Large cohort dose-volume response analysis of parotid gland function after radiotherapy: intensity-modulated versus conventional radiotherapy.

PURPOSE: To compare parotid gland dose-volume response relationships in a large cohort of patients treated with intensity-modulated (IMRT) and conventional radiotherapy (CRT).

METHODS AND MATERIALS: A total of 221 patients (64 treated with IMRT, 157 with CRT) with various head-and-neck malignancies were prospectively evaluated. The distribution of tumor subsites in both groups was unbalanced. Stimulated parotid flow rates were measured before and 6 weeks, 6 months, and 1 year after radiotherapy. Parotid gland dose-volume histograms were derived from computed tomography-based treatment planning. The normal tissue complication probability (NTCP) model proposed by Lyman was fit to the data. A complication was defined as stimulated parotid flow ratio <25% of the pretreatment flow rate. The relative risk of complications was determined for IMRT vs. CRT and adjusted for the mean parotid gland dose using Poisson regression modeling.

RESULTS

One year after radiotherapy, NTCP curves for IMRT and CRT were comparable with a TD(50) (uniform dose leading to a 50% complication probability) of 38 and 40 Gy, respectively. Until 6 months after RT, corrected for mean dose, different complication probabilities existed for IMRT vs. CRT. The relative risk of a complication for IMRT vs. CRT after 6 weeks was 1.42 (95% CI 1.21-1.67), after 6 months 1.41 (95% CI; 1.12-1.77), and at 1 year 1.21 (95% CI 0.87-1.68), after correcting for mean dose.

CONCLUSIONS

One year after radiotherapy, no difference existed in the mean dose-based NTCP curves for IMRT and CRT. Early after radiotherapy (up to 6 months) mean dose based (Lyman) models failed to fully describe the effects of radiotherapy on the parotid glands.