In vivo induction of functional Fc gammaRI (CD64) on neutrophils and modulation of blood cytokine mRNA levels in cancer patients treated with G-CSF (rMetHuG-CSF).

Neutrophils from 13 children who received G-CSF for the collection of peripheral blood progenitors while they were in haematological steady state were studied at various times after G-CSF injection for Fc gammaR expression (Fc gammaRI or CD64, Fc gammaRII or CD32, and Fc gammaRIII or CD16) and for their ability to exert antibody-dependent cell cytotoxicity (ADCC) through Fc gammaRI. Changes in IFNgamma, IL8, IL10, MCP1 and TNF alpha mRNA levels in peripheral blood cells were also studied 4 h and 24 h after the first G-CSF injection. Fc gammaRI expression increased strongly after 24 h and then remained at the same level throughout treatment. In contrast, Fc gammaRIII expression sharply decreased at day 1 and diminished even further thereafter. No change in Fc gammaRII was observed. ADCC exerted by neutrophils through Fc gammaRI started to increase after 24 h with the peak level at day 5. Cytokine mRNA analyses indicated a reproducible and strong increase of IL8 mRNA (11/13 children) after 24 h, whereas the changes in the mRNA levels of the other cytokines tested were more heterogenous (TFNgamma: three; IL10: six; MCP1: five: TNF alpha: four, of the 13 children). Therefore this study opens the way to an optimized therapeutic schedule for the combined use of G-CSF and monoclonal antibodies in adjuvant immuno-intervention.

Generation and characterization of a mouse single-chain antibody fragment specific for disialoganglioside (GD2).

Anti-disialoganglioside (GD2) monoclonal antibodies (MAbs) have been used in vivo for immunolocalization and in phase I and II trials to target disseminated neuroblastoma, the most common extracranial solid tumor in children. However, the efficacy of these first-generation MAbs is likely to be improved by using engineered anti-GD2 antibodies. The generation of single-chain antibody fragments (scFv) could be very helpful as these molecules can be further modified to produce recombinant molecules with pre-defined properties such as immunotoxins, chimeric, or bispecific antibodies. Thus, a scFv directed against GD2 (scFv 7A4) was cloned, sequenced, and expressed. Its binding properties were characterized and compared to that of the parental MAb 7A4. Nucleotide sequence analysis of the scFv 7A4 indicated that its VH region belongs to the V region IIID subgroup and the V kappa to the V region II subgroup. The scFv 7A4 bound to GD2+ neuroblastoma cell lines but not to GD2- cell lines or to GD2- cells isolated from peripheral blood. ELISA and thin-layer chromatography (TLC) indicated that it retained the anti-GD2 specificity, and exhibited a slight cross-reaction with GD3 as the parental MAb. This scFv makes it possible to develop new useful reagents through genetic engineering for adjuvant tumor therapy.

Fc receptors as targets for immunotherapy.

Human membrane and soluble Fc epsilon receptors (Fc epsilon RI, Fc epsilon RII/CD23) and Fc gamma receptors (Fc gamma RI/CD64, Fc gamma RII/CD32, Fc gamma RIII/CD16) have been implicated in a number of diseases. Their functional roles such as capture and clearance of immune complexes, antibody-dependent cell cytotoxicity, or cytokine or inflammatory mediator release, make them potential targets for immuno-intervention. In the present review, we will describe how membrane and soluble human Fc epsilon R and Fc gamma R have been already used as targets/tools for immuno-interventions by using monoclonal and bispecific engineered antibodies. Some therapeutic uses of these molecules both in cancer, infectious, and auto-immune diseases are presented.

In vitro killing of neuroblastoma cells by neutrophils derived from granulocyte colony-stimulating factor-treated cancer patients using an anti-disialoganglioside/anti-Fc gamma RI bispecific antibody.

Neutrophils isolated from cancer patients treated with granulocyte colony-stimulating factor (G-CSF) express high levels of Fc gamma RI. They exhibited an efficient killing of GD2+ neuroblastoma cells in the presence of an antidisialoganglioside (GD2) mouse monoclonal antibody (MoAb; 7A4, IgG3 kappa). However, this cytotoxicity was totally blocked by human monomeric IgG. In contrast, a bispecific antibody (7A4 bis 22/MDX-260), prepared by chemically linking an F(ab') fragment of 7A4 with an F(ab') fragment of an anti-Fc gamma RI MoAb, 22, which binds outside the Fc binding domain, triggered antibody-dependent cell cytotoxicity, even when neutrophils were preincubated with human monomeric IgG. F(ab')2 22 MoAb abrogated the MDX-260 killing without affecting that of 7A4. The 3G8 MoAb, directed against the Fc gamma RIII binding site, did not inhibit the cytotoxicity induced by either antibody. Thus, these results indicate that G-CSF-activated neutrophils exert their cytotoxic effect against neuroblastoma cells through Fc gamma RI and not Fc gamma RIII, and that the saturation of the high affinity Fc gamma RI by monomeric IgG can be overcome by the use of bispecific antibodies binding epitopes outside the IgG Fc gamma RI binding site. A combined administration of such bispecific antibodies and G-CSF may be, therefore, an efficient therapeutic approach to trigger tumor lysis by cytotoxic neutrophils in vivo.

In vivo targeting of human neuroblastoma xenograft by anti-GD2/anti-Fc gamma RI (CD64) bispecific antibody.

Antidisialoganglioside (GD2) monoclonal antibodies can target in vitro and in vivo neuroblastoma cells. However, their in vivo use is limited by the presence of high levels of circulating IgG which hamper the recruitment of effector cells through the high affinity Fc gamma RI (CD64). A bispecific Fab' x Fab' antiGD2/antiFc gamma RI antibody (7A4 bis 22), which binds outside the IgG binding site of Fc gamma RI, was therefore developed. This antibody binds both human GD2+ neuroblastoma and Fc gamma RI+ activated macrophages in vitro. It can localise a GD2 positive neuroblastoma xenografted on Nu/Nu mice. Scintigraphy tumour/muscle ratios showed that targeting with this antibody has an excellent selectivity for the tumour over normal tissues. Furthermore, although its whole body clearance is more rapid than that of the 7A4 parental antibody over the first 48 h, its selective tumour uptake is similar, as shown by immunoscintigraphy imaging. Thus, such a bispecific antibody may represent an efficient tool for in vivo therapy of neuroblastoma through its ability to recruit Fc gamma RI+ effector cells even in presence of circulating IgG and to bind concomitantly GD2+ tumour cells.