Cloning and recombinant virus production

HIV-1 RNA was extracted using the QIAamp Viral RNA Mini Kit (Qiagen, Germantown, Maryland, USA) from 140 μl of patients’ plasma. The complementary deoxyribonucleic acid was generated using the SuperScript III RT enzyme (Invitrogen, Life Technologies, Woburn, Massachusetts, USA) with gene-specific primer 6231R, as described by us (Thermo Scientific, Waltham, Massachusetts, USA) [15]. The first-round PCR was performed using the high-fidelity KAPA HiFiHotStart Ready Mix (2×) (KAPA Biosystem, Wilmington, Massachusetts, USA) with primers 0682F and 6231R primers. The second-round PCR was performed using 0702F-BssHII and 5798R-SalI which has BssHII and SalI restriction sites, respectively. The amplified product was restriction digested followed by gel purification using the QIAquickGel Extraction Kit (Qiagen). The gag-pol fragment (HXB2 : 0702-5798) was cloned in pNL43Δgag-pol plasmid following digestion with BssHII and SalI (New England Biolab, Ipswich, Massachusetts, USA) and ligation with T4 DNA ligase (New England Biolab). The recombinant viruses were produced by transfecting the plasmids into the 293T cell line using FuGENE HD Transfection Reagent (Promega, Madison, Wisconsin, USA). All the molecular clones were sequenced bidirectionally. No primary DRM was observed in any of the clones. M50I was present in seven of the sequences from the HIV-1B (n = 3) and the HIV-1C (n = 4) strains.

Back to Top | Article Outline

Drug sensitivity assay

The drug sensitivity assay (DSA) was done by determining to which extent the five INSTIs inhibited the replication of our 24 recombinant viruses in the TZM-bl cells. Briefly, serial dilutions of the drugs spanning from 10 to 0.0001 μmol/l were added in triplicate in 96-well plates in complete DMEM media containing TZM-bl cells, followed by infection with a reference virus (NL43) or the corresponding patient-derived recombinant viruses (HIV-1B: n = 6; HIV-1C: n = 14; HIV-1A1: n = 1; HIV-1CRF_01AE: n = 1; HIV-1CRF_AG: n = 2) at a multiplicity of infection of 0.05 IU/cells in the presence of 10 μg/ml concentration of diethylaminoethyl. Virus replication was quantified by measuring Renilla luciferase activity (relative light units) using Bright-Glo Luciferase Assay System (Promega). Drug concentrations required for inhibiting virus replication by 50% (EC50) were calculated by a dose–response curve using nonlinear regression analysis (GraphPad Prism, version 5.01; GraphPad Software, La Jolla, California, USA). Each assay was performed using two to four biological replicates in triplicate.

Back to Top | Article Outline

Statistical analysis and data visualization

Statistical analysis was performed using an in-house R script or in GraphPad Prism, version 5.01 (GraphPad Software). The clustering analysis of the EC50 fold-drug change of the patients-derived recombinant viruses against the control virus NL4-3 was performed using Qlucore Omics Explorer version 3.2 (Qlucore AB, Lund, Sweden). Comparisons of the EC50 were conducted by the either Wilcoxon signed-rank test (for paired data), student t test and Mann–Whitney U test (for unpaired data) (between two groups) or analysis of variance (ANOVA) using the Kruskal–Wallis test by ranks (among three groups) where appropriate. A P value less than 0.05 considered significant.

Back to Top | Article Outline


In-vitro inhibition of strand transfer and 3′-end processing activities

The IC50-ST for strand transfer inhibition of RAL, EVG, DTG, CAB and BIC is shown in Fig. 1a. RAL had higher IC50-ST than the other INSTIs in all subtypes (P < 0.001). The drug-specific comparison showed that RAL had comparable potency (IC50-ST) for 01_AE and 02_AG but better potency for HIV-1C (P = 0.047) when compared with HIV-1B. In contrast, EVG had relatively higher IC50-ST for HIV-1C (P = 0.016) and 02_AG (P = 0.02). DTG showed lower IC50-ST for HIV-1C than HIV-1B (P = 0.003). For CAB, the non-B subtypes showed lower IC50-ST (P = 0.043, 0.027 and 0.026 for HIV-1C, 01_AE and 02_AG, respectively) compared with HIV-1B. For BIC, no difference was found between HIV-1B and HIV-1C, but a lower IC50-ST was found for 01_AE and 02_AG (P = 0.017 and 0.045, respectively) compared with HIV-1B.

