Disialoganglioside GD2 on human neuroblastoma cells: target antigen for monoclonal antibody-mediated cytolysis and suppression of tumor growth

A murine monoclonal antibody 14.18 specifically recognizes disialoganglioside GD2, the major ganglioside expressed on the surface of human neuroblastoma cells. This monoclonal antibody (Mab) is of immunoglobulin G3 isotype, has an affinity constant (KA) of 3.5 X 10(8) M-1, and reacts preferentially with tumor cells and fresh frozen tumor tissues of neuroectodermal origin in enzyme-linked immunosorbent assay and immunoperoxidase assays, respectively. Mab 14.18 effectively lyses a number of human neuroblastoma cell lines by two distinct mechanisms, i.e., antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity. There is a good correlation between the average number of antibody-binding sites per neuroblastoma cell and the amount of cell lysis observed in complement-dependent cytotoxicity and antibody-dependent cellular cytotoxicity. In addition, Mab 14.18 suppresses establishment as well as growth of progressively growing, established human neuroblastoma tumors in nude mice when injected 24 h and 9 days, respectively, after the initial s.c. inoculation of tumor cells. These data suggest that Mab 14.18 can mediate tumor cell killing in vivo and in vitro and may thereby prove useful for immunotherapy of human neuroblastoma.     

Anti-GD3 monoclonal antibody analysis of childhood T-cell acute lymphoblastic leukemia: detection of a target antigen for antibody-mediated cytolysis

We have demonstrated in previous studies significant quantitative differences in the ganglioside content of leukemia cell membranes within immunological subclasses of acute lymphoblastic leukemia (ALL): The disialoganglioside GD3 (disialolactosylceramide) is increased in lymphoblasts with a T-cell immunophenotype compared to non-T-ALL blasts. Utilizing an indirect immunofluorescence assays with a monoclonal antibody to GD3(R24), pretreatment leukemic lymphoblasts from 80 children with ALL were assayed for GD3 expression. GD3 was observed in 75% of leukemic samples in which lymphoblasts exhibited a T-cell phenotype, whereas none of the 33 non-T-ALL samples tested exhibited GD3. Correlation between the expression of GD3 and various antigenic determinants of T-cell differentiation was restricted to CD2; 75% of CD2-negative T-cell ALL blasts failed to express GD3. Anti-GD3 immunoreactivity to T-ALL samples was not restricted to R24 in that two other monoclonal anti-GD3 antibodies were similarly reactive to T-ALL blasts. In vitro incubation of T-cell lymphoblasts with the anti-GD3 antibodies, R24 and C281 and human serum resulted in significant cytotoxicity, and R24 also mediated antibody-dependent cellular cytotoxicity by normal effector cells. Cytotoxicity was specific for those T-ALL blast cell populations which reacted with anti-GD3 as assessed by immunofluorescence microscopy. Since immunoreactivity with monoclonal antibody to GD3 was exclusively observed in a large population of immunophenotypically defined T-cell leukemic lymphoblasts, these studies suggest a possible immunodiagnostic and immunotherapeutic potential for anti-GD3 monoclonal antibodies in T-cell lymphoblastic malignancies.     

Biosynthesis of procalcitonin in small cell carcinoma of the lung

Immunoreactive calcitonin (CT) secreted by DMS 53, a cell line derived from human small cell carcinoma of the lung, consists almost entirely of molecular species larger than the mature hormone (Mr 3,420). Messenger RNA isolated from DMS 53 cells and nude mouse tumors was translated in wheat germ systems, and the products were precipitated with CT-specific antisera. Analyses of the translation products by electrophoresis on 15% polyacrylamide-sodium dodecyl sulfate gels indicated synthesis of a Mr 16,500 preprohormone that was reduced to Mr 14,500 by cotranslation with microsomal membranes. Immunoprecipitation of CT from media from pulse-labeled cultures revealed two major products (Mr 16,500 and Mr 14,500) and up to three minor secreted polypeptides (Mr 9,400, 8,400, and 6,800). Intracellular CT from cell homogenates appeared almost entirely as a single major product (Mr 14,500) and possibly 3-4 minor components (Mr 16,500; 9,200, 8,400, and 6,800). No glycosylated forms of CT were demonstrable by lectin binding methods or labeling attempts with tritiated sugars. The presence of multiple CT species in DMS 53 cells suggests significant post-translational processing of the larger precursor molecules and the accumulation and secretion by small cell carcinoma of the lung of several intermediate immunoreactive forms via a glycosylation-independent secretory pathway.     

