Program Information
Using ROC Analysis to Determine the Optimal Lung Dosimetric Parameter as the Risk Factor of Survival for SBRT in NSCLC Patients
H Yao*, W Wang , F Kong , J Jin , D Vile , Y Liu , K Shiue , A Cerra-Franco , G Bartlett , T Lautenschlaeger , Indiana University- School of Medicine, Indianapolis, IN
Presentations
SU-K-FS2-17 (Sunday, July 30, 2017) 4:00 PM - 6:00 PM Room: Four Seasons 2
Purpose: Stereotactic body radiation therapy (SBRT), the standard of care for inoperative early stage non-small cell lung cancer (NSCLC), provides an outstanding tumor control with minimum toxicity. However, the overall survival (OS) remains suboptimal. We have demonstrated that mean lung dose is a significant risk factor for OS. To understand the underlying mechanism of why lung dose affect OS, this study aimed to determine the optimal lung dosimetric parameter that best correlates with OS.
Methods: A total of 245 early-stage NSCLC patients with SBRT between July 2002 and June 2015 were included in the study. Lung dosimetric parameters were computed as the generalized equivalent uniform dose (gEUD) of various lung definitions including ipsilateral lung (IL), contralateral lung (CL) or total lungs (TL) excluding gross tumor volume (GTV) and planning target volume (PTV). In addition, gEUD with “a” = 0.1 to 6.0 were also computed. Area under curve (AUC) was used to assess model accuracy of dosimetric parameters to select the optimal lung dosimetric parameter.
Results: The gEUDs of variously defined lungs were significant for OS (P<0.05). The AUCs for gEUD of IL-GTV, IL-PTV, CL-GTV, TL-GTV and TL-PTV were 0.64, 0.61, 0.59, 0.61 and 0.61, respectively. The AUCs for gEUD of IL-GTV with “a”= 0.1,0.5,1.0,1.5,2.0,2.5,2.8,3.0,3.5,4.0 and 6.0 were 0.56, 0.64,0.65,0.64,0.63,0.64,0.67,0.66,0.66,0.64 and 0.61 respectively. “a” =2.8-3.5 appeared to have the highest AUC, i.e. the best predictive value. The superiority of IL-GTV over others were significant throughout the course of follow-up.
Conclusion: The gEUD with a =2.8-3.5 for ipsilateral lung excluding GTV appears to be the best lung parameter to predict the survival of patients treated with SBRT. This result suggests that the lung dosimetry around the GTV affects the survival in NSCLC SBRT. Future SBRT plan should use technology to minimize gEUD of IL-GTV to improve survival.
Contact Email: