Room: Karl Dean Ballroom C
Purpose: There is limited evidence on regional dose-dependent changes on perfusion (PERF) and FDG uptake in the normal lung during functional lung avoidance radiotherapy. We characterized patient-specific dose-response on longitudinal perfusion SPECT/CT and FDG PET/CT in normal lung for NSCLC patients undergoing functional lung avoidance radiotherapy.
Methods: Eleven stage IIB-IIIB NSCLC patients enrolled on the FLARE-RT phase II interventional trial underwent MAA-SPECT/CT and FDG-PET/CT treatment-position scans at baseline and three months post-treatment. Planning 4DCT scans were rigidly co-registered to the pre/post-treatment SPECT/CT and FDG PET/CT scans. Tumor-subtracted lung dose was converted to EQD2 and discretized into 5 GyEQD2 bins in which mean dose and percent change between pre- and post-RT PERF/FDG uptake were calculated. Perfusion uptake was normalized to total tumor-subtracted lung uptake. Dose-response data were parameterized by sigmoid logistic function to estimate minimum uptake (%Î”Uptakeáµ?â?±â?¿), maximum uptake (%Î”Uptakeáµ?áµƒË£), dose midpoint (Dáµ?â?±áµˆ), and dose-response slope (b). Pairwise differences between PERF and FDG dose-response parameters were evaluated by Wilcoxon signed-rank tests and Spearman rank correlations.
Results: Dose-response goodness-of-fit was similar for PERF and FDG-uptake curves (R2 = 0.93). The median %Î”Uptakeáµ?áµƒË£ was significantly higher for FDG compared to PERF (130% vs. 50%, p=0.01). The median PERF-slope (b = 1.62) was lower compared to the FDG-slope (b = 6.15, p=0.03). The median PERF %Î”Uptakeáµ?â?±â?¿ was -6% compared to 10 % in FDG (p=0.02). The median FDG-dose midpoint Dáµ?â?±áµˆ was similar to PERF Dáµ?â?±áµˆ (31 GyEQD2 vs. 26 GyEQD2, p=0.59). Increased FDG-uptake weakly correlated with decreased PERF-uptake (Spearman Ï? = 0.07-0.25).
Conclusion: Differences in normal lung dose-response parameters on perfusion and FDG imaging, along with weak rank correlation, suggest potentially complementary information. Multiparametric characterization of regional lung changes in response to radiation can support patient risk stratification and enrich functional lung avoidance planning techniques.
Funding Support, Disclosures, and Conflict of Interest: This work was supported by NIH/NCI R01CA204301.
IM/TH- Image Analysis (Single modality or Multi-modality): Computer-aided decision support systems (detection, diagnosis, risk prediction, staging, treatment response assessment/monitoring, prognosis prediction)