CHALLENGES IN DEVELOPING PHARMACEUTICAL LIPOSOME FORMULATIONS
In the early stages of clinical development, research scientists typically determine the requirements for the drug product (e.g., therapeutic window, frequency of administration, patient population constraints) and the various formulation and delivery options that may best meet those diverse product requirement specifications. If a liposome formulation is utilized, it should be designed so that the in vivo rate of drug release results in drug levels falling within the therapeutic window until the next administration event: high enough to provide therapeutic benefit but not so high that pernicious side effects result. An in vitro release (IVR) assay can be utilized to evaluate various liposome compositions and rank order their relative release rates; however, it is unlikely that the results of the IVR assay alone can be used to select a single composition to move into clinical development. Instead, the IVR assay is typically used to exclude liposome formulations that are unacceptable (e.g., possess a high initial burst or do not release at all) and narrow down the list of compositions to be evaluated in a preclinical setting. If the relative release rates in the in vivo preclinical setting mirror that in the IVR assay, then the IVR assay may be utilized more fully to fine-tune formulation compositions and understand the effects of manufacturing changes (e.g., an increase in the scale of operations or a change in a process step) or CMC changes (e.g., alternative excipient suppliers) on the release profile . Preclinical PK data can be used to further exclude undesirable formulations, but depending upon the choice of animal model and the route of delivery, PK data in animals may not be fully predictive of the disposition in humans. Because of this uncertainty, the lead formulation(s) will need to be tested in human clinical studies to provide assurance that the in vivo PK meets the design requirements.
The FDA has specific recommendations for additional content to be provided in the regulatory submission of a liposomal product due to its unique technical aspects . These include characterization of the following physicochemical properties: the liposome structure, liposome integrity, liposome morphology and lamellarity, the encapsulated drug, the viscosity, the surface charge or zeta potential, the drug leakage rate, the vesicle size distribution, the lipid composition, and the IVR of drug using an appropriate physiological medium . Assurance also needs to be provided that the physicochemical properties do not change over the shelf life of the liposomal product .
Focusing first on the IVR methodology, the test method should measure the amount of the released (free) drug over a time scale to cover the complete release of the drug [61-63]. In practice, this may be challenging for a liposomal product because it can often be difficult to differentiate the free drug from that remaining encapsulated in the liposomes by routine analytical methods. There are three general methodologies that have been developed to measure the IVR of a drug from liposomal products . For drugs that fluoresce, the change in fluorescence can be translated into a drug release rate—this is termed an in situ method as the amount of fluorescence is directly measured without further sample manipulation. The advantage of the in situ IVR method is its simplicity (rapid data output), but unfortunately, most drugs do not have properties that are amenable to measurement in situ. Thus, a number of specific dissolution or IVR test methodologies have been developed for liposomal products [61-63]. A second class of methods utilizes membrane dialysis to physically separate the released drug from the encapsulated drug [61,63]. Dialysis methods are appropriate when the rate of drug release from the liposomes is slow compared to the time to diffuse across the membrane—otherwise, the measured release rate may not reflect the true release rate from the liposomes . The third class of IVR methods is the “sample and separate” method: samples are periodically removed from the IVR vessel and the free drug is separated from the encapsulated drug (often by chromatography, centrifugation, or filtration), thus allowing for quantitation by HPLC or another analytical assay .
The liposome formulation should also be robust to the delivery method and procedure; for example, a formulation that requires dilution for IV administration should be evaluated to ensure that it retains its physical properties at the lower (diluted) concentration and in contact with the materials in the IV bag and line. For an inhaled liposomal product, the liposomal formulation will be inhaled into the lung. The in vivo release in the lung fluid may be very different from that for an IV product that is injected into the bloodstream. The relative dilution into the bodily fluids and the composition of the biological milieu are different; for example, the dilution in the lung fluid (ca. 50 mL) is one hundredth that of dilution into the bloodstream (ca. 5 L). The site of deposition in the lung, for example, peripheral versus bronchial airways, and the presence of lung disease will also affect the fluid volume and composition . Ideally the IVR assay should be developed using an appropriate simulated physiological medium or human plasma to mimic the in vivo situation . However, there is no recognized standard simulated lung fluid [61-63]. Many of the simulated lung fluids in the published literature do not contain components that are naturally present in the lung, which may interact with liposomes, for example, proteins, surfactant, or lipids, and result in modulation of the rate of drug release [61-63]. For many of these simulated lung fluids, the IVR data may have little relevance to the in vivo release kinetics .
