Therapeutic antitumor efficacy of anti-CD137 agonistic monoclonal antibody in mouse models of myeloma

Abstract

Purpose: Eradication of post-treatment residual myeloma cells is needed to prevent relapses, and immunostimulatory monoclonal antibodies (mAb) such as anti-CD137, CTLA-4, CD40, etc., which enhance the immune response against malignancies, represent a means of achieving this purpose. This study explores anti-CD137 mAbs for multiple myeloma treatment in preclinical models of the disease because they safely augment tumor immunity and are in clinical trials for other cancers.

Experimental design: The antitumor effect of anti-CD137 mAb on mouse plasmacytomas derived from HOPC and NS0 cell lines was studied and compared with that of anti-CTLA-4, anti-CD40, and anti-ICAM-2 mAbs. The antitumor effect of anti-CD137 mAb was also examined in a mouse syngeneic disseminated myeloma (5TGM1) model, which more closely resembles human multiple myeloma. Depletions of specific cell populations and gene-targeted mice were used to unravel the requirements for tumor rejection.

Results: Agonistic mAb against CD137 and blocking anti-CTLA-4 mAb showed activity against i.p. HOPC tumors, resulting in extended survival of mice that also became immune to rechallenge. Anti-CD137 mAbs induced complete eradications of established s.c. NS0-derived tumors that were dependent on IFN-gamma, natural killer cells, and CD8(+) T lymphocytes. Natural killer cells accumulated in tumor draining lymph nodes and showed increased IFN-gamma production. Antitumor efficacy of anti-CD137 mAb was preserved in CD28-deficient mice despite the fact that CD28 signaling increases the expression of CD137 on CD8(+) T cells. Importantly, anti-CD137 mAb treatment significantly decreased systemic tumor burden in the disseminated 5TGM1 model.

Conclusions: The immune-mediated antitumor activity of anti-CD137 mAb in mouse models holds promise for myeloma treatment in humans.

Fig.1. Mice treated with anti-CD137 mAb showed increased survival to HOPC plasmacytoma and acquire long-lasting and tumor-specific immunity. The therapeutic effect of anti-CD137 treatment was independent from direct targeting of HOPC myeloma cells

(A) To compare the relative efficacy of the different immunostimulatory mAbs on the treatment of HOPC tumors, BALB/c mice (16 per group) were i.p. injected with 5×10⁵ HOPC viable cells on day 0, and on days 4 and 7 were i.v. treated with control rat IgG or the indicated mAbs at the dose of 100μg. These mice were examined weekly for palpable abdominal tumors or ascites. Mice survival was plotted using the Kaplan-Meier method and analyzed for significance using the log-rank test. Pooled data from two identical experiments are shown. (B) HOPC tumor cells were evaluated for MHC I, MHC II, CD80, CD86, ICAM-2, CD40, CD137 and CTLA-4 expression by FACS. The grey area represents the fluorochrome-tagged isotype control antibody and the white area represents the relevant antibody. (C) To examine whether a long-lasting and tumor-specific immunity could be generated, mice that had been cured of HOPC tumors with either anti-CTLA-4 or anti-CD137 mAbs were re-challenge 4 month later with HOPC and CT26 cells. Tumor cells (5×10⁵ per mouse) were s.c. inoculated in the opposite flanks of long-term surviving mice from experiment (A) as represented in the scheme. Mice were monitored for tumor growth and compared to naïve age-matched mice. Tumor sizes were assessed by measuring (in millimeters) perpendicular diameters of tumors and the results are expressed as tumor area. These experiments were repeated at least twice yielding similar results. Representative data are shown.

Fig.2. Potent therapeutic effects of the agonistic anti-CD137 mAb in mice bearing NS0 non-immunoglobulin secreting plasmacytomas accompanied by CTL induction;

(A) To evaluate the relative efficacy of anti-CD137 mAb on prolonging survival in of NS0 tumor-bearing mice, NS0 cell (5×10⁵) were i.p. injected into syngeneic BALB/c mice (12 per group) and antibodies (100 μg/mouse) were given i.v. on days 4 and 7 after tumor injection. Survival was plotted using the Kaplan-Meier method and analyzed for significance using the log-rank test. (B) NS0 tumor cells were evaluated for expression of the indicated surface markers by flow cytometry. The grey area represents the isotype control antibody and the white area represents the relevant antibody. (C) Evaluation of therapeutic effect of anti-CD137 mAb assessed on established subcutaneous tumors by sequential measures of tumor areas (fraction of surviving tumor free mice is provided in each graph). BALB/c mice (6 per group) received a s.c. injection of 5×10⁵ NS0 cells on day 0, and on days 9, 11, 13 and 15 mice were treated i.p with mAb 2A or a control IgG at 100μg per injection. Statistical analyses were performed by the T test. This experiment was repeated at least three times yielding similar results. Representative data are shown and compiled data for statistical analysis are presented in a separate graph. (D) Cell suspensions from spleens and tumor draining lymph nodes of mice cured from NS0 sc tumors by anti-CD137 mAb or naïve mice were restimulated in cocultulture with iradiated NS0 cells (1: 25) for 5 days and tested in ⁵¹Cr-release assays (mean±SEM from six independent mice) against CT26 and NSO cells (left) and for upregulation of intracellular IFNγ and surface CD107a in gated CD3+CD8+ splenocytes by FACS (right). Percentage of double positive cells expressed as the mean±SEM from 5-day cocultures prepared from 6 different mice are shown inside the corresponding dot plots.

