Tumor-specific Binding of Radiolabeled Monoclonal

Jun

Antibody

YOSHIDA,

Toshihiko

Kenichiro

SUGITA,

Masanori

in Glioma

WAKABAYASHI, Hisao

TADOKORO*

SEO**, and

Sadayuki

Patients

Masaaki

Motoo

G-22

MIZUNO,

OSHIMA*, SAKUMA*

Departments of Neurosurgery and *Radiology, Nagoya University School of Medicine, Nagoya; **Department of Endocrinology and Metabolism, Research Institute of Environmental Medicine, Nagoya University, Nagoya

Abstract

Iodine-131-labeled G-22 monoclonal antibody F(ab')2 fragment reacting specifically with a glioma-associated surface glycoprotein was administered to 12 glioma patients to investigate its use in radioimag ing of intracranial gliomas. No immediate or delayed side effects were attributable to antibody injec tion. Nine patients received the radiolabeled complex intravenously. The images of low-grade gliomas were generally poor and disappeared within 4 days. High-contrast images were obtained beyond the 7th day in high-grade gliomas except one case in the pineal region. Three patients received intraven tricular or intratumoral administration. Clear images of all tumors were demonstrated from the 2nd until later than the 7th day. One patient with cerebrospinal fluid (CSF) dissemination of brainstem glioma demonstrated negative CSF cytology after intraventricular administration. Keywords:

monoclonal

antibody,

glioma,

radioimaging

Introduction Patients with malignant glioma have a very poor prognosis because of the aggressive infiltration into normal brain tissue .3,13,14 Successful treatment of this formidable disease requires early and accurate diagnosis of the tumor and targeted therapy for selec tive killing of tumor cells. Monoclonal antibody (MCA) specifically reacting to tumor cells is a very promising reagent for this purpose. 11,12.16)Our laboratory has produced a variety of MCAs against glioma-associated antigens and one, designated G 22, has already entered clinical use. Previously, we established a solid-phase radioimmunoassay for G 22 MCA reactive antigen and demonstrated that quantitative determination of antigen in the cerebrospinal fluid (CSF) is useful in the diagnosis and monitoring of human glioma.17) We also reported an experimental study on the radioimaging of human glioma xenografts with radiolabeled G-22 MCA.15) Received 1991

April

25,

1991;

Accepted

September

20,

Here, we report an investigation of radiolabeled G-22 MCA F(ab')2 fragment in the radioimaging diagnosis of malignant glioma. Materials

and

Methods

This study included 12 patients with primary brain tumors, seven males and five females aged from 4 to 61 years. Histological examination revealed glioblastoma in six patients, anaplastic astrocytoma in two, and low-grade astrocytoma in four. All pa tients had previously received surgery. Radiolabeled MCA investigation was directed at recurrence in eight and residual glioma in four cases. Eight tumors were located in the cerebral hemispheres, two in the brainstem, and one in each of the thalamus and pineal region. G-22 mouse MCA reacting with a surface glycopro tein with a molecular weight of 67,000 Dalton was used. 12,16)The antigen is expressed in many human glioma cell lines and human glioma tissue, but not in normal adult brain. 12,16) Several independent laboratories have shown MCA to be non-pyrogenic and sterile. The present study was approved by the

ethical and isotope committees at Nagoya Univer sity. Informed consent was given by each patient. The F(ab')2 fragment of G-22 MCA was labeled with iodine-131 (31I) (Daiichi Radioisotope Laboratories, Chiba) using a modified chloramine T technique described previously *") More than 95% of the radiolabeled complex was in a sterile filtrate at administration. All patients received 1 ml of Lugol's iodine solution daily, start ing 5 days prior to radiolabeled antibody injection, and continuing through the scanning period, i.e., up to 7 days, blocking the uptake of free 131Iby the thyroid. All patients were tested for hypersensitivity to mouse immunoglobulin G (IgG), by an intrader mal injection of 10,ug of G-22 MCA F(ab')2 frag ment. 0.3-1 mg, 14.8-91.0 MBq of the complex was dissolved in 200 ml physiological saline and given intravenously over 30 minutes in nine patients; 0.5-1.5 mg, 37.0-68.8 MBq was dissolved in 1 ml physiological saline and given through an Ommaya reservoir, intraventricularly in two cases and intra tumorally in one. The patients were imaged on a 35 cm field of view GCA-501 S gamma-camera (Toshiba Medical System Co., Ltd., Tokyo) fitted with a high-energy col limator 2-3 hours after the 1311-labeledMCA infu sion and then daily for 1 week. Lateral and frontal images of the skull, and the whole body anterior im age in a few cases, were obtained.

