Introduction
As succinctly stated by former FDA commissioner, Dr. Scott Gottlieb, and CDER Director, Dr. Janet Woodcock, “One of today’s most important tools for modernizing the pharmaceutical industry is a process known as continuous manufacturing…”.1 A critical challenge in the pharmaceutical industry is the balance between quality and throughput – the solution, continuous manufacturing. Over the past 10 years, notable advances in manufacturing have led to continuous manufacturing systems with a major emphasis on tablet production and API synthesis. Nevertheless, there is another area that has a scarcer focus, which is continuous manufacturing of complex formulations such as liposomal nanoparticles and lipid nanoparticles (LNPs). Although there are only a handful of liposomal drug products approved by the FDA, there is still a bright future for liposomes as delivery vehicles for antineoplastic drugs (e.g. doxorubicin) and LNPs for nucleic acid delivery (e.g. nucleic-acid vaccines and CRISPR technology). Accordingly, at UConn in the lab of Dr. Diane Burgess, we developed a continuous processing system for nanoparticles, which was funded by the US FDA under the 21st Century Cures Act. We are currently designing our system to be cGMP-ready and will perform qualification runs on doxorubicin-loaded liposomes. The work presented here is a high-level overview of our continuous processing approach.
About Doxorubicin-Loaded Liposomes
Liposomal doxorubicin is used as an anticancer therapy for ovarian cancer, AIDs-related Kaposi’s sarcoma and multiple myeloma. One can think of doxorubicin-loaded liposomes as a way to deliver a high-payload to a tumor site, with the added benefits of increased circulation time and reduced cardiotoxicity when compared to free doxorubicin.2 Doxorubicin-loaded liposomes conventionally have a single lipid bilayer with a doxorubicin nanocrystal in the intraliposomal space.3 However, the process to make doxorubicin-loaded liposomes is actually more complicated than one may think. It is important to note that having a processing system that can make these nanoparticles actually enables the encapsulation of an abundance of small molecule, amphipathic weak acid/base drug candidates. Such a processing system opens up the field to new liposomal drug products delivering single or multiple payloads.
In order to form these particles, it is important to understand how doxorubicin hydrochloride can actually be loaded into these nanoparticles. In a nutshell, the first step is to make empty liposomes and the second step is to load the empty liposomes through a mechanism known as “remote loading”. This mechanism is based on establishing a salt gradient, such as with ammonium-sulfate (many other types of salts work as well) between the intraliposomal space and the extraliposomal space, where the intraliposomal salt concentration must be much greater than the extraliposomal salt concentration. The high intraliposomal salt concentration is the “loading battery” that drives the free-doxorubicin into the intraliposomal space. At sufficiently high intraliposomal salt concentrations (e.g. > 150 mM), a rod-like, doxorubicin nanocrystal forms.4 Of course, other structures can also form, such as circular nanocrystals and U-shaped crystals.5 Moreover, under certain conditions, the rod-like nanocrystal may lead to more spherical liposomes or elongated liposomes. A question that may arise is, do these structures impact efficacy or clinical adverse events? The answer to that is still up to debate, but many studies do show that generic forms of liposomal doxorubicin have a variety of crystal morphologies and particle size distributions,6 which begs the question, are these really the same? Conventional batch manufacturing approaches are most likely the result of the inconsistencies of these products. With a batch process, there is difficulty in producing liposomes with nanocrystals that have a uniform morphology. In some products, we observe a mixture of elongated and near-spherical nanoparticles, whereas other formulations have mostly only elongated nanoparticles. Using a continuous manufacturing process over a batch process, we have the added benefit of being able to control the crystal morphology, making more uniform populations of nanoparticles resulting in a more consistent product.
