Abstract
Purpose: An exciting form of ventilation imaging is being developed based on 4‐dimensional computed tomography (4DCT) data. Because 4DCTs are routine for thoracic radiation therapy, calculating ventilation maps from 4DCT data does not add extra dosimetric or monetary cost. Studies have discussed designing treatment plans to avoid highly‐ventilated areas of the lung. However, the hypothesis that using ventilation in treatment planning can reduce thoracic toxicity remains to be tested. The purpose of our work was to determine whether incorporating ventilation‐based functional information could improve prediction for clinical thoracic toxicity. Methods: The study used 96 lung cancer patients, 17 of whom developed radiation pneumonitis (CTCAE 3.0 Grade>=3). Pre‐treatment 4DCT data, spatial registration, and a density‐change based model were used to compute pre‐treatment ventilation maps for each patient. Using dose and ventilation, we calculated dose‐volume and dose‐function histograms, mean lung doses (MLD), and ventilation‐weighted MLD (fMLD). The ventilation‐weighted dosimetric values were compared between the pneumonitis group and the non‐pneumonitis group using a two‐sample t‐test. The ventilation‐based functional data were incorporated into a predictive dose‐response model and maximum likelihood was used to determine whether incorporating functional data could significantly improve the model fit to clinical toxicity data. Results: Specific patient examples illustrate that using functional dose metrics can lead to better estimates of toxicity. The fMLD was higher for patients that developed pneumonitis (20.8 Gy) than for those that did not (18.9 Gy), however, the results were not significant (p=0.251). Although fMLD improved model fit over using MLD, the difference in fits was not found to be significant (p=0.312). Conclusions: This is the first study that has attempted to link irradiating ventilated portions of the lung to clinical outcomes. Although promising data were presented, the results were not statistically significant. Future work will include more patients and further functional information with perfusion imaging.
Original language | English (US) |
---|---|
Pages (from-to) | 3948 |
Number of pages | 1 |
Journal | Medical Physics |
Volume | 39 |
Issue number | 6 |
DOIs | |
State | Published - Jun 2012 |
Externally published | Yes |
ASJC Scopus subject areas
- Biophysics
- Radiology Nuclear Medicine and imaging