Prevalence of genotypic resistance to nucleoside analogues and protease inhibitors in Spain. The ERASE-2 Study Group.

OBJECTIVE: To examine the prevalence of resistance mutations to nucleoside reverse transcriptase inhibitors (NRTI) and protease inhibitors (PI) in a representative HIV-1 population in Spain. METHODS: A cross-sectional study was conducted including 601 HIV-infected patients who attended 20 Spanish hospitals in June 1998. Drug resistant mutations were examined using hybridization line probe assays (LiPA). The 6 bp insert at position 69 and the codon 75 mutant were examined by sequencing analysis in specimens lacking reactivity to 69/70 and 74 bands on LiPA, respectively. RESULTS: Primary resistance to NRTI was recognized in nine out of 52 (17%) naive individuals, whereas primary resistance to PI was found in seven out of 126 (6%) PI-naïve patients. The codons most frequently involved in NRTI resistance were at positions 70 (66%), 184 (44%), 215 (33%), and 41 (11%), whereas the most common PI resistance mutation was at codon 82 (6/7 subjects). In pre-treated patients, the overall prevalence of resistant genotypes was 72.9% for NRTI and 27.2% for PI. The most frequent NRTI mutations occurred at codons 184 (38.5%), 215 (30.1%), and 41 (22.5%), whereas the most frequent PI mutations in pre-treated subjects were found at positions 82 (15.8%) and 84 (11.4%). Overall, patients who began triple combinations as initial therapy showed a lower number of key resistance mutations than those who began highly active antiretroviral therapy (HAART) after being exposed to NRTI for a period of time (mean number of mutations, 0.1 versus 1.8, P< 0.05). Codon 75 mutant was found in three out of 387 patients (0.7%), whereas no insertions at codon 69 were recognized. CONCLUSION: The prevalence of primary genotypic resistance to NRTI and PI in Spain was 17% and 6%, respectively. Zidovudine, lamivudine, indinavir and ritonavir were the drugs most frequently affected. These data support the use of resistance testing prior to the introduction of first-line antiretroviral therapies in Spain. Among pre-treated subjects, drug resistance genotypes were less prevalent in those who began HAART as initial therapy.

Usefulness of genotypic analysis of resistance to nucleoside analogues in the clinical setting.

The introduction of drug resistance testing during clinical care of patients infected with the human immunodeficiency virus (HIV) is still a matter for discussion. Thirty-seven HIV-infected subjects who had positive plasma viraemia despite undergoing antiretroviral therapy were tested for the presence of genotypic mutations linked to resistance to nucleoside analogues. Thirty (81.1%) individuals showed one or more mutations conferring drug resistance. Eight harboured mutants at codons 215 and 41, which confer resistance to zidovudine and are associated with the worst prognosis. One individual carried the codon 69-SSS insertion, which confers multidrug resistance. Seven (18.9%) subjects did not show any mutation, but all of them failed to adhere to the treatment regimen and/or experienced diarrhea that likely caused malabsorption of the drugs. Clinicians referred to the resistance profile when they adjusted or prescribed subsequent combination regimens for their patients.

Multiple drug resistance genotype causing failure of antiretroviral treatment in an HIV-infected patient heavily exposed to nucleoside analogues.

A 37-year-old homosexual man began antiretroviral combination therapy with didanosine (ddI), lamivudine (3TC) and indinavir (IDV) after being exposed previously to zidovudine (ZDV), ddI and 3TC in different sequential regimens. The patient's viral load did not fall below a detectable level despite his adherence to drug therapy, which was considered optimal. Stavudine (d4T) was prescribed in the third month of treatment instead of ddI without any evident improvement in the treatment response. A point mutation nested PCR assay showed that the patient carried a virus with a codon Q151M mutation, which confers multiple drug resistance to nucleoside analogues. Genetic sequence analysis showed that, despite none of the classically associated mutations to Q151M being present at the beginning of treatment, continuous genetic evolution under selective drug pressure allowed the virus to accumulate mutations at codons 62, 74 and 116 over time. As expected, the CD4+ cell count declined during the study period, and the viral load remained detectable.

