⁶⁴Cu-SARTATE PET Imaging of Patients with Neuroendocrine Tumors Demonstrates High Tumor Uptake and Retention, Potentially Allowing Prospective Dosimetry for Peptide Receptor Radionuclide Therapy

Abstract

Imaging of somatostatin receptor expression is an established technique for staging of neuroendocrine neoplasia and determining the suitability of patients for peptide receptor radionuclide therapy. PET/CT using ⁶⁸Ga-labeled somatostatin analogs is superior to earlier agents, but the rapid physical decay of the radionuclide poses logistic and regulatory challenges. ⁶⁴Cu has attractive physical characteristics for imaging and provides a diagnostic partner for the therapeutic radionuclide ⁶⁷Cu. Based on promising preclinical studies, we have performed a first-time-in-humans trial of ⁶⁴Cu-MeCOSar-Tyr³-octreotate (⁶⁴Cu-SARTATE) to assess its safety and ability to localize disease at early and late imaging time-points. Methods: In a prospective trial, 10 patients with known neuroendocrine neoplasia and positive for uptake on ⁶⁸Ga-DOTA-octreotate (⁶⁸Ga-DOTATATE) PET/CT underwent serial PET/CT imaging at 30 min, 1 h, 4 h, and 24 h after injection of ⁶⁴Cu-SARTATE. Adverse reactions were recorded, and laboratory testing was performed during infusion and at 1 and 7 d after imaging. Images were analyzed for lesion and normal-organ uptake and clearance to assess lesion contrast and perform dosimetry estimates. Results: ⁶⁴Cu-SARTATE was well tolerated during infusion and throughout the study, with 3 patients experiencing mild infusion-related events. High lesion uptake and retention were observed at all imaging time-points. There was progressive hepatic clearance over time, providing the highest lesion-to-liver contrast at 24 h. Image quality remained high at this time. Comparison of ⁶⁴Cu-SARTATE PET/CT obtained at 4 h to ⁶⁸Ga-DOTATATE PET/CT obtained at 1 h indicated comparable or superior lesion detection in all patients, especially in the liver. As expected, the highest early physiologic organ uptake was in the kidneys, liver, and spleen. Conclusion: ⁶⁴Cu-SARTATE is safe and has excellent imaging characteristics. High late-retention in tumor and clearance from the liver suggest suitability for diagnostic studies and for prospective dosimetry for ⁶⁷Cu-SARTATE peptide receptor radionuclide therapy, and the half-life of ⁶⁴Cu would also facilitate good-manufacturing-practice production and distribution to sites without access to ⁶⁸Ga.

PET/CT using ⁶⁸Ga-DOTA-octreotate (⁶⁸Ga-DOTATATE), marketed commercially as NETSPOT (Advanced Accelerator Applications) or other ⁶⁸Ga-labeled somatostatin analogs (SSAs), is becoming the gold standard for neuroendocrine neoplasia (NEN) diagnosis and staging (1). ⁶⁸Ga-DOTATATE is particularly important in determining the suitability of patients for peptide receptor radionuclide therapy (PRRT) (2,3). By virtue of its higher sensitivity for NEN than conventional imaging techniques, including ¹¹¹In-pentetreotide (OctreoScan; Mallinckrodt Pharmaceuticals) SPECT/CT, ⁶⁸Ga-DOTATATE PET/CT has been shown to have a significant impact on patient management decisions (4). The efficacy and safety of PRRT depends on delivering the highest possible dose to the tumor deposits while sparing organs, particularly the kidneys, from radiation toxicity (5–7). Although quantitative SPECT/CT techniques can provide verification of the radiation dose delivered from each cycle of PRRT (8), ideally, dose estimation should be performed before therapeutic administration. Accurate prospective dosimetry would allow prescription of an administered activity that maximizes therapeutic efficacy within the tolerance of normal tissues. In the absence of such information, most facilities providing PRRT administer a standardized activity of 7–8 GBq of ¹⁷⁷Lu-labeled SSA or 2–4 GBq of ⁹⁰Y-labeled SSA. This strategy has generally been successful in providing moderate response rates with low toxicity as well as median progression-free and overall survival durations that have been encouraging both in single-center studies (9) and in the only randomized control trial performed to date, the NETTER-1 trial of ¹⁷⁷Lu-DOTATATE versus dose-escalated SSA therapy (10). However, we believe that the therapeutic index may be further improved by an individualized rather than an empiric approach (11). Because of the so-called sink effect, changes in the burden of disease and tumor avidity as treatment progresses can lead to marked variability in radiation delivery to both lesions and normal organs (12). The quantitative capability of PET/CT provides the opportunity to assess uptake and retention of radioactive peptides in tumor and normal tissues and, thereby, to provide truly individualized prospective dosimetry. This capability was first demonstrated using ⁸⁶Y-SSA PET before ⁹⁰Y-SSA PRRT (13). Although ⁶⁸Ga has been widely used as the diagnostic pair of ¹⁷⁷Lu, it has a half-life of only 68 min, which is insufficiently long to model the retention and clearance kinetics required for accurate dosimetry estimation in individual patients. This is particularly the case for organs with significant clearance over time, which include the kidneys and liver. Of longer-lived positron-emitting radioisotopes, the physical half-life of ⁶⁴Cu (12.7 h) provides an opportunity to assess the clearance kinetics out to, and potentially beyond, 24 h after tracer administration. This could also provide a diagnostic pair both to assess suitability of patients for ⁶⁷Cu PRRT and to enable predictive dosimetry. ⁶⁷Cu is a β-emitting radionuclide with favorable physical characteristics for therapeutic application (14).

