Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (9)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Isaacs, J. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Isaacs, J. D.
Related Collections
Right arrow Rheumatoid Arthritis
Right arrow Systemic Lupus Erythematosus and Autoimmunity
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Rheumatology 2001; 40: 724-738
© 2001 British Society for Rheumatology

Review

From bench to bedside: discovering rules for antibody design, and improving serotherapy with monoclonal antibodies

The 1999 Michael Mason Prize Essay

J. D. Isaacs

Rheumatology and Rehabilitation Research Unit/Molecular Medicine Unit, Clinical Sciences Building, St James's University Hospital, Leeds LS9 7TF, UK


   Abstract
Top
 Abstract
Introduction
 The fate of the...
 Immunogenicity
 The ‘first-dose’...
 Summary
 References
 
Anti-T-cell monoclonal antibodies (mAbs) form a unique class of therapeutic agent. Their precise specificity offers tremendous potential for the treatment of autoimmune and inflammatory diseases but also prevents meaningful preclinical animal studies. In particular, adverse reactions to therapy may be unanticipated, and the first administration of a novel T-cell mAb to a patient thus marks the beginning of a unique experiment. By comparing clinical parameters and laboratory measurements, small-scale pilot studies can provide detailed information about mAb biology that both predicts and suggests solutions to the complications of therapy. In this essay I illustrate this concept with reference to three specific areas: lymphocyte depletion, mAb immunogenicity and cytokine-release syndromes. In each case, systematic clinical and laboratory science has improved our understanding of the problem and suggested solutions; most of these solutions have been or are being adopted. Thus, small, open studies are an essential step in the development of novel mAbs, provide an ideal platform for the study of mAb biology, and serve as an early warning system for potential adverse effects.

KEY WORDS: Monoclonal antibody, Immunotherapy, Effector function, Lymphopenia, Immunogenicity, Humanization, First-dose reaction, Cytokine release reaction, Mutagenesis.


   Introduction
Top
 Abstract
 Introduction
The fate of the...
 Immunogenicity
 The ‘first-dose’...
 Summary
 References
 
Animal studies leave little doubt as to the potential therapeutic power of T-cell-directed monoclonal antibodies (mAbs) in transplantation and autoimmune disease. Their application can not only prevent illness but, more importantly, halt and reverse ongoing immunopathology. Perhaps the most impressive aspect of their use is that these outcomes can be achieved with relatively brief courses of therapy [1]. The mechanisms underlying such potent effects are not yet entirely clear, but it seems that therapy is associated with the development of regulatory lymphocytes which keep autoreactive cells in check [2].

There are already hints that equally powerful effects may be attainable in man. Thus, over the past 5 yr we have witnessed apparently permanent modulation of severe, refractory human immunopathology following brief courses of mAb therapy [3–6]. In contrast to these anecdotal achievements in vasculitis and ocular disease, however, success has been more limited in commoner conditions, such as rheumatoid arthritis (RA) [7–9]. There are a number of potential reasons for this, which have been well rehearsed [10, 11], and many are currently being addressed by our own group and by others. For example, in ongoing studies we are examining the effects of increasingly intensive regimes of CD4 mAb therapy in RA and, additionally, the need to combat inflammation prior to anti-T-cell therapy [12, 13]. Similarly, in a forthcoming study we intend to address the potential importance of CD8 lymphocytes in active RA. However, the major obstacle to successful studies is, paradoxically, the specificity of mAbs. Because of this we cannot simply take a mAb which has been effective in a mouse model of RA and administer it to our patients: it would almost certainly fail to bind the equivalent human target. This is a limitation because we do not understand how mAbs achieve their immunomodulatory effects in animals. Thus, we cannot examine a mAb which is efficacious in animal models and, based on its biological profile, design a similar mAb for human therapy: we do not know which aspects of this profile are important. Until such information becomes available, human mAb therapy will continue along a somewhat pragmatic path: we will test novel agents as they become available, not knowing whether, for example, one CD4 mAb is more or less likely than the next to be an effective immunomodulatory therapy. At present, the only way to find out is to compare them in the clinic. (A potential solution is to test novel mAbs in primates, which share many antigenic specificities with humans, but this approach raises its own ethical issues.)

So, to the topic of this essay. When T-cell mAbs are administered to laboratory animals they appear extremely safe. For example, one of the prevailing arguments for the relative inefficacy of CD4 mAbs in man is that the doses administered are too small: 1 g of mAb seems a large dose to administer to our patients, yet if we scale up the quantity required to modulate severe immunopathology in mice, we arrive at figures in excess of 100 g. Even so, laboratory mice do not appear to suffer from such massive amounts of treatment, and it has therefore been assumed that these are relatively safe drugs. There are obvious loopholes to this argument (not least the fact that laboratory mice tend not to complain!), and an alternative view might be as follows. If antibodies have evolved through millions of years to become the exquisitely specific, highly efficacious proteins that they are, then they should possess fairly powerful biological effects. It follows that if we infuse large amounts of monospecific mAbs into patients, then perhaps we should expect adverse as well as beneficial outcomes. As with efficacy, however, this cannot be assessed directly in animal studies—in this case not only will the mAb variable (V)-region fail to recognize its target, but the constant (C)-region (again highly evolved, and encompassing a variety of natural isotypes and polymorphic allotypes) may not ‘dock’ appropriately with murine effector mechanisms (complement and Fc receptors). This essay focuses on our initial experiences using mAbs for treatment of autoimmune disease and, in particular, on three specific side-effects: target cell depletion, cytokine release syndromes and immunogenicity. In each case, in vitro models and assays led to suggestions for improving tolerability which have now been incorporated in second- and third-generation agents. The important message is that, to obtain the best results from modern biological therapies, clinical research must be supported by excellent laboratory science. Furthermore, with the appropriate clinical/laboratory interface, much can be learned from small, observational, open studies.


   The fate of the target cell
Top
 Abstract
 Introduction
 The fate of the...
Immunogenicity
 The ‘first-dose’...
 Summary
 References
 
When mAbs were first applied as therapies in animal models of autoimmune disease, depletion of targeted lymphocytes was the order of the day [14]. The rationale was that if an autoreactive immune system could be ablated, a new immune system may develop in its place which would not necessarily share the same autoaggressive characteristics. Indeed, such strategies were highly successful in animal models, although later experiments suggested that non-depleting mAbs were equally effective [15]. Of note, lymphocyte depletion did not seem to adversely affect animals housed in conventional ‘dirty’ facilities, suggesting that immunosuppression was not a major hazard of therapy. Consequently, one of the first mAbs to be developed as a potential immunomodulator in humans was originally selected because of its promise as a lymphocytotoxic agent [16]. The mAb, directed at an antigen now known as CD52 but at the time as the Campath-1 antigen (Campath because the mAb was made in the Cambridge University Department of Pathology), was christened CAMPATH-1 (upper-case to denote the antibody as opposed to the antigen). CD52 is present on lymphocytes, natural killer (NK) cells and monocytes [17]. The original hybridoma was generated by immunizing a rat with human lymphocytes and so the original mAbs were of rat origin. The rat IgM version (CAMPATH-1M) was exceptionally potent in complement-mediated lysis (CML) in vitro, and the rat IgG2b (CAMPATH-1G) in both CML and antibody-dependent cell-mediated cytotoxicity (ADCC, an in vitro assay traditionally used as a surrogate marker for the capacity of a mAb to kill target cells in vivo via Fc{gamma} receptors) [18].

