Application of Quantitative SPECT-CT reconstructions with 99mTc-sestamibi renal imaging of patients with cT1 renal masses.

Objectives: CT and MRI are unable to reliably distinguish between benign and malignant renal tumors, with approximately 15-20% of resected lesions resulting in a benign pathology. The most common subtype of benign renal lesions are oncocytomas and related oncocytic/chromophobe tumors (HOCT), which contain cells with high concentrations of mitochondria. Recent studies with ^99mTc-sestamibi have shown the ability to discriminate between oncocytoma (^99mTc-sestamibi-avid) and renal cell carcinoma (RCC, ^99mTc-sestamibi non-avid) when CT and MRI could not. In these studies, semi-quantitative maximum uptake ratios of tumor to renal background were used to increase diagnostic confidence. Quantitative SPECT-CT (QSPECT) is an emerging method that implements corrections for image degrading factors that yield physically repeatable quantitative uptake in target and reference organs. Here we applied QSPECT methodology to ^99mTc-sestamibi renal imaging in order to improve separation of lesions into “hot” and “cold” tumors.

Methods: For the QSPECT reconstruction, images were reconstructed using an iterative OSEM algorithm. Attenuation, resolution, and scatter were compensated for during reconstruction by modeling these effects in both the forward and back projection steps. Resolution was modeled using distance-dependent Gaussian functions, and scatter was modeled using the effective source scatter estimation technique. Attenuation maps were generated from low-dose contemporaneously acquired CT scans. Protocol optimization was performed by reconstructing representative hot and cold tumor types with multiple subsets and iterations using the QSPECT method. Two board certified nuclear medicine physicians then selected the optimal parameters for reconstruction. Previously reported uptake ratios utilizing conventional reconstruction methods in 49 prospectively imaged patients with clinical T1 renal masses were included in the analysis and compared to the new optimized QSPECT images. Uptake ratios on the QSPECT reconstructions were measured for comparison. Additionally, tumors were again classified as “hot” or “cold” based on the optimized QSPECT dataset.

Results: Of the 49 patients imaged, 9 were classified as hot and 40 were classified as cold using both conventional and QSPECT reconstruction methods. However, mean uptake ratios in cold tumors were significantly lower when measured on the QSPECT images (QSPECT mean uptake ratio of 0.209 vs. conventional uptake ratio of 0.276; mean absolute change of - 0.067, SE= 0.013, P value= 0.001). Mean uptake ratios in hot tumors were not significantly different between the two methods (QSPECT mean uptake ratio of 0.922 vs. conventional uptake ratio of 0.935; Mean absolute change of - 0.013, SE= 0.030, P-value= 0.68). The difference between mean hot and cold tumor uptake ratios measured 0.713 with the QSPECT method and 0.659 with conventional imaging. This resulted in an increased separation between hot and cold tumors.

Conclusion: We recently published a prospective clinical trial using ^99mTc-sestamibi SPECT-CT imaging to distinguish between RCC and benign renal oncocytoma/HOCT. Results of this trial demonstrated that a maximum ratio of tumor activity to background renal activity could aid in the differentiation of benign from malignant lesions. There are now proposals to include ^99mTc-sestamibi imaging into the workflow of solid renal mass characterization. By applying QSPECT methods we show increased separation of uptake ratios between hot and cold tumors, possibly increasing the diagnostic confidence when characterizing these lesions. Furthermore, this methodology will serve as a framework on which to build more robust quantitative biomarkers such as SPECT SUV. These biomarkers may provide more diagnostic certainty and improved classification of tumor subtypes to help guide patient management and avoid unnecessary surgeries. Research Support: