Human immunodeficiency virus-1 B and non-B subtypes with the same drug resistance pattern respond similarly to antiretroviral therapy.

We analysed the 12-week virological response to protease inhibitor (PI) or non-nucleoside reverse transcriptase inhibitor (NNRTI) therapy in 1108 patients carrying B or non-B human immunodeficiency virus (HIV)-1 subtypes with matched resistance mutation patterns. Response rates were not significantly different for non-B and B subtypes stratified for treatment status (51.5% vs. 41.5% in naïve patients; 46.7% vs. 38.7% in experienced patients) or regimens (46.9% vs. 39.7% with PI; 56.7% vs. 40% with NNRTI). No difference in response was detected in patients harbouring B and non-B subtypes with any resistance profile. Further studies are advisable to fully test this approach on larger datasets.

Changing patterns in HIV-1 non-B clade prevalence and diversity in Italy over three decades.

BACKGROUND: HIV-1 non-B subtypes have recently entered Western Europe following immigration from other regions. The distribution of non-B clades and their association with demographic factors, over the entire course of the HIV-1 epidemic, have not been fully investigated in Italy. METHODS: We carried out a phylogenetic analysis of HIV-1 pol sequences derived from 3670 patients followed at 50 Italian clinical centres over nearly three decades. RESULTS: Overall, 417 patients (11.4%) carried non-B subtypes. The prevalence of non-B strains increased from 2.6% in 1980-1992 to 18.9% in 1993-2008 (P<0.0001) in a subset of 2479 subjects with a known year of diagnosis. A multivariate analysis on a subset of 1364 patients for whom relevant demographic data were available indicated that African ethnicity, heterosexual route of infection and year of diagnosis were independently associated with non-B HIV-1 infection (P ≤ 0.0001). All pure subtypes, except for clade K, and seven circulating recombinant forms were detected, accounting for 56.6 and 34.1% of the non-B infections, respectively. The F1 subtype was the most prevalent non-B clade among Europeans and was acquired heterosexually in half of this patient population. Unique recombinant forms accounted for 9.4% of the non-B sequences and showed a B/F1 recombination pattern in one-third of cases. CONCLUSIONS: The circulation of non-B clades has significantly increased in Italy in association with demographic changes. Spread of the F1 subtype and B/F recombinants appears to predominate, which may result in a redistribution of the relative proportions of the different strains, and this could lead to overlapping epidemics. Thus, the HIV-1 landscape in Italy may in future be distinct from that of the rest of Europe.

Tracing the HIV-1 subtype B mobility in Europe: a phylogeographic approach.

BACKGROUND: The prevalence and the origin of HIV-1 subtype B, the most prevalent circulating clade among the long-term residents in Europe, have been studied extensively. However the spatial diffusion of the epidemic from the perspective of the virus has not previously been traced. RESULTS: In the current study we inferred the migration history of HIV-1 subtype B by way of a phylogeography of viral sequences sampled from 16 European countries and Israel. Migration events were inferred from viral phylogenies by character reconstruction using parsimony. With regard to the spatial dispersal of the HIV subtype B sequences across viral phylogenies, in most of the countries in Europe the epidemic was introduced by multiple sources and subsequently spread within local networks. Poland provides an exception where most of the infections were the result of a single point introduction. According to the significant migratory pathways, we show that there are considerable differences across Europe. Specifically, Greece, Portugal, Serbia and Spain, provide sources shedding HIV-1; Austria, Belgium and Luxembourg, on the other hand, are migratory targets, while for Denmark, Germany, Italy, Israel, Norway, the Netherlands, Sweden, Switzerland and the UK we inferred significant bidirectional migration. For Poland no significant migratory pathways were inferred. CONCLUSION: Subtype B phylogeographies provide a new insight about the geographical distribution of viral lineages, as well as the significant pathways of virus dispersal across Europe, suggesting that intervention strategies should also address tourists, travellers and migrants.

Primary HIV-1 resistance in recently and chronically infected individuals of the Italian Cohort Naive for Antiretrovirals.

