The role of three dimensional functional lung imaging in radiation treatment planning: the functional dose-volume histogram.

PURPOSE: During thoracic irradiation (XRT), treatment fields are usually designed to minimize the volume of nontumor-containing lung included. Generally, functional heterogeneities within the lung are not considered. The three dimensional (3D) functional information provided by single photon emission computed tomography (SPECT) lung perfusion scans might be useful in designing beams that minimize incidental irradiation of functioning lung tissue. We herein review the pretreatment SPECT scans in 86 patients (56 with lung cancer) to determine which are likely to benefit from this technology. METHODS AND MATERIALS: Prior to thoracic XRT, SPECT lung perfusion scans were obtained following the intravenous injection of approximately 4 mCi of 99mcTc-labeled macro-aggregated albumin. The presence of areas of decreased perfusion, their location relative to the tumor, and the potential clinical usefulness of their recognition, were scored. Patients were grouped and compared (two-tailed chi-square) based on clinical factors. Conventional dose-volume histograms (DVHs) (DVFHs) are calculated based on the dose distribution throughout the computed tomography (CT)-defined lung and SPECT-defined perfused lung, respectively. RESULTS: Among 56 lung cancer patients, decreases in perfusion were observed at the tumor, adjacent to the tumor, and separate from the tumor in 94%, 74%, and 42% of patients, respectively. Perfusion defects adjacent to the tumor were often large with centrally placed tumors. Hypoperfusion in regions separate from the tumor were statistically most common in patients with relatively poor pulmonary function and chronic obstructive pulmonary disease (COPD). Considering all SPECT defects adjacent to and separate from the tumor, corresponding CT abnormalities were seen in only approximately 50% and 20% of patients, respectively, and were generally not as impressive. Following XRT, hypoperfusion at and separate from the tumor persisted, while defects adjacent to the tumor improved in several patients. In four patients who achieved a complete response scored by CT with chemotherapy prior to XRT, persistent hypoperfusion was present at and adjacent to the tumor site in three. Among 30 patients with cancers not arising in the lung (14 breast, 12 lymphoma, 4 others), perfusion defects were seen in only 4 (2 adjacent and 2 apart). Recognition of decreases in perfusion mainly impacted on treatment planning for a few patients with poor pulmonary function and limited target volumes. DVFHs have been useful in beam selection for patients with marked perfusion heterogeneities. CONCLUSIONS: Lung perfusion scans provide functional information not provided by CT scans that can be useful in designing radiation treatment beams that minimize incidental irradiation of the function regions of the lung. This approach appears to be most helpful in patients with gross intrathoracic lung cancer, especially those with small targets and relatively poor pulmonary function. One limitation of this approach is that some of the defects adjacent to the tumor site reperfuse following treatment, indicating that these scans identify perfusion rather than potential perfusion. Three dimensional functional data can be used to generate DVFHs that may be more predictive of the physiological consequences of the radiation than conventional DVHs. Additional work is currently underway to test this hypothesis.

Duke Scholars

Author Ronald J. Jaszczak Radiology