Actively controlled local drug delivery using conductive polymer-based devices

The systemic administration of pharmaceutical or bioactive agents has become the gold standard in clinical treatment for a vast range of diseases and disorders. It has also been widely accepted that these treatments are often associated with undesirable side effects. One significant cause of unwanted effects is the need for high systemic doses to achieve a therapeutic benefit. While system-wide delivery of a therapeutic is necessary for some conditions, such as sepsis, in many conditions, such as a localized tumor, the drug travels throughout the whole circulatory system in order to reach one target area. Since this target tissue must receive therapeutic levels of the drug for it to be effective, substantial amounts of healthy tissue are also exposed to the same levels of the drug. The mechanism of action for treatment of many drugs inherently facilitates off-target effects in other exposed tissues and organs. A well-known example is systemic chemotherapy, where the therapeutic drugs are designed to prevent cells from dividing and growing, inhibiting tumor progression, and ultimately killing the cancer cells. However, in healthy tissues, these drugs can cause similar effects, resulting in undesirable side effects ranging from hair loss to suppression of the immune system.^1,2

Enhancing therapeutic efficacy through the elimination of off-target effects underpins the motivation for implanted drug delivery platforms.^3–5 These devices have been used to target conditions where the negative effects of systemic drug administration far outweigh the invasiveness of an implant. The placement of these devices enables highly localized administration of drugs directly to the target site, enhancing drug bioavailability.⁶ The sustained release of drug enabled by these systems also promotes a more stable local concentration profile when compared to bolus delivery via systemic administration. The tissue exposure to continuous levels of therapeutic agents can improve the efficacy of treatment^3,4 and reduce costs to public health systems.⁵ Implantable platforms have seen clinical translation in the form of passively releasing carriers such as heparin eluting vascular stents,⁷ anesthetic releasing polymers,⁸ and hormone releasing implants,⁹ in which a drug is carried and released by passive diffusion. These platforms have all demonstrated clinical benefits through the local release of drug; however, since they rely on passive processes, such as diffusion or carrier degradation, there is limited control over the dosing and release profile. In the absence of a method for reloading the drug carrier, these devices primarily present with a singular bolus (or burst) release of the drug, yielding an inherently limited effective lifetime. As such, the therapeutic effects of passive drug delivery systems are often not sustained for the required treatment time.

A significant step toward broadening the clinical impact of implanted drug delivery platforms lies in enabling both temporally and spatially controlled drug releases through active devices. The rapidly developing field of organic bioelectronics has facilitated the development of a range of platforms capable of controlled release.¹⁰ These devices primarily rely on conductive polymers (CPs) such as polypyrrole (PPy), poly(3,4-ethylenedioxythiophene) (PEDOT), or polyaniline (PANI) to facilitate drug delivery. CPs can be used to move therapeutics into target tissues via charge mediated control of drug mobility, actuation of polymer carriers to facilitate volumetric changes, leading to drug elution, or development of ionic concentration gradients.¹¹ These electroactive polymers provide a robust technology platform by combining controllable electrical performance with flexible processing approaches typical of polymer systems. This has led to a wide range of CP-based platforms being developed for the active control of drug release. CPs have the ability to bind and release charged molecules to their backbone through electrostatic interactions, which can be controlled via the reduction-oxidation (redox) state of the CP. This mechanism imparts the ability to actively control drug release upon electrical or electrochemical stimuli.^12,13 Molecules can be incorporated within CP materials via molecular entrapment and subsequently released using redox-controlled volumetric actuation. The platforms in which CPs have been used fall broadly into three categories: simple CP only structures,^14–21 CP composites,^22–25 and CP mediated ion pumps.^26,27

CP only structures are the most straightforward approach for drug delivery from a CP using a solid-state film and subsequent electrochemical potentials to elute the drug in a controlled manner. Typically, the drug is loaded during polymerization and becomes incorporated either as a dopant, if the drug is appropriately charged, or coincorporated by entrapment as the CP precipitates and nucleates on the electrode surface, as illustrated in Fig. 1(a). The controlled release of therapeutic agents from PPy¹⁴ and PEDOT^15–17,28 materials has been demonstrated through the use of applied electric potential. The applications of these platforms range from the release of anti-inflammatory drug dexamethasone (Dex)^15,16 to a chemotherapeutic treatment with the release of oleanolic acid.¹⁸ The most commonly explored drug release platform in this field is the release of Dex, as it is available as a charged molecule, enabling direct incorporation into the CP as a dopant anion. The quantity of drug released from PEDOT/Dex based platforms has been shown to increase with the film thickness, while drug release efficiency has been demonstrated to vary across different stimulation paradigms (i.e., cyclic voltammetry, DC potential, or biphasic current pulsing).¹⁶ One challenge in the implementation of these platforms is the potential delamination of the CP during drug release.^11,12 However, several groups have developed approaches that minimize or eliminate CP delamination through chemical coupling or surface structuring.^29,30 Critically, simple CP films or coatings have a major limitation in their drug carrying capacity.^28,31 That is, the volume of the films does not allow for incorporation and release of therapeutic quantities of many of the target drugs that have been explored. Attempts to overcome this limitation have led to nano- and microscale structuring of CPs to improve both the drug carrying capacity and access to the drug within the CP matrix.

Micro- and nanostructured CPs enable improved access to and transfer of mobile drugs due to an increased accessible interfacial surface area [Fig. 1(b)].³² Structuring is often achieved by polymerizing the CP around a degradable template, such as an electrospun fiber or microbead. One promising example demonstrated electropolymerization of PEDOT on Dex loaded poly(lactic-coglycolic acid) (PLGA) fibers. The controlled release of Dex was achieved with voltage application to the PEDOT nanotubes after dissolution of PLGA.¹⁹ An example of a microfiber approach was presented by Esrafilzadeh et al.²⁰ using layered CPs with inner fibers fabricated from wet-spun PEDOT:poly(styrene sulfonate) and an outer layer of electropolymerised PPy loaded with the drug ciprofloxacin, incorporated as a dopant.²⁰ Controlled drug release was achieved electrochemically, with confirmed bioactivity in vitro. Nanoporous PPy films have been achieved using polystyrene (PS) nanobeads as a template for electrodeposition.²¹ PPy was loaded with the model drug fluorescein and incorporated as a dopant. After degradation of the PS beads, it was demonstrated that these nanoporous films displayed significantly increased release efficiency compared to nonstructured, planar PPy films. The increased surface area of these structures can impart greater amounts of passive drug release, and Massoumi et al.³³ have reported addressing this through bilayering to restrict uncontrolled drug release. Ultimately, as these platforms rely predominantly on the properties of CPs, their mechanical properties, being relatively brittle and friable, are not well suited to chronic drug delivery. Although the safety of these systems has been well characterized,¹² the potential for material breakdown with drug release remains a barrier for further translation.

CP composites, including conductive hydrogels (CHs),³⁴ have also been explored as drug delivery platforms that can facilitate higher drug loading capacity through the hydrogel matrix [Fig. 1(c)].^22,35 In these systems, the hydrogel component is formed first, and the CP is subsequently polymerized within the hydrogel volume. The drug is usually incorporated during the formation of the hydrogel, but it is possible to incorporate the drug during CP polymerization. Delivery of the drug can be either electrochemical, if there is an ionic interaction with the CP component, or through actuation of the composite. CP systems have been used extensively as actuators because cycling between reduction and oxidation states is known to result in volumetric changes. This is inherent to the charge transfer mechanism that imparts conductivity to a CP. As the potential applied to the CP is switched, the movement of doping anions out of and back into the film causes changes in chain arrangements, hydrophilicity, and consequently material volume. When applied to a volume restricted secondary polymer such as a hydrogel, the CP constriction results in expulsion of mobile molecules. This mechanism has enabled control of drug delivery for CP composites. A CH consisting of poly(dimethylacrylamide-co-4-methacryloyloxy benzophenone-co-4-styrenesulfonate (PDMAAp) in which PEDOT was grown showed on-demand release of both the model drug fluorescein and the anti-inflammatory Dex. Accurate control of drug timing and dosing was achieved by applying different types of electrical trigger signals and intensities.²² Additionally, CHs made of chitosan-graft-PANI copolymer/dextran (OD) loaded with amoxicillin and ibuprofen have displayed a repeatable on-off pulse drug release via electrical stimulation.²³ Other CHs based on PANI/polyacrylamide²⁴ and PANI/polyvinyl alcohol²⁵ have also been used to demonstrate tailored drug release profiles by fine tuning electrical stimulation and hydrogel properties such as cross-linking density and ratio of hydrogel to CP in the composite. In vivo local release of drugs has been achieved upon voltage application of an injectable CH containing PPy nanoparticles loaded with chemotherapeutic agent daunorubicin.³⁶ The increased drug loading capacity of CHs addresses a major limitation of solid-state CP-based platforms. However, this increased capacity comes at the cost of increased difficulty in preventing passive and burst release. The inherent water permeability of a hydrogel facilitates strong diffusion gradients, and placement within tissues results in an initial loss of drug that is difficult to control. This can lead to undesirable release profiles of the delivery platform and ultimately a reduction in the localization of the drug delivery.

