 |
|
|
| |
Tenofovir & Nuke Resistance Mutations & Patterns
|
| |
Reported by Jules Levin
14th Intl HIV Drug Resistance Workshop
June 2005
Quebec City, Quebec, Canada
In the May 2005 issue of Antimicrobial Agents and Chemotherapy, Mark Wainberg has a review of "Changing Patterns in NRTI and NtRTI Mutations". The discussion is directly relevant to the studies reported at the Workshop. There was discussion at the Workshop among researchers that new & different HIV drug resistance patterns are emerging, and that characterizing and understanding HIV drug resistance may be changing due to newly identified mutations that and patterns of mutations. The abstracts reviewed below reflect these changes. Link to Wainberg's article:
http://www.natap.org/2005/HIV/061605_06.htm
In addition, in the July 1 2005 issue of Jnl of Clinical Infectious Diseases, Andrew Zolopa published his study originally presented at the IAC in Thailand: "Accuracy, Precision, and Consistency of Expert HIV Type 1 Genotype Interpretation: An International Comparison (The GUESS Study)". Zolopa surveyed resistance experts. Genotypes from clinical specimens were sent to an international panel of 12 resistance experts. The experts predicted the exact phenotype fold change category correctly 44% of the time, but they varied widely by antiretroviral drug (range, 25%-74%). Agreement among experts in predicting phenotype fold change category ranged widely depending on the drug (median agreement, 42% [range, 28%-74%]). Experts agreed on treatment recommendations in a median of 79% of instances, and recommendations were consistent over time, with blinded retesting. Zolopa concluded: Although their ability to predict phenotype from a genotype varied for individual antiretroviral drugs, this expert panel had a high degree of agreement in deriving treatment recommendations from the genotype.
Here are brief summaries of studies presented at the Workshop & reported below in this report.
"We recently demonstrated the association of 15 unreported mutations in HIV-1 reverse transcriptase (RT) with the nucleoside reverse transcriptase (NRTI)-treatment and NRTI-resistance mutations. Our study gives evidence of RT mutational patterns more complex than currently known, that regulate positively and/or negatively NRTI-resistance. This strongly suggests that mutations beyond those currently known to confer resistance should be considered to define more precise algorithms able to predict resistance to antiretroviral drugs."
"The K70E RT mutation was selected in 10% of the antiretroviral-naive subjects receiving ABC/3TC/TDF regimen with both baseline and post baseline (pre-therapy switch) genotype. This mutation was also previously reported to be selected in adefovir passage experiments and to be associated with adefovir resistance."
"The K70E RT mutation is more prevalent in NRTI-resistant viruses and has become more common in clinical samples since the introduction
of tenofovir into clinical use. It has been reported that K70E modestly reduces tenofovir susceptibility in recombinant viruses."
"After extended in vitro passage with tenofovir and emtricitabine, no Q151M or T69 insertion virus was observed. A mixture of three viruses - M184I, K65R and K65R+M184V - was observed with the M184I occurring first. The initial presence of distinct M184I and K65R viral genomes may relate to the effect of the M184V/I mutation to increase susceptibility to tenofovir."
"The non-TA NAMS K65R, L74V and M184V significantly enhance susceptibilities to NNRTIs and to ZDV. Discrete differences between these mutations exist such that EFV susceptibility is enhanced by K65R>L74V and M184V. ZDV susceptibility is enhanced by M184V>L74V>K65R."
"Although several neucleoside reverse transcriptase inhibitors (NRTI) (notably DDI, ABC) select K65R and L74V mutations, the double mutant K65R+L74V is rare in clinics. Our data confirm that the presence of K65R+L74V double mutant is rare in clinics (0.4% in the studied population). However, patients harbouring this double mutant exhibit a high level plasma viral load (7130-1x106 copies/ml), which is in disagreement with the in vitro studies reporting that this
double mutant is highly attenuated for replication and renders the reverse transcriptase dysfunctional."
"It has been suggested that mutations at codons G196/Q207/H208/ R211/L214 may facilitate drug-resistance and compensate for M184V induced reversion of zidovudine resistance..... These findings suggest significant interactions occur between key reverse transcriptase resistance-mutations and changes at codons G196/Q207/H208/R211/L214....The strong positive association between H208Y and drug-experience, especially with lamivudine, zalcitabine and indinavir, and the presence of M184V and TAMs suggests a resistance or compensatory
role for H208Y. Q207E/A, R211K, L214F, and to a lesser extent G196E, showed an overall negative association with drug-experience, were either positively
(G196E, Q207E/A, R211K) or negatively (L214F) associated with M184V, and clustered with the TAM1 or TAM2 pathways."
Involvement of novel HIV-1 reverse transcriptase mutations in the highly ordered regulation of NRTI resistance
F Ceccherini-Silberstein1, V Svicher1, T Sing2, M Santoro1, N Beerenwinkel 2, F Gago4, A Bertoli1, F Forbici3, MC Bellocchi3, P Narciso3, A d'Arminio
Monforte5, A Antinori3 and CF Perno1,3
1University of Rome "Tor Vergata", Rome, Italy
2Max Planck Institute for Informatics, Saarbrucken, Germany
3INMI "L Spallanzani", Rome, Italy
4University of Alcala, Madrid, Spain
5University of Milan, Italy
BACKGROUND: We recently demonstrated the association of 15 unreported mutations in HIV-1 reverse transcriptase (RT) with the nucleoside reverse transcriptase (NRTI)-treatment and NRTI-resistance mutations.
The aim of this study was to characterize the role of such mutations in the regulation of NRTI-resistance.
METHODS: 1906 HIV-1 subtype B pol sequences from 551 drug-naive patients and 1355 NRTI-treated patients, were collected and analysed. Significant dendrograms were identified by hierarchical-clustering and bootstrap-analysis, and the multidimensional-scaling plot was obtained via Sammon-mapping.
Sequences from Stanford HIV Drug-Resistance Database were also analysed to assess the association of mutations with NRTI-reduced susceptibility (Virco).
RESULTS: Twelve mutations (K20R, V35M, T39A, K43E/N/Q, K122E, G196E, E203D/K, H208Y, D218E), positively associated with NRTI-treatment, strongly clustered with known NRTI-resistance mutations on divergent evolutionary pathways. In particular, T39A, K43E/Q, K122E, E203K, H208Y clustered (bootstrap-value=0.98) with TAM1-cluster (M41L+L210W+T215Y). The co-presence of K43E+K122E+E203K+H208Y in this cluster was associated with a 42-fold increase of AZT-resistance.
In addition, the co-presence of K43E and H208Y in TAM-1-cluster was also associated at failure with 3.7-fold increase in viraemia. Differently, D218E
clustered (bootstrap-value=0.99) with TAM2-cluster (D67N+K70R+K219Q+T215F) and its presence in the cluster determined 2.1-fold greater in viraemia at failure, but no association with an increase of AZT or
D4T-resistance was found.
In contrast, two mutations (I50V, R83K), negatively associated with NRTI-treatment, showed antagonistic correlations with NRTI-resistance mutations and were associated at failure with increased susceptibility to specific NRTIs. In particular, I50V showed antagonism with the 3TC-selected mutation M184V (P=0.015), and was associated with fourfold decrease in M184V/3TC resistance.
R83K showed antagonism with TAM2-cluster (P<0.001), and was associated with twofold decrease in both AZT- and D4T-resistance. Finally, the
polymorphism F214L was positively associated with TAM2-cluster (P<0.001), and negatively associated with TAM1-cluster (P<0.001).
Structural analysis showed that many of the correlated positions were
within 8A distance of each other, confirming the role of local environment for interactions among mutations.
CONCLUSIONS: Our study gives evidence of RT mutational patterns more complex than currently known, that regulate positively and/or negatively
NRTI-resistance. This strongly suggests that mutations beyond those currently known to confer resistance should be considered to define more precise algorithms able to predict resistance to antiretroviral drugs.