any side effects during or after antibody administra tion. Radioactive MCA administration achieved a positive tumor image in all cases corresponding to that identified by computed tomography (CT), magnetic resonance (MR) imaging, technetium-99m diethylenetriaminepenta-acetic acid (99mTc-DTPA) brain scanning, and positron emission tomography (Table 1). However, there were variations in uptake ratio or duration affecting the tumor imaging and radiation intensity, respectively. These variations cor related well with the administration route and the tumor histology. Intraventricular or intratumoral ad ministration resulted in high-contrast imaging of all tumors (two glioblastomas and one anaplastic astrocytoma) from the 2nd day until after the 7th day. However, the image quality depended on the tumor histology when the drug was injected in travenously. Low-grade gliomas had generally poor images which disappeared within 4 days, while good images of high-grade gliomas were obtained beyond the 7th day except one case in the pineal region. The best images after intravenous injection were ob tained on the 4th day. A therapeutic response to locally injected radiolabeled MCA was observed in one case of brainstem glioblastoma with CSF dis semination. Tumor cells detected in the CSF disap peared, although the solid tumor remained unaf fected. Representative

Case

Reports

Results Case No patient

Table 1

had a positive

skin test or demonstrated

1311-labeled G-22 MCA F(ab')2 fragment

radioimaging

2: This

glioblastoma

47-year-old in the right

female frontal

had lobe.

a recurrent Eight

months

Fig. 1 Case 2. Frontal (left) and lateral views (center) on day 4 after intravenous injection of 131 I-labeled G-22 F(ab')2 fragment, showing the uptake of activity in the right frontal lobe corresponding to the tumor site (arrow) on T1-weighted MR image (right).

Fig. 2 Case 5. Lateral view (left) on day 4 after in travenous injection of 131I-labeledG-22 F(ab')2 fragment, showing the uptake of activity in the left parieto-occipital lobe corresponding to the lateral view (right) 2 hours after 99mTc-DTPA injection.

Fig. 3 Case 11. Frontal view (left) on day 4 after in traventricular injection of 131I-labeled G-22 F(ab')2 fragment, showing the uptake of activi ty in the brainstem corresponding to the tumor site on the postcontrast CT scan (right). 0: Ommaya reservoir, T: tumor.

prior to admission, partial removal of a glioblastoma in the left frontal lobe, followed by adjuvant therapy with nitrosourea (MCNU) and radiation (60 Gy), achieved complete disappearance. The contralateral recurrence was confirmed with intravenous radio labeled G-22 MCA F(ab')2 fragment (41.1 MBq). The clear tumor images corresponded well with MR findings (Fig. 1). Case 5: This 46-year-old male had in 1983 under gone removal of an anaplastic astrocytoma in the left parieto-occipital lobe and postoperative adju vant multimodal therapy including repeated injec tion of lymphokine-activated killer (LAK) cells into the intratumoral cavity. In 1989, clinical deteriora tion with tumor progression was observed. Brain

scans with 99mTc-DTPA (740 MBq) and 13'I-labeled G-22 F(ab')2 fragment (62.9 MBq) obtained almost identical tumor images (Fig. 2).

