(692d) Design of a Microinjection Device for Injection of in Situ Gelling Hydrogels for Ophthalmic Drug Delivery

Design of a microinjection device for injection of in situ gelling hydrogels for ophthalmic drug delivery

Scott Campbell*, Jun Yang**, Wen-I Wu**, P. Ravi Selvaganapathy**, and Todd Hoare*

*Department of Chemical Engineering, McMaster University, Hamilton, ON, Canada

***Department of Mechanical Engineering, McMaster University, Hamilton, ON, Canada

E-mail: campbesb@mcmaster.ca  |  hoaretr@mcmaster.ca

INTRODUCTION: The primary causes of vision loss in developed nations are diseases associated with the posterior eye, such as age-related macular degeneration, diabetic retinopathy, posterior uveitis, and retinitis due to glaucoma.^1–3 Thus, the development of materials that can effectively treat such ailments would be highly beneficial. However the posterior region of the eye is a particularly difficult target tissue due to anatomic and physiologic limitations.² Treatments involving injections to the back of the eye have proven quite successful, but the required frequent injections greatly increases the risk of complications over time.² Hydrogels present a potential solution to this issue, as they can facilitate prolonged delivery of a drug to the eye and thus limit the number of required injections for effective treatment.⁴ Furthermore, unlike most drug release vehicles, hydrogels can be designed to effectively match the refractive index of the vitreous humour and be injectable in situ, exploiting the rapid chemical reaction of complementary groups grafted to two polymers (such as aldehyde-mediated crosslinking of alcohol-, amine-, or hydrazide-functionalized polymers or  thiol-mediated crosslinking with vinylsulfone or acryloyl groups) to form the crosslink structure.^5,6 Crosslinked networks may also find applications as vitreous replacement materials for patients suffering from opacification of the vitreous, commonly observed upon aging.

We have developed several in situ-gelling injectable hydrogels using aldehyde-hydrazide chemistry to produce hydrolytically degradable hydrazone crosslinks.^7,8  When polymers with these two functional groups come into contact, they rapidly form a hydrogel, entrapping the desired drug to be released in the process. These hydrogel materials have promising tunable drug delivery characteristics for a variety of drugs, can have refractive indices similar to that of the vitreous, and slowly degrade in physiological conditions.⁷ However, to assess the in vivo capabilities of these hydrogels in mouse, rabbit, and ultimately human eyes, very small amounts of each reactive material (1 – 10 μL) must be injected.  While this is routinely done with single component systems, the additional requirement of mixing the two reactive polymers upon injection (to prevent premature gelation) poses a significant additional challenge with administering these injectable hydrogels to the point that no suitable injection system currently exists. In response, this work describes the fabrication of a novel microinjector that: (1) effectively mixes the two precursor polymers from separate microchannels upon injection; (2) controllably and precisely injects volumes in the 1 - 10 μL range through a narrow gauge needle suitable for ophthalmic applications; and (3) rapidly injects these materials to prevent gelation, and blockage within the needle. The developed dual-barrel microinjection system is characterized in terms of its optimal channel length, the precision of the injected volumes, the properties of the ejected polymer solutions/hydrogels after injection, and its use in initial in vivo studies.

EXPERIMENTAL: The in situ-injectable hydrogels are typically prepared by the mixing a hydrazide-functionalized polymer (A-Polymer) and an aldehyde-functionalized polymer (B-Polymer) (both have a MW <32 kDa) upon injection using a double barrel syringe (Figure 1). The double-barrelsyringe has two compartments: one with hydrazide functionalized NIPAM and the other with aldehyde functionalized dextran in PBS solutions. Upon injection, these two solutions interact in the mixing channel and needle, forming hydrazone bonds that crosslink the polymer to form a hydrogel.

