Model-based calculation of thyroid gland normal tissue complication probability in head and neck cancer patients after radiation therapy

Aim Primary hypothyroidism is one of the late complications that can occur after radiation therapy for malignant tumors in the head and neck region. The aim of this retrospective study was to show the validity of the Lyman–Kutcher–Burman (LKB) normal tissue complication model for thyroid gland based on clinical results. Methods Thyroid function was evaluated by measuring thyroid-stimulating hormone and free thyroxine serum levels before radiation therapy, 3 months after the beginning of radiation therapy, and afterwards at each follow-up visit. Cumulative incidence was calculated using the Kaplan–Meier method. Dose–volume histogram, total dose, fractionation schedule, total duration of the treatment, and other parameters were used for normal tissue complication probability calculation based on the LKB model. The model was evaluated after fitting with the three sets of parameters for grade 2 hypothyroidism: 1) “Emami,” where n = 0.22; m = 0.26, and D 50 = 80 Gy; 2) “mean dose,” where n = 1; m = 0.27, and D 50 = 60 Gy; and 3) “Lyman EUD,” where n = 0.49; m = 0.24, and D 50 = 60 Gy. A value 3.0 Gy was used for α/β ratio Results Eighty-three patients treated with volumetric modulated arc therapy for head and neck cancers at the University Hospital Martin, Slovakia, from January 2014 to July 2017, were included in the retrospective study. Median follow-up was 1.2 years. Cumulative incidence of hypothyroidism grade 2 or higher after 12 and 24 months was 9.6 and 22.0%, respectively. Normal tissue complication probability values calculated with mean dose and Lyman EUD parameters showed the best correlation with our clinical findings. Conclusion Empirically based modelling of normal tissue complication probability was valid for our cohort of patients. With carefully chosen parameters, the LKB model can be used for predicting the normal tissue complication probability value.