Antibody no bar to enfuvirtide's anti-HIV effect
By David Douglas
NEW YORK (Reuters Health) - Existing antibodies to glycoprotein 41 (gp41) of HIV may cross-react with the new HIV fusion inhibitor enfuvirtide (Fuzeon, formerly T20), but this does not impair the efficacy or safety of therapy, according to researchers from Europe and North America.
Enfuvirtide is a synthetic peptide based on a motif within gp41 that regulates the binding of HIV-1 to host-cell membranes. In the December 15th issue of the Journal of Infectious Diseases, Dr. Sharon Walmsley of Toronto General Hospital and colleagues note that "most HIV-1 infected patients express antibody to gp41, which has the potential to cross-react with enfuvirtide."
To investigate whether such reactions might affect treatment, the researchers examined data from 2 phase III clinical trials of enfuvirtide involving more than 900 HIV patients.
Subjects were randomized to receive 90 mg enfuvirtide subcutaneously twice daily in combination with optimized background treatment or to optimized background treatment alone.
At baseline, most patients had detectable levels of the gp41 antibody. In 78% of enfuvirtide patients, levels fell by at least 30%. This was true of only 43% of control patients.
Baseline antibody status did not influence virological response. "In fact," Dr. Walmsley told Reuters Health, "those people who responded best to the drug had lower levels of antibody--a reflection of a decreased viral load."
Overall, the researchers conclude that "there was no evidence that enfuvirtide cross-reactive gp41 antibody affects the efficacy and safety of enfuvirtide."
Nevertheless, Dr. Walmsley also pointed out that it was not possible to "differentiate between antibody formed against the gp 41 of the virus and antibody against the drug."
J Infect Dis 2003;188:1827-1833.
Enfuvirtide (T-20) Cross-Reactive Glycoprotein 41 Antibody Does Not Impair the Efficacy or Safety of Enfuvirtide
The Journal of Infectious Diseases 2003;188:1827-1833
Sharon Walmsley,1 Keith Henry,2 Christine Katlama,5 Mark Nelson,7 Antonella Castagna,8 Jacques Reynes,6 Bonaventura Clotet,9 James Hui,3 Miklos Salgo,4 Ralph DeMasi,3 and John Delehanty3
University of Toronto, Toronto, Ontario, Canada; 2Hennepin County Medical Center, Minneapolis, Minnesota; 3Trimeris, Durham, North Carolina; 4Roche Diagnostics, Nutley, New Jersey; 5Pitié-Salpêtrière Hospital, Paris, and 6CHU Gui de Chauliac, Montpellier, France; 7Chelsea and Westminster Hospital, London, United Kingdom; 8Clinic of Infectious Diseases, Istituto de Recerca de la SIDA, RCCS Ospedale San Raffaele, Milan, Italy; 9Hospital Universitari Germans Trias i Pujol and IRSI-Caixa Foundation, Badalona, Barcelona, Spain
The present study investigated the effect of enfuvirtide cross-reactive glycoprotein 41 (gp41) antibody on the efficacy or safety of enfuvirtide in patients participating in 1 of 2 24-week phase 3 clinical trials (T-20 vs. optimized regimen only [TORO] 1 and TORO 2).
Serum samples from human immunodeficiency virusinfected patients receiving enfuvirtide plus optimized background (OB) and from patients receiving OB only were evaluated for enfuvirtide cross-reactive gp41 antibodies. Most patients had detectable levels of antibody at baseline; 78% of patients treated with enfuvirtide plus OB had a 30% decrease in level of antibody, compared with 43% of patients treated with OB only.
Baseline antibody status did not influence virological responses to enfuvirtide-containing treatment. Favorable virological responses were more common among patients who experienced a 30% decrease from baseline than among those who experienced either an increase or a lesser decrease. A decrease in virus load correlated with a decrease in level of antibody. Safety was unaffected by the presence of positive antibody at any time point or change in level of antibody.
There was no evidence that enfuvirtide cross-reactive gp41 antibody affects the efficacy or safety of enfuvirtide.
Enfuvirtide (T-20) is a human immunodeficiency virus (HIV)1 fusion inhibitor, the first member of a new class of antiretroviral agents for the treatment of HIV infection. The efficacy and safety of chronically administered enfuvirtide in HIV-1infected adults have been demonstrated in multiple phase 2 and 3 clinical trials.