The capacity of CAMPATH-1 mAbs to kill cells in vivo had already been exploited in the therapy of lymphoreticular malignancies [19], when it was decided to use them to treat autoimmune disease. CAMPATH-1H, a ‘humanized’ version of the mAb [20] with a human IgG1 Fc region, was first administered to RA patients in June 1991 [21] and trials continued locally and internationally (under the direction of Wellcome and Burroughs Wellcome) until 1994 [22–24]. Placebo-controlled studies were never performed, and therefore the transient improvement in symptoms associated with therapy were never formally confirmed. As predicted, the mAb was extremely potent at target cell depletion, lymphopenia appearing in peripheral blood with doses as low as 1 mg [22]. The duration of lymphopenia was, however, unexpected and disconcerting. This lasted from months to years and particularly affected the CD4+ subset of peripheral blood T cells (Fig. 1Go) [25]. At the time, this was attributed to a defect in lymphocyte reconstitution, perhaps intrinsic to the RA disease process itself or related to concurrent or subsequent therapies. Recent evidence suggests that it merely reflected the inability of an involuted adult thymus to permit maturation of marrow-derived lymphocyte precursors. Thus, it is now recognized that similar kinetics of reconstitution are seen in other situations, such as bone marrow and autologous stem cell transplantation when performed after adolescence [26]. (This is an important observation, given the current trend for autologous stem cell transplantation in RA, which may result in equally dramatic and long-lasting lymphopenia.)



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 1. Long-term lymphopenia following therapy with CAMPATH-1H. Forty-one patients received a total dose of 100–400 mg of CAMPATH-1H over 5 or 10 days. The figure illustrates peripheral blood lymphocyte subsets at various time points after therapy. There was no difference between dosing cohorts and data were therefore pooled. *CD4 (P=0.0001 at day 178) and CD8 (P=0.03 at day 178) counts were different from baseline at all time points. CD19 was different from baseline to day 87 (P=0.03). CD14 was different from baseline to day 31 (P=0.0001). CD16 was different from baseline at day 3 (P=0.0001). Two-sample t-test. From [24].

 
Subsequent to our work with CAMPATH-1H, prolonged lymphopenia was also seen with a CD4 mAb, cM-T412. This was a chimaeric mouse/human mAb, also of hIgG1 isotype, which was employed in a number of controlled and uncontrolled studies in RA [7, 8, 27–30]. cM-T412-associated lymphopenia exemplified the uncertainties facing immunotherapists at that time. A number of rodent CD4 mAbs had already been administered to RA patients with variable but always transient lymphocyte depletion [10] and, although a mAb of hIgG1 isotype was expected to deplete more potently, another hIgG1 CD4 mAb did not [11]. Thus, the lack of preclinical testing resulted in some surprising biological outcomes of mAb therapy, many of which were not predictable or readily understandable.

The relevance of prolonged peripheral blood CD4+ lymphopenia in this context, particularly when the synovium remains T-cell-replete [31], is unclear. Our own long-term follow-up of these patients has failed to identify a high infection risk [32], but in the era of HIV-1, and in the absence of any clear long-term effects on the RA disease process, CAMPATH-1H (and cM-T412) were dropped as potential therapies for RA. In other circumstances, however, dramatic effects were witnessed in small, uncontrolled studies of CAMPATH-1H and we continue to use this mAb in the therapy of refractory vasculitis [3, 4, 33], multiple sclerosis [34], various forms of eye disease [5, 6, 35] and occasionally in other conditions [36].

Although CAMPATH-1H was initially selected as a depleting mAb on the basis of in vitro lymphocytotoxicity, at that time there was little evidence linking in vitro tests such as CML and ADCC with in vivo activities of mAbs. For example, a contemporaneous paper could show no relationship between CD4+ lymphocyte depletion in vivo in mice and CML and ADCC in vitro for a panel of CD4 mAbs [37]. Furthermore, other studies showed that rules generated for one antigen did not necessarily translate to another [38] or to the same antigen at a different density [39], and increasingly population polymorphisms were being identified which might alter mAb biology in vivo [40]. These data, together with the human CD4 experience referred to above, encouraged us to develop a preclinical model that would help us to understand how mAbs deplete target cells in in vivo. We expected that this would help us to generate rules for designing mAbs which were tailor-made for particular therapeutic scenarios.

Our model was simple [41]. First, using recombinant molecular techniques, we developed a large panel of chimaeric mAbs specific for the mouse CD8 antigen. These shared the same V-region but possessed a range of wild-type and mutated C-regions. Each underwent extensive in vitro testing to characterize its biological activities [41]. They were then administered, in parallel, to groups of thymectomized CBA/ca mice. Their in vivo effects were monitored using flow cytometry to document the number of CD8+ lymphocytes remaining in peripheral blood at subsequent time points. Because the animals had been thymectomized, they were unable to reconstitute lysed cells, enabling us to differentiate between transient sequestration and cytotoxicity. (Ironically, the long-term depletion seen in thymectomized mice paralleled the effects of CAMPATH-1H and cM-T412 in patients; as suggested above, in humans the thymus involutes after adolescence, resulting in a relative ‘physiological’ thymectomy.) We selected the CD8 antigen for these experiments in the knowledge that this was a sensitive mAb target in mice [42] and because limited quantities of chimaeric mAb were likely to be available from our small-scale laboratory cultures of transfectomas.