Selection of the HIV-1 reverse transcriptase mutation K70E in antiretroviral-naive subjects treated with tenofovir/abacavir/lamivudine therapy
L Ross1, P Gerondelis1, Q Liao1, B Wine1, M Lim1, M Shaefer1, A Rodriguez2, K Limoli3, W Huang3, NT Parkin3, J Gallant4, R Lanier1
1GlaxoSmithKline, RTP NC 27709 USA
2Univ of Miami, Miami, FL, USA
3ViroLogic, Inc., South San Francisco, CA, USA
4Johns Hopkins Univ, Baltimore, MD, USA
BACKGROUND: The high incidence of virologic nonresponse observed with the treatment-naive subjects of the ESS30009 tenofovir/abacavir/lamivudine study
arm has been associated with selection for K65R and M184V reverse transcriptase (RT) coding changes (Gallant et al. 2003, 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy, Abstract H-
840).
Here, the identification of additional treatment-emergent coding changes possibly representing alternative resistance pathways associated with the observed
non-response to this regimen is described.
METHODS: Plasma HIV drug susceptibility and population genotyping were performed by ViroLogic. HIV clonal analysis was performed at GlaxoSmithKline.
RESULTS: There were 81 subjects with both baseline and post-baseline (pre-therapy switch) genotypes on the TDF arm. Available on-therapy timepoints with
genotype ranged from weeks 4 through 30. The most commonly selected IAS USA-defined NRTI mutations detected at any timepoint, by population genotype,
were M184I or V (78/81, 96%), K65R/K (43/81, 53%), and/or Y115F/Y (9/81, 11%). However, the K70E mutation was also selected in plasma-derived virus from 8/81 subjects (10%) at timepoints between week 8 to week 20 (P<0.001). At the time of mutation selection, all subjects were either experiencing viral
rebound or had never responded to therapy.
By population genotype, the K70E mutation (all as K/E mixtures) was selected along with M184I/V (8/8), K65R (7/8, all mixtures), and/or Y115Y/F (2/8). No subject on either arm (n=318) had detectable K70E at baseline.
Clonal analysis was done for one sample with a population genotype of K65K/R+K70K/E+ Y115F+ V118I+M184V (V118I present at baseline). K70E
(15/22, 68%) was never detected on the same clones with K65R (6/22, 27%), while Y115F (21/22, 95%) was selected with either mutation.
CONCLUSIONS: The K70E RT mutation was selected in 10% of the antiretroviral-naive subjects receiving ABC/3TC/TDF regimen with both baseline and post baseline (pre-therapy switch) genotype. This mutation was also previously reported to be selected in adefovir passage experiments and to be associated with adefovir resistance, and/or with selection in heavily antiretroviral-
experienced subjects, it was also selected in one treatment-naive patient treated with TDF/3TC/EFV, and in one antiretroviral-experienced subject treated with ABC/3TC/TDF/ddI.
These data suggest an alternate pathway for tenofovir resistance, while the clonal data suggests antagonism and exclusivity between K65R and K70E at the genomic level.
Adefovir-associated HIV-1 RT mutation K70E in the age of tenofovir
R Kagan1, L Ross2, M Winters3, T Merigan3, P Heseltine1 and M Lewinski1
1Quest Diagnostics Nichols Institute, San Juan Capistrano, CA,
USA 2GlaxoSmithKline, RTP, NC, USA
3Centre for AIDS Research, Stanford University, Stanford, CA, USA
BACKGROUND: The rare K70E RT adefovir-associated resistance mutation has been reported in two recent cases of virological failure on tenofovir-containing
regimens. The use of tenofovir has significantly increased since its introduction at the end of 2001. We investigated the prevalence of this mutation in clinical
specimens submitted for reference laboratory testing in 2002-2004.
METHODS: The Quest Diagnostics database of subtype B HIV-1 RT sequences includes 42 109 deidentified clinical samples from population-based sequencing
from 1998-2001 and 92 036 sequences from 2002-2004. We tabulated frequencies of K70E and several RT mutations identified in conjunction with
K70E in two recent case reports. Mixed mutant/wild type substitutions were excluded from pairwise frequency analyses.
