Abstract

Background

Transmitted resistant HIV may revert to wild-type in the absence of drug pressure due to reduced replication capacity (RC). We observed eight therapy-naive patients infected with HIV harbouring four mutations at nucleoside reverse transcriptase inhibitor (NRTI) resistance-related positions: M41L, T69S, L210E and T215S. If partial reverted resistance patterns like these are detected at baseline, concerns for more extensive resistance in the quasispecies often directs selection of first-line combination antiretroviral therapy (cART) towards more complex regimens.

Methods

Genotypic resistance testing and phylogenetic analysis was performed using pol sequences of 400 therapy-naive patients and 1322 patients with at least one NRTI-related mutation. Reverse transcriptase (RT) genes were cloned into a reference strain and RC was investigated.

Results

Phylogenetic analysis showed that all eight patients are part of a transmission cluster (bootstrap value 92%). The patients resided in three distinct geographical regions and were either homosexually or heterosexually infected. Prior negative serology and analysis of base ambiguity demonstrated circulation for at least 7 years. In vivo evolution showed a mixture with wild-type (T215S/T) in only one untreated patient more than 6 years after diagnosis. The reverted resistance pattern did not confer a substantial reduction in RC compared with wild-type, explaining its persistence in vivo and long-term circulation in the population. Four patients started cART; three of them received quadruple cART. All patients showed good virological and immunological response.

Conclusions

Long-term circulation transcending distinct regions and transmission groups suggests reversion occurred in previous hosts in the transmission chain. Identification of clusters using epidemiological data and active-partner tracing may broaden therapeutic options in cases of transmitted resistance.

Introduction

Despite prevention efforts and availability of combination antiretroviral therapy (cART), HIV incidence in industrialized countries has not decreased. Approximately 10% of newly diagnosed patients are infected with a strain harbouring at least one drug-resistance mutation.1–3 Recently we showed that men who have sex with men (MSM) are particularly at risk for contracting such a virus.2 Reports on transmission networks in Switzerland and Canada show that transmission of resistance often occur in clusters.4,5

We observed eight therapy-naive patients infected with a virus with four mutations at nucleoside reverse transcriptase inhibitor (NRTI) resistance-related positions: M41L, T69S, L210E and T215S. Mutations at codons 41, 210 and 215 are thymidine analogue mutations (TAMs) and mutations at position 69 are also frequently observed in patients exposed to thymidine analogues. T215S is a known revertant of the resistance mutation T215Y/F and L210E is a potential revertant of resistance mutation L210W.

In general, drug-resistant strains confer reduced replication capacity (RC) compared with wild-type.6,7 After transmission, RC may be improved by (incomplete) reversion to wild-type in the absence of drug pressure in the newly infected host.8 Therefore it is often assumed that transmitted drug resistance reflects direct infection from drug-experienced individuals.

When such a reverted pattern is detected in naive patients, the possibility that reversion occurred in these particular patients is usually taken into account. Concerns for more extensive resistance in the quasispecies then results in the choice of more complex initial regimens with a higher genetic barrier, increased pill burden, more frequent toxicity and elevated costs. However, recent studies have shown that drug-resistant variants can persist even despite a low RC via various mechanisms in the absence of therapy.8,9 As a result, treatment-naive individuals can be a source of onward transmission of drug resistance,5,10 minimizing the risk of more extensive resistance in the quasispecies. We studied viral characteristics, phylogenetics and in vivo HIV evolution to gain insight into the characteristics of an infection with this partial reverted resistant strain.

Methods

Eight patients with this specific resistance profile (M41L-T69S-L210E-T215S) were identified in 2004–10. All patients participate in the ATHENA observational cohort, which was approved by the national institutional review board. Clinical, virological and therapy-related data were collected. HIV-RNA genotypic resistance analysis of the pol gene was performed in routine clinical practice. Genotypic sensitivity scores (GSSs) were determined as the sum of scores for each drug in the regimen based on the predicted level of resistance [1, susceptible; ½, (potential) low-level or intermediate resistance; and 0, high-level resistance] in the Stanford HIVdb algorithm version 6.2.

Phylogenetic analysis

The pol sequences of the eight patients and baseline sequences of newly diagnosed treatment-naive HIV patients in our centre in 2004–10 (n = 400) and subsequently all sequences obtained in our centre that showed one or more mutations at NRTI resistance-related positions in 2000–10 (n = 1322) were aligned using ClustalW. A maximum likelihood phylogenetic tree was calculated with the Tamura–Nei model of evolution using bootstrap analysis with 1000 replicates (MEGA 5.0).

Estimation of duration of infection

Data on previous negative serological testing were collected to give definitive insight into the window of infection. In addition, base ambiguity of pol was used, which has proven to be a useful tool to estimate the duration of infection. The cut-off of 0.5% ambiguous nucleotides was chosen based on data from the Swiss cohort indicating that >0.5% ambiguity provides strong evidence against a recent infection event 1 year prior to sampling.11

RC and phenotypic drug susceptibility

Viral RNA was isolated from five plasma samples of patients 3, 4, 5 and 6 and five randomly selected patients infected with wild-type. Nested PCR amplicons containing the N-terminal part of the reverse transcriptase (RT) gene were generated and cloned into an HXB2 backbone and sequenced.12 Recombinant viruses were generated by transfecting 293T cells with these HIV constructs using lipofectamine. RC was determined in peripheral blood mononuclear cells; cells were infected for 2 h with 40 ng of p24, washed twice, resuspended and maintained for 14 days. Once daily 300 μL of cell-free virus supernatant was harvested for p24 analysis. Drug susceptibility was determined in a multicycle assay.13

Sequencing of proviral DNA

DNA was isolated from 450 μL of full blood.14 PCRs were performed to amplify the N-terminal part of RT using the Expand High Fidelity PCR System (Roche).12 Three positive amplification reactions were pooled and sequenced.

Results

Phylogenetic analysis showed that all eight patients with this particular resistance profile (M41L-T69S-L210E-T215S) are part of one transmission cluster (Figure 1; bootstrap value 92%). To ensure clustering was not based on the presence of resistance-associated mutations only, the phylogenetic analysis was repeated after manual reversion of the mutations to wild-type, resulting in the same cluster (bootstrap value 89%). Phylogenetic analysis including all sequences harbouring NRTI resistance mutations from different geographical regions did not reveal additional patients in the cluster or a potential common source of the resistant strain.

Figure 1.

Phylogenetic tree. Maximum likelihood phylogenetic tree of therapy-naive patients (2004–10) (n = 400). The box represents the transmission cluster of eight patients. The phylogeny of the cluster is shown in more detail with bootstrap values. Labels at the tips of the trees correspond to the patient numbers in Figure 2.

Figure 1.

Phylogenetic tree. Maximum likelihood phylogenetic tree of therapy-naive patients (2004–10) (n = 400). The box represents the transmission cluster of eight patients. The phylogeny of the cluster is shown in more detail with bootstrap values. Labels at the tips of the trees correspond to the patient numbers in Figure 2.

The eight patients encompassing the transmission cluster were further characterized (Table 1). All patients were of Dutch ancestry and resided in three distinct regions in the country. Although the main route of transmission was homosexual contact, two patients reported being infected by heterosexual contact. Patients 5 and 6 are known to be partners, as are patients 7 and 8. The CD4 counts at the time of HIV-1 diagnosis ranged from 180 to 1129 (median 670) cells/mm3 and the plasma viral load (VL) ranged from 5510 to >100 000 (median 72 800) copies/mL.

Table 1.

Patient characteristics

Patient Sex Age (years) Route of transmission Date of diagnosis (dd-mm-yyyy) Baseline CD4 (cells/mm3Baseline VL (copies/mL) Start of cART Regimen GSS 
50 homosexual 19-06-2006 1129 34 600 October 2010 TDF/FTC, MVC, ATV/r 3.5 
34 heterosexual 26-01-2009 512 5510    
52 homosexual 25-08-2005 470 81 000    
64 heterosexual 26-11-2003 180 55 000 June 2004 ABC/3TC, NVP 2.5 
43 homosexual 09-09-2010 890 88 100    
48 homosexual 16-09-2010 690 135 000    
43 homosexual 14-10-2009 680 86 800 October 2011 TDF/FTC, LPV/r, EFV 3.5 
41 homosexual 21-07-2008 660 64 600 July 2010 TDF/FTC, LPV/r, EFV 3.5 
Patient Sex Age (years) Route of transmission Date of diagnosis (dd-mm-yyyy) Baseline CD4 (cells/mm3Baseline VL (copies/mL) Start of cART Regimen GSS 
50 homosexual 19-06-2006 1129 34 600 October 2010 TDF/FTC, MVC, ATV/r 3.5 
34 heterosexual 26-01-2009 512 5510    
52 homosexual 25-08-2005 470 81 000    
64 heterosexual 26-11-2003 180 55 000 June 2004 ABC/3TC, NVP 2.5 
43 homosexual 09-09-2010 890 88 100    
48 homosexual 16-09-2010 690 135 000    
43 homosexual 14-10-2009 680 86 800 October 2011 TDF/FTC, LPV/r, EFV 3.5 
41 homosexual 21-07-2008 660 64 600 July 2010 TDF/FTC, LPV/r, EFV 3.5 

M, male; F, female; TDF/FTC, tenofovir/emtricitabine; MVC, maraviroc; ATV/r, ritonavir-boosted atazanavir; ABC/3TC, abacavir/lamivudine; NVP, nevirapine; LPV/r, ritonavir-boosted lopinavir; EFV, efavirenz.

Data on the time of HIV diagnosis and prior negative testing allowed us to determine the period of circulation of this particular strain. The strain was first detected in 2003, when patient 4 tested positive for HIV. The other patients were diagnosed in the period 2005–10. We are certain, based on prior negative serological HIV testing in late 2008, that patient 7 was infected in 2009. Therefore we can conclude that the period of circulation is at least 6 years (Figure 2). Based on the low baseline CD4 count and the percentage of ambiguous nucleotides exceeding 0.5%, it is likely that patient 4 was chronically infected at the time of diagnosis in 2003. Therefore we assume that this strain has been circulating in the Dutch population for at least 7 years.

Figure 2.

Epidemiological curve of time of diagnosis. +, first positive HIV test; −, last negative HIV test; c, base ambiguity >0.5%, suggesting no recent infection. The confirmed period of circulation is 6 years, based on date of diagnosis of patient 4 and the last negative HIV test of patient 7.

Figure 2.

Epidemiological curve of time of diagnosis. +, first positive HIV test; −, last negative HIV test; c, base ambiguity >0.5%, suggesting no recent infection. The confirmed period of circulation is 6 years, based on date of diagnosis of patient 4 and the last negative HIV test of patient 7.

In addition, we studied the in vivo evolution of the viral population in the four patients who did not initiate cART (follow-up 18–77 months). The resistance pattern persisted in all but one patient, in which a mixture with wild-type (T215T/S) was observed more than 6 years after diagnosis. Sequences of proviral DNA at the time of diagnosis and after follow-up (1.5–4 years after diagnosis) were available in two patients and no reversion was observed.

We hypothesized that the RC of this strain may play a role in persistence in treatment-naive patients and circulation within the population. We compared the RC of the five patient-derived recombinant RT cluster clones with five randomly selected patient-derived wild-type RT strains and four reference strains (HXB2 and viral strains that harbour various mutations at codon 184 of RT). As expected, reference strain 184T was the least fit and HXB2 wild-type conferred the highest RC. The five RT clones representing the cluster all replicated as efficiently as HXB2 wild-type, demonstrating that strains with this RT reversion pattern are not substantially reduced in RC (data not shown).

To date, four patients have received cART (Table 1). Three patients received quadruple cART, to ensure a GSS >3. One patient received a combination of abacavir, lamivudine and nevirapine (GSS = 2.5). If calculation was based on the most likely pre-reversion resistance pattern (M41L-T69S-L210W-T215Y/F), the GSS dropped 0.5 points for all patients. However, all four patients achieved and maintained undetectable VLs and all had a good immunological response (mean CD4 count 593 cells/mm3). Retrospective phenotypic drug susceptibility testing showed that the RT cluster strain was susceptible to all tested NRTIs (lamivudine, zidovudine, abacavir and tenofovir) (data not shown).

