Neurocognitive and radiological changes after cranial radiation therapy in humans and rodents: a systematic review.

BACKGROUND: Radiation-induced brain injury is a common long-term side effect for brain cancer survivors, leading to a reduced quality of life. Although there is growing research pertaining to this topic, the relationship between cognitive and radiologically detected lesions of radiation-induced brain injury in humans remains unclear. Furthermore, clinically translatable similarities between rodent models and human findings are also undefined. The objective of this review is to then identify the current evidence of radiation-induced brain injury in humans and to compare these findings to current rodent models of radiation-induced brain injury. METHODS: This review includes an examination of the current literature on cognitive and radiological characteristics of radiation-induced brain injury in humans and rodents. A thorough search was conducted on PubMed, Web of Science, and Scopus to identify studies that performed cognitive assessments and magnetic resonance imaging techniques on either humans or rodents after cranial radiation therapy. A qualitative synthesis of the data is herein reported. RESULTS: A total of 153 studies pertaining to cognitively or radiologically detected radiation injury of the brain are included in this systematic review; 106 studies provided data on humans while 47 studies provided data on rodents. Cognitive deficits in humans manifest across multiple domains after brain irradiation. Radiological evidence in humans highlight various neuroimaging-detectable changes post-irradiation. It is unclear, however, whether these findings reflect ground truth or research interests. Additionally, rodent models do not comprehensively reproduce characteristics of cognitive and radiological injury currently identified in humans. CONCLUSION: This systematic review demonstrates that associations between and within cognitive and radiological radiation-induced brain injuries often rely on the type of assessment. Well-designed studies that evaluate the spectrum of potential injury are required for a precise understanding of not only the clinical significance of radiation-induced brain injury in humans, but also how to replicate injury development in pre-clinical models.

Feasibility of a mini-pig model of radiation-induced brain injury to one cerebral hemisphere.

BACKGROUND: Radiation-induced brain injury is a common concern for survivors of adult and pediatric brain cancer. Pre-clinically, rodent models are the standard approach to evaluate mechanisms of injury and test new therapeutics for this condition. However, these rodent models fail to recapitulate the radiological and histological characteristics of the clinical disease. METHODS: Here we describe a hemispheric mini-pig model of radiation-induced brain injury generated with a clinical 6 MV photon irradiator and evaluated with a clinical 3T MRI. Two pairs of Yucatan mini-pigs each received either 15 Gy or 25 Gy to the left brain hemisphere. Quality of intensity modulated radiation therapy treatment plans was evaluated retrospectively with parameters reported according to ICRU guidelines. The pigs were observed weekly to check for any outright signs of neurological impairment. The pigs underwent anatomical MRI examination before irradiation and up to 6 months post-irradiation. Immediately after the last imaging time point, the pigs were euthanized and their brains were collected for histopathological assessment. RESULTS: Analysis of the dose volume histograms showed that 93% of the prescribed dose was delivered to at least 93% of the target volume in the left hemisphere. Organs at risk excluded from the target volume received doses below clinical safety thresholds. For the pigs that received a 25 Gy dose, progressive neurological impairment was observed starting at 2 months post-irradiation leading to the need for euthanasia by 3-4 months. On MRI, these two animals presented with diffuse white matter pathology consistent with the human disease that progressed to outright radiation necrosis and severe brain swelling. Histology was consistent with the final MRI evaluation. The pigs that received a 15 Gy dose appeared normal all the way to 6 months post-irradiation with no obvious neurological impairment or lesions on MRI or histopathology. CONCLUSION: Based on our results, a mini-pig model of radiation-induced brain injury is feasible though some optimization is still needed. The mini-pig model produced lesions on MRI that are consistent with the human disease and which are not seen in rodent models. Our data shows that the ideal radiation dose for this model likely lies between 15 and 25 Gy.