|Home | About | Journals | Submit | Contact Us | Français|
68Ga-DOTATATE is a radiolabeled peptide-based agonist that targets somatostatin receptors overexpressed in neuroendocrine tumors. Here, we present our results on validation of organic matrix 68Ge/68Ga generators (ITG GmbH) applied for radiosynthesis of the clinical doses of 68Ga-DOTATATE (GalioMedixTM).
The clinical grade of DOTATATE (25 µg±5µg) compounded in 1MNaOAc at pH=5.5 was labeled manually with 514±218MBq (13.89±5.9 mCi) of 68Ga eluate in 0.05 N HCl at 95 °C for 10 min. The radiochemical purity of the final dose was validated using radio-TLC. The quality control of clinical doses included tests of their osmolarity, endotoxin level, radionuclide identity, filter integrity, pH, sterility and 68Ge breakthrough.
The final dose of 272±126MBq (7.35±3.4 mCi) of 68Ga-DOTATATE was produced with a radiochemical yield (RCY) of 99%±1%. The total time required for completion of radiolabeling and quality control averaged approximately 35 min. This resulted in delivery of 50% ± 7% of 68Ga-DOTATATE at the time of calibration (not decay corrected).
68Ga eluted from the generator was directly applied for labeling of DOTA-peptide with no additional pre-concentration or pre-purification of isotope. The low acidity of 68Ga eluate allows for facile synthesis of clinical doses with radiochemical and radionuclide purity higher than 98% and average activity of 272 ± 126 MBq (7.3 ± 3 mCi). There is no need for post-labeling C18 Sep-Pak purification of final doses of radiotracer. Advances in knowledge and implications for patient care.
The clinical interest in validation of 68Galabeled agents has increased in the past years due to availability of generators from different vendors (Eckert-Ziegler, ITG, iThemba), favorable approach of U.S. FDA agency to initiate clinical trials, and collaboration of U.S. centers with leading EU clinical sites. The list of 68Ga-labeled tracers evaluated in clinical studies should growth because of the sensitivity of PET technique, the simplicity of the shakebake approach for the dose preparation and reliability of 68Ge/68Ga generators. Our studies have confirmed the reproducible elution profile, and high reliability of ITG GmbH generators required for routine doses preparation according to FDA recommendations.
There is a growing interest in use of generator-produced gallium-68 (68Ga) for molecular imaging of diseases [1–19]. This is primarily due to superior spatial resolution of 68Ga positron emission tomography (PET) compared to single photon emission computed tomography (SPECT); advantageous nuclear properties of the 68Ga isotope; and the commercial availability of clinically-useful 68Ge/68Ga generators, which allow for efficient on-demand production of the radionuclide . Three types of 68Ge/68Ga generators are now available from three separate manufacturers (Eckert-Ziegler GmbH (Germany); iThemba (South Africa); and ITG GmbH (Germany)); and are currently available for clinical and research use in the European Union (EU) and the United States (US). Each of these generators shares similar theoretical-design characteristics, but differ in the composition of the immobilizing matrix for 68Ga parent radionuclide, germanium-68 (68Ge). Although the elution times of the various stationary phases are similar (a few minutes), differences in the stationary-phase matrix result also in differences in the strength of hydrochloric acid (HCl) that is used for 68Ga elution and in the potential for co-elution (breakthrough) of 68Ge and other metallic impurities. [2,16,19,20]. While the matrix of iThemba and Eckert-Ziegler generators is based on tin oxide (SnO2) and titanium oxide (TiO2), respectively, the adsorbent used in ITG generators consists of organic matrix immobilized on silica-resin [2,21–25]. The adsorbent inside the column consists of derivative of dodecyl gallate (Lauryl G) attached to HPLC-grade silica gel. This novel stationary phase allows for elution of 68Ga using a significantly lower concentration of hydrochloric acid (0.05 M HCl) than is required by other generators . In addition, the silica-based resin has lower affinity to metals such as Zn, Fe, and Ti, which putatively minimizes co-elution of metal contaminants in 68Ga eluate. Herein, we describe a simplified protocol for radiosynthesis of chelator-modified somatostatin analog 68Ga-DOTATATE using the organic-matrix-based 68Ge/68Ga generators (ITG GmbH). We validated the efficacy and reliability of these generators based on their performance during synthesis of more than 100 doses of 68Ga-labeled tracer validated in clinical trial (IND 117289).