Organic electronic ion pumps (OEIPs) are CP-based devices that use CPs as an ion selective membrane for drug transfer [Fig. 1(d)].^26,27 This technology utilizes electrophoretic transport of small charged drugs through a CP film. A common configuration uses the chemically polymerized CP, PEDOT:PSS, as a thin film channel from a source reservoir into a target electrolyte. There are a range of alternate configurations that also use ion selective membranes coupled to CPs and overoxidised CPs to control delivery of drug molecules with varying size and charge properties. Ultimately, studies have shown that this delivery method allows for precise spatiotemporal controlled drug release. OEIPs have been utilized to successfully deliver neurotransmitters such as glutamate (Glu), aspartate (Asp), and γ-amino butyric acid (GABA), with a release rate controlled by an applied voltage and capabilities for on-off switching. Controlled release of Glu was confirmed both in vitro by intracellular calcium recordings and in vivo by stimulation of a specific cell type.²⁶ This device platform can also be combined with microfluidics for on-demand GABA delivery from a reservoir and has been used successfully in vivo in an epilepsy model. This example, OEIP, was integrated into a neural interface electrode to trigger GABA pumping upon detection of pathological neural activity to obtain a closed-loop seizure control.²⁷ The demonstration of the closed-loop control of targeted drug delivery makes this platform one of the significant for translation into a clinical setting. Despite the positive results, OEIP drug delivery has been targeted at chronic conditions (epilepsy and neurodegeneration) and, in these conditions, therapy is required for patient lifetimes. While drug release can be controlled and the use of reservoirs have enabled high loading of drug molecules, these devices retain a limited long-term therapeutic efficacy due to the finite volume of the reservoir. Approaches are being developed to enable refilling of reservoirs, but this will be associated with limitations on implant location, repeat surgery, or the need for percutaneous channels that present infection risks. It is, however, perceived that OEIPs could be effectively used in short-term drug delivery applications, such as chemotherapy, where treatment is usually required for under 5 years.

Although CPs represent a strong foundation for the development of localized drug delivery platforms, many of the challenges surrounding precision control of drug delivery remain. Owing to the electroactive nature of CP-based platforms, their successful operation to date has largely focused on the use of charged drug molecules. Although many biologically charged drug molecules are used clinically, this limitation hinders the overall translatability of these devices and renders them appropriate for only some disease states. For example, cancerous tumors are best treated with specific drugs that are related to the tissue type in which they occur and the stage of tumor progression. Many chemotherapy drugs are neutrally charged or cationic and therefore are not able to utilize the doping and dedoping drug release mechanism of a CP. It is possible to induce charged states on noncharged drug molecules through pH interactions and modifications to the molecules; however, this can lead to unwanted effects on the surrounding tissue or reduction in therapeutic efficacy. Therefore, in order for localized CP delivery platforms to become widely used, future work needs to focus on CP systems that can operate independent of the drug net charge. Both OEIPs and CP mediated actuation approaches can be used with noncharged drugs, as seen in Fig. 2. Importantly, electrical actuation of a CP component can release drugs not only in response to a mechanically induced pressure within the matrix but also through dopant flux that results in changes to the charge balance of the matrix. Both mechanisms ‘carry’ noncharged molecules out of the CP at a faster rate than passive diffusion.¹¹ OEIPs and actuation approaches have the most promise for enabling the active transport of noncharged drugs using a CP based device. These approaches may also be expanded to introduce the concept of dual drug delivery where different drug types (for example, one charged and the other uncharged) can be delivered from the same device by controlling the applied potential and therefore the mechanism of release.

