VOLUME EFFECTS IN RHESUS-MONKEY SPINAL-CORD

Purpose: An experiment was conducted to test for the existence of a volume effect in radiation myelopathy using Rhesus monkeys treated with clinically relevant field sizes and fractionation schedules. Methods and Materials: Five groups of Rhesus monkeys were irradiated using 2.2 Gy per fraction to their spinal cords. Three groups were irradiated with 8 cm fields to total doses of 70.4, 77, and 83.6 Gy. Two additional groups were irradiated to 70.4 Gy using 4 and 16 cm fields. The incidence of paresis expressed within 2 years following the completion of treatment was determined for each group. Maximum likelihood estimation was used to determine parameters of a logistic dose response function. The volume effect was modeled using the probability model in which the probability of producing a lesion in an irradiated volume is governed by the probability of the occurrence of independent events. This is a two parameter model requiring only the estimates of the parameters of the dose-response function for the reference volume, but not needing any additional parameters for describing the volume effect. Results: The probability model using a logistic dose-response function fits the data well with the D-50 = 75.8 Gy for the 8-cm field. No evidence was seen for a difference in sensitivities for different anatomical levels of the spinal cord. Most lesions were type 3, combined white matter parenchymal and vascular lesions. Latent periods did not differ significantly from those of type 3 lesions in humans. Conclusion: The spinal cord exhibits a volume effect that is well described by the probability model. Because the dose response function for radiation myelopathy is steep, the volume effect is modest. The Rhesus monkey remains the animal model most similar to humans in dose response, histopathology, and latency for radiation myelopathy.