AIDS, antiretroviral drugs, human immunodeficiency virus, mutations, pol gene, resistance

Transcription

1 ORIGINAL ARTICLE Peptide insertions in reverse transcriptase pol gene of human immunodeficiency virus type 1 as a rare cause of persistent antiretroviral therapeutic failure Véronique Schneider 1,Jérôme Legoff 2, Laurent Bélec 2,3, Nathalie Delphin 1, Corinne Dutreuil 1, Ali Kara-Mostefa 1, Willy Rozenbaum 4 and Jean-Claude Nicolas 1 1 Laboratoire de Virologie, Hôpital Tenon, 2 Laboratoire de Virologie, Hôpital Européen Georges Pompidou, 3 Unité INSERM U430 (Immunopathologie humaine), Institut de Recherche Biologique des Cordeliers, and Université Pierre & Marie Curie (Paris VI), 4 Service des Maladies Infectiuses et Tropicales, Hôpital Tenon, Paris, France ABSTRACT Peptide insertions in codons of the reverse transcriptase (RT) pol gene were detected in 11 (2.7%) of 414 genotypic analyses performed in a hospital cohort of 2900 outpatients with human immunodeficiency virus type 1 (HIV-1) infection. The duration of antiretroviral treatment (bi- or tri-therapy) before the detection of insertions ranged from 12 to 60 months. Dipeptide insertions were detected in ten patients, of which the most frequent was serine serine. A monopeptide insertion was diagnosed once. The amino-acid composition patterns of insertions varied with time in five of the 11 patients. Peptide insertions were always associated with various patterns of pre-existing or appearing resistance mutations in the RT pol gene to different antiretroviral drugs. Genotypic-guided treatment resulted in virological and immunological improvement in two patients. In contrast, the remaining patients did not respond to any of the various antiretroviral regimens prescribed. Furthermore, various patterns of resistance mutations developed to the prescribed antiretroviral drugs, with AIDS-related conditions leading to death in two patients. It was concluded that peptide insertion in this region of the HIV-1 RT pol gene constitutes a rare cause of persistent therapeutic failure, and that management of such patients remains challenging despite successive genotypic analyses aimed at detecting mutations conferring antiretroviral drug resistance. Keywords AIDS, antiretroviral drugs, human immunodeficiency virus, mutations, pol gene, resistance Original Submission: 17 December 2002; Revised Submission: 21 April 2003; Accepted: 30 April 2003 Clin Microbiol Infect 2004; 10: INTRODUCTION Selection of drug-resistant human immunodeficiency virus type 1 (HIV-1) mutant variants is a major cause of antiretroviral treatment failure. Resistance to HIV reverse transcriptase (RT) inhibitors usually results from base-pair mutations leading to amino-acid substitutions in the RT pol gene. The substitution mutation Q151M, Corresponding author and reprint requests: J. Legoff, Pharm D, Laboratoire de Virologie, Hôpital Européen Georges Pompidou, 20, rue Leblanc, Paris Cedex 15, France Tel: Fax: jerome.legoff@hop.egp.ap-hop-paris.fr or insert-deletion mutations conferring multidrug resistance (MDR), may occur in a minority of cases [1 7]. MDR-related insertions generally consist of two amino-acids inserted following codon 69 of the RT pol gene, most frequently serine serine (S S) and serine glycine (S G), and these are associated with high phenotypic resistance to various nucleoside RT inhibitor (NRTI) analogues [6]. Management of patients carrying MDR variants remains challenging [2 4,7], but the outcome of peptide insertions has, until now, been poorly documented. Previously reported follow-ups of patients carrying variants with peptide rearrangements in the pol gene have been limited to < 1 year [6,7]. The present study analysed the clinical, virological and immunological outcome of a cohort of 11 patients carrying Ó 2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases

2 128 Clinical Microbiology and Infection, Volume 10 Number 2, February 2004 HIV-1 variants showing rearrangements in codons of the RT pol gene region over a period of 2 6 years. PATIENTS AND METHODS Study population and sample processing The Service des Maladies Infectieuses et Tropicales at Rotschild Hospital, Paris, follows a cohort of 2900 HIV-infected outpatients. Between February 1996 and June 1999, routine genotypic analyses for drug-resistance mutations were performed for 414 patients. Eleven (2.7%) patients whose HIV-1 variants showed peptide rearrangements in the RT pol gene were included in a prospective follow-up. Plasma was separated from EDTA-anticoagulated blood and stored in aliquots at ) 80 C until used for further analyses. Virological and immunological surrogate markers HIV-1 RNA in plasma was quantified by branched-chain DNA signal amplification. During the study period, three generations of assays were used successively (Quantiplex HIV-RNA v.1.0, v.2.0 and v.3.0; Chiron Diagnostics, Emeryville, CA, USA), with detection limits of , 500 and 50 copies ml, respectively. Absolute numbers of CD4 T-lymphocytes were determined in whole blood by flow cytometry with a FACScan cytofluorometer and Cellquest software (Becton Dickinson, Mountain View, CA, USA). CD4 T-cell counts and plasma HIV RNA levels were measured at baseline, and periodically thereafter. Virological or immunological failures following therapy were defined as a viral load of > copies ml, or as an absolute number of CD4 T cells of < 200 ll, respectively, after antiretroviral therapy for 6 months. A significant variation in plasma viral load or in CD4 T-cell count was defined as a variation of at least 0.5 log copies ml, or cells L, between two determinations separated by at least 6 months. Sequencing of reverse transcriptase and protease Reverse transcriptase and protease gene sequences were obtained from plasma cell-free HIV-1 RNA, as described previously [6,8], but with slight modifications in the primer sets used. In brief, HIV-1 virions in 1 ml of plasma were pelleted by centrifugation at g for 60 min. The pellets were resuspended in 140 ll of RNAse-free sterile distilled water, and RNA was recovered by selective binding to a silica-based membrane in a spin column (High Pure RNA Isolation Kit; Boehringer, Mannheim, Germany). RNA was reverse-transcribed to cdna, and the protease p32 pol gene was amplified by nested PCR with sets of specific primers. Two RT gene fragments (RT1: codons 6 152; RT2: codons ) were amplified with the previously published forward internal primers A and B [9], and the reverse internal primers 5 -GGTGATCCTTTCCATCC-3 for RT1 and 5 -TCATTGGACAGTCCAGCT-3 for RT2. The protease gene was amplified by nested PCR with forward and reverse external primers 5 -ATAATCCACCTATCCCAGTAGGAGG- AAAT-3 and 5 -GGTGATCCTTTCCATCC-3, respectively; the forward and reverse internal primers were 5 -AGA- AGAGAGCTCCCAGGTTTGG-3 and 5 -AGTCTCAATAGG ACTAATGGG-3, respectively. The final PCR products were purified, followed by direct sequencing of amplicons with an automated sequencer. Two opposing strands from each PCR product were sequenced using the 5 and 3 inner primers and aligned with Sequencer Navigator v software (Applied Biosystems, Warrington, UK). The sequences were proof-read manually and aligned with the HIV-1 subtype B consensus sequence from the Los Alamos HIV sequence database for comparison. Primary and secondary mutations known to be associated with antiretroviral resistance were recorded [10]. Other mutations, considered natural polymorphisms of the protease gene [11] or which have not been previously described, were not recorded. HIV-1 subtyping HIV-1 subtypes were assessed by the BLAST program ( Sequences yielding ambiguous results were further compared to reference sequences by phylogenetic analysis using the parsimony method [12]. RESULTS Study population Eleven (2.7%) of 414 patients for whom a resistance genotype was determined showed peptide rearrangements