J Neurosurg 77:757-762, 1992

Antitumor activity against established intracerebral gliomas exhibited by cytotoxic T lymphocytes, but not by lymphokine-activated killer cells FRANK P. HOLLADAY, M D., TERESA HEITZ, AND GARY W. WOOD, Ptt.D. Departments ~f Pathologj' and Oncology, and Surgery (Neurosur,~eEv Division), University of Kansas Medical Center, Kansas City, Kansas

u- Specific immune responses against malignant brain tumors have been difficult to demonstrate. Moreover, immunotherapy has met with little success, despite using lymphocytes with high levels of cytotoxicity against brain tumor cells. Lymphokine-activated killer (LAK) cells that nonspecifically kill brain tumor cells are produced by stimulating resting precursors with high concentrations of interleukin-2 (IL-2). Cytotoxic T lymphocytes that specifically kill brain tumor cells are produced by stimulating antigen receptor-positive immune-cell precursors with tumor cells. In an attempt to gain insight into immune cell function against brain tumors, the present study compared the in vitro and in vivo activities of LAK cells and cytotoxic T lymphocytes produced against RT2, a fast-growing rat glioma cell line. Lymphokine-activated killer cells were produced by stimulating normal rat spleen cells with 1000 unks of IL-2, and RT2-speeific cytotoxic T lymphocytes were produced by priming them in vivo with RT2 and Corynebacterium pan'urn and restimulating primed spleen cells with RT2 in vitro. Lymphokine-activated killer cells were highly cytotoxic for a panel of syngeneic and allogeneic brain tumor and non-brain tumor target cells, including RT2, as measured in a 4-hour S~Crrelease assay. Cytotoxic T lymphocytes were highly cytotoxic only for syngeneic brain tumor target cells. Lymphokineactivated killer cells and cytotoxic T lymphocytes were tested for in vivo antitumor activity against intracerebral RT2 by intravenous adoptive transfer of activated lymphocytes. Untreated rats died in approximately 2 weeks. Lymphokine-activated killer cells plus IL-2 failed to affect survival when treatment was initiated as early as 1 day following tumor inoculation. Cytotoxic T lymphocytes and IL-2 administered as late as Day 5 rejected progressing intracerebral tumor. Thus, although both cytotoxic T lymphocytes and LAK cells exhibited high levels of in vitro killing ofglioma cells, only cytotoxic T lymphocytes rejected progressing intracerebral tumors. KEY WORDS

cytotoxicity

lymphokine-activated killer cell

9 T lymphocyte glioma natural killer cell

HREE major types of cytotoxic lymphocytes with antitumor activity have been described. One of those cell populations, natural killer (NK) cells, expresses cytotoxic activity in its resting state. 92~ Natural killer cells from animals or humans with no known previous exposure to tumor cells will kill selected types of tumor targets. It is believed that NK cells constitute part of the body's system of natural resistance. A second population of cytotoxic cells, lymphokine-activated killer (LAK) cells, is produced by stimulating normal lymphocytes with high concentrations of interleukin-2 (IL-2), ~-~-~As with NK cells, which provide a high proportion of the LAK cell precursors, LAK cells need not be exposed to tumor cells prior to testing. Lymphokineactivated killer cells are characterized by their ability to kill a wide variety of lymphoid and non-lymphoid tumor target cells in vitro in a non-antigen-specific and

T

,L Neurosurg. / Volume 77/November, 1992

immunotherapy

non-major histocompatibility gene product-restricted fashion s.,,.z~.z_,Lymphokine-activated killer cells exhibit in vivo antitumor activity against a variety of experimental and human tumors, s'22 A third population of cells, cytotoxic T lymphocytes, is produced only in response to antigen-specific stimulation of antigen receptor-positive cytotoxic T lymphocyte precursors. Tumor antigen-specific cytotoxic T lymphocytes are efficiently produced by priming lymphocytes in vivo with tumor ceils, then stimulating the primed lymphocytes in vitro with the same or antigenically related cellsJ -7-'~'2~ _,7In contrast to NK cells and LAK cells, cytotoxic T lymphocytes kill only tumor cells that express the antigen(s) used for immunization, and cytotoxic T lymphocyte recognition is class I major histocompatibility gene complex-restricted. 2~ The generation of cytotoxic T lymphocytes is dependent on IL757

F. P. H o l l a d a y , T. Heitz, a n d G. W. W o o d 2 production, a requirement that is fulfilled by antigenspecific stimulation of T-helper lymphocytes.4-7->-> _,7 Tumor immunity is mediated by cytotoxic T lymphocytes. -~ Malignant gliomas progress rapidly and are almost invariably fatal despite aggressive therapy employing various combinations of surgery, irradiation, chemical agents, and biological response modifiers. ~'3~5~-'4~~ Adoptive cellular immunotherapy using LAK cells has been ineffective against both human and experimental gliomas, ~2~4"~7"J~-'9~ Since LAK cells are effective against other types of tumors ~2-"and brain tumor cells are not resistant to LAK cell killing, the masons for the inability of LAK cells to act against progressing brain tumors, even after direct administration to the tumor bed, are unclear. Preliminary studies in this laboratory demonstrated that cytotoxic T lymphocytes can act against intracembral gliomas. Since cytotoxic T lymphocytes and LAK cells have been reported to kill brain tumor target cells in vitro,~t,29it was of interest to compare directly their in vitro and in vivo antitumor effects under identical, well-defined experimental conditions. Materials and Methods

Animals Male Fischer 344 (F344) rats, each weighing 150 to 200 gm, were obtained and maintained in the animal care facility of the University of Kansas Medical Center. Tumor Cell Lines Two rat brain-tumor cell lines were used in this study: 1) RT2, a malignant glioma induced in an F344 rat by an avian sarcoma virus; 2 and 2) C6,* a glioma induced in a Wistar-Furth rat with ethylnitrosouma. 3 Five nonbrain-tumor cell lines were used: 1) 3M2N, a mammary squamous cell carcinoma induced in an F344 rat with 3-methyl-2-naphthylamine;-~ 2) CSE, a fibrosarcoma induced in an F344 rat with nickel sulfide;t 3) Yac-1, a murine lymphoid cell line that was used in cytotoxicity assays as an indicator of NK cell activity; 4) DAUDI, a murine lymphoid cell line that was used as an indicator of LAK cell activity; and 5) GIAd, a gastrointestinal adenocarcinoma that was radiation-induced in a male Holsten rat.* The RT2, 3M2N, and CSE cells were passaged in vivo and, at 6-month intervals, re-established in culture. All tumors were maintained in tissue culture using RPMI 1640 medium, supplemented with 10% fetal calf serum and antibiotics. Protocol.for Generating LAK Cells To generate LAK cells from normal lymphocytes, a * C6 glioma and GIAd cell lines obtained from American Type Culture Collection, Bethesda, Maryland. 3M2N and CSE cell lines obtained from Mason Research Institute, Philadelphia, Pennsylvania. 758

single-cell suspension of spleen cells was prepared by mechanical dissociation of spleens from untreated rats. A total of 5 x 105 spleen cells were mixed with 1 • 10~ Cetus units of recombinant human IL-2. The mixture was cultured in 1.0 ml of Dulbecco's minimum essential medium, supplemented with antibiotics, 5 x I0-' M 2-mercaptoethanol, and 10% fetal calf serum/well (complete medium) in 24-weU tissue culture plates. Cells were routinely cultured for 6 days, harvested from plates, pooled, washed, counted, and tested for cytotoxicity.

Protocol for Priming Rats to Tumor-As,,ociated Antigen Normal F344 rats were injected subcutaneously with 5 • 106 irradiated (2500 rad) RT2 tumor cells and 25 ug of Corynebacterium parvum into the left and fight groin, left and fight hind footpads, and sternum. Cells were also injected into the peritoneal cavity. The tumor cell/adjuvant mixture was evenly distributed between the multiple sites. Protocol for Cytotoxic T Lymphocyte Generation In Vitro To generate cylotoxic T lymphocytes from primed cytotoxic T lymphocyte precursors, 5 x l0 s spleen cells from RT2-primed rats were mixed with 1 x 104 irradiated (2500 tad) RT2 tumor cells and 10 Cetus units of recombinant human IL-2. The mixture was cultured in 1.0 ml of complete medium in 24-well tissue culture plates. Cells were routinely cultured for 6 days, harvested from plates, pooled, washed, counted, and tested for cytotoxicity. Assay for Cytotoxicity To test lymphocyte populations for cytotoxic activity, a 4-hour 5.Cr release assay was employed. Effector cells were obtained directly from 6-day cultures of rat spleen cells. Cells were washed twice and suspended in medium. Variable numbers of effector cells were added to a constant number of target cells in 96-well roundbottom plates to obtain the desired effector-to-target cell ratios. Target cell concentration was 1 • 105 ceils/ ml, in 100 ul/well. The cultures were incubated at 37~ in 5% CO2/95% air for 4 hours. After the incubation period, plates were centrifuged for 5 minutes at 400 G and 100-ul aliquots of supernatants were removed and assessed for released isotope in a gamma counter. Spontaneous release was obtained by incubating the target cells in culture medium alone, and maximum release was obtained by culturing target cells with 4% cetrimide. Three replicate cultures were used at each point. Percentage-specific 5~Cr release was calculated using the following formula: %specific CR release = (sample cpm - spontaneous cpm) 100 x (maximum epm - spontaneous epm)" Cytotoxicity experiments were repeated three times with reproducible results. J. Neurosurg. / Volume 77 / November, 1992

I m m u n o t h e r a p y o f gliomas

Brain Tumor Immunotherapy Model To test lymphocytes for activity against progressing brain tumors, 1 • 103 RT2 tumor cells from tissue culture were injected into the left cerebral hemisphere. Tumor cells in 5 ul of medium were injected to a depth of 4 mm through a burr hole drilled with a No. 23 needle. The reproducibility of the injection procedure was demonstrated by the low standard deviations in the mean survival times of the control rats. Routinely, on Day 5 after tumor inoculation, tumor-bearing rats were injected intravenously with test lymphocytes. Interleukin-2 therapy, which involved a 5-day course of 1 • 105 Cetus units of IL-2/day, administered intraperitoneally, was initiated simultaneously on Day 5. In some experiments, adoptive cellular therapy was initiated on Day 1 following initiation of tumor growth. Surviving rats were allowed to live for 3 months to insure that the treatment produced tumor rejection and did not simply delay tumor recurrence. Statistical Analysis Statistical analysis of the differences between the levels of cytotoxicity was performed using the NewmanKeuls' multiple comparison test following one-way analysis of variance (ANOVA) for repeated measures using BMDP statistical software.J; Because the data were obtained as percentages, arcsine transformation was performed prior to ANOVA. Statistical analysis of the differences in survival time between treatment groups was performed using the Wilcoxon matchedpairs signed-rank test. Results

Cytotoxicity of Spleen Cells From Normal Rats Direct cytotoxicity testing of cells from normal rats revealed that unstimulated spleen cells exhibited only low levels of cytotoxicity (< 20% at the 100:1 effectorto-target cell ratio) against the various target cells used in this study (Fig. 1A). Normal spleen cells cultured with low concentrations of IL-2 (10 Cetus U/ml) proliferated vigorously and cells harvested from those cultures also exhibited only low levels of cytotoxicity against the various target cells. The cytotoxicity profile for cells generated in response to low concentrations of IL-2 was indistinguishable from that of the uncultured spleen cell population. When normal spleen cells were cultured with high concentrations of IL-2 (1000 Cetus U/ml), a high degree of proliferation was observed, and cells harvested from the cultures exhibited a high degree of cytotoxicity against all of the target cells tested (Fig. 1B). This cell population was the only one tested that exhibited more than baseline levels ofcytotoxicityagainst DAUDI cells, the LAK cell target. It was also the only cell population that exhibited uniformly high levels of cytotoxicity BMDP statistical software (1988 version) supplied by University of California, Berkeley,California.

J. Neurosurg. / Volume 77/November, 1992

against RT2, C6, 3M2N, and CSE cells. These results demonstrated that none of the target cells was inherently resistant to lymphocyte-mediated cytotoxicity.

