Elsevier

Journal of Controlled Release

Volume 100, Issue 2, 24 November 2004, Pages 211-219
Journal of Controlled Release

In vivo release from a drug delivery MEMS device

https://doi.org/10.1016/j.jconrel.2004.08.018Get rights and content

Abstract

A drug delivery microelectromechanical systems (MEMS) device was designed to release complex profiles of multiple substances in order to maximize the effectiveness of drug therapies. The device is based on micro-reservoirs etched into a silicon substrate that contain individual doses of drug. Each dose is released by the electrochemical dissolution of the gold membrane that covers the reservoir. The first in vivo operation of this device was reported in this study. Subcutaneous release was demonstrated in rats using two tracer molecules, fluorescein dye and radiolabeled mannitol, and one radiolabeled chemotherapeutic agent, carmustine (BCNU). BCNU was chosen because of the need to improve the direct delivery of chemotherapy to malignant tumors. The spatial profile of fluorescein dye release from the drug delivery device was evaluated by fluorimetry, the temporal profile of 14C labeled mannitol release was evaluated by liquid scintillation counting, and the temporal profile of 14C labeled BCNU release was evaluated by accelerator mass spectrometry (AMS). Release profiles obtained from injected controls were compared with those from activated devices. The in vivo dye release results showed high concentration of fluorescein in the flank tissue surrounding the devices 1 h after activation. The 14C labeled mannitol released from the drug delivery devices was rapidly cleared (1 day) from the rat urine. In vivo release of 14C labeled BCNU from activated devices showed slightly slower kinetics than the injected and in vitro controls, and the time to reach the steady-state plasma 14C concentration was on the order of 1 h. All these results demonstrated the capability of this drug delivery device to achieve localized delivery of various compounds with well-defined temporal profiles.

Introduction

Most controlled release implants are capable of delivering only a single compound at a constant rate although it is well known that the rate of a drug's delivery and its interactions with other chemicals have a significant influence on its effectiveness [1]. Recent advances in the fabrication of microelectromechanical systems (MEMS) offer unique opportunities to create novel implantable drug delivery systems to maximize the efficacy of drug therapies [2], [3]. A variety of microfabricated devices and components have been designed to release drugs of different dosages and with different delivery pattern and duration to address various clinical needs. For example, microfluidic devices that incorporate micropumps, valves and flow channels have been investigated to deliver drugs, proteins and genes [4], [5]. Microfabricated porous membranes have been used for drug encapsulation [6] and microparticles have been used for aerosol and gene delivery [7], [8].

An implantable drug delivery MEMS device has been developed that is capable of delivering multiple substances either in series or in parallel [9]. The device is based on a silicon substrate into which micro-reservoirs are etched. Each reservoir is filled with a dose of drug and covered with a gold membrane (Fig. 1). The membrane electrochemically corrodes into soluble gold chloride when a positive potential is applied, and the drug in the reservoir is free to dissolve or diffuse into solution [10]. Potential advantages of this device include the variety of deliverable compounds, capability of complex release profiles, and precise dosing. Limitations include the small size of the micro-reservoirs and the need for surgical implantation and explantation. This kind of device may be particularly useful for the delivery of hormones, powerful painkillers, chemotherapeutic drugs, and other potent drugs.

One such compound of potential use is 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU), otherwise known as carmustine, an alkylating agent that binds DNA to form interstrand cross-links. Biodegradable polymers incorporated with BCNU (Gliadel®) have been used for the sustained release of BCNU for brain tumor therapy. Although Gliadel has prolonged survival for brain tumor patients, the tumors eventually progress causing death [11]. Local delivery of BCNU combined with other chemotherapeutic agents, such as O6-benzylguanine, or immunotherapeutic agents, such as interleukin-2, can enhance the efficacy of BCNU therapy. Laboratory studies have indeed demonstrated improved survival on experimental tumor models [12], [13]. Effective delivery of the mutli-drug combination requires precise control of the temporal and spatial profiles of the agents, which is generally difficult to achieve using biodegradable polymers. A MEMS drug delivery device, on the other hand, will enable release of multiple compounds in complex release profiles to maintain drug concentrations within their therapeutic window. Moreover, the impermeable gold membranes can effectively protect the drugs contained in the microreservoirs from hydrolytic degradation before release. It is likely that the same therapeutic effect can be achieved using the MEMS device with lower drug dosage requirement compared to the drug impregnated polymer wafers so that systemic toxicities will be decreased.

Understanding the release kinetics of BCNU from this drug delivery device is important for its in vivo operation. Available pharmacokinetic data using 14C labeled BCNU indicate that the compound and its metabolites are accumulated in the liver, kidney and lungs, and that approximately 78% of a radioactive dose injected intraperitoneally, subcutaneouly, or taken orally, is excreted from mice within 24 h. The plasma 14C concentration, though only a small portion (1–5%) of the dose, remains constant in the plasma for relatively long duration with a 67-h half life [14]. It is therefore appropriate to measure pulses of BCNU release from the drug delivery device using plasma samples to obtain the temporal release kinetics. This approach however is limited by the relatively low sensitivity of the conventional liquid scintillation counting (LSC) with respect to the low plasma 14C level. An alternative tool of isotope detection is the accelerator mass spectrometry (AMS), a highly sensitive method for quantifying extremely low concentrations of radioisotopes with ultrahigh precision [15], [16]. Additional advantages of AMS include low radioactivity and small sample size requirements.

