Commercial and Regulatory Development Considerations for Nanomedicines Donna Cabral-Lilly and Lawrence D. Mayer |
CONTENTS
21.2 Regulatory Guidance and Reflection Papers
21.3 Preclinical Safety Studies
21.4 Chemistry Manufacturing and Controls
21.4.2 Nanomedicine Drug Product
21.4.2.1 Characterization of Nanomedicine Drug Product: Chemical/Content Specifications
21.4.2.2 Physicochemical Characterization of Nanomedicines
21.4.2.4 Changes to Manufacturing
21.4.2.5 In Vivo Characterization of Nanomedicine Performance
21.5 Timing for Development Activities
Preclinical proof-of-concept studies typically provide the scientific rationale for further development of a drug candidate in a clinical setting and are usually conducted as basic research using good scientific practices with the goal of publication and enhancing basic knowledge in the field of study. Translating this promising research into a full drug pharmaceutical/clinical development program requires activities aimed at determining the safety of the drug product and its components, the efficacy, and most relevant to this review, the ability to relate these biological effects to specific chemical and physical attributes of the final product (Zamboni et al. 2012). These pharmaceutical/clinical development studies are based on the laws governing human and veterinary medicines, as well as the expectations of the regulatory agencies responsible for marketing approval.
For the most part, regulatory requirements for a drug delivery system–based nanomedicine are the same as those for a conventional drug of the same class. There are, however, significant additional required studies for this class of pharmaceuticals; some are general and applicable to different types of delivery systems, while others address a specific type such as for topicals or pulmonary delivery platforms.
This chapter discusses the regulatory development pathway for injectable human medicines that use a nanotechnology carrier, such as for liposomal/nanoparticle products and colloidal iron/gold conjugate products. For the purposes of this review, such nanomedicines are defined as drug-containing macromolecular particulate assemblies in the size range of 20 nm to <10 μm, with the majority of discussion topics being most relevant to intravenously administrated products with a size range of 20–200 nm. Emphasis will be placed on considerations for chemistry, manufacturing, and controls (CMC) as well as aspects of pharmacology/toxicology requirements. A general timeline, from the Investigational New Drug (IND)/Clinical Trial Application (CTA) enabling studies, through marketing application submission, is also provided.
21.2 REGULATORY GUIDANCE AND REFLECTION PAPERS
In the United States, the development and marketing of drug products is regulated by the Food and Drug Administration (FDA) whose authority is established through the Code of Federal Regulations (CFR). The sections addressing drugs for human use can be found in 21 CFR, parts 1–100, 200s, and 300s. The Center for Drug Evaluation and Research of FDA provides a comprehensive library of guidance documents on its website (www.fda.gov/Drugs) that provide procedural and technical instructions for each stage of the product development cycle. In this context, a guidance exists specific for liposomal drug products (FDA 2015), in which many of the key elements are also applicable to other “particulate” nanomedicines such as micelles, nanoparticles, dendrimers, and colloidal particles. In 2007, FDA published a report of the findings of a Nanotechnology Task Force (FDA 2007) that highlighted the need to understand the physical and chemical properties of the delivery system and its components, which will require development of new analytical methods specific for each drug product. In 2012, FDA commissioner Margaret Hamburg summarized FDA approach to regulating nanotechnology products including nanomedicines (Hamburg 2012).
Similarly, in Europe, the European Medicines Agency (EMA) held a series of meetings on the use of nanotechnologies in medicine and published a general reflection paper in 2006 (EMEA 2006). The overall considerations mirror those of the FDA, where characterization of the drug product and justification of the appropriateness of the analytical methods used are expected. EMA has also published reflection papers on four types of intravenously administered nano-based carriers including liposomes (EMA 2013b), block copolymer micelles (EMA 2013a), surface coatings (e.g., PEGylation) (EMA 2013c), and iron-based nanocolloidal products (EMA 2015).
All of the basic product characterization regulatory criteria for conventional aqueous-based injectable pharmaceutical products (e.g., content, purity/related substances, pH, sterility, endotoxin content, particulate content, and sterility) apply to nanomedicines. However, given the macromolecular particulate assembly features of nanomedicines, significant attention must also be paid to the physical disposition of the administered drug in the milieu of the particle excipients, as well as how these physical attributes affect drug release from the carrier (bioavailability), in addition to the nascent biodistribution/PK properties of the carrier itself. These attributes provide an added level of biophysical characterization that must be considered not only for regulatory (CMC) purposes but also to ensure that drug product providing reliable exposure, efficacy, and toxicology profiles can be reproducibly manufactured in a commercial setting.