Membrane antigen in small cell carcinoma of the lung defined by monoclonal antibody SM1

Mouse myeloma cells were fused with spleen cells from BALB/c mice immunized with a cell line derived from human small cell carcinoma (SCC) of the lung. The cloned hybridoma SM1 produced antibody that was reactive with the surface membrane of SCC cell lines and SCC tumors but not with the membrane of several non-SCC cell lines and tumors. SM1 ascites fluid was used to screen for reactivity of the antibody with other human cancer cell lines, tumor tissues, and normal tissues. SM1 antibody was found to be unreactive with neuroblastoma, adrenal carcinoma, melanoma, and bronchial carcinoid. Reactivity was detected with some breast carcinoma cell lines but not with breast cancer tissue specimens. In the same individual, the antibody was reactive with SCC lung tumor and SCC metastatic to the liver but not with normal tissues, including bronchus, lung parenchyma, liver, kidney, and brain. Human erythrocytes and marrow cells were also unreactive. Since SM1 detects an antigen that is present in greatest amounts on the surface membrane of SCC of the lung, this antibody may be useful in tracing the lineage patterns of human lung cancers.     

Heterogeneity of cell surface antigen expression of human small cell lung cancer detected by monoclonal antibodies

Using immunohistochemistry, radiobinding, and indirect immunofluorescence assays, seven distinct cell surface antigens, detected by monoclonal antibodies, were analyzed for the degree of homogeneity or heterogeneity of antigen expression on a panel of human small cell lung cancers. The panel included 7 tumors taken directly from patients, 21 established cell lines (9 of which were derived from different metastatic sites of 3 patients), and 33 clonal derivatives of 3 lines. With all assays, considerable heterogeneity of antigen expression between tumors from different patients was observed. In both fresh tumors and in cell lines, as well as in cell lines established from different metastatic sites in an individual patient, we observed intratumor heterogeneity finding antigen positive and negative cells and variation in antigenic density, by immunohistochemistry and indirect immunofluorescence assays. Antigenic expression was not cell cycle dependent. In addition, when cell lines or patient samples expressing antigen positive and antigen negative tumor cells were cloned, heterogeneity of antigenic expression was still present in the clonal lines. This suggests that either the expression of the antigen was not heritable and/or the ability to regenerate antigenic heterogeneity is an intrinsic property of the tumor cells. The heterogeneity of antigen expression on lung cancer cells has significant implications for the use of these and other monoclonal antibodies in the study and therapy of lung cancer.     

Bombesin/gastrin-releasing peptide receptor: a potential target for antibody-mediated therapy of small cell lung cancer.

PURPOSE: Bombesin/gastrin releasing peptide (BN/GRP) is a growth factor for small cell lung cancer (SCLC). The receptor (R) for BN/GRP is overexpressed on SCLC cells and other solid tumors. BN/GRP and its receptor form an autocrine loop to promote tumor growth. We developed a novel immunotherapeutic approach targeting cell surface BN/GRP-R on SCLC cells and an immune trigger molecule on host immune effector cells to direct immune effector cells to SCLC cells and mediate targeted cancer cell destruction. Targeted immunotherapy combined with chemotherapy enhanced cell killing. EXPERIMENTAL DESIGN: We designed a synthetic BN/GRP antagonist (Antag 2) and constructed a bispecific molecule (BsMol), H22xAntag 2 (humanized monoclonal antibody) for FcgammaRI. We tested the binding of the BsMol to several SCLC cell lines, its ability to mediate cytotoxicity of SCLC by IFN-gamma-activated human monocytes with chemotherapy, and BsMol-mediated immunotherapy in an animal model of SCLC human xenograft. RESULTS: Common chemotherapy (cisplatin, etoposide, and paclitaxel) inhibited thymidine uptake into SCLC cells in a dose-dependent pattern. Antibody-dependent cellular cytotoxicity mediated by the BsMol inhibited thymidine uptake into SCLC cells and was largely dependent on E:T cell ratio. When SCLC cells were treated with antibody-dependent cellular cytotoxicity followed by exposure to chemotherapy agents an additional 25-40% inhibition of thymidine uptake into SCLC cells was observed consistently. With BsMol and IFN-gamma-activated human monocytes, tumor burdens were reduced significantly in immunodeficient mice bearing human SCLC xenografts. CONCLUSIONS: Combined chemotherapy and immunotherapy targeting BN/GRP-R with a BsMol significantly enhances targeted SCLC cell killing.     