While studies have shown a clear increase in drug release and concentration of therapeutics at the target tissue site when localized delivery is used, there is still minimal comprehensive information on how diffusion properties of the drug within the tissue impact therapeutic efficacy. Diffusion is a rate limiting phenomenon that not only occurs at the interface of device and tissue but is also an intrinsic property of the tissue itself. Systemic delivery commonly uses the circulatory system to ensure that drugs penetrate the target tissue; however, localized delivery devices are usually placed adjacent to or penetrating the target tissues without compromising associated circulatory structures. This means that while the release of drug may be controlled and in contact with appropriate tissues, its diffusion through these tissues is uncontrolled, being both spatially and temporally limited by the natural fluid transfer properties. Localized delivery systems ensure the availability of the drug, but rarely influence the mobility of the drug, and thus, the therapeutic benefit may be reduced. This challenge can be addressed through device design (penetrating or needle structures), manipulation of tissues to improve drug diffusion (such as electroporation), or design of smart drugs that can penetrate cell membranes (actively transported through cell receptor motifs). However, each of these approaches can only be addressed through individual applications, as tissues types are highly variable and disease conditions can also alter tissue properties and drug transport. Another possible method for enhancing postdelivery drug diffusion is the use of electrical potential to drive the drug toward or away from specific locations of the tissue.⁴⁰ This method requires more complex devices with additional electrodes [as depicted in Fig. 2(b)], charged molecules, and the application of charge to drive directed ion flux. Conveniently, CPs are materials that can deliver substantial amounts of charge while maintaining a low voltage within the tissue environment. While CPs provide innovative pathways for influencing drug diffusion within tissues, there has been little focus on this aspect of drug delivery and substantial development is needed to understand the most effective approaches.

The most significant challenge for many drug delivery applications is the desired chronic time scale of delivery and how to effectively and safely carry enough drug to meet supply over that time scale. One major benefit of OEIPs is their ability to be linked to a microfluidic reservoir. However, this also incorporates risk from a clinical perspective, as it is necessary to ensure that in the case of a reservoir failure, the contained amount of drug will not be toxic to the local tissues. In many applications, a device with an attached reservoir is either impossible or infeasible to implant. In the applications where a reservoir is not possible, CP-based devices benefit from the ability to actively control drug release, both for delivery and containment, in order to release only the amount of drug necessary. Coupled with CH-based coatings and their large drug carrying capacity, the potential for prolonged active release can be realized. Going forward, CP-based devices that cannot be connected to an external reservoir can still benefit from high initial drug loading capacity with fine-tuned active release profiles, which can inhibit any long-term passive diffusion.

Finally, there is a need to translate the drug delivery materials and electronics into an implantable system that is tolerated by the local tissues. While added complexity may arise from the need for reservoirs and supporting electronics, there is an extensive range of technologies within the implantable bionics field from which implantable packaging approaches can be drawn.⁴¹ Electronics packaging with hermetic encapsulation of humidity sensitive components has been established for a number of stimulating devices, including pacemakers, cochlear implants, and deep brain stimulators.⁴² A bioelectronic drug delivery device, which has only a small number of electrodes, could be feasibly attached to a commercially available stimulator system. However, these inorganic and rigid components would need to be located at a distance from the target tissue, where foreign body processes do not impact the therapeutic efficacy of the device. This is a common approach used in bionics; for example, cochlear implant electronics are placed subcutaneously, where they are both distant from the cochlea but also accessible by RF coils to ensure continuous powering.⁴³ In line with controlling the foreign body response, the local drug delivery components must also be well tolerated to minimize inflammatory reactions. Of significant benefit is that CP systems, in particular composites of CPs, are organic materials, which are orders of magnitude softer than conventional metallic electronics.³⁴ It is well documented that the mechanical and chemical properties of these materials produce a reduced inflammatory response and a minimal long term scar tissue encapsulation.^44,45 As such, it is expected that CP drug delivery systems will be well tolerated across chronic implant lifetimes.

CP-based devices present unique opportunities for addressing controlled localized drug delivery toward clinical applications. Although these CP-based systems offer attractive solutions, they still face many hurdles on the road toward translation from research to clinical use. Despite having demonstrated in vitro and in vivo efficacy, little headway has been made into clinical-based investigations of CP-based drug delivery systems. This is, in part, due to industry and regulatory reluctance toward adoption of modern functional materials. Overcoming the regulatory hurdle necessitates a field-wide paradigm shift toward translation of new materials, acknowledging the significant benefits that can be provided over conventional technologies.⁴⁶ Ultimately, there must be a concerted effort from researchers, industry, and clinicians alike to facilitate patient benefits from active implantable devices based on CPs. It is anticipated that CP based drug delivery will improve recipient quality of life through the application of the correct dose levels to the target tissues while reducing the impact of systemic drug administration on patient health and, subsequently, the ongoing cost to health care systems. If the capacity to control both drug release and retention while having more than one addressable drug within a given device can be realized, CP based drug delivery systems can facilitate a personalized bioelectronic medicine approach to the treatment of a wide range of conditions.