We have open-sourced the design files for Otto, the Franz diffusion cell autosampler that we built to automate skin permeation studies. This means we’re sharing the 3D-printable models with the world, for free. Anyone may print, modify and share these files with others — with attribution. The files are released under the CC BY 4.0 licence.
Abdelghany TM, Vo N, Vukajlovic D, Smith E, Wong JZ, Jackson E, Hilkens CMU, Lau WM, Ng KW, Novakovic K. Engineering and in vitro evaluation of semi-dissolving, hydrogel-forming polymeric microneedles for sustained-release drug delivery. Int J Pharm. 2025:125932. https://doi.org/10.1016/j.ijpharm.2025.125932
In our latest paper, we describe a microneedle formulation that utilises two polymeric domains: a soluble one and an insoluble one. The insoluble domain is chemically crosslinked and traps the soluble polymer, along with the drug, within it. This combination creates a microneedle array patch that can release a drug for over 2 months.
It can contain a significantly larger dose than microneedles where the drug is contained within the microneedle tips only (e.g., detachable microneedles). The drug reservoir in the backplate makes this possible to support extended release. It uses one-pot synthesis, a mild hydroalcoholic solvent system and mild temperatures to aid manufacturability and drug stability.
For the first time, we were able to see, on video, how the microneedles released the drug and swell as they hydrated. These videos are buried in the supplementary files for the paper, but I thought it worthy of sharing more widely here:
In this collaborative paper—our second with the Malaysian team led by Dr Choon Fu Goh—we examine some commercially available cosmetic products and ask what lessons we can learn from them to enhance pharmaceutical microneedle product translation and commercialisation.
We have known, for a long time, that the regulatory hurdles for pharmaceutical products are much greater than those for cosmetic products. Still, it’s interesting to see how cosmetic microneedle products have surged years (if not decades) ahead of their pharmaceutical counterparts, particularly in the Asian market. A low regulatory hurdle could spur innovation, but it could equally grow fad. How can one tell which it is? We examined a selection of commercially available cosmetic microneedle products to find out, and report our findings in this paper.
This has been an interesting paper to work on. I have admired Goh’s tenacity collecting microneedle patches from pharmacies on his various international trips across Asia for this study. Last year, I hosted him in Newcastle to conduct parts of the study, including some microscopy work and the optical coherence tomography (OCT) analysis on microneedle penetration in ex vivo pig skin. It’s rewarding to see those efforts pay off.
Citation: Chan HKY, Archbold L, Lau WM, Ng KW. Validating Otto: a Franz diffusion cell autosampler to automate in vitro permeation studies, Journal of Pharmaceutical Sciences, 2025:103837. https://doi.org/10.1016/j.xphs.2025.103837
We have a new paper out. This one is close to my heart because I personally spent many hands-on hours developing Otto.
Who’s Otto?
Otto is a Franz diffusion cell (FDC) autosampler robot. It replaces manual sampling and refilling of FDCs in a skin (and cornea, mucosal membrane, etc.) permeation experiment. Those who have worked with FDCs before would know how fiddly, time-consuming and labour-intensive that is. It’s a job most suited for a robot.
But Otto is about more than us trying to avoid menial labour. It’s about the quality of the science, too.
Let me rephrase that — it’s primarily about the quality of the science.
For a long time, we have noted many skin drug absorption studies that include unusually large sampling gaps of ≥16 hours, presumably because the researchers were unable to collect samples outside normal working hours. This sampling gap could allow the drug to accumulate in the FDC receptor chamber and, consequently, underestimate drug absorption due to sink condition being violated. We have faced similar logistical challenges ourselves as local rules prevent some researchers from working out of normal working hours. A FDC autosampler would solve these challenges, but we have not been able to afford any of the few commercial FDC automation systems available. When COVID-19 hit, and lab access was further restricted, we finally found the impetus and time to build the FDC autosampler we had always needed, for less than £500, and retrofitted it to our existing FDCs.
Thus, Otto was born.
Otto is a Franz diffusion cell (FDC) autosampler robot, adapted from the Creality Ender 3 Pro 3D printer. This picture depicts an automated skin permeation experiment using FDCs, in which Otto handled FDC sampling and refilling fully unattended. See the full paper for details on the number labels. Image reproduced under the CC BY 4.0 licence.
We have spent the last 2 years validating Otto’s performance. In this paper, we demonstrate that the sampling gap indeed led to violation of sink condition and underestimation of drug absorption. We further show that Otto improved data quality by avoiding the sampling gap. We have benchmarked Otto’s precision and accuracy against a trained researcher. We are pleased to report that it outperforms the researcher on both counts.
Otto is better than the commercial offerings in many ways. It is built on open-source technologies, using inexpensive consumables and 3D-printed custom parts, and is therefore fully customisable. It has a small footprint of just 50 cm × 46 cm. It can be retrofitted to generic FDCs and can collect up to 100 samples per experiment, fully unattended. The samples are collected directly into high-performance liquid chromatography (HPLC) autosampler vials, so it integrates seamlessly with downstream analysis without any further liquid handling, nor modification to the FDCs or analytical equipment.
Logistical, human resource and financial constraints continue to grip many research organisations long after COVID-19 restrictions have ended. Otto should prove itself a valuable asset in many research labs seeking to retrofit an automation solution to their existing FDCs.