RESULTS: The prevalence of K70E increased from 0.15% (65 sequences, 1998-2001, pre-tenofovir period) to 0.25% (234 sequences, 2002-2004; _2=12,
P<0.0005). The increase was greatest in the population with ≥1 predicted NRTI resistances (0.19%-0.46% _2=34; P<0.0001). NRTI resistance was predicted for
88% of K70E viruses but only for 49% of the non-K70E viruses (_2=139; P<0.0001). The tenofovir-associated mutation K65R was not detected in K70E-containing sequences.
Previously published clonal analysis of a K70E virus (ESS30009) identified mutations M184V, V118I and Y115F, while in this database, M184V and V118I were found in 35% and 7.5% of K70E sequences respectively, not significantly different from viruses lacking K70E. Y115F was found in15.7% of K70E viral sequences, more frequently than in resistant sequences lacking K70E (OR=8.6; 95% CI 7.7-9.4, P<0.0001).
Other mutations found at higher frequencies with K70E were Y181C, H221Y, Y62V, Q151M, V75I and D67G (ORs 2.9-4.6, P<0.0001).
CONCLUSIONS: The K70E RT mutation is more prevalent in NRTI-resistant viruses and has become more common in clinical samples since the introduction
of tenofovir into clinical use. It has been reported that K70E modestly reduces tenofovir susceptibility in recombinant viruses and decreases RT specific activity
and processivity, whereas Y115F increases the fidelity of the RT enzyme. Thus the more frequent occurrence of Y115F in K70E viruses suggests a compensatory role.
In vitro resistance selection with tenofovir and emtricitabine
NA Margot1, J Waters2 and MD Miller1
1Gilead Sciences Inc., Foster City, CA, USA
2Gilead Sciences Inc., Durham, NC, USA
BACKGROUND: Emtricitabine and tenofovir select in vitro and in vivo for the HIV RT mutations M184V/I and K65R, respectively. As both compounds are available
in a fixed-dose combination tablet, we assessed the in vitro resistance profile of the dual formulation.
METHODS: Resistance selection experiments using tenofovir and emtricitabine alone or in combination were performed in MT-2 cells with HIV IIIb or LAI
and serially increasing drug concentrations. Viruses were analysed genotypically by population sequencing and single genome sequencing (SGS). Phenotypic
analyses in MT-2 cells were also conducted.
RESULTS: Selection with emtricitabine readily gave rise to an M184I mutant virus which changed to M184V at higher emtricitabine concentrations.
Selection with tenofovir lead to K65R after multiple passages which later gained S68N but did not further reduce susceptibility. Co-selection with emtricitabine
and tenofovir yielded an M184I virus first and then transitioned to a K65R virus at intermediate drug concentrations (eight to 16-fold EC50).
At higher emtricitabine concentrations, a K65R+M184V double mutant was selected with high-level resistance to emtricitabine, but lower resistance to tenofovir than K65R alone. SGS of intermediate mixed viral populations
showed that M184I and K65R were initially present on distinct genomes and then eventually were present on the same genome. A novel K65N+K70R double
mutant was also observed in a separate culture which resulted in intermediate emtricitabine resistance relative to K65R or M184V/I. Finally, after >1 year of
selection in up to 100 µM tenofovir and emtricitabine, there was no development of multidrug resistance genotypes Q151M or T69 insertions.
CONCLUSIONS: After extended in vitro passage with tenofovir and emtricitabine, no Q151M or T69 insertion virus was observed. A mixture of three viruses - M184I, K65R and K65R+M184V - was observed with the M184I occurring first. The initial presence of distinct M184I and K65R viral genomes may relate to the effect of the M184V/I mutation to increase susceptibility
to tenofovir. In two clinical studies of patients treated with tenofovir and emtricitabine plus either efavirenz or lopinavir, 5 out of 434 patients (1.1%)
have developed M184V/I while no patients have developed K65R, consistent with the in vitro observations of slower development of K65R with this combination.
Differences in enhanced susceptibilities to NNRTIs and to zidovudine (ZDV) in site directed mutants (SDMs) bearing K65R, L74V or M184V
C Chappey, E Coakley, K Limoli, J Whitcomb and
N Parkin
ViroLogic Inc, South San Francisco, CA, USA
BACKGROUND: We have previously demonstrated that non-nucleoside reverse transcriptase inhibitor (NNRTI) hypersusceptibility (HS) is associated with thymidine analog mutations (TAMs). We recently described NNRTI HS in clinical isolates the NRTI mutations K65R, M184I/V, and T69D/N/S. Here we use site directed mutants (SDMs) to define the NNRTI susceptibility patterns of specific nonTA NAMs and compare these to zidovudine (ZDV) susceptibility
patterns
METHODS: NL4-3 SDMs were created with the single mutations K65R, L74V and M184V. Multiple replicates of each SDM, K65R (n=18), L74V (n=20)
and M184V (n=17), were tested in a single cycle phenotyping assay (PhenoSense). The mean fold change (FC) values were compared to 20 replicates of the NL4-3 reference using the t-test.
RESULTS: For the three NNRTIs the mean FCs for the NL4-3 reference ranged 0.95-1.01. Relative to reference all SDMs had lower mean FCs for each NNRTI (P<0.0001). The mean FC (SD) for EFV with K65R, L74V and M184V was 0.56 (0.03), 0.67 (0.02) and 0.65 (0.03), respectively. The mean FC (SD) for NVP
with K65R, L74V and M184V was 0.53 (0.05), 0.59 (0.06) and 0.67 (0.05), respectively. The mean FC (SD) for DLV with K65R, L74V and M184V was 0.54
(0.03), 0.68 (0.06) and 0.65 (0.05), respectively. For EFV, K65R showed lower mean FC than both L74V (P<0.0001) and M184V (P<0.0001). L74V showed
lower mean FC to NVP than to EFV or DLV (P<0.0001 for both). By contrast the mean ZDV FC (SD) with M184V, L74V and K65R was 0.33 (0.04), 0.49 (0.07) and 0.60 (0.05), respectively. The mean ZDV FC was greater for M184V than L74V (P<0.0001) and was greater for L74V than K65R (P<0.0001).
CONCLUSIONS: The non-TA NAMS K65R, L74V and M184V significantly enhance susceptibilities to NNRTIs and to ZDV. Discrete differences between
these mutations exist such that EFV susceptibility is enhanced by K65R>L74V and M184V. ZDV susceptibility is enhanced by M184V>L74V>K65R. These
observations may have relevance to models exploring the mechanisms of reverse transcriptase inhibitor HS.
The coexistence of K65R+L74V on the same HIV genome is rare but possible as evidenced by cloning analysis
M Henry, C Tourres, P Colson and C Tamalet
Timone University Hospital, Marseille, France
BACKGROUND: Although several neucleoside reverse transcriptase inhibitors (NRTI) (notably DDI, ABC) select K65R and L74V mutations, the double mutant K65R+L74V is rare in clinics. The reason for this rarity is still debated. In order to gain further insight, we searched the Marseille database for the frequency
of K65R/L74V pattern.
METHODS: Retrospective analysis of the Marseille database that contains 7151 sequences from 3201 patients. The pol gene from patients' viruses harbouring
the double mutant (pure species or genotypic mixtures) were amplified, sequenced and cloned in pGEM T easy.
RESULTS: Twelve out of 3201 patients (0.4%) harboured the double mutant either as K65R+L74V or as K65K/R+L74V/I/L by population-based sequencing.
Four patients harbouring a wild-type protease devoid of resistance mutations were selected. Two patients received DDI then TDF; the two other patients
received TDF+ABC or DDI as part of the combination regimen.
All four patients were virological non-responders (plasma viral load from 7130-1x106 copies/ml). Cloning and sequencing (average of 18 clones: 12-27 clones) notably revealed that: Patient 1 harboured 100% clones: K65R+L74V and a K65R+L74V bulk sequence. Patient 2 harboured 40% clones: K65R+L74I; 18% clones: K65R+L74I/L+T215Y and a K65KR/L74IL/ T215TNSY bulk sequence. Patient 3 harboured 46% clones: K65R+L74V; 54% clones: K65R+74L and a K65R+L74LV+215T bulk sequence.
In Patient 4, no clone harboured the K65R/L74V double mutant whereas 28.5% of the clones were K65R+T215Y+74L with a K65KR/L74LV/T215CY
bulk sequence.