Evolution of genotypic resistance to nucleoside analogues in patients receiving protease inhibitor-containing regimens.

The prevalence of genotypic resistance to nucleoside analogues (NA) was examined using a line probe assay (LiPA, Innogenetics, Spain) and a point mutation assay to test for codon 151 polymorphism in plasma from 34 individuals who had been exposed to NA for longer than 1 year. The testing was repeated in the same population after 6 months of being on a new potent protease inhibitor (PI)-containing antiretroviral combination. Only nine (47%) of the 19 patients initially carrying the codon 41 mutation restored zidovudine wild-type (WT) virus population. Similarly, eight (33%) out of 24 carrying the codon 215 mutation restored the wild-type variant. Two subjects carrying codon 74 didanosine mutation reverted to wild-type genotype, as well as two (18%) out of 11 harbouring the codon 184 lamivudine-resistant variant. To conclude, the extent to which drug recycling might be of benefit in subjects showing a restoration of genotypic sensitivity to former drugs needs to be explored.

Presence of genotypic resistance in nucleoside analogue-treated HIV-1-infected patients with undetectable viral load.

Patients harbouring drug-resistance viruses usually suffer a rise in serum viraemia after a variable period of time. We have investigated the relationship between the appearance of resistant genotypes and the viral load of each patient after treatment. Our objective was to assess the association between human immunodeficiency virus (HIV) RNA plasma levels and the number of drug resistance-associated point mutations after treatment. A total of 150 patients from three reference centres in Spain (Madrid, Barcelona and Seville) from a previous study (Erase Study) were included. Patients had at that time undergone antiretroviral treatment with nucleoside analogues for at least 1 year (zidovudine/didanosine; zidovudine/zalcitabine; zidovudine/zalcitabine/lamivudine; zidovudine/didanosine/lamivudine). In this study, plasma viraemia in these patients was quantified and a line probe assay was used to determine the genotype of the virus. Viral load was significantly higher in patients harbouring virus with more than three mutations than in those individuals who harboured wild-type strains (P < 0.05). Surprisingly, when patients with viral load < 500 copies/ml (13/150) were analysed, only two carried wild-type strains, whereas three had virus with more than three point mutations. The viral load of six samples was assayed using an ultrasensitive test (detection limit < 20 copies/ml). Of the three samples where viral load was < 20 copies/ml, one patient harboured wild-type virus, whereas two carried mutant virus strains. These results suggest that even in patients with undetectable viral loads by conventional methods, viral replication may continue and mutations develop. Therefore, standard values of plasma viraemia for measuring the effectiveness of the treatment should be reconsidered when patients are on antiviral regimens of just two or three nucleoside analogues.

Efficiency of drug resistance genotypic tests in specimens with low HIV viral load.

The early recognition of resistance to antiretroviral agents could allow a rapid switch in therapy and therefore avoid the accumulation of mutations and reduce the risk of cross-resistance. However, the efficiency of genotypic tests in specimens with low viral load (VL) is severely compromised since human immunodeficiency virus (HIV) RNA in these samples often goes unrecognized. The frequency of results provided by a line probe assay (LiPA, Murex), a commercially available drug resistance test and a home-made point mutation assay (PMA) for recognizing the codon 151 multidrug-resistance mutation was examined in 664 plasma samples stratified with respect to VL values. Overall, 421 (63%) samples could be interpreted by both LiPA and PMA. The sensitivity decreased as plasma VL lowered: 89% for samples with VL > 10,000 HIV RNA copies/ml, 77% for those with VL between 500 and 10,000 HIV RNA copies/ml and 37% for specimens with VL < 500 HIV RNA copies/ml. A good agreement existed comparing the sensitivity of the home-made PMA and LiPA. Although the former tends to produce more results, the difference did not achieve statistical significance. Our results support that new, more sensitive, HIV RNA extraction methods need to be implemented for the rapid recognition of drug-resistant mutants in patients experiencing an early rebound in plasma viraemia.

