Comparison of in vitro glioma cell cytotoxicity of LAK cells from glioma patients and healthy subjects

Transcription

1 J Neurosurg 69: , 1988 Comparison of in vitro glioma cell cytotoxicity of LAK cells from glioma patients and healthy subjects VIDAR BOSNES, M.D., AND HENRY HIRSCHBERG, M.D. Institute of Transplantation Immunology and Department of Neurosurgery, The National Hospital, Oslo, Norway Peripheral blood mononuclear cells from 11 glioma patients and 11 healthy control subjects were cultured in medium containing recombinant interleukin-2 for a period of 5 days. The cytotoxicity of these lymphokineactivated killer (LAK) cells was tested on chromium-5 I-labeled freshly prepared allogeneic glioblastoma cells, and on the cell lines K562 (natural killer cell (NK)-sensitive) and Daudi (NK-resistant). Peripheral blood mononuclear cells from all subjects showed high levels of cytotoxicity against these targets. There was no significant difference between the patients and the control group when LAK cytotoxicity was compared. Thus, although glioma patients are known to have depressed immunological reactivity, the cytotoxic capacity of LAK cells derived from glioma patients is similar to that of LAK cells from healthy control subjects. However, the glioma patients had significantly reduced numbers of mononuclear ceils in their peripheral blood, possibly due to steroid treatment. Therefore, the volume of blood required to generate the same number of LAK cells was approximately three times larger from the glioma patients than from control subjects. KEY WORDS 9 glioma 9 lymphokine-activated killer cells 9 cytotoxicity 9 interleukin-2 YMPHOKINE-ACTIVATED killer (LAK) cells are generated when peripheral blood mononuclear cells (PBMC) from normal subjects or cancer patients are stimulated with interleukin-2 (IL-2) for 2 days or more. 8 These cells show strong cytotoxicity against a wide range of target cells, notably against malignant cells. It has been demonstrated that LAK cells can be generated from the PBMC of glioma patients, l'13 and clinical trials of immunotherapy using LAK cells in the treatment of malignant glioma have also been started? 1,12,14 The fact that LAK cells can be generated from glioma patients is interesting in view of the suppression of cellular immune functions that is known to occur in these patients, as evidenced by impaired delayed cutaneous hypersensitivity, 4,16 reduced numbers of circulating lymphocytes, 5,~6 depressed natural killer cell (NK) activity, 2 and decreased in vitro proliferative responses against mitogens, 6'~7-z4 common microbial antigens, 4 and allogeneic lymphocytes. 3 The purpose of the present study was to investigate whether the suppression of cellular immune functions seen in glioma patients affects the cytotoxicity of their IL-2-activated PBMC. Clinical Material and Methods Patients and Normal Control Subjects The 11 patients included in this study underwent operative debulking for astrocytoma or glioblastoma (Table 1). None had received prior irradiation or chemotherapy, but all were receiving steroids (dexamethasone, 4 to 6 mg daily) at the time of evaluation. Donors of normal lymphocytes were healthy laboratory staff or patients operated on for cervical spinal disorders of a nonmalignant nature. Preparation of Cytotoxic Cells Mononuclear cells were isolated from the peripheral blood by Ficoll-metrizoate flotation of defibrinated blood, and were cultured for 5 days at 1 106/ml in RPMI 1640 medium supplemented with 10% human serum and 50 units/ml of recombinant IL-2 (ALA-125) in upright Falcon 3013 flasks.* On Days 2 and 4, half * Ficoll-metrizoate (Lymphoprep) supplied by Nyco, Oslo, Norway; IL-2 supplied by AMGen Biologicals, Amersham, United Kingdom. 234 J. Neurosurg. / Volume 69/August, 1988

2 LAK cytotoxicity against gliomas TABLE 1 Clinical summary in 11 glioma patients Case Age (yrs), Symptoms Location Histology No. Sex D~s Steroids* Dose/Day(m~ 1 33, F seizures, headache rt parieto-occipital astrocytoma > , M seizures rt frontal astrocytoma , M seizures, dysphasia, personality change It temporo-occipital glioblastoma , F headache, visual hallucinations rt occipitoparietal glioblastoma , M headache, personality change rt frontal glioblastoma , M seizures It parietal glioblastoma , M dysphasia, rt hemiparesis It temporoparietal glioblastoma , M seizures rt parieto-occipital glioblastoma , M It hemiparesis, personality change rt frontal glioblastoma , M seizures, headache rt frontal astrocytoma (recurrence) , M seizures It hemiparesis It frontal glioblastoma 3 16 * The number of days of treatment with dexamethasone (before a blood sample was obtained for the generation of lymphokine-activated killer cells) and the average dose during this period. of the medium was replaced with fresh medium supplemented with IL-2. Control cultures without IL-2 were included in all experiments. Before the assay, the effector cells were washed and resuspended in medium supplemented with 20% human serum. Cytotoxic Assay The target cells used were fresh cryopreserved glioblastoma cells, K562 erythroleukemia cells (susceptible to lysis by NK's) and Daudi (Burkitt's lymphoma) cells (NK-resistant). One glioblastoma patient who was not included as a LAK cell donor in this study served as the source for allogeneic glioblastoma target cells, which were prepared as described previously? The cell lines were kept in culture in RPMI 1640 medium supplemented with 10% fetal calf serum and subcultured three times weekly. Target cells (4 x 106) were incubated in 4 ml of medium supplemented with 20% human serum and 5 x 106 Bq Na25~CrO4 18 hours before the assay. On the morning of the assay the target cells were washed three times at 0*C and resuspended in medium supplemented with 20% human serum. Then 2000 chrornium-51 (5~Cr)-labeled target cells in 100 gl medium were added to varying numbers (160,000, 40,000, 10,000, or 2500) of effector cells in round-bottomed 96-well microtiter plates. The plates were incubated for 4 hours and harvested with the Skatron superuatant harvesting systemt as previously described. 1~ The percentage of specific chromium release was calculated by the formula: experimental release - spontaneous release x 100%, maximal release - spontaneous release based on the mean count per minute of triplicate wells for the experimental release, and the mean count per t Supernatant harvesting system manufactured by Skatron, A/S, Lierbyen, Norway. minute of 12 replicates for the spontaneous and maximal release. Mononuclear Cell Yield Results The number of mononuclear cells obtained per milliliter of blood after Ficoll-metrizoate flotation was significantly lower in the patients ( / ml, mean _ standard deviation) than in the healthy control subjects ( /ml; p < 0.005, Student's t-test) (Fig. 1 left). The cell yield after 5 days of culture in IL-2-containing medium, as judged by the FIG. 1. Left: Yield of peripheral blood mononuclear cells (PBMC) per ml of blood in 11 glioma patients (P, circles) and 11 healthy control subjects (N, triangles) after Ficoll-metrizoate flotation. Right: Yield of mononuclear cells from 11 glioma patients (P, circles) and 11 healthy control subjects (N, triangles) following culture in medium containing interleukin-2. The data are expressed as the ratio of the number of cells collected on Day 5 to the number of cells put into culture. J. Neurosurg. / Volume 69/August,

3 V. Bosnes and H. Hirschberg FIG. 2. Cytotoxicity against allogeneic glioblastoma cells exerted by interleukin-2 (IL-2)-activated peripheral blood mononuclear cells (PBMC, Panel A) or by PBMC that had been cultured for 5 days without IL-2 (Panel B). The cytotoxic cells were derived from 11 glioma patients (circles) and 11 healthy control subjects (triangles). FIG. 3. Cytotoxicity against K562 cells exerted by interleukin-2 (IL-2)-activated peripheral blood mononuclear cells (PBMC, Panel A) or by PBMC that had been cultured for 5 days without IL-2 (Panel B). The cytotoxic cells were derived from 11 glioma patients (circles) and 11 healthy control subjects (triangles). ratio of the number of cells collected on Day 5 to the number of cells put into culture, was also lower in the patients (ratio 0.64:1 _ 0.37:1) than in the control subjects (ratio 0.79:1 _ 0.24:1)(Fig. i right). However, this difference was not statistically significant. As a consequence of these two effects, the volume of blood required to generate 106 IL-2-activated cells was considerably larger in the patients ( ml) than in the control subjects (1.3 _ 0.5 ml; p < 0.002, Student's t- test). Cytotoxicity The cytotoxicity was then compared against allogeneic glioma cells of LAK ceils from 11 glioma patients and 11 healthy control subjects. The results are given in Fig. 2. The cytotoxicity was also compared against K562 cells (Fig. 3) and against Daudi cells (Fig. 4). All patients and control subjects showed strong cytotoxicity against these three targets. At an effector to target (E:T) ratio of 20:1, the mean cytotoxicity against allogeneic glioma cells was 37% _ 15% in the patients and 42% + 14% in the control group (Fig. 2). At the same E:T ratio, the cytotoxicity against K562 was 79% _ 8% and 83% _ 7% (Fig. 3), and the cytotoxicity against Daudi cells was 77% _ 8% and 80% _ 8% (Fig. 4), in patients and control subjects, respectively. At none of the E:T ratios tested was there any significant difference between patients and control subjects in their cytotoxicity against these targets (Student's t-test). In some individuals in both groups, there was cytotoxicity against K562 in the control cultures (PBMC cultured for 5 days without IL-2). This presumably represents residual NK activity, since there was no cytotoxicity against Daudi ceils or allogeneic glioma cells in the unstimulated cultures. Discussion This study demonstrated that LAK cells derived from glioma patients show similar levels of cytotoxicity to those derived from healthy individuals. This is a surprising finding, considering the compromised status of cellular immunity in glioma patients. 2-6'16'17'24 Glioma patients have depressed NK activity, 2 and the NK population is believed by many investigators to contain the precursor cells which develop into LAK cells upon IL- 2 stimulation. 9'19'21 One might therefore have expected the suppression of NK activity in glioma patients to have led to a suppression of LAK activity. It appears that IL-2-activated cytotoxicity is a quite robust immunological function, which is unaffected in patients who in other respects are immunosuppressed. A marked reduction was found in the number of circulating mononuclear cells (judged by the yield after Ficoll-metrizoate centrifugation). This is consistent with previous reports 5'16 of lymphocytopenia in glioma patients who did not receive corticosteroids. In the present study, all patients received dexamethasone, which is known to cause a profound reduction in the number of circulating lymphocytes. 7'2~ It is therefore reasonable to regard the reduced number of PBMC seen in this study as a compound effect of the steroid treatment and the disease itself. The decreased LAK yield/ml of blood does limit the number of LAK cells that it is possible to obtain from glioma patients. This may be overcome by obtaining blood samples from the patients in a steroid-free interval. Additionally, the LAK cells could be expanded through a culture period longer than the currently used 3 to 7 days, as suggested by Ochoa and coworkers.18 In their Phase I clinical trial, Jacobs and coworkers ~1'12'14 used to 1 X 101~ 236 J. Neurosurg. / Volume 69/August, 1988

4 LAK cytotoxicity against gliomas FIG. 4. Cytotoxicity against Daudi ceils exerted by interleukin-2 (IL-2)-activated peripheral blood mononuclear cells (PBMC, Panel A) or by PBMC that had been cultured for 5 days without IL-2 (Panel B). The cytotoxic cells were derived from 11 glioma patients (circles) and 11 healthy control subjects (triangles). LAK cells, administered by direct injection into the brain tissue surrounding the cavity left after debulking. In comparison, Rosenberg and coworkers 22,23 have used cumulative doses of 1 x 101~ to 2 x 10 tt cells administered by repeated intravenous infusions over a 2- to 3-week period in patients with disseminated, nonglial cancer. Successful immunotherapy of glioblastoma might call for cell numbers in this range, possibly involving repeated local injections. Kitahara, et al, t5 treated five patients with malignant glioma by repeated injections of autologous T lymphocyte lines into the tumor bed via an Ommaya reservoir placed in the subgaleal space.~s Similar schemes could be envisaged using LAK cells. In summary, our results indicate that although LAK cells derived from glioma patients are highly cytotoxic against glioma cells, the practical implementation of LAK cell therapy in glioma patients might be limited by low yields of LAK ceils. Further studies are necessary to optimize the methods for production of LAK cells from glioma patients. Acknowledgment The technical assistance of Inga Skafladottir is greatly appreciated. References 1. Bosnes V, Hirschberg H: Human interleukin-2 activated cytotoxic cells kill autologous glioma cells in vitro. J Neurooncol 6:85-92, Braun DP, Penn RD, Harris JE: Regulation of natural killer cell function by glass-adherent ceils in patients with primary intracranial malignancies. Neurosurgery 15: 29-33, Brooks WH, Caldwell