Cvtotoxicity of Spleen Cells From R T2-Primed Rats Spleen cells from rats primed with RT2 and C. parvum were not significantly (p > 0.05) more cytotoxic than unprimed spleen cells. Culturing spleen cells from primed rats with 10 Cetus units of IL-2/ml failed to generate cytotoxic cells (Fig. 1C). In contrast, culturing spleen cells from primed rats with 1000 Cetus units of IL-2-generated cells that exhibited a level of cytotoxicity similar to high IL-2-stimulated unprimed spleen cells. When cells from RT2-primed rats were cultured with RT2 tumor cells and 10 Cetus U/ml of IL-2, a high degree of proliferation occurred, and the harvested cells were highly cytotoxic for syngeneic (RT2) but not allogeneic (C6) brain tumor target cells (Fig. I D). Those cytotoxic cells exhibited only low levels of cytotoxicity against C6, 3M2N, CSE, and DAUDI cells. At all effector-to-target cell ratios, the degree of killing of syngeneic brain tumor targets by the specifically stimulated cytotoxic cells was significantly greater than cytotoxicity exhibited by any of the control effector cell populations (p _ 0.001) but was not significantly (p > 0.05) greater than the degree of killing of brain tumor targets by Ik-2-activated cytotoxic cells. Natural Killer Cell Activity Unstimulated spleen cells, normal and RT2-primed spleen cells cultured with Ik-2, LAK cells, and lymphocytes generated in the manner of specific cytotoxic T lymphocytes demonstrated a significantly (p _< 0.05) higher killing of Yac- 1 cells, the NK cell target, than of other targets, kymphokine-activated killer cells and lymphocytes generated in the manner of specific cytotoxic T lymphocytes exhibited significantly (p < 0.001) greater killing of Yac-1 cells than did unstimulated spleen cells and normal and RT2-primed spleen cells cultured with IL-2. There was no relationship between Yac-I cell killing and the in vivo efficacy of LAK cells or cytotoxic T lymphocytes (see below). Immunotherapy With LAK Cells The mean survival time for untreated rats or rats bearing IL-2-treated RT2 cells was 14 to 17 days (Fig. 2 k'/?). Treatment of RT2 tumor-beating rats with up to 1 to 2 x 10s LAK cells and systemic IL-2 had no significant effect (p _> 0.05) on rat survival (Fig. 2 left). Treatment of rats with a similar number of IL-2-activated LAK cells and IL-2 on Day 5 of tumor progression also had no effect on rat survival (Fig. 2 left). Immunotherapy With RT2-SpecOqc Cvtotoxic T L yrnphoo,tes All RT2 glioma-bearing rats treated with 1 to 2 x l0 s RT2-specific cytotoxic cells lived for at least 3 months (Fig. 2 right). Surviving rats exhibited no evi759

F. P. Holladay, T. Heitz, and G. W. Wood

FIG. 1. Graphs showing cytotoxicity assay of splenic lymphocytes. For a description of cell lines, see Materials and Methods section. A: Lymphocytes obtained from unprimed rats. B: Lymphocytes obtained from unprimed rats stimulated for 6 days in vitro with 1000 U/ml ofinterleukin-2. C: Lymphocytes obtained from rats primed with RT2 tumor cells and Corynebacterium parvum and stimulated for 6 days in vitro with 10 U/ml of interleukin-2. D: Lymphocytes obtained from rats primed with RT2 tumor cells and C. parvum and stimulated for 6 days in vitro with RT2 tumor cells and 10 U/ml of interleukin-2.

dence of neurological dysfunction, and no residual tumor was histologically evident at autopsy. Discussion Brain tumors constitute a difficult test for immunological approaches to therapy. Humoral and cellular immune responses have rarely been demonstrated in tumor-bearing individuals, and immunity has been difficult to demonstrate in experimental brain tumor models, particularly when animals have been exposed to tumor cells by the intracerebral route. Perhaps more important, tumors growing in the brain are believed to be relatively inaccessible to cellular immune responses generated systemically. For example, patients whose systemic tumor has been successfully treated with biological response modifiers often will succumb to brain metastases. ~ Moreover, various forms of immunotherapy that are effective against experimental tumors in other sites are ineffective against the same tumors growing in the brain. Nevertheless, brain tumors have been a frequent target for immunological studies because they grow as a single focus and rarely metastasize. Cellular I m m u n o t h e r a p y

