Quantitative and qualitative investigations of salivary gland function in dependence on irradiation dose and volume for reduction of xerostomia in patients with head-and-neck cancer

Background and Purpose: Radiation treatment of head-and-neck tumors mostly leads to a damage to the salivary glands and a consequential permanent loss of saliva. The aim of this investigation was to establish a modern three-dimensional conformal radiotherapy (3D-CRT) to show a decrease in severe xerostomia in contrast to the proven conventional technique (K-RT) with photons and electrons. Patients and Methods: Between April 2002 and September 2003, 32 patients (25 male, seven female, mean age: 58 years) with malignant tumors of the head and neck were included - after surgery or in case of inoperability with curative intent - in a prospective, nonrandomized study. 10/32 patients (31%) received K-RT with photons and electrons, and 22/32 patients (69%) 3D-CRT (six to eight photon portals). The quantity of saliva was measured as stimulated saliva flow rate (ml/5 min) prior to treatment, at the end, and 1, 6, and 12 months after termination of treatment. To find out the resulting mean dose of both parotid glands for every patient in Gray (D-mean doses), the D-mean doses of the ipsilateral and the contralateral. parotid gland, determined by dose-volume histograms (DVHs), were averaged over. For calculation of the NTCP (normat tissue complication probability), the logistic model was used. Results: In the trend the stimulated salivary flow rates were higher in the group with 3D-CRT than in the group with K-RT during the whole observation period (at 10 weeks after the start of radiotherapy 3D-CRT vs. K-RT with 1.56 +/- 1.6 vs. 0.82 +/- 1.2 ml/5 min; p < 0.1). The patients treated with the K-RT had, on average, significantly higher averaged D-mean values than those irradiated with 3D-CRT (p < 0.012). Patients, who were irradiated with 3D-DRT for tumors of the larynx or hypopharynx, showed, on average, significantly Lower D-mean values than patients, who were treated with 3D-DRT because of oral cavity or oropharynx carcinomas or with K-RT irrespective of the primary tumor site (p < 0,003). The resulting dose for 50% comptication probabitity (TD50) of the salivary glands was 36.9 Gy (30.9-43.5 Gy; 95% confidence interval.). The gradient kappa of the curve Located in point TD50 was 7.7 (4.8-15.8; 95% confidence intervaL). Conclusion: Basically, 3D-CRT seems to be suitable as a standard for all patients with carcinomas of the oral cavity, oro- and hypopharynx. Especially in patients with tumors Located in the Larynx and hypopharynx, averaged D-mean doses of both parotids during irradiation can be reached, to conserve salivary flow rates, which are simiLar to baseline flow rates.