Immunoperoxidase staining for carcinoembryonic antigen in small cell carcinoma of the lung

Diagnostic pretreatment biopsies in 53 patients with small cell carcinoma of the lung (SCCL) were examined retrospectively for presence of carcinoembryonic antigen (CEA). The unlabelled antibody-enzyme (PAP) technique was utilized. All patients had been treated with intensive chemotherapy. The patients were divided into 3 groups according to the immunoperoxidase staining reactions of the biopsies. Thirteen (24.5%) patients had biopsies that were strongly positive for CEA while22 (41.5%) had slightly positive and 18 (34%) had negative biopsies. The patients with biopsies strongly positive for CEA had a significantly longer survival as compared with patients with negative staining results (P < 0.05). No significant correlation was observed between immunoperoxidase staining for CEA and morphological subtypes of SCCL according to the WHO 1977 classification.     

HLA class I and neuroendocrine antigen expression in multidrug resistant small cell lung carcinoma cell lines

Antigenic marker expression was studied in a series of eight small cell lung carcinoma (SCLC) cell lines, according to their histological subtype, classic or variant. These lines were obtained from human tumors xenografted into nude mice, originally derived from heterotransplanted tumor biopsy samples. We looked at an altered expression of HLA class I antigens, a battery of neuroendocrine antigens and the P-glycoprotein (Pgp) responsible for MDR1 encoded multidrug resistance, as markers of tumor malignancy. Three cell lines out of four of the classic subtype and two cell lines out of four of the variant subtype showed a lack or a low expression of HLA class I antigen. Recombinant interferon gamma (rIFN-gamma) treatment (100 U/ml, for 48 h) increased HLA class I expression of the cell lines differently, but did not induce an imbalance between HLA-A and HLA-B molecules as described in other tumor models. Neuroendocrine antigens were tested in six out of these eight lines, using a family of monoclonal antibodies developed against the cell membrane antigens of low passage cell lines derived from pleural effusions (de Leij et al, Cancer Res 45: 2192-2200, 1985). Globally, these antigens were more highly expressed in classic subtypes of SCLC. Neuroendocrine antigens corresponding to MOC-21 and MOC-32 monoclonal antibodies were weakly expressed in variant forms. Pgp expression was detectable with the JSB1 monoclonal antibody on the three variant SCLCs out of the six lines. Comparing two cell lines originated from the same patient before and after therapy, we showed that neuroendocrine reactivity to MOC-21 and MOC-32 was lost simultaneously with a gain of Pgp expression, and with a classic to variant histological transition. With regard to the clinical evolution, HLA class I expression and stimulation by rIFN-gamma was not related to malignancy. It appears that for variant forms, a low expression of neuroendocrine antigens detected by MOC-21 and MOC-32 monoclonal antibodies and a high level of Pgp predict for a poor prognosis.     

Stimulation of the aberrant expression of a paraneoplastic antigen, recoverin, in small cell lung cancer cell lines

Recoverin, a retina-specific Ca2+-binding protein, is one of the paraneoplastic antigens (PNAs) which are normally present in neurons, but can also be aberrantly expressed in malignant tumors localized outside the nervous system. In this study, we have analyzed 16 small cell lung carcinoma (SCLC) and 12 non-small cell lung carcinoma cell lines and found that none of them is capable of expressing recoverin in vitro. However, two small cell lung carcinoma lines, NCI-H69 and NCI-H82, became recoverin-positive after cultivation in the presence of butyrate. Recoverin expression in the butyrate-treated cells has been detected by immunoblotting with polyclonal (monospecific) antibodies against recoverin and confirmed by the analysis of recoverin mRNA expression. To our knowledge, this work is the first to demonstrate stimulation of the aberrant expression of recoverin in cancer cell lines in vitro. This result opens the way to investigation of the mechanisms underlying the aberrant expression of recoverin, as well as other paraneoplastic antigens, in tumor cells.     