The risk of acquiring HIV-1 drug resistance at time of infection has become a public health problem following the widespread use of antiretroviral drugs in developed countries. Although a number of studies have reported data regarding the prevalence of HIV-1 primary resistance in developed countries over the past years, limited knowledge is available regarding the proportion of mutations related to drug resistance in antiretroviral naive subjects with chronic HIV-1 disease. In this study, we evaluated the prevalence of mutations in the reverse-transcriptase (RT) and protease region both in a representative group of recently HIV-1 infected subjects (n=68) and a cohort of chronically-infected HIV-positive patients (n=347) enrolled in the Italian Cohort of Antiretroviral Naive patients (I.CO.NA.). In recently infected individuals, the overall prevalence of mutations for nucleoside RTI (NRTIs) was 10/68 (14.7%). The distribution of mutations by calendar year were 0, 1 in 1996, 9, 3 in 1997 and 1, 0 in 1998 for NRTIs and protease inhibitors (PIs) respectively. Thymidine associated mutations were identified in six subjects (8.8%), five of whom had one mutation [41L, 70K (n=2), 215Y] and one had two mutations (67N+219Q). Four subjects (5.9%) showed the changes associated with resistance to lamivudine (184V or 118I). No non nucleoside-RTI (NNRTI) mutations were present in the study period. Primary PIs mutations (two 46L and two 82I) were present in four subjects (5.9%). Of note, mutations related to resistance to more than one class of antiretrovirals were present in one (1.5%). Among patients with chronic infection a large proportion (88.5%) carried no mutations in RT region, 11.5% individuals carried one or more mutations associated with resistance to NRTI (7.8%), or NNRTI (4.9%), with 4 patients carrying mutations to both classes. Among mutations associated with high-level resistance to RTI, T215Y was found in only 2 patients, M184V in 2 cases, T69D in another case, and K103N in only 1 patient, for a total of 6 patients (one carrying both T215Y and M184V) (1.7%). Primary mutations associated with substantial resistance to PIs were found in only 5/347 patients (1.4%); all the other patients carried only secondary mutations. Prevalence of mutations associated with high-level resistance to antiretroviral drugs is stable in recently infected individuals and low in patients with established HIV infection. The potential impact of transmitted mutations on the response to first regimen in individuals carrying transmitted mutations needs to be assessed by prospective studies.

Role of CCR5, CCR2 and SDF-1 gene polymorphisms in a population of HIV-1 infected individuals.

The finding that in addition to CD4 molecule HIV-1 uses, CCR5 or CXCR4 receptors to enter target cells prompted the research to identify polymorphisms in coreceptor genes affecting disease progression. In this study we analyzed the prevalence of CCR5-delta32, CCR2-641 and SDF1-3'A alleles in a highly selected group of 42 Long-Term Nonprogressors (LTNPs) compared to 112 subjects with a typical course of HIV-1 infection (TPs) and 117 healthy controls (HCs). In addition, we correlated CCR5, CCR2 and SDF-1 genotypes with molecular indexes of HIV-1 replication, cell-free RNA and both unspliced (US) and multiply spliced (MS) intracellular transcripts, to investigate the role of the mutant alleles in determining a long-term nonprogressive course of HIV-1 disease. Our results indicate a significantly higher prevalence of CCR5-delta32 allele in LTNPs compared to TPs (p=0.0434), while the proportions of CCR2-64I and SDF1-3'A alleles were comparable between the two groups. However, SDF-1 wild type LTNP subjects showed significantly lower levels of HIV-1 genomic RNA, US and MS transcripts than SDF1-3'A heterozygous ones (p=0.0021, 0.016, 0.0031, respectively), whereas both CCR5 and CCR2 wild type individuals had similar rates of viral replication compared to CCR5-delta32 and CCR2-64I heterozygous ones. CCR5, CCR2 and SDF-1 combined genotypes were also studied and this analysis did not identify a specific protective cluster of alleles in LTNPs. Taken together, our results indicate that genetic background involving CCR5, CCR2 and SDF-1 alleles may play a limited role in the natural history of HIV-1 infection.

Secondary mutations in the protease region of human immunodeficiency virus and virologic failure in drug-naive patients treated with protease inhibitor-based therapy.

The role of mutations in protease (PR) and reverse-transcriptase (RT) of human immunodeficiency virus (HIV) in predicting virologic failure was assessed in 248 antiretroviral-naive HIV-positive patients who began a PR inhibitor-containing antiretroviral regimen. Genotypic testing was performed on plasma samples stored before the start of therapy. Twenty-seven patients (10.9%) had mutations in the RT, 5 (2%) carried primary mutations in the PR, and 131 (52.8%) showed only secondary PR mutations. Virologic failure at week 24 occurred in 62 (25.0%) of 248 patients. There was a statistically significant correlation between virologic failure and the number of PR mutations (P= .04, chi(2) test). Mutations at codons 10 and 36 of PR (present in 39.3% and 40.0% of patients in whom treatment failed, respectively) were identified by stepwise logistic regression as the strongest predictors of virologic failure (odds ratio, 2.20; 95% confidence interval, 1.30-3.75; P= .004). If confirmed in independent studies, this result may justify the increased use of HIV genotyping in drug-naive patients requiring antiretroviral therapy.