CONCLUSION: Our data confirm that the presence of K65R+L74V double mutant is rare in clinics (0.4% in the studied population). However, patients harbouring this double mutant exhibit a high level plasma viral load (7130-1x106 copies/ml), which is in disagreement with the in vitro studies reporting that this
double mutant is highly attenuated for replication and renders the reverse transcriptase dysfunctional.
Our findings suggest possible compensatory mechanisms allowing for the coexistence of K65R+L74V, which remains to be established. In addition, surprisingly, this clonal analysis revealed that in two patients, K65R+T215Y which are known to be antagonistic can, in some rare cases, coexist in the same genome.
Mutations at reverse transcriptase (RT) codons G196, Q207, H208, R211 and L214 are associated with drug experience and specific RT mutation patterns
AM Geretti1, C Sabin1, D Dunn2 and G Nebbia1, on behalf of the UK Collaborative Group on HIV Drug Resistance, and the UK Collaborative HIV Cohort (CHIC) Study Group
1Royal Free Hospital and Royal Free & University College Medical
School, London, UK
2Medical Research Council Clinical Trial Unit, London, UK
BACKGROUND: It has been suggested that mutations at codons G196/Q207/H208/ R211/L214 may facilitate drug-resistance and compensate for M184V induced reversion of zidovudine resistance.
AIM: To determine whether mutations at codons G196/Q207/H208/R211/L214 are associated with drug-experience and the presence of M184V and thymidine analogue mutations (TAMs).
METHODS: The CHIC clinical database was linked to the HIV drug-resistance database (1997-2003; drugnaive=2548; drug-experienced=4839), containing data from six clinics in London/Brighton.
RESULTS: The most common mutations were G196E, Q207E/A, H208Y, R211K, L214F. Multivariable analyses showed a significant positive association with drug-experience for H208Y [aOR 24.9 (7.9-78.5), P=0.0001], and a significant negative association for Q207E/A [aOR 0.82, (0.73-0.93), P=0.001], R211K [aOR 0.70 (0.63-0.78), P=0.0001], and L214F [aOR 0.28 (0.25-0.31), P=0.0001]. G196E was weakly associated with drug-experience [aOR 0.85 (0.70-1.02), P=0.08].
In multivariable analyses, Q207E/A was associated with exposure to nelfinavir [aOR 1.3 (1.10-1.60), P=0.003] and saquinavir [aOR 0.79 (0.65-0.97), P=0.02]; H208Y with lamivudine [aOR 3.15 (1.87-5.31), P=0.0001], zalcitabine [aOR 2.2
(1.40-3.40), P=0.0005] and indinavir [aOR 1.82 (1.16-2.88), P=0.01]; R211K with stavudine [aOR 1.17 (1.04-1.32), P=0.009]; L214F with lamivudine
[aOR 0.76 (0.67-0.86), P=0.0001] and zalcitabine [aOR 1.21 (1.00-1.45), P=0.05]. G196E and R211K were positively associated with M184V, M41L,
L210W, T215Y/F; Q207A/E was positively associated with M184V, D67N, K70R, K219Q/E, and negatively associated with M41L, L210W, T215Y/F; H208Y was
positively associated with M184V and any TAM; L214F was negatively associated with M184V, D67N, K70R, T215Y/F, K219Q/E. G196E, H208Y, and
L214F occurred more commonly in subtype-B, Q207E/A in subtype C; R211K was equally distributed between subtypes.
The prevalence of the mutations declined between 1996 and 2003 (P=0.0001).
CONCLUSIONS: The strong positive association between H208Y and drug-experience, especially with lamivudine, zalcitabine and indinavir, and the presence of M184V and TAMs suggests a resistance or compensatory
role for H208Y. Q207E/A, R211K, L214F, and to a lesser extent G196E, showed an overall negative association with drug-experience, were either positively
(G196E, Q207E/A, R211K) or negatively (L214F) associated with M184V, and clustered with the TAM1 or TAM2 pathways.
These findings suggest significant interactions occur between key reverse transcriptase resistance-mutations and changes at codons G196/Q207/H208/R211/L214. Further studies are warranted to determine the impact on virological responses.
| |
| |
| |
| |
|
|
|
|
|