HD, Mortara RH: Immune responses in patients with gliomas. Surg Neurol 2: , Brooks WH, Netsky MG, Normansell DE, et al: Depressed ceu-mediated immunity in patients with primary intracranial tumors. Characterization of a humoral immunosuppressive factor. J Exp Med 136: , Brooks WH, Roszman TL, Mahaley MS, et al: Immunobiology of primary intracranial tumours. II. Analysis of lymphocyte subpopulations in patients with primary brain tumours. Clin Exp Immunol 29:61-66, Elliott LH, Brooks WH, Roszman TL: Activation of immunoregulatory lymphocytes obtained from patients with malignant gliomas. J Neurosurg 67: , Fauci AS: Mechanisms of corticosteroid action of lymphocyte subpopulations. II. Differential effects of in vivo hydrocortisone, prednisone and dexamethasone on in vitro expression of lymphocyte function. Clin Exp Immunol 24:54-62, Grimm EA, Mazumder A, Zhang HE, et al: Lymphokineactivated killer cell phenomenon. Lysis of natural killerresistant fresh solid tumor cells by interleukin 2-activated autologous human peripheral blood lymphocytes. J Exp Med 155: , Herberman RB, Hiserodt J, Vujanovic N, et al: Lymphokine-activated killer cells activity. Characteristics of effector cells and their progenitors in blood and spleen. Immunol Today 8: , Hirschberg H, Skare H, Thorsby E: Cell mediated lympholysis: CML. A microplate technique requiring few target cells and employing a new method of supernatant collection. J Immunol Meth 16: , Jacobs SK, Wilson DJ, Kornblith PL, et al: Interleukin-2 and autologous lymphokine-activated killer ceils in the treatment of malignant glioma. Preliminary report. J Neurosurg 64: , Jacobs SK, Wilson DJ, Kornblith PL, et al: Interleukin-2 or autologous lymphokine-activated killer cell treatment of malignant glioma: Phase I trial. Cancer Res 46: , Jacobs SK, Wilson D J, Kornblith PL, et al: In vitro killing of human glioblastoma by interleukin-2-activated autologous lymphocytes. J Neurosurg 64: , Jacobs SK, Wilson DJ, Melin G, et al: Interleukin-2 and lymphokine-activated killer (LAK) cells in the treatment of malignant glioma: clinical and experimental studies. Nenrol Res 8:81-87, Kitahara T, Watanabe O, Yamaura A, et al: Establishment of interleukin-2 dependent cytotoxic T lymphocyte cell line specific for autologous brain tumor and its intracranial administration for therapy of the tumor. J Neurooneol 4: , Mahaley MS Jr, Brooks WH, Roszman TL, et al: Immunology of primary intracranial tumors. Part 1: Studies of cellular and humoral general immune competence of brain-tumor patients. J Neurosurg 46: , Neuwelt EA, Kikuchi K, Hill S, et al: Immune responses in patients with brain tumors. Factors such as anti-convulsants that may contribute to impaired cell-mediated immunity. Cancer 51: , Ochoa AC, Gromo G, Alter B J, et al: Long-term growth of lymphokine-activated killer (LAK) cells: role of anti- CD3, /3-IL 1, interferon--/ and -/3. J lmmunol 138: , Ortaldo JR, Mason A, Overton R: Lymphokine-activated killer cells. Analysis of progenitors and effectors. J Exp Med 164: , 1986 J. Neurosurg. / Volume 69/August,

5 V. Bosnes and H. Hirschberg 20. ParriUo JE, Fauci AS: Mechanisms of corticosteroid action on lymphocyte subpopulafions. III. Differential effects of dexamethasone administration on subpopulations of effector cells mediating cellular cytotoxicity in man. Clin Exp Immunol 31: , Phillips JH, Lanier LL: Dissection of the lymphokineactivated killer phenomenon. Relative contribution of peripheral blood natural killer cells and T lymphocytes to cytolysis. J Exp Med 164: , Rosenberg SA, Lotze MT, Muul LM, et al: Observations on the systemic administration of autologous lymphokine-activated killer cells and recombinant interleukin-2 to patients with metastatic cancer. N Engl J Med 313: , Rosenberg SA, Lotze MT, Muul LM, et al: A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and inter- leukin-2 or high-dose interleukin-2 alone. N Engl J Med 316: , Roszman TL, Brooks WH: Immunobiology of primary intracranial tumours. III. Demonstration of a qualitative lymphocyte abnormality in patients with primary brain tumours. Clin Exp Immunol 39: , 1980 Manuscript received October 5, Dr. Bosnes is a research fellow supported by the Norwegian Cancer Society. Support for Dr. Hirsehberg was provided from the Norwegian Society for Fighting Cancer. Address reprint requests to: Vidar Bosnes, M.D., Institute of Transplantation Immunology, Rikshopitalet, N-0027 Oslo 1, Norway. 238 J. Neurosurg. / Volume 69/August, 1988