The 68Ge/68Ge generators (1110 MBq, 30 mCi ITG GmbH) used in these clinical studies were manufactured under GLP conditions and have a manufacturer-guaranteed shelf life of 6 months (or 125 elutions). Generators were first used for clinical studies for 6 months — and were also further evaluated for research applications by examination of their performance over an extended period of 8 months.
All steps from elution of the generator to labeling of clinical kits were performed manually. An understanding of the elution behavior of 68Ga was developed by use of a fractional elution approach in which 1 mL volume fractions of the eluate were collected sequentially and measured for 68Ga content. Using this approach it was determined that more than 90% of the total radioactivity generator-produced 68Ga could be collected in the second and third 1mLfractions, while marginal amounts of Ga activity were observed in the first and fourth fractions (Fig. 1). These main fractions were used without further purification or processing for radiolabeling of DOTATATE.
The organic-matrix based generators (loaded at the manufacturer site with 1110 MBq, 30 mCi of 68Ge) produced during the first elutions 932–954 MBq (25.1–25.8 mCi of 68Ga that corresponds to 84%–86% yield for a decay-corrected generator (80% without decay correction; Fig. 2). These values were consistent with the specifications of the generator provided by ITG, which guarantees over 80% of activity eluted from the generator at the time of calibration. A similar elution profile was reported for TiO2-based generators (Eckert-Ziegler) and SnO2-based generators (iThemba) . The technical specifications of the ITG GmbH generators are shown in Table 1. According to the manufacturer’s information, the 68Ge content in 68Ga eluate should not exceed 0.005% at the time of calibration. The 68Ge breakthrough was at least a factor of two lower in final doses of 68Ga-DOTATATE without the need for further purification via a C18 Sep-Pak column, as is often used for final purification of chelator-modified 68Ga-labeled peptides. Further, our results indicate that there is no need to purify or pre-concentrate 68Ga eluate to achieve acceptable radiochemical quality control objectives (e.g., radiochemical purity, specific activity, reaction efficiency).
In the course of normal laboratory operations, generators are often eluted twice daily, and as expected, the eluted 68Ga radioactivity is a function of the ingrowth time allowed between elution runs. Our standard procedure permits 4 h time breaks between elutions, which allowed for regeneration over 90% of 68Ga radioactivity (Table 2).
We synthesized over 100 doses of 68Ga-DOTATATE over a period of more than one year. The clinical doses of DOTATATE (25 µg ± 5 µg) were compounded in 1MNaOAc at pH = 5.5 and labeled with an average of 514±218MBq (13.89±5.9mCi) of 68Ga using a reaction temperature of 95 °C for 10 min (Fig. 3). Reactions were conducted in the presence of ascorbic acid used as a free radical species scavenger. Our IND protocol permits release of 148–592 MBq (4–16 mCi) of 68Ga-DOTATATE produced with a radiochemical yield (%RCY) of ≥ 95% (percent incorporation), as determined by iTLC. In our hands, radiosynthesis of 68Ga-DOTATATE proceeded with average RCY of 99% ± 1% and led to release of 272 ± 126 MBq (7.35 ± 3.4 mCi) of 68Ga-DOTATATE in 9.1 ± 0.3 mL. The presence of metallic impurities in the final dose or eluate was not determined in these studies. This test was not required to complete the quality control of the final dose of 68Ga-DOTATATE.
The total time required for completion of radiolabeling and quality control averaged approximately 35 min. This resulted in delivery of 50% ± 7% of 68Ga-DOTATATE at the time of calibration (not decay corrected) compared to the starting activity of this radiotracer at the end of synthesis (EOS) (Fig. 4).
For the clinical studies presented here, doses of 68Ga-DOTATATE produced during a single radiolabeling reaction (using the 1110 MBq, 30 mCi generator) were used for PET/CT imaging of 1–3 patients. The new generators (loaded at the manufacturer site with 1110 MBq, 30 mCi of 68Ge) can produce 932–954 MBq (25.1–25.8 mCi) of 68Ga during the first elutions. This activity is sufficient to formulate a dose of 20.4–21.01 mCi of 68Ga-tracer that can be dispensed into 2–3 doses of 68Ga-DOTATATE. Since this generator can be eluted every 4 h, our experience suggests that up to 4 patients per day can be imaged with radiotracer, under the procedures in the present study, without stretching generator limits.
The quality control requirements and their specifications as listed in the 68Ga- DOTATATE certificate of analysis (COA) are shown in Table 3. Recently, the Society of Nuclear Medicine and Molecular Imaging (SNMMI) Clinical Trials Network (CTN) published a harmonized release specification document for 68Ga-DOTATOC. The COA analysis of 68Ga-DOTATATE adheres to the recommendations of the SNMMI CTN release criteria, as of the writing of this manuscript.
For the production of the clinical doses of 68Ga-DOTATATE, 68Ge breakthrough was monitored over a period of 6 months. The 68Ge content in the final doses of radiotracer was 0.0026% ± 0.002%. Our IND clinical protocol permits the 68Ge content in the clinical doses to be equal to or less than 0.01% (Fig. 5). The limit of 68Ge breakthrough allowed in US clinical trials of Ga68-radiotracers is higher than the values recommended by the European Pharmacopeia . As postulated by Velikyan at el, the low limit of 68Ge (0.001%) accepted by European Pharmacopeia could be increased at least 100 times without compromising the patient’s safety .
During our initial research studies, we monitored 68Ge content directly in 68Ga eluate over the period of 8 months. The average value of 68Ge breakthrough was 0.0032% during the first 6 months of the use of generator and did not exceed 0.0045%when monitored for an additional 2 months (Fig. 6).
Our observations demonstrate that pre-purification and pre-concentration of 68Ga eluate and post-purification of the final dose are not required prior to release of 68Ga-DOTATATE under our IND protocol. On the other hand, this first step of 68Ga pre-concentration may be recommended for labeling reactions proceeding in large volumes. The reduction of 68Ga volume may be mandatory for clinical applications using two different generators connected in series or using very small amounts (<5 µg) of precursor for labeling. Thus, we evaluated currently available methods of pre-concentration, with the goal of selecting one method, which could deliver 68Ga with the highest %RCY (Fig. 7). We modified the published protocols to adapt them for use of lower molarity HCl recommended for ITG generators. The percent adsorption of 68Ga eluted from the generator and concentrated using commercially available ion exchange resins was ≥ 81% for all of the examined resins. On the other hand, desorption of 68Ga (i.e., its elution from resins) varied widely from 54% to 83%. There were minor differences in elution yield of 68Ga for AG-1X8, Strata-XC and SCX, with the highest desorption yields achieved using the AG-1X8 and Strata SCX column. Overall, our results indicated that all of examined resins can be applied for pre-concentration of 68Ga.
The 68Ge content determined in 68Ga eluates collected for these experiments was on the order of 0.0002%, which represents a 10-fold decrease compared to the concentration of 68Ge in generator-eluate measured without the pre-concentration step. Thus, our results suggest that the pre-concentration step can be used to remove about 84% of 68Ge breakthrough (Fig. 8). We further find the efficacies of all ion-exchangers studied to be comparable.
The organic-matrix based 68Ge/68Ga generators produced by ITG GmbH Germany perform consistently and provide a high purity 68Ga that can be directly used for radiopharmaceutical preparation without the need for pre-concentration/purification of the eluted 68Ga. This is likely to be valid for most peptides encountered and is consistent with the acidic conditions required for high efficiency radiolabeling with 68Ga for most chelators currently available commercially. In our studies, ITG GmbH generators delivered consistent and reliable elution profiles with efficiency required for routine doses preparation in clinical studies. The simplified protocol for clinical dose preparation; low level of 68Ge content; and absence of metal contaminants in the eluate as verified by producer, define the ITG 68Ge/68Ga generator as a competitive alternative to other commercially-available generators (Fig. 9). ITG GmbH introduced recently a GMP-grade generator with extended shelf life of 12 months (or 250 elutions).
68Ga (t1/2 = 68 min, β+ intensity = 89% and EC = 11%) was produced by elution of 68Ge/68Ga generators (ITG GmbH Germany) using 4 ml of 0.05 M HCl. Three generators loaded with 1110 MBq (30 mCi) of 68Ge and one generator with 370 MBq (10 mCi) of 68Ge were evaluated in these studies. Reagents used in all described reactions and in the final product preparation were ultra-pure trace-metal-free grade. Commercially available GMP-grade DOTATATE (5 mg, Bachem) was compounded into doses of 25 ± 5 µg peptide dissolved in 100 µL 1 M metals-grade sodium acetate (NaOAc; pH = 5.5; Sigma-Aldrich). Metals grade ascorbic acid (50 mg/mL, Trace SELECT®, Sigma Aldrich) was used as a free-radical scavenger during dose preparation.