in the RT pol gene, and were included in the prospective follow-up. Details of virus subtypes, surrogate markers, duration of follow-up, resistance mutation profiles, and peptide rearrangements in the RT pol gene are given in Table 1. All but one of the HIV-1 variants were of the B subtype. The estimated duration of antiretroviral treatment before the appearance of insertions ranged from 12 months (patients 1 and 9) to 60 months (patient 11), with a median of 36 months (Fig. 1). Most patients were prescribed various bi- and tri-combination therapies at the time of peptide insertion occurrence (Fig. 1). Patient 11 had taken zidovudine (ZDV) monotherapy for > 5 years, followed by zalcitabine monotherapy (Fig. 1). Molecular analysis and temporal evolution of peptide insertions In patient 11, the peptide rearrangement consisted of a single tyrosine, coded by TAC, inserted between codons 67 and 68 of the RT pol gene, in association with the D67G, T69D and K70R

3 V. Schneider et al. HIV-1 pol gene insertions 129 mutations (Table 1). This insertion, and the associated flanking resistance mutations, appeared to be stable during the follow-up. In the ten other patients (1 10), the peptide rearrangements consisted of double amino-acid insertions between codons 68 and 69 of the RT pol gene (Table 1). Pre-insertion RT pol gene sequences obtained in patients 4 7 and 10 showed the wild-type sequence (DSTKW) for codons before the appearance of the insertion, while they showed the mutations T69A, D67N and K70R, respectively, in patients 2, 3 and 9. The dipeptide insertion between codons 68 and 69 of the RT pol gene consisted of alanine serine (A S) in patient 6, of S S in patients 1, 3, 5 and 9, and of serine glutamine (S Q) in patient 7. In these five latter patients, the temporal evolution of the double amino-acid insertions showed stable rearrangements during the course of follow-up (Table 1). In the four remaining patients (2, 4, 8 and 10), the dipeptide insertion evolved with time. In patients 4 and 10, the A S insertion evolved into an alanine valine (A V) insertion. In patient 2, the S S insertion evolved into an A S insertion. Finally, in patient 8, the S-S insertion evolved into a serine cysteine (S C) insertion. In total, 14 different dipeptide insertions were identified in the successively analysed RT pol gene sequences, namely six of S S, four of A S, two of A V, one of S C, and one of S Q. The first aminoacid of the dipeptide insertions was a serine in eight sequences, and an alanine in six sequences. The serine was coded for by different nucleotide triplets, namely TCC (five times), TCT (two), and AGC (one), while the alanine was always coded for by the GCT triplet. The second amino-acid of the dipeptide insertions was, in decreasing frequency, a serine (ten sequences, coded by the AGT triplet), a valine (two sequences, coded by GTT), a glutamine and a cysteine. The sequences between codons flanking the dipeptide rearrangements may be described as follows. The minor ZDV-resistance D67N and K70R mutations disappeared when the dipeptide insertion was identified for the first time. In patients 1, 2, 5 and 9, a D67E mutation was noticed post-insertion. In patient 3, codon 67 was originally an asparagine (N), which mutated to a glutamate (E) at the time of the insertion. A mutation of codon 68 (S68G) was only observed in patient 10. The wild-type amino-acid T69 was mutated in nine of ten patients. In patients 4, 6 and 8, the mutation in codon 69 was T69A. In patients 3 and 7, the mutation was T69S, while in patient 10, it was T69G. In patient 5, codon 69 first mutated to serine, and subsequently to glycine, while in patient 9, a first mutation in codon 69 to alanine was followed by a change to glycine. In patient 2, codon 69 was an alanine before the insertion, which mutated to glycine and then to serine post-insertion. Finally, codons 70 and 71 were wild-type (KW) after selection of the dipeptide insertion in all patients. The viability of the HIV-1 strains containing the amino-acid insertions in the RT pol gene was checked by cell culture of blood mononuclear cells from four patients (1, 3, 9 and 10). The RT pol gene sequences of HIV variants isolated from the culture supernatants were identical to those obtained from the plasma samples. Associated resistance genotype Various patterns of resistance mutations were found consistently in patients whose HIV variants developed peptide insertions. Before amino-acid insertions, resistance mutation profiles were determined for eight patients. All these patients had received ZDV and had HIV variants showing mutations associated with ZDV resistance; in particular, the major resistance mutation T215Y was found in all cases. Minor ZDV resistance mutations comprised M41L (n ¼ 5), L210W (n ¼ 5), D67N (n ¼ 1) and K70R (n ¼ 1). Interestingly, the mutated codons 67N (patient 3) and 70R (patient 9) reverted to 67E and wild-type 70K codons, respectively, at the time of insertion. Before peptide insertion, lamivudine (3TC)-treated patients 2 and 7 had the major 3TC-resistance mutation M184V, which then disappeared when 3TC treatment was stopped. At the time of detection of peptide insertions, all patients had HIV variants harbouring the major ZDV resistance mutation T215Y F, frequently associated with the minor resistance mutations M41L and L210W. 3TC-treated patients (1, 6 and 10) developed the M184V mutation at a time close to that of insertion appearance. After detection of peptide insertions, the M184V I mutation appeared in patients 3 and 5, while it disappeared in patient 10 even though 3TC treatment was continued. The major didanosine (ddi) resistance mutation, L74V, appeared postinsertion in patient 1, who had then been taking ddi

4 130 Clinical Microbiology and Infection, Volume 10 Number 2, February 2004 Table 1. Demographic and biological findings in 11 patients showing HIV-1 variants with amino-acid insertions in reverse transcriptase pol gene Patient Sex/ HIV Age a subtype Followup b CD4 T-cell count/ll Plasma HIV-1 load (copies/ml) Antiretroviral therapy: sucessive combinations during the follow-up c Resistance mutations in RT pol gene d Resistance mutations in protease gene d Insertion Codons of RT pol gene e 1 M 32 B M ZDV fi ZDV+ddI fi ZDV+3TC fi 3TC+d4T+IDV M41L, A62V, M184V, 7 NA f NA 3TC+d4T+IDV M41L, M184V, TC+d4T+IDV M41L, A62V, M184V, 12 NA TC+ddC+RTV+SQV M41L, M184V, 31 1 > d4t+nfv+nvp fi ZDV+3TC+ddC +SQV+NFV fi ZDV +3TC+ABC+EFV M41L, A62V, L74V, K103N, M184V, G190A, 32 NA NA d4t+3tc+abc+efv M41L, L74V, V75M, K103N, M184V, G190A, ND S S DSSSTKW ND S S DSSSTKW L10I, K20I, M46I, I54V, L63P, A71V, V82T, L90M L10I, K20I, M36I, I54V, L63P, A71V, S S DSSSTKW S S DSSSTKW V82T, L90M ND S S ESSSTKW L10I, K20I, M36I, I54V, L63P, A71V, V82T, L90M S S ESSSTKW 2 M 39 B M0 53 NA ZDV+ddI M41L, ND Absent DSAKW TC+d4T+IDV M41L, M184I, ND Absent DSAKW TC+d4T+IDV M41L, M184V, ND S S DSSSAKW TC+d4T+IDV M41L, A62V, M184I, ND S S DSSSAKW 10 NA TC+d4T+ddI+IDV M41L, A62V, M184V, TC+d4T+ddI+IDV M41L, A62V, M184I, ND S S DSSSAKW (50%) g DSSSGKW (50%) g A S ESASGKW L10I, M46I, A71V, V77I, L90M d4t+rtv+sqv M41L, A62V, ND A S ESASGKW (50%) g ESASSKW (50%) g A S ESASSKW NA d4t+rtv+sqv M41L, A62V, L10I, M46I, G48V, I54V, L63P, A71V, V77I, L90M 3 M 42 B M ZDV fi ZDV+ddI fi ZDV+3TC M41L, T215Y ND Absent NSTKW ZDV+3TC M41L, T215Y ND S S ESSSTKW ZDV+3TC+IDV M41L, T215Y ND S S ESSSTKW ZDV+3TC+IDV M41L, A62V, T215Y ND S S ESSSTKW TC+d4T+NFV M41L, A62V, M184I, T215Y ND S S ESSSTKW 21 NA TC+d4T+NFV M41L, A62V, M184I, T215Y ND S S ESSSTKW TC+d4T+NFV M41L, A62V, M184I, T215Y ND S S ESSSSKW d4t+abc+efv M41L, A62V, L100I, K103N, ND S S ESSSSKW M184V, T215Y

