The anti-mesothelin monoclonal antibody amatuximab enhances the anti-tumor effect of gemcitabine against mesothelin-high expressing pancreatic cancer cells in a peritoneal metastasis mouse model

Pancreatic cancer often has a very poor prognosis, even after complete resection. The recurrence of hepatic and peritoneal metastases is an important prognostic factor; therefore, the development of improved adjuvant therapy is urgently required. Mesothelin is a cell surface glycoprotein whose expression is restricted to a variety of cancer types, including pancreatic cancer. This expression pattern makes mesothelin an attractive target for cancer therapy, and several agents targeting mesothelin are currently in clinical trials. Here, we used the chimerized high-affinity anti-mesothelin monoclonal antibody amatuximab to investigate its effect on peritoneal metastasis. We used the AsPC-1 pancreatic cancer cell line engineered to express Gaussia luciferase (Gluc), (AsPC-1-Gluc) for in vivo experiments. Results showed that while amatuximab was not directly cytotoxic on an AsPC-1-Gluc tumor cells in a peritoneal metastasis model, it prevented the formation of tumor growth. In combination therapy with gemcitabine, amatuximab exhibited synergistic killing. Our results suggest that blockade of mesothelin by amatuximab may be a useful strategy for preventing the peritoneal dissemination of pancreatic cancer under an adjuvant setting.

Immunofluorescence and Flow Cytometry (FCM). The techniques of immunofluorescence staining, FCM and analysis, and FCM data analysis have been described in detail previously (6). Immunofluorescence was measured by FCM analysis using the Los Alamos National Laboratories flow system.
Cell Lines, The murine myelomonocytic leukemia, WEHI-3, has been carried in suspension culture in our laboratory for several years and was originally developed in the BALB/c mouse at the Walter and Eliza Hall Research Institute, Melbourne, Australia (7). The IL-2-sensitive indicator cell, HT-2, has been described previously (8). The origin, MHC restriction, and antigen specificity of the two T-T hybridomas used as responder cells in this study have been described in detail elsewhere (2,9). 1 Briefly, AODK-3.4 produces IL-2 only in the presence of ovalbumin (OVA) and an accessory cell that is I-A ~ positive; AODH-7.1 responds to human gamma globulin (HGG) plus an accessory cell that is positive for the LE d cistron product. All the T-T hybridoma cell lines were provided by Dr. J. Kappler and Dr. P. Marrack.
Accessory Cell Function Assay System. The assay system used to measure accessory cell function has been described in great detail elsewhere (2)J Briefly, the assay consists of placing 2 X 105 T-T hybrid cells together with various tests populations of accessory cells, with or without antigen. The T-T hybrid cells will generate IL-2 only in the combined presence of a syngeneic accessory cell and the appropriate specific antigen. After 24 h, the culture supernatants of each test well were titrated by twofold serial dilution to a final dilution of 1/64. An IL-2-sensitive indicator cell, HT-2, was added to each well, and proliferation was assayed visually or by thymidine incorporation.
Ia Induction. The supernatant used to induce cell surface Ia on WEHI-3 was generated by treating rat spleen cell suspension cultures with a 48-h pulse of concanavalin A (Con A), as previously described (10). This initial Con A-induced supernatant was diluted to a 50% concentration in Duibecco's minimum essential medium (DMEM) (buffered for 12% CO2) + 10% fetal calf serum (FCS) (referred to as rat Con A S/N). WEHI-3 was seeded in this media at a concentration of 1 × 105 cells/ml to 2 × l0 s cells/ml and incubated for 48 h at 37°C. Subsequently the cells were washed three times in sterile basic salt solution and recuhured at 2 × 10 cells/ml in factor-free DMEM 10% FCS for 16-24 h. Cells were washed once and used in the T-T hybrid assay.

