Lack of evidence for increased tolerance of rat spinal cord with decreasing fraction doses below 2 Gy.

The radiation tolerance of the spinal cord, both in man and in rats, has been shown to depend strongly on the size of the dose per fraction. With fraction doses down to about 2 Gy, the spinal cord tolerance can be predicted by a modified Ellis formula: D approximately N0.43. More recently alternative isoeffect formulas were based on the linear-quadratic (LQ) model of cell survival where the effect of dose fractionation is characterized by the ratio alpha/beta which varies from tissue to tissue. For the spinal cord, as well as for other late responding tissues, the ratio alpha/beta is small, in contrast to most acutely responding tissues. Both the Ellis-type formula, and to a lesser extent the LQ-model, predict a continuously increasing tolerance dose with decreasing fraction size. From the LQ model, the concept of "flexure dose" has been derived, which proposes the limit of effective fractionation to be about 0.1 alpha/beta. At this dose per fraction no significant further gain in tolerance would be detected. From previous experiments on the rat cervical spinal cord with doses per fraction down to about 2 Gy, the ratio alpha/beta was determined to be 1.7 Gy, and the LQ-model would predict a rise in tolerance with a reduction in fraction size to far below 2 Gy. Based on these predictions clinical studies have been initiated assuming a significantly increased tolerance by reduction of fraction size to about 1 Gy. However, in the present experiments no evidence was found for such an increase in tolerance with fraction sizes below 2 Gy.