Fig. 1

Fig. 1

INSTIs also inhibit 3′-end processing activity albeit at a lower efficiency than strand transfer activity. Hence, we determined IC50–3P for all the IN against all the drugs (Fig. 1b). All INSTIs had a ∼10-fold higher IC50–3P for the 3′-end processing reaction compared with the strand transfer reaction.

Back to Top | Article Outline

Drug sensitivity assay

The ex-vivo DSA was performed using recombinant HIV-1B (n = 6), HIV-1C (n = 14) and A-like (A1, 01_AE and 02_AG; n = 4) viruses, derived from INSTI naive patients (Fig. 2a). When the results of all recombinant viruses were analyzed, the EC50 for RAL and EVG {RAL: median [interquartile range (IQR)]: 5.31 (1.56–6.70)}; EVG: 2.82 (1.69–5.39)] were significantly higher (P < 0.05, for all comparisons) than those for DTG, CAB and BIC [DTG 2.14 (1.3–2.56); CAB 1.68 (1.34–2.55); BIC 1.07 (0.22–2.53)]. The EC50 of BIC were lower than for DTG (P = 0.01). No difference was observed between BIC and CAB. Clustering analysis of the fold-change values identified specific clustering for RAL and EVG vs. other INSTIs, but not any subtype-specific clustering (Fig. 2b). Although performing ANOVA among the group (Fig. 2c–g), no statistical significance has been observed among the groups in RAL (Fig. 2c), EVG (Fig. 2d), CAB (Fig. 2f) and BIC (Fig. 2g). However, in DTG, there was statistical difference as A-like viruses gave better ex-vivo potency than HIV-1B (P = 0.019) and HIV-1C (P = 0.011) (Fig. 2e).

Fig. 2

Fig. 2

Back to Top | Article Outline

Analysis of the integrase sequences from the patient cohort

The presence of primary integrase DRM was analyzed in 270 INSTI-naive and ART experienced individuals, infected with HIV-1B (n = 92), C (n = 82), 01_AE (n = 25), 02_AG (n = 15), A1/A2 (n = 22), other pure subtypes (n = 13) and recombinant forms [CRFs or unique recombinant forms (URFs)] (n = 21). One HIV-1B patient had a major primary INSTI-DRM, T66I (Supplementary Digital Content 1, In addition, 17 of 270 (6.3%) patients had at least one major accessory DRM: T97A (n = 3), E157Q (n = 12), or A128T (n = 2). The A128T is a nonpolymorphic mutation selected in vitro by EVG, but does not appear to reduce INSTI susceptibility ex vivo. The naturally occurring polymorphism M50I was found in 49 (18%) of the INSTI-naive patients. None of the INSTI-naive patients had nucleoside reverse transcriptase Inhibitor (NRTI) mutations; however, three individuals had non-nucleoside reverse-transcriptase inhibitors (NNRTIs), DRM (V106M, K103N, G190A).

Thirteen of the 17 INSTI-naive patients initiated an INSTI-based regimen with either DTG (n = 11) or RAL (n = 2). Among them, seven initiated 2 NRTI + DTG once-daily (OD), six switched from a 2 NRTI + 1 NNRTI regimen to an INSTI containing ART (DTG OD n = 3; DTG twice-daily n = 1) with either abacavir/lamivudine or tenofovir disoproxil/emtricitabine. The reasons for switching the regimen was either failing on 2 NRTI + 1 NNRTI therapy (n = 4) or due to side effects with undetectable viremia (n = 2). The patients who failed an NNRTI (n = 4)-based regimen had one to three regimens prior to initiation of the INSTI. Among the NNRTI failure patients, two patients had K103N and V106M/Y188CH mutation, respectively. No NRTI or protease inhibitors mutations were found. The HIV-1 RNA load was also checked after 6 months. Two RAL-treated patients with E157Q and T97A mutations, respectively, failed to suppress viremia at month 6. Eight of the nine patients on DTG, who had a major accessory mutation just before switch, showed low-level viremia at month 6. One of the two patients who switched with suppressive viremia, maintained undetectable viral load, but one (with T97A) had a viral blip at month 6 (57 copies/ml) (Fig. 3a). Sequencing of samples collected 1216 and 1038 days earlier from these two patients showed that they had E157Q and the T97A mutations, respectively. Taken together, 11 of 13 INSTI-naive patients with INSTIs DRM before initiating RAL or DTG did not reach undetectable viremia after 6 months.