Fig.3. Absolute requirements of IFN-γ, NK cells, and CD8⁺ T lymphocytes for eradication of NS0 plasmacytomas after anti-CD137 mAb treatment

(A) Involvement of CD4⁺, CD8⁺ T cells and NK cells in the eradication of tumors after anti-CD137 treatment as in Fig. 2C was assessed. BALB/c mice, in groups of 6, bearing s.c. NS0 tumors were injected i.p. with either anti-CD4 or anti-CD8β mAbs or i.v with anti-Asialo GM1 antiserum. A total of 200μg per dose of each mAb were injected into recipient mice for depleting CD4⁺ and CD8⁺ T cells and 50 μl per dose of anti-Asialo GM1 were administered for depleting NK cell. Both CD4 and CD8β specific mAbs and anti-Asialo GM1 antiserum were administered as described in Materials and Methods. Fraction of surviving tumor free mice is provided in each graph. In the graph corresponding to NK cell depletion an inset is provided that shows specific lysis (mean±SEM) in ⁵¹Cr release assays demonstrating the sensitivity of NS0 cells to killing by activated DX5⁺ NK cells isolated from the spleens of Rag1^−/− mice that had been pretreated 18h earlier with 50 μg of poly I:C ip. % of lysis were compared to those achieved against YAC-1 cells and P815 targets. (B) To determine whether IFN-γ was required for eradication of NS0-derived tumors after anti-CD137 mAb treatment, WT or IFN-γ^−/− (IFN-γ K.O.) BALB/c mice (6 per group) were inoculated s.c. with 5×10⁵ NS0 viable cells and then treated with either anti-CD137 mAb or control antibody. Alternatively, tumor-bearing mice (n = 6) were treated with anti-CD137 mAb, and were subsequently given 200 μg of neutralizing anti-IFN-γ as described in Materials and Methods. Statistical analyses were performed using the t test. These experiments were performed at least twice, yielding similar results. Representative data are shown and compiled data for statistical analysis are presented in separate graphs for A and B.

Fig.4. NK cells are increased and activated in tumor draining lymph nodes (TDLNs) while CD8 T cells predominate in the tumor rejecting infiltrates

(A) BALB/c mice (6 per group) received a s.c. injection of 5×10⁵ NS0 viable cells on day 0, and on days 9, 11, 13 and 15 were treated i.p. with either anti-CD137 mAb or a control IgG at 100μg per injection. The average percentages and total cell numbers of NK cells (DX5⁺ CD3⁻ cells) in TDLN were determined by flow cytometry. Pooled data from 2 experiments are shown. Statistical analyses were performed by the T test. (B) NK cells were evaluated for CD69 expression by flow cytometry. Histograms show CD69 expression on this cell subset. The percentages ± SEM of positive cells are indicated. Data are representative of 2 independent experiments. (C) Intracellular expression of IFN-γ by NK cells recovered from TDLNs of either control IgG or anti-CD137-treated-tumor-bearing mice was examined. The mononuclear cells from TDLNs were stimulated with PMA/Ionomycin in vitro for 5 hours and subsequently stained for intracellular IFN-γ. The percentages ± SEM of NK cells producing IFN-γ are indicated. In panels (B) and (C) shaded histograms represent isotype control antibody and white histograms represent the relevant antibody. (D) percentage± SEM of CD4 (CD3+CD4+), CD8 (CD3+CD8+) and NK cells (CD3−DX5+) in lymphoid cell suspensions obtained from disaggregated NS0 tumor nodules. The lesions were excised on day 19 from mice that had been treated with anti-CD137 mAb or control antibody on days 9,11,13,15 after tumor cell injection. Absolute numbers also showed clear increases in intratumoral CD8 T cell counts and decreases in CD4 T cell counts (data not shown).

Fig. 5. CD28 signals up-regulate CD137 expression on CD8⁺ T cells but anti-CD137 mAb treatment of NS0 tumors is independent from CD28 function

(A) Dot plot analysis of CD137 expression on CD8⁺ splenocytes after 96 h in vitro activation with anti-CD3 mAb or anti-CD3 + anti-CD28 as indicated. (B) Comparison of subcutaneous growth of individual tumors derived from NSO treated either with rat IgG or anti-CD137 mAb (100μg on days 9, 11, 13, 15) in Balb/c WT mice or in CD28^−/− mice as indicated in the figure the fraction of tumor-free surviving mice is provided. Representative results from two similarly performed are shown.

Figure 6. Anti-CD137 mAb treatment significantly decreases tumor burden in a disseminated multiple myeloma model

To investigate the anti-tumor efficacy of anti-CD137 mAb in a model of myeloma with widespread skeletal involvement, 5TGM1-GFP myeloma cells (10⁶) were i.v. injected into syngeneic C57BL/KaLwRijHsd mice (≥ 8 per group) and then randomly assigned to four groups which received either vehicle, bortezomib, control rat IgG or anti-CD137 mAb for 4 weeks (see Materials and Methods for dosing protocol). (A) Fraction of mice with detectable EGFP⁺ foci in the indicated organs of mice immediately after sacrifice on day 30. Tissues were optically imaged, as described under Materials and Methods, in antero-posterior (A-P) and postero-anterior (P-A) orientations and then scored. Data represent the % of mice in each group with ≥ 2 fluorescent foci in both orientations in ≥3 bones. (B) Representative pictures of green fluorescence emitted (A-P and P-A views) of mice from (A) that had been eviscerated during necropsy and the corresponding pictures form the explanted organs. In the case of anti-CD137 mAb, the mice photos shown are from representative animals displaying either fluorescence under the threshold of detection (representing completely eradicated tumor/no residual disease) or reduced but detectable fluorescence (representing some residual disease) as indicated in the figure (c) Serum concentrations (mean ± SEM) of the 5TGM1 monoclonal paraprotein (IgG_2bκ) from mice in the different treatment groups measured 30 days after tumor cell inoculation.