68Ge/68Ga generator was eluted manually using 0.05 M HCl (Ultrapure trace-metal-free) by fractional elution (1 mL/fraction) or by collecting the entire 4-ml volume of eluate. The radioactivity content of 68Ga eluted from generator and 68Ge content in the final dose [%] were monitored over a period of 6 months for doses prepared for clinical use. This time period was extended for another two months (total of 8 months) for the generators used for research applications.
cGMP-produced DOTATATE (25 ± 5 µg), DOTA-D-Phe1Tyr3-octerotide, where DOTA is 1, 4, 7, 10-teteraazacyclododecane-1, 4, 7, 10-teteraacetic acid), (Bachem) was dissolved in 100 µL 1 M sodium acetate (NaOAc; pH = 5.5) and reacted with 68Ga (6–24 mCi; 222–888 MBq as 68GaCl3) in 2 ml of fractionated eluate. The reaction mixture was incubated for 10 min at 95 °C in a temperature-controlled heating block. After completion of radiolabeling, 0.4 mL of 1 M phosphate buffer (PBS, pH = 7.5) and 2 mL of sterile water for injection (SWFI) were added to the reaction to adjust osmolality of the clinical dose. The dose of 68Ga-DOTATATE was filtered by 0.22 µm Millipore filter and diluted using 5 ml sterile water for injection (SWFI) to reach a final volume of 10 mL. The final quality control protocol for release included tests to determine osmolarity; sterility; radiochemical purity; radionuclide identity; and endotoxin level.
Radiochemical purity of 68Ga-DOTATATE was determined by Instant Thin Layer Paper Chromatography (iTLC; Whatman) using 0.2 M sodium citrate (pH = 5.5) as mobile phase and a BioRadScan as an iTLC reader; RF (free 68Ga) = 0.8–1.0, RF (68Ga peptide) = 0.0–0.3. The radiochemical yield of 68Ga-DOTATATE was confirmed using radio-HPLC (Shimadzu, binary pump; C18 column, 5 µm, 4.6 mm × 150 mm, Waters) and mobile-phases (deionized-distilled water (ddH2O) + 0.1% TFA (mobile phase A) and acetonitrile + 0.1% trifluoroacetic acid (TFA) (mobile phase B). The gradient profile (flow rate 1.0 mL/min) was as follows: 0–1 min 10% B; 1–10 min 10%–50% B, 10–12 min 50% B; 12–15 min 50%–10%; 15–20 min 10% B. HPLC injections of the dose were done using auto-sampler (injection volume 50 µl). The retention time of the radiolabeled peptide was monitored using a standard inline gamma-detector (Bio-Rad).
The osmolarity of 68Ga-DOTATATE was determined using Osmometer model 3250 (Advance Instruments). The final dose of 68Ga-DOTATATE was examined for the presence of endotoxin using 10× dilution of radiotracer and the Endosafe®-PTS™reader. The integrity of the Millipore filter was confirmed using a standard bubble point test.
The sterility of final dose of radiotracer was determined by addition of 0.1mL of 68Ga-DOTATATE added to thioglycolate medium (FTG) and tryptic soy broth and incubation for 14 days at temperature of 30–37 °C and 20–25 °C, respectively. Media were monitored on a daily basis. The sterility of the dose was confirmed when there was no visual evidence of growth in the tested media.
68Ge breakthrough was expressed as a percent of the radioactivity of the parent radionuclide (68Ge) measured at 48 h post elution relative to the radioactivity of daughter radionuclide (68Ga) in the eluate at the time of elution as presented previously . All measurements were performed using a PerkinElmer Gamma Counter (Wizard 1475).
Radionuclide identity was determined based on the decay properties of the 68Ga (t1/2 = 68min). Briefly, the radioactivity of a 20–100 µCi QC-aliquot of 68Ga was measured at 0 min and 10 min post-radiolabeling, using Biodex Atomlab 50 dose calibrator, and compared to theoretical decay of 68Ga. The values recorded within 10% of the known half-life were considered as acceptable for release of the dose.