Our group has recently reported the synthesis and preclinical evaluation of ⁶⁴Cu-MeCOSar-Tyr³-octreotate (⁶⁴Cu-SARTATE) (15) based on a cage amine ligand called sarcophagine (Sar). An advantage of the MeCOSar, when compared with DOTA, is that copper (II) sarcophagine complexes are more stable than copper (II) DOTA complexes. In a murine xenograft model, the uptake of ⁶⁴Cu-SARTATE in somatostatin receptor–expressing tumors after 2 h was high, at 63.0 ± 15.0 percentage injected dose (%ID)/g, and remained high after 24 h, at 105 ± 27.1 %ID/g. This high retention at late time-points suggests superior potential for personalized PRRT dosimetry planning and therapeutic application. Accordingly, we have performed a single-center open-label, phase 0–1 investigation of ⁶⁴Cu-SARTATE in patients with World Health Organization grade 1 or 2 NEN. Our hypotheses were that ⁶⁴Cu-SARTATE can be safely administered to humans and will facilitate identification of somatostatin receptor–expressing tissues using PET/CT scanning, with absorbed radiation doses that are appropriate for clinical use but with sufficient late retention to enable prospective radiation dosimetry.

MATERIALS AND METHODS

The trial protocol was approved by our institutional ethics committee as a first-time-in-humans study and was registered with the Therapeutic Goods Administration as an Australian Clinical Trial Notification (2015/0320). All patients provided written informed consent. The study was pragmatically designed to include up to 10 participants. Interim safety analyses were planned after 1, 2, and 5 study subjects had completed all required investigations. The trial flow-chart is displayed in Figure 1.

FIGURE 1.

Study flow chart.

The primary objectives of this study were to estimate the rate of occurrence of adverse clinical, biochemical, or hematologic events after ⁶⁴Cu-SARTATE administration; the %ID and SUVs of ⁶⁴Cu-SARTATE found in organs of interest at early and late time-points after administration of the investigational product; and absorbed organ doses (expressed as μGy/MBq of administered ⁶⁴Cu-SARTATE) and whole-body dose (expressed as mSv/200 MBq of administered activity).

The secondary objectives of this study were to determine whether ⁶⁴Cu-SARTATE PET/CT scans at 1, 4, and 24 h demonstrate known sites of malignancy with tumor-to-background ratios that are equivalent to, or greater than, those achieved using routine clinical ⁶⁸Ga-DOTATATE acquisitions at 1 h, thereby enabling same-day diagnostic evaluation, and, if tracer is retained in lesions on the day after administration, to support its use for prospective radiation dosimetry calculations as an aid in the development of ⁶⁷Cu-SARTATE as a therapeutic agent. Reporting clinician preference for image quality was also assessed.

Eligibility criteria were as follows: signed informed consent; an age of 18 y or older; life expectancy of at least 8 wk; biopsy-proven World Health Organization G1 or G2 (Ki-67 index, 0%–2% and 3%–20%, respectively) NEN; at least 1 site of active somatostatin receptor–positive malignancy, as demonstrated on ⁶⁸Ga-DOTATATE PET/CT scanning performed as part of routine clinical care; a creatinine clearance of more than 60 mL/min as estimated by the Cockcroft–Gault formula (using actual body weight); and Eastern Cooperative Oncology Group performance of 0–2.

The exclusion criteria were pregnancy or breastfeeding; known sensitivity or allergy to SSAs; interventional treatment received for NEN in the interval between ⁶⁸Ga-DOTATATE PET/CT and ⁶⁴Cu-SARTATE PET/CT scanning; treatment with long-acting SSAs within 28 d before test agent administration; treatment with short-acting SSAs within 24 h before test agent administration; a QTc interval of more than 0.44 s as measured by screening electrocardiography; any serious medical condition that the investigator felt may interfere with the procedures or evaluations of the study; unwillingness or inability to comply with the protocol or a history of noncompliance; and inability to grant informed consent.

Safety Analysis

Safety evaluations were performed 1 d and 1 wk after tracer administration. These included physical examination, full blood examination, urea and electrolytes, liver function tests, and electrocardiography. Patient-reported adverse events were recorded during administration of the radiotracer and after each scan and reported using the Common Terminology Criteria for Adverse Events, version 4.0, as published by the National Cancer Institute.

PET/CT

The PET/CT scans were acquired on a Discovery 690 scanner (GE Healthcare), which has time-of-flight acquisition and a 64-slice CT scanner. CT scans were performed from the vertex to the lower thighs for the purposes of attenuation correction, localization of lesions, and estimation of tumor volumes. These findings were correlated with recent contrast-enhanced CT when available. A low-dose, unenhanced technique was used (140 kVp, 40–200 mAs, GE Healthcare SmartmA), with a final reconstructed slice thickness of 3.27 mm. Our standard ⁶⁸Ga-DOTATATE PET/CT acquisition protocol was used. This involves a weight-based, intravenous administration of 150–300 MBq of radiopeptide with acquisition of 6–8 bed positions (3 min per bed position). PET scans were reconstructed using the ordered-subset estimation method with 2 iterations and 20 subsets with a 3.25-mm filter in a 128 × 128 matrix over a 55-mm reconstructed transaxial field of view.

⁶⁴Cu-SARTATE was prepared using a modified version of a previously described method (15). Briefly, ⁶⁴Cu(II) (500–800 MBq, 0.02 M HCl) was added to SARTATE (20 μg) in a solution of 4% ethanol in 0.1 M ammonium acetate and a 1 mg/mL solution of gentisic acid, sodium salt (5 mL). The reaction mixture was incubated for 30 min at room temperature and then passed through a Strata-X (Phenomenex, Inc.) cartridge (30 mg). Cartridge-retained ⁶⁴Cu-SARTATE was rinsed with saline for injection before elution with ethanol into a vial containing saline for injection. The contents of this vial were filtered through a 0.22-μM filter. ⁶⁴Cu-SARTATE was recovered in 60%–80% radiochemical yield with more than 95% radiochemical purity. The average peptide mass administered was 6.4 ± 4.7 μg (range, 3.6–19.5 μg). After intravenous administration of approximately 200 MBq of ⁶⁴Cu-SARTATE (mean, 192 ± 24 MBq; range, 125–209 MBq), PET scans were acquired over the same axial extent with 3-dimensional acquisition of 8 bed positions (1.5 min per bed position for the 30- and 60-min scans, 2.0 min for the 240-min scans, and 3 min for the 24-h scan). PET scans were reconstructed with 2 iterations, 18 subsets, and a 7-mm gaussian smoothing kernel to mitigate image noise. All scans were performed at a mean of 14 ± 9 d (range, 6–27 d) of the ⁶⁸Ga-DOTATATE PET/CT that determined eligibility for the study.

Analysis of ⁶⁴Cu-SARTATE Scans

Qualitative analysis of each subject’s PET/CT scans was performed independently by 2 qualified nuclear medicine physicians masked to the clinical details of the patient. The ⁶⁸Ga-DOTATATE and ⁶⁴Cu-SARTATE scans obtained at approximately 1 h were presented side by side in a deidentified manner with the physicians indicating the image quality of each scan on a 5-point scale. They then indicated if any suspected lesions were identified on one scan but not seen on the other. The number and location of such discordant lesions were recorded. After this direct comparison of each scan type at the 1-h time point, the series of 4 individual ⁶⁴Cu-SARTATE image sets acquired from 30 min to 24 h after injection were displayed in a random order against each other and the ⁶⁸Ga-DOTATATE images. The physicians were asked to score image quality for each time point and to indicate whether any lesions were apparent on one scan but not the other.

SUV was estimated for up to 5 reference lesions (with no more than 3 per organ) on ⁶⁸Ga-DOTATATE obtained at 1 h after injection using volume-of-interest software (MIM Encore; MIM Software Inc.). If able to be identified, the same lesions were analyzed on the ⁶⁴Cu-SARTATE scans at 1, 4, and 24 h. Further, SUV_mean estimations for normal liver, lungs, spleen, kidneys, bone marrow, thyroid, and pituitary gland were also obtained for ⁶⁴Cu-SARTATE at 1, 4, and 24 h. Lesion-to-liver ratios were calculated for reference deposits at 1, 4, and 24 h to determine the optimal time for detection of hepatic lesions, which are common among patients with metastatic NEN (16).

For radiation cross-dose analysis, the source organs considered were the adrenals, brain, upper and lower large intestine contents, small intestine contents, stomach contents, heart contents, kidneys, liver, lungs, muscle, pancreas, red marrow, spleen, thyroid, urinary bladder contents, and the remainder of the body. Organs were contoured on the 30-min PET/CT series, and each subsequent time point was fused by rigid and deformable image registration (17). For each contoured region, the mean activity concentration was recorded, and the total region activity was determined on the basis of standard MIRD adult source-organ volumes. Three-phase time–activity curves were generated through an automated routine (18), and integrated cumulated activity values (MBq⋅h/MBq) were converted to organ self- and cross-doses using OLINDA dose factors (S values) (19). Mean organ doses, effective dose contribution, and total effective dose were based on International Commission on Radiological Protection Publication 89 tissue-weighting factors.