Quality by Design for Liposomal Formulations
When it comes to processing of any pharmaceutical, it is important in the early research and development stages that one clearly understands how material attributes and processing parameters impact quality attributes. Following Quality by Design principles, an in-depth understanding of the “envisioned” drug product will lead to forming functional relationships between the material attributes and processing parameters, which become critical as the project moves from development to manufacturing. With respect to lipid-based nanoparticles, some questions around the physicochemical properties may include, for example, what is the lipid stability? What type of lipid should one choose – from a stability, economic and sourcing perspective? What is the target specification of the nanoparticle and what types of analytical techniques should be implemented to assess the quality attributes?
The Control Strategy
In the following section, I will outline some of the initial stages of a control strategy for liposomal formulations. A full explanation of a control strategy can be found in ICH Q117. As a first step, one needs to establish a quality target product profile and start listing quality attributes. For lipid-based nanoparticles, these include particle size, polydispersity, lamellarity, lipid concentration, API concentration, drug encapsulation percentage, et alia. The second step is to then perform some form of quality risk assessment, which can be based on experimental evidence and literature reviews. Next, one should establish a set of controls on material attributes and process parameters, followed by establishing functional relationships between the two on quality attributes (Figure 1). For liposomes, material attributes include the type of lipid (e.g. phospholipids vs sphingolipids), the lipid hydrocarbon chain length, the lipid composition, salt compositions, pH, lipid headgroup pKa, etc. Processing parameters would be dependent on the specific process and may vary widely. Common liposomal processing techniques include thin-film hydration, homogenization and organic solvent injection, where solvent injection may be performed using a co-axial turbulent jet, microfluidic cassettes or other mixing strategies.
Our strategy is based on the ethanol injection approach where we form a co-axial turbulent jet in co-flow. This approach is highly effective at controlling the particle size and maintaining a tight particle size distribution at a high-throughput.8 This scalable approach effectively enables researchers to use the system, while the same technology can be implemented from clinical trial material to manufacturing, albeit higher flowrates may be required for manufacturing. For an ethanol injection process, some processing parameters that are of importance include ethanol flow rate, aqueous phase flow rate, temperatures of both phases and degree of dissolved gases.
Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special offers
from American Pharmaceutical Review – all delivered right to your inbox! Sign up now!
Design of Experiments (DOE) is an important tool that can be used to establish functional relationships between material attributes and process parameters, with quality attributes as outputs. These outputs can then be used in models (either for predictive or screening purposes) to look for potential interaction and higher order terms of the inputs. For instance, when looking at the particle size (d.nm) as a model output, the aqueous phase flow rate, aqueous temperature and type of lipid typically have higher order terms and interaction terms, indicating a complicated relationship between a material attribute and two process parameters. Once all of the major relationships are established, one can start to form a knowledge space with a design space and normal operating region within.
Processing Via a Turbulent Jet in Co-Flow
As outlined in the FDA Draft Guidance “Quality Considerations for Continuous Manufacturing”,9 continuous manufacturing has the benefit of improving the manufacturing process by integrating processes, following QbD principles and implementing process analytical technology. The latter of which is critical from both a monitoring and control point of view. The processing system that we developed consists of three main stages: (1) a liposome formation stage; (2) a concentrating/dilution/buffer exchange stage; and (3) an active loading stage. We incorporated Process Analytical Technology (PAT) at each stage, with automated controls and monitoring capabilities. At the liposome formation stage, pre-dissolved lipid in ethanol is injected into an aqueous phase forming a turbulent jet with controlled flow characteristics. These intermediates then enter a degassing unit and a secondary dilution port. The degassing unit is important for reducing any dissolved gas prior to entering the dilution port. After dilution, the nanoparticles are analyzed via an online particle size analyzer, where particle sizing data can be used to control process parameters and for monitoring. As an example, a 45-minute run is shown in Figure 3, where both particle size (d.nm) and polydispersity were monitored every 20-25 seconds. Deviations in particle size from the specification can be used in feedback mechanisms to return the particle size back within the target range. After the nanoparticles are formed, particles are passed into a single-pass tangential flow filtration unit and are continuously concentrated/diluted during the entire run, while maintaining a controlled state. Following the TFF stage, the particles are passed into an active loading stage, where the doxorubicin is added into the flow stream and we are able to achieve high drug encapsulation (typically >95% drug encapsulation) in our continuous process. A soft sensor UV-Vis spectrometer using a multivariate, predictive algorithm is used to determine the doxorubicin encapsulation (Figure 3). After this stage, the material undergoes bioburden reduction and purification, at which point the drug substance is stored and ready for a fill line upon testing for quality.
