Immunotherapy of metastases enhances subsequent chemotherapy.

In many multimodal therapies of cancer, postsurgical chemotherapy is administered before immunotherapy for treatment of micrometastatic disease. This sequence may not be the most efficacious. Experiments in which strain 2 guinea pigs bearing syngeneic L10 hepatocarcinomas were given immunotherapy showed that infiltrating immune effector cells not only were tumoricidal but disrupted the characteristically compact structure of metastatic foci. When cytotoxic drugs were administered at the peak of this inflammatory response, the survival rate of the guinea pigs increased significantly. We conclude that postsurgical immunotherapy can enhance the effect of cytotoxic drugs administered subsequently.

Monoclonal antibodies in the lymphatics: selective delivery to lymph node metastases of a solid tumor.

After subcutaneous injection, monoclonal antibodies directed against a tumor can enter local lymphatic vessels, pass to the draining lymph nodes, and bind to metastases there. Lymphatic delivery of antibody to early metastases is more efficient than intravenous administration, and the lymphatic route can be used to image smaller metastatic deposits. Perhaps more important, the lymphatic route minimizes binding of antibodies to circulating tumor antigens and to cross-reactive antigens present on normal tissues. Antibodies inappropriate for intravenous use because of binding to normal tissues may therefore be useful against lymph node metastases when injected subcutaneously or directly into lymphatic vessels.

Mechanism of action of BCG-tumor cell vaccines in the generation of systemic tumor immunity. II. Influence of the local inflammatory response on immune reactivity.

The intradermal injection of a vaccine composed of 10(7) X-irradiated syngeneic hepatocarcinoma line 10 (L10) cells admixed with 10(8) Mycobacterium bovis strain BCG into inbred Sewall Wright strain 2 guinea pigs induces a local inflammatory reaction and effectively immunizes against a contralateral challenge with viable L10 cells. The relationship between the local inflammatory reaction and the generation of tumor immunity was studied. Immunization against L10 was most effective when both L10 cells and BCG were injected into the same site, less effective when they were injected into separate sites with common lymphatic drainage, and not effective when they were injected at totally separate sites. Enzymatic dissociation of dermal vaccination sites revealed that vaccination with BCG and L10 cells combined induced a significantly greater inflammatory response than did vaccination with BCG alone or L10 cells alone; the inflammatory response was also greater than the combination of these individual responses, suggesting that BCG and L10 cells interacted synergistically in the elicitation of an inflammatory response. Sites receiving a combined vaccine of BCG and L10 cells were infiltrated rapidly by inflammatory cells, and surgical excision of these sites as early as 24 hours after vaccine administration did not affect significantly the development of immune responsiveness. However, vaccination sites induced by the injection of BCG and L10 cells at separate but adjacent sites were slowly infiltrated by inflammatory cells, and surgical removal of these sites within 96 hours of vaccination inhibited later immune responsiveness. Quantitative cellular analysis of these inflammatory reactions showed that inflammation was related to tumor-reactive immunity such that the greater the initial inflammatory process, the greater the resistance to tumor challenge.

Mechanism of action of BCG-tumor cell vaccines in the generation of systemic tumor immunity. I. Synergism between BCG and line 10 tumor cells in the induction of an inflammatory response.

The intradermal injection of a vaccine composed of 10(7) X-irradiated, syngeneic hepatocarcinoma line 10 (L10) cells admixed with 10(8) Mycobacterium bovis strain BCG into inbred Sewall Wright strain 2 guinea pigs induced a local acute and then chronic inflammatory response. Cellular analysis of enzymatically dispersed dermal vaccination sites and regional lymph nodes revealed quantitative differences between the cellular infiltrate induced by a mixed BCG-tumor cell vaccine and the inflammation induced by either 10(8) BCG alone or 10(7) tumor cells alone. Analysis of dermal sites from days 1 through 4 following vaccination showed that sites receiving the BCG-tumor cell vaccine contained twofold to fourfold more cells than did those induced by BCG alone and more than fourfold the number of cells induced by L10 cells alone. These results indicated that there was a synergistic interaction between BCG and L10 cells in the induction of an inflammatory response at the site of vaccination. Analysis of the superficial distal axillary lymph nodes draining the site of vaccination revealed a similar synergistic interaction between BCG and L10, such that lymph nodes draining vaccination sites had a greater total cellular content than that induced by either BCG or L10 cells administered separately. When two other tumors, also syngeneic to strain 2 guinea pigs, were tested for their ability to interact synergistically with BCG to elicit an inflammatory response, it was found that the antigenically distinct hepatocarcinoma line 1 tumor also shared this characteristic with L10, whereas the L2C B-cell leukemia did not.