Statistical Analysis

All statistical analyses were performed in the R statistical software package, version 3.4.2, using standard and validated statistical procedures. All statistical analysis results and their interpretation were independently reviewed by a qualified statistician. The %ID in each organ of interest (liver, lungs, spleen, kidneys, bone marrow, thyroid, pituitary gland) across the 10 patients at each time point are reported using standard descriptive baseline statistics, including means, SD, medians, and ranges for continuously valued variables, and cell counts and percentages for categorically valued variables. Statistical methods consisted of production of Bland–Altman plots and Wilcoxon tests for paired data, performed using the base package of the R language for statistical computing and commonly used add-on packages (gdata, Hmisc, plyr, R2wd, and utils). P values of less than 0.05 on paired analyses were considered significant.

RESULTS

In total, 10 patients were recruited between June and October 2015. The clinical characteristics of these patients are included in Table 1.

TABLE 1.

Baseline Characteristics

Variable	Statistic/level	Count/value
Sex (n)	F	4
	M	6
Age (y)	Mean	63.2 (SD, 10)
	Median	62 (range, 46–78)
	Interquartile range	57.2–71.5
Baseline ECOG	0	6 (60%)
	1	4 (40%)
Baseline EGFR	≤70	3 (30%)
	>70	7 (70%)
Weight (kg)	Mean	87.7 (SD, 16)
	Median	90.5 (range, 60–112)
	Interquartile range	74.6–98
Height (cm)	Mean	173.2 (SD, 7)
	Median	173 (range, 163–185)
	Interquartile range	168–176.5

ECOG = Eastern Cooperative Oncology Group performance; EGFR = estimated glomerular filtration rate (mL/min/1.73 m²).

Safety of ⁶⁴Cu-SARTATE

No grade 3 or 4 adverse events were reported. A single grade 2 event was recorded (transient wheezing). However, this was considered unrelated to treatment. Five grade 1 events were recorded: 2 were considered unrelated to the investigational agent (nausea and petechial rash), 2 were considered possibly related (flushing and leukopenia), and 1 (lymphopenia) was considered probably related. Except for the 2 hematologic reactions, the adverse events were of short duration (<7 d). The patient with lymphopenia (patient 9) had previously received PRRT and had lymphocyte counts persistently in the lower end of the reference range since treatment. These dipped to 0.95 × 10⁹/L (reference range, 1.00–4.00 × 10⁹/L) at 1 wk after the scan before returning to the low reference range within 4 wk. The patient with leukopenia (patient 6) had a normal baseline white cell count that dropped to 3.9 × 10⁹/L (reference range, >4.00 × 10⁹/L) at 6 wk after the scan before returning to the reference range by follow-up at 3 mo.

%ID and SUV in Reference Organs and Tissues

The %ID in reference tissues and organs are detailed for each imaging time-point in Table 2. As expected, there was relatively rapid clearance of blood-pool activity, slower clearance of renal activity, and intermediate clearance of hepatic activity. High retention was seen in tumor and organs with high somatostatin receptor expression, including the pituitary and spleen.

TABLE 2.

Mean and SD of ⁶⁴Cu-SARTATE %ID for Normal Organs

Organ	30 minutes	1 hour	4 hours	24 hours
Adrenals	0.19 (0.13)	0.17 (0.12)	0.19 (0.14)	0.17 (0.16)
Brain	0.23 (0.06)	0.17 (0.10)	0.22 (0.06)	0.16 (0.06)
Lower large intestine	0.42 (0.18)	0.44 (0.21)	0.47 (0.19)	0.69 (0.37)
Small intestine	4.40 (1.38)	4.32 (0.99)	3.78 (1.37)	2.90 (0.83)
Stomach	1.69 (1.26)	1.73 (1.22)	0.68 (0.87)	0.82 (0.61)
Upper large intestine	1.60 (1.59)	1.77 (1.65)	1.44 (1.19)	1.34 (0.59)
Heart	0.76 (0.37)	0.66 (0.31)	0.48 (0.30)	0.37 (0.19)
Kidneys	6.40 (1.59)	5.22 (1.51)	4.33 (0.91)	3.49 (0.68)
Liver	15.02 (2.06)	13.90 (2.30)	11.79 (3.12)	6.79 (2.87)
Lungs	0.99 (0.18)	0.90 (0.25)	0.66 (0.18)	0.46 (0.12)
Muscle	2.27 (0.48)	2.13 (0.44)	1.82 (0.51)	1.32 (0.36)
Pancreas	0.52 (0.25)	0.49 (0.31)	0.50 (0.39)	0.41 (0.24)
Red marrow	7.15 (1.39)	6.28 (1.39)	5.62 (1.38)	4.44 (0.99)
Spleen	4.68 (1.90)	4.94 (1.96)	5.41 (2.07)	4.06 (1.90)
Thyroid	0.05 (0.02)	0.04 (0.02)	0.04 (0.02)	0.04 (0.03)
Urinary bladder	5.27 (3.55)	5.72 (4.71)	2.13 (2.85)	1.49 (1.11)
Remainder of body	55.62 (11.63)	56.77 (12.44)	52.58 (14.02)	42.34 (13.45)

SDs are in parentheses.

