Two tumor models of curative adoptive chemoimmunotherapy using tumor-infiltrated spleen cells with potent antitumor cytotoxicity stimulated by antigen-sharing tumors.

Previously we have established curative protocols for adoptive chemoimmunotherapy (ACIT) of mice bearing different plasmacytomas that are known to bear cross-reacting antigens: (a) the cure of mice bearing an early-stage, nonpalpable MOPC-315 tumor by a very low dose of cyclophosphamide (10 mg/kg) and cultured MOPC-315-tumor-infiltrated (TI) spleen cells (25 x 10(6)) and (b) the cure of mice bearing a late-stage, relatively drug-resistant, highly metastatic RPC-5 tumor with cyclophosphamide (100 mg/kg) and cultured RPC-5 TI spleen cells (25 x 10(6) - 50 x 10(6)). In both models, the spleen cells were obtained from mice bearing a late-stage tumor and were cultured for 5 days in the presence of polyethyleneglycol 6000 and autochthonous tumor cells as a source of tumor antigen. Here we show that RPC-5 tumor cells could substitute for MOPC-315 tumor cells in the 5-day culture of MOPC-315 TI spleen cells so that they became curative in ACIT for mice bearing an early-stage MOPC-315 tumor. Similarly, MOPC-315 tumor cells could substitute for RPC-5 tumor cells in the 5-day culture of RPC-5 TI spleen cells so that they became curative in ACIT of mice bearing a late-stage RPC-5 tumor. In addition, RPC-5 TI spleen cells cultured with either MOPC-315 or RPC-5 tumor cells were effective in curing all mice bearing an early-stage MOPC-315 tumor by ACIT. However, MOPC-315 TI spleen cells whether cultured with MOPC-315 or RPC-5 tumor cells, were much less effective than cultured RPC-5 TI spleen cells in curing mice bearing a late-stage RPC-5 tumor by ACIT (although the survival of these mice was extended significantly). Interestingly, whereas RPC-5 TI spleen cells cultured with either MOPC-315 or RPC-5 tumor cells were as effective as MOPC-315 TI spleen cells cultured under the same conditions in lysing MOPC-315 tumor cells in vitro, MOPC-315 TI spleen cells that had been cultured with either MOPC-315 or RPC-5 tumor cells exerted a much weaker in vitro cytotoxic T lymphocyte activity against RPC-5 tumor cells than did RPC-5 TI spleen cells that had been cultured under the same conditions.

Cure of mice bearing a late-stage, highly metastatic, drug-resistant tumor by adoptive chemoimmunotherapy.

We show here that in contrast to BALB/c mice bearing a late-stage, large MOPC-315 plasmacytoma, BALB/c mice bearing a late-stage, large RPC-5 plasmacytoma were not cured by cyclophosphamide therapy (15, 50, 100 or 200 mg/kg). However, most BALB/c mice bearing a late-stage RPC-5 tumor were cured by cyclophosphamide therapy (100 mg/kg) in conjunction with adoptive immunotherapy using tumor-infiltrated spleen cells (TISpC) that had been cultured with inactivated RPC-5 tumor cells plus polyethylene glycol 6000, even though this protocol was not effective for the therapy of mice bearing a barely palpable, early-stage RPC-5 tumor. Only a few of the mice that were cured of a late-stage RPC-5 tumor following adoptive chemoimmunotherapy (ACIT) were resistant to a subsequent challenge with RPC-5 tumor cells. However, the challenged mice that had developed progressively growing tumors could then be cured by cyclophosphamide alone when the tumor became large, even though this treatment was not curative for mice bearing a tumor of similar size but not previously treated by ACIT. Thus, the cure by ACIT of BALB/c mice bearing a lethal, late-stage RPC-5 tumor with extensive metastases provides a novel experimental tumor model for investigating the mechanisms by which a chemotherapeutic drug and adoptive cellular immunotherapy can cooperate in causing the complete regression of a large tumor load.

Advantages of adoptive chemoimmunotherapy with polyethylene glycol-cultured, antigen-activated, tumor-infiltrated spleen cells for the complete eradication of lethal MOPC-315 plasmacytomas.

The incorporation of polyethylene glycol-6000 (PEG) into the culture media of tumor-infiltrated spleen cells (TISpC) and MOPC-315 stimulator tumor cells at a responder to stimulator cell ratio of 30/1 had been shown to lead to the appearance of CD8+ T-cells that were effective in adoptive chemoimmunotherapy (ACIT) of mice bearing a barely palpable MOPC-315 tumor (J. A. Wise, M. B. Mokyr, and S. Dray, Cancer Res., 49:3613-3619, 1989). Here we show that in the presence of substantially fewer added stimulator tumor cells (responder to stimulator cell ratio, 100/1), the inclusion of PEG in the cultures of TISpC also enhanced the appearance of cells that were highly effective in curing such mice by ACIT. Moreover, these PEG-cultured TISpC were more effective in ACIT than TISpC cultured in the presence of an optimal concentration of recombinant interleukin-2 (60 IU/ml). The potency of the tumor-eradicating activity of the PEG-cultured TISpC in ACIT was further illustrated by their ability to cause the complete regression of a large (20-22 mm) s.c. MOPC-315 tumor in conjunction with a dose of drug that by itself did not cause tumor regression. PEG-cultured TISpC that were effective against MOPC-315 tumor cells in an antigen-specific manner. In fact, PEG-cultured TISpC were more effective than recombinant interleukin-2-cultured TISpC, not only in ACIT, but also in their ability to lyse MOPC-315 tumor cells in vitro. Thus, a direct specific lytic activity against the tumor by cytotoxic T-lymphocytes is the apparent mechanism through which the complete regression of the large tumor burden is brought about by the PEG-cultured TISpC. Finally, we suggest that the incorporation of PEG to render ineffective lymphoid cells effective in ACIT may offer some advantages compared with the incorporation of recombinant interleukin-2 and may be suitable for protocols to generate human cytotoxic cells for cancer therapy when there are relatively low numbers of available tumor cells.

Serine proteinase inhibitors produced by human melanoma cell lines.

The human melanoma cell lines M21 and MSM-M2 are shown to produce two similar competitive inhibitors of trypsin, a serine proteinase. These proteinase inhibitors inhibited the serine proteinases trypsin and kallikrein with similar efficiency but did not inhibit plasmin (a serine proteinase) or papain (a thiol proteinase). Active synthesis of the inhibitors during cell culture was indicated by the requirement for cell viability, the increase in inhibitory activity of the supernatant with time, and the incorporation of 35S-methionine into the inhibitors. The two inhibitors were stable to heat (70 degrees C) and extremes of pH. Their molecular weights were estimated at 670 and 250 kD, respectively. A screening of the supernatants of five other human melanoma cell lines by HPLC showed that they all released a similar trypsin inhibitory factor not detected in human or bovine serum. The isolation of these proteinase inhibitors facilitates a study of their putative role in tumor growth.

Inhibitors of lymphocyte activation secreted by human melanoma cell lines.

The concentrated supernatants of nine human melanoma cell line cultures were analyzed for the presence of factors that inhibit in vitro immunological reactions. All cell lines secreted a factor that inhibited LPS-induced proliferation of murine B cells; eight cell lines released a factor that inhibited PHA-induced proliferation of murine T cells; all of the three cell lines investigated secreted a factor that inhibited the allogeneic stimulation of BALB/c spleen cells by mitomycin-C-treated C57BL spleen cells. Further analysis of the M21 cell supernatant indicated that its continuous presence was required for the inhibition of the PHA but not of the LPS response suggesting a different mechanism of action. High levels of PHA, but not of LPS, could overcome the inhibitory effect of M21 supernatants. Fractionation of M21 supernatants by sucrose gradient centrifugation, DEAE chromatography and gel filtration suggested that the anti-PHA and the anti-LPS activities were due to different factors. These factors differed from a serine proteinase inhibitor that is also released by M21 cells. Since these inhibitory factors may have a role in protecting a melanoma tumor from attack by the immune system of the host, their consideration may be helpful in designing protocols for therapy which include methods to boost the antitumor responses of the host.

Lysis of antigenically unrelated tumor cells mediated by Lyt 2+ splenic T-cells from melphalan-cured MOPC-315 tumor bearers.

We have previously shown that mice cured of a large MOPC-315 tumor following low-dose melphalan (L-phenylalanine mustard, L-PAM) therapy can exert, upon challenge with MOPC-315 tumor cells, an antitumor effect against innocent bystander tumor cells present within the same tumor site (Barker, E., and Mokyr, M.B. Cancer Immunol. Immunother., 25: 215-224, 1987). Here we show that T-cells are important for the MOPC-315-induced rejection of MOPC-104E tumor cells present within the same site. To further characterize the innocent bystander killing activity exerted by L-PAM-cured MOPC-315 tumor bearers upon stimulation with MOPC-315 tumor cells, we established the in vitro conditions under which lymphoid cells from L-PAM-cured MOPC-315 tumor bearers can exert an antitumor effect against innocent bystanders. Specifically, we established that spleen cells from mice that just completed the rejection of a large MOPC-315 tumor following low-dose L-PAM therapy can, upon stimulation with MOPC-315 tumor cells, bring about the killing of antigenically unrelated tumor cells in a 12-h 51Cr release assay. The magnitude of lysis of EL4 and WEHI 22.1 tumor cells by MOPC-315 in vitro-immunized (IVI) spleen cells from L-PAM-cured MOPC-315 tumor bearers can be substantially enhanced upon reexposure of the spleen cells to MOPC-315-associated antigens during the 12-h 51Cr release assay. The lysis of innocent bystander tumor cells by these MOPC-315-IVI spleen cells was found to be mediated by T-cells of the Lyt 2 and not the L3T4 phenotype. These Lyt 2+ T-cells did not appear to mediate their lytic activity for innocent bystander tumor cells via effector macrophages, since a drastic reduction in macrophage frequency among the MOPC-315-IVI spleen cells just prior to assessing the lytic activity of the spleen cells did not reduce, but actually enhanced, the magnitude of EL4 lysis. In addition, a Lyt 2+ T-cell clone derived from mice cured of a large MOPC-315 tumor by a low dose of drug was capable, upon stimulation with MOPC-315 tumor cells, of exerting a potent lytic effect against EL4 and WEHI 22.1 tumor cells in the 12-h 51Cr release assay. Thus, Lyt 2+ T-cells independent of effector macrophages are responsible for lysis of innocent bystander tumor cells by MOPC-315-IVI spleen cells from L-PAM-cured MOPC-315 tumor bearers.

Enhancement of the effectiveness of Lyt-2+ T-cells for adoptive chemoimmunotherapy by short-term exposure of tumor-bearer spleen cells to polyethylene glycol and/or melphalan.

Uncultured tumor-infiltrated spleen cells (TISpC) from mice bearing large (20-22 mm) s.c. MOPC-315 plasmacytomas were previously shown to be ineffective in bringing about the cure of mice bearing a nonpalpable (Day 4) tumor that had been treated with a subcurative dose (10 mg/kg) of cyclophosphamide (i.e., adoptive chemoimmunotherapy, ACIT) (M. B. Mokyr, J. C. D. Hengst, and S. Dray, Cancer Res., 42:974-979, 1982). Here we show that TISpC cultured for 5 days in the presence of inactivated MOPC-315 stimulator cells acquire some effectiveness in curing mice by ACIT, and this effectiveness is greatly enhanced if polyethylene glycol 6000 (PEG) is also added to the culture. The Lyt 2+ T-cells, and not the L3T4+ T-cells, are responsible for the effectiveness of the cultured TISpC in ACIT. In fact, the L3T4+ T-cells are apparently not required even during culture of TISpC for the generation of Lyt 2+ T-cells effective in ACIT. Although the TISpC cultured with MOPC-315 cells and PEG contained approximately twice as many Lyt 2+ cells as did TISpC cultured without PEG, the increase in the activity of the former cells is not due simply to the increase in the percentage of Lyt 2+ cells, but is most likely due to an increase in the percentage and/or activity of Lyt 2+ cells with specificity for MOPC-315-associated antigens. The effectiveness of TISpC cultured with MOPC-315 stimulator cells and PEG in ACIT can be enhanced even further by pretreatment of these cells with the immunomodulating agent melphalan (0.5 nmol/ml) prior to culture initiation. Thus, the above methods of culture render ineffective lymphoid cells effective in ACIT and are suitable for evaluation in protocols for human cancer therapy.

Cyclophosphamide and melphalan as immunopotentiating agents in cancer therapy.

The murine plasmacytoma, MOPC-315, has been used as a tumor model to investigate the immunopotentiating effect of a low dose of cyclophosphamide (CY) or melphalan (L-PAM). Each drug was shown to shift the balance in mice bearing a late-stage tumor from a state of immunosuppression to that of potent T-cell-dependent antitumor immunity against tumor-associated antigens. The resultant immunity eradicated the extensive tumor burden not already eradicated by the direct tumoricidal activity of the drug and brought about the cure of the mice. The immunity responsible for tumor eradication, as well as the immunity responsible for the resistance of the cured mice to further tumor challenge, was mediated by the Lyt 2 subset of T-cells which contains cytotoxic T-cells. The principle of using a low dose of drug to selectively decrease suppressor cell activity so as to allow the development of antitumor immunity with the aid of autologous tumor vaccine or interleukin-2 has been exploited successfully by clinicians in therapeutic protocols for human melanoma.

Tumoricidal and immunomodulatory activities of drugs and implications for therapy of mice bearing a late stage MOPC-315 plasmacytoma.

The effectiveness of a relatively low dose of cyclophosphamide (15 mg/kg CY), melphalan (2.5 mg/kg L-PAM) or the monofunctional form of CY (150 mg/kg MoCY) for the cure of mice bearing a large primary s.c. MOPC-315 tumor and extensive metastases has been shown to be dependent on the cooperation of the drugs' tumoricidal activity with T-cell-dependent antitumor immunity, the latter facilitated by the drug's immunomodulatory activity. Here, we have compared the curative effectiveness of three additional drugs: methyl nitrosourea (MNU), hydroxyurea (OH-urea) and bis-chloroethyl nitrosourea (BCNU). Among these drugs, only a relatively low dose of BCNU (15-20 mg/kg) was effective in curing most mice (85%) bearing a large, late stage tumor. A higher dose of BCNU (40 mg/kg, LD10) was much less effective. After an optimal dose of BCNU, the proliferative capacity of the tumor cells 24 h after therapy was reduced by greater than 97%. However, viable tumorigenic cells were still present in the primary tumor and enhanced T-cell-dependent antitumor immunity was necessary for their eradication. The cured mice were resistant to tumor rechallenge. When a low curative dose of L-PAM was followed by OH-urea, the therapeutic effectiveness was not affected, but when this dose of L-PAM was followed by a high nontoxic dose of MNU (100-150 mg/kg), the therapeutic effectiveness was diminished even though MNU was highly tumoricidal (i.e. greater than 99% inhibition of proliferative activity). Thus, BCNU appears to be similar to CY, L-PAM and MoCY in its mechanism of MOPC-315 tumor eradication. The alkylating activity of CY, L-PAM, MoCY and BCNU appears to be critical for their combined tumoricidal and immunomodulatory effects. Since BCNU is the simplest of these four drugs with respect to metabolic pathway, a further study with BCNU and related constructs may shed some light on the biochemical mechanisms of their mode of action. At least one reason for the ineffectiveness of OH-urea or MNU at either low or nontoxic high doses was poor tumoricidal or immunomodulatory activity, respectively. Thus, it seems important to consider both the tumoricidal and immunomodulatory activities of drugs when developing regimens for effective chemotherapy.

Effect of low-dose cyclophosphamide therapy on specific and nonspecific T cell-dependent immune responses of spleen cells from mice bearing large MOPC-315 plasmacytomas.

Some T cell-dependent immune parameters were examined in mice bearing a large MOPC-315 plasmacytoma before and after treatment with a low dose (15 mg/kg) of CY. Prior to CY therapy, spleen cells from mice bearing a large MOPC-315 tumor were depressed in their ability to generate an in vitro cytotoxic response to the MOPC-315 tumor, to a different syngeneic plasmacytoma, MOPC-104E, and to an allogeneic thymoma, EL4. The spleen cells of these mice were also depressed in their ability to proliferate in response to the T cell mitogen PHA. Following CY therapy, the spleen cells generated an enhanced anti-MOPC-315 cytotoxic response by day 2, and the level of this response continued to increase so that by day 7, it was greatly enhanced and was much greater than the response of normal spleen cells. The recovery of the cytotoxic responsiveness to the antigenically related MOPC-104E tumor after CY therapy followed a similar pattern. In contrast, the spleen cells of these animals remained depressed in their cytotoxic response to the antigenically unrelated EL4 thymoma for at least 11 days after CY therapy. Although the anti-EL4 response recovered by day 14, the level of antitumor cytotoxicity generated did not exceed that generated by normal spleen cells. The PHA response remained greatly depressed in CY-treated MOPC-315 tumor bearers, even 14 days after the chemotherapy. Thus, at a time following low-dose CY therapy, when potent T cell-dependent antiplasmacytoma immunity had completed the eradication of a large MOPC-315 tumor burden not eliminated through the direct effect of the drug, the T cell-dependent response to an unrelated tumor and to PHA remained depressed.

A serine proteinase inhibitor produced by an HTLV I virus-transformed human T lymphocyte line.

A previous report from our laboratory indicated that a proteinase inhibitor is produced by rabbit T lymphocytes. We now report that a human T cell line, C91/PL, produces a proteinase inhibitor which inhibits the enzymatic activity of trypsin and kallikrein. This newly identified proteinase inhibitor (LPI 1) did not inhibit the enzymatic activity of four other serine proteinases (thrombin, plasmin, chymotrypsin, or pancreatic elastase), a thiol proteinase (papain), or a carboxyl proteinase (pepsin). Active synthesis of LPI 1 by the C91/PL cell line was shown by the appearance of similar levels of inhibitory activity in sequential cell supernatants, lack of appearance of inhibitor in supernatants of cells killed by heat or sodium azide or of viable cells in the presence of cyclohexamide, and incorporation of a radiolabeled amino acid into newly synthesized inhibitor. Although both the inhibitor of rabbit origin and of human origin are proteins produced by T cells and have similar inhibitory specificity, important differences were observed: LPI 1 is sensitive to boiling and the two inhibitors migrate differently upon electrophoresis in substrate-containing polyacrylamide gel. Furthermore, LPI 1 was produced by a cell line of the T4 phenotype which had been established by in vitro viral transformation of human cord blood lymphocytes with HTLV 1 whereas the inhibitor of rabbit origin was produced by normal splenic T cells. Three other human T cell lines of the T4 phenotype, MOLT-13, KE-37, and HPB-ALL, from patients with acute lymphoblastic leukemia did not produce a proteinase inhibitor. Thus, the production of proteinase inhibitors does not appear to be a general characteristic of human T cell lines nor of the T4 subset. Proteinase inhibitors produced by T cells may have an immunoregulatory role in proteinase-mediated physiological processes.

Interplay between the toxic effects of anticancer drugs and host antitumor immunity in cancer therapy.

The results accumulated thus far illustrate that the therapeutic efficacy of many anticancer drugs depends not only on the direct tumoricidal/tumoristatic activity of the drug but also on the contribution to tumor eradication of antitumor immunity which emerges after the chemotherapy. When a low dose of anticancer drug is employed, it reduces the tumor burden to a lesser extent than a high dose of drug. Consequently, in order for the low dose of drug to be as effective as a high dose of drug, antitumor immunity has to control a larger tumor burden than that controlled by the immune system following high-dose chemotherapy. This was shown to happen in several experimental tumor models wherein the low dose of drug greatly potentiated host antitumor immunity while the high dose of drug either potentiated host antitumor immunity to a lesser extent, did not potentiate it at all, or actually exerted a suppressive effect on host antitumor immunity. As a result of the immunopotentiating activity of the low dose of drug, tumor-bearing animals which did not exhibit concomitant antitumor immunity due to the inhibitory activity of suppressor cells, developed a very potent antitumor immunity shortly after the chemotherapy. The immunopotentiating effect of the low dose of drug was attributed in these situations to drug-mediated selective elimination of suppressor cell activity and possibly also to drug-mediated enhancement in the activity of T cells of the helper phenotype through elevation in IL-2 production. Antitumor immunity can also facilitate the therapeutic effectiveness of high-dose chemotherapy. Therefore, it is important to determine the conditions under which high-dose chemotherapy can also potentiate host antitumor immunity. From the rodent data available, the picture emerges that the maturity of antitumor immunity at the time of the chemotherapy is an important factor in determining if the high dose of drug leads to immunosuppression or immunopotentiation. When a high dose of drug is administered to tumor bearers with a fully developed antitumor immune response, it is more likely to lead to potentiation of host antitumor immunity than when the high dose of drug is administered to tumor bearers that have not yet achieved the full maturity of their antitumor immune response. In order to achieve the maximal enhancement of antitumor immunity in hitherto immunosuppressed tumor bearers, the suppressor-cell pool should be more sensitive to the toxic effects of the anticancer drugs than are cells involved in immune-tumor eradication.(ABSTRACT TRUNCATED AT 400 WORDS)

Ability of cyclophosphamide in the absence of cross-linking activity to exert the immunomodulatory effect required for the cure of mice bearing a large MOPC-315 tumor.

We have previously shown that mice bearing a late-stage, large primary MOPC-315 plasmacytoma and extensive metastases can be cured by a low dose of the bifunctional alkylating drug, cyclophosphamide (BiCY) (J.C.D. Hengst et al., Cancer Res., 40: 2135-2141, 1980). Here we show that therapy with the monofunctional form of cyclophosphamide (MoCY) can also cure such mice. However, a dose of at least 150 mg of MoCY per kg is required to approximate the curative effectiveness of the lowest curative dose of BiCY, i.e., 15 mg/kg. This need for a 10-fold higher dose of MoCY is due, at least in part, to the 10-fold lower direct tumoricidal and/or tumoristatic activity of MoCY compared to BiCY. Consequently, a 10-fold higher dose of MoCY is required to directly reduce the tumor burden to the level reduced by 15 mg of BiCY per kg. Other than dose, the therapy of the mice with 150 mg of MoCY per kg was similar in its essential features to that shown previously for therapy with 15 mg of BiCY per kg (J.C.D. Hengst et al., Cancer Res., 40: 2135-2141, 1980; J.C.D. Hengst et al., Cancer Res., 41:2163-2167, 1981; Q-W. Ye et al., Cancer Immunol. Immunother., 16:162-169, 1984; Q-W. Ye and M.B. Mokyr, Cancer Res., 44: 3873-3879, 1984; M.B. Mokyr and S. Dray, Cancer Res., 43: 3112-3119, 1983), namely: (a) the drug does not directly eradicate all tumor cells; (b) host T-cell-dependent antitumor immunity is also required for the curative effect; (c) the therapy of tumor bearers leads to the rapid appearance of an augmented antitumor immune potential in their hitherto immunosuppressed spleen; and (d) the cured mice are resistant to a subsequent challenge with at least 300-fold the minimal lethal tumor dose. Thus, cross-linking is not an essential property for the immunomodulatory activity of BiCY nor for its direct antitumor effect. However, in the presence of cross-linking activity, a much lower dose of drug is effective.

Melphalan-induced enhancement of tumor cell immunostimulatory capacity as a mechanism for the appearance of potent antitumor immunity in the spleen of mice bearing a large metastatic MOPC-315 tumor.

Exposure of MOPC-315 cells from the primary tumor nodule to a low concentration (0.5 nmol/ml) of melphalan (L-phenylalanine mustard; L-PAM) rendered the tumor cells capable of bringing about the generation of a potent primary antitumor cytotoxic response. Accordingly, the level of antitumor cytotoxicity generated by normal spleen cells immunized in vitro with L-PAm-treated tumor cells was at least five-fold greater than the level generated in response to untreated tumor cells. The marked superiority of L-PAM-treated tumor cells over untreated tumor cells in bringing about the generation of antitumor cytotoxicity was evident over a wide range of responder to stimulator cell ratios. The higher level of antitumor cytotoxicity exhibited by normal spleen cells immunized with L-PAM-treated tumor cells as compared with untreated tumor cells was not merely the result of direct drug-mediated tumoricidal activity, thereby reducing the number of tumor cells present which can act as cold target cell inhibitors during the 51Cr release assay. This is apparent from the observation that the level of antitumor cytotoxicity generated in response to a given percentage of stimulator tumor cells pretreated with 0.5 nmol L-PAM/ml, a drug concentration associated with retention of 60% tumor cell proliferative capacity, is substantially greater than that generated in response to less than half that percentage of untreated stimulator tumor cells. Moreover, stimulator tumor cells exposed to a fully antiproliferative concentration of L-PAM brought about the generation of a higher level of antitumor cytotoxicity than stimulator tumor cells exposed to mitomycin C at a concentration which inhibited the proliferation of the tumor cells to the same extent as the L-PAM. A low concentration of L-PAM which was effective in rendering isolated tumor cells from the primary tumor nodule capable of bringing about the generation of antitumor cytotoxicity was also effective in inducing the appearance of potent antitumor immune potential in tumor bearer splenic cells containing metastatic tumor cells.

Cyclophosphamide-mediated enhancement of antitumor immune potential of immunosuppressed spleen cells from mice bearing a large MOPC-315 tumor.

We have utilized 4-hydroperoxycyclophosphamide (4-HP-CY) as a probe for the immunomodulatory activity of the metabolites of cyclophosphamide (CY) since 4-HP-CY hydrolyzes spontaneously in aqueous solution to the same metabolites as those formed after in vivo conversion of CY by microsomal enzymes. Exposure of immunosuppressed MOPC-315 tumor bearer spleens to a low concentration of 4-HP-CY (0.1-3.0 micron) resulted in augmented antitumor immune potential. The level of antitumor immune potential exhibited by 4-HP-CY-treated tumor bearer spleen cells was not further augmented but was actually reduced by depletion of glass-adherent cells, a procedure which is effective in removing the cells known to have immunosuppressive activity (i.e. metastatic tumor cells and macrophages) from the spleen of untreated MOPC-315 tumor-bearing mice. In fac, 4-HP-CY was superior to depletion of glass-adherent cells in augmenting the antitumor immune potential of immunosuppressed tumor bearer spleen cells. When cells from the primary tumor nodule were incubated with a low concentration of 4-HP-CY which only marginally inhibited their proliferation, the drug completely abolished the suppressive activity of the cells for in vitro generation of antitumor cytotoxicity by normal spleen cells. Moreover, a high level of antitumor cytotoxicity developed when normal spleen cells were cultured in vitro with 4-HP-CY-treated tumor cells at a wide range of ratios of spleen cells to tumor cells. Thus, in the MOPC-315 tumor model, metabolites of CY eliminate the inhibitory effectiveness of splenic suppressor cells and induce the appearance of immunopotentiating activity. The results obtained with 4-HP-CY in vitro provide support for the hypothesis that low-dose CY therapy of mice bearing a large MOPC-315 tumor leads to the appearance of augmented antitumor immune potential in their hitherto immunosuppressed spleen cells as a result of the in situ immunomodulatory effect of the drug on cells in the spleen.

Suppression of antitumor immunity by macrophages in spleens of mice bearing a large MOPC-315 tumor.

We had shown previously that progression of MOPC-315 plasmacytoma growth is associated with an increase in the percentage of macrophages in the spleen as well as a decrease in the ability of tumor-bearer spleen cells to mount an antitumor cytotoxic response upon in vitro immunization. Here we provide evidence that macrophages in the MOPC-315 tumor-bearer spleen are responsible at least in part for the suppression of the generation of antitumor cytotoxicity. Accordingly, removal of most macrophages by depletion of phagocytic cells or Sephadex G-10-adherent cells from spleens of mice bearing a large tumor resulted in augmented antitumor immune potential. Also, Sephadex G-10-adherent spleen cells from tumor-bearing (but not normal) mice drastically suppressed the in vitro generation of antitumor cytotoxicity by normal spleen cells. The suppressive activity of these adherent cells did not reside in contaminating suppressor T cells, since it was not reduced by treatment with monoclonal anti-Thy 1.2 antibody plus complement. The Sephadex G-10-adherent cell population from the tumor-bearer spleen suppressed the in vitro generation of antitumor cytotoxicity against autochthonous tumor cells but not against allogeneic EL4 tumor cells, and hence the suppression was apparently specific. The suppressive activity of the Sephadex G-10-adherent cell population from tumor-bearer spleens was overcome by treatment of the tumor-bearing mice with a low curative dose of cyclophosphamide. This immunomodulatory effect of a low dose of the drug in overcoming the suppression mediated by the Sephadex G-10-adherent cell population enables the effector arm of the immune system of tumor-bearing mice to cooperate effectively with the drug's tumoricidal activity in tumor eradication.

Melphalan-mediated potentiation of antitumor immune responsiveness of immunosuppressed spleen cells from mice bearing a large MOPC-315 tumor.

Administration of a low dose of L-PAM (0.75 mg/kg) to mice bearing a large SC MOPC-315 tumor and extensive metastases led to the development of augmented antitumor immune potential in their hitherto immunosuppressed spleen cells. Such drug-induced potentiation of antitumor immune responsiveness appeared by day 2 after chemotherapy, and it could not be further enhanced but was actually reduced by depletion of glass-adherent cells, a procedure which is effective in depleting the cells known to have inhibitory activity (i.e., macrophages and metastatic tumor cells). To establish that L-PAM can lead to selective in situ abrogation of the inhibitory effectiveness of the splenic macrophages and metastatic tumor cells, we demonstrated that incubation of immunosuppressed tumor-bearer spleen cells with a low concentration of L-PAM in vitro also resulted in augmented antitumor immune potential that could not be further augmented by depletion of glass-adherent cells. L-PAM-mediated enhancement of the antitumor immune potential of immunosuppressed tumor bearer spleen cells was due at least in part to the effects of the drug on the splenic metastatic tumor cells. Isolated tumor cells treated with a low concentration of L-PAM were not only devoid of inhibitory activity for the primary in vitro antitumor immune response by normal spleen cells, but actually manifested a strong immunostimulatory capacity. Thus, L-PAM given at a low dose enhances the development of potent antitumor immunity which brings about the eradication of a large tumorigenic load that remains after the drug has been cleared from the circulation.

Some advantages of curing mice bearing a large subcutaneous MOPC-315 tumor with a low dose rather than a high dose of cyclophosphamide.

Mice bearing a large s.c. MOPC-315 tumor can be cured by a dose of cyclophosphamide (CY) ranging from 15 to 200 mg/kg. However, the low (15 mg/kg) and the high (200 mg/kg) doses mediate tumor eradication via different mechanisms. Tumor eradication by the low dose of drug requires the cooperation of the toxic effect of the drug and T-cell-dependent antitumor immunity. On the other hand, tumor eradication by the high dose of drug does not require the participation of antitumor immunity but depends primarily on the tumoricidal activity of the drug. Spleen cells from tumor-bearing mice treated with the low dose of CY exhibit an augmented antitumor immune potential, whereas spleen cells from tumor-bearing mice treated with the high dose of CY exhibit suppressed antitumor immune potential. More importantly, tumor-bearing mice treated with the low dose of drug are able to reject a challenge with 300 times the minimal lethal tumor dose given 1, 6, or 31 days after CY therapy, whereas mice treated with the high dose of drug are unable to reject such a challenge given within the same time intervals after CY therapy. Moreover, when mice bearing a large tumor are treated with the high dose of CY and subsequently challenged again with tumor cells to establish a Day 4 nonpalpable tumor, this tumor is less responsive to cure by combined chemoimmunotherapy than is a Day 4 nonpalpable tumor established in normal mice. Thus, although the high dose of CY can cure most mice bearing a large-size MOPC-315 tumor, it not only does not result in antitumor immunity, but it actually reduces the effectiveness of chemoimmunotherapy for a second tumor challenge. In contrast, mice cured with the low dose of CY exhibit long-lasting potent antitumor immunity.

Increase in the effectiveness of melphalan therapy with progression of MOPC-315 plasmacytoma tumor growth.

Following inoculation with 1 X 10(6) MOPC-315 tumor cells, a single injection of a very low dose of melphalan (L-PAM, L-phenylalanine mustard), 0.75 mg/kg, cured most of the mice bearing a day 11 large primary tumor (20 mm) and metastases, but failed to cure mice bearing a day 4 nonpalpable tumor. Treatment of mice bearing a nonpalpable tumor with the very low dose of drug compromised the ability of the mice to respond effectively to the same low dose of drug when the tumor became large (day 12). However, a nonpalpable tumor could be eradicated by treatment of tumor bearers with a low dose of L-PAM, if it was present concomitantly with a large tumor on the contralateral side. A high dose of L-PAM, 15 mg/kg, cured mice bearing either a nonpalpable or a large tumor. The eradication of the tumor induced by the high dose of L-PAM appeared to be due solely to the tumoricidal effect of the drug. On the other hand, the eradication of the tumor by the low dose of L-PAM also required the participation of antitumor immunity of the host, since subsequent injection of antithymocyte serum abrogated the curative effect of the drug in most mice. Mice cured by a high dose of L-PAM were not resistant to subsequent lethal tumor challenge. In contrast, mice cured by the low dose of L-PAM were able to reject a tumor challenge of 300 times the minimal lethal tumor dose. The results obtained with L-PAM therapy are similar to the results that we had previously reported with cyclophosphamide therapy. Thus, the timing of therapy with a low dose of drug for mice bearing a MOPC-315 tumor is critical for successful therapy. Moreover, the selection of a low dose rather than a high dose of drug to eradicate a large tumor offers the advantage that it results in long-lasting potent antitumor immunity as a consequence of the participation of host antitumor immunity in the eradication of the tumor.

Mode of action of polyethylene glycol 6000 in potentiating the in vitro generation of antitumor cytotoxicity by MOPC-315 tumor bearer spleen cells.

Some of the possible mechanisms by which polyethylene glycol (PEG) augments the ability of MOPC-315 tumor bearer spleen cells to mediate in vitro antitumor cytotoxicity were evaluated. The level of antitumor cytotoxicity obtained in 5-day cultures of tumor bearer spleen cell suspensions correlated inversely with the percentage of Trinitrophenol (TNP)-rosettable cells (presumably metastatic tumor cells) present in the spleen. The kinetics of decrease in the percentage of TNP-rosettable cells coincided with the appearance of antitumor cytotoxicity. In addition, PEG was shown to interfere with the ability of viable tumor cells to suppress the in vitro generation of antitumor cytotoxicity in normal spleen cells cultured with mitomycin C-treated tumor cells. However, the decrease in the content of TNP-rosettable cells and the concurrent increase in the level of antitumor cytotoxicity were not due to direct cytotoxic and/or cytostatic effects of PEG on tumor cells. Spleen cells cultured in the presence of PEG had an increased rate of [3H]thymidine incorporation and proliferation compared to spleen cells cultured in the absence of PEG. However, the PEG-induced decrease in the percentage of TNP-rosettable cells either preceded or occurred at the same time that the PEG-induced increase in spleen cell number was observed. Therefore, spleen cell proliferation can at best explain only partially the PEG-induced decrease in the content of TNP-rosettable cells, and other mechanisms for the decrease must be considered.