Section snippets

General procedure

A two-part study examined the release of chemical substances from a drug delivery MEMS device. First, the release of two tracer molecules was studied on devices implanted subcutaneously in rats. A voltammetric profile sufficient to activate the device in vivo was established; the spatial profile of fluorescein release and temporal profile of mannitol release were examined. The second part of the study examined the temporal release kinetics of 14C labeled BCNU from devices.

Device fabrication and packaging

The device fabrication

Fluorescein and mannitol release

The voltammetric profile necessary to activate the devices in vivo differs from that in saline because proteins adsorb onto the gold surface in vivo and therefore slow corrosion kinetics. Different voltammetric profiles were applied in the fluorescein release study to test their efficiency of opening anode gold membranes in vivo. Optical microscopy examination of the explanted devices showed that the most reliable membrane opening occurred with both a cathodic cleaning cycle and a 20-min square

Discussion

The results of this study indicate that the subcutaneous use of this drug delivery MEMS device for delivery of drug is feasible and that the in vivo chemical release was successful. The capability of this device to achieve localized delivery of various compounds with well-defined temporal profiles was demonstrated.

A sufficient voltammetric profile for in vivo activation and complete opening of the device anode membranes was determined. This profile included a cathodic cleaning cycle to remove

Conclusions

A drug delivery MEMS device was used in this study to deliver tracer molecules as well as a therapeutic agent in vivo. Their spatial and temporal release profiles were evaluated using fluorimetry, scintillation counting and AMS. The results clearly demonstrated the capability of this drug delivery device to achieve localized delivery of various compounds with well-defined temporal profiles. Moreover, they provided useful BCNU release kinetics information for the ongoing efficacy study that uses

Acknowledgements

This work was funded by the U.S. National Institute of Health grant no. 1 R24 AI47739-01 and U19-CA52857. R.S.S. was supported by a U.S. National Science Foundation fellowship. All microfabrication work was carried out at the Microsystems Technology Laboratory at MIT. The AMS work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under Contract W-7405-Eng-48 and in collaboration with the National Center for

References (26)

  • J.T. Santini et al.

    A controlled-release microchip

    Nature

    (1999)
  • J.T. Santini et al.

    Microchips as controlled release drug delivery devices

    Angew. Chem., Int. Ed.

    (2000)
  • L.D. Rhines et al.

    O6-Benzylguanine potentiates the antitumor effect of locally delivered carmustine against an intracranial rat glioma

    Cancer Res.

    (2000)
  • Cited by (119)

    • AI in microfabrication technology

      2023, A Handbook of Artificial Intelligence in Drug Delivery
    • Materials and device design for advanced phototherapy systems

      2022, Advanced Drug Delivery Reviews
      Citation Excerpt :

      However, with the recent development of injectable optical devices, it is possible to deliver light deep into the living body, so not only tumors formed on the surface of the skin or organs, but also cancers located in deep places, can be effectively treated through photo-dynamic therapy with the photosensitizer [188]. In addition, if a drug delivery system (DDS) that can spatiotemporally control the photosensitizer concentration in the lesion site is used, it is expected that a small dose of drug can be used to effectively treat cancer, and prevent side effects caused by excessive dose of the photosensitizers [189,190]. In the following part, we investigate a drug delivery system that can potentially be used for injection of the photosensitizers.

    • Materials and technologies for multifunctional, flexible or integrated supercapacitors and batteries

      2021, Materials Today
      Citation Excerpt :

      Active drug delivery devices allow precise control over the rate of a drug release. The role of a battery (Fig. 7b) in the active device is to power its circuitry and on-board electronics to release the drug either via electrothermal (where passage of a current ruptures a thin film) [398] or electrochemical (where passage of a current oxidizes and dissolves a thin film) [399] routes. In an alternative (passive) design with a much lower fraction of circuit components, a galvanic cell could be used to passively induce formation of gases (e.g., H2 bubbles during spontaneous Zn oxidation and hydrogen reduction), which activate a drug release in vivo [400].

    • Zero-order drug delivery: State of the art and future prospects

      2020, Journal of Controlled Release
      Citation Excerpt :

      For example, one MEMS device has achieved pulsatile delivery of parathyroid hormone to treat osteoporosis; this system has undergone human trials and has been shown to have the same efficacy as the current standard of care, daily subcutaneous injections over 20 days [264]. Other active microchips have been used to achieve rapid delivery of vasopressin to treat various cardiac pathologies [275] and temozolomide to treat glioblastoma [276–278]. It is clear that active devices are capable of complex drug release profiles; however, their ability to achieve zero-order drug release is relatively unexplored.

    View all citing articles on Scopus
    View full text