The decline in CD4+ T lymphocytes as a function of the duration of HIV infection, age at seroconversion, and viral load.

BACKGROUND: The rate of CD4+ T-lymphocyte decline seen in HIV-infected patients is very variable. Although older patients, a longer duration of HIV infection, and a high level of plasma viraemia, have been associated with a faster fall in CD4+ T-cells, the relationship between these variables is still not well known. PATIENTS AND METHODS: In a cross-sectional study that included a total of 107 patients of known age and date at HIV seroconversion, the current CD4+ T-cell count and plasma viraemia were examined. Patients were not taking antiretroviral drugs, nor had received immunizations nor were suffering any intercurrent infections at the time of the study. RESULTS: The mean duration of HIV infection was 8.6+/-2.9 years. The mean CD4+ T-lymphocyte count was 367+/-264 x 10(6)/l. Mean plasma viraemia was 4.3+/-0.9 logs. In a linear regression model, the current CD4+ T-cell count was explained in 21.7% by the duration of HIV infection, while the level of plasma viraemia justified separately up to 37.0%. When both parameters were combined, they explained 58.8%. of the CD4+ lymphocyte values. In this model, a variation of one logarithm in the plasma viraemia had six times greater effect on the number of CD4+ lymphocytes than each year of HIV infection. When the age at seroconversion was added to the model, the CD4+ cell count allowed the explanation of up to 62.2% of cases. CONCLUSIONS: The age at seroconversion, the duration of HIV infection, and the level of plasma viraemia independently and substantially influence the current CD4+ lymphocyte count in HIV-infected subjects. However, other variables should exist (e.g. virus syncytium-inducing phenotype, host immunogenetic repertoire, etc.), contributing to explaining the different rate of CD4+ T-cell decline seen in HIV-infected subjects.

[Interrelationship between the duration of HIV infection, viral load and CD4 positive lymphocyte count]. / Interrelación entre la duración de la infección por el virus de la inmunodeficiencia humana, la viremia plasmática y la cifra de linfocitos CD4+.

BACKGROUND: The decline in CD4+ lymphocytes occurs at different rates in patients with HIV infection. A longer duration of HIV infection and a higher level of viral replication, represented by the viral load, are associated with a lower CD4+ lymphocyte count. However, the interelationship between these variables is still not well known. PATIENTS AND METHODS: 107 HIV-infected patients for whom the date of infection was known, were included in a transversal study, in which the CD4+ lymphocyte count and the plasma viral load were analysed, the last using an isothermal amplification method (NASBA). Patients were not receiving antiretroviral drugs or suffered intercurrent infections at the time of the study. RESULTS: The mean duration of HIV infection was 8.6 +/- 2.9 years. The mean CD4+ lymphocyte count was 366 +/- 264 x 10(6)/l. The mean plasma viraemia was 4.3 +/- 0.9 logs. In a linear regression model, the CD4+ lymphocyte count was explained in 21.7% of cases by the duration of HIV infection, meanwhile the viral load justified up to 36.2 of CD4+ cell variability. When both parameters were combined, up to 58.4% of CD4+ lymphocyte values were explained. In this model, changes of 1 log in viral load had a 4-fold higher effect on the CD4+ cell count than each year of HIV infection. CONCLUSIONS: The duration of HIV infection and, particularly the viral load strongly influences the current CD4+ lymphocyte count, although other variables should exist (virus with syncytium-inducing phenotype, age of the patient and his immunegenetic repertoire) influencing the different decline seen in CD4+ T-cells.

Prevalence of genotypic resistance to nucleoside analogues in antiretroviral-naive and antiretroviral-experienced HIV-infected patients in Spain.