Tumor-associated membrane sialoglycoprotein on human small cell lung carcinoma identified by the IgG2a monoclonal antibody SWA20

A mouse IgG2a monoclonal antibody, SWA20, defining a tumor-associated cell surface antigen on small cell carcinoma of the lung (SCC) was generated. The reactivity of the antibody with cell lines was examined by indirect immunofluorescence staining and solid phase radioimmunoassay and the reactivity with tissues by immunoperoxidase staining. The antibody reacts with a proportion of small cell carcinoma cell lines (4 of 8) and tissues (7 of 12), but not with other pulmonary or extrapulmonary cell lines (0 of 30) or tumor tissues (0 of 78). The antibody was unreactive with primary cultures of normal bronchial epithelial cells, RBC, and WBC. Immunoperoxidase staining of normal tissues showed rare antigen-positive cells in suprabasal layers of bronchial epithelium and less than 10% of positive cells in colon epithelium. Immunoblots of SCC extracts demonstrated antibody reactivity with a doublet band at Mr 40,000, a broader band at Mr 100,000, and a band at Mr 180,000. The antigen was not present in crude lipid extracts of SCC cells. Solid phase radioimmunoassays and immunoblots showed binding competition with the lectin Triticum vulgaris, sensitivity of the antigen to neuraminidase, and a partial sensitivity to treatment with periodate. The antigen was coexpressed on SCC cell lines with the antigen sGP90-135 defined first by antibody LAM8 (R. Waibel, C. J. O'Hara, and R. A. Stahel. Cancer Res., 47:3766-3770, 1987) but differed from it by lack of reactivity with Lea-positive saliva and partial resistance to periodate treatment. There was no binding competition between radiolabeled antibodies SWA20 and LAM8 to SCC target cells. The IgG2a antibody SWA20 identifies a previously undescribed tumor-associated surface membrane antigen, sGP100, expressed selectively on a proportion of SCC.     

Human lung carcinoma monoclonal antibody specific for the Thomsen-Friedenreich antigen

A monoclonal antibody, RS1-114, was raised against the human adenocarcinoma of the lung cell line A549. By studying the reactivity of RS1-114 with A549 cells following chemical and enzymatic treatments, it was shown that the epitope is a galactose-containing carbohydrate, which is devoid of sialic acid. Hemagglutination of desialylated RBCs, enzyme-linked immunosorbent assay studies with glycoprotein antigens before and after desialylation, and competition studies using peanut agglutinin indicate that monoclonal antibody RS1-114 recognizes the Thomsen-Friedenreich antigen, a cryptic determinant on human erythrocytes which can be exposed by neuraminidase treatment. It is expressed in an unhidden form on a large percentage of carcinomas and is therefore an important human tumor marker. RS1-114 is reactive with cryptic determinants of the Thomsen-Friedenreich antigen on white blood cells as well as red blood cells, and it reacts with unhidden determinants on human tumor cell lines. The number of binding sites on carcinoma cells is further increased by neuraminidase treatment. By immunohistochemical staining, it was shown that 75% of the human tumors tested are reactive with RS1-114. These include tumors of the breast, colon, lung, kidney, ovary, and rectum.     

Effects of retinoic acid and bromodeoxyuridine on human melanoma-associated antigen expression in small cell lung carcinoma cells.

The dispersed neuroendocrine system includes cells with different embryological derivations, sharing a common neuroendocrine (NE) program, as indicated by the expression of NE markers, some of which are shared antigenic determinants. We report here that the small cell lung carcinoma cells NCI-H69 express the two human melanoma-associated antigens (HMAA) NGA/LS62 an LS109. Incubation of NCI-H69 cells with maturational inducers, such as retinoic acid and bromodeoxyuridine (BrdU), upregulated the expression of both HMAA. Exposure to BrdU for 4 weeks induced the appearance of a different phenotype in subpopulations of NCI-H69 cells, which became epithelioid, substrate-adherent, grew in monolayer and continued to express NE-associated antigens in variable amount. The shift in phenotype was not reversible after BrdU withdrawal and was maintained for at least 6 months in continuous culture. The substrate adhesion of NCI-H69 cells was paralleled by a change in NGA glycosylation pattern, thus suggesting a possible functional role for NGA in cell substrate adhesion/recognition.     

Characterization of an epithelial and a tumor-associated human small cell lung carcinoma glycoprotein antigen

The small cell carcinoma (SCC) antigens recognized by LAM2 and LAM8 antibody were characterized by comparison of their tissue expression and analysis of their biochemical composition. LAM2, but not LAM8 antigen could be demonstrated in lipid extracts of SCC cells. By immunohistochemical staining the SCC antigen LAM2 was shown to be an epithelial type membrane antigen. Immunoblotting experiments and competition solid phase radioimmunoassays showed LAM2 antigen to be a native conformation of a glycoprotein with major bands at Mr 100,000-120,000 and a minor band at Mr 210,000. L-Fucose was a dominant part of the epitope which appeared to be closely related to the carbohydrate epitope of the blood group antigen H(O). The tumor-associated membrane antigen LAM8 was shown to be a glycoprotein with major bands at Mr 90,000-135,000 and a minor band at Mr 200,000. Neuraminic acid was the predominant part of the carbohydrate epitope. LAM8 antibody recognized a structure in the saliva of Lea positive probands, but untreated and neuraminidase-treated SCC extracts were unreactive with anti-Lea antibody. Anti CA 19-9 (sialo-Lea antigen) and LAM2 antibodies did not compete for LAM8 binding in direct radioimmunoassays. The sialo-GP90-135 antigen recognized by LAM8 antibody therefore is likely to represent a novel tumor antigen.     