5 V. Schneider et al. HIV-1 pol gene insertions F 39 B M ZDV fi ZDV+ddC M41L, ND Absent DSTKW ZDV+ddC fi ZDV+3TC M41L, A62V, ND A S DSASAKW d4t+3tc+idv M41L, A62V, L10I, M46I, L63P, A V DSAVAKW A71V, V77I NA ZDV+ddC+3TC+IDV M41L, A62V, L10I, M46I, L63P, A V DSAVAKW A71V, V77I, L90M d4t+rtv+sqv A V DSAVAKW fi d4t+3tc+rtv +SQV fi d4t+3tc+ NFV fi ABC+NVP+ NFV fi ABC+NVP+ RTV+SQV fi ABC +NVP+IDV M41L, A62V, L74I, Y181R, G190A L10I, M46I, L63P, A71V, G73S, V77I, I84V L90M 5 M 29 B M0 127 NA ZDV M41L, T215Y ND Absent DSTKW 5 NA NA ZDV+ddI M41L, T215Y ND S S DSSSSKW 7 40 NA ZDV+ddC M41L, T215Y ND S S DSSSSKW ZDV+3TC M41L, M184V, T215Y ND S S DSSSGKW ZDV+3TC+IDV fi 3TC+d4T+IDV ZDV+3TC+IDV fi 3TC+d4T +RTV+SQV TC+d4T +ABC+NVP M41L, V75M, K103N, M184I, T215Y M41L, V75M, M184I, T215Y L10I, I54V, V82T S S DSSSGKW M41L, V75M, M184I, T215Y ND S S ESSSGKW ND S S ESSSGKW 6 M 31 B M0 132 NA ZDV ND Absent DSTKW ddi fi ddc M41L, M184V, T215Y No mutation A S DSASAKW fi ZDV+3TC 7 F 40 A M ZDV fi ZDV+ddC fi ZDV+3TC fi ZDV+3TC+IDV M184V, ND Absent DSTKW ZDV+3TC+IDV M41L, M184V, ND Absent DSTKW ZDV+3TC+IDV M41L, M184V, L120W, T215F ND S Q DSSQSKW d4t+ddi+nfv fi d4t+ddi M41L, Y181C, L120W, T215F ND S Q DSSQSKW +RTV+SQV ddi+nvp +RTV+SQV M41L, Y181C, G190A, L120W, T215F 8 M 58 B M ZDV fi ZDV +ddc fi ZDV+3TC L10V, K20I, L24I, M36I, M46I, L63P, A71V, I84V S Q DSSQSKW M41L, No mutation S S DSSSAKW ZDV+3TC+IDV M41L, ND S S DSSSAKW S S DSSSAKW ZDV+3TC+IDV M41L, L10I, M46L, I54V, L63P, A71V, V82A ZDV+3TC+IDV M41L, ND S S DSSSAKW S C DSSCAKW TC+d4T+RTV +SQV fi 3TC +d4t+nvp M41L, V75A, Y181C, G190A, L120W, T215F L10V, M46L, I54V, L63P, A71V, V82A, L90M

6 132 Clinical Microbiology and Infection, Volume 10 Number 2, February 2004 Table 1. continued Patient Sex/ HIV Age a subtype Followup b CD4 T-cell count/ll Plasma HIV-1 load (copies/ml) Antiretroviral therapy: sucessive combinations Resistance mutations during the follow-up c in RT pol gene d Resistance mutations in protease gene d Insertion Codons of RT pol gene e 9 M 35 B M0 253 NA ZDV fi ZDV+ddI K70R, T215Y ND Absent DSTRW ZDV+ddI A62V, L210W, T215Y V77I S S DSSSAKW d4t+3tc A62V, L210W, T215Y ND S S DSSSGKW d4t+3tc A62V, L210W, T215Y ND S S ESSSGKW 10 M 51 B M0 NA NA ZDV T215Y ND Absent DSTKW NA ZDV+ddI L210W, T215Y ND Absent DSTKW NA ZDV+ddI fi ZDV+3TC M41L, M184V, ZDV+3TC M41L, M184V, ZDV+3TC+IDV fi d4t+3tc+idv M41L, L10I, L63P, A71V, V77I, V82A, L90M d4t+3tc+rtv+sqv M41L, L10I, G48V, A71V, V82T, L90M ND A S DSASGKW ND A S DSASGKW A V DGAVGKW A V DGAVGKW d4t+3tc+nfv M41L, ND A V DGAVGKW ND Y GYSDRW 11 M 30 B M0 NA ZDV fi ZDV+ddC fi ddc+idv T69D, K70R, T215F, K219Q TC+d4T+IDV T69D, K70R, T215F, K219Q TC+d4T+IDV T69D, K70R, T215F, K219Q TC+d4T+IDV T69D, K70R, M184V, T215F, K219Q ddi+d4t+rtv+sqv A62V, T69D, K70R, T215F, K219Q ND Y GYSDRW M46I, L63P, V77I Y GYSDRW M46I, L63P, I84V Y GYSDRW ND Y GYSDRW a M, male; F, female. Age is given in years at beginning of follow-up. b Duration of the follow-up in months. The follow-up represents retrospective evaluation of plasma samples kept frozen, and prospective evaluation of new collected plasma samples. M0, first month of follow-up. c Antiretroviral therapy taken before CD4 + T-cell counts, plasma HIV-1 RNA loads and resistance genotype determinations. d All amino-acid changes related to major or minor resistance are given, as described [10]. e Peptide sequences of the RT pol gene in codons Mutations not associated with a known genotype conferring resistance to reverse transcriptase inhibitors are shown in bold. f NA: not available. g In brackets: frequency of each sequence. ZDV, zidovudine; ddi, didanosine; ddc, zalcitabine; d4t, stavudine; 3TC, lamivudine; ABC, abacavir; NVP, nevirapine; EFV, efavirenz; IDV, indinavir; RTV, ritonavir; SQV, saquinavir; NFV, nelfinavir; ND, not determined; RT, reverse transcriptase; A, alanine; C, cysteine; D, aspartic acid; E, glutamic acid; G, glycine; K, lysine; N, asparagine; S, serine; R, arginine; Q, glutamine; T, threonine; V, valine; W, tryptophan; Y, tyrosine; S S, serine serine; A S, alanine serine; A V, alanine valine; S C, serine cysteine; S Q, serine glutamine.