Case 11: This 5-year-old boy with the diagnosis of brainstem glioma underwent a suboccipital craniec tomy for posterior fossa decompression and a biopsy of the tumor. Histological examination revealed a glioblastoma. Treatment with human interferon-# (HuIFN-f3), ACNU, and radiation achieved no neurological or neuroradiological improvement, and CSF dissemination of the tumor soon occurred. Radiolabeled G-22 MCA F(ab')2 fragment (68.8 MBq) was injected into the lateral ventricle through an Ommaya reservoir, demonstrating tumor images in the brainstem clearly (Fig. 3). Tumor cells in the

CSF disappeared, remained negative.

and

repeated

CSF

cytology

tests

Discussion

MCAs or their conjugates with toxin, drug, or isotopes have been safely used in humans for more than decade.') Epenetos et al. 6)first used a 123Ior 1311_ labeled MCA called 9A against epidermal growth fac tor receptor as a useful agent for diagnosis and treat ment by intracarotid infusion. Similar clinical studies used a variety of radiolabeled MCAs against glioma-associated antigen: UJ 13A,10) MUC2-63,2) and Mel-14') reacting with neuroectodermal cell sur face antigens, and 81C6 reacting with tenascin.18I The radionuclides 1311,1231,1251,and indium-111 were used for diagnosis (radioimmunodetection) and 131I and yttrium-90 for treatment (targeted radiation therapy) of gliomas. These studies revealed the poten tial for demonstrating tumors in contrast to normal brain tissue by intravenous or intra-arterial ad ministration of radiolabeled whole IgGs or F(ab')2 fragments. The average tumor:normal brain an tibody uptake ratio was 3-13:1 with UJ13A,'0) 2-18:1 with Mel-14,') and 25:1 with 81C6.18 In contrast, the specific to non-specific antibody binding ratio was 74:1 with intrathecal administration, and signifi cantly better than those obtained after systemic ad ministration. 4) In the present study, all patients with astrocytoma or glioblastoma showed an uptake of 131I-labeledG 22 MCA F(ab')2 fragment corresponding to the anatomical distribution of the brain tumors as seen by CT and/or MR imaging. High-grade gliomas demonstrated a high uptake, while low-grade gliomas had only a low uptake, corresponding with our previous study") where solid-phase radioim munoassay detected antigen at concentrations greater than 8 ng/ml in the CSF, and a CSF diagnosis of glioblastomas more than 2 cm in diameter was possible. The antigen level in the CSF tended to correlate with the tumor size and histology, with glioblastomas demonstrating higher levels than astrocytomas. Therapeutic injection of radiolabeled MCA has usually used intracarotid or intrathecal rather than systemic administration because of the better uptake ratio. Coakham et al.5) developed intrathecal radioimmunotherapy using 1311-labeled MCAs especially for neoplastic meningitis. Clinical and/or CSF responses were achieved in 1/2 pineal tumors, 3/10 medulloblastomas, 2/3 melanomas, 1/5 car cinomas, 5/6 leukemias, and 1/1 lymphoma.4) In our series, one case had a similar CSF response.

However, such therapy has never achieved a reduc tion in solid glioma size. We suggest that MCAs are better used as immunotoxins for selective transfer of cytotoxic agents to tumor cells. We previously reported that G-22 MCA could be coupled to liposomes without changing the specifici ty toward glioma cells. The MCA then binds with glioma cells, significantly increasing the cellular up take of liposomal contents.') Furthermore, selective and significant cytotoxicity toward glioma cells is ob tained when a chemotherapeutic agent (methotrex ate) or a growth-inhibiting cytokine gene (HuIFN i) is incorporated into the liposome.7'9> Based on these previous and the present studies, we are now developing targeted chemotherapy and/or gene therapy using local infusion of G-22 coupled lipo somes containing methotrexate or HuIFN-,6 gene. References 1)

2)

3) 4)

5)

6)

Behnke J, Mach J-P, Buchegger F, Carrel S, Delaloye B, de Tribolet N: In vivo localisation of radiolabelled monoclonal antibody in human glioma. Br J Neurosurg 2: 193-197, 1988 Bergh J, Nilsson S, Liljedahl C, Sivolapenko G, Maripuu E, Stavrou D, Epenetos A: In vivo imaging and treatment of human brain tumors utilizing the radiolabeled monoclonal antibody MUC2-63. An ticancer Res 10: 655-660, 1990 Bloom HJG: Combined modality therapy for in tracranial tumors. Cancer 35: 11-12, 1975 Coakham HB, Kemshead JT, Papanastassiou V, Piger BL, Lashford L, Moseley R: Targeted therapy for CNS tumors with monoclonal antibodies (MAbs), in: Abstract of International Symposium on Ad vances in Neuro-oncology. 1990, p 24 Coakham HB, Richardson RB, Bourne SB, Davis AG, Kemshead JT: Antibody guided radioim munolocalization and therapy for malignant men ingitis. Br J Cancer 52: 566, 1985 Epenetos AA, Courtenay-Luck N, Pickring D, Hooker G, Durbin H: Antibody guided irradiation of brain glioma by arterial infusion of radioactive monoclonal antibody against epidermal growth fac tor receptor and blood group A antigen. Br Med J 290: 1463-1466, 1985

