In vitro characterization of immune modulating drug-eluting immunobeads towards transarterial embolization in cancer

Ayele H. Negussie, Andrew S. Mikhail, Joshua Owen, Natalie Hong, Camella J. Carlson, Yiqing Tang, Kendal P. Carrow, Michal Mauda‐Havakuk, Andrew L. Lewis, John W. Karanian, William F. Pritchard, Bradford J. WoodScientific 2022 · DOI

Abstract

Hepatocellular carcinoma (HCC) is an aggressive liver cancer with limited effective treatment options. In this study, we selected TLR agonists imiquimod (IMQ), gardiquimod (GARD), GS-9620 and DSR 6434, and a small molecule checkpoint inhibitor, BMS-202, for characterization of drug loading and release from radiopaque embolic beads (DC Bead LUMI) for potential use in image-guided transarterial embolization (TACE) of HCC. The maximum drug loading capacity and amount of drug released over time were determined by high performance liquid chromatography and compared with the commonly used anthracycline, doxorubicin hydrochloride (Dox). Maximum drug loading was 204.54 ± 3.87, 65.28 ± 3.09, 65.95 ± 6.96, 65.97 ± 1.54, and 148.05 ± 2.24 mg of drug per milliliter of DC Bead LUMI for Dox, GARD, DSR 6434, IMQ, and BMS-202, respectively. Fast loading and subsequent rapid release in saline were observed for IMQ, GARD, and DSR 6434. Compared to TLR agonists, slower loading and release were observed for Dox and BMS-202. The maximum loading capacity of BMS-202 and Dox in DC Bead LUMI exceeded the maximum theoretical loading capacity of the beads expected from ionic interaction alone suggesting additional drug-bead or drug-drug interactions may play a role.

Introduction

Hepatocellular carcinoma (HCC) remains a significant clinical challenge as the most prevalent primary liver cancer. A cornerstone of treatment for intermediate-stage HCC is Transarterial Chemoembolization (TACE). This procedure utilizes embolic beads to occlude the tumor's blood supply while providing a localized, high-dose delivery of chemotherapeutic agents like doxorubicin.

As the therapeutic landscape shifts toward immunotherapy, there is growing interest in delivering immune-modulating agents—such as Toll-like receptor (TLR) agonists and checkpoint inhibitors—locally to minimize systemic toxicity and enhance anti-tumor immune responses. This study investigates the feasibility of creating "immunobeads" by loading these potent modulators onto radiopaque DC Bead LUMI microspheres, potentially enabling a new "immuno-TACE" paradigm.

Methods and Materials

The study evaluated five immune-modulating candidates: four TLR agonists (IMQ, GARD, GS-9620, DSR 6434) and one small-molecule checkpoint inhibitor (BMS-202). Doxorubicin (Dox) served as the control. The primary delivery platform was DC Bead LUMI, a radiopaque hydrogel microsphere containing iodine for visualization and negatively charged sulfonate groups for drug sequestration.

Characterization involved:

  • Elemental Analysis to quantify sulfonate binding sites.
  • pKa Measurements to determine the ionization state of the drugs.
  • High-Performance Liquid Chromatography (HPLC) to measure loading and elution kinetics in saline.
  • Microscopy to assess bead morphology and drug aggregation.

Results

Drug Loading Capacity

The loading experiments revealed significant differences between the drug classes. Doxorubicin and BMS-202 achieved the highest loading levels, reaching 204.54 mg/mL and 148.05 mg/mL, respectively. Notably, these two drugs exceeded 100% of the theoretical ionic capacity of the beads, suggesting that non-ionic interactions, such as hydrophobic forces and drug-drug stacking, contribute to their high affinity for the bead matrix.

Relative percentage of drug loaded into DC Bead LUMI for various immune modulators compared to doxorubicin.

In contrast, the TLR agonists (IMQ, GARD, and DSR 6434) showed significantly lower loading capacities, plateauing around 60-70% of the relative drug load compared to the theoretical maximum.

