Otelixizumab
Chemistry
Otelixizumab (or ChAglyCD3) was developed by Herman Waldman and was subsequently developed by Tolerx in collaboration with GSK (Cambridge, USA).[15] The monoclonal antibody was derived from the rat anti-CD3 antibody YTH 12.5, and consists of the α rat/human λ light chain and γ1 heavy chain, which lacks the CD2-domain glycosylation site.[15–17] As a consequence it is aglycosylated in the Fc domain, limiting its ability to bind to complement or Fc receptors (FcR). It is directed against human CD3ε, which is a part of the CD3–TCR complex present on T lymphocytes.
Pharmacokinetics & Pharmacodynamics
The pharmacokinetics of otelixizumab are in agreement with a one-compartment model with a first order elimination.[18] Nonlinearity was manifested at high concentrations (Km = 0.968 µg/ml).[19] The estimated elimination rate constant was 0.457 per day and the half-life was 1.52 day. The distribution volume was estimated at 7.56 l and the clearance was 3.45 l/d.[18] In another report, the distribution volume was estimated at 13.9 l.[19] Dose-dependent accumulation of otelixizumab was demonstrated and modeling of the data indicated that the terminal half-life was approximately 1.5 days.[18]
In diabetic subjects, the otelixizumab serum concentrations resulting in a 50% decrease in peripheral CD4+ cells and CD8+ cells were 0.0187 and 0.0120 µg/ml, respectively.[19] The rapid, dose-dependent decline of lymphocytes after administration of otelixizumab followed by a return to baseline after drug clearing is suggestive of a redistribution phenomenon and argues against apoptosis as an important contributor to T-cell decline.[19] Furthermore, the authors demonstrated that there were few otelixizumab-CD3–TCR complexes at the T-cell surface showing rapid modulation and saturation of CD3–TCR at the cell surface.[19]
High-dose administration of otelixizumab can result in headache and/or vomiting. In a recent study, four of nine patients receiving 24 mg of otelixizumab experienced these symptoms,[18] whereas none of the 71 patients administered six consecutive infusions of 8 mg/d experienced infusion-related reactions.[18]
Mechanism of Action
The effects of anti-CD3 antibodies on the immune system are diverse and incompletely understood at this moment; TCR blockade and internalization, shifting the Th1/Th2 balance, induction of T-cell anergy or hyporesponsiveness and induction of Tregs have all been reported.[20–31]
Anti-CD3 monoclonal antibodies exert their effects through modulation of the CD3–TCR complex through a process called antigenic modulation. Ligation of the CD3–TCR complex results in blockade and internalization of the receptor complex.[32] Under normal conditions, there is a constant internalization and recycling to the cell surface of the CD3–TCR complex.[33,34] Upon T-cell activation by anti-CD3 antibodies, the internalization of the CD3–TCR is increased.[35] Furthermore, engaged receptors are degraded through lysosome- and proteosome-dependent mechanisms after T-cell activation by anti-CD3 antibodies.[35–37] This alteration in TCR-mediated signal transduction prevents normal immunological synapse formation (and associated events such as calcium influx) necessary for normal T-cell activation.[21,38] Furthermore, ligation of the TCR outside the antigenic region leads to preferential activation of the SRC-family kinase FYN resulting in the downstream inactivation of T cells.[21,22]
Administration of anti-CD3 antibodies results in T-cell activation. In addition, apoptosis and redistribution of effector T cells have been reported. A recent study demonstrated that the administration of anti-CD3 antibodies led to rapid redistribution of T cells resulting in transient lymphopenia.[39,40] Anti-CD3 antibodies have also been reported to increase the susceptibility of effector T cells to Fas-mediated apoptosis through the induction of CD95 and CD95 death-inducing signaling complex following ligation of CD3 by the antibody. As CD95 is an activation marker, this phenomenon explains why effector T cells at the site of inflammation are the most affected by anti-CD3 antibody administration.[29,41] For otelixizumab, apoptosis only affects a limited proportion of lymphocytes as lymphocyte numbers return rapidly to normal levels after clearance of the drug.[19]
