Application of analytic methodologies for image quantification in neuroendocrine tumor therapy with 177Lu-DOTA

Neuroendocrine tumors have annual incidence of 1 to 2 cases per one hundred thousand inhabitants. The 177Lu-DOTA-octreotate treatments in 3 or 4 cycles has been effective in controlling disease progression and, in some cases, promote tumor remission. To estimate radiation side effects in healthy organs, image quantification techniques have been broadcast for individualized patient dosimetry. In this paper, image data processing methods are presented to allowing comparisons between different image conjugate views, combined with attenuation correction and system sensitivity. Images were acquired 24, 72 and 192 h after administration of 74 GBq of 177Lu-DOTA using a dual-head gamma camera detection system and they were evaluated with ImageJ software. 4 female patients underwent to two cycles of treatment. The kidneys, liver and whole-body regions of interest were separately assessed by 4 techniques for counts method and 12 techniques for pixel intensity method, considering the main photopeak separately and aided by the attenuation correction map and adjacent windows to photopeak energy. The pixel intensity method was combined with mathematical correction for pixels with null value. The results obtained by the two methods were strongly correlated (r>0.9) (p<0.001). The paired t-test accepted the null hypothesis of compatibility between the two methods (with and without attenuation correction map) (p<0.05), but rejected it when the adjacent windows were combined. No significant tumor reduction (p>0.05) was found between the treatment cycles. In conclusion, the pixel intensity method is faster and allows macros, minimizing operator error, and may optimize dosimetry in tumor therapies with 177Lu-DOTA-octreotate. (author)