Friedhelm Helling

Phase 1 trial of immunological adjuvant QS-21 with a GM2 ganglioside-keyhole limpet haemocyanin conjugate vaccine in patients with malignant melanoma

Increasing doses of saponin fraction QS-21 were administered as immunological adjuvant in a Phase 1 trial with a constant dose of the melanoma ganglioside GM2 covalently attached to keyhole limpet haemocyanin (KLH). Twenty-eight patients with AJCC Stage III or IV melanoma who were free from disease after surgery were treated with six vaccinations administered subcutaneously over a 5-month period. Local and systemic reactions were QS-21 dose-related. Doses of < or = 100 micrograms induced mild local tenderness and inflammation at vaccination sites lasting 2-4 days and occasional brief low-grade fever and malaise, but no significant incapacitation. The 200 micrograms dose induced low-grade fever and malaise after 30% of vaccinations and local reactions as large as 20 cm in diameter were seen in all patients, resulting in discomfort with usage of the injected extremity for 5-10 days. The titres of IgM and IgG antibodies against GM2, and IgG antibodies against KLH, were highest at the 100 and 200 micrograms QS-21 doses. No antibodies against QS-21 were detected. This trial identifies the 100 micrograms dose of QS-21 as the optimal well tolerated dose for induction of antibodies against both the melanoma ganglioside/GM2 and the protein KLH in melanoma patients.

Tumor cell reactivity mediated by IgM antibodies in sera from melanoma patients vaccinated with GM2 ganglioside covalently linked to KLH is increased by IgG antibodies

Abstract Natural IgM antibodies against the melanoma cell-surface ganglioside GM2, and IgM antibodies induced by vaccination with GM2 adherent to bacillus Calmette-Guerin, have been correlated with increased disease-free and overall survival in melanoma patients in previous phase I and II clinical trials. A vaccine containing GM2 covalently attached to keyhole limpet hemocyanin (KLH) plus the immunological adjuvant QS-21 now induces higher-titer, longer-lasting IgM antibodies against GM2 and has recently entered phase III clinical trials. For the first time this new vaccine also induces IgG antibodies against GM2 in the majority of immunized patients. With regard to immunity against bacteria, IgM antibodies have been described to be 1000-fold more effective than IgG antibodies at opsonification, complement-mediated cytotoxicity and protection from bacterial challenge. Though IgG antibodies have the theoretical advantage of being able to mediate antibody-directed cell-mediated cytotoxicity (ADCC), they may inhibit complement mediated IgM effector mechanisms against melanoma cells. Our goal was to confirm the functional characteristics of the anti-GM2 IgM and IgG antibodies induced by vaccination and to determine the impact that IgG antibodies might have on IgM antibody reactivity with GM2-positive tumor cells. Post-immunization sera from seven immunized patients were separated by size-exclusion chromatography into IgM and IgG fractions and a variety of serological assays were performed with the individual fractions and their combinations. Assays identifying specific IgM or IgG reactivity demonstrated partial inhibition by the opposite fraction. However, when the endpoint was complement-mediated lysis or overall antibody binding, which may more faithfully predict in vivo complement-mediated opsonification and lysis, the combinations of IgM and IgG fractions consistently demonstrated higher reactivity than either fraction alone. In addition, ADCC was induced in all seven patients. The results were the same whether the sera were obtained after 2 months or 2 years of immunizations. These findings suggest that IgG antibodies induced by the GM2-KLH plus QS-21 vaccine will not inhibit and should further augment the clinical impact of induced IgM antibodies.

GD3/proteosome vaccines induce consistent IgM antibodies against the ganglioside GD3

The gangliosides of melanoma and other tumours of neuroectodermal origin are suitable targets for immune intervention with tumour vaccines. The optimal vaccines in current use contain ganglioside plus bacillus Calmette-Guérin and induce considerable morbidity. We have screened a variety of new adjuvants in the mouse, and describe one antigen-delivery system, proteosomes, which is especially effective. Highly hydrophobic Neisserial outer membrane proteins (OMP) form multimolecular liposome-like vesicular structures termed proteosomes which can readily incorporate amphiphilic molecules such as GD3 ganglioside. The optimal GD3/proteosome vaccine formulation for induction of GD3 antibodies in the mouse is determined. Interestingly, the use of potent immunological adjuvants in addition to proteosomes augments the IgM and IgG antibody titres against OMP in these vaccines but GD3 antibody titres are unaffected. The application of proteosomes to enhance the immune response to GD3 extends the concept of the proteosome immunopotentiating system from lipopeptides to amphipathic carbohydrate epitopes such as cell-surface gangliosides. The demonstrated safety of meningococcal OMP in humans and the data in mice presented here suggest that proteosome vaccines have potential for augmenting the immunogenicity of amphipathic tumour antigens in humans.

Increased tumor cell reactivity and complement-dependent cytotoxicity with mixtures of monoclonal antibodies against different gangliosides

Melanomas and other cancers of neuroectodermal origin express multiple cell-surface gangliosides in patterns that vary significantly even within the same tumor type. Monoclonal antibodies (mAb) against four of these gangliosides (GM2, GD2, 9-O-acetyl-GD3 and GD3) were tested alone and in combination on 14 tumor cell lines (7 melanomas, 3 neuroblastomas, 3 sarcomas and 1 astrocytoma) using flow cytometry and complement-dependent cytotoxicity (CDC) assays. Increased tumor cell recognition and CDC resulting from the combination of three or four mAb were found in 14/14 tested cell lines, and this was most striking when each mAb was used at suboptimal concentration. At these concentrations, the average mean fluorescence intensity of the 14 cell lines with individual mAb was between 3.0 and 6.8 and increased to 10.8 and 18.8 with the three- and four-mAb mixtures. The average percentage CDC-specific release with individual mAb was 2.0%–8.3%, and 12.3% and 16.6% with the three- and four-mAb combinations. The number of cell lines showing significant mean fluorescence intensity and CDC increased from 2–8/14 with single mAb to 13–14/14 with the mixtures of three or four mAb. Our experimental results support the rationale for active immunization with a polyvalent ganglioside vaccine or passive therapy with a combination of mAb to different gangliosides in patients with tumors of neuroectodermal origin. In addition, our studies have demonstrated that 9-O-acetyl-GD3 is a surprisingly effective target for immune attack, although it is a minor constituent of these cells.

Ganglioside conjugate vaccines : immunotherapy against tumors of neuroectodermal origin

AbstractGangliosides are known to be suitable targets for immune attack against cancer but they are poorly immunogenic. Active immunization with ganglioside/BCG or liposome vaccines results in moderate titer IgM antibody responses of short duration. Covalent attachment of poorly immunogenic antigens to immunogenic proteins is a potent method for inducing an IgG antibody response. GD3, a dominant ganglioside on malignant melanoma, was modified by ozone cleavage of the double bond in the ceramide backbone, an aldehyde group introduced and used for coupling via reductive amination to ε-aminolysyl groups of proteins. Utilizing this method, GD3 conjugates were constructed with:1.Synthetic multiple antigenic peptide (MAP) constructs expressing 4 repeats of a malaria T-cell epitope;2.Outer membrane proteins (OMP) ofNeisseria meningitidis;3.Cationized bovine serum albumin;4.Keyhole limpet hemocyanin (KLH); and5.Polylysine. In addition, conjugates containing only the GD3 oligosaccharide were synthesized. All constructs were tested for antigenicity using anti-GD3 antibody R24, and for immunogenicity in mice. Serum antibody levels were analyzed by ELISA and immune thin-layer chromatography. Results in the mouse show a significant improvement in the IgM antibody response and a consistent IgG response against GD3 using GD3-KLH conjugates. Other carrier proteins and the use of GD3 oligosaccharide were significantly less effective. If improved immunogenicity and clinical benefit with conjugate vaccines can be demonstrated in patients with melanoma, this approach may be applicable to patients with other tumors of neuroectodermal origin, including gliomas, glioblastomas, astrocytomas, and neuroblastomas.