Two general approaches to adoptive cellular immu760

notherapy, one employing nonspecific effector cells (LAK cells) and the other employing specific cytotoxic cells (cytotoxic T lymphocytes), have enjoyed success against histopathologically distinct tumors in different body locations, although to date neither approach has been effective against brain tumors. Recent studies in our laboratory demonstrated that it is possible to produce cytotoxic T lymphocytes reactive against malignant brain tumor cells, t~ and, more important, preliminary studies demonstrated that those polyclonal cytotoxic cell populations are active against progressing gliomas in vivo. ~~ Since our preliminary studies had demonstrated that both LAK cells and cytotoxic T lymphocytes were highly cytotoxic against brain tumor targets in vitro, we decided to compare the in vivo efficacy directly under comparable experimental conditions. Making these direct comparisons was important. When a population of cytotoxic cells fails to affect tumor progression in vivo, as has been the case for LAK cells against brain tumors, it is difficult to determine whether the lack of an effect is caused by an inherent resistance of the progressing tumor to cellular immunotherapy or to some technical deficiency in the delivery of cytotoxic cells. The current study confirmed previous observations that, despite susceptibility to J. Neurosurg. / Volume 77 / November, 1992

Immunotherapy of gliomas

FIG. 2. Survival data for different rat groups. Left: Rats injected intracerebrally with 103 RT2 tumor cells, then treated on Day 1 either with interleukin-2 (IL-2, open circles), for a mean survival time of 16 days (20 rats), or with >_ 108 lymphokine-activated killer (LAK) cells and IL-2 (filled circles), for a mean survival time of 15 days (20 rats). Data were pooled from two experiments. Right: Rats injected intracerebrally with 103 RT2 cells, then treated on Day 5 either with IL-2 (open circles), for a mean survival time of 15 days (22 rats), or with _> 108 cytotoxic T lymphocytes (CTL's) and systemic IL-2 (filled circles). All 24 rats in the second group survived for at least 90 days. Data were pooled from four experiments.

LAK cell killing in vitro, brain tumor cells were resistant to LAK cell therapy in vivo. Thus, the ability to kill tumor cells was not sufficient by itself to imbue cytotoxic cell populations with therapeutic efficacy. However, the study demonstrated that brain tumors are not inherently resistant to immune cells in vivo because similar numbers of immunologically specific cytotoxic cells were therapeutically effective.

Interleukin-2 These studies were important as controls for the therapeutic effects of the specific immune effector populations. It was particularly important to control for the effects of IL-2 because cytotoxic T lymphocytes generated in vitro in the presence of IL-2 were therapeutically effective only when combined in vivo with systemic I L-2.1~Natural killer cell activity was increased when primed lymphocytes were exposed to low concentrations of IL-2 plus tumor cells and to high concentrations of IL-2 alone, yet increased NK cell activity did not correlate with in vivo therapeutic activity. Although IL-2 has been shown to have therapeutic efficacy by itself against certain tumors, IL-2 alone and IL-2 combined with nonspecific cytotoxic cells had no effect on brain tumor progression. Cytotoxic T L ymphocytes This study has added significance because, although there have been several previous reports of LAK cell activity against brain tumor cells, there are few reports of the generation of cytotoxic T lymphocytes with reactivity against malignant gliomas. The RT2-specific effector cells generated in the current study are described as cytotoxic T lymphocytes because they exhibited all of the typical characteristics of cytotoxic T lymphocytes: 1) they could only be generated using a specific immunization protocol; j~ 2) killing was specific and major histocompatibility gone complex productrestricted; and 3) the cytotoxic cell populations were J. Neurosurg. / Volume 77/November, 1992

composed almost exclusively of T lymphocytes, l~ All rat spleen cell populations tested exhibited low levels of cytotoxicity against the various tumor targets. The current study is important because the RT2specific cytotoxic cells were able to reject progressing intracerebral gliomas following intravenous administration. We interpret those data to mean that the brain is not protected from adoptive cellular immunotherapy. This is particularly important with gliomas such as RT2 that diffusely infiltrate the brain parenchyma and are incurable by other means. The data suggest that effector ceils must migrate from the blood vessels and infiltrate the tumor mass in numbers sufficient to eradicate the tumor or activate secondary effector mechanisms. An alternative explanation for the data is that the transferred effector cells have systemic effects that result in tumor regression. Additional studies employing labeled effector cells are in progress to resolve this issue. The observation that effector cell populations were only active against brain tumors when administered along with systemic IL-2 suggests that the administered cell populations must undergo further IL-2-stimulated proliferation in vivo. An alternative possibility is that IL-2 increased vascular permeability, thereby increasing the number of effectors entering the tumor. The susceptibility of brain tumor cells both to in vitro cytotoxicity and to specific rejection in vivo demonstrates that brain tumors can express antigens that render them susceptible to immunotherapy. Those observations raise the possibility that human central nervous system malignancies may be treatable using comparable protocols.

Acknowledgments We thank Dr. Suyu Shu, The University of Michigan, for helpful criticism and Dr. Masahiro Chiga, Director, Clinical Laboratories, University of Kansas Medical Center, and Dr. Paul O'Boynick, Division of Neurosurgery, University of 761

F. P. Holladay, T. Heitz, and G. W. Wood Kansas Medical Center, for their support. We thank Cctus Corporation for providing us with recombinant human 1I-2. References 1. Barba D, Saris SC, Holder C, et al: Intratumoral LAK cell and interleukin-2 therapy of human gliomas, a Neurosurg 70:175-182, 1989 2. Beckman WC Jr, Powers SK, Brown JT, et al: Differential retention of rhodamine 123 by avian sarcoma virusinduced glioma and normal brain tissue of the rat in vivo. Cancer 59:266-270, 1987 3. Benda P, Lightbody J, Sato G, et al: Differentiated rat glial cell strain in tissue culture. Science 161:370-371, 1968 4. Cheever MA, Greenberg PD, Fefer A: Specific adoptive therapy of established leukemia with syngeneic lymphocyles sequentially immunized in vivo and in vitro and nonspecifically expanded by culture with interleukin 2. a Immunol 126:1318-1322, 1981 5. Greenberg PD: Therapy of murine leukemia with cyclophosphamide and immune Lyt-2+ cells: cytolytic T cells can mediate eradication of disseminated leukemia. J Immunol 136:1917-1922, 1986 6. Greenberg PD, Cheerer MA, Fefer A: Eradication of disseminated routine leukemia by chemoimmunotherapy with cyclophosphamide and adoptively transferred immune syngeneic Lyt-l+2 lymphocytes. 3 Exp Med 154: 952-963, 1981 7. Greenberg PD, Klarnet JP, Kern DE, et at: Therapy of disseminated tumors by adoptive transfer of specifically immune T cells. Prog Exp Tumor Res 32:104-127, 1988 8. Grimm EA, Owen-Shaub L: The IL-2 mediated amplification of cellular cytotoxicity. J Cell Biochem 4B: 335-339, 1991 9. Herberman RB: Natural killer cells. Annu Rev IMed 37: 347-352, 1986 10. Holladay FP, Heitz T, Chert YL, et al: Successful treatment of a malignant rat glioma with cytotoxic T lymphocytes. Neurosurgery (In press, 1992). 11. Holladay FP, Lopez G, De M, et al: Generation of cytotoxic immune responses against a rat glioma by in vivo priming and secondary in vitro stimulation with tumor cells. Neurosurgery 311:499-505, 1992 12. Jacobs SK, Wilson DJ, Kornblith PL, et al: Interleukin-2 and autologous lymphokine-activated killer cells in the treatment of malignant glioma. Preliminary report. J Neurosurg 64:743-749, 1986 i3. Kaye AH: Adjuvant treatment of malignant brain turnouts. Aust NZ J Surg 59:831-833, 1989 14. Kruse CA, Lillehei KO, Mitchell DH, et al: Analysis of interleukin 2 and various effector cell populations in adoptive immunmherapy of 9L rat gliosarcoma: allogeneic cytotoxic T lymphocytes prevent tumor take. Proe Natl Acad Sei USA 87:9577-9581, 1990 15. Lillehei KO, Mitchell DH, Johnson SD, et al: Long-term follow-up of patients with recurrent malignant gliomas treated with adjuvant adoptive immunotherapy. Neurosurgery 28:16-23, 1991 16. Mahaley MS, Urso MB, Whaley RA, el al: lmmunobiology of primary intracranial tumors. Part I0: Therapeutic efficacy of interferon in the treatment of recurrent

762

gliomas. J Neurosurg 63:719-725, 1985 17. McCutcheon IE, Baranco RA, Katz DA, et al: Adoptive immunotherapy of inlraccrebra[ metastases in mice. J Neurosurg 72:102-109, 1990 18. Merchant RE, Merchant LH, Cook SHS, et al: lntralesional infusion of lymphokine-acti~ated killer (LAK) cells and recombinant interleukin-2 (rlL-2) for the treatment of patients with malignant brain tumor. Neurosurgery 23:725-732, 1988 19. Mitchell MS: Relapse in the central nervous system in melanoma patients successfully treated with biomodulators. J Clin Oncol 7:1701-1709, 1989 20. North RJ: The murine antitumor immune response and its therapeutic manipulation. Adv lmmunol 35:89-155, 1984 2l. Ortaldo JR: Comparison of natural killer and natural cytotoxic cells: Characteristics, regulation, and mechanism of action. Pathol Immunopathol Res 5:203-218, 1986 22. Rosenherg SA: The development of new immunotherapies for the treatment of cancer using interleukin-2. Ann Surg 208:121 - 135, 1988

23. Sakai K, Chang AE, Shu S: Effector phenotype and immunologic specificityof T cell-mediated adoptive therapy for a routine tumor that lacks intrinsic immunogenicity. Cell Immunol 129:241-255, 1990 24. Shapiro WR: Treatment of neuroectodermal brain tumors. Ann Neurol 12:231-237, 1982 25. Shu S, Chou T, Rosenberg SA: Generation from tumorbearing mice of lymphocytes with in vivo therapeutic efficacy, a lmmunol 139:295-304, 1987 26. Shu S, Chou T, Rosenberg SA: In vitro sensitization and expansion with viable tumor cells and interleukin 2 in the generation of specific therapeutic effector cells, a lmmunol 136:3891-3898, 1986 27. Shu S, Chou T, Sakai K: Lymphocytes generated by in vivo priming and in vitro sensitization demonstrate therapeutic efficacy against a murine tumor that lacks apparent immunogenicity, a Immunol 143:740-748, 1989 28. Townsend A, Bodmer H: Antigen recognition by class Irestricted T lymphocytes. Annu Rev Immunol 7:601-624, 1989 29. Tzeng JJ, Barth RF, Clendenon NR, et al: Adoptive immunotherapy of a rat glioma using lymphokine-activated killer cells and interleukin 2. Cancer Res 50: 4338-4343, 1990 30. Walker MD, Alexander E Jr, Hunt WE, et at: Evaluation of BCNU and/or radiotherapy in the treatment of ariaplastic gliomas. A cooperative clinical trial. J Neurosurg 49:333-343, 1978 31. Walker MD, Green SB, Byar DP, el al: Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 303:1323-1329, 1980 Manuscript received January, 16, 1992. Accepted in final form April 13, 1992. Address reprint requests to. Gary W. Wood, Ph.D., Department of Pathology, University of Kansas Medical Center, 391h and Rainbow Boulevard, Kansas City, Kansas 661067410.

J. Neurosurg. / Volume 77 / November. 1992

Antitumor activity against established intracerebral gliomas exhibited by cytotoxic T lymphocytes, but not by lymphokine-activated killer cells.

Specific immune responses against malignant brain tumors have been difficult to demonstrate. Moreover, immunotherapy has met with little success, desp...
1MB Sizes 0 Downloads 0 Views