Development of a model of chronic lymphocytic leukemia in inbred strain 2 guinea pigs.

A new transplantable leukemia (KSL) of unknown etiology that arose in a female Sewall-Wright strain 2 guinea pig is described. KSL cells morphologically resembled medium to large lymphocytes, displayed surface Ia antigen and receptors for complement and for the Fc portion of immunoglobulin, and synthesized surface immunoglobulin (IgM). These characteristics suggest a leukemia of B-lymphocyte origin. KSL cells were shown to be sensitive in vivo to both cyclophosphamide and 1,3-bis(2-chloroethyl)-1-nitrosourea; however, neither drug effectively prevented eventual recurrence of the disease. KSL leukemia was also shown to be distinct from another guinea pig lymphatic leukemia (L2C) with respect to cell morphology, antigenicity, and in vivo growth rate. In this last respect, KSL appeared more closely related to the chronic lymphocytic leukemias. Thus KSL is the first chronic lymphocytic leukemia in the guinea pig to be characterized; it is also the only guinea pig model of lymphatic leukemia that is distinct from L2C leukemia currently available for study.

Isolation of tumoricidal macrophages from lung melanoma metastases of mice treated systemically with liposomes containing a lipophilic derivative of muramyl dipeptide.

The systemic administration of multilamellar liposomes--composed of phosphatidylcholine and phosphatidylserine (7:3 mol ratio), containing the immunomodulator, muramyl tripeptide phosphatidylethanolamine (MTP-PE)--into C57BL/6N mice bearing the syngeneic B16-BL6 melanoma was associated with the eradication of spontaneous lung and lymph node metastases. Immunofluorescence and electron microscopic analyses revealed that 24 hours after the tumor-bearing mice were given iv injections of liposomes, 15% of the alveolar macrophages and 5% of the metastasis-associated macrophages contained phagocytosed liposomes. However, only macrophages isolated from lungs or metastases of mice given injections of liposomes containing MTP-PE (treatment success), but not macrophages from mice treated with empty liposomes (treatment failure), were tumoricidal against the target cells in vitro. These data provide direct evidence that the regression of established metastases, after treatment of tumor-bearing mice with liposomes containing MTP-PE, was associated with tumoricidal macrophages residing within the metastases.

Characterization of a murine ovarian reticulum cell sarcoma of histiocytic origin.

We have studied the M5076 tumor, a transplantable murine reticulum cell sarcoma that arose spontaneously in the ovary of a C57BL/6 mouse. This tumor displays functional and ultrastructural characteristics indicating that it is of macrophage origin. Cells from the M5076 tumor are phagocytic, form rosettes with sheep red blood cells, mediate antibody-dependent cellular cytotoxicity against 51Cr-labeled red blood cells, and display macrophage-like cytotoxicity against syngeneic tumor target cells but do not exhibit any natural killer cell activity. The tumor cells possess lysozyme, nonspecific esterase, and phosphatase activities comparable to that seen in rodent macrophages. Ultrastructural examination revealed phagocytic vacuoles and a lack of tight junctions typical of macrophage morphology. Karyotype analysis showed that M5076 tumor cells are hypodiploid with a high percentage (greater than 80%) of metacentric chromosomes that serve as an excellent marker for identification of these tumor cells.

Guinea pig line 10 hepatocarcinoma model: characterization of monoclonal antibody and in vivo effect of unconjugated antibody and antibody conjugated to diphtheria toxin A chain.

Monoclonal antibodies were raised against the guinea pig line 10 (L10) hepatocarcinoma, and an IgG1-producing hybridoma (D3) was selected for further study. D3 is a true monoclonal antibody as demonstrated by two-dimensional gel electrophoresis. Radioimmunoassays on live cells revealed no cross-reactivity with normal tissues or with the line 1 hepatocarcinoma which was used as a control. Membrane immunofluorescence assays demonstrated similar specificity. Immunoperoxidase staining of cryostat sections of tumor and normal tissues of both adult animals and fetuses showed that the D3 monoclonal antibody reacted primarily with the L10 tumor, but some cross-reactivity with smooth muscle, placenta, fetal skeletal muscle, and fetal liver was also demonstrated. Radioimmunoprecipitation of detergent extracts of iodinated L10 cells showed that the antigen is present on the cell surface as a dimer of Mr 290,000 (unit size, Mr 148,000). Therapy studies with unconjugated D3 antibody demonstrated a minor dose-dependent effect on tumor growth. D3 antibody conjugated to the A chain of diphtheria toxin (10(-7) M) was cytotoxic to 100% of L10 cells in vitro. Animals treated with a single 1-mg i.v. injection of this immunoconjugate on Day 7 following the intradermal injection of 10(5) tumor cells demonstrated a highly significant inhibition of tumor growth compared to control animals and those treated with unconjugated antibody.

Prevention of human prostate tumor metastasis in athymic mice by antisense targeting of human angiogenin.

PURPOSE: Angiogenin is a potent positive mediator of neovascularization, a process required for both primary tumor growth and metastasis. In the present study, the effect of a fully phosphorothioated antisense oligodeoxynucleotide, designated JF2S, targeting the AUG translation initiation codon region of human angiogenin, on human prostate tumor development and metastasis in athymic mice was examined. EXPERIMENTAL DESIGN: JF2S was evaluated for its capacity to affect in vitro synthesis of angiogenin and subsequent tumorigenicity of transiently transfected prostate tumor cells in mice. In vivo treatment experiments were then conducted in which JF2S was used to prevent formation of tumors in an ectopic model and metastasis in an orthotopic model. RESULTS: Transient transfection of tumor cells with JF2S inhibited both angiogenin gene expression in vitro and tumorigenicity of these transfected cells in athymic mice. In therapy experiments, local treatment with JF2S completely protected mice from developing prostate tumors after s.c. injection of PC-3 human prostate tumor cells (P < 0.0001, survivor analysis). Most importantly, systemic prophylactic administration of JF2S prevented, in 47% of mice, formation of regional iliac lymph node micrometastases arising from primary tumors growing in the more natural orthotopic prostate setting (P = 0.0003, Fisher's exact test). Furthermore, total protection from regional metastasis occurred in those mice in which JF2S treatment successfully diminished human angiogenin expression in vivo. Tumor-associated angiogenesis was also impaired by JF2S treatment. When therapy was delayed until all of the mice harbored primary tumors in the prostate, the incidence of regional metastasis was still significantly decreased (P < 0.005, survivor analysis). CONCLUSIONS: These findings demonstrate that human prostate cancer establishment and spread in athymic mice is extremely susceptible to targeted disruption of tumor-derived human angiogenin gene expression. Therefore, angiogenin is a valid target against which to devise preventative strategies for prostate cancer metastasis.

Antigen retrieval in formalin-fixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections.

We describe a new approach for retrieval of antigens from formalin-fixed, paraffin-embedded tissues and their subsequent staining by immunohistochemical techniques. This method of antigen retrieval is based on microwave heating of tissue sections attached to microscope slides to temperatures up to 100 degrees C in the presence of metal solutions. Among 52 monoclonal and polyclonal antibodies tested by this method, 39 antibodies demonstrated a significant increase in immunostaining, nine antibodies showed no change, and four antibodies showed reduced immunostaining. In particular, excellent immunostaining results were obtained with a monoclonal antibody to vimentin as well as several different keratin antibodies on routine formalin-fixed tissue sections after pre-treatment of the slides with this method. These results showed that after antigen retrieval: (a) enzyme predigestion of tissues could be omitted; (b) incubation times of primary antibodies could be significantly reduced, or dilutions of primary antibodies could be increased; (c) adequate staining could be achieved in long-term formalin-fixed tissues that failed to stain by conventional methods; and (d) certain antibodies which were typically unreactive with formalin-fixed tissues gave excellent staining.

Macrophages in cancer metastases and their relevance to metastatic growth.

The dissemination of malignant cells throughout the body to form secondary growths is a highly complex process dependent upon both host and tumor cell properties. One potential defensive system which could influence the outcome of metastasis is the mononuclear phagocyte system. Although macrophages have been observed in essentially all primary and metastatic tumors examined regardless of histologic type or anatomic location, the biological significance of these observations is far from clear. What is clear, however, is that in most cases the effects of macrophages on tumor cells are not sufficient to alter progressive growth. Therefore, the presence of macrophages within metastases does not necessarily signify a protective host defensive response. On the other hand, macrophage-induced regression of established metastases can occur in vivo under certain conditions in which tumor-bearing mice are treated systemically with macrophage-activating agents. When mice bearing metastases are treated with macrophage-activating agents contained within liposomes, metastasis-associated macrophages are activated to the tumoricidal state and the metastatic lesions are simultaneously eradicated. Such a process of macrophage activation may have implications in determining successful approaches to the therapy of disseminated cancer.

Induction of syngeneic tumour-specific immunity by liposomes reconstituted with L2C tumour-cell antigens.

Liposomes composed of phosphatidylcholine and phosphatidylserine were reconstituted by detergent dialysis with 3 M K Cl extracts of L2C tumour cells. Liposomes containing L2C antigens were as antigenic as intact tumour cells in the elicitation of delayed-type hypersensitivity responses in strain-2 guinea-pigs previously immunized against L2C tumours. Soluble L2C antigens were devoid of immunoprotective activity, whereas the reconstituted liposomes were capable of protecting animals against up to approximately 100 times the minimal lethal dose of tumour cells. Moreover, the reconstituted liposomes were as antigenic and immunoprotective as viable (irradiated) cells.

Antitumor and phagocytic activities of rat alveolar macrophage subpopulations separated on a discontinuous gradient of bovine serum albumin.

Alveolar macrophages (AM) lavaged from the lungs of normal F344 rats were separated on a discontinuous density gradient of bovine serum albumin (BSA) into four fractions designated as fraction A (20/25% BSA interface), fraction B (25/30%), fraction C (30/35%); and fraction D (35%/pellet). The abilities of these four fractions to form rosettes with opsonized sheep red blood cells (SRBC), to phagocytize these SRBC, and to become tumoricidal in response to macrophage-activating agents in vitro were examined. Fractions A and D had greater abilities than fractions B and C to form rosettes and to phagocytize opsonized SRBC, and a good correlation was found between these two activities in the four fractions. In contrast, the four AM fractions were equally susceptible to activation stimuli, such as lipopolysaccharide (LPS), Nocardia rubra cell wall skeleton (N-CWS), macrophage activating factor (MAF), muramyl dipeptide (MDP), or a mixture of MAF and MDP in vitro to become cytotoxic to syngeneic mammary adenocarcinoma cells.

More on the relevance of animal tumor models: immunogenicity of transplantable leukemias of recent origin in syngeneic strain 2 guinea pigs.

The immunogenicity of four transplantable leukemias was evaluated in syngeneic Sewall-Wright strain 2 guinea pigs. The L76 and KSL leukemias are considered appropriate models for the assessment of immunogenicity because of their recent origin and unknown (natural) etiology. The chemically induced K77 leukemia of recent origin and the long-passaged L2C leukemia of unknown etiology could be classified as experimental tumors prone to artifactual immunogenicity. Our studies clearly show that naturally occurring guinea pig leukemias are potentially immunogenic, although to a lesser degree than experimental leukemias. These findings are in contrast to previous studies in mouse and rat tumor models, which showed that naturally occurring tumors were essentially nonimmunogenic, thereby raising questions about the relevance of animal tumor models to human cancer.

Synergistic effects of active specific immunotherapy and chemotherapy in guinea pigs with disseminated cancer.

Sewall Wright strain 2 guinea pigs bearing pulmonary metastases of the syngeneic line 10 (L10) hepatocarcinoma were treated with a vaccine composed of 10(7) bacillus Calmette-Guérin admixed with 10(7) x-irradiated L10 tumor cells beginning 10 days after tumor inoculation. Although this treatment failed to cure most of the guinea pigs of their metastatic disease, histologic examination of the pulmonary tumors in the vaccinated guinea pigs provided evidence of a cell-mediated hypersensitivity response that disrupted the normally compact architecture seen in control tumors. When a monoclonal antibody against the L10 tumor was injected i.v. to evaluate the vascular permeability of the tumors, significantly more antibody localized in tumors of vaccinated guinea pigs than in tumors of untreated controls. These results suggested that blood-borne substances could be delivered more efficiently to L10 metastases after the tumor-bearing guinea pigs had been treated with vaccine. To determine whether such increased vascular permeability would enhance the antitumor effects of chemotherapeutic agents, combined immunotherapy and chemotherapy studies were performed. Although cyclophosphamide treatment by itself did not cure L10-bearing guinea pigs, cyclophosphamide used in conjunction with prior immunotherapy increased the survival rate of animals to more than twice that of animals treated with immunotherapy alone (74 vs 33%). These results suggest that one mechanism by which active specific immunotherapy enhances chemotherapy of disseminated tumors is by rendering tumor foci more permeable to subsequently administered cytotoxic drugs.

Immunocytochemical and enzymatic detection of lysozyme in human colon carcinoma cell lines.

Six of a total of 14 human colon carcinoma cell lines produce and secrete lysozyme in vitro. Three also produce the enzyme when propagated in vivo in athymic mice. None of the lysozyme positive cells stained in a manner typical of Paneth cells. Additionally, lysozymes from all six colon lines possess identical molecular weights (approximately 14,000 daltons).

Partial purification and characterization of a vascular permeability factor secreted by a human colon adenocarcinoma cell line.

The established human colon adenocarcinoma cell line, HT-29, secretes a vascular permeability factor (VPF) in vitro. The factor has been purified from serum-free conditioned medium by acidification, cation-exchange, and reverse-phase and anion-exchange high-performance liquid chromatography. The VPF is a non-glycosylated acidic protein of apparent molecular weight 45,000, and is at least 4 orders of magnitude more potent than histamine as an inducer of vascular permeability. Its biological activity is unaffected by soybean trypsin inhibitor or by inhibitors of histamine, kinins, prostaglandins, or acid proteases. The VPF induces vascular permeability within minutes, which suggests a direct effect on the endothelial cell.

Angiogenin antagonists prevent tumor growth in vivo.

A noncytotoxic neutralizing monoclonal antibody (mAb), 26-2F, to human angiogenin (Ang), a potent inducer of neovascularization, has been reported to prevent or delay the establishment of HT-29 human tumor xenografts in athymic mice. In the present study the tumor model was modified to increase sensitivity to Ang antagonists to facilitate further investigations and comparisons of their capacity to inhibit tumor growth. An increase in the percentage of tumor-free mice from 10-25% to 65% is observed in this modified model after treatment with mAb 26-2F. An additional neutralizing mAb, 36u, that interacts with a different epitope on Ang similarly prevents the appearance of tumors, both alone and in combination with mAb 26-2F. In those tumors that develop in mice treated with these agents, the number of vascular elements is reduced. Actin, an Ang antagonist that unlike the mAbs binds both human and mouse Ang, also prevents the establishment of tumors while exhibiting no toxic effects at daily doses > 50 times the molar amount of circulating mouse Ang. Ang antagonists also inhibit the appearance of tumors derived from two other Ang-secreting human tumor cell lines--i.e., A549 lung adenocarcinoma and HT-1080 fibrosarcoma. These results demonstrate that inhibition of the action of Ang is an effective therapeutic approach for the treatment of malignant disease.