Rodent Brain Tumor Models for Studies Focusing on Boron Neutron Capture Therapy

Introduction

Rat and mouse brain tumor models have led to the development of a number of therapeutic approaches for the treatment of human glioblastomas (GBMs). Historically, the first GBM models were generated by administering nitrosoureas to rats or mice to induce brain tumors either in the animals themselves or in their progeny. Among these, the most widely used have been the C6, 9L, and F98 rat gliomas, and the GL 261 mouse glioma. All four were derived from glial cells and will be briefly described hereunder. For more detailed information on these tumor models, readers are referred to a recently published review. ¹

Rat C6 Glioma

The C6 rat glioma has been the most widely used of all rodent brain tumor models. It was produced in an outbred Wistar rat ^2,3 and, therefore, is highly immunogenic (Table 1). Although it was originally classified as an astrocytoma, its accession by the American Type Culture Collection (ATCC# CCL-107) led to its reclassification as a glial tumor. The C6 glioma expresses a number of genes also expressed in human GBMs (Table 1). However, despite its induction in outbred Wistar rats, the C6 glioma model continues to be used for a wide variety of studies, but very few related to boron neutron capture therapy (BNCT). ⁴

Rat 9L Gliosarcoma

The 9L gliosarcoma has been the second most widely used experimental rat brain tumor model. It was produced by intravenous (i.v.) injection of methylnitrosourea (MNU) into Fischer rats. ⁵ The 9L tumor cells can be propagated in tissue culture, which has made it a very useful model for both in vitro and in vivo studies. When 9L cells are implanted intracerebrally (i.c.) into syngeneic Fischer rats, they develop into rapidly growing tumors that are composed of spindle-shaped cells with a sarcomatoid appearance, hence the name, gliosarcoma. The tumor margins are sharply delineated with little obvious invasion into the contiguous normal brain. ⁶

The 9L model has been used widely to investigate many therapeutic modalities, including BNCT. The most important of these studies focused on the potential use of boronophenylalanine (BPA) as a boron delivery agent for BNCT of patients with GBMs and were carried out by Coderre et al ^7,8 at the Brookhaven National Laboratory (BNL). BPA first had been used clinically by Mishima et al ⁹ for BNCT of patients with cutaneous melanomas. Coderre et al at BNL first established that BPA could be used as a boron delivery agent for tumors other than melanomas. ^7,10

Subsequently they carried out studies in Fischer rats that had been implanted i.c. with 9L cells and then given BPA followed by BNCT. This established its therapeutic efficacy as a boron delivery agent for brain tumors. This study quickly was followed by a clinical trial at BNL to treat patients with GBMs ¹¹ and then to the widespread clinical use of BPA for BNCT. Although impressive therapeutic results, including apparent cures of tumor-bearing animals, have been obtained, this tumor is highly immunogenic. ¹² This characteristic must be kept in mind when using this model to evaluate the efficacy of novel therapies for brain tumors. The success obtained in experiments using BNCT ^7,8,12 and in gene therapy ^13,14 was in part due to an effective host antitumor immune response.

F98 Glioma

The third most widely used rat brain tumor model, the F98 glioma (ATCC #CRL-2397), was produced by the i.v. administration of N-ethyl-N-nitrosurourea (ENU) to a pregnant Fischer 344 rat, the progeny of which developed brain tumors. ¹⁵ A clone of these tumors was designated F98. From its in vivo histopathology, it was classified as an anaplastic or undifferentiated glioma, ¹⁵ which has been described in detail elsewhere. ⁶ The F98 glioma simulates human GBMs in a number of important ways, including its highly invasive pattern of growth and weak immunogenicity, which have made it particularly useful in evaluating the efficacy of a variety of therapeutic agents (Table 2).

The F98 glioma model is refractory to a number of therapeutic modalities, including photon irradiation. It has been used extensively by Barth and Yang and their coworkers to evaluate the efficacy of BNCT using sodium borocaptate and BPA. ¹⁶ They also have produced transfectants of the F98 glioma expressing human wildtype epidermal growth factor receptor (EGFR) or EGFR_VIII and evaluated molecular targeting of these receptors ¹⁷ using boronated monoclonal antibodies.

In addition, they have evaluated another binary therapeutic modality that combines i.c. carboplatin and photon irradiation. These latter studies have yielded the longest survival times ever obtained with this model. ¹⁸ Tzeng et al carried out extensive studies on its immunogenicity and concluded that it was weakly immunogenic. ^19,20 No other studies were carried out to evaluate its immunogenicity until 10 years later when Volovitz et al ²¹ reported that it was moderately immunogenic. Be that as it may, it is the best rat brain tumor model currently available to assess a variety of therapeutic modalities, including BNCT.

The GL261 Mouse Glioma

The most frequently utilized murine glioma cell line is the GL261, which was produced by serial i.c. injections of methylcholanthrene into C57BL/6 mice. ²² The i.c. inoculation of GL261 cells reliably results in tumors that are invasive but do not metastasize or spontaneously regress after subcutaneous injection into syngeneic mice. Histologically, they resemble ependymoblastomas but display many features of human GBMs. ²³ This model has been used extensively in research to study both glioma stem cells and their response to therapies, including most recently BNCT. ²⁴ However, since the GL261 is highly immunogenic, the enhanced survival times may be due to a tumor-specific immune response, which must be taken into account when using the GL261 in therapy studies.

Conclusions

Rat brain tumor models have been the most widely used to evaluate a number of boron delivery agents for BNCT. ²⁵ However, most frequently in vitro studies are the starting point for evaluating the toxicity and selective uptake of potential boron delivery agents before in vivo animal studies are initiated. If sufficiently promising data were obtained from these in vitro studies, then the next step would be their in vivo evaluation: first, toxicological studies; second, biodistribution studies; and third, therapy studies in tumor-bearing animals.

The rat brain tumor models discussed in this short review range from weakly to strongly immunogenic and the latter may yield deceptively longer survival times that would be expected with a weakly immunogenic tumor. Although the strongly immunogenic murine GL261 glioma occasionally has been used for BNCT studies, ²⁴ recently a number of murine models have been developed that have been genetically engineered ¹ to express a human immune system. Potentially these murine models could be useful in BNCT studies if human brain tumor xenografts were used.

Since one of the great advantages of BNCT is its selectivity for tumor versus normal cells, this could spare immune effector cells. This could result in a concomitant tumor-associated immune response that could increase the survival times of tumor-bearing mice, and this could be extrapolated to humans. However, it must be kept in mind that if it were not therapeutically effective in treating rat or mouse gliomas, it is unlikely to be successful in treating humans with gliomas.