PARP-1 Targeting With the Novel Radiotracer [18F]FTT in Pancreatic Neuroendocrine Tumors

Pancreatic neuroendocrine tumors are a rare disease entity for which there are relatively limited treatment options. Due to the nature of this malignancy, preclinical models have also been challenging to develop. It is known that the majority of these tumors harbor a mutation in MEN1, DAXX or ATRX, all of which are thought to be associated with abnormal DNA repair mechanisms, including homologous recombination. Defects in homologous recombination often render a tumor susceptible to treatment with a PARP inhibitor but the activity of PARP inhibition in pancreatic neuroendocrine tumors has not been well explored. Investigators at Penn have developed 1-(4-(2-Fluoroethoxy)phenyl)-8,9-dihiydro-2,7,9a-triazabenzo[cd]azulen-6(7H)-one, also known as [18F]FluorThanatrace or [18F]FTT, which is a radiolabeled, selective PARP-1 inhibitor that can be visualized by PET imaging. As such, the rationale for this trial is to evaluate [18F]FTT in pancreatic neuroendocrine tumors to determine their level of uptake and to perform accompanying molecular studies to evaluate PARP-1 expression in associated tissue specimens as well as to correlate PET findings with underlying mutational status.

1-(4-(2-Fluoroethoxy)phenyl)-8,9-dihiydro-2,7,9a-triazabenzo[cd]azulen-6(7H)-one, also known as [18F]FluorThanatrace or [18F]FTT is a positron emitting radiopharmaceutical that has been studied in animals for selective measurement of the in vivo inhibition of the PARP-1 nuclear enzyme with positron emission tomography/computed tomography (PET/CT). Zhou et al. described synthesis of a series of radiolabeled benzimidazole carboxamide analogs that could be easily labelled with 18F and their inhibition potency against PARP-1 was determined1. Newly synthesized PARP-1 inhibitors were assessed for their ability to inhibit active PARP-1 using the method described by Putt and Hergenrother2 (Table 1). The results showed that tricycle benzamide analogs had higher inhibition potency than their respective benzimidazole analogs. The analogs with a fluoroethoxy substituent had three times higher inhibition potencies than the respective analogs having a fluoroethyl triazole group. From this, the most potent inhibitor, 12, was selected for 18F-labeling.

Pancreatic neuroendocrine tumors (PNETs) account for approximately 10% of all pancreatic neoplasms and their incidence is rapidly rising3. Patients with these tumors are often treated with temozolomide (TMZ) and capecitabine (CAP) chemotherapy, the former being a DNA damaging agent that generates DNA adducts requiring repair via the direct and base excision repair mechanisms4. DNA repair abnormalities have been identified as one of the key cancer mechanisms implicated in PNETs5. Mutations in MEN1 and either DAXX or ATRX are seen in 44% and 43% of PNETs respectively, and preclinical data demonstrate that these mutations are associated with yet another DNA repair mechanism, homologous recombination (HR)6-8. With multiple DNA repair mechanisms being involved, simultaneous targeting of these mechanisms may be an opportunity to enhance the treatment of PNETs. Specifically, adding a PARP inhibitor to target the HR mechanism may improve treatments via a synthetic lethality approach.

Tumors with defects in HR are known to be sensitive to PARP inhibition and use of these agents has become standard for the treatment of some breast, ovarian and pancreatic cancers9-13. In an in vitro model utilizing BON-1 cells (a PNET cell line), administration of a PARP inhibitor in combination with TMZ results in enhanced suppression of cell growth as compared to TMZ alone. In a model of BALB/c mice with liver BON-1-Luc tumors, administration of TMZ resulted in a decrease in bioluminescence as well as a decrease in tumor volume and this effect was further enhanced when a PARP inhibitor was co-administered with TMZ14. These data suggest a role for PARP inhibition in PNET, but the mechanism for this is poorly understood.

Several preclinical studies have shown menin, the protein product of the MEN1 gene, to be an important regulator of HR8 and this has been explored using the QGP1 PNET cell line. In an MEN-1 knockout model generated from these cells, higher levels of DNA double-strand breaks are observed via Comet assay as compared to wildtype. MEN1 loss is also associated with a decrease in BRCA2 as QGP1 cells treated with siMEN1 (thereby suppressing MEN1) results in significantly reduced BRCA2 mRNA expression and BRCA2 protein levels. When MEN1 knockout cells were treated with the PARP inhibitor talazoparib, a significant decrease in cell growth was observed as compared to untreated cells. Similarly, when QGP1 cells were treated with siMEN1 and exposed to talazoparib, there was a significant decrease in cell growth compared to PARP untreated cells15. The presence of a DAXX or ATRX mutation presents another mechanism by which HR may be implicated. These mutations are associated with the alternative lengthening of telomeres (ALT) phenotype, a mechanism for telomere maintenance that is utilized in many PNETs that is believed to be a HR dependent process5. Overall, these data suggest a role for HR in PNETs and that targeted treatment with PARP inhibition may be a promising approach.