Impact of Newly Available Drugs on Clinical Progression in Patients with Virological Failure after Exposure to Three Classes of Antiretrovirals

Objective

To study the prognosis of HIV-infected patients with virological failure after exposure to three classes of antiretroviral drugs (ARVs).

Design

Cohort study. Setting: French Hospital Database on HIV.

Patients

Patients previously exposed to at least two nucleoside reverse transcriptase inhibitors (NRTIs), two protease inhibitors and one non-NRTI, with viral load (VL) values of >5000 copies/ml after the exposure criteria were met and a new treatment initiated between 1998 and 2001 with VL >5000 copies/ml.

Main outcome measures

Risk of new AIDS-defining-events (ADEs) or death from first introduction of a drug never used before occurring between 1998 and 2001 defined as baseline.

Results

The main baseline characteristics of the 1092 patients were: previous ADE in 49% of cases, median CD4 cell count 181 μl, median VL 4.9 log10 copies/ml, median duration of ARV therapy 5.0 years and previous exposure to a median of nine ARVs. The crude progression rates were 20.1/100 patient-years among patients included in 1998, 15.1 in 1999, 11.1 in 2000 and 8.6 in 2001. After adjustment for baseline characteristics, the calendar year of inclusion was associated with the risk of clinical progression ( P<0.001). When the types of newly available drugs used at baseline or during follow-up were introduced into the model, year of inclusion was no longer associated with the risk of clinical progression ( P=0.42), while exposure to amprenavir/r, lopinavir/r, abacavir or tenofovir was associated with a lower risk.

Conclusions

The clinical prognosis of heavily pretreated patients experiencing virological failure improved between 1998 and 2001, mainly thanks to the use of newly available drugs with more favourable resistance profiles.

Slow viral dynamics of hepatitis C virus genotype 4

Patients infected with hepatitis C virus genotype 4 (HCV-4) respond to interferon alpha (IFN) as poorly as those infected with genotype 1. However, there is no information on the viral dynamics of HCV-4. Interferon-ribavirin treatment was administered to untreated patients infected with HCV-4. Viral load was assessed with Versant 3.0. Viral dynamics parameters were estimated based on the bi-phasic model for HCV during IFN treatment. Viral kinetics of HCV-4 follow a bi-phasic decline pattern also. The mean effectiveness of IFN in blocking production of HCV-4 was a decline of 77.8% during the first day of treatment. The half-life of free virions was estimated at 3.5 h and that of infected cells from 1.9 to over 70 days. The viral dynamics parameters of HCV-4 appear similar to those of HCV-1 and slower than those of HCV-2. HCV-4 infected patients should be grouped with those with HCV-1 when therapeutic schemes are considered in relation to genotype.

Persistence of HIV-1 resistance in lymph node mononuclear cell RNA despite effective HAART

OBJECTIVE

To analyse the presence of genotypic and phenotypic resistance in lymph node mononuclear cells from patients with sustained undetectable plasma HIV-1 RNA with highly active antiretroviral therapy.

DESIGN

Cross-sectional study on 27 HIV-infected patients receiving triple therapy for a mean period of 232.8 +/- 22.1 weeks.

METHODS

HIV-1 RNA was measured in plasma and lymph node cells using PCR. Reverse transcriptase and protease genes were sequenced from HIV-1 RNA obtained from lymph node cells and from peripheral blood mononuclear cell proviral DNA using a commercially available kit (TruGene). Phenotypic resistance was assessed by using a recombinant virus assay (AntiVirogram).

RESULTS

Mutations were not found in lymph node mononuclear cell RNA in six out of nine patients on first-line regimens although they were detected in 15 out of 18 who received prior suboptimal combinations. Phenotypic resistance was confirmed in most of these cases. These patterns of resistance were closely related to patients' history of antiretroviral therapy and genotypic analysis of plasma HIV-1 RNA taken just before initiation of the current regimen. In half the patients analysed, resistance mutations found in lymph nodes were not always detected in archival proviral DNA from blood cells. Mean levels of HIV-1 RNA in lymph node cells were not different in patients exhibiting resistance compared with those harbouring wild-type viruses.

CONCLUSION

These data demonstrate that resistant HIV-1 is produced in lymphoid tissues for prolonged periods despite effective therapy. The mechanism could represent a release from previously infected cells rather than new cycles of cellular infection.

HIV-1 induction-maintenance at the lymph node level: the "Apollo-97" Study

OBJECTIVE

To assess the effects of five-drug combination therapy on HIV-1 load in lymph nodes and subsequent maintenance with four and three drugs.

METHODS

Ten pharmacotherapeutically naive patients received a combination of zidovudine, lamivudine, didanosine, ritonavir, and saquinavir for 24 weeks, then zidovudine, lamivudine, didanosine, and saquinavir for the next 24 weeks, and finally zidovudine, lamivudine, and saquinavir for the last 24 weeks. HIV-1 RNA in lymph nodes was measured using quantitative polymerase chain reaction (PCR) at baseline, after 12, 24, 48, and 78 weeks. Plasma HIV-1 RNA, proviral DNA in peripheral blood mononuclear cells (PBMCs), circulating lymphocyte subsets, and protease inhibitor levels in blood were also regularly measured. Genotypic resistance was assessed in the different compartments in 2 patients who were failed by therapy.

RESULTS

HIV-1 RNA decreased in lymph nodes in 9 patients and was stable in 1 despite initial control of plasma replication <20 copies/ml in each patient. Lymph node levels rebounded in 1 patient at week 72 as a result of lack of adherence and remained stable in the 8 others despite maintenance regimens. This represents a mean drop of -3.17 log in lymph nodes for the 8 patients maintaining undetectable viremia at 72 weeks. In the patient with stable lymph node viral RNA, selection of the M184V mutation was demonstrated at this level before detection in plasma and low blood saquinavir levels were found throughout the study. Continuous improvements in immune parameters were observed in all cases, although PBMC proviral DNA levels either showed a continuous decrease or stabilized to a plateau.

CONCLUSIONS

More complex regimens do not perform better in lymph nodes than classic triple therapy. The persistence of HIV-1 RNA in lymph nodes could be related with cellular resistance mechanisms rather than an insufficient potency of the regimens.

Phenotypic and genotypic resistance to nucleoside reverse transcriptase inhibitors in HIV-1 clinical isolates

OBJECTIVE

To assess phenotypic and genotypic cross-resistance to nucleoside reverse transcriptase inhibitors in patients treated with a combination including zidovudine, who were switched to a combination including stavudine.

METHODS

We analysed 24 clinical HIV-1 isolates from 12 patients before and several months after therapeutic switching. Plasma HIV-1 RNA was measured using quantitative polymerase chain reaction (Roche). Genotypic resistance was measured by sequencing the reverse transcriptase gene from plasma HIV-1 RNA. Phenotypic resistance was measured using a recombinant assay (Virco).

RESULTS

Patients were treated with a combination including zidovudine for a mean (+/- SEM) period of 21.8 +/- 3.5 months and had a plasma viral load of 4.1 +/- 0.2 log HIV-1 RNA copies/mL (time 1). After a mean period of 19.3 +/- 1.6 months following the therapeutic change, the plasma viral load was 3.6 +/- 0.1 log copies/mL (time 2). At time 1, genotypic resistance to zidovudine was found in all cases (41L: four cases; 41L, 215Y: five cases; 41L, 210W, 215Y: two cases; K70R: one case) with a mean 6.6 +/- 1.6-fold increase in the median inhibitory concentration (IC50) to zidovudine and 1.7 +/- 0.4-fold to stavudine. At time 2, genotypic resistance to zidovudine was found in 11 out of 12 cases (41L: two cases; 41L, 215Y: six cases; 41L, 210W, 215Y: two cases; M41L, D67N, L210W, T215Y: one case) with a mean 18.9 +/- 8.8-fold increase in the IC50 to zidovudine and 1.4 +/- 0.4-fold to stavudine

CONCLUSIONS

In this clinical series of patients with suboptimal control of plasma HIV-1 RNA using a combination including zidovudine, the presence of zidovudine-related mutations was associated with a decreased phenotypic sensitivity to this drug. Despite persistent HIV-1 replication, switching to stavudine was associated with a further decrease in phenotypic sensitivity to zidovudine but not to stavudine after 19 months. These data suggest that stavudine remains a possible option in zidovudine-experienced patients.