7 V. Schneider et al. HIV-1 pol gene insertions 133 Fig. 1. Schematic representation of treatment history and codons of the RT pol gene for 11 patients whose HIV-1 variants had peptide rearrangements. Black circles depict nucleoside reverse transcriptase inhibitor; black triangles depict non-nucleoside reverse transcriptase inhibitor; black squares depict protease inhibitor. Antiretroviral drug regimens are shown in detail in Table 1. Vertical lines indicate therapeutic shift. Open circles indicate strains without insertions. Grey circles indicate insertion-containing strains. Triangles, diamonds and squares in grey indicate genetic evolution in codons Open arrows indicate time of first diagnosis of insertion rearrangement. for 11 months. The major stavudine (d4t) resistance mutation V75M A appeared in patients 5 (V75M) and 9 (V75A) after 6 and 19 months, respectively, of post-insertion d4t treatment. Response to treatment and clinical-biological outcome The successive types of therapeutic associations of antiretroviral drugs are depicted in Fig. 1 and listed in detail in Table 1. The variations in CD4 T-cell count and HIV-1 RNA load in plasma are shown for each patient in Fig. 2. Before an insertion was diagnosed, all patients were in therapeutic failure as defined by both virological (viral load of > copies ml) and immunological (CD4 count of < 200 ll) criteria, except for patient 9 whose CD4 T-cell count ranged from 200 to 350 ll, and whose circulating viral load was copies ml at the time of detection of insertions (Fig. 2). Two patients (6 and 9) showed a favourable immunological, virological and clinical evolution following triple antiretroviral treatment commenced after peptide rearrangement discovery. Patient 6 was in treatment failure when the insertion was discovered, with a viral load of > copies ml and a CD4 T-cell count of 12 ll. After starting antiprotease treatment (indinavir, 4 months; indinavir plus d4t, 13 months; indinavir plus d4t plus 3TC, 19 months), the viral load dropped below detectable limits (< 50 copies ml) and the CD4 T-cell count increased from 12 ll to 400 ll within 3 years. At discovery of the dipeptide insertion, patient 9 had a viral load of copies ml and a CD4 T-cell count of 290 ll. With triple therapy (indinavir plus two NRTIs, 3 months the indinavir was stopped because of side-effects to this drug; nelfinavir plus two NRTIs, 12 months; efavirenz plus two NRTIs, 7 months), the viral load dropped to < 50 copies ml and the CD4 T-cell count increased from 290 ll to 600 ll within 2 years. In contrast, the other nine patients showed therapeutic failure. Nine patients (1 5, 7, 8, 10 and 11) received NRTI treatment in combination with indinavir after the insertions were discovered. These patients were then treated unsuccessfully with various antiretroviral combinations comprising ritonavir plus saquinavir, or nelfinavir, or one of the non-nucleoside RT inhibitors (NNRTIs) nevirapine or efavirenz. Major protease inhibitor resistance mutations (M46I, G48V, V82A T, T84V and L90M) accumulated in most patients during the course of follow-up. Six patients (1, 3, 4, 5, 7 and 8) selected HIV variants with major resistance

8 134 Clinical Microbiology and Infection, Volume 10 Number 2, February 2004 Fig. 2. Temporal evolution of circulating CD4 T-cell count (CD4 as black diamonds in cells ll) and HIV-1 RNA viral load (VL as open circles in log copies ml) for 11 patients whose HIV-1 variants contained peptide rearrangements in the RT pol gene. Vertical lines represent the time of the first discovery of insertion rearrangements. mutations to NNRTIs: L100I (1 6), K101E (1 6), K103N (3 6), Y181C (2 6) and G190A (4 6). No significant increase (> 0.5 log) in plasma HIV-1 RNA load or decrease in the CD4 T-cell count was observed in these patients post-insertion. In patients 1 5 and 10, the triple antiretroviral therapy was not associated with a significant decrease (> 0.5 log in two successive samples). In patients 7, 8 and 11, a transitory decrease in circulating HIV-1 RNA load occurred for 15 months, 14 months and 6 months, respectively. During the follow-up period, patient 9 remained stable, without virological or immunological failure. Patient 6 showed favourable virological and immunological responses. All remaining patients remained in virological and immunological failure; all were suffering from at least one disease indicative of AIDS during the follow-up period. Patients 5 and 10 died 40 and 27 months, respectively, after the discovery of peptide insertions. DISCUSSION In the present study, mono or dipeptide insertions in the fingers subdomain of the HIV-1 RT pol gene constituted a rare and unstable event, occurring in the context of antiretroviral therapeutic failure. Long-term follow-up and management of patients carrying HIV variants with

9 V. Schneider et al. HIV-1 pol gene insertions 135 peptide rearrangements in codons of the RT pol gene showed clearly that such genetic change is associated frequently with various patterns of resistance mutations to antiretroviral drugs and with sustained therapeutic failure. These observations raise the issue of peptide insertion rearrangements in the RT pol gene as a rare cause of persistent therapeutic failure in patients treated with antiretroviral drugs. In our institution, rearrangement insertions in the RT pol gene that conferred MDR were detected in plasma HIV-1 RNA from a minority (2.7%) of patients in therapeutic failure. Similar low prevalences have been reported previously in hospital cohorts in which HIV-1 RT pol gene rearrangements could be found in % of multidrug-experienced patients for whom an HIV genotypic resistance assay was carried out [5 7,13,14]. In the present series, the most frequent dipeptide insertion was S S (6 14); other dipeptide insertions were A S, A V, S C and S Q. Monopeptide insertions were detected in < 10% of HIV variants with RT pol gene rearrangements. Peptide insertions were seen in HIV variants of both A and B subtypes. The peptide insertions appeared after variable treatment durations of not less than 1 year, in agreement with a previous report [4]. The insertions were observed only in patients whose initial antiretroviral treatment was either monotherapy or bitherapy with nucleoside analogues. However, peptide insertions in the RT pol gene have also been described in patients who received triple therapy initially with ZDV plus 3TC plus indinavir [3]. Thus, peptide insertion selection is not associated with a unique combination of antiretroviral drug classes or regimen. The predominance of dipeptide over monopeptide insertions, and the patterns of peptide rearrangements in the RT pol gene, are similar to series published previously [6]. The findings further indicate that the amino-acid composition pattern of insertions may vary with time in about half of patients [4]. Peptide insertions were always associated with various patterns of preexisting or appearing resistance mutations to various antiretroviral drugs located in the RT pol gene or in codons flanking the dipeptide rearrangements, as also described previously [6]. As hypothesised previously [15], the variability and instability of the insertion, and the flanking amino-acids, may be associated with a progressive selection of variants with increased fitness. Taken together, these observations emphasise that codons constitute an unstable, highly plastic region of the RT pol gene that is prone to extensive genetic heterogeneity, giving rise to peptide rearrangements, nucleotide substitutions or resistance mutations. Despite several antiretroviral shifts, most of the patients in this study remained in virological and immunological failure over 2 6 years. These findings indicate that the management of patients carrying HIV variants with peptide insertions in the RT pol gene remains challenging [6,7]. Dipeptide insertions in codon 69 have been found to be associated with moderate to high levels of phenotypic resistance to several NRTIs, even in the absence of mutations in the RT pol gene known to confer resistance to these drugs, thereby demonstrating that such rearrangements are effective MDR inserts [6,7]. Furthermore, the numerous associated resistance mutations and the insertion are thought to act synergically to increase resistance [4]. The extensive cross-resistance to NRTIs, and the rapid selection of resistance to protease inhibitors and NNRTIs when prescribed, as in the patients described above, render the choice of antiretroviral regimens extremely difficult. The proposed therapeutic options and genotypicguided treatment allowed effective virological and immunological improvement in two patients (6 and 9) despite cross-resistance to NRTIs. Both patients had a long-term beneficial response, with an undetectable viral load and a progressive increase in CD4 T-cell count, after the introduction of indinavir in patient 6, and of indinavir followed successively by nelfinavir and an NNRTI in patient 9. These patients received two new antiretroviral drugs, including one protease inhibitor shortly after the appearance of an insertion. Patient 6 received d4t and indinavir in addition to 3TC, and patient 9 received zalcitabine (ddc) and indinavir, followed 3 months later by ddc and nelfinavir in addition to d4t. Antiretroviral shifts consisted of only one drug change in patients 3, 5, 8 and 10, while a protease inhibitor already formed part of the antiretroviral therapy in patients 1, 2, 7 and 11 before the detection of amino-acid insertions. In contrast to patients 6 and 9, most of the patients studied did not respond, either virologically or immunologically, to any of the various antiretroviral regimens prescribed. All selected various patterns