OBJECTIVE: To determine the prevalence of genotypic resistance to nucleoside analogues (NA) in a large group of HIV-infected individuals in Spain, some of whom had no previous treatment with antiretroviral drugs (antiretroviral-naive) and some of whom had such experience (antiretroviral-experienced). SETTING: Cross-sectional study in out-patient clinics in three reference hospitals for HIV/AIDS located in Barcelona, Madrid and Seville, Spain. PATIENTS AND METHODS: Primary mutant genotypes were examined in plasma HIV RNA collected from 150 antiretroviral-naive subjects, half in 1993 and the other half in 1997. Furthermore, drug resistance mutations were analysed in plasma collected from another 150 antiretroviral-experienced patients who had received 2 NA for longer than 1 year, either in sequence as monotherapy or as combination therapy. A line probe assay was used for recognizing mutations conferring resistance to zidovudine (ZDV), didanosine (ddI), zalcitabine (ddC), and lamivudine (3TC). A point-mutation nested-PCR assay was used for examining a codon 151 mutation associated with multiple drug resistance. RESULTS: One or more mutations associated with primary resistance to NA were seen in 10 antiretroviral-naive (13.3%) patients in 1993 and in nine (12%) in 1997. In all but two cases, they were associated with ZDV resistance. In contrast, all but six (96%) of the antiretroviral-experienced subjects harboured drug-resistant mutant viruses. The codon 184 mutation (associated with resistance to 3TC) was detected in 92% of patients treated with 3TC, but also in 18% of those treated with only ddI or ddC. The codon 215 mutation was found in 67.3% of patients who had been exposed to ZDV; the codon 69 mutation was found in 15% of patients treated with ddC; and the codon 74 mutation was found in only 7.2% of patients treated with ddI. Finally, the codon 151 multidrug resistant mutation was found in four (2.7%) of 150 patients with a long-term exposure to NA. CONCLUSIONS: Overall, the prevalence of drug-resistant HIV-1 genotypes was 12.7% in antiretroviral-naive patients, most of whom had ZDV-resistant mutants. There is no evidence of an increase during the last 5 years. However, multidrug-resistant HIV genotypes are currently circulating in Spain.

[Efficacy of ritonavir in HIV positive patients pretreated with nucleoside analogues]. / Eficacia del ritonavir en pacientes VIH+ previamente tratados con análogos de nucleósidos.

The virologic and immunologic efficacy of ritonavir, an HIV protease inhibitor, was examined in 14 HIV-infected patients with a mean CD4+ cell count of 183 x 10(6)/l at baseline. All had been exposed to nucleoside analogs for longer than 6 months; and ritonavir was added to previous ongoing antiretroviral drugs. A mean reduction of 1.5 logs was seen in plasma viral load 4 weeks after began ritonavir, and 9 (64.3%) of 14 patients achieved undetectable levels (< 4,000 HIV-RNA copies/ml). A mean increase of 222 x 10(6) cells was observed in the CD4+ count. Nine (64.3%) individuals reported side effects, although they did not force to stop the medication. In conclusion, the addition of ritonavir seem to be a reasonable alternative strategy in patients with advanced HIV disease and heavy previous exposure to nucleoside analogs.

[Efficacy of combining zidovudine plus didanosine in patients with and without previous exposure to zidovudine]. / Eficacia de la combinación de zidovudina y didanosina en pacientes con y sin exposición previa a zidovudina.

BACKGROUND: The virologic and immunologic efficacy of the combination of zidovudine (ZDV) and didanosine (DDI) was examined in 27 HIV-infected patients. METHODS: 27 HIV-infected patients, 15 of whom were naive for antiretroviral drugs. The remaining 12 subjects had been exposed to ZDV for at least 16 weeks. Mean plasma viral load was of 4.7 logs and 4.5 logs, respectively. All patients had less than 350 x 10(6)/l CD4+ T lymphocytes at baseline. RESULT: Significant reductions in viremia (> 0.5 logs) were seen at the first and third months, respectively, in 71.4% and 64.3% of naive patients. In contrast, it occurred only in 10% and 0%, respectively, of pre-treated patients. Moreover, the CD4 count only increased significantly (> 15%) in naive individuals. Changes in serum p24 antigenemia were not significant in both groups neither at the first nor at the third months of began therapy. CONCLUSION: The antiviral efficacy of the ZDV plus DDI combination is limited and transient in patients with advanced HIV disease, and disappears in subjects already exposed to ZDV.

