Production and Quality Control of [177Lu]Lu-PSMA-I&T: Development of an Investigational Medicinal Product Dossier for Clinical Trials

Since prostate cancer is the most commonly diagnosed malignancy in men, the theranostic approach has become very attractive since the discovery of urea-based PSMA inhibitors. Different molecules have been synthesized starting from the Glu-urea-Lys scaffold as the pharmacophore and then optimizing the linker and the chelate to improve functional characteristics. This article aimed to highlight the quality aspects, which could have an impact on clinical practice, describing the development of an Investigational Medicinal Product Dossier (IMPD) for clinical trials with [177Lu]Lu-PSMA-I&T in prostate cancer and other solid tumors expressing PSMA. The results highlighted some important quality issues of the final preparation: radiolabeling of PSMA-I&T with lutetium-177 needs a considerably longer time compared with the radiolabeling of the well-known [177Lu]Lu-PSMA-617. When the final product was formulated in saline, the stability of [177Lu]Lu-PSMA-I&T was reduced by radiolysis, showing a decrease in radiochemical purity (<95% in 24 h). Different formulations of the final product with increasing concentrations of ascorbic acid have been tested to counteract radiolysis and extend stability. A solution of 20 mg/mL of ascorbic acid in saline prevents radiolysis and ensures stability over 30 h.


Introduction
Prostate cancer (PCa) is the most commonly diagnosed malignancy in men worldwide and remains one of the leading causes of cancer-related deaths. Prostate-specific membrane antigen (PSMA) is a type II membrane glycoprotein with an extensive extracellular domain (44−750 amino acids) and plays a significant role in prostate carcinogenesis and progression [1][2][3][4].
PSMA expression correlates with the malignancy of the disease, being further increased in metastatic and hormone-refractory patients [5]. As a consequence, PSMA has attracted attention as a target for molecular imaging and targeted radioligand therapy, especially in metastatic castration-resistant prostate cancer (mCRPC).
Since the discovery of urea-based PSMA inhibitors in 2001 [6], a variety of PSMAtargeted radioligands for imaging prostate cancer was developed. Briefly, most of the relevant molecules are structured by three main components: a pharmacophore, usually X-urea-Glu (XuE)-scaffold, in our case X = K; a linker to enhance affinity by interaction with receptor hydrophobic pocket; and a chelator to bind the radionuclide. Other types of PSMA inhibitors are the ones labeled with fluorine-18 ([ 18 F]F-PSMA-1007, ([ 18 F]F-DCMPyL, and others), the chelator is in those cases replaced by organic structures with a leaving group to support an SN2 reaction. The "gold standard" ligand for PSMA imaging is Glu-NH-CO-NH-Lys(Ahx)-68 Ga-HBED-CC ([ 68 Ga]Ga-PSMA-HBED-CC) characterized by a high-affinity chelator for Gallium-68 (logK = 38.5) that seems to interact advantageously with the lipophilic part of the PSMA binding pocket [7]. Unfortunately, the HBED-CC chelator is unsuitable for radiolabeling with therapeutic radiometals such as yttrium-90 and lutetium-177, and therefore new theranostic compounds were designed to bind gallium-68, as well as yttrium-90 and lutetium-177.
Substitution of HBED-CC with DOTA and systematical modification of the side chain with the introduction of a naphtylic linker have led to PSMA-617 [8] with excellent pharmacokinetic properties, high binding affinity and internalization, prolonged tumor uptake, and high tumor-to-background ratio, which are extremely important for both imaging quality and therapy.
In order to use a radiopharmaceutical in human applications, it should be manufactured under Good Manufacturing Practice (GMP) or National Regulations. In Italy, the reference quality assurance system is "Norme di Buona Preparazione dei Radiofarmaci per Medicina Nucleare (NBP-MN)" [11], which is a GMP-like quality assurance system dealing with no-profit clinical trials.
To submit the clinical study to the Italian Medicines Agency (AIFA), an Investigational Medicinal Product Dossier (IMPD) needs to be produced for [ 177 Lu]Lu-PSMA-I&T according to European Medicines Agency EMA guideline [12]. This guideline aims to address the documentation on the chemical and pharmaceutical quality of investigational medicinal products (IMPs), including radiopharmaceuticals, to ensure their quality, safety, and efficacy.
The aim of this paper is to describe the process validation as well as the analytical methods, along with establishing acceptance criteria for [ 177 Lu]Lu-PSMA-I&T according to the purpose of obtaining an IMPD.
At the moment the ligand PSMA-I&T is commercially available from ABX and there are no patents that prevent us from working with PSMA-I&T.

Results
IMPD for [ 177 Lu]Lu-PSMA-I&T was prepared according to the EMA guideline [12]. IMPD includes the most up-to-date information relevant to the clinical trial available at the time of submission of the clinical trial application. It essentially consists of two parts, the first dedicated to the drug substance and the second dedicated to the investigational medicinal product under test.

Drug Substance
Two drug substances have been identified: the ligand PSMA-I&T and the precursor [ 177

Drug Substance
Two drug substances have been identified: the ligand PSMA-I&T and the precursor [ 177 Lu]LuCl3. The chemical structure of PSMA-I&T is showed in Figure 1.
Lutetium-177 was produced according to the following nuclear reaction: [ 176 Yb(n,γ) 177 Yb → (β -) → 177 Lu] The neutron thermal flux was between 1013 and 1016 cm −2 s −1 . The nuclear reaction is no-carried added (n.c.a). The n.c.a. reaction results in a very high specific activity (≥3000 GBq/mg), in comparison with the lower specific activity (500 GBq/mg) when the Lutetium-177 is obtained by neutron irradiation of Lutetium-176. Moreover, the "direct production" leads to formation of a radionuclidic impurity (Lutetium-177m), not present in the Lutetium-177 produced by ytterbium-176. Metallic impurities concentration was very low, the sum of both metallic as well as other radionuclidic impurities cannot interfere with labeling nor with radiochemical and radionuclidic purity.
The specifications for the release of lutetium-177 chloride are indicated in Table 1.
Lutetium-177 was produced according to the following nuclear reaction: The neutron thermal flux was between 1013 and 1016 cm −2 s −1 . The nuclear reaction is no-carried added (n.c.a.). The n.c.a. reaction results in a very high specific activity (≥3000 GBq/mg), in comparison with the lower specific activity (500 GBq/mg) when the Lutetium-177 is obtained by neutron irradiation of Lutetium-176. Moreover, the "direct production" leads to formation of a radionuclidic impurity (Lutetium-177m), not present in the Lutetium-177 produced by ytterbium-176. Metallic impurities concentration was very low, the sum of both metallic as well as other radionuclidic impurities cannot interfere with labeling nor with radiochemical and radionuclidic purity.
The specifications for the release of lutetium-177 chloride are indicated in Table 1.

Investigational Medicinal Product (IMP) under Test Description and Composition of the IMP
The IMP consists of a description of the [ 177 Lu]Lu-PSMA-I&T solution, among other things stating the range of radioactivity (17,980-29,790 MBq), at the end of synthesis (EOS), which, in this case, is also considered Activity Reference Time (ART); the final volume is 17-25 mL.
The radioactive concentration is between 1057 and 1192 MBq/mL. IMP is formulated as a multidose drug with the components described in Table 2.  Figure 2, which shows the difference between the retention time of [ 177 Lu]Lu-PSMA-I&T and the precursor PSMA-I&T (6.72 and 7.01, respectively). The largest peak in the UV chromatogram is gentisic acid. The radiometric detector evidences the high radiochemical purity of the product, since only the peak of [ 177 Lu]Lu-PSMA-I&T is present.

Acceptance Criteria
Acceptance criteria, specifications, and release timing were chosen in compliance with the general texts and monographs of the current European Pharmacopoeia and are summarized in Table 3. The product should meet the acceptance criteria for all the established quality parameters. The administered patient dose ranged from 5500 to 7400 MBq according to the patient conditions. Typical radiometric and UV chromatograms of [ 177 Lu]Lu-PSMA-I&T syntheses are shown in Figure 2, which shows the difference between the retention time of [ 177 Lu]Lu-PSMA-I&T and the precursor PSMA-I&T (6.72 and 7.01, respectively). The largest peak in the UV chromatogram is gentisic acid. The radiometric detector evidences the high radiochemical purity of the product, since only the peak of [ 177 Lu]Lu-PSMA-I&T is present.  All the tests, except sterility, were carried out before the release.

