Scaled-up radiolabelling of DOTATATE with ⁶⁸Ga eluted from a SnO₂-based ⁶⁸Ge/⁶⁸Ga generator

Abstract

A scaled-up radiolabelling and improved post-labelling purification procedure for [⁶⁸Ga]DOTATATE is reported, using a more than 1 year old SnO₂-based 1850 MBq ⁶⁸Ge/⁶⁸Ga generator (initially double-loaded with 3700 MBq ⁶⁸Ge) as a source of ionic ⁶⁸Ga. The elution method of choice comprised elution with 0.6 M HCl in a single 4 mL fraction, containing up to 95% of the total eluted ⁶⁸Ga activity. The unpurified fraction was directly used for labelling after pH adjustment with 2.5 M sodium acetate. Labelling efficiencies were determined at 90–95 °C at various reaction times and reaction volumes of up to 5.7 mL, using either 30 μg or 50 μg DOTATATE. Only the latter amount resulted in consistently high labelling efficiency in excess of 95%. Post-labelling purification, carried out on Sep-Pak C18, showed that 50% ethanol in saline was a superior desorption eluant than 100% ethanol. The highest and most consistent decay-corrected radiochemical yields (89%) were obtained using 50 μg DOTATATE and a 20 min reaction time.

Introduction

The diagnostic applications of ⁶⁸Ga-labelled DOTA-conjugated peptides for positron emission tomography (PET) in oncology have been well documented (Hofmann et al., 2001, Henze et al., 2001, Meyer et al., 2003a, Meyer et al., 2004, Maecke et al., 2005, Breeman et al., 2005). The somatostatin analogue DOTA-D-Phe¹–Tyr³-Octreotide (DOTATOC) appears to be the gold standard for ⁶⁸Ga-based PET peptide radiopharmaceuticals (Maecke et al., 2005), but DOTATATE, in which the terminal Thr(ol) amino acid of DOTATOC is replaced with a Thr unit, has also become an important peptide analogue for clinical application (Fani et al., 2008).

The ⁶⁸Ge/⁶⁸Ga generator has become a well established and reliable source of ⁶⁸Ga, making the latter completely cyclotron-independent and routinely available. Various sorbent materials for such generators have been documented over the years, such as organic anion exchange resins (Neirinckx and Davis, 1980), and more recently, composite sorbents such as nanoceria-polyacrylonitrile (Chakravarty et al., 2010). The most commonly used commercially available ⁶⁸Ge/⁶⁸Ga generator in the recent times has been based on a TiO₂ solid support (Zhernosekov et al., 2007). Recently, a SnO₂-based generator, produced by iThemba LABS, South Africa, has also become commercially available and its characteristics have been described in a paper by De Blois et al. (2011). In order to achieve optimal elution efficiencies with this generator, it has to be eluted with a more acidic eluant (0.6 M HCl) as opposed to the 0.1 M HCl required for the TiO₂-based generator. DOTA-peptide labelling is conducted at pH values between 3 and 5 (Meyer et al., 2003a, Meyer et al., 2004, Breeman et al., 2005). Therefore, when more acidic ⁶⁸Ga-containing eluates are used in peptide labelling, labelling recipes need to be adapted.

Due to the presence of metal contaminants in ⁶⁸Ga solutions that could interfere with labelling, as well as the possible breakthrough of ⁶⁸Ge from generators, the pre-purification of eluates has become an important part of peptide labelling systems (Meyer et al., 2003a, Meyer et al., 2003b, Meyer et al., 2004, Zhernosekov et al., 2005, Zhernosekov et al., 2007). Another advantage of the pre-purification process is the volume reduction of ⁶⁸Ga solutions, as the labelling at nanomolar peptide concentration levels requires small reaction volumes to maximise labelling yields (Meyer et al., 2003a, Meyer et al., 2004, Zhernosekov et al., 2007). Various concentration methods have been described, such as anion (Meyer et al., 2003b, Meyer et al., 2004, Velikyan et al., 2008) and cation (Zhernosekov et al., 2005, Zhernosekov et al., 2007) exchange chromatography. Each of these methods has its advantages and disadvantages; however, the design of labelling systems could become complicated when such pre-purification devices are incorporated into a system. The use of merely a post-labelling purification device, such as a solid phase C18 cartridge, could probably be a more practical option, provided that de-sorption yields are optimised. Concerns about possible complexation of the four-valent radionuclidic impurity ⁶⁸Ge by DOTA-conjugated ligands, regardless the chemical form in which Ge exists, are unjustified. It has been reported that the DOTA moiety forms stable complexes with several 2+ and 3+ charged metals only (Meyer et al., 2004, Heppeler et al., 1999). Furthermore, it has been reported that four-valent metals such as Zr⁴⁺ and Hf⁴⁺ are not competitors for the incorporation of trivalent cations such as In³⁺, Y³⁺ and Lu³⁺ into DOTA (Breeman et al., 2003). Literature reports that C18 purification reduces the ⁶⁸Ge content of labelling mixtures more than 100-fold (Decristoforo et al., 2007) and give further impetus to our statement. Another technique to overcome problems associated with voluminous eluates is their fractionation (Breeman et al., 2005), in which the generator is eluted in such a way that the maximum amount of activity is recovered in the smallest possible volume. This technique, however, might result in the over-complication of labelling systems and/or loss of usable activity in un-used fractions.

With all these in mind, we embarked on a study to develop a simple, practical method to prepare ⁶⁸Ga-labelled DOTATATE on a relatively large scale, using partially fractionated, unpurified ⁶⁸Ga eluates obtained from a SnO₂-based generator. In order to accomplish this, we set out the following objectives: to maximise the amount of eluted activity available for the labelling process, to adapt existing labelling recipes in order to accommodate a more acidic eluate and to scale up labelling reactions at the same time without compromising radiochemical yields, to optimise reaction time in order to achieve optimal effective yields and to improve existing solid phase C18 purification techniques in order to optimise recovery yields and to bring about an injection-ready post-purified product.