RADIATION BIOLOGY

Co-chairs: Dennis Shrieve, M.D., Ph.D., and Philip J. Tofilon, Ph.D.

Participants: C. Norman Coleman Brian Fuller Glenn Gobel Daphne A. Haas-Kogen Peter Inskip

Larry Kun Frederick F. Lang, Jr. Bertrand Liang Jay Loeffler Lorraine Marin

Minesh Mehta Raymond Mulhern Judith J. Ochs Ed Oldfield Libby Stevenson

STATEMENT OF THE PROBLEM

Clinical Radioresistance of Primary Glial Tumors

Radiation therapy is a major component of the treatment of many primary and metastatic brain tumors. Standard therapy for glioblastoma multiforme and other primary malignant astrocytomas consists of radiotherapy after the fullest possible surgical resection has been performed. Radiotherapy has long been known to be the single most active treatment for these tumors; doses up to about 60 Gy yield dose-related increases in survival.

Despite this therapeutic benefit, however, in nearly all patients such tumors recur within the volume of tissue receiving high-dose radiation, and eventually these patients succumb to local disease within a median period of about 12 months. Attempts to escalate radiation doses above 60 Gy have not yielded a significant further advantage in survival, probably owing to toxicity related to the volumes of normal brain receiving doses in excess of tolerance. Conformal therapies (brachytherapy or radiosurgery), designed to "boost" the local dose, may allow increased survival. Even after these therapies, however, local failure is the rule and the risk of radionecrosis is extremely high in patients surviving for more than 6 months. Even in cases of "microscopic residual disease," these doses of radiation are inadequate to prevent progression of disease.

In vitro studies of the inherent radiosensitivity of cell lines derived from human glioblastoma multiforme have not demonstrated remarkable radioresistance of these cells. It is highly probable that the "neural environment" and microenvironment of the in situ tumors, and not just the special characteristics of tumor cells in culture, contribute to the remarkable radioinsensitivity of gliomas.

Normal Tissue Toxicity After Radiation

Necrosis and Edema

Radiation doses higher than 60 Gy may produce vasogenic edema and necrosis in some patients with glioblastoma multiforme. Escalation of the radiation dose above this level poses a significant risk of necrosis. These risks are greater than in patients with lower-grade astrocytic tumors or non-glial tumors, which are associated with better prognosis and longer survival.

Functional Deficits

Functional deficits in patients after radiotherapy are probably more common than is currently reported. These deficits include mental retardation in patients irradiated as infants, learning disabilities in older pediatric patients, and memory or cognitive deficits in adults. Whole-brain radiotherapy for metastatic disease can result in a range of neurocognitive outcomes, ranging from little or no deficit to full-blown dementia. The factors contributing to the development of neurocognitive deficits are poorly understood. These deficits have severe effects on quality of life for patients and their families.

CHALLENGES AND QUESTIONS

• Lack of understanding of mechanisms of tumor radioresistance and normal tissue toxicity--Little is known of the basic mechanisms by which radiation kills brain tumor (or other) cells. Two main types of cell death, however, are thought to be important: mitotic, or clonogenic, death (loss of the ability to divide) and apoptosis (programmed cell death). There is evidence that malignant glioma cells do not undergo significant apoptosis after irradiation. Clearly, in patients, many cells escape both modes of cell killing after radiotherapy. The "neural environment" appears to play an important role in this clinical radioresistance, as may microenvironmental factors. In addition, the fundamental processes involved in the development of normal tissue toxicity after radiation are not understood.

• Lack of integration of neurobiology with radiobiology--There would be a great advantage to integration of basic and brain tumor neurobiology and brain tumor-related radiobiology.

• Unavailability of appropriate models--Appropriate animal models are not available, hindering progress. The brain is a unique organ, and its milieu is crucial to the behavior of central nervous system tumors. Cell culture studies are insufficient to allow understanding of the interactions between tumor and normal tissue, which is key to studying the mechanisms of resistance and thereby to improving treatment. It is important to describe in situ the physiology of these tumors, which is probably involved to a great degree in their resistance. Imaging is available, but interactions between normal and tumor cells must be described and modeled. Better models--specifically, orthotopic tumor models rather than subcutaneous models, and practical models for studying late damage to the brain--are necessary to study these interactions.

• Low enthusiasm for development of drugs or modulators of radiosensitivity--There is little enthusiasm within the pharmaceutical industry for the development of drugs or modulators of radiosensitivity for brain tumors. Incentives and encouragement for industry involvement in brain tumor research are needed. The National Cancer Institute (NCI) is planning to establish a screening program to test drugs in the clinic for their potential as radiosensitizers.

• Inability to test new drugs designed to work with radiation--It is difficult to test new drugs designed to work with radiation. Combination development at preclinical stages currently does not occur. The Food and Drug Administration focuses on single agents rather than on combined-modality therapies. Furthermore, many companies are apprehensive about studying their agents in combination with radiation because of concern that toxicity could be, or appear to be, enhanced. This means that, owing to regulatory complications, pharmaceutical companies will not study combined modalities as first-strike therapies.

RESEARCH AND SCIENTIFIC PRIORITIES

Priority 1: Overcome radioresistance of primary brain tumors.

• Delineate the mechanisms of inherent radioresistance.

• Define the influence of the neural environment on radioresistance of brain tumors.

• Identify molecular targets for modulation of brain tumor radiosensitivity.

• Develop hypothesis-driven combinations of radiation therapy and modulators to overcome resistance in clinical practice.

• Develop and validate appropriate models.

Priority 2: Overcome normal tissue toxicity (necrosis/edema versus functional deficits). • Delineate the molecular, cellular, and physiological processes leading to radiation-induced toxicity.

• Define the influence of neurodevelopmental stage on these processes.

• Delineate the interactions between tumor and normal cells in the development of radionecrosis.

• Develop hypothesis-driven interventional strategies (radioprotection).

• Develop and validate appropriate models.

Priority 3: Establish clinical indicators of radiation response of tumor and normal tissue. • Develop imaging modalities to assess tumor response and early changes that are predictive of late sequelae.

• Identify serum, cerebrospinal fluid, or tissue markers of tumor or normal tissue response.

• Develop methods of target "credentialing"--identification of target molecules, evidence of modification of the target molecule, and measurement of desired effect.

• Develop high-throughput techniques to assess the efficacy of modulators of radiosensitivity (e.g., microarray technologies).

• Establish the use of clinical correlates to validate preclinical studies.

• Develop predictors of sensitivity (tumor versus normal tissue).

RESOURCES NEEDED • NCI program for the development of drugs, sensitizers, and/or modulators of radiosensitivity

• Ability to test drugs in combination with radiation

• NCI-sponsored workshop bringing together neurobiologists and radiobiologists to discuss strategies for investigating brain tumor radioresistance and radiation-related toxicity