The pre-concentration of 68Ga was done using ion-exchange resins according to published protocols, which were modified slightly to accommodate the use of lower molarity of HCl [29–34]. To select the most appropriate separation method, we tested several anion-exchange (Bio-Rad AG1-X8, Wax Oasis Waters, Macherey-Nagel 30-PS-HCO3 Chromafix) and cation-exchange resin (Phenomenex Strata XC, Bio-Rad AG50W-X8) and determined their efficacy for 68Ga pre-concentration.
68Ga eluted from the generator using 4 mL of 0.05 M HCl was mixed with 4 mL of 9.5 M HCl (30%) and loaded on the pre-equilibrated Oasis WAX anion-exchange resin (30 µm, 30 mg). The flow through volume was discarded and the resin was washed with 1 mL of 5 M HCl to remove potential metal contaminants. The purified 68Ga was recovered from the resin soaking and slow washing with 500 µL of ddH2O . Alternatively, after washing the resin with 1 mL of 5MHCl, two extra rinses were added to decrease the acidity of the final-concentrated fraction of 68Ga. The first rinse of the resin included the use of 1 mL of 5 M NaCl, followed by 1 mL of ethanol, which served to reduce concentration of Cl−. Finally, the concentrated 68Ga was released from the resin using 0.5 mL of ddH2O .
The strong anion exchange resin, AG-1-X8 resin (Bio-Rad, 200–400 dry mesh) was pre-condition and activated by washing with 1 mL of 4 M HCl, 1 M HCl, ddH2O and again 4 M HCl. 4 mL of concentrated 9.5 M HCl (30%) was mixed with 68GaCl3 in 0.05 M HCl (4 mL), in order to form [68GaCl4]−, and loaded together on AG-1-X8 resin. The flow through volume was discarded and the resin was purged with air and washed out manually—firstwith 1 mL of 5MHCl, followed by 1 mL of 5 M NaCl, and finally 0.5 mL ddH2O to collect concentrated 68Ga [29,34].
The 30-PS-HCO3 anion resin (SPE cartridge Chromafix, Machanery-Nagel, Germany) was pre-equilibrated with 1 mL of 9.5 M HCl (30% HCl), followed by 10 mL of ddH2O [29,34]. 68Ga (5 mL in 0.05 M HCl) was mixed with 4 mL of 30% HCl to reach a final concentration of 5 M HCl. This acidic solution of 68Ga was loaded on the 30-PS-HCO3 anion resin. The cartridge was washed first with 1mL of 5MHCl, and then two extra rinses were added to decrease acidity of final concentrated fraction 68Ga. First rinse of the resin included the use of 1 mL of 5 M NaCl, followed by 1 mL of ethanol. The concentrated 68Ga was released from the resin using 0.5 mL of ddH2O.
40 mgof the AG50W-X8 resin (Bio-Rad, 200–400 dry mesh size) was loaded into mini-column (1 cm inner diameter). The resin was pre-equilibrated by washing with 1 mL of ddH2O, followed by 1 mL of 8 M HCl. Concentrated HCl (4 ml) was mixed with Ga68 eluate (4 ml) and loaded on the resins, followed by washing with 1 mL saturated 5MNaCl. The pre-concentrated 68Ga was collected by final slow elution of resin with 0.5 mL of deionized water [29,30].
Strata-SCX column (Varian, Bond Elut-SCX, 100 mg, 1 mL column) was pre-equilibrated by washing with 1 mL of 5.5 M HCl, followed by 10 mL of ddH2O. A 20 µL volume of 5.5 M HCl was added to 5 mL of Ga68 eluate in 0.05 M HCl and loaded on the Strata-SCX column. The flow through was discarded and the cartridge was washed slowly with 1 mL of 5 M of NaCl to collect concentrated 68Ga eluate .
I.T. and D.R. designed and conducted all experiments and have an equal contribution. S.M., K.Z., S.T. provided assistance throughout experimentation. M.S. provided conceptual assistance on 68Ga purification strategies. M.S., S.M., K.Z., D.R, S.T, E.D., and A.M. reviewed the manuscript. I.T. analyzed the data and wrote the manuscript, with editorial assistance of M.S.
Authors would like to thank Arthur E. Camp (Iso-Tex Diagnostics, Inc.), Jim (Jaime) Simon (IsoTherapeutics Group LLC) and Russell Hildebrandt (Southwest South Texas Nuclear Pharmacy Inc.) for help in setting up production of clinical kits and guidance on their quality control validation.