岩立 康男

A slowly proliferating cell fraction in tumors shows reduced sensitivity to cell cycle-dependent anticancer agents. To understand the molecular basis of drug resistance observed in brain tumors, we examined the relationship between alteration of p16, a cyclin dependent kinase inhibitor whose functions are frequently lost in many human gliomas, and chemosensitivity of tumor cells to various kinds of anticancer agents. Alterations of the p16 gene that include mutation(s) and homozygous deletion as well as p16 protein expression level, were examined in 56 specimens of astrocytic tumors. Their in vitro chemosensitivities to 30 kinds of anticancer agents were analyzed with flow cytometry which detects drug-induced cell death. We found that the alterations were correlated with increased sensitivity to antimetabolite anticancer agents but not with other kinds of agents, including alkylating agents, antibiotics, topoisomerase inhibitors and antimicrotubule agents. The present results suggest that p16 plays a role in determining chemosensitivity of brain tumors, depending on pharmacological mechanisms of anticancer agents. Proper understanding of the molecular machinery which regulates the chemosensitivity may contribute to the choice of anticancer agents for individual patients.

Astrocytic tumours often become resistant to a variety sf chemotherapeutic agents in advanced stages and frequently possess mutations in the p53 tumour-suppressor gene. Previous studies using established cell lines to investigate the relation between mutated p53 genes and altered resistance to anti-cancer agents brought inconsistent results. In this report, we examined the status of the p53 gene in 56 astrocytic tumour specimens by single-strand conformation polymorphism and their in vitro chemosensitivity to 30 different kinds of anticancer agents. The chemosensitivity was determined by drug-induced cell death using flow cytometry. We found that the mutated p53 gene correlated with increased resistance to DNA-damaging agents but the sensitivity to anti-microtubule agents was independent of the mutation, suggesting a clinical significance of the status of p53 gene in astrocytic tumours and a rational application of anti-microtubule agents to the patients with p53-mutated astrocytic tumours.

We examined a possible antitumor response against 9L rat gliosarcoma cells induced by the expression of the herpes simplex virus-thymidine kinase (HSV-TK) gene and the ganciclovir (GCV) system. Based on the amount of the major histocompatibility complex (MHC) class I antigens expressed on 9L cells transduced by the HSV-TK gene (9L/HSV-TK) we selected two clones (clones H and L), which represent high and low expressors of class I antigens, respectively. By means of serial magnetic resonance imaging we followed the change of tumor volumes of each clone in syngeneic rats, and found that the intracranial tumor growth was inversely correlated with the expression of MHC class I antigens, although in vitro growths of the clones remained unchanged. Moreover, histological examination revealed significant lymphocyte infiltration in the 9L/HSV-TK tumor of high MHC expression but not in the wild-type tumor. The therapeutic effect of GCV on them was not different, but we observed a prolonged survival of the rats which had eliminated 9L/HSV-TK clone L tumors by the treatment of GCV and were rechallenged with the same cells compared with the survival of naive rats inoculated with clone L cells. These data collectively suggest that the immune response operates even in the brain previously described as an immunologically privileged site.

We have investigated clonal variations on the in vivo bystander effect of herpes simplex virus-thymidine kinase (HSV-TK)/ganciclovir (GCV) system, using an intracranial tumor model with 9L rat gliosarcoma cells. For this purpose, we established three 9L clones transduced with HSV-TK gene (9L-TK), each of which had similar TK activity but showed different tumor growth rate in vivo. The use of GCV in vitro confirmed previous reports that the bystander effect was manifest in a confluent culture condition, but not in a low cell density condition. In contrast, the therapeutic benefit of the in vivo bystander effect varied among each 9L-TK clone, and had a positive correlation with the in vivo growth rate of the clone. Thus, the bystander effect seems to reflect the growth rate of TK-positive cells, and these data raise a crucial point for applying the bystander effect to clinical trials.

Loss of p53 function is involved in tumorigenesis of various human cancers, but the relation between mutation of the p53 tumor-suppressor gene and the chemo- and radiosensitivity of tumors remains unclear. Mutated p53 gene in malignant glioma is often associated with progression and recurrence of malignancy, and these events are closely linked with increased resistance to both chemotherapy and radiation. We have examined the status of the p53 gene in malignant gliomas obtained from 34 patients (glioblastoma: 29 cases, anaplastic astrocytomas: 5 cases). The chemosensitivities of these specimens using 28 kinds of anti-cancer agents were determined using an in vitro assay system. Overall, 12 mutated cases of p53 gene were found in malignant glioma samples. The mean numbers of effective agents were 0.58 for the tumor samples with p53 mutations and 5.00 for tumors without mutations. Our data indicate that p53 gene mutation predisposes to decreased cell killing via chemotherapy in malignant gliomas. (C) 1996 Wiley-Liss, Inc.

Intra-arterial (IA) chemotherapy has achieved no obvious clinical superiority as a treatment for glioblastoma multiforme despite the many theoretical advantages. The clinical courses of 38 patients who underwent surgery and radiotherapy with IA 1-(4-amino-2-methyl-5-pyrimidinyl)-methyl-3-(2-chloroethyl)-3-nitrosourea hydrochloride (ACNU) and cisplatin were reviewed. Tumor regrowth was evaluated by comparison of contrast-enhanced areas on computed tomographic scans. The initial response rate was 19 of 32 patients evaluated, and the median survival time (MST) for all 38 patients was 53 weeks. Local recurrence was observed in 20 patients, and distant recurrence (areas more than 3 cm from the original tumor margin) was observed in 15 patients. The MST was 59 weeks for patients without distant recurrence, and 42 weeks for patients with distant recurrence (statistically not significant). Adjuvant IA ACNU and cisplatin chemotherapy did not improve the survival time. An important clinical feature was the high incidence of distant recurrence, in contrast to experience with other conventional therapy regimens. Distant recurrence, without extended survival, may suggest insufficient control of tumor regrowth. © 1995, The Japan Neurosurgical Society. All rights reserved.

To assess whether therapeutic efficacy is related to the intra-arterial (IA) mannitol infusion prior to ACNU and cisplatin (CDDP) for malignant brain tumors, the survival time of patients with and without mannitol infusion was compared. Ninety-eight patients were randomly assigned to either a mannitol infusion group (group A) or a non-mannitol infusion group (group B); 34 with malignant gliomas (18 in group A and 16 in group B) and 64 with brain metastases (36 in group A and 28 in group B). During radiotherapy, ACNU and CDDP at a dose of 100 mg/body were given through the common carotid artery at a rate of 20 mg/min. In group A, 50 ml of 20% mannitol was injected intra-arterially at a rate of 50 ml/min immediately prior to the injection of chemotherapeutic agents. Of the patients with malignant gliomas, the median survival time (MST) was 52 weeks for all 34 cases, 68 weeks for group A, and 47 weeks for group B. Survival analysis showed no significant differences between the two treatment groups. Of the patients with brain metastases, the MST was 40 weeks for all 64 cases, 47 weeks for group A, and 24 weeks for group B; the survival time was significantly longer in group A as compared to group B (p< 0.05). This study has demonstrated that, for the patients with brain metastases, IA mannitol infusion provided a survival benefit in the IA chemotherapy employing ACNU and CDDR In contrast, IA mannitol infusion offered no survival benefit to the patients with malignant gliomas.