Therapy of colorectal carcinoma with monoclonal antibodies (MAb17-1A) alone and in combination with granulocyte monocyte-colony stimulating factor (GM-CSF)

Addition of GM-CSF to trastuzumab stabilises disease in trastuzumab-resistant HER2+ metastatic breast cancer patients

Background:

One of the proposed mechanisms of trastuzumab-induced regression of human epidermal growth factor receptor 2-positive (HER2+) tumours includes facilitation of antibody-dependent cell-mediated cytotoxicity (ADCC). Granulocyte-macrophage colony-stimulating factor (GM-CSF) mediates ADCC. We presented our pilot study of adding GM-CSF to trastuzumab in patients with trastuzumab-resistant HER2+ metastatic breast cancer.

Methods:

Patients with HER2+ metastatic breast cancer that progressed after trastuzumab +/− chemotherapy were continued on trastuzumab 2 mg kg^–1 intravenous weekly and GM-CSF 250 μg m^–2 subcutaneous daily. Patients were assessed for response every 8 weeks. Treatment was continued until disease progression or intolerable toxicity.

Results:

Seventeen patients were evaluable (median age 48 years, range 27–75 years). The median number of metastatic sites was 2 (range 1–3); the most common site was the liver (n=10). The median number of prior regimens for metastatic disease was 2 (range 1–5). No objective disease response was observed, but five patients (29%) had stable disease for a median duration of 15.8 (range 10–53.9) weeks. The most common adverse event was rash at the injection site. No grade 4 or irreversible adverse event was seen.

Conclusion:

The addition of GM-CSF to trastuzumab alone had a modest clinical benefit and acceptable safety profile in heavily pretreated patients with trastuzumab-resistant HER2+ metastatic breast cancer.

G- and GM-CSF in oncology and oncological haematology

Administration of G- and GM-CSF increases the neutrophil counts in a number of clinical situations. GM-CSF shows the additional effect of increasing the number of monocytes and eosinophil granulocytes. Both G- and GM-CSF affect of neutrophil functions, in the case of GM-CSF there are some potentially negative effects on neutrophil migration and adhesiveness. The clinical relevance of the various effects on mature haematopoietic cells is not fully understood. Clinical data with G-CSF treatment indicate that increased levels of neutrophil granulocytes following cytotoxic chemotherapy may translate into clinical benefit such as a decreased rate of neutropenic infection and an increased cytotoxic chemotherapy dose even though the data are conflicting and the risk of "laboratory cosmetics" is apparent. Regarding treatment with GM-CSF following chemotherapy, the clinical benefit is unclear. The clinical benefit of GM-CSF-induced monocytes and eosinophils is unknown. G- and GM-CSF accelerates neutrophil recovery following autologous or allogeneic BMT. The influence on neutropenic infections is, however, less impressive. Pretreatment with G- or GM-CSF increases the yield of peripheral stem cell harvest, thereby reducing the number of leukaphereses needed. Transplantation of G- and GM-CSF primed autologous peripheral stem cells tends to reduce the period of post-transplant cytopenia, particularly thrombocytopenia, in comparison with traditional ABMT. In patients with MDS, G- and GM-CSF appear to increase the number of neutrophil granulocytes and there is some evidence that patients with severe infectious problems will benefit from this treatment. However, little influence was seen on the main clinical problems with these patients, which are anaemia and thrombocytopenia. In conclusion, G- and GM-CSF are two different proteins with different properties in vivo and in vitro. GM-CSF has, compared with G-CSF, more complex pharmacological effects and a more trouble-some side-effect profile. Early clinical development indicates that both compounds have a substantial influence on the levels of certain blood cells. Whether the increases in different blood cells translate into long-term clinical benefit for greater patient groups is the focus of ongoing research. The effects of G- and GM-CSF may be potentiated by other cytokines, an area which is presently being explored.

Presentation of native TROP-2 tumor antigens to human cytotoxic T lymphocytes by engineered antigen-presenting cells

Professional antigen-presenting cells (APC), e.g. dendritic cells, express immuno-proteasome components and process proteins for MHC presentation differently from non-immune cells. Thus, they induce reactivities against sets of peptides that do not overlap with those generated by non-professional APC, i.e., tumor cells, and stimulate cytotoxic T lymphocytes (CTL) that may not recognize them. The goal of this work was to establish a system for antigen presentation and in vitro stimulation of human CTL using "tumor-cell-like" engineered APC. Murine fibroblasts were transfected with human HLA Class I alleles, together with the B7.1, ICAM-1 and germ-line TROP2 genes. The last encodes a cell surface glycoprotein widely expressed by human cancers. Unseparated peripheral blood mononuclear cells from HLA Class I-matched individuals were stimulated in vitro by the engineered APC. These efficiently induced the activation and proliferation of antigen-specific HLA-restricted CTL lines and clones. The Trop-2-specific CTL demonstrated high specific cytotoxicity against the appropriate transfected target cells. They also efficiently lysed MCF-7 human tumor cells expressing endogenous HLA-A2.1, Trop-2 together with ICAM-1. These results demonstrate that Trop-2 is a target molecule recognized by human CTL. Moreover, they demonstrate that non-immune engineered APC efficiently process and present native tumor-specific proteins in the context of human MHC Class I, and stimulate the growth and cytotoxicity of specific anti-tumor CTL.

Histamine as an adjunct to immunotherapy

Interleukin-2 and interferon-alpha have been used as therapeutic options in the treatment of certain malignancies such as metastatic malignant melanoma, acute myelogenous leukemia, and renal cell carcinoma. However, the outcome with these agents has been less than optimal. While experiments in vitro would lead one to believe that these agents would be useful therapeutic alternatives, the situation in vivo is confounded by the fact that the microenvironments of the tumor and surrounding tissue are infiltrated with monocytes and macrophages, which suppress the cytotoxic activity of T cells and natural killer cells. The mechanism by which this occurs is through the generation of reactive oxygen species that are responsible for apoptosis by both T cells and natural killer cells. Histamine abrogates this suppression, thus restoring the cytotoxicity of these cells. Therefore, the addition of histamine to regimens containing cytokines is expected to optimize cytokine therapy. Clinical trials with these regimens are under way in the treatment of metastatic malignant melanoma, acute myelogenous leukemia, and renal cell carcinoma. Results published thus far indicate that the addition of histamine to cytokine therapy is both safe and efficacious in the treatment of these diseases.

Histamine: a novel approach to cancer immunotherapy

The functions of intratumoral lymphocytes in many human malignant tumors are inhibited by reactive oxygen species (ROS), generated by adjacent monocytes/macrophages (MO). In vitro data suggest that immunotherapeutic cytokines such as interleukin-2 (IL-2) or interferon-alpha (IFN-alpha) only weakly activate T cells or natural killer (NK) cells in a reconstituted environment of oxidative stress and that inhibitors of the formation of ROS or scavengers of ROS synergize with IL-2 and IFN-alpha to activate T cells and NK cells. In this review, we focus on the immunoenhancing properties of histamine, a biogenic amine. Histamine inhibits ROS formation in MO via H2-receptors; thereby, histamine protects NK cells from MO-mediated inhibition and synergizes with IL-2 and IFN-alpha to induce killing of NK cell-sensitive human tumor cells in vitro. Histamine also optimizes cytokine-induced activation of several subsets of T cells by affording protection against MO-inflicted oxidative inhibition. The putative clinical benefit of histamine as an adjunct to immunotherapy with IL-2 and/or IFN-alpha is currently evaluated in clinical trials in metastatic malignant melanoma and acute myelogenous leukemia.

Immunomodulation therapy in colorectal carcinoma

There has been much progress in the understanding of the relationship between the immune system and colorectal cancer. This has led to the use of immunomodulatory therapy in the adjuvant and palliative treatment of the condition. Although attempts at the use of non-specific immunomodulation with agents such as levamisole, cimetidine, alpha interferon and Bacillus Calmette-Guerin (BCG) have not produced significant clinical benefits when tested in randomized trials in both the adjuvant setting and for metastatic disease, promising results are being obtained with more specific therapy. Edrecolomab [corrected], a murine monoclonal antibody targeting the 17-1A antigen on malignant colorectal cells has produced a reduction in relapse and mortality rates when used as adjuvant treatment following surgery for Dukes' C colon cancer. Active specific therapy with autologous tumour vaccine administered with BCG has produced similar benefits in Dukes' B cancer. Both 3H1 anti-idiotypic antibody against carcinoembryonic antigen and 105AD7 antibody to gp72 glycoprotein have demonstrated in-vitro and in-vivo immune activation against tumour. Non-randomized studies postulate prolongation of survival using these antibodies in advanced disease. These agents are all currently being tested in randomized studies powered to detect meaningful survival differences and clinical benefit. Immune therapy offers the potential of low toxicity therapy in colorectal cancer and may have a role as an adjunct to conventional chemotherapy.

Histamine and cytokine therapy

Interleukin-2 (IL-2) and interferon-alpha (IFN-alpha) are potent activators of natural killer (NK) cells and other anti-tumor effector cells, but the results obtained in clinical trials with these cytokines have proved disappointing in many forms of cancer. It may be that IL-2 and IFN-alpha are often not sufficiently effective because intratumoral monocytes/macrophages (MO) inhibit the cytokine-induced activation of cytotoxic effector lymphocytes such as NK-cells at the site of tumor growth. An essential part of this inhibitory signal is conveyed by MO-derived reactive oxygen species (ROS), which potently inhibit NK-cell-related functions, including the constitutive and cytokine-induced cytotoxicity against tumor cells. Histamine, a biogenic amine, inhibits ROS formation in MO; thereby, histamine synergizes with IL-2 and with IFN-alpha to induce killing of NK-cell-sensitive human tumor cells in vitro. Furthermore, treatment of tumor-bearing mice with histamine potentiates cytokine-induced killing of NK-cell-sensitive murine tumor cells in vivo. In ongoing clinical trials, histamine has been added to IL-2 or IFN-alpha in immunotherapy of human neoplastic disease. The results of two pilot trials in metastatic melanoma suggest that the addition of histamine to IL-2/IFN-alpha prolongs survival time and induces regression of tumors, such as liver melanoma, which are considered refractory to immunotherapy with IL-2 or IFN-alpha. In acute myelogenous leukemia (AML), histamine and IL-2 have been given in order to protect patients in remission against relapse of leukemic disease. The potential benefit of histamine therapy in melanoma and AML will be evaluated in randomized trials.

