(T1430-02-07) Exploring the Potential Influence of Physicochemical Characteristics on Biodistribution of Iron-Carbohydrate Nanoparticles Using a Chicken Embryo Model

Purpose: Iron-carbohydrate nanomedicines complex drugs that are widely used to treat iron deficiency globally. Despite many decades of clinical use, the structure-function relationship that influences in vivo biodistribution is not fully established. The purpose of this study was to systematically evaluate the in vivo biodistribution profiles of different ferric carboxymaltose analogs with alterations in physicochemical characteristics. The goal of the project is to better understand the relationship of Critical Quality Attributes (CQAs) relevant for iron carbohydrate nanoparticle biodistribution.

Methods: Three different ferric carboxymaltose analogs were synthesized in house compared to the reference listed drug. Extensive physicochemical characterization was performed for each analog including dynamic light scattering, zeta potential and molecular weight (Figure 1). Biodistribution was evaluated in an established chicken embryo model and compared to vehicle (normal saline) injection. Fertilized White Leghorn eggs were incubated at 37.5°C with 50% relative humidity for 14 days. On embryonic development day 14 the eggs were cannulated and blood vessels are localized through the shell. A small hole through the eggshell is made and the experimental compound is directly injected intravenously (I.V.) with 30G syringes. For each product a 400 µg dose was administered and eggs within each group were sacrificed at various time points and the following organs were harvested: liver, spleen, heart, kidneys, brain and lung. Iron concentrations in the organs were measured by ICP-MS (mg iron/g tissue) and area under the concentration-time curve (AUC) was calculated for each organ using the trapezoidal rule. Regression models were employed to assess iron concentration over time using generalized least squares, with variance modeled as a power of the concentration. Data points that deviated from the mean by more than 3 standard deviations were identified as outliers and subsequently excluded from the analysis. The Confidence Intervals were computed based on 10,000 parametric bootstrap replications. For each timepoint, the fold change was computed, and significance was assessed using p-values derived from bootstrap replications. Statistical significance was inferred through a comparison of fold change in AUC values.

Results: For all studied analogs the highest increase in concentration of iron was observed in the liver and spleen followed by the kidney (Figure 2) with negligible increase of iron concentrations in lung, brain and heart. Analog 1 and Analog 3 showed biodistribution patterns distinct from the reference listed drug. AUC for analog 1 was significantly higher in spleen (p < 0.01) and kidney (p < 0.001), whereas for analog 3 AUC was significantly lower in liver (p < 0.001) but higher in spleen (p < 0.001).

Conclusion: The ferric carboxymaltose analogs synthesized demonstrated differences in deposition of iron in the liver and spleen in an established chicken egg embryo model indicating differences in the nano-bio interface. These data indicate that structural changes for ferric carboxymaltose final products may impart different biodistribution profiles and should be further studied in relevant animal models to better elucidate the structure-function relationship of this class of complex drugs. More sophisticated, orthogonal characterization methods may be required to ensure manufacturing quality for complex drugs.

Acknowledgements: AA, BF and RD are employees of CSL Vifor

Physicochemical Characteristics

Iron biodistribution in the liver, spleen and kidney