Abundant expression of fibronectin is a major feature of leukemic dendritic cells differentiated from patients with acute myeloid leukemia.

Dendritic cells (DCs) are the most potent antigen-presenting cells responsible for the initiation of primary immune responses, playing a key role in eliciting effective antitumor immune responses. We reported previously that leukemic blasts from selected patients with acute myeloid leukemia (AML) were able to differentiate in vitro into cells with DC features. In order to identify genes differentially expressed in leukemia-derived DCs (AML-DCs), a polymerase chain reaction (PCR)-based subtraction approach was applied using cDNA from AML-DCs and monocyte-derived DCs from healthy donors as competitors. In the 548 sequences analyzed, 80% corresponded to fibronectin (FN) gene fragments. Overexpression of FN in AML-DCs was demonstrated both by semiquantitative PCR analysis and by immunostaining. In addition, we could show that FN was secreted by AML-DCs. Indeed, FN overexpression was already detectable in AML blasts of M4 and M5 subtype, and was significantly induced during DC differentiation after culture. Although the molecular events leading to overexpression of FN and the in vivo relevance of this phenomenon remain to be resolved, leukemic DCs appear to have specific patterns of differentiation, warranting stringent biological and cellular monitoring for the development and testing of leukemic DC-based immunotherapeutic strategies.

Immunisation of metastatic cancer patients with MAGE-3 protein combined with adjuvant SBAS-2: a clinical report.

Fifty-seven patients with MAGE-3-positive measurable metastatic cancer, most of them with melanoma, were vaccinated with escalating doses of a recombinant MAGE-3 protein combined with a fixed dose of the immunological adjuvant SBAS-2, which contained MPL and QS21. The immunisation schedule included 4 intramuscular (i.m.) injections at 3-week intervals. Patients whose tumour stabilised or regressed after 4 vaccinations received 2 additional vaccinations at 6-week intervals. The vaccine was generally well tolerated. Among the 33 melanoma patients who were evaluable for tumour response, we observed 2 partial responses, 2 mixed responses and 1 stabilisation. Time to progression in these 5 patients varied from 4 to 29 months. In addition, a partial response lasting 10 months was observed in 1 of the 3 metastatic bladder cancer patients included. None of the tumour responses described above involved visceral metastases. Immunological responses to the vaccine will be reported separately.

Biological and clinical developments in melanoma vaccines.

The identification of antigens recognised on human tumours by autologous T-lymphocytes has opened the way for vaccination strategies involving defined tumour antigens. These vaccinations are therapeutic, i.e. they involve patients with detectable disease. Tumour regressions have been observed in a minority of melanoma patients in Phase I/II trials. Some of these regressions have been complete and long lasting. Improving the efficacy of therapeutic vaccines will critically depend on their capacity to trigger a robust immune response, on the development of appropriate methods to monitor these antitumour immune responses to vaccination and on a better understanding of the mechanisms used by tumours to escape immune attack. Finally, the initiation of large randomised Phase III trials will determine the impact of these vaccines on melanoma treatment.

MAGE-A genes are not expressed in human leukemias.

Immunotherapy is promising to improve the prognosis of human leukemias, at least as adjuvant treatment. Tumor-associated antigens such as antigens encoded by MAGE-A1, -A2, -A3, -A4, -A6 and -A12 genes might provide tools in this field. We demonstrated recently that the presentation peptides encoded by MAGE-A genes might make leukemic blasts suitable targets to cytolytic T lymphocytes. We reported previously negative data of MAGE-A1 gene expression in hematological malignancies, but in further studies positive results of MAGE-A gene expression were published in some subtypes of hematological malignancies such as T leukemia, myeloma and Hodgkin's disease. This led us to enlarge the screening of MAGE-A gene expression in human leukemias. In the RT-PCR screening of a large panel including 154 patients, only weak signal were detected in a few samples. We conclude that MAGE-A genes are not expressed in human leukemias.

Expression of gene MAGE-A4 in Reed-Sternberg cells.