The microfluidic injector was designed to operate analogous to this macroscale system. The A- and B-polymer solutions are contained within separate reservoirs that are connected to the same pneumatic pressure source. Upon the application of a given pressure pulse, the polymers will travel through a microfluidic channel of a given length. To effectively mix the two polymer components, a staggered herringbone mixer originally described by Stroock et al. (2002) will be utilized which generates transverse flows to induce microvortexes and promotes mixing.⁹ After navigating through the mixing channel, the mixed polymer solutions will be ejected from the microfluidic injector via a 30 ½ (or higher) gauge needle, which are typically used for in vivo ocular injections. A simplified schematic of the microinjection device is shown in Figure 2.

RESULTS: The initial prototype is shown in Figure 3, which depicts the initial device design and the staggered herringbone structure. The channels have a width of 200 μm and a height of 100 μm with the herringbone grooves having a height of 23 μm and spacing of 50 μm.

A syringe pump was set to generate an equal flow rate at both the inlets. DI water mixed with methylene blue dye was flowed in the central inlet while DI water was flown in the other two inlets.  A comparison of mixing in microchannels with and without the herringbone design was made, as shown in Figure 4. It shows that complete mixing occurs at 17.5 mm in the channel with the herringbone design. On the other hand in the case of a simple microchannel the two streams are still delineated from each other and the mixing is only due to diffusional effects. The degree of mixing was shown to increase with channel length as well.

 Next, the mixing channel design was used with polymer A and B solution to determine whether effective mixing occurs. The interaction of two polymer solutions in these channels can be observed as the two polymer solutions are transparent and  become opaque upon gel formation. As shown in Figure 5, the herringbone design promotes faster mixing and gel formation in these microchannels.

The point of gelation can be used to determine the placement of the outlet. The optimal length will depend on the polymer solution viscosities, the polymer concentration in the initial solutions, the applied flow rate, and the gelation rate of the polymers (which depends on the degree of functionalization and polymer properties). Tests on the final material after it is ejected from the needle at varying degrees of mixing will be discussed, aiming to determine the optimal channel length for a given flow rate. For a chosen solution, the structure of the channels will be altered to lower the channel lengths and allow for the insertion of neddles to the channel outlet. The level of control over the injected volume will be reported, along with the preliminary in vivo testing of the device and their comparison with conventional injections.

CONCLUSIONS: The development of a device that could inject precise small amounts of in-situ gelling hydrogel precursors allows for the in vivo assessment of injectable hydrogels as ocular drug delivery materials, enabling potentially significant improvements over current therapies in terms of treating diseases of posterior eye.

REFERENCES: (1)  Myles, M. E.; Neumann, D. M.; Hill, J. M. Advanced Drug Delivery Reviews 2005, 57, 2063–79.  (2)  Sheardown, H. Future Medicinal Chemistry 2012, 4, 2123–2125.  (3)  Fitzpatrick, S. D.; Jafar Mazumder, M.; Lasowski, F.; Fitzpatrick, L. E.; Sheardown, H. Biomacromolecules 2010, 11, 2261–2267.  (4)  Patenaude, M.; Hoare, T. Biomacromolecules 2012, 13, 369–378.  (5)  Van Tomme, S. R.; Storm, G.; Hennink, W. E. International Journal of Pharmaceutics 2008, 355, 1–18.  (6)  Overstreet, D. J.; Dutta, D.; Stabenfeldt, S. E.; Vernon, B. L. Journal of Polymer Science Part B: Polymer Physics 2012, 50, 881–903.  (7)  Patenaude, M.; Hoare, T. ACS Macro Letters 2012, 1, 409–413.  (8)  Campbell, S. B.; Patenaude, M.; Hoare, T. Biomacromolecules 2013, 14, 644–653.  (9)  Stroock, A. D.; Dertinger, S. K. W.; Ajdari, A.; Mezic, I.; Stone, H. a; Whitesides, G. M. Science 2002, 295, 647–651.

ACKNOWLEDGEMENTS: This research is funded by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Ontario Graduate Fellowship, and the J.P. Bickell Foundation (Medical Research Grant).