Increasing prevalence of non-clade B HIV-1 strains in heterosexual men and women, as monitored by analysis of reverse transcriptase and protease sequences.

OBJECTIVE: We evaluated the prevalence of HIV-1 non-clade B over time in a formerly clade B-restricted area. Protease and reverse transcriptase regions of the pol gene were used for phylogenetic and recombination analysis and for clade assignment to HIV-1 A-D, F-H, J, and K strains of the M group. METHODS: The pol gene of 349 HIV-1 patients belonging to the Italian Cohort Naive for Antiretrovirals (ICONA) were genotypically analyzed to study the prevalence of antiretroviral-associated resistance mutations. All HIV-1 pol sequences and 32 HIV reference strains were analyzed, including the reference strains for the major HIV-1 subtypes. The non-clade B sequences according to the HIV-1 Subtyping Tool program were further studied by a bootscan analysis (SimPlot) to investigate the likelihood of recombination between subtypes. RESULTS: Phylogenetic analysis detected 19 of 349 (5.4%) non-clade B subtypes. The proportions of patients carrying non-clade B virus before and after 1997 were 1.9% and 8.4%, respectively (p =.008). Among whites, heterosexual infection and female gender were significantly associated with the presence of non-clade B subtypes (p =.001 and.005, respectively). Non-clade B HIV-1 was harbored by 14.5% of the heterosexuals who were found to be HIV-1 positive after 1997, 60% of whom were women. Bootscan analysis identified four strains as F, two as A, one as C, one as G, and 11 (57.9 %) as non-clade B recombinant subtypes. CONCLUSION: Detection of HIV-1 subtypes and intersubtype recombinants in a previously clade B-homogeneous area indicates that the HIV-1 epidemic is evolving in Italy and that heterosexuals and women are at increased risk of infection with non-clade B HIV-1 subtypes. Sequences inferred from the pol gene yield to establish the subtype of circulating HIV-1 strains. As a consequence, genotyping of pol gene for testing resistance to antiretrovirals warrants concomitant surveillance of non-clade B subtypes.

Prevalence of multiple dideoxynucleoside analogue resistance (MddNR) in a multicenter cohort of HIV-1-infected Italian patients with virologic failure.

We evaluated the prevalence of both Q151M and 6-bp insert at position 69 of RT region responsible for multiple dideoxynucleoside analogue-resistant (MddNR) HIV-1 variants in 177 patients who failed to respond to combination therapy. Patients had received protease inhibitors (PI) and/or nonnucleoside reverse transcriptase inhibitors (NNRTIs) after a long-term experience with nucleoside reverse transcriptase inhibitors (NRTIs) (including zidovudine monotherapy). Two of 177 patients (1.1%) showed the specific complex of Q151M mutation, while 4 (2.3%) had the 69 6-bp insert. Mutations that belong to the 151 set in the absence of the pivotal Q151M substitution were detected in as many as 3.9% of the patients. One patient exhibited a 69S [VG] insert that has not been previously phenotypically characterized. This HIV-1 isolate had high levels of resistance to all NRTIs except stavudine. MddNR is an emerging problem after sequential therapy with this class of compounds among HIV-1-infected patients. Either didanosine (ddI) or zidovudine (ZDV) monotherapy allowed the emergence of MddNR variants containing Q151M complex. Monotherapy with ZDV and ddI or subsequent treatments with various NRTI combinations were the common background in the patients with the 69 insert. The overall prevalence of MddNR (3.4%) in Italy is comparable with that observed in several other European countries (3.4%-6.5%). These data suggest that patients failed by NRTI regimens should be analyzed for the presence of both patterns of MddNR.

Quantitative evaluation of the recombinant HIV-1 phenotype to protease inhibitors by a single-step strategy.