Loading and Release Kinetics

The kinetics of drug uptake also varied. While IMQ, GARD, and DSR 6434 loaded rapidly, they also exhibited a "burst release" profile in saline, indicating weak binding. BMS-202 and Doxorubicin demonstrated a more sustained loading profile and a slower, controlled release, which is preferable for maintaining therapeutic concentrations within the tumor environment over an extended period.

Loading kinetics of immune-modulating drugs into DC Bead LUMI over time (A: long-term, B: short-term).

Discussion

The high loading capacity of BMS-202 and Doxorubicin highlights the importance of multi-modal interactions within the DC Bead LUMI matrix. The presence of iodine in these beads increases hydrophobicity and binding site density but also introduces steric hindrance, which can slow the diffusion of larger molecules. The failure of most TLR agonists to bind stably suggests that their interaction with sulfonate groups is insufficient for reliable delivery via this platform. Furthermore, the decomposition of GS-9620 during preparation underscores the chemical stability challenges inherent in drug-device combinations.

Conclusion

This in vitro characterization identifies BMS-202 as a promising candidate for immuno-TACE, showing loading and release profiles comparable to the gold-standard doxorubicin. While the TLR agonists tested showed less favorable kinetics, they provide a baseline for future chemical modifications to improve bead affinity. Future research should transition to in vivo models to evaluate the therapeutic efficacy and safety of BMS-202-loaded immunobeads in treating HCC.

Summary updated

Abstract

Hepatocellular carcinoma (HCC) is an aggressive liver cancer with limited effective treatment options. In this study, we selected TLR agonists imiquimod (IMQ), gardiquimod (GARD), GS-9620 and DSR 6434, and a small molecule checkpoint inhibitor, BMS-202, for characterization of drug loading and release from radiopaque embolic beads (DC Bead LUMI) for potential use in image-guided transarterial embolization (TACE) of HCC. The maximum drug loading capacity and amount of drug released over time were determined by high performance liquid chromatography and compared with the commonly used anthracycline, doxorubicin hydrochloride (Dox). Maximum drug loading was 204.54 ± 3.87, 65.28 ± 3.09, 65.95 ± 6.96, 65.97 ± 1.54, and 148.05 ± 2.24 mg of drug per milliliter of DC Bead LUMI for Dox, GARD, DSR 6434, IMQ, and BMS-202, respectively. Fast loading and subsequent rapid release in saline were observed for IMQ, GARD, and DSR 6434. These drugs could also be partially removed from the beads by repeated washing with de-ionized water suggesting weak interaction with the beads. Aggregation of IMQ was observed in water and saline. GS-9620 partially decomposed in the solubilizing solution, so loading and release were not characterized. Compared to TLR agonists, slower loading and release were observed for Dox and BMS-202. Potential factors influencing drug loading into and release from DC Bead LUMI including steric hinderance, hydrophobicity, drug pKa, and the electrostatic nature of the beads are discussed. The maximum loading capacity of BMS-202 and Dox in DC Bead LUMI exceeded the maximum theoretical loading capacity of the beads expected from ionic interaction alone suggesting additional drug-bead or drug-drug interactions may play a role. Slightly more release was observed for BMS-202 at early time points followed by a slower release compared to Dox. Further study of these drug-bead combinations is warranted in search of new tools for locoregional delivery of immune-modulating agents for treatment of HCC via drug-eluting bead chemoembolization.
BibTeX
@inproceedings{negussie2022vitro,
  author = {Ayele H. Negussie and Andrew S. Mikhail and Joshua Owen and Natalie Hong and Camella J. Carlson and Yiqing Tang and Kendal P. Carrow and Michal Mauda‐Havakuk and Andrew L. Lewis and John W. Karanian and William F. Pritchard and Bradford J. Wood},
  title = {{In vitro characterization of immune modulating drug-eluting immunobeads towards transarterial embolization in cancer}},
  booktitle = {Scientific},
  year = {2022},
  doi = {10.1038/s41598-022-26094-1},
}
Figure