SU‐GG‐J‐139: Quantitative PET Imaging of Heterogeneous Tumors: The Dosimetric Effect of Patient Motion on Image‐Based Dose Painting Plans

• Published in: Medical Physics Vol. 37; no. 6; pp. 3177 - 3178
• Main Authors: McCall, K, Bowen, S, Jaskowiak, C, McNall, M, Rice, S, Jeraj, R
• Format: Conference Proceeding Journal Article
• Language: English
• Published: American Association of Physicists in Medicine 01-06-2010
• Subjects: Cancer; Computed tomography; Localization effects; Lungs; Medical imaging
• ISSN: 0094-2405; 2473-4209
• DOI: 10.1118/1.3468363

Purpose: Dose‐painting requires accurate quantification of tumor characteristics and precise localization of molecular targets within tumors. Imaging with PET provides surrogate markers of molecular activity within sub‐tumor volumes; however, quantification is limited by many physical factors, including patient motion during image‐acquisition. We have evaluated the dosimetric impact of neglecting respiration‐motion management during PET image‐acquisition. Methods and Materials: PET/CT images were used to generate dose‐painting plans of 60Gy base‐dose to the entire tumor volume (PTV) and 10Gy boost‐dose redistributed within sub‐regions of elevated metabolic activity. Spatial distributions of metabolic activity were derived from: (1) heterogeneous phantoms with variable amplitudes of respiration‐motion, and (2) respiration‐gated and non‐gated [18F]FDG PET images of lung‐cancer patients. For each phantom and patient, a reference plan was generated from PET image where minimal motion occurred. For lung patients, these reference plans corresponded to respiration‐gated PET of the end‐exhalation phase. “Motion‐blurred” dose‐painting plans were generated from non‐gated PET images; however, the PTV contour, boost volume, and integral dose were kept constant. Results: Phantom studies characterized the relationship between dosimetric errors and amplitude of motion. For typical respiration‐motion (2cm superior‐inferior, 1cm anterior‐posterior) 10% of PTV voxels showed dose errors exceeding 3Gy (30% of boost‐dose). Errors were reduced below 3Gy, when motion amplitude was lcm or less. For lung‐cancer patients, the use of motion‐blurred images resulted in dosimetric errors (exceeding 3Gy) within 5% of PTV voxels. Additionally, dose differences up to ±10Gy were observed. Conclusion: Respiration‐motion during PET acquisition resulted in dosimetric errors throughout PTV for dose‐painting plans in phantoms and lung‐cancer patients. Magnitudes of these errors were correlated to the amplitude of residual motion within PET images. Motion‐management techniques (e.g. gated acquisition, 4D‐PET) are necessary to reduce uncertainty in defining heterogeneous dose‐distributions and improve efficacy of image‐based dose‐painting.