| |
Impact of Coinfection With HIV-1 and GB Virus C in Patients Receiving a Ritonavir-Boosted HAART Regimen: A Substudy to the MaxCmin1 Trial
|
| |
| |
JAIDS Journal of Acquired Immune Deficiency Syndromes: Volume 40(3) 1 November 2005 pp 378-380
Tillmann, Hans L MD*; Kaiser, Thorsten MD*; Fox, Zoe MSc; Staszewski, Schlomo MD; Antunes, Francesco MD; Monforte, Antonella d'Arminio MD; Vernazza, Pietro MD; Hill, Andrew PhD#; Dragsted, Ulrik B PhD; Lundgren, Jens D MD, DMS; on behalf of the MaxCmin1 Trial Group
*Medizinische Hochschule, Hannover, Germany; CHIP, Hvidovre University Hospital, Copenhagen, Denmark; J. W. Goethe University Hospital, Frankfurt, Germany; Hospital Santa Maria, Lisbon, Portugal; Ospedale L. Sacco, Milan, Italy; Kantonsspital, St. Gallen, Switzerland; #University of Liverpool, United Kingdom
.....In this study we could neither observe any influence of GBV-C on any of the prognostic markers of HIV-1 infection at baseline nor on the response to HAART in the prospective data. This was also independent of patients being protease inhibitor or antiretroviral naive or experienced (data not shown)..... It is possible that this positive effect is blunted by the very strong effect of HAART itself. Whether certain subgroups of HAART-treated patients have a beneficial effect of being GBV-C coinfected remains to be shown....
Letter To the Editor:
In 1995 the GB virus C (GBV-C) was identified. Although it is related to the hepatitis C virus, there are currently no persuasive data to suggest that this virus causes hepatitis or other illnesses in humans.1,2
It has repeatedly been reported that GBV-C viremia is associated with a more benign course of the natural history of HIV-1 infection, as has been found in most of the studies but not in all (as reviewed by Kaiser and Tillmann3). Highly active antiretroviral therapy (HAART) has completely changed the course of HIV-1 disease, leading to both declining morbidity and mortality.4 A preliminary analysis indicated that the beneficial effect of GBV-C on HIV-1 infection is also seen during HAART,5 which again has been confirmed by some but not other groups.3
The MaxCmin1 trial evaluated the safety and efficacy of indinavir (IDV) and saquinavir (SAQ) both in combination with low-dose ritonavir (r) as part of HAART regimens in a heterogeneous HIV-1-infected population.6 The main study conclusions were that the study regimens had equal immunologic and virologic efficacy, but more treatment-limiting adverse events were observed in the IDV/r arm. This provided an opportunity within the context of a randomized controlled trial to assess the role of GBV-C on efficacy outcome measures.
METHODS
The methodology of the MaxCmin1 trial has been described elsewhere.6 In brief, the trial enrolled 317 HIV-1-infected adult patients from clinics in European countries, Argentina, and the United States, of whom 306 patients (97%) initiated their assigned study regimen. Patients were antiretroviral treatment (ART) naive (25%) or ART experienced (75%); follow-up was 48 weeks. Week 12 samples were available for GBV-C testing from 258 of the 306 patients (84%) who initiated the assigned treatment (137 in the IDV/r arm and 121 in the SAQ/r arm).
GBV-C RNA was determined using a real-time polymerase chain reaction protocol on a LightCycler (Roche Molecular Systems, Basle, Switzerland) after RNA isolation with the MagNA Pure (Roche Molecular Systems, Basle, Switzerland).
The predefined measures of interest for this substudy were changes in CD4 cell count (at predefined study visits experiencing a CD4 cell increase of ≥100) and HIV-1 RNA (proportion <50 and <400 copies/mL at preplanned study visits), the proportion of patients experiencing protocol-defined virologic failure,6 and the proportion of patients experiencing adverse events of grade 3 or 4.
Statistical Analysis
Fisher exact or X2 tests were used to compare categorical variables, and continuous variables were analyzed using Student t tests and Kruskal-Wallis tests depending on the shape of the distribution. No differences were found when comparing treatment arms for GBV-C-positive or GBV-C-negative patients (data not shown); therefore treatment arms were combined and results are presented accordingly. Cox analysis was performed and Kaplan-Meier plots were produced for the time-to-event analyses containing sufficient numbers of events (n > 25). Logistic regression analysis was used to determine predictors of having high HIV-1 RNA at preselected timepoints. Two-sided statistical analyses were performed on an intention-to-treat (ITT) basis using a 5% level of significance, in STATA Statistical Software, Version 7 (StataCorp., College Station, TX).
RESULTS
Prevalence of GBV-C
Eighty-seven of the 258 patients (34%) were GBV-C positive. We did not observe any significant differences in gender, age, HIV-1 exposure group, Centers for Disease Control status, protease inhibitor experience, or any other baseline characteristics according to GBV-C serostatus at week 12 (data not shown).
CD4 Cell Increase
No statistical differences were observed for GBV-C-positive vs. GBV-C-negative patients in median (interquartile range [IQR]) baseline CD4 cells; 260 (134, 480)/mm3 vs. 280 (151, 440)/mm3, P = 0.46, or median CD4 cell nadirs 109 (50, 180)/mm3 vs. 120 (39, 224)/mm3, respectively, P = 0.63. GBV-C-positive compared with GBV-C-negative patients had similar CD4 cell counts at all follow-up visits, and similar median (IQR) increase in CD4 cells from baseline; 72.5 (-31.0, 138.3)/mm3 vs. 73.5 (-4.5, 154.8)/mm3 over 48 weeks of follow up in GBV-C-positive vs. GBV-C-negative patients, respectively. In patients who experienced an increase of ≥100 CD4 cells from baseline, the median (IQR) time to this increase was 12.1 (4.2, 35.3) and 13.7 (4.1, 28.1) weeks, for GBV-C-positive and GBV-C-negative patients, respectively (log rank P = 0.85).
HIV-1 RNA Decline and Virologic Failure
At baseline the HIV-1 RNA was marginally higher in GBV-C-positive patients (data not shown), but there were no differences in either the frequency of patients with HIV-1 RNA ≦400 copies/mL (GBV-C positive vs. GBV-C negative) or HIV-1 RNA ≦50 copies/mL. HIV-1 RNA decreased throughout follow-up in 157 of 166 patients (94.5%) with detectable viral load at baseline; 4 patients with and 5 without GBV-C did not experience a decrease in their HIV-1 RNA. There was no difference in the magnitude of the HIV-1 RNA decline according to GBV-C RNA status either overall or when restricted to those with detectable viral load at baseline.
Of the 258 patients included in this substudy, 55 experienced protocol-defined virologic failure (35 in GBV-C-negative patients and 20 in GBV-C-positive patients). After adjustments for baseline HIV-1 RNA levels, GBV-C serostatus was not found to predict virologic failure: Cox relative hazards (RH) 1.2, 95% CI: 0.7 to 2.0, P = 0.60. An HIV-1 RNA ≥400 copies/mL at baseline was the only significant predictor of HIV-1 RNA ≥400 copies/mL at the end of follow-up (odds ratio 9.9, 95% CI: 4.1 to 24.4, P < 0.0001) in a multivariate logistic regression analysis after adjustment for the CD4 cell count and GBV-C-status (GBV-C RNA positive vs. GBV-c RNA negative).
This remains the same even if viral failure is defined as >1000 or 5000 copies/mL (data not shown).
Adverse Events
Adverse events of grade 3 and 4 were not significantly different between GBV-C-positive and GBV-C-negative patients (data not shown). A total of 88 patients experienced an adverse event of grade 3 or 4 (28 or 32% of GBV-C-positive patients and 60 or 35% of GBV-C-negative patients), with 10 (11.6%) of GBV-C-positive and 23 (13.4%) of GBV-C-negative patients experiencing a multiple number of grade 3 or 4 adverse events, respectively (Fisher exact test, P = 0.99) (data not shown). There were no significant differences according to time to develop a grade 3 or 4 adverse event (log-rank P = 0.60) between GBV-C-positive and GBV-C-negative patients.
DISCUSSION
The present study investigated the influence of GBV-C on outcome measures in patients coinfected with HIV-1 and GBV-C in a randomized trial that assessed the safety and efficacy of 2 ritonavir-boosted HAART regimens.
Whereas studies in the pre-HAART era mostly universally showed a correlation of GBV-C viremia and a more beneficial course of HIV-1 disease, studies in the HAART era have been more conflicting.3
In this study we could neither observe any influence of GBV-C on any of the prognostic markers of HIV-1 infection at baseline nor on the response to HAART in the prospective data. This was also independent of patients being protease inhibitor or antiretroviral naive or experienced (data not shown).
Several shortcomings were accepted for this study that need to be overcome in subsequent studies. One is the short follow-up of only 1 year, as a moderate influence of GBV-C on the course of HIV-1 disease under HAART (both virologic and immunologic changes) might become evident only with longer follow-up. Furthermore, 2 studies have shown that the beneficial influence of GBV-C is lost, once GBV-C is no longer detectable in the blood.7-9 As we only had month 3 samples available, we cannot exclude that some patients have changed their GBV-C status on treatment. Thus in a future study, a sample prior to initiation and at the end of the study should be tested. This seems justified, especially in view of newer in vitro data further supporting an inhibition of HIV-1 by GBV-C.10-12
Further studies are required in advanced and less advanced disease stages. We still cannot exclude that GBV-C might be beneficial during the early period of HIV-1 infection but might no longer be beneficial once the disease has progressed.
Finally, the positive influence of GBV-C in HIV-1 infection appears relatively well established, taking into account that all reports show at worst a neutral effect, and most show a beneficial influence even though sometimes a nonsignificant one.13 It is possible that this positive effect is blunted by the very strong effect of HAART itself. Whether certain subgroups of HAART-treated patients have a beneficial effect of being GBV-C coinfected remains to be shown.
|
|
| |
| |
|
|
|