Comparison of Same-Day ⁶⁸Ga-DOTATATE and ⁶⁴Cu-SARTATE Images

A comparison of the baseline ⁶⁸Ga-DOTATATE and serial ⁶⁴Cu-SARTATE images in all patients is displayed in Figure 2 using an upper threshold SUV of 20. The quality of images obtained at 1 h with ⁶⁴Cu-SARTATE was judged to be comparable to that of images obtained using ⁶⁸Ga-DOTATATE in 9 of 10 patients. One patient had superior definition of small liver lesions on ⁶⁸Ga-DOTATATE, which was judged as being diagnostically superior by 2 of 3 masked readers. However, the 4-h image in this patient was felt to be comparable to—and the 24-h ⁶⁴Cu-SARTATE was judged by all readers to be superior to—the ⁶⁸Ga-DOTATATE PET/CT obtained at 1 h. The SUV_max of the most intense lesion and mean of up to 5 reference lesions selected on the basis of baseline ⁶⁸Ga-DOTATATE (median, 4 lesions; mean, 3.3 lesions; range, 1–5 lesions) for each individual patient were significantly lower for ⁶⁸Ga-DOTATATE than for ⁶⁴Cu-SARTATE at both 1 h and 4 h (Table 3). Although the SUV_max of lesions did not change significantly on ⁶⁴Cu-SARTATE between 4 and 24 h, the progressive increase in lesion-to-liver ratio suggests that delayed imaging would provide optimal lesion detection (Table 4). Figure 3 provides an example of improved lesion detection on late imaging.

FIGURE 2.

Comparison of tracers. Maximum-intensity projections obtained at 1 h for ⁶⁸Ga-DOTATATE (GaTate) are compared with those obtained at 30 min, 1 h, 4 h, and 24 h for ⁶⁴Cu-SARTATE. Although ⁶⁴Cu-SARTATE images were deemed comparable at 1 h in 9 of 10 patients (patient 6 was the exception), all were considered equivalent or superior to ⁶⁸Ga-DOTATATE at both 4 h and 24 h.

TABLE 3.

Comparison of Lesion and Normal-Organ Uptake and Retention for ⁶⁸Ga-DOTATATE and ⁶⁴Cu-SARTATE

Measure	68Ga-DOTATATE	64Cu-SARTATE	Difference	95% CI	P*
Median at 1 h
Max lesion SUVmax	31.65	41.90	12.45	5.1, 21.4	0.010
Mean patient SUVmax	23.61	33.32	10.35	5.4, 15.3	0.002
Liver SUVmean	7.10	7.45	0.30	−0.2, 0.8	0.24
Spleen SUVmean	20.40	28.25	9.05	4.0, 14.9	0.004
Kidney SUVmean	13.10	20.98	6.80	5.4, 12.1	0.002
Median at 1 h (68Ga-DOTATATE) or 4 h (64Cu-SARTATE)
Max lesion SUVmax	31.65	50.50	17.10	9.6, 36.3	0.004
Mean patient SUVmax	23.61	34.80	12.99	6.3, 27.9	0.004
Liver SUVmean	7.10	5.90	−1.05	−2.2, 1.0	0.19
Spleen SUVmean	20.40	35.55	17.50	7.7, 102.2	0.020
Kidney SUVmean	13.10	18.05	4.86	2.1, 7.0	0.004

^*Paired Wilcoxon test on 10 patients.

CI = confidence interval; max lesion = most intense lesion; mean patient = mean of up to 5 reference lesions.

TABLE 4.

Comparison of Lesion-to-Liver Ratios of ⁶⁸Ga-DOTATATE and ⁶⁴Cu-SARTATE

Ratio 1	Median	Ratio 2	Median	Difference	95% CI	P*
68Ga-DOTATATE to liver (1 h)	3.92	64Cu-SARTATE to liver (1 h)	5.45	1.35	0.7, 2.2	0.004
68Ga-DOTATATE to liver (1 h)	3.92	64Cu-SARTATE to liver (4 h)	6.70	3.86	1.5, 6.4	0.002
64Cu-SARTATE to liver (4 h)	6.70	64Cu-SARTATE to liver (24 h)	16.69	6.75	3.4, 10.3	0.002

^*Paired Wilcoxon test on 10 patients.

CI = confidence interval.

FIGURE 3.