OBJECTIVE: To develop and optimize a fast and quantitative recombinant strategy for evaluating the HIV-1 phenotype to protease inhibitors (PI). DESIGN AND METHODS: A non-replicative HIV-1 molecular vector (designated pdelta prodelta env) capable of expressing exogenous HIV-1 protease-encoding sequences was developed in this study. The HIV-1 protease sequences were amplified from either viral isolates or plasma samples (both from 21 HIV-1-infected individuals, 19 of whom were failing different anti-HIV-1 combination treatments) and cloned in the pdelta prodelta env backbone. The HIV-1 recombinant phenotype to PI was determined directly after transfection of viral chimeric clones by measuring protease activity and calculating a percentage sensitivity index (SI%; the ratio between the results from each clone and those from a PI-sensitive reference strain). RESULTS: The SI% values obtained from the recombinant clones paralleled the IC50 results of the viral isolates and documented different degrees of resistance and cross-resistance to PI, compatible, with few exceptions, with the respective genotype. Interestingly, an inverse correlation between SI% values and the presence of primary mutations for resistance to PI (P = 0.0038 and P = 0.0414, for indinavir and ritonavir, respectively) and a difference in SI% between samples harbouring an increasing number of mutations (indinavir, P = 0.022; ritonavir, P = 0.0466) were observed. CONCLUSION: The data substantiate the reliability of the novel strategy for a fast (5 day) quantitative evaluation of HIV-1 phenotype to PI, and indicate that this method may contribute to the understanding of mechanisms of virus resistance to PI.

Prevalence of transmitted nucleoside analogue-resistant HIV-1 strains and pre-existing mutations in pol reverse transcriptase and protease region: outcome after treatment in recently infected individuals.

We retrospectively studied 38 Italian recently HIV-1-infected subjects who seroconverted from 1994 to 1997 to investigate: (i) the prevalence of nucleoside reverse transcriptase inhibitors (NRTI)-related mutations at primary infection; (ii) the proportion of naturally occurring mutations in reverse transcriptase (RT) and protease regions of patients naive for non-nucleoside RT inhibitors (NNRTIs) and protease inhibitors (PIs); (iii) the drug-susceptibility to NRTIs and PIs in subjects with NRTI- and/or PI-related mutations; and (iv) the outcome of seroconverters treated with various NRTIs or NRTI/PI regimens. Baseline HIV-1 plasma viraemia and absolute CD4 count at baseline could not be used to distinguish patients with NRTI- and/or PI-related pre-existing mutations from those with wild-type virus (P = 0.693 and P = 0.542, respectively). The frequency of zidovudine-related mutations was 21% in the study period. The response to treatment was not significantly different in subjects with or without genotypic zidovudine-related mutations at primary infection (P = 0.744 for HIV-1 RNA and P = 0.102 for CD4 cells). Some natural variation (2.6%) was present within regions 98-108 and 179-190 of RT involved in NNRTI resistance. The high natural polymorphism in the protease region present in our patients was similar to that reported by others. In our study some PI-associated substitutions, thought to be compensatory in protease enzymatic function, could confer intermediate to high PI-resistance. As discrepancies between genotypic and phenotypic results may exist in recent seroconverters, our data suggest that the role of transmitted NRTI- and PI-resistant variants remain to be fully elucidated in vivo.

Prevalence of HIV-1 resistant strains in recent seroconverters.

Twenty-nine HIV-1 recently infected subjects were retrospectively studied to investigate both the prevalence of nucleoside reverse transcriptase inhibitors (NRTI)-related mutations at primary infection and the proportion of naturally occurring mutations in protease inhibitor (PI)-naive patients. Neither HIV-1 plasma viremia nor CD4 absolute count at baseline could distinguish patients with NRTI pre-existing mutations from those with wild-type virus. An increasing proportion of ZDV-related mutations was observed over time with an overall frequency of 20.7% in the study period. Only 1 out of 6 patients (16.7%) with ZDV-related mutations showed a phenotypically ZDV resistant isolate. A striking proportion of polymorphic changes was present in the protease region of pol gene in newly infected individuals. As many as 80% of seroconverters presented at least one naturally occurring substitution. Some PI-associated substitutions, thought to be compensatory in protease enzymatic function, could confer intermediate to high PI-resistance. Their role following PI administration remains to be elucidated. Our data suggest that the choice of drugs should be oriented by both genotypic and phenotypic evaluations to tailor individual regimens in seroconverters.

Susceptibility to PNU-140690 (Tipranavir) of human immunodeficiency virus type 1 isolates derived from patients with multidrug resistance to other protease inhibitors.

In our study we examined the anti-human immunodeficiency virus type 1 (anti-HIV-1) activity of a novel HIV-1 protease inhibitor, PNU-140690 (tipranavir), against patient-derived isolates resistant to multiple other protease inhibitors (PIs). The aim of our experiments was to investigate the genotypes and the in vitro phenotypes of drug resistance of PNU-140690. We carried out drug susceptibility tests with peripheral blood mononuclear cells and a fixed amount of infectious virus (1,000 50% tissue culture infective doses) to determine the 50% inhibitory concentration (IC(50)) and IC(90), PCR assays for the detection of drug resistance mutations in RNA in plasma, and direct sequencing of PCR products. Phenotypic resistance to PIs was invariably related to genotypic mutations. The substitutions among the amino acid residues of the protease included L10I, K20R, L24I, M36I, N37D, G48V, I54V, L63P, I64V, A71V, V77I, V82A, I84V, and L90M. Isolates from all of the patients had developed a maximal degree of resistance to indinavir, ritonavir, and nelfinavir (IC(50)s, >0.1 microM). We also compared these mutations with the amino acid changes previously described in association with in vivo tipranavir administration. The mutations included the following: I15V, E35D, N37D, R41K, D60E, and A71T. Infections with IIIB, 14aPre, and N70 were inhibited by an average drug IC(90) of 0.18 +/- 0.02 microM in multiple experiments. The average mean +/- standard error of mean IC(90) for the entire group of multidrug-resistant isolates derived from the mean values for two culture wells with p24 antigen supernatant appeared to be 0.619 +/- 0.055 microM (range, 0.31 to 0.86 microM). Tipranavir retained a sustained antiviral activity against PI-MDR clinical isolates and might be useful in combination regimens with other antiretroviral agents for patients who have already failed other PI-containing therapies.

Identification of two distinct subsets of long-term nonprogressors with divergent viral activity by stromal-derived factor 1 chemokine gene polymorphism analysis.

Stromal-derived factor (SDF)-1, the natural ligand for CXCR4, is present in a common polymorphic variant defined by a G-->A transition in the 3' untranslated region of the gene. In persons infected with human immunodeficiency virus type 1 (HIV-1), the homozygous genotype (SDF1-3'A/3'A) has been postulated to interfere with the appearance of T-tropic syncytium-inducing strains. The polymorphism of SDF1 was correlated with HIV-1 phenotype, plasma viremia, and unspliced and multiply spliced specific transcripts in 158 virologically characterized HIV-1-infected patients (39 recent seroconverters, 75 typical progressors, and 44 AIDS patients) and in 42 HIV-1-infected long-term nonprogressors (LTNPs). Analysis of SDF1 allele distribution revealed that SDF1-3'A/3'A status is associated with low CD4 cell count (P=.0449) but not with a specific HIV-1 phenotype. In LTNPs, SDF1-+/+ condition defined a subset of persons with lower HIV-1 replication than in heterozygous subjects. The low viral activity in SDF1-+/+ LTNPs suggests that other factors play a major role in vivo in determining the course of HIV-1 infection.

Homozygous delta 32 deletion of the CCR-5 chemokine receptor gene in an HIV-1-infected patient.

BACKGROUND: Recent research has found that entry of non-syncytium-inducing (NSI), monocyte-macrophage-tropic HIV-1 isolates requires binding to both CD4 and CCR5 receptors, and that delta 32/delta 32 homozygous individuals are protected against infection. OBJECTIVE: To analyse the polymorphism of CCR-5 gene in HIV-1-infected and uninfected subjects. DESIGN AND METHODS: CCR-5 sequences were amplified by polymerase chain reaction (PCR) from DNA of peripheral blood mononuclear cells. Samples from 152 HIV-1-infected subjects and 122 uninfected controls were tested for the detection of the 32 base-pair deletion. HIV-1 phenotype was determined by viral isolation and MT-2 evaluation. RESULTS: The wild-type/delta 32 heterozygous and delta 32/delta 32 homozygous conditions were represented in 10.7 and 0.8% of healthy controls and in 9.8 and 0.7% of HIV-1-infected subjects, respectively. Of note, the delta 32/delta 32 deletion of the CCR-5 gene was detected by PCR and sequencing confirmed in a patient with progressive infection harbouring a clade B virus with SI phenotype. CONCLUSIONS: delta 32/delta 32 homozygosity for the CCR-5 gene does not confer absolute protection against HIV-1 infection, suggesting that either macrophage-tropic viral strains could use coreceptors other than CCR-5 or infect independently of the presence of a functional CCR-5 coreceptor. Alternatively, primary infection sustained by T-cell-tropic isolates, although exceptional, may occur.

Comparable biological and molecular determinants in HIV type 1-infected long-term nonprogressors and recently infected individuals.

Human immunodeficiency virus type 1 (HIV-1) isolability, rate of replication, phenotype, plasma viremia, and specific intracellular transcripts were cross-sectionally analyzed in 61 HIV-1-seropositive individuals to evaluate the correlations between the virological and molecular correlates of protection and progression in different clinical subsets: recently infected subjects (RIs), long-term nonprogressors (LTNPs), late progressors (LPs), and typical progressors (TPs). Comparison of the major virological and molecular features of HIV-1 infection has defined distinct profiles for different subsets of patients. LTNPs or RIs, as well as LPs or TPs, exhibited similar titers of coculture p24 antigen; the differences between the former and the latter were statistically significant at all the time points tested (p = 0.0001; 0.0003 and 0.0001). Whereas LTNPs and RIs revealed comparable low levels of indexes of viral replication, LPs and TPs showed higher genome and mRNA copy numbers (p = 0.0004 and p = 0.0008, respectively). We demonstrated close biological and molecular similarities between RIs and LTNPs on the one hand, and LPs and TPs on the other. In LTNPs both viral biological properties and viral load are important determinants of the course of the disease.

Plasma viremia and virus phenotype are correlates of disease progression in vertically human immunodeficiency virus type 1-infected children.

OBJECTIVE: To analyze the relationships among HIV-1 plasma viremia, phenotype and CD4 T cell counts in vertically infected children. METHODS: Plasma viremia was quantified in 37 vertically infected children at different stages of the disease by a standardized molecular assay. Virus isolation and non-syncytia-inducing or syncytia-inducing (SI) HIV-1 phenotype evaluation were performed in parallel. RESULTS: HIV-1 RNA genomes were found to be significantly different in CDC clinical classes N, A, B and C (P = 0.0135) and in immunologic classes 1, 2 and 3 (P = 0.0110). None of the children in Class N or A harbored HIV-1 isolates with SI phenotype, whereas SI primary isolates were detected in 2 of 7 (29%) and 7 of 10 (70%) Class B and C children, respectively. Similarly SI variants were present in only 9 of 13 children in immunologic Class 3 (70%). When stratified according to the increasing severity of virologic status, the children showed a significant difference (P = 0.0458) in viral burden. CONCLUSIONS: Clinical symptoms, the most dramatic being reduction in the number of CD4 lymphocytes, and the highest plasma viremia levels were observed in the children in whom fast replicating, highly cytopathic SI variants were isolated. These data extend the virologic characterization of vertically HIV-1 infected children and suggest that both the plasma viremia levels and phenotype of primary isolates are viral correlates of disease progression in vertically infected children.

Viral load, viral phenotype modification, zidovudine susceptibility and reverse transcriptase mutations during the first 6 months of zidovudine monotherapy in HIV-1-infected people.

We studied 14 zidovudine-naive, HIV-1-infected patients attending an infectious diseases clinic in Milan during zidovudine therapy for 6 months. We monitored CD4 cell counts, immune complex-dissociated p24 antigen, viral phenotype and viral load in plasma. The virus infecting a subset of patients was examined for zidovudine susceptibility and zidovudine resistance-associated mutations. A significant correlation was established between the increase in the CD4 cell count and the decrease in viral load (Spearman's coefficients < -0.5). Patients who were p24 antigen positive had a higher viral load (P < 0.005 at baseline and after 6 months of therapy). Patients with non-syncytium-inducing (NSI) virus had higher CD4 cell counts over time than those with syncytium-inducing (SI) virus. We also examined the viral load in relation to viral phenotype. The median viral load in patients with NSI virus was higher than in SI controls at baseline, but not after 3 and 6 months of therapy. Sequential isolates of HIV-1 were obtained from nine patients and tested for resistance to zidovudine by monitoring the drug susceptibility and the reverse transcriptase-encoding sequence. Amino acid changes at codons 70 and 215 were present in some but not all isolates with zidovudine-resistant phenotype in vitro. It was possible to perform a correlation between zidovudine susceptibility and zidovudine-associated pol gene mutations only at the 6-month time point (Spearman's coefficient = 0.076). SI phenotype was associated with the development of a decreased zidovudine susceptibility. A correlation between zidovudine-associated pol gene mutations and SI phenotype was detected at the 6-month time point.