7) Kito A, Yoshida J, Kageyama N, Kojima N, Yagi K: Liposomes coupled with monoclonal antibodies against glioma-associated antigen for targeting chemotherapy of glioma. J Neurosurg 71: 382-387, 1989 8)

9)

Miller RA, Levy R: Response of cutaneous T cell lym phoma to therapy using hybridoma monoclonal an tibody. Lancet 2: 226-230, 1981 Mizuno M, Yoshida J, Sugita K, Inoue I, Seo H, Hayashi Y, Koshizaka T, Yagi K: Inhibition of glioma cells transfected with the human-interferon

gene by liposomes coupled with tibody. Cancer Res 50: 7826-7829,

10)

11) 12)

13)

14)

a monoclonal 1990

an

Richardson RB, Davies AG, Bourne SP, Staddon GE, Kemshead JJ, Coakham HB: Radioim muno localization of brain tumors: Biodistribution of radiolabeled monoclonal antibody UJ13A. Eur J Nucl Med 12: 313-320, 1986 Stavrou D: Monoclonal antibodies in neuro-on cology. Neurosurg Rev 13: 7-19, 1990 Wakabayashi T, Yoshida J, Seo H, Kazo K, Murata Y, Matsui N, Kageyama N: Characterization of neuroectodermal antigen by a monoclonal antibody and its application in CSF diagnosis of human glioma. J Neurosurg 68: 449-455, 1988 Walker MD, Green SB, Byar DP, Alexander E, Batzdorf U, Brooks WH, Hunt WE, MacCarty CS, Mahaley MS, Mealey J, Owens G, Ransohoff J, Robertson JT, Shapiro WR, Smith KR, Wilson CB, Strike TA: Randomized comparisons of radiotherapy and nitrosoureas for the treatment of malignant glioma after surgery. N Engl J Med 303: 1323-1329, 1980 Yoshida J, Kobayashi T, Kageyama N: Multimodali ty treatment of malignant glioma: Effect of chemotherapy with ACNU and immunotherapy with N-CWS. Neurol Med Chir (Tokyo) 24: 19-26, 1984

15)

Yoshida J, Mizuno M, Inoue I, Wakabayashi T, Sugita K, Seo H, Chiba K: Radioimaging of human glioma xenografts with 123I labeled monoclonal an tibody G-22 against glioma associated antigen. J Neurooncol 8: 221-229, 1990

16) Yoshida J, Wakabayashi T, Kito A, Kageyama N, Murata Y, Seo H, Kojima N, Yagi K: Clinical applica tion of monoclonal antibodies against glioma-associ ated antigen. Prog Exp Tumor Res 30: 44-56, 1987 17)

18)

Yoshida J, Yamamoto R, Wakabayashi T, Nagata M, Seo H: Radioimmunoassay of glioma-associated antigen in cerebrospinal fluid and its usefulness for the diagnosis and monitoring of human glioma. J Neurooncol 8: 23-31, 1990 Zalutsky M, Moseley R, Coakham HB, Coleman E, Bigner DD: Pharmacokinetics and tumor localization of 131I-labeled anti-tenascin monoclonal antibody 81C6 in patients with glioma and other intracranial malignancies. Cancer Res 49: 2807-2813, 1989

Address reprint requests ment of Neurosurgery, Medicine, 466, Japan.

65

to:

J. Yoshida, M.D., Depart Nagoya University School of

Tsurumai-cho,

Showa-ku,

Nagoya