[Combined antiretroviral therapy for prevention of vertical HIV-1 transmission]. / Terapia antirretroviral combinada para la prevención de la transmisión vertical de la infección por el VIH-1.

A 25-year old woman with rapid HIV disease progression had been receiving zidovudine (ZDV) for two years, when she became pregnant. She had a high viral load and carried out zidovudine-resistant viral strains. For these reasons, and with the main objective to maximally reduce viremia, the association of DDI to ZDV was introduced a few weeks before delivery. The virological follow-up for one year has confirmed the lack of HIV infection in the child. Combined antiretroviral therapy during the last weeks of pregnancy might be considered for the prevention of vertical transmission of HIV in cases of high risk of newborn infection, without adding relevant toxicity.

Measurement of human immunodeficiency virus type 1 plasma virus load based on reverse transcriptase (RT) activity: evidence of variabilities in levels of virion-associated RT.

Virus load based on levels of functional reverse transcriptase (RT) was measured in plasma from 50 human immunodeficiency virus (HIV) type 1-infected persons, in 87 samples from 10 HIV-1 seroconversion panels, and in 100 uninfected persons by use of Amp-RT, an ultrasensitive RT assay. Of the 50 clinical samples, 38 (76%) were Amp-RT positive, while all uninfected controls were negative. Pearson's correlation coefficient of RNA and RT levels was .73 for all samples, .86 for seroconversion samples, and .49 for clinical samples. Calculated ratios of RT activity to virion RNA varied widely during both early and late stages of infection. Mean RT:RNA ratios in 8 seroconversion panels and in 12 (34.3%) of 35 individual clinical samples were significantly lower than the ratio for a reference virus. However, ratios were stable in individual seroconversions over time. These data demonstrate that RT activity can be used to quantitate plasma virus load and provide evidence of different levels of virion-associated RT among HIV-1-infected persons.

Short-term efficacy and safety of stavudine in pretreated HIV-infected patients.

Most of the information available on stavudine (d4T) comes from studies in patients with advanced human immunodeficiency virus (HIV) disease to whom stavudine was administered as monotherapy. Herein, we summarize the results of adding 40 mg stavudine twice daily to previous therapies in patients with mild to advanced immunological disease (mean CD4 T cell count 178 cells/mm3; range 6-480 cells/mm3). In an intention-to-treat, prospective, open trial, 64 patients (84.4% men; mean age 35.2 years) were analysed. Their average time on previous antiretroviral therapy was 19.8 months (range 6-52). Plasma HIV RNA load fell by a mean of 0.64 and 0.74 log at 1 and 3 months, respectively, after the start of stavudine therapy (P <0.001 Sign rank test). The CD4 cell count increased by a mean of 25.1 cells/mm3 in the third month (P = 0.002 Sign rank test). Antiviral activity was independent of the CD4 cell count at baseline, but more pronounced declines in viral load were seen in patients with shorter periods of previous antiretroviral therapy and in those in whom stavudine was combined with didanosine or lamivudine rather than zidovudine. Ten (15.6%) patients discontinued the drug during the first 6 months of treatment because of the development of toxicity (neuropathy in six cases, hepatitis in two, oedema in one and rash in another); all but one of them had CD4 counts < 200 cells/mm3. Another two patients stopped treatment voluntarily. The remaining 52 patients tolerated the drug well for the first 6 months and had a high level of compliance with treatment. In conclusion, stavudine is generally well tolerated and has significant antiretroviral activity when it is administered to patients with extensive previous treatment with multiple reverse transcriptase (RT) inhibitors. It should be expected that the short-term favourable effects of stavudine on laboratory markers will further translate into a reduced progression of disease and improved survival.