Adoptive transfer of anti-CD3-activated CD4+ T cells plus cyclophosphamide and liposome-encapsulated interleukin-2 cure murine MC-38 and 3LL tumors and establish tumor-specific immunity.

The infusion of anti-CD3-activated murine T cells plus interleukin-2 (IL-2) exerts antitumor effects against several tumors in murine immunotherapy models. This study compares the therapeutic efficacy of anti-CD3-activated CD4+ or CD8+ T-cell subsets, when given with cyclophosphamide (Cy) and liposome-encapsulated IL-2 (L-IL2) in a murine model. C57BL/6 mice bearing subcutaneous (S.C.) MC-38 colon adenocarcinoma, 3LL Lewis lung carcinoma, or 38C13 lymphoma for 7 to 14 days were pretreated with low-dose intraperitoneal (I.P.) Cy before intravenous (I.V.) injection of anti-CD3-activated T cells or T-cell subsets. Cell administration was followed by I.P. administration of L-IL2 for 5 days. Mice receiving activated CD4+ T cells showed significantly reduced tumor growth or complete remissions with prolonged disease-free survival in MC-38, 3LL, and 38C13. The timing of Cy doses in relation to adoptive transfer was critical in obtaining the optimal antitumor effect by CD4+ cells. Injecting Cy 4 days before the infusion of CD4+ cells greatly enhanced the antitumor effect of the CD4+ cells and improved survival of the mice compared with other Cy regimens. C57BL/6 mice cured of MC-38 after treatment with CD4+ T cells developed tumor-type immunologic memory as demonstrated by their ability to reject rechallenges with MC-38, but not 3LL. Similarly, mice cured of 3LL tumors rejected rechallenges of 3LL, but not MC-38. The immunologic memory could be transferred with an I.V. injection of splenocytes from mice cured of MC-38 or 3LL. No cytotoxic T-lymphocyte activity was detected in T cells or T-cell subsets from mice cured of MC-38 or 3LL. Increased IL-2 and interferon-gamma (IFN-gamma) production was observed from CD4+ subsets in cured animals when stimulated in vitro with the original tumor, but not with an unrelated syngeneic tumor. These results suggest that tumor-specific immunity can be achieved in vivo with anti-CD3-stimulated CD4+ T cells in this cellular therapy model.

Antitumor response to recombinant murine interferon gamma correlates with enhanced immune function of organ-associated, but not recirculating cytolytic T lymphocytes and macrophages.

The mechanism of therapeutic activity for recombinant murine interferon-gamma (rMu IFN gamma) in the treatment of metastatic disease was investigated by comparing effector cell augmentation with therapeutic activity in mice bearing experimental lung metastases (B16-BL6 melanoma). Effector cell functions in spleen, peripheral blood, and lung (the tumor-bearing organ) were tested after 1 week and 3 weeks of rMu IFN gamma administration (i.v. three times per week). Natural killer (NK), lymphokine-activated killer (LAK), cytolytic T lymphocyte (CTL) activities against specific and nonspecific targets, and macrophage tumoristatic activity were measured. rMu IFN gamma demonstrated immunomodulatory activity in most assays of immune function. The optimal therapeutic protocol of rMu IFN gamma (2.5 x 10(6) U/kg, three times per week) prolonged survival and decreased the number of pulmonary metastatic foci. This therapeutic activity was correlated with specific CTL activity from pulmonary parenchymal mononuclear cells (PPMC), but not from spleen or blood. Macrophage tumoristatic activity in PPMC also correlated with therapeutic activity, but activity in alveolar macrophages did not. However, therapeutic activity did not correlate with NK or LAK activity at any site. These results demonstrate that the optimal therapeutic protocol is the same as the optimal immunomodulatory dose for pulmonary CTL and macrophage activities. Furthermore, while immunological monitoring may help to optimize treatment protocols, current monitoring procedures that use readily accessible sites, particularly peripheral blood, may not accurately predict the therapeutic efficacy of biological response modifiers in clinical trials.

Effect of tumor burden and route of administration on the immunotherapeutic properties of polyinosinic-polycytidylic acid stabilized with poly-L-lysine in carboxymethyl cellulose [Poly(I,C)-LC].

We examined the immunomodulatory and therapeutic activities of poly(I,C)-LC. Mice received a subcutaneous (s.c.) injection of sufficient numbers of MBL-2 lymphoma cells to produce in 1 week either a high or low tumor burden. A week after tumor cell injection, poly(I,C)-LC treatment was initiated; the agent was administered intraperitoneally (i.p.) at 5 mg/kg twice a week or at 2.5 or 0.5 mg/kg every day or as an intravenous (i.v.) injection at 0.5, 0.05, or 0.005 mg/kg three times a week. Poly(I,C)-LC treatment significantly increased antitumor effector cell functions in a variety of organs (including spleen, lungs, and peritoneum), as shown by increased killing of MBL-2 cells in vitro and increased tumor cell killing by natural killer cells and macrophages. Furthermore, prolongation of survival correlated with peritoneal macrophage tumoricidal activity when poly(I,C)-LC was given i.p. and with pulmonary effector cell function (including natural killer, cytolytic T-lymphocyte and macrophage tumoricidal activity) when the agent was administered i.v.

Immunoregulation in cancer-bearing hosts. Down-regulation of gene expression and cytotoxic function in CD8+ T cells.

The causes of the decreased immune responsiveness in tumor-bearing hosts are incompletely understood. The impact of a decreased immune response in cancer patients on the clinical response in immunotherapy trials has not been evaluated. The present report demonstrates a marked decrease in the therapeutic efficacy of adoptively transferred T lymphocytes obtained from murine hosts bearing tumor for greater than 30 days [late tumor-bearing mice (TBM)] as compared with normal mice and mice bearing tumor for less than 21 days (early TBM). In vitro analysis of the functions of the T lymphocytes from late TBM showed an apparently normal proliferative response to anti-CD3 and IL-2 with adequate lymphokine production from CD4+ cells, but a significant decrease in the cytotoxic function of CD8+ cells. The decreased cytotoxicity was not because of cell-mediated suppression. The expression of granzyme B mRNA was significantly delayed and decreased in magnitude in CD8+ cells from late TBM. Culture supernatants from two unrelated tumor cell lines were able to inhibit the cytotoxic activity of normal CD8+ cells in vitro. The tumor-derived suppressive factor is not transforming growth factor-beta (TGF-beta), but it has not been further characterized. The data suggest that one potential mechanism responsible for immunologic defects in patients with large tumor burdens is a tumor-induced defect that compromises the function of CD8+ effector T cells.

Protective, restorative, and therapeutic properties of recombinant colony-stimulating factors.

Pretreatment of mice with recombinant murine (rM) colony-stimulating factor-granulocyte-macrophage (CSF-gm) or recombinant human (rH) CSF-g provides partial protection from the lethal effects of ionizing radiation or the alkylating agent cyclophosphamide (CTX). In addition, these agents can significantly prolong survival if administered following lethal doses of irradiation or CTX. To induce protective activity, cytokines were injected 20 hours before lethal irradiation or CTX administration. To accelerate recovery from lethal irradiation, the cytokines must be administered shortly following irradiation, and the induction of maximal levels of activity is dependent on chronic administration. In contrast, because of their longer half-lives, accelerated recovery from alkylating agents requires a delay of at least 24 to 48 hours to allow complete clearance of CTX before administration of a CSF. Studies quantitating peripheral blood leukocytes and bone marrow cellularity as well as colony-forming units per culture (CFU-C) frequency and CFU-C per femur revealed a significant correlation between these parameters and the ability to survive lethal irradiation. This is a US government work. There are no restrictions on its use.

Myelostimulatory activity of recombinant human interleukin-2 in mice.

In a series of studies designed to extend our understanding of interleukin-2 (IL-2) and to study the effect of biologic response modifiers on bone marrow, we observed that administering recombinant human (rH) IL-2 to normal mice resulted in an increase in the frequency of colony-forming units-culture (CFU-C) in bone marrow. In addition, rH IL-2 was able to accelerate host recovery from cyclophosphamide (CTX)- or radiation-induced bone marrow depression and peripheral blood leukopenia. Not only can rH IL-2 accelerate, in a dose-dependent manner, the return of bone marrow, peripheral blood cellularity, and CFU-C frequency to normal levels following cytoreduction by CTX or irradiation, but it also significantly increases CFU-C frequency to greater than normal levels. Furthermore, rH IL-2 can significantly prolong survival of animals receiving a lethal dose of irradiation or CTX. Thus, multiple mechanisms are responsible for the synergistic therapeutic activity associated with rH IL-2 and CTX. rH IL-2 does not act only as an immunomodulatory agent in the presence or absence of suppressor T cells, but also accelerates host recovery from cytoreductive agents, resulting in decreased leukopenia and perhaps resistances to secondary infection. Thus, rH IL-2 plus chemotherapy may increase therapeutic activity against neoplastic disease, not only by adding immune stimulation to the direct antitumor effect of the drug but also by allowing delivery of higher, more effective doses of chemotherapy.

Correlation of immunomodulatory and therapeutic activities of interferon and interferon inducers in metastatic disease.

The mechanism of therapeutic activity of recombinant murine interferon-gamma (rMu IFN-gamma) and the IFN inducer polyinosinic-polycytidylic acid solubilized with poly-L-lysine in carboxy methyl cellulose (pICLC) in treating metastatic disease was investigated by comparing effector cell augmentation with therapeutic activity in mice bearing experimental lung metastases (B16-BL6 melanoma). Effector cell functions in spleen, peripheral blood, and lung (the organ with tumor) were tested after 1 and 3 weeks of rMu IFN-gamma or pICLC administration (intravenous, three times a week). In these studies, natural killer (NK), lymphokine-activated killer (LAK), cytolytic T lymphocytes (CTL) (against specific and nonspecific targets), and macrophage tumoricidal and tumoristatic activities were measured. rM IFN-gamma and pICLC had therapeutic activity and immunomodulatory activity in most assays of immune function examined. Specific CTL activity of pulmonary parenchymal mononuclear cells (PPMC), but not in splenocytes or peripheral blood lymphocytes (PBL), during week 3 and not during week 1, correlated with the therapeutic activity of rMu IFN-gamma and of pICLC. Macrophage tumoricidal activity in PPMC, but not in alveolar macrophages, also correlated with the therapeutic activity of rMu IFN-gamma, but the opposite was true for the therapeutic activity of pICLC. NK activity of PPMC, but not of splenocytes or PBL, during week 1 correlated with the therapeutic activity of pICLC; in contrast, NK activity at any site did not correlate with the therapeutic activity of rMu IFN-gamma. LAK activity at any site did not correlate with the therapeutic activity of either agent.

Systematic preclinical study on the therapeutic properties of recombinant human interleukin 2 for the treatment of metastatic disease.

The availability of recombinant human interleukin 2 (rH IL 2) has resulted in its clinical utilization both as a single agent and in combination with lymphokine-activated killer cells. In this report, we discuss the effects of rH IL 2, administered by various routes, on effector cell function, pharmacokinetics and bioavailability, and therapeutic activity. Studies of the pharmacokinetics of in vitro natural killer (NK) cell augmentation by rH IL 2 revealed that a short exposure to high levels of rH IL 2 can augment NK cell activity; however, a prolonged exposure (greater than 12 h) was required to augment NK cell activity at lower doses of rH IL 2. These observations suggested that chronic administration of rH IL 2 might improve immunomodulatory and therapeutic activity. This hypothesis was supported by the results of studies in which we treated experimental and spontaneous metastasis, which revealed that the daily i.p. administration of rH IL 2 resulted in significantly greater therapeutic activity than administration three times/week. The therapeutic protocol for daily i.p. administration had a biphasic dosage optimum, such that low dose therapeutic activity was observed at approximately 100-1000 units/animal in the treatment of experimental metastases or 10 to 100 units/animal in the treatment of spontaneous metastases. There was a second dosage optimum at greater than or equal to 100,000 units/animal rH IL 2 delivered i.p. on a daily basis. Intermediate doses had no significant therapeutic activity. Additional studies revealed that low dose therapeutic activity was not observed in nude mice. In contrast, therapeutic activity was observed in nude mice at high doses of rH IL 2 suggesting that low dose activity was associated with a T-cell-mediated effect, whereas high dose activity may have been mediated by NK or lymphokine-activated killer-like cells. This observation was in agreement with the dose response for T-cell adjuvant activity supporting the hypothesis that low dose therapeutic activity was T-cell associated, because adjuvant activity was observed when rH IL 2 was given daily at approximately 100 units/animal for 3 days, and higher doses had no activity or had a suppressive effect. Because we were concerned about the pharmacological aspects of rH IL 2 treatment, we also examined its therapeutic properties after continuous administration i.p. by osmotic pumps. Under these conditions, therapeutic activity was observed after administration of 600 units/h, whereas lower or higher doses did not have significant therapeutic activity.

Toxicity of tumor necrosis factor is synergistic with gamma-interferon and can be reduced with cyclooxygenase inhibitors.

In recent studies, we have demonstrated that recombinant human tumor necrosis factor (rH TNF), as a single agent, has only minimal therapeutic activity for the treatment of metastatic disease, but when combined with recombinant murine gamma-interferon (rM gamma-IFN), we observed significantly more therapeutic activity than when either agent was administered alone. However, this combination also resulted in increased toxicity. Thus, we undertook a systematic toxicologic study of rH TNF alone or in combination with rM gamma-IFN. Briefly, the toxicity was similar to the generalized Shwartzman's reaction seen during endotoxin shock, with multifocal microthrombi and ischemic necrosis as sequelae. Lesions were observed in the lungs, liver, gastrointestinal tract (preferentially in the duodenum and cecum), testes or uterus, and bone marrow. Our results suggest that TNF (either directly administered or induced in situ) and its induction of arachidonic acid metabolites form one element of toxicity in this model. This conclusion is supported by studies revealing that the toxicity of rH TNF in combination with rM gamma-IFN can be reduced by inhibitors of the cyclooxygenase/lipoxygenase pathway.

In vivo immunologic selection of class I major histocompatibility complex gene deletion variants from the B16-BL6 melanoma.

The mechanism by which tumor allografts escape host immunologic attack was investigated. B16-BL6 cells (the bladder 6 subline of the B16 melanoma) (H-2b) were transfected with a gene (Dd) encoding an allogeneic class I major histocompatibility complex antigen. Clones that expressed Dd antigen were injected into the footpads of nonimmune syngeneic mice, syngeneic immune mice, and nude mice. Under conditions of immunologic selection a clone that contained multiple copies of the transfected gene formed variants that lacked the transfected gene. Primary tumors and pulmonary metastases of immunized mice and pulmonary metastases of nonimmunized mice had lost the Dd gene and, in most cases, all of the associated plasmid. In contrast, in immunodeficient nude mice, primary tumors and pulmonary metastases retained the Dd gene and the associated plasmid. Deletion of genes encoding cell surface antigens may be one of the mechanisms by which allogeneic tumors escape immunologic attack.

Immunomodulatory and immunotherapeutic properties of recombinant gamma-interferon and recombinant tumor necrosis factor in mice.

These studies were designed to examine the immunodulatory and immunotherapeutic properties of recombinant murine interferon gamma (rM IFN-gamma) and recombinant human tumor necrosis factor (rH TNF). We report that rM IFN-gamma activated murine natural killer cells and macrophages in a dose-dependent manner in vivo. The rM IFN-gamma, which demonstrated a bell-shaped therapeutic response curve, must be administered at specific doses and schedules to produce optimal therapeutic activity. Optimal activity was observed after i.v. administration of 50,000 U/animal rM IFN-gamma three times per week. In contrast, rH TNF produced its major therapeutic activity in the treatment of metastatic disease after i.v. but not i.p. administration. The therapeutic effects of rH TNF were as great in these in vivo systems as those of rM IFN-gamma. Furthermore, rH TNF had additive therapeutic activity when administered in conjunction with suboptimal doses of rM IFN-gamma. Unlike rM IFN-gamma, rH TNF did not activate natural killer cells in vivo or in vitro but did augment in vivo and in vitro macrophage tumoricidal activity. It also had synergistic cytostatic properties with rM IFN-gamma for some murine tumor cell lines in vitro. High levels of rH TNF were readily detected in the serum with a half-life of approximately 30 min after i.v. administration. In contrast, only minimal serum TNF activity occurred after i.p. administration, suggesting that i.v. administration may more efficiently facilitate systemic therapeutic activity. In summary, rH TNF and rM IFN-gamma have therapeutic activity for metastatic disease as individual agents and additive therapeutic activity when used in combination. Furthermore, it appears that in addition to therapeutic potential as cytostatic agents, the immunomodulatory properties of rH TNF have a role in its therapeutic properties.

Preclinical approaches to the treatment of metastatic disease: therapeutic properties of rH TNF, rM IFN-gamma, and rH IL-2.

The present studies were undertaken to examine the immunotherapeutic properties of recombinant murine interferon-gamma (rM IFN-gamma), recombinant human tumour necrosis factor (rH TNF), and recombinant human interleukin-2 (rH IL-2) in preclinical metastasis models. It was observed that these cytokines have disparate mechanisms of therapeutic activity as well as different optimal therapeutic protocols. Thus, not only is the dose important to the therapeutic activity of each of the agents; so also is the route of administration, schedule of administration, duration of administration, and sequence of administration. The rM IFN-gamma has a narrow window of activity, with a bell-shaped therapeutic response with a dosage optimum at 50,000 U/animal of rM IFN-gamma administered 3 times per week. In contrast, rH IL-2 has optimal therapeutic activity for the treatment of metastatic disease after i.p. as compared to i.v. administration. This appears to be associated with the serum pharmacokinetics, since longer serum concentrations are achieved following i.p. administration although lower serum levels are also achieved. RH IL-2 has a biphasic dose-response curve for therapeutic activity with optima from 100 to 1000 U/animal and at doses greater than 100,000 U/animal. The lower doses appear to be associated with T cell augmentation whereas the higher doses are associated with NK cell or LAK cell augmentation. RH TNF has therapeutic activity for the treatment of metastatic disease after i.v. but not i.p. administration. High levels of rH TNF are readily detected in the serum following i.v. administration, with a serum half-life of approximately 30 min. In contrast, only minimal serum TNF activity is observed after i.p. administration, suggesting that this may be the origin of the increased therapeutic activity following i.v. administration. Furthermore, rH TNF has additive therapeutic activity when administered in conjunction with suboptimal doses of rM IFN-gamma. Unfortunately, the additive therapeutic activity of rM IFN-gamma and rH TNF is also associated with increased toxicity. However, in preliminary experiments it was found that the b.i.d. administration of aspirin at 25 mg/kg resulted in decreased toxicity. In summary, the recombinant cytokines provide a challenge both preclinically and clinically to the development of optimal therapeutic protocols, and suggest that close attention must be paid to the dose, route, schedule, duration, and sequence of their administration.

Immunomodulatory and therapeutic properties of bestatin in mice.

In this report, we describe the immunomodulatory properties and therapeutic efficacy of bestatin. Macrophage activation, but not natural killer cell augmentation, was observed both in vitro and in vivo. Immunostimulation of T-cell activity was observed in assays of allogeneic mixed lymphocyte response, but cytotoxic effector cells did not develop after an allogeneic mixed lymphocyte-tumor cell culture. Bestatin also had T-cell adjuvant activity when it was admixed with a suboptimal vaccine composed of irradiated tumor cells. We observed significant therapeutic activity against preexisting experimental and spontaneous metastases when bestatin was administered at high doses per animal for 4 weeks.

Replication of Venezuelan equine encephalomyelitis virus in vitro. II. Viral growth response to selected nutritional additives in suspension cultures.

An attempt was made to identify nutritional additives that influence the replication of Venezuelan equine encephalomyelitis virus in suspension cultures grown in a defined serum-free medium. Proline, serine, and choline enhanced titers of the virulent PES strain; the progeny population, however, possessed a virulence character that was somewhat different from that of the PES inocula. These nutritional supplements did not appreciably influence the titers of the attenuated 9t and 20t viral strains. When both the PES and 20t strains were employed as a mixed inoculum in culture, the presence of the latter strain appeared to interfere with the growth of the PES strain and reduced its response to the medium supplements.

Gamma-irradiated Venezuelan equine encephalitis vaccines.

The efficacy of Formalin-inactivated Venezuelan equine encephalitis (VEE) vaccine has been reported to be low for man. Although a live VEE vaccine has been shown to be highly effective for the protection of laboratory workers, local and systemic reactions have occurred in approximately 20% of inoculated individuals. Therefore, studies were initiated in an attempt to produce an inactivated vaccine of high potency with low toxicity. Inactivated VEE vaccines were prepared by exposing virus suspensions to 8 x 10(6) or 10 x 10(6) r of gamma radiation. Irradiated VEE vaccines prepared from virus suspensions produced in Maitland-type chick embryo (MTCE) cell cultures and in monolayer cultures of human diploid strain WI-38 cells were highly immunogenic for mice and guinea pigs. Guinea pigs vaccinated with a series of three inoculations of vaccine (MTCE) survived challenge with at least 10(8.4) mouse intracerebral 50% lethal doses of VEE virus. Irradiated vaccines induced high levels of serum-neutralizing and hemagglutinin-inhibiting antibodies in guinea pigs and rabbits. These findings suggest that ionizing radiation may be effective in the preparation of an inactivated VEE vaccine.

Inactivation of Venezuelan Equine Encephalomyelitis Virus by gamma-Radiation.

Exposure of Venezuelan equine encephalomyelitis (VEE) virus (at -70 C) to 6 x 10 r gamma-radiation (Co) resulted in loss of lethality for young adult mice and guinea pigs, and loss of capacity to produce plaques or cytopathic effects in tissue culture. The suckling mouse was more sensitive for detecting live virus in radiated suspensions than was the adult mouse or guinea pig. Live virus was demonstrable in preparations exposed to 6 x 10 r but not in suspensions exposed to 8 x 10 r and more. The rate of inactivation of VEE virus by gamma-radiation was an exponential function of the dosage.