Discussion

This study is the first to provide evidence of long-term circulation of a resistant HIV-1 strain among therapy-naive patients based on prior negative serological testing. We showed that this variant has been circulating for at least 7 years within a cluster of therapy-naive patients. Clusters with drug-resistant virus have been described before. Buskin et al.15 observed the transmission of multiclass drug-resistant HIV within a cluster of MSM for >2 years. In 2009 Hue et al.16 showed the circulation of resistant lineages within untreated populations. Based on assessment of the most recent common ancestor, they estimated circulation for up to 8 years.16

In contrast to acquired drug-resistant HIV-1 variants in treated patients, transmitted drug-resistant variants in treatment-naive patients can persist for a long time in the absence of drug pressure.9 This is surprising since the fitness costs of some drug-resistance mutations present in our cluster strain have been well established.6 However, we showed that the RT mutations in this cluster did not have an evident negative impact on the RC in vitro, which is in line with the slow reversion to wild-type in vivo.

Previous studies have shown that the persistence of drug-resistant mutations may be explained by selection of compensatory mutations that restore the RC of the virus.17 In our cluster, we detected a previously described compensatory mutation, 60I, in the HIV strain of one patient. It is possible that other mutations in this particular strain play a role as well.

The transmission clusters in the analysis of Hue et al.16 included almost exclusively MSM and were restricted to one geographical area in the UK. We report circulation of drug resistance among therapy-naive individuals transcending distinct regions throughout the country and different transmission groups. Our dataset covered only a subset of all newly diagnosed patients in the Netherlands. Taking into account the long-term circulation in different regions and risk groups, it is likely that transmission of drug resistance among therapy-naive patients is even more extensive than we have shown here. The existence of reservoirs of drug resistance in therapy-naive patients has important public health implications and underlines the importance of baseline genotypic testing and the need for active testing policies to interrupt transmission chains.

Circulation of particular resistance patterns in therapy-naive patients may also have consequences for clinical practice. Due to concerns for more extensive resistance in the quasispecies most clinicians initiated complex quadruple cART for the patients in our cluster.

Although we cannot exclude the possibility of one drug-experienced source infecting all patients individually in the cluster, we could not identify such a source by phylogenetic analysis of the available sequences. However, the long-term circulation in different regions and risk groups makes the hypothesis of one source infecting all patients unlikely. We infer that reversion did not take place in these therapy-naive patients, but in a previous host earlier in the transmission chain. We showed that this particular strain was phenotypically susceptible to all NRTIs that are frequently used in clinical practice. Furthermore, the patient on triple therapy achieved and maintained a good virological and immunological response for more than 7 years. Therefore we conclude that identification of clusters and active-partner tracing may provide insight into the risk of more extensive resistance patterns in the quasispecies and prevent excessive use of antivirals.

Funding

This work was supported in part by the Aids Fonds (grant number 2007034).

Transparency declarations

None to declare.

Acknowledgements

This work was presented by L. M. H. at the Tenth European Meeting on HIV and Hepatitis – Treatment Strategies and Antiviral Drug Resistance, Barcelona, 2012, and at the Seventh International Workshop on HIV Transmission – Principles of Intervention, Washington, DC, 2012. Data are published as abstracts in Reviews in Antiviral Therapy & Infectious Diseases 2012_2 (Abstract O-03) and in Reviews in Antiviral Therapy & Infectious Diseases 2012_7 (Abstract O-11).

We thank Leander Theissen for his contribution to DNA sequencing. We also want to acknowledge the Utrecht Genoteam, especially Danielle van Heumen, for performing genotypic analysis.

References

1
Wensing
AM
van de Vijver
DA
Angarano
G
, et al.  . 
Prevalence of drug-resistant HIV-1 variants in untreated individuals in Europe: implications for clinical management
J Infect Dis
 , 
2005
, vol. 
192
 (pg. 
958
-
66
)
2
Frentz
D
van de Vijver
DA
Abecasis
A
, et al.  . 
Transmission of HIV resistant to non-nucleoside RT inhibitors is rising in MSM in Europe
Brussels, Belgium
European AIDS Clinical Society
 