Superior lesion detection at 4 h. High lesion contrast on ⁶⁴Cu-SARTATE images at 4 h (right) better defines regional nodal disease than ⁶⁸Ga-DOTATATE images at 1 h (left) in patient with large pancreatic primary tumor but slightly greater small-bowel activity, indicating some hepatobiliary excretion.

Delayed Image Quality

Image quality remained high at 24 h after administration of ⁶⁴Cu-SARTATE (Fig. 2). The delayed images demonstrated clearance of activity from the liver relative to tumor resulting in qualitatively improved image quality. This finding reflected primarily washout of liver activity, as the median SUV_max of lesions at 24 h was not significantly different from that at 4 h (34.8 vs. 33.3, P = 0.16).

Radiation Dosimetry

Dose estimates for major organs are presented in Table 5. The mean effective dose was 8.7 ± 1.6 mSv per administration (mean injected activity, 192 MBq) or 0.0454 mSv/MBq. Accordingly, an administered activity of 200 MBq would give an effective whole-body dose of approximately 9 mSv plus that related to the CT component of the study.

TABLE 5.

Absorbed Organ Doses in Comparison with ⁶⁴Cu-DOTATATE Data of Pfeifer et al. (24)

Organ	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5	Patient 6	Patient 7	Patient 8	Patient 9	Patient 10	Mean	SD	Pfeifer mean
Adrenals	4.14E−01	1.81E−01	4.88E−02	1.96E−01	8.19E−02	5.67E−02	3.18E−02	2.36E−01	1.55E−01	2.91E−01	1.69E−01	1.22E−01	1.37E−01
Brain	1.53E−02	1.48E−02	2.19E−02	1.43E−02	9.45E−03	1.51E−02	1.39E−02	1.03E−02	1.20E−02	1.15E−02	1.39E−02	3.50E−03	1.27E−02
Breasts	1.38E−02	1.37E−02	2.07E−02	1.18E−02	8.86E−03	1.33E−02	1.18E−02	9.68E−03	1.09E−02	1.08E−02	1.25E−02	3.31E−03	1.32E−02
Gallbladder wall	2.78E−02	2.79E−02	3.60E−02	2.33E−02	1.91E−02	2.43E−02	2.36E−02	1.94E−02	2.14E−02	2.14E−02	2.44E−02	5.06E−03	3.96E−02
LLI wall	6.47E−02	3.24E−02	6.74E−02	7.06E−02	3.29E−02	4.44E−02	5.68E−02	2.90E−02	3.13E−02	5.35E−02	4.83E−02	1.63E−02	4.32E−02
Small intestine	6.83E−02	5.53E−02	7.92E−02	8.36E−02	5.83E−02	8.59E−02	9.53E−02	6.16E−02	4.98E−02	6.88E−02	7.06E−02	1.49E−02	6.55E−02
Stomach wall	6.92E−02	5.43E−02	3.85E−02	3.61E−02	4.13E−02	3.62E−02	2.58E−02	2.19E−02	4.87E−02	7.50E−02	4.47E−02	1.73E−02	1.93E−02
ULI wall	5.33E−02	5.55E−02	7.02E−02	3.91E−02	4.39E−02	1.24E−01	7.74E−02	4.79E−02	5.37E−02	5.92E−02	6.24E−02	2.45E−02	2.18E−02
Heart wall	2.66E−02	2.65E−02	3.69E−02	1.92E−02	1.62E−02	2.11E−02	2.33E−02	1.68E−02	2.98E−02	2.02E−02	2.37E−02	6.41E−03	1.86E−02
Kidneys	1.48E−01	1.88E−01	2.76E−01	2.20E−01	1.87E−01	2.42E−01	2.10E−01	1.82E−01	1.97E−01	1.73E−01	2.02E−01	3.66E−02	1.39E−01
Liver	1.20E−01	1.21E−01	1.14E−01	8.15E−02	8.14E−02	5.21E−02	7.88E−02	7.01E−02	7.56E−02	7.22E−02	8.67E−02	2.34E−02	1.61E−01
Lungs	2.75E−02	2.60E−02	2.89E−02	2.10E−02	1.71E−02	2.30E−02	2.39E−02	1.92E−02	2.03E−02	2.00E−02	2.27E−02	3.85E−03	1.67E−02
Muscle	1.68E−02	1.63E−02	2.43E−02	1.48E−02	1.10E−02	1.71E−02	1.50E−02	1.20E−02	1.37E−02	1.36E−02	1.55E−02	3.70E−03	1.90E−02
Ovaries	2.13E−02	1.96E−02	3.03E−02	2.07E−02	1.44E−02	2.44E−02	2.07E−02	1.53E−02	1.64E−02	1.79E−02	2.01E−02	4.68E−03	1.92E−02
Pancreas	9.40E−02	1.89E−01	7.24E−02	9.91E−02	4.48E−02	8.65E−02	6.95E−02	6.85E−02	1.05E−01	7.01E−02	8.99E−02	3.91E−02	9.27E−02
Red marrow	4.87E−02	3.99E−02	6.23E−02	5.03E−02	3.23E−02	4.12E−02	4.34E−02	3.40E−02	4.47E−02	3.74E−02	4.34E−02	8.82E−03	2.65E−02
Osteogenic cells	4.76E−02	4.26E−02	6.74E−02	4.57E−02	3.09E−02	4.36E−02	4.16E−02	3.31E−02	4.05E−02	3.69E−02	4.30E−02	1.01E−02	3.35E−02
Skin	1.29E−02	1.28E−02	1.98E−02	1.14E−02	8.36E−03	1.30E−02	1.12E−02	9.16E−03	1.02E−02	1.03E−02	1.19E−02	3.19E−03	1.22E−02
Spleen	2.47E−01	3.05E−01	3.33E−01	2.01E−01	2.64E−01	2.55E−01	3.97E−01	5.96E−01	6.20E−01	3.92E−01	3.61E−01	1.44E−01	1.15E−01
Testes	1.41E−02	1.38E−02	2.18E−02	1.30E−02	9.04E−03	1.55E−02	1.24E−02	9.78E−03	1.09E−02	1.13E−02	1.31E−02	3.64E−03	1.36E−02
Thymus	1.55E−02	1.53E−02	2.34E−02	1.33E−02	9.84E−03	1.49E−02	1.33E−02	1.07E−02	1.25E−02	1.21E−02	1.41E−02	3.77E−03	1.49E−02
Thyroid	5.24E−02	3.68E−02	8.06E−02	2.29E−02	1.98E−02	5.25E−02	4.32E−02	2.01E−02	3.72E−02	2.41E−02	3.90E−02	1.92E−02	1.41E−02
UB wall	6.73E−02	6.03E−02	7.66E−02	1.27E−01	5.59E−02	2.16E−01	7.38E−02	4.91E−02	5.30E−02	6.61E−02	8.46E−02	5.13E−02	3.70E−02
Uterus	2.05E−02	1.93E−02	2.95E−02	2.08E−02	1.41E−02	2.56E−02	1.98E−02	1.49E−02	1.60E−02	1.73E−02	1.98E−02	4.81E−03	1.89E−02
Total body	2.22E−02	2.17E−02	3.01E−02	1.97E−02	1.53E−02	2.10E−02	2.00E−02	1.68E−02	1.88E−02	1.80E−02	2.04E−02	4.06E−03	2.50E−02