Granulocyte-macrophage colony-stimulating factor inhibits tumor growth during the postoperative period

BACKGROUND

Granulocyte-macrophage colony-stimulating factor (GM-CSF) may have important antineoplastic properties because it induces macrophage tumoricidal activity in vitro. We examined the inhibitory effect of GM-CSF on tumor growth in a murine carcinoma model and whether this inhibitory effect would persist during the postoperative period. Potential macrophage-mediated mechanisms were studied.

METHODS

The effect of GM-CSF on macrophage function in vitro was assessed by measuring superoxide anion and interleukin-6 production, percentage phagocytosis of Candida albicans, and percentage Ia expression. GM-CSF's effect on tumor volume was assessed first in a murine tumor model and second to examine whether these effects also occurred during the postoperative period in the same model after laparotomy. Macrophage function in the latter study was assessed by measuring superoxide anion, cytotoxicity, and tumor necrosis factor production.

RESULTS

GM-CSF treatment was associated with a decrease in tumor volume on day 4 after the initiation of GM-CSF treatment (0.93 +/- 0.08 cm3 for control versus 0.34 +/- 0.08 cm3 for GM-CSF; p < 0.05). This effect was also seen after laparotomy (1.07 +/- 0.2 cm3 for laparotomy+saline versus 0.16 +/- 0.04 cm3 for laparotomy+GM-CSF, p < 0.05). In vivo macrophage function showed increased superoxide anion, cytotoxicity, and tumor necrosis factor-alpha production from macrophages obtained from GM-CSF treated animals compared with saline treated controls.

CONCLUSIONS

Tumor growth is inhibited by GM-CSF treatment, and this effect also occurs after laparotomy. Thus, GM-CSF may have a therapeutic role in the treatment of the tumor bearing host after operation.

Cytotoxic lymphocyte-mediated immunotherapy

Cytotoxic CD8+ T cells and NK cells play a role in the elimination of some viruses, graft rejection, antitumour responses, immunoregulation and some autoimmune diseases. The central importance of these cells in each of these immune responses and the therapeutic potential they offer, when effectively targeted, has justified continued interest in their function. Molecular biology has dominated the recent study of cytotoxic lymphocyte function, allowing the characterisation of recognition structures on cytotoxic lymphocytes, the definition of two distinct mechanisms of cytotoxicity and the determination of their relevance in vivo. Biological and genetic experimental approaches which exploit the targeted cytolytic activity of lymphocytes are now being developed for immunotherapy. A greater knowledge of the biology of cytotoxic lymphocytes when adoptively transferred, the development of engineered monoclonal antibodies and the characterisation of novel endogenous tumour cell antigens, has us on the brink of using these cells to greater therapeutic advantage. This article reviews ongoing efforts to characterise the mechanism of action of cytotoxic lymphocytes and outlines the progression of approaches designed to enhance the anti-tumour activity of these cells.

Chemotherapy and immunotherapy of colorectal cancer

More than 50% of the patients with large bowel cancer develop disseminated disease and invariably succumb. Adjuvant chemotherapy with 5-FU and levamisole have been shown to be more efficient than 5-FU alone or in combination with cytostatics. The combination of 5-FU, leukovorin and methotrexate induces prolonged survival with a good quality of life in metastatic colorectal cancer (CRC). During the last decade tumor immunotherapy has been an alternative facilitated by isolation and large scale production of cytokines and monoclonal antibodies. The mouse monoclonal antibody (MAb) 17-1A recognizes a tumor-associated antigen (TAA), present in high concentrations on the surface of gastrointestinal tumor cells. Injections of MAb 17-1A in patients with metastatic CRC induced generation of anti-idiotypic (ab2) in 90% and anti-anti-idiotypic (ab3) antibodies in 47% of the treated patients. The development of ab3 correlated significantly with survival (mean 80 weeks) while ab3- patients survive only 38 weeks. One of 52 patients treated with MAb 17-1A is a complete remission after 66 months, 3 had minor regression and 6 had a stable disease (19% RR). Based on in vitro findings showing increased antibody-dependent cellular cytotoxicity (ADCC) by the combination of granulocyte-macrophage colony stimulating factor (GM-CSF) and MAb 17-1A, 16 CRC patients have been treated with subcutaneously injections of GM-CSF for 10 days and intravenous infusions of MAb 17-1A at day 3. Two of 16 are in CR, 1 in MR and 3 in SD (37.5% RR). Minor side-effects were registered. A further development of immunotherapy of CRC might imply vaccination by injection of specific human anti-idiotypic antibodies (ab2) which mimics the nominal antigen, in order to induce a specific immunity.