Radiation dose-volume effects on growth hormone secretion.
Mots clés
Abstrait
OBJECTIVE
Growth hormone (GH) deficiency is a known consequence of central nervous system irradiation. The relationship between the dose to the hypothalamus and the time to onset of clinically significant GH deficiency is unknown. Conformal radiotherapy (CRT) techniques allow for a more accurate determination of hypothalamic dosimetry. We correlated the dosimetry of the hypothalamus and the peak GH value after CRT in children with localized primary brain tumors.
METHODS
The arginine tolerance/L-dopa test was performed before (baseline) and repeated 6 and 12 months after CRT in 25 children (median age 4.8 years) with ependymoma (n = 15) or low-grade (n = 8) or high-grade (n = 2) astrocytoma. None had evidence of GH deficiency (arginine tolerance/L-dopa peak GH level >10 ng/mL [10 microg/L]) at baseline. Peak GH levels were modeled as a function of time after CRT and volume of the hypothalamus receiving a dose within the specified intervals of 0-20 Gy, 20-40 Gy, and 40-60 Gy. The model was used to predict the change in the peak GH levels over time (0-12 months) and fit under the assumption that the integral effect of irradiation was a linear sum of the products of the volume receiving a particular dose and the impact of that dose.
RESULTS
The peak GH level declined during the 0-12 months after CRT (p < 0.0001). GH deficiency was observed in 11 children at 6 months and a total of 20 children at 12 months. As expected, the effect of the dose interval 0-20 Gy was smaller than the 20-40-Gy dose interval; the largest effect was noted with the dose interval 40-60 Gy. The peak GH level may be predicted using the following estimating equation within the time limit of 0-12 months: GH(t)=Exp[ln(bGH)-(0.00058V(0-20 Gy)+0.00106V(20-40 Gy)+0.00156V(40-60 Gy))x t], where bGH is the baseline peak GH level, V(0-20 Gy), V(20-40 Gy), and V(40-60 Gy) is the percent-volume of the hypothalamus irradiated from 0 to 20 Gy, 20 to 40 Gy, and 40 to 60 Gy, respectively, and t is time after irradiation. When included in the model, the rate of decline in the peak GH response also was influenced by hydrocephalus and tumor location.
CONCLUSIONS
The peak GH response within 12 months after CRT depends on hypothalamic dose-volume effects and may be predicted on the basis of a linear model that sums the effects of the entire distribution of dose. The modeled effects may be used to optimize radiotherapy and minimize and treat GH deficiency.
