(A-24) Modeling of Morphine and RTX in a Porous Media for Intrathecal Drug Delivery

Introduction:: Severe pain management remains a challenging and arduous issue for patients battling advanced cancer. Current solutions are high-dose opioids that act on the central nervous system (CNS) and alter the processing of pain signals. However, many side effects, including excessive sedation, cognitive impairment, and tolerance, reduce the efficacy of these treatments. Therefore, there is a strong medical need for a non-opioid interventional pain treatment. Resiniferatoxin (RTX), a novel agent and potent capsaicin analog, activates the nociceptive ion channel- transient receptor potential vanilloid 1 (TRPV1). When given as a single intrathecal injection, it fragments nociceptive axons, which provides long-term pain relief [1]. We partnered with the NIH in an ongoing study of porcine specimens to develop a model illustrating the behavior of RTX intrathecal drug delivery [2]. The goal of this study is to predict the distribution of RTX into the intrathecal subarachnoid space of the cauda equina after injection. In this initial stage of the model, our goal is to use finite element analysis to determine whether one-dimensional dispersion is sufficient to describe the spatial spread of RTX after the intrathecal injections.

Materials and Methods::

To represent the cauda equina in the spinal canal, in COMSOL Multiphysics® we adopt a simple cylindrical model (Figure 1), which has been utilized by other drug delivery models for similar analytical studies [4,5]. To ensure the accuracy of our model, we compare our computational data with existing experimental data from an intrathecal morphine infusion study in farm-bred pigs [6].
It is common in biological models to use porous media for tissues [7]. Using the diffusivity of morphine in water (D_w), we design three different scenarios with an effective diffusivity multiplier, α, to compare with our analytical solutions. First, we model the behavior of morphine in water only such that α₁=1. Second, we acknowledge that the cauda equina has a porosity hindering upward diffusion, implying α₂< 1. We varied α to verify our results. A detailed list of parameters used is denoted in Table 1.

Results, Conclusions, and Discussions::

After plotting our results in MATLAB (figure 2), we found that the resulting concentrations from our analytical computations for the last timepoint validated the concentration output in COMSOL for a given distance, z, up the spinal model. We also noted that decreasing α hindered diffusion in both methods, meaning when accounting for porosity only, morphine’s distribution distance decreased. This is justified because porosity decreases the available space in the column for the drug to travel.

Finally, we note that porosity is not the only mechanism directing dispersion. Mechanical mixing heavily influences dispersion of drugs in the spine. To account for this, we will manipulate D_eff to match the interpolated porcine data from the study, predicting that α₃>1.

Once we have justified our model with the morphine experimental data, the goal is to use it with our drug of interest, RTX. When we receive imaging data, we plan to segment a geometry that is more representative of the spinal anatomy. After the injection experiments are conducted, we will then use the data to find the experimental dispersion coefficient of RTX to use with our model.

Acknowledgements (Optional): :

References (Optional): : [1] M. J. Iadarola and A. J. Mannes, “The vanilloid agonist resiniferatoxin for interventional-based Pain Control,” Current Topics in Medicinal Chemistry, vol. 11, no. 17, pp. 2171–2179, 2011. doi:10.2174/156802611796904942
[2] M. R. Sapio et al., “Pain control through selective chemo-axotomy of centrally projecting TRPV1+ sensory neurons,” Journal of Clinical Investigation, vol. 128, no. 4, pp. 1657–1670, 2018. doi:10.1172/jci94331
[3] COMSOL Multiphysics® v. 6.1. www.comsol.com. COMSOL AB, Stockholm, Sweden.
[4] [1] Y. Gong, Y. Ji, and Z. Qi, “ BENG 221 Modeling of Intrathecal Drug Delivery ,” thesis, University of California San Diego, 2016
[5] C. Gutiérrez-Montes et al., “Modelling and direct numerical simulation of flow and solute dispersion in the spinal subarachnoid space,” Applied Mathematical Modelling, vol. 94, pp. 516–533, 2021. doi:10.1016/j.apm.2021.01.037
[6] S. H. Flack, C. M. Anderson, and C. Bernards, “Morphine distribution in the spinal cord after chronic infusion in Pigs,” Anesthesia &amp; Analgesia, vol. 112, no. 2, pp. 460–464, 2011. doi:10.1213/ane.0b013e318203b7c0
[7] J. A. Rey, “Image-Based Computational Models of Tumor Mechanics and Perivascular Mass Transport in the Rat Brain,” University of Florida, 2022.