LLI = lower large intestine; ULI = upper large intestine; UB = urinary bladder.

Data are mGy/MBq.

DISCUSSION

There is increasing evidence that changing either the radionuclide or the chelating agent can alter the affinity of SSAs for somatostatin receptors (20). Accordingly, the performance of novel radiopharmaceuticals cannot be reliably predicted by the receptor-binding affinity of the peptide (21). Additionally, when a long-lived radioisotope is used, retention of activity at sites of disease is important both for diagnostic and for therapeutic purposes. The radionuclides ⁶⁴Cu and ⁶⁷Cu are an attractive theranostic pair. The main advantages of ⁶⁴Cu as a diagnostic agent are its favorable positron energy, cyclotron production, and a half-life that enables both distribution to sites without on-site radiochemistry and delayed imaging after administration to patients (22). Imaging of tissue clearance kinetics will aid prospective dosimetry calculations for its therapeutic pair, ⁶⁷Cu, which also has favorable physical decay characteristics (14). These include a β-energy that is similar to that of ¹⁷⁷Lu (580 keV vs. 497 keV) but with a significantly shorter half-life (2.58 d vs. 6.71 d), providing a higher dose-rate.

The first study to use ⁶⁴Cu to image neuroendocrine tumors was completed by Anderson et al. in 2001 (23). Eight subjects with a history of neuroendocrine tumors were imaged by conventional scintigraphy with ¹¹¹In-pentetreotide, followed by PET imaging with ⁶⁴Cu-TETA-octreotide. PET images were collected at times ranging from 0 to 36 h after injection. Although a formal analysis of temporal changes in tracer uptake in tumor deposits was not performed, the authors noted that lesions were more clearly visualized on the early images than on the delayed images and that the best imaging time appeared to be approximately 4–6 h after tracer injection. A subsequent study completed by Pfeifer et al. (24), investigated the biodistribution and image quality of ⁶⁴Cu-DOTATATE in 14 subjects with histologically confirmed neuroendocrine tumors, again with comparison to ¹¹¹In-pentetreotide. PET/CT scans were acquired at 1, 3, and 24 h after administration. Compared with conventional scintigraphy, ⁶⁴Cu-DOTATATE PET/CT identified additional lesions in 6 of 14 (43%) patients. However, the images presented in this paper also suggest poor visualization of lesions at 24 h. A more recent paper from the same group compared ⁶⁴Cu-DOTATATE PET/CT to ¹¹¹In-pentetreotide SPECT/CT in 111 patients but imaged patients with the PET tracer only at 1 h after tracer administration (25). Although this emulates the time at which ⁶⁸Ga tracers are typically imaged, it does not leverage the potential advantages of the longer half-life of ⁶⁴Cu. To our knowledge, no direct comparison of ⁶⁴Cu-DOTATATE and ⁶⁸Ga-DOTATATE in humans has been reported.