One IVR assay was developed specifically for an inhaled liposomal product utilizing bovine serum as the release vehicle . Bovine serum was chosen because it is biological in composition, it is inexpensive to obtain in large quantities from established suppliers, and the components in serum may be relevant to the release mechanisms in the lung . Most serum proteins including albumin, trypsin, and ovalbumin do not induce release from liposomes . The components in serum that do have a destabilizing effect on liposomes, causing drug release, are lipoproteins and apolipoproteins [64,65]. All serum lipoproteins and apolipoproteins induce drug release from liposomes and these components are present in lung fluid at about half the concentration in serum and thus are relevant as release agents for inhaled liposomal products [63,64]. In this IVR assay, after the liposomes are diluted into bovine serum and incubated for varying times at 37°C, the released drug is separated from
FIGURE 8.1 Schematic of the centrifugal filtration device to separate free drug (dots) from liposomally encapsulated drug (circles containing dots). (Reprinted from Cipolla, D. et al., J. Pharm. Sci., 103(1), 314, 2014. With permission.)
the encapsulated drug using centrifugal filtration  (Figure 8.1). This IVR assay has been used to discriminate between liposome formulations with different release rates [26,27,30,66]. However, the drawback to utilizing bovine serum or any other biological source as a release medium is at least twofold: there is the possibility of a supply disruption and the potency may change batch to batch, for example, if the amount of lipoproteins and apolipoproteins (which are not typically measured or reported) were to change in the bovine serum.
During pharmaceutical development, the critical quality attributes (CQAs) of a liposomal product, like all pharmaceutical products, must be identified to establish product quality . The CQAs of the drug product represent the physical, chemical, biological, or microbiological properties or characteristics that should be monitored and verified to remain within an appropriate range to provide assurance of product purity, strength, and performance at the time of release and on stability. In addition to the standard CQAs that are typical for most pharmaceutical products, a liposomal product may require additional CQAs to ensure maintenance of liposome integrity, vesicle size, and drug release rate, among other functional attributes .
Analytical assays need to be established to verify that the drug product’s CQAs remain within the acceptable range over the shelf life required to cover manufacturing, distribution, and use of the product by the patient—most marketing organizations typically require a shelf life of at least 18 months and preferably 2 years or more to ensure that the supply chain can be effectively managed. A number of analytical methods and procedures have been developed that are specific to liposomal products and many are described in these references [68-71].
In addition to verifying that the active pharmaceutical ingredient retains its purity within acceptable levels, for liposome products, the functional excipients (e.g., lipids) must also be monitored to ensure that degradation products are not formed that compromise the safety or performance of the drug product . While lipid oxidation is a possible degradation route , the use of high-purity excipients and the elimination of oxidants from the manufacturing processes can typically address this concern [32,70]. For many liposome products, lipid hydrolysis remains the primary degradation mechanism that limits shelf life [69,70,73,74]. Lipid hydrolysis can be either base or acid catalyzed to form free fatty acid and lysophospholipid; thus, most liposome products are formulated near neutral pH to minimize lipid hydrolysis . However, liposome formulations may be able to accommodate a significant amount of lipid hydrolysis (e.g., 10%) without compromise to the liposome integrity or permeability . The presence of cholesterol in the lipid bilayer counteracts the destabilizing effect of lysophospholipids so liposomal products containing cholesterol may be able to accommodate even greater lipid hydrolysis without deleterious effects on functionality . The safety of the lipid degradation products will also need to be verified for each liposomal product.