Injectable "Liquid Brachytherapy" for Treating Pancreatic Cancer in a Porcine Model

Introduction: Pancreatic cancer is a leading cause of cancer deaths in the United States, with a 5-year survival rate of only 11%. The development of effective therapies for pancreatic cancer is hence an area of huge unmet need. Motivated by this, we have developed a local radiotherapy that destroys tumors from within using elastin-like polypeptide (ELP) technology. ELPs are recombinant biopolymers that exhibit lower critical solution temperature phase behavior. Above a sequence-defined transition temperature, an ELP will undergo liquid-liquid phase separation and transition into a viscous, gel-like coacervate. Our approach to treating pancreatic cancer uses an ELP that is covalently conjugated to the therapeutic radionuclide iodine-131 (131I). When injected intratumorally, it immediately transitions into a depot upon heating to body temperature, undergoes radiation-induced crosslinking, and retains nearly all injected radioactivity within the tumor. With numerous in vivo studies demonstrating efficacy in mice, we now report the first successful administration of this class of biopolymer within a porcine model that recapitulates the delivery challenges posed in human pancreatic cancer. Using interventional radiology, we performed a CT image-guided injection of ELP brachytherapy labeled with iodine-124 (124I)—a positron-emitting radionuclide—to visualize deposition of the ELP depot within pancreatic space and confirm retention of the depot, and nearly all radioactivity, over the following six days. Our study demonstrates the translatability and promise of our localized “liquid brachytherapy” platform, laying the groundwork for future use in the clinic.

Materials and

Methods: As concentration increases, the phase transition temperature of an ELP will decrease, and vice versa. To enable injection of ELP into the pancreas of a large animal, it is essential that the ELP does not coacervate inside the long needle and clog it. We first injected ELP at various concentrations through a catheter in a 37C incubator to simulate body temperature. Once we identified ELP concentrations that could be injected without clogging the catheter ex vivo, we radiolabeled them with 131I and injected 1000, 400, and 300 µM ELP into BxPC3-luc2 tumors in mice. We confirmed deposition of activity by measuring whole-body radioactivity daily and after 1 week, we excised tumors and thyroids and compared radioactivity in each with the whole-body value. We then performed our porcine experiment using the lower concentration, optimized ELP formulation.

A 40 kg male Yorkshire pig was positioned prone on a CT scanner. From a scout CT scan, a lateral route from the skin into the splenic lobe of the pancreas was planned. A 22g Chiba needle was advanced into the target portion of the pancreas under intermittent CT guidance. Intervening bowel, liver, and spleen was displaced using hydrodissection. When the treatment needle was advanced into the pancreas, 1 cc of ELP containing 300 μCi of 124I activity was injected. The needles were then removed, and high-resolution PET/CT was performed to assess for successful 124I-ELP deposition at t=0, 3, and 144 hrs (six days) following administration.

Results, Conclusions, and Discussions: In our previous mouse studies, where the conjugate was directly injected though a needle into tumors, we chose an ELP concentration of 1000 µM. When testing this concentration ex vivo, the ELP immediately clogged the catheter and needle upon heating (Table 1). As concentration was decreased, we observed that concentrations below 400 µM could be injected with no perceptible resistance, leading us to test formulations at 1000, 400, and 300 µM in our in vivo mouse study. Compared to the control, both new formulations exhibited near-identical depot-forming behavior as well as retention of injected activity for one week following injection (Figure 1A). At that time, we harvested the tumor and thyroid from the mice and found no significant differences in measured activities in any of the formulations (Figure 1B). Based on this, we chose to inject an ELP concentration of 300 µM in our pig experiment.

Percutaneous injection (Figure 2) of the 124I-labeled ELP into the pig pancreas was successful, with the PET/CT scan at t=0 hrs demonstrating deposition of the injected ELP within the local site (Figure 3). At t=3 hrs, a follow-up scan confirmed retention of the ELP depot at this site. Six days following this, we performed a final scan and observed the 124I-labeled ELP depot in the same position within the pancreas tissue, demonstrating successful depot formation and retention. We observed no other detectable activity under PET, besides a small amount of 124I activity within the thyroid of the animal (Figure 4), an expected finding based on previous experiments in mice.

The results of this experiment represent a first-in-pig approach to injecting biopolymer brachytherapy. To our knowledge, this is the first example of an attempt to deliver injectable, persistent radiotherapy within an internal organ that cannot be reached with conventional rod- or seed-brachytherapy. Future work includes treating pancreatic tumors in a transgenic Oncopig model using therapeutic 131I-ELP brachytherapy, with eventual translation towards clinical studies.

Acknowledgements (Optional): The researchers thank the North Carolina Biotechnology Center (NCBC) as well as the Cinelli Family Foundation for their support of these studies.