Unlike conventional antiretroviral agents that work intracellularly, enfuvirtide is a peptide that exerts its antiviral activity outside the cell. It is a 36-aa synthetic peptide with a primary sequence derived from the naturally occurring motif (residues 643678) within the HR2 domain of the HIV-1HXB2 gp41 transmembrane glycoprotein. This protein plays a role in enabling fusion and entry of HIV into CD4 cells.
Enfuvirtide inhibits de novo infection and cell-to-cell virus transmission by binding specifically to a region of gp41 that regulates the fusion of HIV-1 to host-cell membranes.
The envelope glycoproteins of HIV-1, gp41 and gp120, are recognized by the immune system as foreign and elicit humoral antibody responses after natural infection or immunization with recombinant proteins. Antibody responses to gp41 and gp120 occur soon after acute infection and can exert neutralizing effects. This finding has raised the hope that it should be possible to develop an effective vaccine based on recombinant gp120 or recombinant gp160 (the unprocessed precursor of gp120 and gp41).
Since it is identical to a fragment of gp41, enfuvirtide may cross-react with existing antibodies to gp41 or may elicit its own humoral response and become neutralized by anti-enfuvirtide antibodies. Cynomolgus monkeys that previously had not been exposed to HIV-1, simian immunodeficiency virus, or enfuvirtide mounted a substantial antibody response when given enfuvirtide. On the basis of these findings, it was thought that enfuvirtide might elicit a humoral response in humans.
Most HIV-1infected patients express antibody to gp41, which has the potential to cross-react with enfuvirtide. Although enfuvirtide-specific antibodies did not reduce plasma levels of the drug in cynomolgus monkeys, the same situation cannot be assumed to occur in HIV-infected humans. Antibodies elicited in response to other exogenously administered agentssuch as insulin, factor VIII,
erythropoietin, granulocyte-macrophage colony-stimulating factor, and interferons (IFNs)have modified the biological effects and/or pharmacokinetics of these agents in humans. Neutralizing and nonneutralizing anti-IFN antibodies have been detected and are implicated in the failure of IFN therapy.
Given these observations, it is important to determine the immune response to long-term administered enfuvirtide in humans and to establish whether gp41 antibodies or antibodies induced during treatment with enfuvirtide influence its efficacy or safety in HIV-1infected patients. The present study estimates the incidence of cross-reactive gp41 antibody in HIV-infected patients at study entry, the incidence of changes in serum gp41 antibody status during treatment with either enfuvirtide plus optimized background (OB) or OB only, the presence
and effect of serum gp41 antibody at study entry, and the change of level of antibody during treatment, on selected efficacy and safety end points.
DISCUSSION by authors
The present study has investigated whether enfuvirtide cross-reactive gp41 antibody influences the performance of enfuvirtide during therapy in phase 3 clinical trials. Antibody was assessed by the presence or absence of antibody at baseline and week 24, by the maximum level of antibody achieved, and by incremental changes in level of antibody during treatment. Results demonstrate that enfuvirtide cross-reactive gp41 antibody does not negatively affect the efficacy or safety of chronically administered enfuvirtide.
Serum from almost all patients had some evidence of positive levels of enfuvirtide cross-reactive gp41 antibody at baseline (77.4% in the enfuvirtide-plus-OB group and 74.2% in the OB-only group), most likely due to naturally occurring antibody to gp41 in HIV-infected patients. The level of enfuvirtide cross-reactive gp41 antibody was nonquantifiable in 20%25% of patients in both treatment groups. It
was rare (1.3% in both treatment groups) for a patient to test negative for enfuvirtide cross-reactive gp41 antibody at baseline, which highlights the ability of patients' immune systems to recognize gp41.
Greater HIV RNA suppression was associated with a decrease in levels of enfuvirtide cross-reactive gp41 antibody. This suggests that antiretroviral therapy, particularly with enfuvirtide, in this patient population can affect levels of gp41 antibody by presumably reducing antigen burden. It would therefore follow that the decrease in level of antibody would be significantly greater in the enfuvirtide-plus-OB group than in the OB-only group, since enfuvirtide-plus-OB treatment had a superior effect on suppression of HIV-1 RNA in the TORO 1 and TORO 2 trials.
Maximum changes from baseline in level of antibody showed a clear relationship with efficacy, according to all measures of virological response. Enfuvirtide-plus-OB recipients who experienced a maximum decrease in level of antibody of 30% from baseline showed a greater mean decrease in HIV-1 RNA load from baseline and fewer virological failures, and there was a higher proportion of those patients
with l.0 log10 decrease in HIV-1 RNA load from baseline, compared with enfuvirtide-plus-OB recipients who experienced either less change (i.e., ±30%) or a maximum increase in level of antibody of 30% from baseline.
Increases in the level of enfuvirtide cross-reactive gp41 antibody were uncommon in both treatment groups, suggesting that chronic exposure to enfuvirtide is unlikely to elicit a marked humoral response. Since enfuvirtide cross-reactive gp41 antibody has no effect on the pharmacokinetics of enfuvirtide, treatment-emergent increases in level of antibody is unlikely to compromise the efficacy of enfuvirtide.
None of the safety assessments revealed any clear evidence to suggest that the presence of enfuvirtide cross-reactive gp41 antibody influences the safety of enfuvirtide during 24 weeks of therapy. Results from the TORO trials indicate that enfuvirtide-plus-OB treatment has a comparable safety profile to OB-only treatment. There were few cases in which the incidence of clinical or laboratory toxicity occurred in >5% more patients in the enfuvirtide-plus-OB group than in the OB-only group. One exception was eosinophilia, which occurred in 10.1% of patients receiving enfuvirtide plus OB and in 2.4% of patients receiving OB only, but in only 1.8% and 0.9% of patients, respectively, was the incidence of eosinophila greater than twice the upper limit of normal. Overall, any manifestation of hypersensitivity occurred in 19.3% of patients receiving enfuvirtide plus OB versus 17.4% of patients receiving OB only. The incidences of adverse events, hypersensitivity reactions, or ISRs in enfuvirtide-plus-OB recipients were similar, irrespective of baseline or maximum during-treatment antibody status. Only 1 patient tested negative for enfuvirtide cross-reactive gp41 antibody at baseline and subsequently tested positive for antibody (i.e., seroconverted), with a level of antibody at week 24 of 0.258 g/mL. In addition to an ISR, this patient had diarrhea (study day 1), bronchitis (study day 24), fatigue (study day 33), bronchitis (study day 103), back pain (study day 109), and paresthesia (study day 135), but none of these events suggested an immune-mediated reaction.
The overall incidences of treatment-emergent grade 3 or 4 laboratory abnormalities indicated that the presence of enfuvirtide cross-reactive
gp41 antibody had no apparent influence on the safety of enfuvirtide during 24 weeks of therapy. The differences between treatment groups are mostly attributable to a higher incidence of eosinophilia in the enfuvirtide-plus-OB group than in the OB-only group, which is consistent with observations of specimens obtained from patients receiving enfuvirtide. Hematoxylin-eosin staining of excisional biopsy specimens obtained from patients participating in phase 2 and 3 clinical trials of enfuvirtide have revealed an inflammatory response with conspicuous eosinophils and histiocytes.
Antibodies against gp41 may be elicited to viral debris that is expelled after cell lysis rather than as a response to intact virions. This debris could include incomplete or defective viral particles and peptides in various stages of processing. Since these antibodies may recognize various oligomeric conformations of gp41, they may also bind to enfuvirtide. The immune-based assay used in the present study cannot distinguish between antibodies directed specifically against enfuvirtide and gp41 antibodies that cross-react with enfuvirtide. Irrespective of
the antigen responsible for the cross-reacting antibodies, the findings of the present study are compatible with those of previous studies indicating the low neutralizing antibody activity in HIV-1 infection. Even if antibodies that recognize various oligomeric conformations of gp41 do bind to enfuvirtide, they do not appear to influence its antiviral activity.
The analysis of the combined TORO clinical trial population revealed that there was no meaningful influence on the clinical antiviral activity of enfuvirtide by cross-reacting antibodies, and suppression of HIV-1 RNA was associated with the decrease in level of enfuvirtide cross-reacting gp41 antibody. No association was detected between presence and level of antibody and any safety measure during 24 weeks of treatment with enfuvirtide.