Different results concerning the effects of anti-CD3 treatment on Tregs have been reported: from absent,[42] to modest,[31,43] to restricted to particular niches[31,44] or involving Tregs with an atypical phenotype (CD25low).[45] Some of these issues can be attributed to the mouse models used; for example in the NOD mouse a high proportion of Tregs is contained in the CD25low compartment.[46] Recently, Nishio et al. demonstrated that anti-CD3 treatment resulted in a selective proliferative expansion of particular Treg clonotypes.[47] Whereas normally Tregs are regulated within TCR-specific niches in secondary lymphoid organs and their numbers are kept in check by intraclonal competition, anti-CD3 treatment overcomes these niche limitations in IL-2- and IL-7-dependent manner and results in an 50–60-fold expansion of some Treg clonotypes.[47] Both Tregs and effector T cells proliferated substantially in the first day following anti-CD3 treatment; effector T cells ceased to proliferate after day 1 whereas Tregs continued to expand further for several days.[47] These expanded Tregs operate in a transforming growth factor-β-dependent manner and an increased sensitivity of pathogenic T cells to the effects of Tregs after anti-CD3 treatment has been reported.[31,45,48] Besides expansion, Tregs are relatively more resistant to cell death induced by FcR-nonbinding CD3-specific monoclonal antibodies compared with recently activated effector and naive T cells.[29,41] The abnormal signals through the CD3–TCR complex after ligation of FcR-nonbinding CD3-specific monoclonal antibodies are partially responsible for their tolerogenic effect on T cells, as the induction of tolerance was completely blocked by coadministration of cyclosporine.[30] The combined effects of anti-CD3 treatment on effector T cells and Tregs results in a predominance of Tregs and this accumulation is most pronounced in grafted tissue and the draining lymph nodes.[49–52]
Based on these findings, a two-phase action model has been proposed.[41,47] In the first phase, which occurs concomitantly with antibody administration, pathogenic T cells in the ongoing immune response are eliminated by the anti-CD3 antibody. Although short-lived, this effect might be desirable in limiting acute damage in patients with rapidly evolving disease. The second phase, which results in long-lasting protective effect associated with anti-CD3 treatment, is characterized by the coexistence of effector T cells and TGF- β-dependent Tregs with associated reassortment of the Treg repertoires. Through location in vicinity of the target organ (pancreatic islets and draining lymph nodes) and resetting of Treg niches, Tregs will control the pathogenic effector cells.
Preclinical Studies
Anti-CD3 antibodies have been shown to be able to prevent and reverse established autoimmune diabetes in different animal models.[30,53–56] In the NOD mouse, anti-CD3 antibodies were able to restore lost self-tolerance to β-cells and in this manner reverse recent-onset diabetes mellitus.[30,56] Interestingly, FcR-binding anti-CD3 antibodies were more effective when administered at the time of disease onset compared with administration in very young 4–8 week-old mice, in which there are no signs yet of a destructive cell infiltration.[30] These findings are compatible with the selective destruction of activated T cells as a key protective mechanism early after antibody administration.[30] This beneficial effect of anti-CD3 treatment has also been demonstrated in other animal models of autoimmune disease for example, experimental autoimmune encephalomyelitis[44] and inflammatory bowel disease.[57] In these models there also appears to be a strict therapeutic window in which CD3 treatment is efficient in restoring tolerance.
The study of the humanized anti-CD3 antibodies has been hampered by the fact that anti-CD3 antibodies are strictly species-specific. To circumvent this problem, recently a transgenic mouse model expressing the ε chain of the human CD3 complex on the NOD background (NOD–huCD3ε) has been developed.[58] In this model, the authors were able to confirm the findings obtained using mouse-specific antibodies in mouse models.[58] When evaluating the effect of antihuman CD3 antibody administration in new-onset diabetes in NOD–huCD3ε mice, 80% of mice recovered normal glycemic control in the absence of any exogenous insulin administration within 1 week of the start of the treatment. By the end of treatment, a complete clearance of insulitis was demonstrated.[58] Moreover, this remission correlated with features of T-cell mediated active tolerance with increased numbers and suppressive capacity of CD4+CD25+FoxP3+ Tregs and decreased numbers of autoantigen-specific CD8+ T cells.[58] The resistance of pathogenic effector cells to Tregs was reversed upon the administration of anti-CD3 antibody and was associated with an increase in serum levels of TGF-β.[58] In our opinion, this model will be valuable in the development and evaluation of novel human anti-CD3 antibodies.
Clinical Studies: Phase I/II
Two FcR-nonbinding anti-CD3 antibodies have been used to protect β-cell mass in patients with recently diagnosed Type 1 diabetes (Table 1).[59–61] Both antibodies showed the potential to rapidly arrest ongoing autoimmunity and possibly shift immunological processes towards tolerance through a short treatment regimen.[59–61]
In a US Phase I and II clinical trial evaluating another FcR-nonbinding anti-CD3 antibody (teplizumab or hOKT3γ1 Ala–Ala) at the time of onset of Type 1 diabetes, anti-CD3 treatment halted the progression of the disease for more than 1 year and beneficial effects on C-peptide production and insulin requirements where noted 3 years post-treatment.[61] Two different Tregs were detected in vivo after anti-CD3 administration; IL-10-producing CD4+ and FOXP3+CD8+ Tregs.[62]
The Belgian Diabetes Registry (BDR) coordinated a European multicenter Phase II randomized, placebo-controlled clinical trial where 80 patients with recent-onset Type 1 diabetes were randomly assigned to receive either intravenous otelixizumab (8 mg) or placebo for 6 consecutive days. By using the hyperglycemic clamp test to evaluate the functional β-cell mass, it was demonstrated that in new-onset Type 1 diabetic patients of this age category the functional β-cell mass at diagnosis averaged only 25% of that in nondiabetic controls. Patients receiving otelixizumab preserved β-cell function more efficiently resulting in higher endogenous insulin production at 6, 12 and 18 months post- treatment.[60,63] This translated in a significant decrease in the insulin requirement of the patients up to 48 months after the single course of otelixizumab.[60,63] The patients with the best response were those who had a higher functioning β-cell mass at the beginning of treatment (4-weeks post-diagnosis); 75% of the subset of patients with a high β-cell mass (higher than the median value of the whole population) required insulin doses of ≤0.25 U/kg/day at 18 months: a dose requirement that is compatible with clinical insulin independency.[60] Moreover, patients below 27 years of age were more likely to experience positive effects on β-cell mass, insulin need, glycemic variability and HbA1c, without an increase in hypoglycemic events. Younger age and higher initial β-cell mass were also predictive of the duration of the beneficial effects of otelixizumab treatment.[63] There was no effect on the development of chronic complications, but this would of course require a much longer follow-up. At the doses used in the BDR study, clear side effects, such as fever, arthralgia and headache, were reported as well as Epstein–Barr virus (EBV) reactivation. However all these side effects were of moderate severity and limited in time.
Side Effects & Safety
The first FDA-approved anti-CD3 antibody (OKT3) was associated with important side effects as this FcR-binding CD3-specific monoclonal antibody is a potent mitogen and releases cytokines following in vivo administration.[64,65] The mitogenic property of anti-CD3 antibodies is dependent on the capacity of the Fc portion to interact with FcRs on phagocytes and natural killer cells and involves the production of IL-1 β and IL-6 by these cells.[66] This cytokine release is transient and associated with a 'flu-like' syndrome characterized by fever, chills, headaches, nausea, vomiting, diarrhea, respiratory distress, aseptic meningitis and hypotension.[67–69] Besides the Fc portion of the antibody, the epitope specificity of the anti-CD3 antibody also influences its mitogenic potential.[70]
FcR-nonbinding CD3-specific antibodies are not mitogenic and do not induce large-scale systemic cytokine release, however, they do result in limited T-cell activation and low levels of cytokines can be detected in patients treated with FcR-nonbinding antibodies.[71–73] This partial activation is a prerequisite to obtain a therapeutic effect of anti-CD3 antibody therapy.[30] So, there is a clear distinction between the mitogenic potential of CD3-specific antibodies, which does not seem to be required for their therapeutic efficacy and the partial signaling/activating capacity, which is required for the therapeutic effect.
As these antibodies will be used in the treatment of Type 1 diabetes, especially in pediatric patients, clinical safety is a major issue. There is a need to reduce the residual side-effects observed in some patients, although a partial activation probably has to be preserved in order to retain the therapeutic efficacy of the anti-CD3 antibody.[59,60,74] In the European trial, transient EBV reactivation was observed 10–20 days after the start of anti-CD3 antibody treatment in patients who had already experienced a primary EBV infection.[74]
Another concern is the development of antibodies directed against the anti-CD3 antibody. In the past, the use of OKT3 was associated with the rapid occurrence of an anti-globulin response (mainly IgM and IgG) and resulted in accelerated clearance and neutralization of this anti-CD3 antibody.[75] Although an extremely rare event, the development of OKT3-specific IgE antibodies-associated risk of anaphylaxis- has been described.[76,77] Studies have demonstrated that the antibody response has remarkable restricted specificity resulting in a situation where the patient immunized with a monoclonal antibody is still responsive to a second monoclonal antibody with the same specificity but a different idiotype.[78] Furthermore, sensitization to OKT3 has never been associated with the occurrence of serum sickness as the amount of immune complexes formed was probably insufficient to allow sustained tissue deposition.
The FcR-nonbinding CD3-specific antibodies have been reported to be less immunogenic than their conventional FcR-binding counterparts.[79,80] In a recent study, anti-otelixizumab antibodies could be detected in 77.5% of patients.[18] Among the 28 drug-treated patients who were tested for neutralizing antibodies, five were negative in the neutralizing assay and 23 were positive.[18] The antibody response was primarily directed against the anti-CD3 CDR regions of otelixizumab and not against any new epitopes exposed by aglycosylation nor against the remaining rat portion of the light chain.[18] Importantly, there appeared to be a correlation between the presence of neutralizing antibodies and clinical outcome as there was a consistent trend toward higher C-peptide levels in the patients who did not have detectable neutralizing antibodies.[18] Other studies have demonstrated that in most patients these antibodies disappear by 6 months and only a small percentage of patients continue to produce them long term.[59–61] Despite their potential neutralizing capacity, as they appeared by 2–3 weeks after the last dose of drug was injected, anti-idiotypic antibodies do not represent a problem for CD3-specific antibody treatment. However, these anti-idiotypic antibodies will possibly represent a potential problem if repeated treatment is needed.
Clinical Studies: Phase II/III
As discussed above, the proof-of-principle studies using both anti-CD3 antibodies (teplizumab and otelixizumab) showed effects on surrogate markers (C-peptide preservation) in newly diagnosed Type 1 subjects, but in particular the placebo-controlled BDR-study using otelixizumab, uncovered important dose-dependent side-effects as described: flu-like symptoms (probably related to cytokine release) and more importantly EBV reactivation. Thus, subsequent multicenter Phase III studies were initiated using much lower doses than those in the BDR study (48 mg otelixizumab) (Table 1).
For teplizumab, the Protégé Phase III follow-up trial was initiated by MacroGenics and Eli Lilly. This is a multinational double-blind randomized controlled trial in newly diagnosed patients with Type 1 diabetes (<12 weeks after diagnosis). In this dose-finding study, more than 500 children (8–17 years of age) and adults (18–35 years of age) were randomized to receive a 14-day course of intravenous teplizumab. The primary end point was a composite end point of total daily insulin dosage and HbA1c at 12 months after teplizumab administration. A planned analysis of 1-year safety and efficacy data revealed no safety issues but unfortunately the primary efficacy end point was not met.[81] However, post-hoc analysis demonstrated that the higher dose group (cumulative dose 17 mg) did show a prolongation of C-peptide positivity, particularly in children. These data are in alignment with 4-year observations for otelixizumab in the BDR trial.[63] Another trial studying tepluzimab in recent-onset Type 1 diabetic patients (AbATE) was reported recently (American Diabetes Association June 2011); where high doses (35 mg) again demonstrated a delay in C-peptide decline, but no additional effect of a second therapy course at 1 year after diagnosis was observed [Herold K, Unpublished Data]. At present, enrollment and dosing of patients in two other ongoing clinical trials of teplizumab in Type 1 diabetes, the Protégé Encore Trial and the SUBCUE trial were terminated. The findings of another trial investigating the potential of teplizumab (Delay) should be reported soon.
Also Tolerx initiated the DEFEND1 and 2 studies in newly diagnosed patients with Type 1 diabetes (<90 days after diagnosis) in a multinational, multicenter setting.[202] In a double-blind, placebo-controlled study, intravenous administration of 3.1 mg otelixizumab was evaluated[60,74] in adolescents (n = 29; 12–17 years of age) and adults (n = 243; 18–45 years of age) (Table 1).[202] The primary end point was stimulated C-peptide during a mixed-meal stimulation test at 12 months after otelixizumab administration. Also here, the analysis at 12 months did not reveal any safety issues, but again the primary end point was not met. For that reason, the DEFEND2 study, investigating the same intravenous dose of otelixizumab but mainly in adolescents, was terminated. However, the GSK-initiated Phase I study on subcutaneous administration of otelixizumab is still going on and results are eagerly awaited.
Immunotherapy. 2011;3(11):1303-1306. © 2011 Future Medicine Ltd.
Comments