10 136 Clinical Microbiology and Infection, Volume 10 Number 2, February 2004 of resistance mutations to both NNRTIs and protease inhibitors following their introduction, and all developed AIDS-related conditions, leading to death in two patients. Salvage therapy in patients harbouring HIV-1 variants with dipeptide insertions should preferably include two new antiretroviral drugs, probably including a protease inhibitor, as in patients 6 and 9, but with no guarantee of success. In conclusion, the insertion of amino-acid(s) in the fingers subdomain of the HIV-1 RT pol gene occurs rarely in patients in therapeutic failure, but is associated with various patterns of crossresistance to most antiretroviral drugs. The management of such patients is challenging, even with repeated genetic determinations of the resistance pattern of the circulating HIV variants. ACKNOWLEDGMENTS We would like to thank Dr Christophe Goujon for determining the HIV subtypes. REFERENCES 1. Larder BA, Bloor S, Kemp SD et al. A family of insertion mutations between codons 67 and 70 of human immunodeficiency virus type 1 reverse transcriptase confer multinucleoside analog resistance. Antimicrob Agents Chemother 1999; 43: de Jong JJ, Goudsmit J, Lukashov VV et al. Insertion of two amino acids combined with changes in reverse transcriptase containing tyrosine-215 of HIV-1 resistant to multiple nucleoside analogs. AIDS 1999; 13: Ross L, Johnson M, Graham N et al. The reverse transcriptase codon 69 insertion is observed in nucleoside reverse transcriptase inhibitor-experienced HIV-1-infected individuals, including those without prior or concurrent zidovudine therapy. J Hum Virol 1999; 2: Winters MA, Coolley KL, Girard YA et al. A 6-basepair insert in the reverse transcriptase gene of human immunodeficiency virus type 1 confers resistance to multiple nucleoside inhibitors. J Clin Invest 1998; 102: Tamalet C, Izopet J, Koch N et al. Stable rearrangements of the beta3-beta4 hairpin loop of HIV-1 reverse transcriptase in plasma viruses from patients receiving combination therapy. AIDS 1998; 12: F Masquelier B, Race E, Tamalet C et al. Genotypic and phenotypic resistance patterns of human immunodeficiency virus type 1 variants with insertions or deletions in the reverse transcriptase (RT): multicenter study of patients treated with RT inhibitors. Antimicrob Agents Chemother 2001; 45: Andreoletti L, Weiss L, Si-Mohamed A et al. Multidrugresistant HIV-1 RNA and proviral DNA variants harboring new dipeptide insertions in the reverse transcriptase pol gene. J Acquir Immune Defic Syndr 2002; 29: Karmochkine M, Si Mohamed A, Piketty C et al. The cumulative occurrence of resistance mutations in the HIV-1 protease gene is associated with failure of salvage therapy with ritonavir and saquinavir in protease inhibitor-experienced patients. Antiviral Res 2000; 47: Larder BA, Boucher CAB. PCR detection of human immunodeficiency virus drug resistance mutations. In: Persing, DH, ed. Diagnostic Molecular Microbiology: Principles and Applications, Vol. 1 Washington DC: American Society for Microbiology; 1993: Hirsch MS, Brun-Vezinet F, D Aquila RT et al. Antiretroviral drug resistance testing in adult HIV-1 infection: recommendations of an International AIDS Society-USA Panel. JAMA 2000; 283: Kozal MJ, Shah N, Shen N et al. Extensive polymorphisms observed in HIV-1 clade B protease gene using high-density oligonucleotide arrays. Nat Med 1996; 2: Korber B, Hahn B, Foley B et al. HIV Sequence Compendium Los Alamos, NM: Theoretical Biology and Biophysics Group T10, Los Alamos National Laboratory; Van Laethem K, Van Vaerenbergh K, Schmit JC et al. Phenotypic assays and sequencing are less sensitive than point mutation assays for detection of resistance in mixed HIV-1 genotypic populations. J Acquir Immune Defic Syndr 1999; 22: Balotta C, Violin M, Monno L et al. Prevalence of multiple dideoxynucleoside analogue resistance (MddNR) in a multicenter cohort of HIV-1-infected Italian patients with virologic failure. J Acquir Immune Defic Syndr 2000; 24: Boyer PL, Lisziewicz J, Lori F et al. Analysis of amino insertion mutations in the fingers subdomain of HIV-1 reverse transcriptase. J Mol Biol 1999; 286: