|Home | About | Journals | Submit | Contact Us | Français|
We characterized the raltegravir (RAL) susceptibility and the integrase (IN)-mediated replication capacity (RC) of RAL-resistant human immunodeficiency virus type 1 from two patients experiencing virologic failure during continuous RAL salvage therapy. The following two distinct outcomes were observed: (i) the selective outgrowth of virus with high-level RAL resistance and high IN-mediated RC leading to significant viral load rebound and (ii) the selection of virus with a slight reduction in RAL susceptibility and low IN-mediated RC resulting in sustained low-level viremia.
Resistance to raltegravir (RAL), the first integrase (IN) inhibitor approved for use in the clinic, is associated with mutations at IN codons 143, 148, and 155 (3). Recent studies indicate that the primary resistance mutations Y143C/R, Q148H/K/R, and N155H represent mutually exclusive and nonoverlapping genotypic resistance pathways (1, 2, 5, 6, 7). The replacement of one resistant pathway by another has been observed in patients experiencing prolonged virologic failure on a RAL-containing regimen (6, 8). Most of these shifts in IN resistance profiles were characterized by the loss of variants containing N155H and the emergence of variants containing Q148R/H or, in a few cases, Y143C/R (6, 8). However, detailed characterizations of viral evolution under continued RAL drug pressure in patients experiencing incomplete viral suppression are limited. In this study, we conducted a comprehensive longitudinal evaluation of RAL susceptibility and IN-mediated replication capacity (RC) in two subjects experiencing virologic failure while receiving RAL-based salvage therapy.
(This research was presented at the 16th Conference on Retroviruses and Opportunistic Infections, Montreal, Canada, 8 to 11 February 2009.)
Two human immunodeficiency virus (HIV)-infected patients were studied, of whom both had received extensive prior treatments for 12 years with multiple classes of antiretroviral drugs and were receiving RAL-based salvage therapy at the time of this study. Patient 1 was infected with subtype B virus and receiving RAL in combination with zidovudine, lamivudine, abacavir, etravirine (ETV), ritonavir-boosted darunavir, and enfuvirtide (ENF). The baseline (BL) HIV type 1 (HIV-1) RNA level and CD4 count were 5.23 log10 copies/ml and 16 cells/mm3, respectively. Patient 2 was infected with a subtype CRF02_AG virus and was receiving RAL in combination with tenofovir, emtricitabine, ETV, ritonavir-boosted lopinavir, and ENF. The BL HIV-1 RNA level and CD4 count were 4.88 log10 copies/ml and 4 cells/mm3, respectively. In both patients, aside from RAL, ETV was the only drug with predicted activity based on the protease and reverse transcriptase genotypes at BL. No ENF resistance mutations were found at this time; however, a previous history of virological failure associated with ENF-resistant virus selected during a first ENF-based regimen was reported for these two patients. Plasma samples were obtained from patient 1 at BL and at week 4 (W4), W7, W11, W22, and W39 of RAL treatment. As previously described, the IN resistance mutations N155H and Q148R were detected at initial treatment failure (W4 and W7) but were later accompanied, and replaced, by Q148H and G140S (W11, W22, and W39) (2). Plasma samples from patient 2 were collected at BL, W16, and W21. Treatment failure in patient 2 was associated with an uncommon RAL-resistant genotype including the T66T/A, L74I/M, and E92E/Q mutations at W16 (2) with the subsequent addition of E138K/N/D and G140R by W21. RAL susceptibility, expressed as the n-fold change (FC) in the drug concentration required to inhibit virus replication by 50% (EC50) relative to a RAL-susceptible reference virus, and IN-mediated RC were determined using the PhenoSense IN assay (4). The PhenoSense IN assay is a modification of the platform PhenoSense technology (9) and was validated to Clinical Laboratory Improvement Amendments (CLIA) specifications for routine patient testing in July 2008. Briefly, IN coding sequences were amplified from patient plasma virus by reverse transcriptase PCR and transferred into an HIV genomic vector containing a luciferase reporter gene in place of the HIV envelope. Cotransfections of human embryonic kidney (HEK 293) cells with HIV-1 genomic vectors containing IN coding sequences derived from patient virus and an amphotropic murine leukemia virus envelope expression vector were performed to produce pseudovirus stocks, which were then used to infect fresh HEK 293 cells in the absence or presence of serial dilutions of IN inhibitor. Susceptibility was calculated by plotting the percent inhibition of virus replication (luciferase activity) versus the log10 drug concentration to derive the EC50. The FC in drug susceptibility was calculated by comparing the EC50 of the sample virus to the EC50 of a wild-type reference strain (NL4-3). The IN-mediated RC was determined and expressed as a percentage of the viral infectivity (luciferase activity) in the absence of drug relative to the NL4-3 reference virus.
Virus populations isolated from patient 1 displayed incremental reductions in RAL susceptibility, with the FC increasing from 0.91 at BL to ~5 at W4 and W7 and >150 at W11, W22, and W39. Mutations Q148R and N155H, detected at W4, were accompanied by modest reductions in RAL susceptibility (5-fold), whereas G140S and Q148H appeared at W11 coinciding with a substantial reduction in RAL susceptibility (>150-fold) (Fig. (Fig.1).1). The decrease in IN-mediated RC at W4 (53%) associated with Q148R and N155H was not observed at W11 and later time points (IN-mediated RC = 90%) when the virus populations were dominated by variants containing the Q148H primary mutation in combination with the secondary mutation G140S (Fig. (Fig.1).1). To further define the emergence and evolution of RAL resistance, we characterized 34 to 46 molecular clones from each of the virus populations. Clonal analyses revealed detailed changes in the composition of virus populations under continued RAL drug pressure (Fig. (Fig.2).2). Initially, at W4, a heterogeneous viral population was observed containing three distinct RAL-resistant variants, Q148K, Q148R, and N155H. A gradual decrease in the proportion of the Q148K, Q148R, and N155H variants occurred between W4 and W11. By W11, the plasma viral population was comprised predominantly of G140S-Q148H variants (74%), which were the only resistant variants detected at later time points (W22 and W39). Representative clones from each time point containing different resistance mutations were characterized phenotypically (Table (Table1).1). The clones bearing G140S-Q148H exhibited a higher level of reduction in RAL susceptibility (FC > 150) than the clones containing mutation N155H or Q148R alone (range of the average FC over all time points, 13- to 39-fold). The clones with G140S-Q148H also displayed higher IN-mediated RC (range of the average IN-mediated RC over all time points, 63 to 69%) than that of the clones with N155H or Q148R alone (range of the average IN-mediated RC over all time points, 0.1 to 40%). Over the time points tested, the RAL FC and IN-mediated RC for each resistant variant generally remained constant, with the exception of N155H clones which exhibited much lower IN-mediated RC at W11 than W4 (Table (Table1).1). The reduction in the IN-mediated replication capacity of N155H clones at W11 was accompanied by a modest reduction in susceptibility (13 to 23). The genetic determinants of these phenotypic changes are not obvious upon simple inspection and require further investigations. Importantly, resistant clones with double mutation G140S-Q148H that had the largest reductions in RAL susceptibility and higher IN-mediated RC emerged and subsequently dominated the viral population under continued RAL drug pressure.
For patient 2, viruses isolated at W16 of the RAL-based therapy contained the uncommon RAL-associated resistance genotypes T66A, L74I/M, and E92Q. Interestingly, this patient exhibited low viremia, approximately 2.40 log10 copies/ml, prolonged over 1 year of follow-up and accompanied by increases in CD4 counts (Fig. (Fig.3A).3A). Only a modest reduction in RAL susceptibility was observed at W16 (FC = 2.3) that was associated with a notable reduction in IN-mediated RC (60%) (Fig. (Fig.3B).3B). By W21, the additional IN mutations E138K/N/D and G140R were detected along with a further reduction in IN-mediated RC (24%); however, these changes were not accompanied by incremental reduction in RAL susceptibility (FC = 1.3) (Fig. (Fig.3B).3B). The IN sequence of the W72 virus did not reveal additional mutations despite 12 months of low-level viremia in the presence of continued RAL pressure.
In summary, we demonstrated with patient 1 that the incomplete suppression of virus replication, in the presence of ongoing RAL drug pressure, initially resulted in a heterogeneous virus population with modest reductions in RAL susceptibility and low IN-mediated RC that was eventually replaced by a more homogeneous population containing highly resistant variants with high IN-mediated RC. These changes in phenotype may help explain the shifts in IN genotype under RAL treatment failure observed here and reported previously (2, 8). In the case of patient 1, virus with single mutations N155H or Q148R emerged early and modest reductions in susceptibility to RAL were associated with significant reductions in virus replication. Two mutations, Q148H and G140S, were required to establish a high level of resistance to RAL and high IN-mediated RC. In contrast, studies of patient 2 suggest that RAL-selected viruses appearing during viral rebound can acquire mutations that diminish virus replication while remaining relatively susceptible to RAL. The individual contributions of low IN-mediated RC and sustained RAL susceptibility to the low-level viremia observed in this patient will require further investigations. Of note, patient 2 was infected with a subtype CRF02_AG strain, and little is known about RAL resistance profiles in non-B subtypes.
Although preliminary, our observations are consistent with the existence of distinct genetic pathways leading to the failure of RAL-containing regimens. Our data further imply that one or more constraints and/or stochastic events may restrict the availability of certain pathways, thereby limiting the ability of certain virus populations from acquiring large reductions in RAL susceptibility despite sustained drug pressure. One likely barrier is the reduction in IN-mediated RC associated with specific RAL resistance mutations, but other as-yet-undefined constraints may also influence the availability of RAL resistance pathways. Consequently, RAL treatment failure may present as a sustained low-level viremia or a complete loss of viral suppression, each with significantly different consequences for the future clinical course of HIV-1 infection.
Published ahead of print on 10 August 2009.