This pilot study evaluated the imaging performance of pretargeted immunological positron emission tomography (immuno-PET) using an anti-carcinoembryonic antigen (CEA) recombinant bispecific monoclonal antibody (BsMAb), TF2 and the [68Ga]Ga-labelled HSG peptide, IMP288, in patients with metastatic colorectal carcinoma (CRC). Patients requiring diagnostic workup of CRC metastases or in case of elevated CEA for surveillance were prospectively studied. They had to present with elevated CEA serum titre or positive CEA tumour staining by immunohistochemistry of a previous biopsy or surgical specimen. All patients underwent endoscopic ultrasound (EUS), chest-abdominal-pelvic computed tomography (CT), abdominal magnetic resonance imaging (MRI) and positron emission tomography using [18F]fluorodeoxyglucose (FDG-PET). For immuno-PET, patients received intravenously 120 nmol of TF2 followed 30 h later by 150 MBq of [68Ga]Ga-labelled IMP288, both I.V. The gold standard was histology and imaging after 6-month follow-up. Eleven patients were included. No adverse effects were reported after BsMAb and peptide injections. In a per-patient analysis, immuno-PET was positive in 9/11 patients. On a per-lesion analysis, 12 of 14 lesions were positive with immuno-PET. Median SUVmax, MTV and TLG were 7.65 [3.98–13.94, SD 3.37], 8.63 cm3 [1.98–46.64; SD 14.83] and 37.90 cm3 [8.07–127.5; SD 43.47] respectively for immuno-PET lesions. Based on a per-lesion analysis, the sensitivity, specificity, positive-predictive value and negative-predictive value were, respectively, 82%, 25%, 82% and 25% for the combination of EUS/CT/MRI; 76%, 67%, 87% and 33% for FDG-PET; and 88%, 100%, 100% and 67% for immuno-PET. Immuno-PET had an impact on management in 2 patients. This pilot study showed that pretargeted immuno-PET using anti-CEA/anti-IMP288 BsMAb and a [68Ga]Ga-labelled hapten was safe and feasible, with promising diagnostic performance. ClinicalTrials.gov NCT02587247 Registered 27 October 2015
Despite the weak immunogenicity of gangliosides, a limited number of highly specific murine monoclonal antibodies (MAbs) were elicited. This study investigated the reactivity and the structure of the VH and V kappa genes of nine hybridomas obtained from independent fusions producing antibodies against disialogangliosides GD2 and GD3 and their O-acetylated derivatives. These antibodies depended on four types of V kappa genes. They were also encoded by VH genes of the J558 family (5 out of 9) and occasionally by VH genes of the S107, 7183, and 3609 families, rearranged with a variety of DH and JH genes. The 8B6 and 7H2 MAbs specific for GD2-O-acetylated, respectively, used the VH gene of the S107 and 7183 families. The length of H chain CDR3 ranged from 8 to 11 amino acids. A set of S107 and 3609 germline genes closed from A/J murine fetal liver and matched with the VH segment of hybridomas 8B6 and 10B8 revealed somatic mutations. Although the relative number of sequences does not preclude any formal conclusions regarding the preferential use of V genes in the immune recognition of carbohydrate structures, our results clearly indicate that MAbs directed to very similar structures as GD2 and GD3 were encoded by different VH and V kappa genes.