Abstracts of the Thirteenth European AIDS Conference, Belgrade, Serbia, 2011. Abstract PS1/5, p. 11
3
Castor
D
Low
A
Evering
T
, et al.  . 
Transmitted drug resistance and phylogenetic relationships among acute and early HIV-1-infected individuals in New York City
J Acquir Immune Defic Syndr
 , 
2012
, vol. 
61
 (pg. 
1
-
8
)
4
Brenner
BG
Roger
M
Moisi
DD
, et al.  . 
Transmission networks of drug resistance acquired in primary/early stage HIV infection
AIDS
 , 
2008
, vol. 
22
 (pg. 
2509
-
15
)
5
Yerly
S
Junier
T
Gayet-Ageron
A
, et al.  . 
The impact of transmission clusters on primary drug resistance in newly diagnosed HIV-1 infection
AIDS
 , 
2009
, vol. 
23
 (pg. 
1415
-
23
)
6
Harrigan
PR
Bloor
S
Larder
BA
Relative replicative fitness of zidovudine-resistant human immunodeficiency virus type 1 isolates in vitro
J Virol
 , 
1998
, vol. 
72
 (pg. 
3773
-
8
)
7
Deeks
SG
Wrin
T
Liegler
T
, et al.  . 
Virologic and immunologic consequences of discontinuing combination antiretroviral-drug therapy in HIV-infected patients with detectable viremia
N Engl J Med
 , 
2001
, vol. 
344
 (pg. 
472
-
80
)
8
Pingen
M
Nijhuis
M
de Bruijn
JA
, et al.  . 
Evolutionary pathways of transmitted drug-resistant HIV-1
J Antimicrob Chemother
 , 
2011
, vol. 
66
 (pg. 
1467
-
80
)
9
Brenner
BG
Routy
JP
Petrella
M
, et al.  . 
Persistence and fitness of multidrug-resistant human immunodeficiency virus type 1 acquired in primary infection
J Virol
 , 
2002
, vol. 
76
 (pg. 
1753
-
61
)
10
Audelin
AM
Gerstoft
J
Obel
N
, et al.  . 
Molecular phylogenetics of transmitted drug resistance in newly diagnosed HIV type 1 individuals in Denmark: a nation-wide study
AIDS Res Hum Retroviruses
 , 
2011
, vol. 
27
 (pg. 
1283
-
90
)
11
Kouyos
RD
von Wyl
V
Yerly
S
, et al.  . 
Ambiguous nucleotide calls from population-based sequencing of HIV-1 are a marker for viral diversity and the age of infection
Clin Infect Dis
 , 
2011
, vol. 
52
 (pg. 
532
-
9
)
12
Huigen
MC
de Graaf
L
Eggink
D
, et al.  . 
Evolution of a novel 5-amino-acid insertion in the β3-β4 loop of HIV-1 reverse transcriptase
Virology
 , 
2007
, vol. 
364
 (pg. 
395
-
406
)
13
Boucher
CA
Keulen
W
van Bommel
T
, et al.  . 
Human immunodeficiency virus type 1 drug susceptibility determination by using recombinant viruses generated from patient sera tested in a cell-killing assay
Antimicrob Agents Chemother
 , 
1996
, vol. 
40
 (pg. 
2404
-
9
)
14
Boom
R
Sol
CJ
Salimans
MM
, et al.  . 
Rapid and simple method for purification of nucleic acids
J Clin Microbiol
 , 
1990
, vol. 
28
 (pg. 
495
-
503
)
15
Buskin
SE
Ellis
GM
Pepper
GG
, et al.  . 
Transmission cluster of multiclass highly drug-resistant HIV-1 among 9 men who have sex with men in Seattle/King County, WA, 2005–2007
J Acquir Immune Defic Syndr
 , 
2008
, vol. 
49
 (pg. 
205
-
11
)
16
Hue
S
Gifford
RJ
Dunn
D
, et al.  . 
Demonstration of sustained drug-resistant human immunodeficiency virus type 1 lineages circulating among treatment-naive individuals
J Virol
 , 
2009
, vol. 
83
 (pg. 
2645
-
54
)
17
Lindstrom
A
Ohlis
A
Huigen
M
, et al.  . 
HIV-1 transmission cluster with M41L ‘singleton’ mutation and decreased transmission of resistance in newly diagnosed Swedish homosexual men
Antivir Ther
 , 
2006
, vol. 
11
 (pg. 
1031
-
9
)