Myocardial perfusion changes in patients treated for left-sided breast cancer and correlation with coronary artery distribution

Abstract

Purpose: To correlate changes in single photon emission computed tomography myocardial perfusion imaging (SPECT MPI) with coronary artery distributions and percent incidentally irradiated left ventricle (LV). To compare two techniques for evaluating post-irradiatory SPECT MPI changes. Materials and Methods: Rest SPECT MPI of the LV was acquired prior to radiotherapy (RT) and at 6 mos. intervals post-RT in 25 patients treated with tangential photon beam irradiation for left-sided breast cancer (BC). The extent (%) of defects with a severity > 1.5 SD below the mean was quantitatively analyzed for the territories of the left anterior descending (LAD), left circumflex (LCX) and right coronary (RCA) arteries based on computer assisted polar map reconstruction (PMR), i.e. bull’s-eye-view. Changes in perfusion were correlated with percent of the SPECT-defined LV incidentally irradiated to > 25 Gy (range: 0-32%). The results from the quantitative PMR-analyses were compared with the standard semi-quantitative visual SPECT MPI analyses by nuclear medicine radiologist (12-segment model; severity score 0-3) (summed rest score-SRS). Results: In the LAD-territory, there was a mean increase in PMR perfusion defects of 13% at 6 mos. and 7 % at 12 mos. compared to baseline (Friedman ANOVA, P=0.006). No differences in perfusion were found in the LCX- or the RCA-territories at 6 or 12 mos. compared to base line. A statistically significant association between perfusion defects in the LAD-territory at 6 mos. and percent irradiated LV was present for the SRS-method (Spearman Correlation, P<0.001), and a trend for association was found in the PMR-analysis (P=0.10). When the results from the PMR- and SRS-analyses of the LAD-territory at 6 mos. were compared, an association in the directional trend of the perfusion changes was found (Gamma Statistics, P=0.001). The results were unchanged when controlled for the potential effects of co-variates. Conclusions: Tangential photon beam RT in patients with left-sided BC was associated with perfusion defects in the vascular distribution corresponding to the LAD. The percent of LV demonstrating perfusion changes appeared to be of a similar magnitude as the percent incidentally irradiated LV volume among the majority of patients, suggesting that the defects at 6-12 mos. could be manifested via small vessel- rather than coronary artery dysfunction. The PMR- and SRS-methods provided similar results in the LAD territory, implying that the PMR-technique could be a useful compliment to the standard SRS-method for objective quantification of intra patient changes over time following BC irradiation. Up-dated results including data from 18 mos. PMR-analyses will be presented at the meeting. Supported by the Department of Defense Breast Cancer Research Grant #BC972695 and the Raychem/Duane Rodgers/John Morrish Fellowship. Thanks to University of North Carolina for the use of PLUNC treatment planning system.