Genes of the MAGE-A family are expressed in several types of solid tumors but are silent in normal tissues with the exception of male germline cells, which do not carry HLA molecules.Therefore, peptides encoded by MAGE-A genes are strictly tumor-specific antigens that can be recognized by CTL and constitute promising targets for immunotherapy. The expression of 6 genes of the MAGE-A family was tested with reverse transcriptase-polymerase chain reaction in lymphoma samples. Among 38 samples of non-Hodgkin lymphoma, 1 anaplastic large cell lymphoma expressed genes MAGE-A1, -A2, -A3, -A4, and -A12, and 1 lymphoepithelioid T-cell lymphoma expressed gene MAGE-A4. Five of 18 samples (28%) from patients with Hodgkin disease expressed gene MAGE-A4. In tissue sections, staining by a monoclonal antibody that recognizes the MAGE-A4 protein was observed in 11 of 53 samples (21%) from patients with Hodgkin disease. In the positive samples, the Reed-Sternberg cells were strongly stained whereas the surrounding cells were not. These results indicate that Hodgkin disease may be a target for specific immunotherapy involving MAGE-A4 antigens.

High frequency of autologous anti-melanoma CTL directed against an antigen generated by a point mutation in a new helicase gene.

We have identified an Ag recognized by autologous CTL on the melanoma cells of a patient who enjoyed an unusually favorable clinical evolution. The antigenic peptide, which is presented by HLA-A28 molecules, is encoded by a mutated sequence in a new gene. This gene, which was named MUM-3, is expressed ubiquitously and shows homology with the RNA helicase gene family. Limiting dilution analysis indicated that at least 0.15% of the blood CD8 T cells were tumor-specific CTL precursors. The MUM-3 Ag was recognized by 90% of these CTL, indicating that it is the dominant target Ag of the tumor-specific CTL response. The high frequency of anti-MUM-3 CTL was confirmed with tetramers of soluble HLA-A28 molecules loaded with the antigenic peptide. MUM-3 tetramers stained 1.2% of blood CD8 cells, a frequency that has never been reported for T cells directed against a strictly tumor-specific Ag. To confirm these results, the CD8 T cells that were clearly labeled with tetramers were restimulated in clonal conditions. About 90% of these cells proliferated, and all the resulting clones proved lytic and MUM-3 specific. By improving the conditions used for the in vitro restimulation of CTL precursors by the tumor cells, the same frequency could be obtained in limiting dilution analysis. These results show that some cancer patients have a high frequency of circulating CTL that are directed against a strictly tumor-specific Ag. These CTL are responsive to restimulation in vitro and are easily detected with tetramers. Such responses may therefore be an achievable goal for therapeutic vaccination with tumor-specific Ags.

Genes encoding tumor-specific antigens are expressed in human myeloma cells.

Genes of the MAGE, BAGE, GAGE, and LAGE-1/NY-ESO-1 families encode antigenic peptides that are presented by HLA class I molecules and that are recognized on human tumors by autologous cytolytic T lymphocytes. These genes are expressed in many solid tumor types but not in normal tissues, except male germline cells. Because the latter cells are devoid of HLA molecules, the derived antigens are strictly tumor-specific and should constitute safe immunogens for cancer immunotherapy. We detected a significant expression of these genes in a high proportion of bone marrow samples from patients with advanced multiple myeloma. This observation provides a basis for clinical trials aimed at inducing a cellular immune response directed at malignant plasma cells in advanced myeloma patients.

Tumor regressions observed in patients with metastatic melanoma treated with an antigenic peptide encoded by gene MAGE-3 and presented by HLA-A1.

Thirty-nine tumor-bearing patients with metastatic melanoma were treated with 3 subcutaneous injections of the MAGE-3.A1 peptide at monthly intervals. No significant toxicity was observed. Of the 25 patients who received the complete treatment, 7 displayed significant tumor regressions. All but one of these regressions involved cutaneous metastases. Three regressions were complete and 2 of these led to a disease-free state, which persisted for more than 2 years after the beginning of treatment. No evidence for a cytolytic T lymphocyte (CTL) response was found in the blood of the 4 patients who were analyzed, including 2 who displayed complete tumor regression. Our results suggest that injection of the MAGE-3.A1 peptide induced tumor regression in a significant number of the patients, even though no massive CTL response was produced.

PRAME, a gene encoding an antigen recognized on a human melanoma by cytolytic T cells, is expressed in acute leukaemia cells.

Gene PRAME was found to encode an antigen recognized on a human melanoma cell line by an autologous cytolytic T-lymphocyte clone. This gene is expressed at a high level in a very large fraction of tumours, such as melanomas, non-small-cell lung carcinomas, sarcomas, head and neck tumours and renal carcinomas. It is therefore a candidate for tumour immunotherapy even though some low expression is found in certain normal tissues. We tested by RT-PCR the expression of PRAME on more than 250 bone marrow or blood samples from patients with a haematological malignancy. Approximately 25% of the acute leukaemia samples were positive. Remarkably, all acute myeloblastic leukaemias that carried the chromosomal translocation t(8;21), which fuses the genes AML1 and ETO, expressed PRAME at a high level.

Mutations of the beta2-microglobulin gene result in a lack of HLA class I molecules on melanoma cells of two patients immunized with MAGE peptides.

Mutations have been identified in the beta2-microglobulin gene of tumor cells of two metastatic melanoma patients who received immunizations with MAGE peptides. One mutation abolishes the start codon whereas the other introduces a premature stop codon. The second beta2-microglobulin allele of both tumors appears to be lost on the basis of sequence data and loss of microsatellite heterozygosity. The lack of beta2-microglobulin gene product results in the absence of HLA class I antigens on the surface of the tumor cells. This may explain why the tumors of both patients progressed despite the immunization treatment and shows the necessity of analyzing in depth the antigen presentation capability of the tumor cells for the interpretation of clinical trials involving anti-tumor vaccination.

Characterization of an antigen that is recognized on a melanoma showing partial HLA loss by CTL expressing an NK inhibitory receptor.

Melanoma lines MEL.A and MEL.B were derived from metastases removed from patient LB33 in 1988 and 1993, respectively. The MEL.A cells express several antigens recognized by autologous cytolytic T lymphocytes (CTL) on HLA class I molecules. The MEL.B cells have lost expression of all class I molecules except for HLA-A24. By stimulating autologous lymphocytes with MEL.B, we obtained an HLA-A24-restricted CTL clone that lysed these cells. A novel gene, PRAME, encodes the antigen. It is expressed in a large proportion of tumors and also in some normal tissues, albeit at a lower level. Surprisingly, the CTL failed to lyse MEL.A, even though these cells expressed the gene PRAME. The CTL expresses an NK inhibitory receptor that inhibits its lytic activity upon interaction with HLA-Cw7 molecules, which are present on MEL.A cells and not on MEL.B. Such CTL, active against tumor cells showing partial HLA loss, may constitute an intermediate line of anti-tumor defense between the CTL, which recognize highly specific tumor antigens, and the NK cells, which recognize HLA loss variants.

Von Recklinghausen neurofibromatosis and hematologic malignancies: 2 case reports in adulthood.

Von Recklinghausen's neurofibromatosis is a hereditary disease predisposing to distinctive malignant hemopathies. These often develop during early childhood and are characterized by particular cytologic subtypes: juvenile chronic myeloid leukemia, monosomy 7-associated myeloproliferative syndrome and myelomonocytic leukemia. The etiopathologic mechanism underlying this association begins to be elucidated: the neurofibromatosis gene behaves like a tumor suppressor gene; its inactivation by mutation results in activation of the corresponding oncogenes. We report here the cases of two late-aged adults with neurofibromatosis: the first developed acute myelogenous leukemia, the second polycythemia vera. Based on a review of the literature, we suggest that, in opposition to childhood, the association between neurofibromatosis and malignant blood diseases is not demonstrated in adulthood.

Structure of the gene of tum- transplantation antigen P35B: presence of a point mutation in the antigenic allele.

Mutagen treatment of P815 tumour cells produces tum- variants that are rejected by syngeneic mice because they express new transplantation antigens. These 'tum-' antigens elicit a cytolytic T lymphocyte (CTL) response but no detectable antibody response. The DNA of tum- variant P35 was transfected into P815 cell line P1.HTR. Transfectants expressing tum- antigen P35B were identified on the basis of their ability to stimulate anti-P35B CTL. This was repeated with a cosmid library and a cosmid carrying the sequence encoding antigen P35B was recovered from a transfectant expressing the antigen. Gene P35B is 6 kb long and contains 11 exons. The sequence shows no homology with the previously identified tum- gene P91A nor with any gene presently recorded in the data banks. The antigenic allele of gene P35B differs from the normal allele by a point mutation located in exon 5. This mutation, which replaces a Ser by an Asn residue, was shown by site-directed mutagenesis to be responsible for the expression of the antigen. A synthetic decapeptide covering the sequence surrounding the tum- mutation rendered P815 cells sensitive to lysis by anti-P35B CTL. Surprisingly, the homologous peptide corresponding to the normal sequence of the gene had the same effect, indicating that this tum- mutation does not exert its effect by generating the aggretope or the epitope of the antigenic peptide. As observed previously with gene P91A, we found that fragments of gene P35B containing only exons 4 and 5, which were cloned in non-expression vectors, transferred efficiently the expression of the antigen.