Room: ePoster Forums
Purpose: Patients with oropharynx cancer were prospectively enrolled on an IRB-approved clinical trial aiming at determining the correlation of two normal tissue complication probability (NTCP) models and various dosimetric indices of the parotid and submandibular glands with the patient reported severity of xerostomia 12 months after radiotherapy.
Methods: 120 patients with favorable risk, HPV-associated oropharyngeal squamous cell carcinoma were treated with de-intensified chemoradiotherapy. All the patients received 60Gy IMRT with concurrent weekly chemotherapy. Xerostomia was assessed based on the PRO-CTCAE score and specifically it was defined at 12 months post-RT as a ≥ 2 point increase from baseline. Individual patient dosimetric data from the contralateral parotid and contralateral glands (parotid and submandibular glands combined) were correlated with xerostomia. The clinical data was fitted by the Lyman-Kutcher-Burman (LKB) and Relative Seriality (RS) NTCP models.
Results: Xerostomia was observed in 40 patients (33%). For both the contralateral parotid and combined glands the dose-volume index V15Gy was found to correlate best with the follow-up data with an AUC value of 0.71. The corresponding AUC values for the mean dose to those structures was 0.68 and 0.67, respectively. For the contralateral parotid, the values of the D50, m and n parameters of the LKB model were 66.9Gy, 0.58 and 0.01, whereas the values of the D50, γ and s parameters of the RS model were 31.0Gy, 0.46 and 1.0. A statistically significant Odds Ratio of 5.7 and 7.8 was found for the contralateral parotid and combined glands at V15Gy ≤ 53% and ≤ 61%, respectively.
Conclusion: V15Gy ≤ 53% to the contralateral parotid and V15Gy ≤ 61% to the combined parotid and submandibular glands were found to significantly reduce the risk for patient reported xerostomia. The dose-response curve of xerostomia could be determined by the fitted parameters of the LKB and RS NTCP models.
NTCP, Bioeffect Dose, Radiobiology
TH- Radiobiology(RBio)/Biology(Bio): RBio- LQ/TCP/NTCP/outcome modeling