Validation of the Analytical Procedures
Validation is the act of proving that any procedure, process, equipment, material, activity, or system actually leads to the expected results, with the aim to contribute to and guarantee the quality of a radiopharmaceutical.
The objective of validation of an analytical procedure is to demonstrate that it is suitable for its intended purpose.
The validation of the analytical procedures, the acceptance limits, and the parameters (specificity, linearity, range, accuracy, precision, quantification, and detection limit) for performing validation of analytical methods was carried out according to the ICH guideline Q2(R1) [13]. For the HPLC determination of chemical purity, nat Lu-PSMA-I&T and PSMA-I&T were used. The high concentration and absorbance of gentisic acid cause it to be by far the highest peak in the UV (205 nm) spectrum of the PSMA-I&T product; we, therefore, chose to perform the validation of the PSMA-I&T analytical method in presence of gentisic acid to simulate closely the condition of the final product formulation. Parameters and acceptance criteria for the validation of the radio-HPLC method and the results obtained are shown in Table 4. Radionuclidic purity, [ 177   Rs = Resolution, CV= coefficient of variation, R 2 = correlation coefficient, LOQ = quantitation limit, LOD = detection limit, n.a. = not applicable.

Batch Analysis and Process Validation
Process Validation should be intended as a means to establish that all the process parameters that bring to the preparation of the intended radiopharmaceutical and their quality characteristics are consistently and reproducibly met.
Process validation was carried out by producing three different batches of [ 177 Lu]Lu-PSMA-I&T on three different days, in the same conditions set for typical routine preparations and in the activity range reported in the acceptance criteria. Each batch was prepared accordingly the validation protocol should be fully characterized from the analytical point of view, with the aim to verify that the product meets the acceptance criteria as for all the established quality parameters.
Parameters were measured by the tests described in Table 3.
The results for three representative batches are shown in Table 5. All the batches used for process validation complied with the acceptance criteria.

Stability
Stability was assessed at 0, 24, and 30 h after the end of the synthesis. The three batches used for process validation were kept at room temperature and then parameters that could change over time such as appearance, radiochemical purity, and pH were reanalyzed after 24 and 30 h. The synthesis does not affect the radionuclidic purity so the radionuclidic purity is the same as observed on the sheet of [ 177 Lu]LuCl 3 .
The analysis of the radiochemical purity was carried out by HPLC and TLC because the TLC method is used to detect the presence of colloidal lutetium-177 [14] that is not detectable with HPLC analysis.
For organizational reasons, it is not possible to administer the drug to the patient, later than 24 h after preparation. Chromatograms are shown in Figure 4; the stability data are shown in Table 6.       [8,9]. The two molecules have the same pharmacophore but different linkers and chelators.

Discussion
The effect of replacing the DOTA chelator with DOTAGA led to an increase in radiolabeling reaction time: PSMA-I&T was incubated with [ 177 [9] and could probably be ascribed to a slower kinetic of incorporation due to different conformational changes in the chelators along with the overall molecular structure of the linker.
The two molecules showed slight differences in solubility characteristics, which, however, did not require substantial changes in the buffer solutions.
The longer incubation time needed to prepare [ 177 Lu]Lu-PSMA-I&T did not affect the impurities profile. In terms of quality parameters, the experimental results from the three batches of [ 177 Lu]Lu-PSMA-I&T fulfilled the specifications. Furthermore, the radiolabeling conditions always led to a very high radiochemical yield.
To prevent radiolysis of the radiopharmaceutical, the radiolabeling was carried out in presence of ascorbic and gentisic acid.
The tests performed for quality controls are commonly used in radiochemical and radiopharmaceutical methods.
HPLC allows the use of two detectors in series for example UV or a mass detector coupled with a radiometric detector.
Acceptance criteria are based on the Eu. Pharm. general monograph "Radiopharmaceutical Preparation" [15] and on National Regulations for preparation of radiopharmaceuticals [11].
Gamma-ray spectroscopy is used for the measurement of the radionuclidic purity of radiopharmaceuticals. The Pharmacopoeia generally states that radiopharmaceuticals should have a radionuclidic purity of at least 99.9% throughout their shelf-life. High purity germanium detectors are required to detect impurities of less than 0.1%. For identification, the same approach was used to compare γ energies lines characteristics of the radionuclide.
The most significant quality issue in the preparation of [ 177 Lu]Lu-PSMA-I&T is the poor stability of the finished product when diluted with saline to a final volume of 17-25 mL, a general practice with other radiopharmaceuticals such as [ 177 Lu]Lu-PSMA-617. Comparative results on the stability of both radiopharmaceuticals are reported in Table 7. [ 177 Lu]Lu-PSMA-I&T was found to be instable in saline solution (<95% radiochemical purity after 24 h). These data emphasize that what is reported in the Supplemental data by Weineisen et al. [9] either does not consider the long-term stability of the preparation or establishes less stringent acceptance criteria compared with this paper (≥97% radiochemical purity). Another aspect to consider is the large difference between the final radioactivities reported in the literature [9] and the radioactivities used in this study. Long-term radiolysis due to high radioactivity concentration should be taken into account. [ 177  To overcome this problem, [ 177 Lu]Lu-PSMA-I&T was formulated in an ascorbic 20mg/mL saline solution to a final volume of 17-25 mL, resulting in a shelf life of 30 h, which is the optimal time interval for the management of patients.

[ 177 Lu]Lu-PSMA-I&T Manufacturing Process and Process Controls
For the radiolabeling of [ 177 Lu]Lu-PSMA-I&T, a manual synthesis was used. The radiosynthesis was carried out in a shielded isolator offering a class A environment with class B pre-chambers (Manuela Beta, COMECER S.p.A, Castelbolognese, Italy) located in a class C cleanroom. PC 1: Accuracy testing of emitting β-sources is performed prior to the introduction of a new measured geometry (e.g., new vial size of lutetium-177).
PC 2: A daily radioactivity check is performed before each production run and the deviation between the read and calculated activity, according to the calibration certificate, must be < ± 10 %.

In Process Controls (PC)
PC 1: Accuracy testing of emitting β-sources is performed prior to the introduction of a new measured geometry (e.g., new vial size of lutetium-177).
PC 2: A daily radioactivity check is performed before each production run and the deviation between the read and calculated activity, according to the calibration certificate, must be <±10%. Since this is a preparation that cannot be subjected to terminal sterilization, the product solution, therefore, has to be subjected to sterile filtration through a sterile filter (pores size less than 0.22 µm).
Filter integrity must be checked by bubble point test before the release. The bubble point test was performed on an Integritest 4 system (Merck Millipore, Merck KgaA, Darmstadt, Germany.
A 1 mL aliquot of the product mixture was sent to the Microbiological Laboratory of the Regional Healthcare, Pievesestina, Cesena, Italy for the sterility tests. The sterility test was performed according to current European Pharmacopoeia Monograph 2.6.1 "Sterility".

HPLC Analysis
nat Lu-PSMA-I&T was purchased from ABX GmbH-Advanced Biochemical Compounds (Radeberg, Germany) as a reference standard.
HPLC analysis was performed on an Ultimate 3000 system equipped by a UV variable wavelength detector RS300 (Thermo Fischer Scientific, Germany) and a radiometric detector (GABI, Raytest, Germany). The system was run by Chromeleon software version 7.2 SR5 (Dionex Sunnyvale, CA, USA).
A multi-step gradient was applied using two solvent A (0.1% TFA in water) and solvent B (0.1% TFA in acetonitrile): 92% A to 40% A in 10 min, then from 40% A to 20% A for a further 3 min, and back to 92% A in 2 min, then stable for 5 min. The flow rate was set at 0.6 mL/min. UV wavelength at 205 nm. Column Oven: 25 • C. Injection volume 20 µL. Retention times of nat -LuPSMA-I&T and PSMA-I&T were 6.730 and 7.013, respectively.
Chemical purity was calculated by comparing the areas of the peaks of the product with a standard solution injected before the analysis of the final product, as a general practice suggested by Pharmacopoeia.

Thin-Layer Chromatography (TLC)
TLC was performed using a TLC Silica Gel 60 (Merck KGaA, Darmstadt, Germany), Approximately 1-2 µL of IMP injection solution was spotted on the plate. The solvent for the development of the TLC plates was ammonium acetate 0.1N and methanol 50:50 v/v. The developed plate was analyzed by autoradiography on MS (MultiSensitive) storage phosphor screens and by a Cyclone Plus Storage Phosphor System (PerkinElmer).

Conclusions
This study demonstrates that [ 177 Lu]Lu-PSMA-I&T can be prepared as a radiopharmaceutical suitable for human use. Clearly defined acceptance criteria, validations plans, and methods for quality control were outlined. We compared two structural related radiopharmaceuticals, which differed in chelator and linker, but not in pharmacophore. The difference resulted in a different metal chelation kinetic, a different solubility, and, above all, a very different stability of the two radiopharmaceuticals. Ascorbic acid functions as a scavenger and can significantly prolong the shelf life of [ 177 Lu]Lu-PSMA-I&T by at least 30 h, and can prolong the time interval where patients can be treated.

Patents
PSMA-I&T is available by ABX and there are no patents.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.