Our strategy was rewarded when we demonstrated that as little as 5 µg of chimaeric mAb could lead to lasting depletion of CD8+ peripheral blood lymphocytes. Both rat and human isotypes interacted with mouse effector systems, resulting in efficient depletion of cells. In our initial experiments, rat IgG1 (rIgG1), rIgG2b, human IgG1 (hIgG1), hIgG2, hIgG3 and hIgG4 were all potent depleters, whereas rIgG2c, hIgA and hIgE were impotent and rIgG2a had an intermediate potency (Fig. 2Go). It was surprising that isotypes that were very poor at harnessing in vitro effector activities (hIgG2, hIgG4) were so potent in vivo and, to an extent, this validated the need for an in vivo model. Equally, however, this provoked criticisms over the value of a heterologous system (human and rat mAbs administered to mice) and raised the possibility that our results were artefactual. We therefore dissected the system further. We demonstrated that an aglycosyl variant of hIgG1, created by site-directed mutagenesis of amino acid residue 297, no longer depleted [41]. The N-linked carbohydrate attached to residue 297 of IgGs was known to be essential for complement activation and Fc{gamma} receptor (Fc{gamma}R) binding [43], suggesting that our initial results were a genuine reflection of mAb–effector mechanism interactions. We then showed that a mutant of hIgG1 that could not bind Clq still depleted, even in mice lacking the C3 component of complement, indisputably excluding complement as an effector pathway in this model [41]. Finally, we demonstrated that mutations within the Fc{gamma}R binding motif (residues 233–238 in the mAb lower hinge) completely arrested depletion by hIgG1 and hIgG4, supporting a critical role for Fc{gamma}Rs in this model (Fig. 3Go). In contrast to depleting mAbs, these mutants ‘coated’ CD8+ lymphocytes for several days [44]. We repeated this work using wild-type and mutant mouse IgG2b (mIgG2b) mAbs, confirming the importance of Fc{gamma}R interactions and the redundancy of complement in a completely homologous system [44]. Although we were unable to improve the depleting potency of mIgG2b by enabling it to bind the mouse high-affinity Fc{gamma}RI, removing its ability to bind mouse Fc{gamma}RII significantly impaired its ability to deplete cells (Fig. 4Go).



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 2. A comparison of chimaeric CD8 mAbs in vivo. mAb (50 µg) was administered to thymectomized mice via a tail vein in three divided doses (days 1–3). The figure shows the percentage of CD8+ peripheral blood lymphocytes (PBL) remaining at subsequent time points. Each line represents an individual mouse. hIgG1–4, rIgG1 and rIgG2b (including ‘parental’ rIgG2b mAb YTS 169.4) caused potent and rapid depletion. hIgA2 and hIgE were inactive, and rIgG2a was intermediate in potency. From [41]. Copyright 1992. The American Association of Immunologists.

 


View larger version (15K):
[in this window]
[in a new window]
  		FIG. 3. hIgG1 and hIgG4 mutated in the Fc{gamma}R-binding motif are impotent in vivo. Groups of four thymectomized mice were administered the mAbs shown on day 0. Depletion of CD8+ peripheral blood lymphocytes (PBL) was measured on day 14 (shown as mean±S.D.). The data are pooled from three experiments. See also Table 2. From [44]. Copyright 1998. The American Association of Immunologists.

 


View larger version (21K):
[in this window]
[in a new window]
  		FIG. 4. mIgG2b depletes via a Fc{gamma}RII-dependent mechanism. Groups of four thymectomized mice were administered the mAbs shown. Depletion of CD8+ peripheral blood lymphocytes (PBL) was measured on day 14 (shown as mean±S.D.). The data are pooled from two experiments. Removing the C1q-binding motif or enabling binding to mFc{gamma}RI does not alter depleting potency (groups D–L). Removing the mFc{gamma}RII-binding motif reduces depleting potency (groups M–O). From [44]. Copyright 1998. The American Association of Immunologists.

 
The model supported our initial hypothesis that in vivo mAb effector function could not necessarily be predicted from in vitro tests. Consequently, no therapeutic mAb should be ascribed particular in vivo biological activities until it has been administered to at least a few patients. This is illustrated by our data from a small study of an hIgG4 version of CAMPATH-1H [45]. In vitro data suggested that a mAb of this isotype would not deplete but our murine work suggested otherwise. We therefore designed a study in which IgG4 CAMPATH-1H was administered to a small number of subjects with RA. We witnessed significant depletion in all patients, possibly related to circulating mAb levels (Fig. 5Go), and presumably mediated by the weak affinity of this isotype for human Fc{gamma}RI (hIgG4 does not activate complement). It is pertinent to note that a number of therapeutic mAbs are currently being produced with an IgG4 isotype on the assumption that they will not deplete. As a consequence of the above data, however, our own current work focuses on the use of aglycosyl and non-Fc{gamma}R-binding Fc-mutated mAbs when depletion is an undesirable biological activity.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 5. A hIgG4 version of CAMPATH-1H depletes in vivo. Patients were administered 12 mg IgG4-C1H on day 0. Peripheral blood lymphocyte (PBL) counts were monitored during the ensuing 48 h. Twelve milligrams of IgG1-C1H (wild-type CAMPATH-1H) was given on day 2. (A) PBL counts. Blood was analysed before and after administration of both mAbs. Each line represents an individual patient. Note that the x axis is not to scale. (B) Relationship between peak serum IgG4-C1H levels and depletion of circulating lymphocytes measured at 48 h. Each point represents an individual patient (r=0.565, P=0.089). IgG4-C1H causes depletion of PBL with a possible relationship to peak circulating mAb levels. From [45].

 


View this table:
[in this window]
[in a new window]
  		TABLE 2. Summary of unwanted consequences of T-cell-directed mAb therapy and potential solutions

 


   Immunogenicity
Top
 Abstract
 Introduction
 The fate of the...
 Immunogenicity
The ‘first-dose’...
 Summary
 References
 
Before the description of antibody humanization [20], all therapeutic mAbs comprised a rodent protein sequence. Thus, they were immunogenic and their administration inevitably resulted in an anti-globulin (anti-Ig) response, directed at either the C-region (anti-isotype response) or the V-region (anti-idiotype response) [46]. This was not usually a problem on first administration, but subsequent courses were less effective because of mAb neutralization and accelerated clearance, and there were occasional reports of anaphylactoid reactions [47]. The genetic engineering process of mAb humanization is best depicted as the complementarity-determining regions (CDRs) of a rodent mAb being ‘grafted’ on to a human mAb framework, in place of the native human CDRs [48]. Because the CDRs dictate binding characteristics, the humanized mAb should retain the target specificity of the rodent parent and, in general, this is the case [20, 49]. With regard to immunogenicity, however, the CDRs may still be seen as foreign by the recipient's immune system, although the hypermutated CDRs of ordinary circulating human polyclonal antibodies should appear very similar. There are also a number of genetic polymorphisms which give rise to human Ig allotypic variants; these could provide a further immune stimulus depending on the allotype of the patient and that of the humanized therapeutic mAb [49].