Conclusion
Continuous processing for downstream processing of drug delivery vehicles such as liposomes enables enhanced control over the physical properties of the nanoparticles. The system developed at UConn is a high-throughput system and has the potential to be connected to an API processing stream, enabling a true end-to-end continuous manufacturing system. To-date, we have successfully performed a 3-hour run at a controlled state and can produce tens of liters of doxorubicin-loaded liposomes to RLD specifications using our current, lab-scale system. Lastly, continuous processing offers major cost savings, reduction in processing time, reduction in footprint and reduction in the number of batches; potentially resulting in higher quality drug products as a more economical approach to manufacturing.
References
- Gottlieb S, Woodcock J. FDA statement on FDA’s modern approach to advanced pharmaceutical manufacturing. https://www.fda.gov/news-events/pressannouncements/fda-statement-fdas-modern-approach-advanced-pharmaceuticalmanufacturing. Updated 2019. Accessed 03/26, 2020.
- Xing M, Yan F, Yu S, Shen P. Efficacy and cardiotoxicity of liposomal doxorubicin-based chemotherapy in advanced breast cancer: A meta-analysis of ten randomized controlled trials. PloS one. 2015;10(7):e0133569-e0133569. https://pubmed.ncbi.nlm.nih.gov/26204517; https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4512701/. doi:10.1371/journal.pone.0133569.
- Abraham SA, Waterhouse DN, Mayer LD, Cullis PR, Madden TD, Bally MB. The liposomal formulation of doxorubicin. Methods in Enzymology. 2005;391:71-97. doi: https://doi.org/10.1016/S0076-6879(05)91004-5.
- Wei X, Shamrakov D, Nudelman S, et al. Cardinal role of intraliposome doxorubicin-sulfate nanorod crystal in doxil properties and performance. ACS omega. 2018;3(3):2508-2517. https://www.ncbi.nlm.nih.gov/pubmed/30023837; https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6044617/. doi: 10.1021/acsomega.7b0123.
- Li X, Hirsh DJ, Cabral-Lilly D, et al. Doxorubicin physical state in solution and inside liposomes loaded via a pH gradient. Biochimica et Biophysica Acta (BBA) - Biomembranes. 1998;1415(1):23-40. doi: https://doi.org/10.1016/S0005-2736(98)00175-8.
- Wibroe PP, Ahmadvand D, Oghabian MA, Yaghmur A, Moghimi SM. An integrated assessment of morphology, size, and complement activation of the PEGylated liposomal doxorubicin products doxil®, caelyx®, DOXOrubicin, and SinaDoxosome. Journal of Controlled Release. 2016;221:1-8. doi: https://doi.org/10.1016/j.jconrel.2015.11.021.
- ICH. ICH Harmonised Tripartite Guideline: Development and manufacture of drug substances (chemical entities and biotechnological/biological entities). https://www.ich.org/page/quality-guidelines. Updated 2012. Accessed 03/26, 2020.
- Costa AP, Xu X, Khan MA, Burgess DJ. Liposome formation using a coaxial turbulent jet in co-fl ow. Pharm Res. 2016;33(2):404-416. https://doi.org/10.1007/s11095-015-1798-8. doi: 10.1007/s11095-015-1798-8.
- U.S. FDA. Guidance for industry: Quality considerations for continuous manufacturing guidance for industry. https://www.fda.gov/regulatory-information/search-fdaguidance-documents/quality-considerations-continuous-manufacturing. Updated 2019. Accessed 03/26, 2020.