Clinical Assessment of 177Lu-DOTATATE Quantification by Comparison of SUV-Based Parameters Measured on Both Post-PRRT SPECT/CT and 68Ga-DOTATOC PET/CT in Patients With Neuroendocrine Tumors

Abstract

Introduction The aim of this study was to study the feasibility of an SUV-based 177Lu-DOTATATE SPECT/CT quantification and to assess the correlation of different parameters with the corresponding measured on 68Ga-DOTATOC PET/CT in patients with well-differentiated gastroenteropancreatic neuroendocrine tumors (WD-GEP-NET) treated with peptide receptor radionuclide therapy. Patients and Methods Patients with WD-GEP-NET who benefited from a pretherapeutic 68Ga-DOTATOC PET/CT and a 177Lu-DOTATATE SPECT/CT after the cycle 1 of peptide receptor radionuclide therapy were prospectively included. SPECT/CT acquisitions were performed on a system calibrated with a conversion factor of 9.48 counts/MBq per second and were reconstructed with an iterative algorithm allowing quantification using the SPECTRA Quant software (MIM Software, Cleveland, OH). For each patient, different SUV parameters were recorded on both PET/CT (Ga parameters) and SPECT/CT (Lu parameters) for comparison: physiological uptakes (liver/spleen), tumor uptake (1–10/patient; SUVmax, SUVmean, SUVpeak, MTV), tumor-to-liver and tumor-to-spleen ratios according to liver/spleen SUVmax and SUVmean (TLRmax, TLRmean, TSRmax, and TSRmean, respectively). Results Ten patients (8 female; 2 male) aged from 50 to 83 years presenting with a metastatic progressive WD-GEP-NET (7 small intestine, 2 pancreas, 1 rectum) were included. Median values of lesional Lu-SUV were significantly lower than the corresponding Ga-SUV (P < 0.001), whereas median values of lesional Lu-MTV, Lu-TLR, and Lu-TSR were significantly higher than the corresponding Ga-MTV, Ga-TLR, and Ga-TSR (P < 0.02). Pearson correlation coefficients were strong for both SUV and MTV parameters (0.779–0.845), weak for TLR parameters (0.365–0.394), and moderate-to-strong for TSR parameters (0.676–0.750). Conclusions Our results suggest the feasibility of 177Lu-DOTATATE SPECT/CT quantification in clinical practice and show a strong correlation of several SUV-based parameters with the corresponding in 68Ga-DOTATOC PET/CT.

Differences in tumor-to-normal organ SUV ratios measured with 68Ga-DOTATATE PET compared with 177Lu-DOTATATE SPECT in patients with neuroendocrine tumors

“…These observations are similar to those of Thuillier et al [14] who performed an SUV-based comparison of 68 Ga-DOTATOC PET with 177 Lu-DOTATATE SPECT in 10 patients with GEP-NET. Although they found a strong correlation of PET and SPECT SUV max , SUV mean and SUV peak measurements in lesions and normal organs, correlations were notably weaker for tumor-to-liver and tumor-to-spleen ratios, which were higher on 177 Lu-SPECT, as in our study.…”

Prognostic Value of Whole-Body PET Volumetric Parameters Extracted from 68Ga-DOTATOC PET/CT in Well-Differentiated Neuroendocrine Tumors

“…Higher SRETVwb (≥39.1ml) and TLSREwb (>306.8g) were correlated with significantly shorter median TTP (TTP=12months; CI95% [10][11][12][13][14][15][16][17][18][19][20][21][22][23] vs not reached for both; p<0.001) and shorter median OS (OS not reached for both; p<0.001). SUVmax was not associated with TTP and OS (p=0.08 and p=0.09, respectively; fig.…”

“…For this study, SUVmax threshold based on the healthy liver uptake was chosen. SUVmax value was assessed by placing a spherical VOI of diameter 3 cm in the right upper lobe of the liver, as previously reported (16). For each lesion, the SRETV and TLSRE were semiautomatically extracted.…”

Population-based input function (PBIF) applied to dynamic whole-body 68Ga-DOTATOC-PET/CT acquisition

“…Patlak reconstructions were performed using 4D nested generalized Patlak expectation-maximization reconstruction in the validation cohort, using both IDIF and the scaled PBIFs to obtain parametric images (5,(29)(30)(31). Mean Ki-Liver, Kispleen, and Ki-tumor values (in ml/min/100 ml) were generated from the different reconstructed parametric images by applying circular 3-and 1-cm diameter region of interests (ROI), respectively, drawn over an uninvaded part of the liver (in the right hepatic lobe) and spleen organs, as previously recommended (32); spherical volume of interests (VOI) drawn over the highest tumor uptake using a fixed threshold method delineating a 3D contour around voxels equal to or greater than 40% of the lesion Ki max, by analogy with the SUV approach. Each ROI and VOI were segmented on the last ninth frame and applied to generate time-activity curves.…”