Initial Safety and Tumor Control Results From a "First-in-Human" Multicenter Prospective Trial Evaluating a Novel Alpha-Emitting Radionuclide for the Treatment of Locally Advanced Recurrent Squamous Cell Carcinomas of the Skin and Head and Neck.

PURPOSE: Our purpose was to report the feasibility and safety of diffusing alpha-emitter radiation therapy (DaRT), which entails the interstitial implantation of a novel alpha-emitting brachytherapy source, for the treatment of locally advanced and recurrent squamous cancers of the skin and head and neck. METHODS AND MATERIALS: This prospective first-in-human, multicenter clinical study evaluated 31 lesions in 28 patients. The primary objective was to determine the feasibility and safety of this approach, and the secondary objectives were to evaluate the initial tumor response and local progression-free survival. Eligibility criteria included all patients with biopsy-proven squamous cancers of the skin and head and neck with either primary tumors or recurrent/previously treated disease by either surgery or prior external beam radiation therapy; 13 of 31 lesions (42%) had received prior radiation therapy. Toxicity was evaluated according to the Common Terminology Criteria for Adverse Events version 4.03. Tumor response was assessed at 30 to 45 days at a follow-up visit using the Response Evaluation Criteria in Solid Tumors, version 1.1. Median follow-up time was 6.7 months. RESULTS: Acute toxicity included mostly local pain and erythema at the implantation site followed by swelling and mild skin ulceration. For pain and grade 2 skin ulcerations, 90% of patients had resolution within 3 to 5 weeks. Complete response to the Ra-224 DaRT treatment was observed in 22 lesions (22/28; 78.6%); 6 lesions (6/28, 21.4%) manifested a partial response (>30% tumor reduction). Among the 22 lesions with a complete response, 5 (22%) developed a subsequent local relapse at the site of DaRT implantation at a median time of 4.9 months (range, 2.43-5.52 months). The 1-year local progression-free survival probability at the implanted site was 44% overall (confidence interval [CI], 20.3%-64.3%) and 60% (95% CI, 28.61%-81.35%) for complete responders. Overall survival rates at 12 months post-DaRT implantation were 75% (95% CI, 46.14%-89.99%) among all patients and 93% (95% CI, 59.08%-98.96%) among complete responders. CONCLUSIONS: Alpha-emitter brachytherapy using DaRT achieved significant tumor responses without grade 3 or higher toxicities observed. Longer follow-up observations and larger studies are underway to validate these findings.

Treatment of solid tumors by interstitial release of recoiling short-lived alpha emitters.

A new method utilizing alpha particles to treat solid tumors is presented. Tumors are treated with interstitial radioactive sources which continually release short-lived alpha emitting atoms from their surface. The atoms disperse inside the tumor, delivering a high dose through their alpha decays. We implement this scheme using thin wire sources impregnated with (224)Ra, which release by recoil (220)Rn, (216)Po and (212)Pb atoms. This work aims to demonstrate the feasibility of our method by measuring the activity patterns of the released radionuclides in experimental tumors. Sources carrying (224)Ra activities in the range 10-130 kBq were used in experiments on murine squamous cell carcinoma tumors. These included gamma spectroscopy of the dissected tumors and major organs, Fuji-plate autoradiography of histological tumor sections and tissue damage detection by Hematoxylin-Eosin staining. The measurements focused on (212)Pb and (212)Bi. The (220)Rn/(216)Po distribution was treated theoretically using a simple diffusion model. A simplified scheme was used to convert measured (212)Pb activities to absorbed dose estimates. Both physical and histological measurements confirmed the formation of a 5-7 mm diameter necrotic region receiving a therapeutic alpha-particle dose around the source. The necrotic regions shape closely corresponded to the measured activity patterns. (212)Pb was found to leave the tumor through the blood at a rate which decreased with tumor mass. Our results suggest that the proposed method, termed DART (diffusing alpha-emitters radiation therapy), may potentially be useful for the treatment of human patients.

Degradation of cell-bound antibodies by tumor cells from C57BL/6 mice.

A kinetic study of the disappearance of sensitizing antibody from metabolizing EL 4 cells in C57BL/6 mice revealed that this disappearance was at least partly due to the degradation of the antibody into dialyzable products. The degradation took place at 37 degrees C, but not at 4 degrees C, and occurred on or inside the cell. The process was specific inasmuch as only those IgG molecules that could bind to the cells were degraded, whereas unrelated third-party antibodies present in the culture medium remained intact. Although nonmalignant cells also have the capacity to degrade cell-bound antibodies, possibly the degradation of proteinaceous structures on antitumor effector mechanisms is, in part, responsible for decreased antitumor immunity.

Development of resistance to MOPC-315 plasmacytoma after intralesional and intraperitoneal melphalan therapy of tumor-bearing BALB/c mice. II. Enhancement of in vitro cell-mediated cytotoxicity by combined chemotherapy-immunotherapy.

As previously reported, tumor-bearing BALB/c mice can be cured by split-course melphalan therapy, with 40-60% of the treated animals developing resistance to subsequent challenge with viable MOPC-315. The present study deals with the identification of effector-cytotoxic cells that may be developed as a result of chemotherapy-induced tumor regression and their possible potentiation by active, specific immunization with melphalan- and glutaraldehyde-treated MOPC-315 plasmacytoma cells. The cytotoxic potential of spleen-derived lymphocytes in treated animals could be demonstrated only after in vitro sensitization against mitomycin-treated MOPC-315 cells. Lymphocyte-mediated cytotoxicity, as measured against syngeneic 51Cr-labeled MOPC-315, could be detected in melphalan-cured animals and was significantly enhanced by active immunization as compared to the cytotoxicity seen in normal and tumor-bearing mice. With the use of M109 syngeneic, unrelated tumor cells as control targets in the assay, no cytotoxicity was detected. Macrophage cytotoxicity was not significantly enhanced in any of the treatment groups described, with these assays performed 6-8 weeks following treatment and cure. When in vitro killing of MOPC-315 targets was tested with the use of peritoneal macrophages harvested shortly following cure of ascitic tumor by ip injected melphalan, the cytotoxic response was significantly enhanced. In conclusion, following chemotherapy-mediated cure of established MOPC-315 tumors, splenic lymphocytes exhibited enhanced antitumor cytotoxicity, which was further augmented by active immunization. Macrophage activation, as measured by direct cytotoxicity against MOPC-315 targets, was found to occur locally and early following the event of melphalan-induced tumor regression.

Oxidative burst-dependent tumoricidal and tumorostatic activities of paraffin oil-elicited mouse macrophages.

Mouse peritoneal macrophages (MPM) from C57BL/6J, BALB/c, A strains, and (BALB/ cfemale X C57BL/ 6Jmale )F1 offspring were treated with the oxidative burst (OB)-stimulant 12-O-tetradecanoylphorbol 13-acetate, and their in vitro tumoricidal, tumorostatic , and OB activities were examined. Paraffin oil-elicited, but not thioglycollate (TG)-elicited, MPM exhibited cytotoxicity only toward Yac-1 cells and cytostatic activity toward Yac-1, EL 4, RBL-5, and RLmale 1 lymphoma cells. This activity was in correlation with the reduced capacity of TG-elicited cells to generate OB products. The toxic effect of such activated MPM was partially inhibited by catalase, superoxide dismutase, cytochrome c, and vitamin E (alpha-tocopherol) and was augmented by horseradish peroxidase and the catalase inhibitor aminotriazole (3-amino-1H-1,2,4-triazole), thus indicating the involvement of oxygen-derived toxic reagents, mainly hydrogen peroxide, in the MPM-mediated damage inflicted on the tumor cells. EL 4 cells incubated with nonstimulated MPM exhibited enhanced growth both in vitro and in vivo, whereas OB-stimulated MPM inhibited the growth of such cells in the same experimental systems.

Immunopotentiating and immunotherapeutic effects of thymic hormones and factors with special emphasis on thymic humoral factor THF-gamma2.

The essential role played by the thymus in the development of the immune response was well documented in many publications. These findings prompted a long series of studies devised to define the factors produced and secreted by thymus cells, which are involved in the development and nature of immunological responsiveness. First experiments done with crude thymus extracts were followed by isolation of purified products and finally by chemical characterization and synthesis of immunologically active thymus-derived peptides. In this article we review the various thymic hormones and factors described, that is, thymosin fractions 5, the thymosins, prothymosin alpha, thymulin (FTS-Zn), thymopoietin, thymostimulin (TP-1), Thymic humoral factor (THF), and THF-gamma2. Studies demonstrating the activity of the various thymic factors in increasing the immunocompetence potential in both in vitro and in vivo conditions are discussed. The immunostimulatory potential of thymic factors was also investigated in experimental models where beneficial therapeutic effects were sought in a situation of immunological malfunction. The last part of the review is dedicated to clinical trials with thymic factors that revealed improvement in the immunocompetence potential in cases of immunodeficiencies, viral infections, and cancer and its correlation with therapeutic effectiveness. It seems that more research is required in order to better define conditions for the use of thymic factors in immunotherapy.

Inhibition of antibody-dependent stimulation of lipoteichoic acid-treated human monocytes and macrophages by polyglycerolphosphate-reactive peptides.

By itself, lipoteichoic acid (LTA) obtained from S. pyogenes, S. aureus, or E. hirae poorly stimulated cytokine production by macrophages, whereas in the presence of anti-polyglycerol phosphate (PGP), the cells secreted significant amounts of IL-6. Two peptides constructed from the deduced sequence of the selected anti-PGP phage-antibody's complementary-determining region 3 of the variable heavy chain (V(H)-CDR3) reacted specifically with PGP. The monomeric form of the peptides markedly inhibited cytokine production by macrophages pretreated with LTA and anti-LTA. In contrast, the polyvalent form of biotinylated peptides complex with streptavidin-induced cytokine production by the LTA-treated macrophages. The data taken together support the concept that cross-linking of macrophage-bound LTA by anti-PGP is required for cytokine release by these cells. Importantly, these studies identified small, PGP-reactive peptides as potential tools in reducing this proinflammatory process.

Surfactant protein D-coated Klebsiella pneumoniae stimulates cytokine production in mononuclear phagocytes.

Encapsulated Klebsiella pneumoniae strains K21a, K10, and K50, all of which contain dimannose sequences in their capsular polysaccharides that are recognized by the mannose receptor of macrophages, stimulated interleukin secretion and cytokine mRNA expression by human monocyte-derived macrophages. By contrast, the corresponding unencapsulated phase variants and the K2 strain, which lack the dimannose sequence, did not. Coating of unencapsulated phase variants of Klebsiella strains with surfactant protein (SP)-D resulted in marked stimulation of cytokine mRNA accumulation. The induction of cytokine mRNA via the mannose receptor occurred only in monocyte-derived macrophages, whereas that caused by SP-D-coated Klebsiella strains occurred in both macrophages and peripheral-blood monocytes. The results suggested that innate immunity against pulmonary pathogens might be mediated by SP-D, which acts as an opsonin to enhance the interaction of macrophages with unencapsulated phase variants originating from the upper respiratory tract, and by macrophage mannose receptors, which recognize encapsulated variants expressing capsular dimannose residues.

A colorimetric microtiter assay for the quantitation of cytokine activity on adherent cells in tissue culture.

A colorimetric microtiter assay was developed for the quantitation of adherent cells in culture, which is based on the staining of cells with a commercially available staining kit. Adherent L929, A375 cell lines and human monocytes were stained with Hemacolor reagents and the color eluted with SDS 0.5% was determined spectrophotometrically with an ELISA plate reader at 630 nm. The method enabled the detection of cells and was linear up to 3 x 10(4) L929 cells/well. The Hemacolor staining assay was compared to the crystal violet staining assay and the MTT reduction assay, and was found to be sensitive, accurate and reproducible, and has the advantage of enabling microscopic inspection of the stained cells prior to color elution. The assay was found to be suitable for the determination of cytotoxic cytokines, and the enumeration of adherent monocytes. This method might be also used for the quantitation of cytotoxic drugs, and the cytophatic activity of viruses.

A simple colorimetric method for the measurement of hydrogen peroxide produced by cells in culture.

A simple, rapid and inexpensive method for the measurement of hydrogen peroxide (H2O2) produced by cells in culture is described. The assay is based on the horseradish peroxidase (HRPO)-mediated oxidation of phenol red by H2O2 which results in the formation of a compound demonstrating increased absorbance at 610 nm. A linear relationship between absorbance at 610 nm and concentration of H2O2 was found in the 1--60 microM (1--60 nmoles/ml) range. Due to the non-toxic character of phenol red and HRPO, the assay permits measurement of H2O2 production and release by macrophages for time intervals of 5--60 min under regular tissue culture conditions. Using this assay, the ability of a number of agents to induce H2O2 release by guinea pig peritoneal macrophages was demonstrated. These agents were: phorbol myristate acetate (PMA), opsonized zymosan, concanavalin A (Con A), wheat germ agglutinin (WGA), N-formyl-methionyl-leucyl-phenylalanine (FMLP) and A23187.

Increased oxidative burst potential exhibited by macrophages during graft-versus-host reactions in mice.

Graft versus host reactions (GVHR) in mice are accompanied by macrophage activation. Since macrophage oxidative burst (OB) was found to increase in activated macrophages (MPs), we examined the OB of peritoneal and spleen MPs from mice during GVHR. Parental spleen lymphocytes, were injected i.p. into F1 hybrids (BALB/cXC57Bl/6) and at different intervals following injection we monitored the number of peritoneal exudate cells (PEC), spleen enlargement, and the OB of both peritoneal and spleen adherent MPs. Macrophage OB was assessed by measuring O-2 and H2O2 production, following stimulation with the phorbol ester TPA. The spleen-to-body weight ratio increased by 40-70% in mice with GVHR during the period of 6-20 days post-injection and decreased to almost normal levels after 27 days. In such mice with GVHR the number of PEC returned to normal (approximately equal to 6 X 10(6) cells/animal) after 20 days compared with 5 days in control mice (injected with F1 spleen cells). Adherent peritoneal MPs from control mice and mice with GVHR exhibited increased OB activity 3 days following treatment. However, in the control mice the OB response returned to normal within 6 days, while in the experimental mice it showed a slight decrease after 6 days, increased again within 9-17 days, and finally decreased 20-27 days after treatment. Adherent spleen MPs from mice with GVHR generated a high response 6 days posttreatment, which gradually returned to the basal level after 17 days. These findings suggest that during GVHR in mice, macrophage OB was potentiated, giving rise to cytotoxic products such as O2- and H2O2, which may contribute to tissue damage during GVHR.

The effect of cyclosporine on macrophage oxidative burst potential during graft-versus-host reactions in mice.

The development of graft-versus-host reactions in mice was characterized by an increase in activated macrophage populations in the peritoneum and spleen of the animals. In the present study we tested the effect of the immunosuppressive agent CsA on the appearance and activity of such macrophages. Parental spleen lymphocytes were injected intraperitoneally into F1 hybrids (BALB/c x C57Bl/6), and 13 days following injection we monitored the number of peritoneal exudate cells (PEC), spleen enlargement, and the oxidative burst (OB) of adherent peritoneal macrophages (APM). Macrophage OB was assessed by measuring O2- and H2O2 production, following stimulation with 12-O-tetradecanoyl-phorbol-13-acetate (TPA). In GVHR mice a 72% increase in their spleen to body weight ratio was observed, whereas in CsA-treated GVHR mice only a 32% increase was evident. Assessment of the effect of CsA on the number and function of peritoneal cells revealed that CsA caused a 73% reduction in the number of infiltrating PEC in GVHR mice. Furthermore, measurements of H2O2 and O2- production by APM revealed that the overall OB capacity of APM from CsA-treated mice was significantly reduced compared to APM from nontreated GVHR mice. CsA had no effect on the number and OB activity of paraffin oil elicited or resident PEC. These results indicate that CsA may prevent recall and activation of macrophages via its effect on T cell lymphokine release and thus may lessen the contribution of macrophage-derived toxic reagents to the damage inflicted by GVHR.

Differential stimulation of mononuclear phagocyte IL 1 production and oxidative burst by tumor-promoting and non-tumor-promoting agents.

Adherent bone marrow, spleen and peritoneal mouse macrophages, as well as human peripheral blood monocytes were exposed in vitro to the phorbol ester derivatives 12-O-tetradecanoyl-phorbol-13-acetate (TPA), phorbol 13-monoacetate (PA), phorbol 12-myristate (PM), phorbol 12,13-diacetate (PDA), phorbol 12,13 dibutyrate (PDBu), TPA-20 aldehyde (TPA-AL), phorbol 12-retinoate 13-acetate (PRA), 4-alpha TPA (alpha-TPA) and to mezerein (MEZ) and aplysiatoxin (APL). The triggered macrophages/monocytes were tested for the production of an IL 1-like activity by the thymocyte proliferation assay and for H2O2 generation in a quantitative method which is based on the H2O2-mediated and horseradish peroxidase-dependent oxidation of phenol red. The results showed that strong first stage and second stage tumor promoters such as TPA, PDBu, PRA, MEZ and APL are also strong stimulators of IL 1 and H2O2 generation, whereas weak tumor promoters exhibited a low, if any, effect at all. The afore-described findings lend support to the idea that chronic inflammatory phagocytes might play a role in the tumor promoting process by furnishing both carcinogenic and growth factors at the site of tumor origin.

The oxidative burst and related phenomena in mouse macrophages elicited by different sterile inflammatory stimuli.

Mouse peritoneal macrophages (MPM) elicited by paraffin oil (PO) or thioglycollate (TG) were compared in their capacity to generate oxidative burst (OB) products such as superoxide ion (O2-) and hydrogen peroxide (H2O2) and lyse target erythrocytes. Strong OB stimulants such as 12-O-tetradecanoyl-phorbol-13-acetate (TPA), opsonized zymosan; and wheat-germ agglutinin (WGA) were found to stimulate O2 and H2O2 production by both PO- and TG-elicited MPM. PO-elicited MPM exhibited a vigorous OB response compared to TG-elicited MPM and showed a higher degree of self iodination in the presence of 125I. Macrophage-mediated cytolysis of 51Cr-labeled red blood cells, following OB stimulation, was revealed by PO-elicited MPM triggered by TPA, opsonized zymosan, WGA, and Concanavalin A. TG-elicited MPM failed to respond to TPA and opsonized zymosan. In the presence of horseradish peroxidase, both PO- and TG-elicited MPM exhibited augmented cytocidal activities upon stimulation with TPA, whereas catalase abrogated the capacity of TPA-stimulated MPM to lyse red blood cells, which may suggest the involvement of H2O2 in the lytic process.

On the interactions between macrophages and developmental stages of Schistosoma mansoni: the cytotoxic mechanisms involved in macrophage-mediated killing of schistosomula in vitro.

In an in vitro cytotoxicity assay, mouse adherent peritoneal exudate macrophages (APEM), harvested 8-10 weeks post Schistosoma mansoni infection caused sizable (greater than 90%) specific killing of schistosomula. This cidal effect was not diminished by the addition of scavengers of oxidative burst products to the cytotoxicity assay, albeit macrophages from schistosome-infected mice produced more H2O2 than did macrophages from non-infected mice. Of inhibitors of lysosomal enzyme function and release added to the cytotoxicity assay, trypan blue (1 mg/ml) fully abolished the schistosomulicidal effect; hydrocortisone (100 micrograms/ml) was partly effective, and gold salts (1 mg/ml) were ineffective. A cidal effect was not apparent in the absence of L-arginine nor in the presence of excess (greater than 400 micrograms/ml) L-arginine, L-lysine or L-ornithine. Arginase (5 U/ml) totally abrogated the schistosomulicidal effect. The findings suggest that a macrophage protein of a lysosomal origin, dependent on arginine for its reaction and/or production, may be involved in the in vitro killing of schistosomula by macrophages from S. mansoni-infected mice.

IL-1, TNF-alpha and IL-2 production by peritoneal and spleen cells from Schistosoma mansoni infected mice and its potentiation by preimmunization with schistosomal antigens and immunostimulants.

In the present study we tested the effect of immunization with schistosome derived antigens such as frozen-thawed schistosomula in combination with either BCG, liposomes or liposomal muramyl tripeptide-phosphatidyl ethanolamine (MTP-PE), on the resistance of mice to infection, and on the function of their macrophages and lymphocytes. Immunization with either F-T schistosomula + BCG or F-T schistosomula + MTP-PE and subsequent infection, resulted in a 2-3-fold increase in adherent peritoneal macrophage-mediated schistosomulicidal activity (SCA). Peritoneal and spleen macrophages from immunostimulant treated and/or immunized animals showed a significant increase in LPS triggered TNF-alpha production, as compared to non-treated controls. The highest increase in TNF-alpha production was achieved after immunization with either F-T schistosomula + BCG or F-T schistosomula + MTP-PE. LPS triggered IL-1 production was elevated in spleen and peritoneal macrophages from F-T schistosomula + BCG treated mice, and also in spleen macrophages treated with F-T schistosomula + MTP-PE. Only immunization with F-T schistosomula + BCG increased ConA-induced spleen lymphocyte proliferation and IL-2 production. Immunization of mice with F-T schistosomula + BCG also induced protection against parasite infection, while F-T schistosomula + MTP-PE failed to do so. Potentiation of antischistosomal resistance seems to require both macrophage and lymphocyte activation which was achieved only when BCG served as an immunostimulant.

Laser activated fluorescence measurements and morphological features: an in vivo study of clearance time of fluorescein isothiocyanate tagged cell markers.

Fourteen BALB/c mice were divided into two groups. One group served as the control and the second group was injected with a squamous cell carcinoma cell line to the tongue. After tumor development (1-4 weeks), mice were injected with a FITC conjugated CD3 marker to their tongues. Immediately after the marker injection, the clearance of the marker was measured using a laser spectroscopy system. The markers were excited by an argon laser at 488 nm and the fluorescence signal was measured as a function of time. A biopsy was taken from every mouse after the procedure and the excised tissue was histologically evaluated. Analysis of clearance times revealed a second order exponential decay for both groups with a slower pace of signal clearance for the sick mice.

Effects of photodynamic therapy with hypericin in mice bearing highly invasive solid tumors.

The tumoricidal properties of photodynamic therapy (PDT) with hypericin (HY) were evaluated in a highly metastatic adenocarcinoma (DA3Hi) and anaplastic squamous cell carcinoma (SQ2) tumors in vivo. Photosensitization of the tumor site with hypericin (HY-PDT) reduced primary tumor development and significantly prolonged the survival of tumor-bearing (TB) mice. Of these two tumors the squamous cell carcinoma emerged as more sensitive to HY-PDT compared with DA3Hi adenocarcinoma both in vitro and in vivo. HY-PDT caused extensive tumor necrosis that was followed by local, intratumoral, and systemic inflammatory reactions. Analyses of cytokine mRNA profiles reveal increases in mRNA levels of expression confined to inflammation-related cytokines both within the tumor and also systemically (measured in spleens). However, there was no evidence for any HY-PDT-induced antitumoral immune reactions. Our results suggest that PDT with hypericin can be considered as a supplementary treatment in the management of some invasive and metastatic cancers such as squamous carcinoma and similar tumors.

Anti-cancer activities of hypericin in the dark.

The potent photodynamic properties of hypericin (HY) elicit a range of light-dependent virucidal and tumoricidal activities. Yet, a relatively low reduction/oxidation potential endows HY with electron accepting and donating properties enabling it to act as both an oxidizing and a reducing agent. HY can thus compete as an electron acceptor from bioenergized reduction/oxidation reactions generating its excitation energy for biological activities from physiological reduction/oxidation reactions in the absence of light. Our studies show that HY can inhibit the growth of highly metastatic murine breast adenocarcinoma and squamous cell carcinoma tumors in culture. Furthermore, we show that HY can interfere with the growth of these tumors in mice reducing tumor size and prolonging animal survival in complete absence of light. While there is no evidence that HY induces apoptosis in these cells in the dark, 3H-thymidine incorporation into DNA was significantly reduced indicating effects that are apparently cytostatic in nature compared to the cytocidal effects of HY with light.

Thymic humoral factor-gamma 2 augments immune cell response and exerts antitumor activity in murine model systems.

The immunomodulatory and antimetastatic/antitumor activity of thymic humoral factor-gamma 2 (THF-gamma 2) was evaluated in BALB/c-mice. Daily subcutaneous applications (7 consecutive days, 20, 200 ng of THF-gamma 2 per injection/mouse) upregulated counts of thymocytes and peripheral blood cells in tumor bearing mice. To check the influence of THF-gamma 2 treatment on the growth of experimental metastases, RAW 117 H10 lymphosarcoma cells or L-1 sarcoma cells were intravenously inoculated into BALB/c-mice to establish liver or lung metastases, respectively. Local tumor growth was induced by subcutaneous injection of L-1 sarcoma cells. THF-gamma 2 was subcutaneously administered daily for 7 consecutive days starting 24 hrs after tumor cell challenge. Organ colonization as well as local tumor growth were investigated on day 14 after tumor cell inoculation and demonstrated a statistically significant (p < 0.05) reduction of experimental liver and lung metastases and local tumor growth for THF-gamma 2 treated mice.