PT - JOURNAL ARTICLE AU - Colin Miller AU - Irene Virgolini AU - Andreas Kjaer AU - Henning Gronbaek AU - Pierre Terve AU - Khokan Shamsi AU - Sandy McEwan AU - Shadfar Bahri TI - A novel read methodology to evaluate the optimal dose of <sup>68</sup>Ga-satoreotide trizoxetan as a PET imaging agent in patients with GEP-NETs in a phase II clinical trial DP - 2020 May 01 TA - Journal of Nuclear Medicine PG - 60--60 VI - 61 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/61/supplement_1/60.short 4100 - http://jnm.snmjournals.org/content/61/supplement_1/60.full SO - J Nucl Med2020 May 01; 61 AB - 60Introduction: 68Ga-satoreotide trizoxetan(also known as 68Ga-NODAGA-JR11, 68Ga-OPS202 or 68Ga-IPN01070) is a novel somatostatin receptor 2 (sstr2) antagonist which has shown to have improved diagnostic performance as a PET imaging agent versus the sstr2 agonist 68Ga-DOTATOC in a single-center Phase 1 study (NCT02162446). This international, open-label, factorial-design, phase II study (NCT03220217; EudraCT: 2016-004928-39) investigated two peptide mass dose ranges and three radioactivity ranges of 68Ga-satoreotide trizoxetan, with a novel read design to minimize the number of participants to acquire the necessary dosing information. The read design and quality score results are presented here. Methods: A total of 24 participants with well-differentiated, metastatic, sstr2-expressing gastroenteropancreatic neuroendocrine tumors (GEP-NETs), were randomized to one of three study arms (n=8 participants per arm). Participants received two different peptide mass/activity range combinations at two consecutive visits 2-3 weeks apart with PET/CT images obtained at 60±10 minutes post intravenous injection. Participants were allocated to 5-20 μg of peptide mass at visit-1 and 30-45 μg at visit-2, with one of three gallium-68 radioactivity ranges (40-80, 100-140 or 160-200 MBq) per visit. The study was approved by each site institutional review board (IRB), and every participant completed an informed consent. As per design, PET/CT images were all sent to an imaging core lab (Keosys, Nantes, France) for efficacy evaluation. A 2+1 read design, with two readers and a third to adjudicate at the organ level on number of lesions identified, was employed to read the PET only and PET/CT images, using contrast enhanced CT (ceCT) as the standard of truth (SoT) (Figure 1; Each reader, R 1 to 4, a discrete independent reader; Reader 4A and Reader 4B are two independent readers adjudicating differences between the two readers in each of the separate reads: Read 1A and 1B, Read 2A and 2B, Read 3A and 3B). A final comparison read was conducted with no adjudication, where the readers were shown both timepoints for each participant contemporaneously in a blinded randomized manner. The readers had to select which image provided “optimal quality”. Each image was assigned a “1” or a “0”. For each pair, one or both images could score a “1”. The scores were summed for each peptide mass dose range and radioactivity range. An interim analysis was performed including 15 participants (safety population) of whom 14 in per protocol. Results: The results of the quality score are shown in Table 1. These results trend in the same manner as the other endpoints. Of note, there was only one discrepant reading where each reader differed in the quality read: One reader preferred one image because of more lesions, the other reader preferred the alternative image because of superior signal-to-noise ratio. Conclusion: A Likert score (1 to 5 scoring system) is often used to compare images and doses during the development of novel diagnostics. The score itself is highly subjective between readers and the conclusions can be difficult to interpret. We have developed a novel read-design which removed the subjectivity of applying a score to an image and making direct comparisons. This information, along with the other results, provided the rationale as to the optimal dose range of 68Ga-satoreotide trizoxetan for future studies.