Overall, humanization was expected to provide good camouflage, although the relatively large quantities of mAb used therapeutically might have tipped the balance in favour of recognition by the recipient's immune system. Thus, when the first humanized therapeutic mAb, CAMPATH-1H, was administered to a cohort of eight RA patients, we had no idea whether it would prove immunogenic [21]. In fact, after a single course of treatment we could not measure any anti-Ig activity in the patients’ serum. This compared favourably with previous experience using the rat parent mAb CAMPATH-1G. The latter had not been used in RA but, when administered to transplant recipients, 11 of 14 became sensitized after a single course of treatment [50]. A second course of therapy with CAMPATH-1H, however, provoked an anti-Ig response in three of four RA patients [21]. These recipients were allotype-matched with the therapeutic mAb and the anti-Ig response was focused entirely on the CAMPATH-1H idiotype. In subsequent trials, CAMPATH-1H has been associated with a low incidence of anti-idiotype responses after single courses of therapy [23, 24]. Furthermore, a patient with refractory systemic vasculitis who requires CAMPATH-1H therapy intermittently for disease control, undergoes plasmapheresis before each course of therapy in order to remove a strong, neutralizing anti-idiotype response [4]. Thus, whilst reducing immunogenicity, humanization has not rendered mAbs invisible to patients’ immune systems.

As a result of this work, we decided to investigate the anti-Ig response in more detail, with a view to defining method(s) for rendering therapeutic immunoglobulins immunologically invisible, perhaps by inducing immunological tolerance to them. Previous work from our laboratory had shown that it was possible to render T cells tolerant to foreign proteins by two distinct methods. One method involved the administration of ‘deaggregated’ protein (produced by centrifugation at high speed for several hours) and the other involved administering the foreign protein simultaneously with an anti-CD4 mAb. Using either method, subsequent administration of the foreign protein on its own did not evoke an immune response [51]. When these methods were applied to cell-binding mAbs, however, it was possible to prevent an anti-isotype but not an anti-idiotype response [52]. Two possible explanations were that the immune responses to cell-binding idiotypes were T-cell-independent, or that they were T-cell dependent but in some way special, rendering them relatively refractory to tolerance induction. CD8 mAbs were particularly immunogenic [52], and we therefore analysed the murine anti-Ig response to two rat anti-mouse CD8 mAbs in more detail [53].

First, we demonstrated that the anti-Ig (anti-idiotype and anti-isotype) response was unequivocally a typical T-cell-dependent immune response. Thus, no anti-Ig response was seen after repeated administration of a CD8 mAb to mice that had no CD4+ lymphocytes (as a consequence of thymectomy followed by depleting CD4 mAb treatment) (Fig. 6Go). Furthermore, the anti-Ig response in unmanipulated mice exhibited priming and class-switching, both classical characteristics of a T-cell-dependent immune response [53].



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 6. The antiglobulin response is T-cell dependent. Groups of eight (A and C) or 12 (B and D) mice were unmanipulated (A), CD4+ T-cell-depleted using the depleting rat anti-mouse CD4 mAb YTS 191.1 (B), thymectomized (C) or thymectomized and depleted (D). Following recovery from these procedures, all mice were immunized on two occasions 10 days apart with 100 µg of YTS 169.4 (rat anti-mouse CD8) and with 200 µl of a 1% solution of sheep red blood cells (SRBC) in PBS. Mice were bled 7 days after the second immunization and their immune responses determined by ELISA. The figures show the responses for individual mice and the geometric mean for each group. The left-hand panel shows the total anti-Ig response, the middle panel the anti-isotype response and the right-hand panel the anti-SRBC response. Where six or more mice shared the same value, a numeral in parentheses indicates the number represented. T-cell-deficient mice (D) failed to raise either an anti-Ig or an anti-SRBC response; mice previously receiving YTS 191.1 (B) made a very small anti-isotype response and a greatly reduced anti-lg response, suggesting tolerance to isotype. From [53].

 
These results suggested that, unless cell-binding idiotypes were in some way unique and did not obey the rules derived for other T-cell-dependent immunogens, it should be possible to tolerize the relevant T cells and thereby prevent an anti-idiotype response. In fact, we were able to induce tolerance to the idiotype of both CD8 mAbs. The critical factor was first to induce robust tolerance to the mAb C-region, although the conditions necessary for this differed from one to the other. For one mAb (YTS 105.18, rIgG2a), co-administration of a CD4 mAb was sufficient (Fig. 7Go), whereas for the other (YTS 169.4, rIgG2b) this protocol actually primed for an anti-idiotype response. In this case, C-region tolerance was first induced (via deaggregation or CD4 mAb co-administration) using an irrelevant, non-cell-binding mAb of the same isotype [53]. With both mAbs, once C-region tolerance was induced, even multiple administrations of therapeutic mAb did not provoke an anti-idiotype response (Fig. 7Go).



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 7. YTS 177.9 (rat IgG2a anti-mouse CD4) tolerizes to YTS 105.18 (rat IgG2a anti-mouse CD8). Groups of mice were administered PBS (A) or YTS 177.9 (B, C) on days -2 (1 mg i.v.), -1 and 0 (1 mg i.p.). Group C received YTS 105.18 (1 mg i.p.) concurrently on day 0. Mice were immunized with YTS 105.18 (100 µg i.p.) on four occasions subsequently (days 42, 52, 80 and 87). The figure shows anti-Ig responses against YTS 105.18 determined by ELISA on day 94. Values are shown for individual mice, and the geometric mean is shown for each group. Mice that received YTS 177.9 and YTS 105.18 made an insignificant response (group C). In addition, mice in group B were tolerant to the rat IgG2a isotype, accounting for their reduced response (not shown). From [53].

 
Although complex, these experiments demonstrated for the first time that tolerance could be induced to cell-binding idiotypes and, consequently, that therapeutic mAbs could be rendered non-immunogenic. The two mAbs investigated differed in their immunogenicity, which presumably related to the number of T-cell epitopes in their heavy and light chains. It is important to realize that all of our experiments were performed in CBA/ca mice, and C-region tolerance may not have been sufficient to prevent anti-Ig responses in mouse strains with a different major histocompatibility complex (MHC) haplotype, in which the CD8 V-regions may also have contributed T-cell epitopes. Translating to the human situation, the administration of a chimaeric mAb (murine V-region, human C-region) to an Ig allotype-matched patient is equivalent to administering a rat mAb to a mouse that has been tolerized to the xenogeneic (rat) C-region. Thus, it should not be surprising that some recipients of a chimaeric CD4 mAb mounted an anti-V-region immune response, whereas others did not [29]. Similarly, some (Ig allotype-matched) recipients of CAMPATH-1H developed an anti-idiotype response [21]. In the latter scenario, the V-region CDRs themselves presumably provided T-cell epitopes (as opposed to V-region framework residues, which remain foreign in chimaeric mAbs).

Because the human MHC is so polymorphic, in order to routinely avoid an anti-Ig response, T-cell epitopes in a therapeutic mAb would need to be silenced in a patient-specific manner. We suggested and tested a possible protocol to achieve this which involved the creation of non-cell-binding variants of a therapeutic mAb. These were designed to contain the same linear T-cell epitopes as the therapeutic mAb but to be less immunogenic by virtue of their inability to bind cells. Our hypothesis proposed that if tolerance to the variants could be induced, this should spread to the parent mAb. If the therapeutic mAb is denoted H2L2 (H, heavy chain; L, light chain) then the two variants are H2K2 and G2L2, where G and K are irrelevant heavy and light partner chains which destroy the mAb’s ability to bind cells (Fig. 8, bottom panel). Using cell fusion and limiting dilution cloning techniques, we produced H2K2 and G2L2 variants of our more immunogenic CD8 mAb. These were indeed tolerogenic, allowing subsequent administration of the cell-binding H2L2 without an anti-Ig response [53], although the general applicability of this concept has yet to be tested with other mAbs or in other mouse strains.

The above work again illustrates the importance of basic laboratory science in suggesting novel approaches for clinical investigators. Figure 8Go summarizes the main aspects of our work pertaining to the anti-globulin response.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 8. Summary of our experiments on the anti-globulin response to therapeutic mAbs.

 


   The ‘first-dose’ reaction
Top
 Abstract
 Introduction
 The fate of the...
 Immunogenicity
 The ‘first-dose’...
Summary
 References
 
The ‘first-dose’ reaction is seen on initial administration of a number of therapeutic mAbs to patients. It was first described during therapy of transplant recipients with CD3-specific mAb OKT3. In this situation it appeared to be secondary to the release of cytokines from T cells, following their partial activation [54, 55]. The classic syndrome occurred approximately 2 h into treatment and comprised chills, fever, nausea, headache and, in severe cases, bronchospasm/chest tightness and hypotension. We witnessed a very similar reaction when treating RA patients with CAMPATH-1H. When CAMPATH-1H was subsequently used to treat multiple sclerosis (MS), however, a more sinister manifestation occurred. In addition to the usual features of the syndrome, MS patients experienced a transient exacerbation of their neurological symptoms and a temporary recrudescence of former features [56]. This usually lasted for around 12 h and patients subsequently recovered fully. Although preventable by prior administration of methylprednisolone, we decided to investigate these reactions further, in the expectation that an understanding of their pathogenesis would allow a more targeted intervention.

First, concurrent laboratory parameters were documented in MS patients experiencing first-dose reactions. The onset of neurological symptoms at 2 h coincided with peak circulating levels of the cytokines tumour necrosis factor-{alpha} (TNF-{alpha}) and interferon-{gamma} (IFN-{gamma}). Shortly afterwards there was a rise in serum interleukin-6 (IL-6) which peaked at 6 h (Fig. 9Go). Fever alone did not reproduce the symptoms in a patient whose temperature was artificially elevated [56]. Visual evoked potentials deteriorated during neurological symptoms and corrected again afterwards, demonstrating a transient slowing of nerve conduction. None of the above features was present when the reaction was prevented by methylprednisolone prophylaxis, providing circumstantial evidence that cytokines were responsible for the neurological symptoms. The transient nature of both symptoms and electrophysiological changes was most in keeping with direct impairment of nerve conduction, which was consistent with published data showing such an effect of TNF-{alpha} [57]. Similar cytokine elevations were seen in CAMPATH-1H-treated RA patients [39].



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 9. The relationship between neurological exacerbations, serum cytokine concentrations and pyrexia in four MS patients receiving their first infusion of CAMPATH-1H. Arrows indicate the onset of neurological exacerbations. See text for details. From [56].

 
It was now necessary to establish an in vitro assay to use as a model for the in vivo reaction. We found that simply incubating heparinized whole blood with therapeutic mAbs resulted in mAb- and time-dependent release of the same three cytokines (TNF-{alpha}, IFN-{gamma} and IL-6) into the plasma [58]. The mAb factors determining cytokine release were specificity and istotype: CD52 mAbs of hIgG1 and rIgG2b isotype were good stimuli, whereas the hIgG4 isotype was less effective and rIgM ineffective. mAbs against other targets, such as CD2 and CD4, did not induce cytokine release, regardless of isotype (Fig. 10Go). Both the time course of cytokine release in vitro (levels becoming measurable at 2 h and peaking at 4–6 h) and the hierarchy amongst different therapeutic mAbs suggested that our in vitro assay reflected the first-dose reaction in vivo. Thus, rIgG2b and hIgG1 CD52 mAbs precipitated moderately severe reactions in vivo, and we witnessed lesser reactions with the hIgG4 version of that mAb [45] and variable, usually minor, reactions with a hIgG1 CD4 mAb [11]. Table 1Go summarizes these findings [59].



View larger version (10K):
[in this window]
[in a new window]
  		FIG. 10. The whole blood assay for studying the first-dose reaction. Aliquots of whole blood were incubated for up to 6 h with rat and humanized mAbs at a final concentration of 10 µg/ml. At selected time points the plasma and cells were separated by centrifugation, and the plasma levels of (a) TNF-{alpha} and (b) IFN-{gamma} were determined by ELISA and immunoradiometric assay, respectively. Squares, CAMPATH-1G (rIgG2b anti-CD52); diamonds, CAMPATH-1H (hIgG1 anti CD52); crossed squares, CAMPATH-1M (rIgM anti-CD52); circles, hIgG4 anti-CD52; triangles, hIgG1 anti-CD4. Six hour samples from CAMPATH-1G and CAMPATH-1H incubations were also positive by bioassay for IL-6. (c) Aliquots of whole blood were incubated with 10 µg/ml of either intact CAMPATH-1H or a pepsin-digested F(ab')2 fragment for 4 h. Plasma TNF-{alpha} was measured by ELISA after separation of the cells and plasma by centrifugation. There is a hierarchy amongst the mAbs for cytokine release (a, b) which is Fc-dependent (c). From [58].

 

View this table:
[in this window]
[in a new window]
  		TABLE 1. Comparison of in vitro cytokine release with extent of first-dose reactions

 
Our assay was now used to dissect the biology of cytokine release. The relationship to mAb isotype seen with CD52 mAbs (which also reflected their depleting capacity) suggested the involvement of Fc{gamma}R. Unlike OKT3, however, most CD52 mAbs were known not to activate T-cells, which made these an unlikely cytokine source. The alternative possibility was that Fc{gamma} receptor-bearing cells, having bound to mAb-coated T-cells, were themselves the cytokine source. The dependency of the in vitro reaction on Fc{gamma}R ligation was confirmed by showing that a F(ab')2 derivative of CAMPATH-1H (hIgG1 anti-CD52) did not cause cytokine release (Fig. 10Go) [58]. The reaction was also shown to be a property of purified lymphocytes but not monocytes (nylon wool non-adherent vs adherent peripheral blood mononuclear cells) [58]. Additionally, cytokine release could be prevented by mAbs to CD16 (Fc{gamma}RIII) and CD11a (LFA-1) but not CD32 (Fc{gamma}RII) or CD11b (Mac-1).

Thus, cytokine release by CD52 mAbs occurred in pure lymphocyte preparations and required the interaction of mAb with CD52, CD16 and CD11a. The precise cytokine source was identified using a number of assays in which fresh cells and paraformaldehyde-fixed cells were coincubated. Purified lymphocytes were incubated with CAMPATH-1H, washed, and then fixed with paraformaldehyde to render them inert. Subsequently, addition of fresh lymphocytes resulted in cytokine release which was inhibitable by anti-CD16 (Fig. 11Go). This experiment was repeated using fluorescently sorted CD3+ T-lymphocytes and CD3-CD16+ NK cells, showing conclusively that the latter were the cytokine source [58]. Dependency on CD11a (see above) suggested the involvement of leucocyte function-associated antigen-1 in the T cell–NK cell interaction. Thus, in contrast to the situation with OKT3, the cytokines responsible for the first-dose reaction to CD52 mAbs are released from ‘accessory’ cells rather than the targeted lymphocytes themselves. This concept is summarized in Fig. 12Go.



View larger version (14K):
[in this window]
[in a new window]
  		FIG. 11. Dissecting the first-dose reaction in vitro. Left-hand columns: 1x106 fresh or formaldehyde-fixed lymphocytes were incubated alone (-) or with 10 µg/ml CAMPATH-1H, and TNF-{alpha} release was measured by L1929 bioassay. CAMPATH-1H provoked TNF-{alpha} release from fresh but not fixed lymphocytes. Right-hand columns: 1x106 fresh lymphocytes were incubated with (+) or without (-) a Fab fragment of a neutralizing anti-CD16 mAb. Fixed lymphocytes (1x106) which had been incubated alone (-) or with 10 µg/ml CAMPATH-1H were then added and TNF-{alpha} release was measured as above. Fresh lymphocytes released TNF-{alpha} following incubation with fixed, CAMPATH-1H-treated lymphocytes. TNF-{alpha} release was prevented if the fresh cells had been incubated previously with anti-CD16. From [58].

 


View larger version (33K):
[in this window]
[in a new window]
  		FIG. 12. Mechanisms of first-dose reactions with (1) OKT3 and (2) Campath-1 mAbs. With the former, cytokines are predominantly released from targeted and activated T cells. Campath-1 mAbs do not usually activate their target cells, but result in activation of a cross-linking NK cell. It is this cell which then releases cytokines. In either situation, preventing mAb from binding Fc{gamma}R will avert a reaction. Note that Campath-1 mAbs are likely to interact with Fc{gamma}R-bearing cells other than NK cells, but our model suggests that it is only the latter cell type that releases cytokines as a result of the interaction.

 
As with the preceding examples, this work suggested a solution to a particular complication of mAb therapy: mAbs which do not interact with Fc{gamma}Rs should not trigger first-dose reactions. It has already been demonstrated that an aglycosyl variant of a humanized CD3 mAb has reduced affinity for Fc{gamma}Rs [60] and provokes just a minor first-dose reaction in transplant recipients [61]. We have also produced an Fc-mutated version of a human CD4 mAb, modelled on our non-depleting murine CD8 mAbs [44], which also has reduced affinity for Fc{gamma}Rs. A CD52 mAb with the same mutated Fc region has impaired ability to release cytokines in the whole blood assay [58]. For immunotherapy of autoimmunity and transplantation, the combination of reduced first-dose reactions and negligible lymphocyte depletion (see above) provide a double advantage to mAbs with these and similar C-regions.


   Summary
Top
 Abstract
 Introduction
 The fate of the...
 Immunogenicity
 The ‘first-dose’...
 Summary
References
 
Small, open clinical studies of novel biological therapies cannot prove therapeutic efficacy, but useful and important information can be gathered from such pilot experiments. In particular, biological effects secondary to interactions between the mAb Fc region and Fc-R-bearing accessory cells were unpredictable in early clinical work. In this essay, I have tried to illustrate that observations derived from such open studies can be used to design relevant in vitro and in vivo models from which important rules can be generated. These rules can then be applied to the engineering and design of second- and third-generation mAbs which subsequently enter the clinic in an iterative process, constantly building on previous experience. Such a process is critical if we are to exploit mAbs to their full potential. Failure to adopt such an approach may result in unforeseen adverse events, premature termination of phase II trials, and a general loss of confidence in biological therapies. Table 2Go summarizes our findings and suggests solutions to common unwanted consequences of T cell-directed mAb therapy.


   Acknowledgments
 
This work could not have taken place without the inspiration and support of my mentors: Professor Herman Waldmann, whose vision and enthusiasm originally attracted me to the field of therapeutic immunology; and Dr Brian Hazleman, whose energy and encouragement stimulated me on to a rheumatological career path. I am also indebted to the Medical Research Council, whose financial support by means of a Clinical Training Fellowship and Clinician Scientist Fellowship enabled me to divide my time between bench and bedside for seven years, and to my fellow laboratory workers, in particular Drs Steve Cobbold, Mark Wing, Thibeau Moreau and Judith Greenwood; and to Dr Geoff Hale and the staff of the Therapeutic Antibody Centre, who produced therapeutic-grade monoclonal antibodies for the studies described in this article.


   References
Top
 Abstract
 Introduction
 The fate of the...
 Immunogenicity
 The ‘first-dose’...
 Summary
 References
 

  1. Cobbold SP, Qin S, Leong LYW, Martin G, Waldmann H. Reprogramming the immune system for peripheral tolerance with CD4 and CD8 monoclonal antibodies. Immunol Rev 1992;129:165–201.[Web of Science][Medline]
  2. Isaacs JD, Waldmann H. Regaining self-control. Regulation and immunotherapy. Br J Rheumatol 1998; 37:926–9.[Free Full Text]
  3. Mathieson PW, Cobbold SP, Hale G et al. Monoclonal-antibody therapy in systemic vasculitis. N Engl J Med 1990; 323:250–4.[Web of Science][Medline]
  4. Lockwood CM, Thiru S, Isaacs JD, Hale G, Waldmann H. Long-term remission of intractable systemic vasculitis with monoclonal antibody therapy. Lancet 1993;341: 1620–2.[Web of Science][Medline]
  5. Isaacs JD, Dick AD, Haynes R et al. Monoclonal antibody therapy of chronic intraocular inflammation using CAMPATH-1H. Br J Ophthalmol 1995;79:1054–5.[Free Full Text]
  6. Newman DK, Isaacs JD, Watson PG, Meyer PAR, Hale G, Waldmann H. Prevention of immune mediated corneal graft destruction with the anti-lymphocyte monoclonal antibody CAMPATH-1H. Eye 1995;9:564–9.
  7. Van der Lubbe PA, Dijkmans BAC, Markusse HM, Nassander U, Breedveld FC. A randomized, double-blind, placebo-controlled study of CD4 monoclonal antibody therapy in early rheumatoid arthritis. Arthritis Rheum 1995;38:1097–106.[Medline]
  8. Moreland LW, Pratt PW, Mayes MD et al. Double-blind, placebo-controlled multicenter trial using chimeric monoclonal anti-CD4 antibody, cM-T412 in rheumatoid arthritis patients receiving methotrexate. Arthritis Rheum 1995;38:1581–8.[Medline]
  9. Levy R, Weisman M, Weisenhutter C et al. Results of a placebo-controlled multicenter trial using a primatized, non-depleting, anti-CD4 monoclonal antibody in the treatment of rheumatoid arthritis. Arthritis Rheum 1996; 39:S122.
  10. Isaacs JD. Monoclonal antibodies in rheumatology. In: Ritter M, Ladyman P, eds. Monoclonal antibodies: Production, engineering and clinical application. Cambridge: Cambridge University Press, 1985:403–28.
  11. Isaacs JD, Burrows N, Wing M et al. Humanised anti-CD4 monoclonal antibody therapy of autoimmune and inflammatory disease. Clin Exp Immunol 1997;110: 158–66.[Web of Science][Medline]
  12. Morgan AW, Hale G, Rebello P et al. Combination therapy with a TNF antagonist and a CD4 monoclonal antibody in rheumatoid arthritis: A pilot study. Arthritis Rheum 1997;40:S81.
  13. Morgan AW, Hale G, Waldmann H, Emery P, Isaacs J. Prolonged combination immunotherapy of RA using CD4 and TNFa blockade—a pilot study. Arthritis Rheum 1998;41:S138.
  14. Qin S, Cobbold S, Benjamin R, Waldmann H. Induction of classical transplantation tolerance in the adult. J Exp Med 1989;169:779–94.[Abstract/Free Full Text]
  15. Qin S, Wise M, Cobbold S et al. Induction of tolerance in peripheral T cells with monoclonal antibodies. Eur J Immunol 1990;20:2737–45.[Web of Science][Medline]
  16. Hale G, Bright S, Chumbley G et al. Removal of T-cells from bone marrow for transplantation: a monoclonal anti-lymphocyte antibody that fixes human complement. Blood 1983;62:873–82.[Abstract/Free Full Text]
  17. Hale G, Xia M-Q, Tighe HP, Dyer MJS, Waldmann H. The Campath-1 antigen (Cdw52). Tissue antigens 1990;35:118–27.[Web of Science][Medline]
  18. Hale G, Clark M, Waldmann H. Therapeutic potential of rat monoclonal antibodies: Isotype specificity of antibody-dependent cell-mediated cytotoxicity with human lymphocytes. J Immunol 1984;134:3056–61.[Abstract]
  19. Dyer MJS, Hale G, Hayhoe FGJ, Waldmann H. Effects of CAMPATH-1 antibodies in vivo in patients with lymphoid malignancies: influence of antibody isotype. Blood 1989;73:1431–9.[Abstract/Free Full Text]
  20. Riechmann L, Clark M, Waldmann H, Winter G. Reshaping human antibodies for therapy. Nature 1988; 332:323–7.[Medline]
  21. Isaacs JD, Watts RA, Hazleman BL et al. Humanised monoclonal antibody therapy for rheumatoid arthritis. Lancet 1992;340:748–52.[Web of Science][Medline]
  22. Weinblatt ME, Maddison PJ, Bulpitt KJ et al. CAMPATH-1H, a humanized monoclonal antibody in refractory rheumatoid arthritis. An intravenous dose-escalation study. Arthritis Rheum 1995;38:1589–94.[Medline]
  23. Matteson EL, Yocum DE, St Clair EW et al. Treatment of active refractory rheumatoid arthritis with humanized monoclonal antibody CAMPATH-1H administered by daily sub-cutaneous injection. Arthritis Rheum 1995; 38:1187–93.[Medline]
  24. Isaacs JD, Manna VK, Rapson N et al. CAMPATH-1H in rheumatoid arthritis—an iv dose-ranging study. Br J Rheumatol 1996;35:231–40.[Abstract/Free Full Text]
  25. Watts RA, Isaacs JD, Hale G, Hazleman BL, Waldmann H. Peripheral blood lymphocyte subsets after CAMPATH-1H therapy in rheumatoid arthritis—a 3 year follow-up. Arthritis Rheum 1994;37:S338.
  26. Mackall CL, Hakim FT, Gress RE. T-cell regeneration: all repertoires are not created equal. Immunol Today 1997;18:245–51.[Web of Science][Medline]
  27. Choy EHS, Chikanza IC, Kingsley GH, Corrigall V, Panayi GS. Treatment of rheumatoid arthritis with single dose or weekly pulses of chimaeric anti-CD4 monoclonal antibody. Scand J Immunol 1992;36:291–8.[Medline]
  28. Moreland LW, Bucy RP, Tilden A et al. Use of a chimeric monoclonal anti-CD4 antibody in patients with refractory rheumatoid arthritis. Arthritis Rheum 1993;36: 307–18.[Medline]
  29. Van der Lubbe PA, Reiter C, Breedveld FC et al. Chimeric CD4 monoclonal antibody cM-T412 as a therapeutic approach to rheumatoid arthritis. Arthritis Rheum 1993;36:1375–9.[Medline]
  30. Moreland LW, Pratt PW, Bucy RP, Jackson BS, Feldman JW, Koopman WJ. Treatment of refractory rheumatoid arthritis with a chimeric anti-CD4 monoclonal antibody. Long-term followup of CD4+ T cell counts. Arthritis Rheum 1994;37:834–8.[Web of Science][Medline]
  31. Ruderman EM, Weinblatt ME, Thurmond LM, Pinkus GS, Gravallese EM. Synovial tissue response to treatment with CAMPATH-1H. Arthritis Rheum 1995;38:254–8.[Medline]
  32. Isaacs JD, Greer S, Symmons D et al. Survival of patients after lymphocytotoxic monoclonal antibody therapy. Arthritis Rheum 1998;41:S141.
  33. Lockwood CM, Thiru S, Stewart S et al. Treatment of refractory Wegener's granulomatosis with humanised monoclonal antibodies. Q J Med 1996;89:903–12.[Abstract]
  34. Moreau T, Thorpe J, Miller D et al. Preliminary evidence from magnetic resonance imaging for reduction in disease activity after lymphocyte depletion in multiple sclerosis. Lancet 1994;344:298–301.[Web of Science][Medline]
  35. Isaacs J, Dick A. Short term immunosuppressive therapy and long term immunoregulation: promises and problems. Br J Ophthalmol 1996;80:1035–6.[Free Full Text]
  36. Isaacs JD, Hazleman BL, Chakravarthy K, Grant JW, Hale G, Waldmann H. Monoclonal antibody therapy of diffuse cutaneous scleroderma with CAMPATH-1H J Rheumatol 1996;23:1103–6.
  37. Alters SE, Sakai K, Steinman L, Oi V. Mechanisms of anti-CD4-mediated depletion and immunotherapy: A study using a set of chimeric anti-CD4 antibodies. J Immunol 1990;144:4587–92.[Abstract]
  38. Bindon CI, Hale G, Waldmann H. Importance of antigen specificity for complement-mediated lysis by monoclonal antibodies. Eur J Immunol 1988;18:1507–14.[Web of Science][Medline]
  39. Michaelsen TE, Garred P, Aase A. Human IgG subclass pattern of inducing complement-mediated cytolysis depends on antigen concentration and to a lesser extent on epitope patchiness, antibody affinity and complement concentration. Eur J Immunol 1991;21:11–16.[Web of Science][Medline]
  40. Greenwood J, Clark M, Waldmann H. Structural motifs involved in human IgG antibody effector functions. Eur J Immunol 1993;23:1098–104.[Web of Science][Medline]
  41. Isaacs JD, Clark MR, Greenwood J, Waldmann H. Therapy with monoclonal antibodies. An in vivo model for the assessment of therapeutic potential. J Immunol 1992;148:3062–71.[Abstract]
  42. Cobbold SP, Jayasuriya A, Nash A, Prospero T, Waldmann H. Therapy with monoclonal antibodies by elimination of T-cell subsets in vivo. Nature 1984; 312:548–51.[Medline]
  43. Tao M-H, Morrison SL. Studies of aglycosylated chimeric mouse–human IgG: role of carbohydrate in the structure and effector functions mediated by the human IgG constant regions. J Immunol 1989;145:2595–601.[Abstract]
  44. Isaacs J, Greenwood J, Waldmann H. Therapy with monoclonal antibodies II: The contribution of Fc{gamma} receptor binding and the influence of CH1 and CH3 domains on in vivo effector function. J Immunol 1998;161:3862–9.[Abstract/Free Full Text]
  45. Isaacs JD, Wing MG, Greenwood JD, Hazleman BL, Hale G, Waldmann H. A therapeutic human IgG4 monoclonal antibody that depletes target cells in humans. Clin Exp Immunol 1996;106:427–33.[Medline]
  46. Isaacs JD. The antiglobulin response to therapeutic antibodies. Semin Immunol 1990;2:449–56.[Medline]
  47. Reiter C, Kakarand B, Rieber EP, Schattenkirchner M, Riethmuller G, Kruger K. Treatment of rheumatoid arthritis with monoclonal anti-CD4 antibody M-T151. Clinical results and immunopharmacologic effects in an open study, including repeated administration. Arthritis Rheum 1991;34:525–36.[Web of Science][Medline]
  48. Isaacs JD. Immunomodulation with monoclonal antibodies. In: Champion R, Pye R, eds. Recent advances in dermatology, vol. 9. Edinburgh: Churchill Livingstone, 1991.
  49. Gorman SD, Clark MR. Humanisation of monoclonal antibodies for therapy. Semin Immunol 1990;2:457–66.[Medline]
  50. Friend PJ, Waldmann H, Hale G et al. Reversal of allograft rejection using the monoclonal antibody CAMPATH-1G. Transplant Proc 1991;23:2253–4.[Medline]
  51. Benjamin RJ, Waldmann H. Induction of tolerance by monoclonal antibody therapy. Nature 1986;320:449–51.[Medline]
  52. Benjamin RJ, Cobbold SP, Clark MR, Waldmann H. Tolerance to rate monoclonal antibodies: Implications for serotherapy. J Exp Med 1986;163:1539–52.[Abstract/Free Full Text]
  53. Isaacs JD, Waldmann H. Helplessness as a strategy for avoiding antiglobulin responses to therapeutic antibodies. Ther Immunol 1994;1:303–12.[Medline]
  54. Chatenoud L, Bach J-F. OKT3 in allogeneic transplantation: clinical efficacy, mode of action, and side effects. In: Burlingham WJ, ed. A critical analysis of monoclonal antibody therapy in transplantation. Boca Raton: CRC Press, 1992.
  55. Parlevliet KJ, ten Berge IJM, Yong S-L, Surachno J, Wilmink JM, Schellekens PA. In vivo effects of IgA and IgG2a anti-CD3 isotype switch variants. J Clin Invest 1994;93:2519–25.
  56. Moreau T, Coles A, Wing M et al. Transient increase in symptoms associated with cytokine release in patients with multiple sclerosis. Brain 1996;119:225–37.[Abstract/Free Full Text]
  57. Brosnan CF, Litwak MS, Schroeder CE, Selmaj K, Raine CS, Arezzo JC. Preliminary studies of cytokine-induced functional effects on the visual pathways in the rabbit. J Neuroimmunol 1989;25:227–39.[Web of Science][Medline]
  58. Wing MG, Moreau T, Greenwood J et al. Mechanism of first-dose cytokine release syndrome by CAMPATH-1H: Involvement of CD16 (Fc{gamma}RIII) and CD11a/CD18 (LFA-1) on NK cells. J Clin Invest 1996;98:2819–26.[Web of Science][Medline]
  59. Wing MG, Isaacs JD, Waldmann H, Compston DAS, Hale G. Ex-vivo blood cultures for predicting cytokine-release syndrome: dependence on target antigen and antibody isotype. Ther Immunol 1995;2:183–90.[Medline]
  60. Bolt S, Routledge E, Lloyd I et al. The generation of a humanized, non-mitogenic CD3 monoclonal antibody which retains immunosuppressive properties. Eur J Immunol 1993;23:403–11.[Medline]
  61. Friend PJ, Hale G, Chatenoud L et al. Phase I study of an engineered aglycosylated humanized CD3 antibody in renal transplant rejection. Transplantation 1999;68: 1632–7.[Web of Science][Medline]
Submitted 18 January 2001; Accepted 19 January 2001


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Rheumatology (Oxford)Home page
J. D. Isaacs
Therapeutic T-cell manipulation in rheumatoid arthritis: past, present and future
Rheumatology, October 1, 2008; 47(10): 1461 - 1468.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (9)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Isaacs, J. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Isaacs, J. D.
Related Collections
Right arrow Rheumatoid Arthritis
Right arrow Systemic Lupus Erythematosus and Autoimmunity
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?