As well as assessing its safety, this first-time-in-humans study was designed to assess whether the high binding-affinity of the MeCOSar chelating agent for ⁶⁴Cu observed in murine models (15) is reproduced in humans. Supporting the preclinical promise of this agent, there was high retention in tumor deposits and substantial hepatic clearance leading to a significant increase in lesion contrast in the liver, as well as excellent ongoing visualization of extrahepatic disease. Image quality and SUV analysis suggests that, for the patient convenience of same-day imaging, acquisition at 4 h would be optimal, although earlier imaging is feasible, being comparable with ⁶⁸Ga-DOTATATE PET/CT in most (9/10) patients. Although less convenient for patients, delayed imaging at 24 h is also feasible, with particularly high lesion-to-liver contrast at this time, potentially improving the sensitivity of detection of disease in this organ, which is a major site of metastases in NEN (26).

Since all patients were selected on the basis of a positive ⁶⁸Ga-DOTATATE scan, the utility of ⁶⁴Cu-SARTATE in patients with low or absent tumor uptake on ⁶⁸Ga-DOTATATE PET/CT could not be assessed. Another potential limitation is that ⁶⁸Ga-DOTATATE PET/CT always preceded the ⁶⁴Cu-SARTATE study. However, reference lesions were all first defined on the ⁶⁸Ga-DOTATATE study and were the most intense lesions on this scan. Accordingly, the likelihood of partial-volume effects was low on the initial scan and, given selection of G1–G2 patients, the likelihood that significant growth through the partial-volume count recovery curve accounts for the increased lesion visualization is low with a mean delay of only 14 d between the scans.

No significant adverse events were seen with ⁶⁴Cu-SARTATE administration. The transient minor reduction in lymphocyte and leukocyte counts in 2 patients did not appear to be related to radiation dose to either the spleen or the red marrow, which were similar to those in other patients. This reduction may represent a normal fluctuation in these counts, given that both patients had counts that were otherwise persistently in the lower end of the reference range, in one case probably related to prior radionuclide therapy. We do not consider these findings likely to affect the suitability of this peptide as a therapeutic agent. The effective whole-body radiation dose was approximately 9 mSv for a 200-MBq (0.0454 mSv/MBq) administered activity. This estimate is in keeping with the long physical half-life and per-decay energy emission of ⁶⁴Cu relative to other PET isotopes, as well as previous dose reports for ⁶⁴Cu-DOTATATE by Pfeifer et al. (0.0315 mSv/MBq) (24). Organ doses were also comparable to the Pfeifer publication, although kidney and spleen doses were higher as indicated by the SUV data. These slight differences between dosimetry data may be related to the biologic affinity of SARTATE versus DOTATATE molecules for somatostatin receptor-2 and altered clearance from normal organs.

CONCLUSION

High uptake and retention of ⁶⁴Cu-SARTATE in tumors and accompanying clearance of activity from the liver provides high-contrast diagnostic images until at least 24 h after injection, increasing the flexibility for timing of diagnostic imaging and the potential for multiple time-point dosimetry estimation before ⁶⁷Cu-SARTATE PRRT. The longer half-life of ⁶⁴Cu than of ⁶⁸Ga also makes the former agent suitable for commercial good-manufacturing-practice production and distribution to remote sites. This preliminary evaluation indicates that the agent is safe and has acceptable radiation dosimetry for human use as a diagnostic agent, with improved imaging characteristics at 4 h compared with 1 h.

DISCLOSURE

This study was funded by Clarity Pharmaceuticals. Rodney Hicks is the recipient of a National Health and Medical Research Council Practitioner Fellowship (APP1108050), which supported this work. Amos Hedt and Matthew Harris are employed by Clarity Pharmaceuticals, the licensee of the intellectual property for SARTATE. Charmaine Jeffery has Clarity Pharmaceuticals share options. Paul Donnelly and Brett Paterson are inventors of, and hold intellectual property in, this area of research, which has been licensed from the University of Melbourne to Clarity Pharmaceuticals. Brett Paterson and Paul Donnelly possess share options in Clarity Pharmaceuticals. Paul Donnelly serves on the Scientific Advisory Board of Clarity Pharmaceuticals. Unrelated to this project, Rodney Hicks has share options in Telix Radiopharmaceuticals that are held on behalf of the Peter MacCallum Cancer Centre. No other potential conflict of interest relevant to this article was reported.