Impact of immune interventions on proviral HIV-1 DNA decay in patients receiving highly active antiretroviral therapy

OBJECTIVE

To measure the evolution of proviral HIV-1 DNA levels in patients receiving highly active antiretroviral therapy (HAART) compared to those treated with HAART plus interleukin-2 (IL-2) and hydroxyurea.

DESIGN

Prospective randomised trial.

METHODS

Twenty-two HIV-1 infected patients were randomly assigned to a five-drug antiretroviral regimen for 72 weeks, with or without IL-2, followed by a three-drug regimen up to week 120 with additional hydroxyurea in patients having received IL-2. HIV-1 DNA levels in peripheral blood mononuclear cells (PBMC) were measured regularly using the Amplicor Monitor kit from Roche Diagnostics (Meylan, France). Potentially infectious HIV-1 was cultured in enhanced conditions from circulating CD4 T cells at week 120.

RESULTS

During the study period of 120 weeks, HIV-1 DNA levels in PBMC decreased by -1.1 log in patients treated with HAART only compared with -1.8 log in patients with additional IL-2 and hydroxyurea. A two-phase decay rate was observed, with an inflexion point at 12 weeks. The second decay was slow, with mean half-lives of 130.1 +/- 21.3 weeks and 95.1 +/- 26.3 weeks for patients on HAART and those receiving additional IL-2 and hydroxyurea, respectively. At week 120, one out of 11 patients with HAART alone compared to six out of 11 in the group with IL-2 and hydroxyurea had undetectable proviral DNA levels and three of them had unsuccessful recovery of replication-competent HIV-1 from blood CD4 T cells.

CONCLUSION

Therapeutic strategies combining HAART and immune interventions have higher potency to decrease the number of infected cells than HAART alone.

Increased mitochondrial toxicity with ribavirin in HIV/HCV coinfection

In two of 15 patients coinfected with HIV and hepatitis C virus who received interferon-alpha plus ribavirin in addition to HAART, we observed multiorgan dysfunction and lactic acidaemia. As ribavirin is a nucleoside analogue, an increased risk of mitochondrial toxicity can be induced in HIV-infected patients already treated with nucleoside analogues, leading to clinical deterioration in some cases.

Pilot study of a combination of highly active antiretroviral therapy and cytokines to induce HIV-1 remission

A pilot study of a combination of highly active antiretroviral therapy (HAART) and cytokines in early HIV-1 infection has been undertaken to test the hypothesis that HIV-1 remission can be reached with this strategy by flushing latently infected viral reservoirs. Ten previously antiretroviral naive patients have received a combination of zidovudine, lamivudine, didanosine, saquinavir, and ritonavir for 72 weeks. Between weeks 12 and 48, three courses of interleukin (IL)-2 (7.5 millions of international units [MUI] twice a day for 5 consecutive days) and 2 courses of gamma-interferon (IFN) (100 microg every other day during 2 weeks) were administered subcutaneously. All patients reached plasma HIV-1 RNA levels < 20 copies/ml within 12 +/- 4 weeks. Transient increases in plasma levels (< 120 copies/ml) were observed during administration of IL-2, but less frequently during gamma-IFN administration. HIV-1 RNA decreased in lymph node cells by approximately 4 log, then remained stable after week 24. A mean drop of -0.8 log in peripheral blood mononuclear cell (PBMC) proviral DNA was observed during the trial. Isolation of potentially infectious HIV-1 was successful in each case by coculture of CD4+ T cells taken at week 72. The 2 patients who stopped therapy at the end of the trial showed rebounding plasma HIV-1 RNA levels within a few weeks. No additional mutations were selected in comparison with those present at baseline in 8 patients. In addition, 2 patients developed new mutations in the reverse transcriptase or protease gene and in 1 case, resistance selection was found in lymphoid tissue HIV-1 RNA but not in latently infected cells. In all cases, a rapid increase in both naive and memory CD4+ T cells was observed, with a reduction in activation markers and preservation of the CD8+CD28+ subset. Consequently, an aggressive regimen of HAART and cytokines administered in early stage disease is associated with a positive effect in terms of proviral load reduction and immune reconstitution but is unable to induce HIV-1 remission, allowing low levels of viral replication to persist in lymphoid reservoirs.