The build instructions for Otto are too extensive to include in this paper, so we will be publishing them separately.
Keng has published a new paper out with Dr Othman Al Musaimi and co-workers. In this paper, we report the development of a self-assembling hydrogel formulation based on peptide sequences derived from elastin.
Elastin is a naturally occurring protein found in many connective tissues in the body, including the skin and blood vessels. These peptide sequences have been selected carefully to promote self-assembly of the hydrogels and confer the desired mechanical properties to the hydrogel. This hydrogel can be an interesting drug delivery system. The ability for the hydrogel to self-assemble at room temperature makes it easier to incorporate drugs into the hydrogel matrix. The mechanical properties will determine the rate at which the drugs can then be released from the hydrogel.
The paper is open access and free to read, so head over there now to read the full text for free.
Hydrogels are a popular drug delivery vehicle. You can encapsulate drugs including large biological macromolecules like proteins in them, to be released in the body. Some hydrogels use chemical crosslinking to create the hydrogel matrix – with the drug in it. Can protein drugs encapsulated this way participate in those crosslinks? What if they do? How will that affect their subsequent release?
We started asking these questions when attempting to deliver protein drugs by encapsulating them in hydrogel-forming microneedle array patches. These were important questions, and now we have the answers.
In this paper, we address these questions using bovine serum albumin as a model drug, and a genipin-chitosan hydrogel as the drug delivery vehicle. Using a combination of empirical fluorometry data and mathematical modelling, we investigate the kinetics of the interactions of the protein drug and genipin (the crosslinker).
We test our microneedles a lot to examine how they fail. The most common technique we use to quantify the mechanical strength of the microneedles is by applying axial compression (crushing the microneedles from the tip towards the base) on a texture analyser, and pinpointing the minimum force that results in microneedle failure. We’re fortunate that our lab is well equipped to perform these tests. We get a live video feed of what’s happening to each microneedle as it’s being compressed. We can also playback the videos and analyse them synchronously with the force-displacement data. This technique has proved invaluable in making sure our microneedles perform as they should.
Hence, when Dr Choon Fu Goh invited me (Keng) to contribute to this paper on the mechanical testing of microneedle strength, I jumped at the opportunity. This is our first paper together and I really enjoyed working on this. I hope the microneedle community finds it useful.
Our hands-free, ‘vac-and-fill’ micromoulding technique prevented air entrapment and bubble formation in viscous formulations when degassed under vacuum. Image from Smith E, et al. Int J Pharm 2024;650:123706. Licence: CC BY 4.0 Deed.
This paper reports the solution to a problem that took us several months to solve. We were trying to mould a microneedle array patch. There are basically two ways to do it: you fill the mould with the liquid formulation and either centrifuge it or degas it under vacuum. Both techniques are widely reported in the literature. They have been designed to force any air out of the microcavities in the mould, so that the formulation can enter them to form the microneedles. We didn’t have the right rotor to go with the centrifuge, so we opted for the vacuum degassing technique, fully expecting it to be a walk in the park. What a disappointment that turned out to be! We discovered that our formulation was too viscous to allow the air to escape. We ended up with a lot of air bubbles trapped in the liquid formulation.
We quickly realised that the vacuum degassing technique reported in the literature had used low polymer concentrations, which meant that their liquid formulations were not as viscous as ours. To micromould the microneedle array patch successfully from our viscous formulation, we had to remove the air first before filling the formulation into the mould. But how would one fill the mould under vacuum?
The answer: a modified syringe, a 3D-printed part, some painstaking calibration, and viola! The paper describes our solution in full, but here’s a peek of the contraption in action.
This is Emma’s first paper and our first together with Dr Katarina Novakovic‘s group. Congratulations, Emma, and thank you team for the hard work!
We have published a review article on microneedles as a technological platform for diagnosing infectious diseases. In this review, we enumerate the infectious diseases that could potentially be diagnosed in the skin, examine the mechanisms of existing microneedle diagnostic technologies, and evaluate their applications in infectious disease diagnosis. This publication is timely given that we’re in the middle of a infectious disease pandemic.
Figure 1: Microneedles inserted into the skin may extract or detect disease markers in situ. Diagnostic tests for infectious diseases should ideally be both specific and rapid.
Any diagnostic test has to be specific to be useful (Figure 1). For a potentially deadly diseases, the more rapid the diagnosis, the better, because it buys precious time for the patient to seek early treatment, which could save lives. However, for infectious diseases that can be transmitted by close contact, it’s also desirable that the patient can administer the test themselves without too much discomfort. Many tests do better in meeting some of these criteria at the expense of other criteria. For example, I took the PCR test for COVID-19. I am sure the test itself was highly specific, but the nasal and throat swabs were uncomfortable. The diagnosis wasn’t exactly ‘rapid’ either — I got my result several days later (mainly due to limited test capacity at that time, but that has improved significantly since). Microneedle devices are painless to administer on the skin, and tests can produce real-time or near-real-time results. Not all of these studies have been on infectious diseases, of course. The technology is still nascent but the potentials are huge.
The review article is currently in press, but a pre-proof is already available for download from Acta Pharmaceutica Sinica B. There have been a number of minor corrections to this pre-proof (mainly typographical and referencing error) which will appear in the final publication, but the pre-proof should satisfy the impatient for now.
Congratulations to everyone involved in putting this publication together.