Results and Discussion
FCM Analysis ofla-induced WEHI-3. Using indirect immunofluorescence and quantitative flow cytometry analysis, we examined WEHI-3 cells for cell surface expression of MHC gene products. As shown in Fig. 1, WEHI-3 tumor cells showed minimum reactivity with the anti-Ia antibodies (panel A, B), but did react significantly with antibodies directed against H-2D a and H-2K a, In contrast, incubation of WEHI-3 in the presence of rat Con A S/N for 48 h induced cell surface Ia determinants, with >90% of the cell population showing positive staining for the I-A and the I-E/C gene products (panels E, F). As shown in panels G and H, the amount of cell surface H-2K a and H-2D a determinants is also significantly increased. Quantitative comparison of the induced vs. uninduced cells indicated an ~3.3-fold increase in surface H-2K d determinants and a 2.5-fold increase in H-2D d determinants. In contrast to the results obtained with rat Con A S/N, appropriate control supernatants from various Con Ainduced, murine T cell tumor preparations consistently failed to elicit Ia induction on WEHI-3. Continual presence of the "Ia inducing factor" (IaIF) is required to maintain the In-positive phenotype; the induced Ia expression is significantly reduced on tumor cells incubated in factor-free media for periods >24 h. More detailed kinetic studies of Ia expression are being pursued using quantitative FCM analysis and metabolic labeling of Ia determinants. Preliminary experiments indicated that P388.DI and WEHI-265 (Abelson virus-induced myelomonocytic leukemia) are also capable of expressing Ia, using this induction procedure. In contrast, certain other lines, such as PU5.1R, WEHI-274, and certain sublines of WEHI-3, did not respond (data not shown). Further studies with these tumors are in progress. Coulter Volume to Fluorescence Intensity Analysis. After treating the WEH]-3 macrophage tumor cell line with rat Con A S/N, it was noted that the cell line seemed to go into proliferative arrest; the doubling time increased from 16 to 48 h, and the cells become visably larger by microscopic examination and coulter volume determination. Therefore, it was of concern to see whether the increase in Ia-positive cells was simply related to an overall increase in cell volume upon treatment with the factor source. Thus, it was possible that induced cells might express more total Ia antigen per cell because of a greater cell surface area, and, as a consequence, might be detected as Ia positive, whereas the smaller uninduced cells might lack sufficient total Ia antigen to be detected. To test this possibility, the relationship of cell volume of fluorescence intensity was determined, using the flow systems of the Los Alamos National Laboratories. This type of analysis has been described in detail elsewhere (4). Fig. 2 shows the mean fluorescence, in volts, plotted vs. the coulter channel fraction. Panels A and B clearly show that there is a dramatic increase in I-A, I-E/C, and H-2K a determinants on induced WEHI-3 cells, compared with uninduced controls for cells in all the coulter channel fractions. Panels C and D show that there is a smaller but significant increase in H-2D d determinants and surface antigens recognized by a macrophagespecific monoclonal antibody, MAC-1, in all coulter channel fractions. Thus, increase in cell surface MHC determinants, as measured by flow cytometry, is not the result of simple increases in cell surface, but indicates increased surface density of these specificities. Similar analysis using monoclonal antibodies against other non-MHCrelated determinants, such as Ly-5, showed no significant cell surface change in density for these markers (unpublished data). Because there was a profound, proliferative arrest observed concomitant with the Ia induction phenomenon, we analyzed the DNA profile of IaIF-treated WEHI-3 cells, using mithramycin staining and flow cytometry analysis, as described previously (6). These studies show that there is a significant loss of cells in the S phase and accumulation of cells in G1 and G2/M. Preliminary experiments using fractions of isoelectric focusing gel separation of rat Con A S/N indicate that the activity responsible for Ia induction is not the activity causing the proliferation arrest (unpublished data).
Accessory Cell Function Analysis. The two representative exDeriments shown in Table   I clearly demonstrate that Ia-induced WEHI-3 activates AODH-7.1 to produce IL-2 in an antigen-dependent fashion, although it does not stimulate the I-Ak-specific cell line AODK 3.4. Further, uninduced WEHI-3 has little or no effect in triggering IL-2 production from AODH-7.1. Approximately 60% of the IL-2 response of AODH-7.1 can be blocked by treating WEHI-3 with anti-Ia.7 monoclonal antibody; the anti-Ia.15 monoclonal reagent has virtually no effect. Blocking the WEHI-3 tumor line with A.TH anti-A.TL completely abrogates the IL-2 production by AODH-7.1, whereas NMS has no effect on the response. We are also presently engaged in  The data presented in this preliminary study clearly demonstrate that selected murine tumors of the monocyte-macrophage lineage can be induced to express serologieally defined, cell surface Ia determinants. The expression of these determinants is paralleled by the ability of the Ia-induced macrophage tumor line to provide accessory cell function in at least one type of MHC-restricted, antigen-dependent in vitro assay system. Further, the accessory cell function can be partially or totahy blocked by various anti-Ia antibodies. Although these observations do not prove any necessary causal relationships between the appearance of cell surface Ia and accessory cell function, they are consistent with the notion that ta molecules play a central role in antigen processing and presentation. This study may provide the first epportunity to initiate the analysis of the mechanism of accessory cell function, using defined, homogeneous populations of tumor ceils of the monocyte-macrophage compartment as antigen-presenting cells.

Summary
This study demonstrates that an uncharacterized soluble factor produced in concanavalin A-induced rat spleen cell suspensions has the capacity to induce the