Fig. 3

Fig. 3

Among the INSTI-experienced patients failing on ART (n = 96), 24% (23/96) had major INSTI DRM (Fig. 3b) and all of them had failed RAL. Four (17%) of the 23 patients were predicted to be fully susceptible to DTG, CAB and BIC. Five patients (21%) with multiple mutations (Q148QR + N155H, G140A + Q148R, E138K + G140S + Q148R, E138K + G140A + S147SG + Q148K and G140S + Q148H + N155H) were predicted to have high-level cross resistance to DTG, CAB and BIC, although no phenotypic assessment has confirmed this observation [2,3]. The DRM profiles were similar in the HIV-1B and non-B subtypes.

Back to Top | Article Outline


An increasing body of data suggests that the efficacy of the ART and the HIV-1 drug resistance pattern varies depending on the HIV-1 subtype. However, whether this holds true for second generation INSTIs is not well studied. Hence, this study was designed to evaluate the potency of INSTIs in different HIV-1 subtypes, using a combination of biochemical and cell culture assays. Overall, our data suggested that all of the three newer INSTIs (DTG, CAB, BIC) had a higher antiretroviral potency against HIV-1B as well as non-B subtypes compared with the older INSTIs (RAL and EVG), and no significant subtype-specific difference in antiretroviral potency of the individual INSTIs was observed.

The integration of HIV DNA into host genome requires 3′-end processing and strand transfer reaction [16]. Similar to previous reports for RAL, EVG [17], DTG [18] and BIC [3], our data showed that the INSTIs are not as efficient in inhibiting 3′- end processing as they are in inhibiting strand transfer activity for all subtypes, as expected. It has been reported that 3′-end processing is as efficient in HIV-1 group O as in HIV-1 group M [19]; however, to our knowledge, no comparison among different HIV-1 subtypes has been done earlier. Our results suggest that the IC50–3P for HIV-1C IN is highest among the tested subtypes for all INSTIs, except for BIC. As the selectivity for strand transfer over 3′-end processing is driven by key enzyme–DNA interactions in the active site [18], it is possible that DNA/IN interactions in HIV-1C integrase differ from other subtypes. This interpretation could possibly be further supported by our results that the RAL IC50-ST for HIV-1B and HIV-1A-like were significantly higher than the IC50-ST of other INSTIs, but similar IC50-ST were obtained for HIV-1C for all INSTIs. It must be pointed out here that the IC50–3P values were determined with one integrase from four different subtypes. Hence, it is possible that these IC50 values may change when several INs from different subtypes are cloned and inhibition assays are conducted. However, IC50–3P values from our independently conducted assays for each subtype were consistent suggesting ‘meaningful’ differences.

In our DSA, we noted that DTG, CAB and BIC had a higher potency than RAL and EVG independent of subtype, which is in line with recent studies on BIC [3] and CAB [2,20]. Our cluster analysis also suggested that no differences existed in activity against the subtypes analyzed for CAB, and BIC, although a better potency for A-like viruses were observed for DTG.

A natural polymorphism M50I was found in 18% of INSTI-naive patients, which is in-line with the prevalence of this mutation in different subtypes in the Stanford HIV drug resistance database, ranging between 3 and 33% (median 10.5%) [14]. Mutation R263K mutation has been shown to confer two-fold to five-fold resistance to DTG, with decreased viral replication and strand transfer activity [21]. In breakthrough selections, M50I emerges after R263K [3] and provides replication advantage to R263K containing virus. A selection of R263K/M50I has been described in virus outgrowth assay for BIC that resulted 2.8-fold reduction of BIC-susceptibility, but M50I alone did not have any effect [3]. Seven of our 24 recombinant viruses had preexisting M50I, but the EC50 for all INSTIs were very low confirming no effect of this mutation alone ex vivo.

Although analyzing a large number of integrase sequences from INSTI-naive patients, we observed only one patient (0.4%) with a primary DRM. Instead, 17 (6.3%) patients had major INSTI accessory mutations (E157Q, T97A and A128T). These accessory mutations should not be interpreted as transmitted drug resistance and the importance of such natural polymorphisms is disputed. However, of 13 patients with these mutations, who were switched to RAL (n = 2) or DTG (n = 11) containing ART, only two had undetectable viremia after 6 months. Obviously, this is an anecdotal observation, and other factors such as decreased adherence may also have contributed. However, the slow decline in viral load is still surprising to us, in view of the usual rapid decline when INSTIs are given and that the rate of undetectable viremia in the Swedish InfCare cohort is 95% after 6 months or more on ART [22].

In conclusion, our study showed that the three newer INSTIs (DTG, CAB and BIC) had higher antiretroviral potency not only in HIV-1B but also in non-B subtypes compared with the first two approved INSTIs RAL and EVG. Moreover, no significant difference among the newer INSTIs was seen with regard to their efficacy against different subtypes. Thus, DTG, CAB and BIC are attractive options for countries where non-B subtype infections are prevalent. Our analyses of patient-derived sequences from INSTI-naive individuals in clinical settings were reassuring with respect to the very low prevalence of primary INSTI DRM, despite the use of RAL and EVG since 2006. Furthermore, our sequencing results suggest that a relatively high rate of major accessory INSTI DRMs requires follow-up for their long-term impact on the outcome of INSTI-based ART.

Back to Top | Article Outline


The study is funded by grants from Swedish Research Council (2016-01675) and Stockholm County Council (ALF 20160074). U.N. acknowledges the support received from Jonas Söderquist's Stipendium for Experimental Virology and Immunology Research-2016. Several reagents were obtained from the NIH AIDS Reagent Program. K.S. acknowledges support received from Bond Life Sciences Center grant (DU108). Part of the cloning and recombinant virus production was supported by National Institute of Health RO1 grant GM118012 (S.G.S, K.S., U.N. and A.S.). CAB and BIC were kindly provided by ViiV Healthcare/GSK and Gilead Sciences, respectively. The authors thank all the patients, nurses and clinicians who supported the InfCare system and Anoop T Ambican for generation of the CIRCOS plot.

Authors contribution: U.N. and A.S. conceived and designed the studies and analysis plan. K.S., L.C.R. and S.G.S. performed, analyzed and supervised the biochemical data. S.G.A. and D.T.N. performed the virological assays supervised by U.N. U.N. performed the data collection and sequence analysis. A.S. and B.H. interpreted the clinical data. U.N. wrote the first draft of the article reviewed by K.S. and A.S. All the authors approved the final version of the article.

Back to Top | Article Outline

Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline


1. Anstett K, Brenner B, Mesplede T, Wainberg MA. HIV drug resistance against strand transfer integrase inhibitors. Retrovirology 2017; 14:36.
2. Yoshinaga T, Kobayashi M, Seki T, Miki S, Wakasa-Morimoto C, Suyama-Kagitani A, et al. Antiviral characteristics of GSK1265744, an HIV integrase inhibitor dosed orally or by long-acting injection. Antimicrob Agents Chemother 2015; 59:397–406.
3. Tsiang M, Jones GS, Goldsmith J, Mulato A, Hansen D, Kan E, et al. Antiviral activity of bictegravir (GS-9883), a novel potent HIV-1 integrase strand transfer inhibitor with an improved resistance profile. Antimicrob Agents Chemother 2016; 60:7086–7097.
4. Hightower KE, Wang R, Deanda F, Johns BA, Weaver K, Shen Y, et al. Dolutegravir (S/GSK1349572) exhibits significantly slower dissociation than raltegravir and elvitegravir from wild-type and integrase inhibitor-resistant HIV-1 integrase-DNA complexes. Antimicrob Agents Chemother 2011; 55:4552–4559.
5. White K, Majka A, Novikov N, Miller M, Tsiang M. The integrase strand transfer inhibitor bictegravir has a long integrase/DNA dissociation half-life. International Congress of Drug Therapy in HIV Infection. Glasgow, UK; 2016. pp. P025.
6. Clotet B, Feinberg J, van Lunzen J, Khuong-Josses MA, Antinori A, Dumitru I, et al. Once-daily dolutegravir versus darunavir plus ritonavir in antiretroviral-naive adults with HIV-1 infection (FLAMINGO): 48 week results from the randomised open-label phase 3b study. Lancet 2014; 383:2222–2231.
7. Walmsley SL, Antela A, Clumeck N, Duiculescu D, Eberhard A, Gutierrez F, et al. Dolutegravir plus abacavir-lamivudine for the treatment of HIV-1 infection. N Engl J Med 2013; 369:1807–1818.
8. Han YS, Mesplede T, Wainberg MA. Differences among HIV-1 subtypes in drug resistance against integrase inhibitors. Infect Genet Evol 2016; 46:286–291.
9. Kantor R, Smeaton L, Vardhanabhuti S, Hudelson SE, Wallis CL, Tripathy S, et al. Pretreatment HIV drug resistance and HIV-1 subtype C are independently associated with virologic failure: results from the multinational PEARLS (ACTG A5175) clinical trial. Clin Infect Dis 2015; 60:1541–1549.
10. Haggblom A, Svedhem V, Singh K, Sonnerborg A, Neogi U. Virological failure in patients with HIV-1 subtype C receiving antiretroviral therapy: an analysis of a prospective national cohort in Sweden. Lancet HIV 2016; 3:e166–e174.
11. Doyle T, Dunn DT, Ceccherini-Silberstein F, De Mendoza C, Garcia F, Smit E, et al. Integrase inhibitor (INI) genotypic resistance in treatment-naive and raltegravir-experienced patients infected with diverse HIV-1 clades. J Antimicrob Chemother 2015; 70:3080–3086.
12. Quashie PK, Oliviera M, Veres T, Osman N, Han YS, Hassounah S, et al. Differential effects of the G118R, H51Y, and E138K resistance substitutions in different subtypes of HIV integrase. J Virol 2015; 89:3163–3175.
13. Aralaguppe SG, Siddik AB, Manickam A, Ambikan AT, Kumar MM, Fernandes SJ, et al. Multiplexed next-generation sequencing and de novo assembly to obtain near full-length HIV-1 genome from plasma virus. J Virol Methods 2016; 236:98–104.
14. Rhee SY, Gonzales MJ, Kantor R, Betts BJ, Ravela J, Shafer RW. Human immunodeficiency virus reverse transcriptase and protease sequence database. Nucleic Acids Res 2003; 31:298–303.
15. Grossmann S, Nowak P, Neogi U. Subtype-independent near full-length HIV-1 genome sequencing and assembly to be used in large molecular epidemiological studies and clinical management. J Int AIDS Soc 2015; 18:20035.
16. Craigie R. HIV integrase, a brief overview from chemistry to therapeutics. J Biol Chem 2001; 276:23213–23216.
17. Marinello J, Marchand C, Mott BT, Bain A, Thomas CJ, Pommier Y. Comparison of raltegravir and elvitegravir on HIV-1 integrase catalytic reactions and on a series of drug-resistant integrase mutants. Biochemistry 2008; 47:9345–9354.
18. Metifiot M, Johnson BC, Kiselev E, Marler L, Zhao XZ, Burke TR Jr, et al. Selectivity for strand-transfer over 3′-processing and susceptibility to clinical resistance of HIV-1 integrase inhibitors are driven by key enzyme–DNA interactions in the active site. Nucleic Acids Res 2016; 44:6896–6906.
19. US Food and Drug Administrator. Antiretroviral drugs used in the treatment of HIV infection. (2016) Available at http:// [Accessed on 13 October 2016].
20. Karmon SL, Mohri H, Spreen W, Markowitz M. GSK1265744 demonstrates robust in vitro activity against various clades of HIV-1. J Acquir Immune Defic Syndr 19992015; 68:e39–e41.
21. Mesplede T, Quashie PK, Osman N, Han Y, Singhroy DN, Lie Y, et al. Viral fitness cost prevents HIV-1 from evading dolutegravir drug pressure. Retrovirology 2013; 10:22.
22. Gisslen M, Svedhem V, Lindborg L, Flamholc L, Norrgren H, Wendahl S, et al. Sweden, the first country to achieve the Joint United Nations Programme on HIV/AIDS (UNAIDS)/World Health Organization (WHO) 90–90–90 continuum of HIV care targets. HIV Med 2017; 18:305–307.

bictegravir; cabotegravir; integrase inhibitor; non-B subtype; resistance

Copyright © 2018 Wolters Kluwer Health, Inc.