Carcinoembryonic antigen for monitoring patients with small cell carcinoma of the lung during treatment

Carcinoembryonic antigen (CEA) was measured at specific intervals in the plasma of 56 patients with small cell carcinoma of the lung. Of these patients, 47 had serial analyses for varying periods during their illness, 42 had pretreatment CEA levels, and 17 of the latter patients had determinations throughout the entire course of their disease. Pretreatment CEA levels were elevated above 2.5 ng/ml for 74% of the 42 patients and above 5.0 ng/ml for 48%. Although exceptions were noted, in general, a direct relationship was found between pretreatment CEA levels and extent of disease or tumor burden. Initial stage of disease, however, was more predictive of survival than was the pretreatment CEA level. With response to therapy, a corresponding decrease in CEA levels occurred for patients with an elevated pretreatment level. For those patients with a pretreatment CEA level below 5.0 ng/ml, an immediate slight increase in level was often seen associated with response and followed by a subsequent fall after one month. A rising CEA level was usually found with recurrence or progression of disease after initial response and occurred frequently prior to clinical evidence of progression. In combination with careful clinical evaluation, serial CEA measurements can aid in assessing tumor changes associated with treatment in patients with small cell carcinoma of the lung particularly at the times of recurrence or disease progression following a partial or complete response.     

The epithelial cell surface antigen 17-1A, a target for antibody-mediated tumor therapy: its biochemical nature, tissue distribution and recognition by different monoclonal antibodies

The antigen defined by MAb 17-1A, raised against colorectal cancer cells and currently used in immunotherapy trials, is a 37 kDa protein containing N-linked glycans as demonstrated by inhibition of glycosylation with tunicamycin. The 17-1A antigen (17-1A Ag) is broadly distributed in normal epithelial tissues and is also found in various types of carcinoma. Quantitative differences in expression between normal and malignant tissues were only apparent in gastric carcinomas. After immunizing mice with fresh colon carcinoma tissue, 4 MAbs were obtained that showed the same tissue reaction pattern as MAb 17-1A and recognized the 17-1A Ag as judged from precipitation and immunoblotting experiments. MAbs M72 and M74 reacted with an anti-idiotypic serum directed against the original MAb 17-1A and were able to block 17-1A binding. In contrast, M77 and M79 are directed against a different epitope. This makes them potentially interesting for passive immunotherapy in regimens involving a combination of MAbs.     

Identification of a surface antigen, common to murine lung and Lewis lung carcinoma, by monoclonal antibody

Lewis lung carcinoma (3LL) is a spontaneous murine lung carcinoma which preferentially metastasizes into the lung. Monoclonal antibodies (MAb) against 3LL cells were prepared by fusing spleen cells from rats immunized with 3LL and the P3-NSI/I-Ag4 (NSI) plasmacytoma. Two hybridomas were selected which secreted antibodies capable of preferential binding to 3LL cells and not to other malignant cells. Immunohistochemical staining of sections prepared from various mouse organs with one of the MAbs (22.2), revealed exclusive staining of the lung tissue. The data suggest that a lung-specific, 3 LL-associated antigen, is detected by the 22.2 MAb on 3LL cells. The 22.2 antigen is a surface molecule, and is stable following trypsin treatment and fixation by glutaraldehyde or picric acid. Cells that were prepared from a subcutaneous tumor or from lung metastases express high amounts of 22.2 antigen compared to cells of the culture 3LL line (3LLC). Upon reinoculation of 3LLC cells into mice there was a gradual increase in 22.2 antigen expression. It was possible to select from the heterogeneous 3LLC cell population several clones which express high amounts of antigen. These clones differed also in their metastatic patterns. However, the metastatic capacity of these clones did not correlate with expression of the 22.2 antigen. Thus, it appears that expression of the 22.2 antigen is not a necessary condition for formation of lung metastases by 3LL.