Encrypted login | home

Program Information

Correlating Toxicity with Voxelwise Quantification of Delivered Dose to SPECT Perfusion and Ventilation in Lung Cancer Patients Receiving Radiotherapy

no image available
D Owen

D Owen1*, P Boonstra2 , C Anderson3 , A Racette4 , M Skipper5 , I El Naqa6 , J Balter7 , S Jolly8 , R Ten Haken9 , F Kong10 , M Matuszak11 , (1) ,,,(2) University of Michigan, Ann Arbor, MI, (3) University of Michigan, Ann Arbor, MI, (4) University of Michigan, Ann Arbor, MI, (5) University of Michigan, Ann Arbor, MI, (6) University of Michigan, Ann Arbor, MI, (7) Univ Michigan, Ann Arbor, MI, (8) University of Michigan, Ann Arbor, MI, (9) University of Michigan, Ann Arbor, MI, (10) Georgia Regents University, Augusta, GA, (11) University of Michigan, Ann Arbor, MI

Presentations

TH-AB-FS4-9 (Thursday, August 3, 2017) 7:30 AM - 9:30 AM Room: Four Seasons 4


Purpose: It has been hypothesized that limiting radiation to functional lung can reduce radiation-induced lung toxicity (RILT), but studies correlating functional metrics and RILT are limited. Here, voxelwise SPECT-based metrics quantifying the dose delivered to relative perfusion and ventilation function in the lung were correlated to RILT rates at various time-points in advanced stage non-small-cell lung cancer (NSCLC) patients.

Methods: Perfusion and ventilation SPECT scans were obtained from a cohort of 60 NSCLC radiotherapy patients, with 60, 56, 39 and 21 scans at pre-, mid-, 3-month post-, and 1-year post-treatment, respectively. A plug-in was developed to correlate SPECT intensities with registered, bio-corrected (EQD2) dose distributions on a voxel-by-voxel basis. Planned dose (D) values, weighted by fractional perfusion or ventilation signal intensity (fVD) were computed for Lungs-GTV, ipsilateral, and contralateral lung voxels, as were generalized equivalent uniform functional doses (gEUfD). RILT was defined as grade 2 radiation injury. Cox proportional hazard model was used to calculate standardized hazard ratios (HR) and significance of relating functional dosimetrics to RILT.

Results: Patients that experienced lung toxicity had higher average fV20 and gEUfD at each time-point. Increasing pre- and mid-treatment perfusion (fV20, HR=2.62, 3.56 and p=0.01, 0.002) and ventilation (fV20, HR=2.37, 3.50 and p=0.008, 0.002) and change from pre- to mid-treatment ventilation (ΔfV20, HR=3.68 and p=0.002) metrics of the ipsilateral lung were correlated with increased risk of RILT. No significant correlations with RILT were observed for functional dosimetrics of the contralateral lung or Lungs-GTV, although strong associations remained between physical Lungs-GTV dose and RILT (V20, HR=3.84 and p=0.008).

Conclusion: Significant correlations between pre- and mid-treatment SPECT-based functional dosimetrics of ipsilateral lung and RILT are observed. These results suggest functional damage to ipsilateral lung is a significant factor in RILT incidence. Future studies aim to investigate incorporation of functional dosimetrics in treatment planning and adaptation.

Funding Support, Disclosures, and Conflict of Interest: This research is partially funded by NIH grants P01-CA059872 and R01-14284. Dr. Martha Matuszak has research funding from Varian Medical Systems that is unrelated to the current work.


Contact Email: