Click here to


Are you sure ?

Yes, do it No, cancel

Fitting NTCP Models to SBRT Dose and Carotid Blowout Syndrome Data

P Mavroidis1*, J Grimm2 , M Cengiz3 , S Das4 , G Yazici5 , G Ozyigit6 , (1) university North Carolina, Chapel Hill, NC, (2) Johns Hopkins University, Baltimore, MD, (3) Hacettepe University, Faculty of Medicine, Ankara, Ankara, (4) University of North Carolina, Chapel Hill, NC, (5) Hacettepe University, Faculty of Medicine, Ankara, Ankara, (6) Hacettepe University, Faculty of Medicine, Ankara, Ankara


(Tuesday, 7/31/2018) 11:00 AM - 12:15 PM

Room: Karl Dean Ballroom A1

Purpose: To determine the values of the parameters of three popular normal tissue complication probability (NTCP) models and derive the shape of the corresponding dose-response relations of carotid blowout syndrome (CBOS) after stereotactic body radiotherapy (SBRT).

Methods: In this study 61 patients with inoperable locally recurrent head and neck cancer were analysed. These patients underwent SBRT using CyberKnife at the Department of Radiation Oncology, Hacettepe University, Ankara, Turkey between 2007 and 2011. For each patient, the ipsilateral internal carotids were retrospectively delineated and their dose volume histograms were calculated and exported. Follow-up data regarding the endpoint of carotid blowout syndrome were retrospectively collected. The clinical data were fitted by the Lyman-Kutcher-Burman (LKB), Relative Seriality (RS) and Logit NTCP models. The data fitting was assessed through the area under the receiver operating characteristic curve (AUC), Akaike information criterion (AIC) and Odds Ratio methods.

Results: Using the maximum dose (Dmax), physical DVH and equivalent dose to 2Gy per fraction, EQD2Gy-DVH as the dosimetric metrics in the NTCP models, the derived LKB model parameters were: 1) D50=45.8Gy, m=0.24, n=n/a; 2) D50=44.8Gy, m=0.28, n=0.01; and 3) D50=115.8Gy, m=0.45, n=0.01, respectively. Regarding the correlation of the NTCP prediction against the actual follow-up data, the Dmax-based NTCP fitting shows a little higher AUC values (0.70), than the DVH-based (0.62-0.68) and the EQD2Gy-DVH-based (0.60-0.61) models. The variation of the AIC values was small between the different models (58.2-60.8).

Conclusion: A maximum dose to the internal carotid of less than 34 Gy was correlated with a statistically significant reduction of the risk for CBOS. Additionally, the values of the three NTCP model parameters and the respective dose-response relations were determined for the endpoint of CBOS. A threshold of the generalized equivalent uniform dose (gEUD) < 34.5 Gy appears to be significantly associated with lower risks of CBOS.


Stereotactic Radiosurgery, NTCP, Dose Response


TH- response assessment : General (most aspects)

Contact Email