Comparative Biodistribution and Radiation Dosimetry of ⁶⁸Ga-DOTATOC and ⁶⁸Ga-DOTATATE in Patients with Neuroendocrine Tumors

Patients

Ten patients (mean age, 62 y; range, 44–75 y; 6 men and 4 women) diagnosed with disseminated neuroendocrine tumors as confirmed by histopathology were included. Three of the patients had pancreatic neuroendocrine tumors, 5 had small-intestinal neuroendocrine tumors, and 2 had lung carcinoids. All patients signed a written informed consent form before inclusion, and the study was approved by the Regional Medical Ethics Board.

Tracer Production and Quality Control

Sodium acetate buffer (pH 4.6), sodium hydroxide (10 M), and doubly distilled hydrochloric acid (Riedel de Haën) were obtained from Sigma-Aldrich Sweden. Trifluoroacetic acid was obtained from Merck. The purchased chemicals were used without further purification. Commercially available DOTATOC and DOTATATE (Eurogentec S.A.) were dissolved in sterile water to give a 1 mM solution. ⁶⁸Ga (half-life, 68 min; β⁺ decay, 89%; electron capture, 11%) was eluted with 0.6 M HCl from a ⁶⁸Ge/⁶⁸Ga-generator (1,850 MBq; IDB Holland BV), where ⁶⁸Ge (half-life, 270.8 d) was attached to a column of an inorganic matrix based on tin dioxide. Production and quality control of the tracers were compliant with good manufacturing practices and were accomplished within 1 h. The fractionation method was used for the labeling (8). The first 1.0-mL fraction was sent to the waste, and the next 1.0 mL, containing over 65% of the total radioactivity, was collected and buffered with 200 μL of acetate buffer and 60 μL of sodium hydroxide to provide a pH of 4.6 ± 0.4. Radioactivity was measured, and 200–300 MBq were taken for the labeling synthesis. Thirty nanomoles of either DOTATOC or DOTATATE were added, and the reaction mixture was heated at 90°C–95°C for 10 min. The crude ⁶⁸Ga-DOTATOC and ⁶⁸Ga-DOTATATE were purified using a C8 (Sep-Pak Light; Waters) reversed solid-phase extraction cartridge. The product was eluted with 0.5 mL of 90% ethanol and formulated with sterile phosphate buffer (pH 7.4), and the resulting solution was passed through a 0.22-μm filter into a sterile injection vial. A sample was taken for determination of identity, radiochemical purity, chemical purity, and pH; for estimation of the peptide content, and for control of sterility and endotoxins.

PET Protocol

PET data were acquired using a Discovery ST (GE Healthcare) PET/CT scanner with an axial field of view of 15.7 cm. After a low-dose CT scan for patient positioning and attenuation correction, a 45-min dynamic PET scan (6 × 10, 4 × 60, 5 × 180, 5 × 300 s) over the abdominal region, including the liver, spleen, adrenals, and kidneys, was started simultaneously with the injection of either 91.4 ± 18.7 MBq (range, 72–120 MBq) of ⁶⁸Ga-DOTATATE or 86.9 ± 16.4 MBq (range, 62–112 MBq) of ⁶⁸Ga-DOTATOC. Three whole-body scans followed (proximal femur to base of skull), starting at 1, 2, and 3 h after injection, with scan durations of 3, 4, and 5 min per bed position, respectively, each preceded by a low-dose CT examination. Images were reconstructed using normalization and attenuation-weighted ordered-subsets expectation maximization (2 iterations, 21 subsets), including corrections for dead time, random coincidences as estimated by singles counting rates, model-based scatter correction (9), attenuation based on a bilinear conversion of Hounsfield units to 511-keV attenuation coefficients, and decay. A 5.4-mm gaussian postprocessing filter was applied. Venous blood samples were drawn after each whole-body scan for measurement of whole-blood radioactivity concentrations. Patients were allowed to urinate between the dynamic scan and each consecutive whole-body examination. Urine was collected and weighed, and radioactivity concentration was measured. Accurate cross-calibration between the PET/CT scanner, the dose calibrator, and the well counters that were used to measure blood and urine activity was verified monthly and was always within 3%. The scan was repeated on the next day with the other tracer. The patients were randomized regarding the order of tracer used on the 2 consecutive examination days.

Volumes of Interest (VOIs)

VOIs were drawn at the same locations on the whole-body images and on the last time frame of the dynamic image series over clearly tumor-free subsets of all clearly identifiable source organs: liver, kidneys, spleen, lungs, small intestine, and adrenal glands. VOIs drawn on the last frame of the dynamic scans were transferred to all earlier time frames. Activity concentration and standardized uptake value (SUV) normalized to body mass and injected amount of radioactivity were determined using the mean activity concentration in VOIs. For adrenal glands, SUV was based on maximum pixel values because of their small volume and corresponding limited recovery.

For measurement of blood radioactivity concentrations during the dynamic scan, 1-cm-diameter circular regions of interest were drawn on 10 consecutive image planes over the descending aorta in the time frame where the first pass of the activity was best visualized and were combined to a single VOI. This VOI was then transferred to all other time frames in the dynamic scan and combined with blood samples collected at later time points to obtain a whole-blood time–activity curve.

Absorbed Dose Calculations

Red marrow radioactivity concentration was assumed to be equal to blood radioactivity concentration (10). Total organ activity was calculated by multiplying the radioactivity concentrations by the organ weights of the adult reference male or female phantom (11). Time-integrated activity in each organ was calculated by trapezoidal integration of the first 45 min of the organ’s time–activity curve, followed by a single-exponential fit to the remaining data points (based on the 3 whole-body images) extrapolated to infinity, all based on non–decay-corrected data. For estimation of urinary bladder contents, all urine measurements were decay-corrected to the time of injection. Then, the bladder content at intermediate time points was estimated assuming linear filling of the bladder until each complete voiding. After the last scan, a 4-h voiding interval with filling rates equivalent to those before the last measured urine sample was assumed. This urinary bladder content time–activity curve was then uncorrected for decay, and time-integrated activity was calculated as the area under this time–activity curve. Residence times were obtained by dividing the time-integrated activity by the injected amount of activity. Small intestine, upper large intestine, and lower large intestine residence times were estimated using the ICRP 30 gastrointestinal tract model (12). Remainder-of-body activity was calculated as the injected activity minus the sum of the activity in all source organs and the activity excreted to urine up to each measurement point. Absorbed doses were calculated using OLINDA/EXM 1.1 (13).

Statistical Analysis

Differences between residence times and absorbed doses for both tracers were assessed using a Wilcoxon matched-pairs signed rank test. Differences between absorbed doses in male and female subjects were assessed using a Mann–Whitney test. A P value below 0.05 was considered significant. No multiple-comparisons correction was applied.