Each nanomedicine drug product will be reviewed by the regulatory agencies on a case-by-case basis, since even within the nanomedicine category of injectable drugs, different classes of drug delivery carriers exhibit unique features that must be evaluated. For example, the bioavailability and plasma half-life of paclitaxel bound to albumin particles may be very different from a taxane derivative covalently linked to PEGylated-copolymer micelles. Even within a specific class of carrier, different drug products may have unique features; for example, a liposomal product may have the active agent completely encapsulated inside the internal aqueous trapped volume (e.g., Doxil®/Caelyx®), or the active agent may be imbedded in the phospholipid membrane (e.g., Ambisome®). Each type of nanomedicine will have similar types of content and impurity quality control (QC) specifications, but each will also have a very distinct “fingerprint” of biophysical attributes such as particle size distribution and morphology, surface charge, aqueous/matrix partitioning, plasma/serum particle stability, and drug release kinetics, since these features will ultimately dictate the efficacy/safety profile in vivo.
Given the diverse physical and pharmacological properties that can be designed into nanomedicines, it is recommended that developers of such products seek regulatory advice early in the development process to gain agreement on required data and to ensure that gaps do not exist that may lead to clinical hold, or long delays, in development. Mechanisms for obtaining guidance from FDA include pre-IND meetings and guidance meetings of type A (stalled development), type B (pre-IND, end of Phase 2, pre-NDA), or type C (other general guidance), among others. Such meetings can be requested for single or multiple topics where the agency will respond to questions posed and may also remind sponsors of requirements not addressed in the questions. For EMA, advice may be obtained from the Scientific Advice Working Party and/or the Innovation Task Force. More recently, FDA and EMA have begun a process where a sponsor can obtain parallel advice from the agencies. This will save time, but does not guarantee that the agencies agree on the responses to all of the questions.
21.3 PRECLINICAL SAFETY STUDIES
As with any new drug, once a candidate formulation is established that can be produced under pharmaceutically acceptable conditions, toxicology studies are required to determine if any overt safety problems exist, as well as to establish basic pharmacokinetics and a safe starting dose before first-in-man studies may begin. Normally, these studies are done in two animal species with at least one being a non-rodent mammal, and several guidelines are available to aid in study design, based on disease indication and route of administration. If an active agent is a new molecular entity (NME), additional in vitro studies to determine carcinogenicity and genotoxicity will likely also be required. For these studies, final drug product supplies should be prepared using equipment and processes that will be utilized for production of Phase 1 clinical supplies and ideally, at the same Good Manufacturing Practices (GMP) manufacturer.
All such studies must be conducted according to Good Laboratory Practices (GLPs) (see 21CFR part 58). GLP requirements include that the personnel conducting the study are qualified by education and training, facilities are adequate and maintained, equipment is qualified and calibrated, and testing facilities and controls are adequate.
Written procedures must be in place, followed, and documented to assure that the laboratory is compliant with GLP standards. The study itself must be conducted according to a preapproved protocol that describes the objective, experimental design (e.g., number of animals, dose levels, and number of doses), samples to be collected, methods of analysis, and how results will be reported. Deviations from the approved protocol must be documented and approved, and a final report is written and approved. Analytical methods that measure drug content in plasma, urine, feces, and tissues must be validated per applicable guidelines (see FDA and International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use [ICH] guidelines on bioanalytical testing). Complete, accurate documentation is required, and records and samples are retained to at least the extent required in the regulations. A quality assurance unit reviews the study documents and certifies compliance with GLP standards. Full reports and summaries are submitted to FDA with the initial IND application, while only the summaries are submitted in the CTA in Canada and European countries.
For intravenous (i.v.) nanomedicine drug products, a delivery vehicle control arm is typically included in the preclinical animal studies to demonstrate the safety of the carrier itself. This is particularly important if the delivery vehicle is composed of novel excipients or exhibits physical structures that have not been tested previously in man. In such cases, more extensive toxicity studies may be warranted where chronic dosing for cumulative and/or reproductive toxicities are evaluated, as well as immunogenic toxicity effects. Dose-escalation (low, medium, high) toxicity assessments using single administration help to determine the dose to use for subsequent multiple-dose studies. Data obtained include observations of animal condition, weight, blood chemistries, plasma levels of active agent and known metabolites over time, immune reactions, tissue distribution, and full necropsy. The starting dose for the first-in-man trial (Phase 0 or Phase 1) is typically determined from the highest dose level that shows no adverse events in the most sensitive animal species.
Drug delivery carriers can often contain excipients for which no information on the safety of the compounds exists. It is likely that, at a minimum, separate carcinogenicity and genotoxicity studies will be required for each novel excipient. Drug carriers that are nonbiodegradable, as the intact carrier or components of a carrier, are of particular concern since they may remain in the body for long periods of time with the potential for adverse cumulative effects. Multiple-dose studies may be required, designed to show rates and sites of accumulation, as well as studies addressing immunogenicity, cumulative toxicity, dose dependence of toxicity, dose dependence of immunogenicity, and routes of elimination. Excipients never before administered to man may need long-term data in a suitable animal model, before human clinical studies can begin.
21.4 CHEMISTRY MANUFACTURING AND CONTROLS
The last 10–15 years have seen significant advances in technologies used to manufacture and test drug delivery–based nanomedicine products. Manufacturing equipment is now controlled by microprocessors capable of balancing temperature, pressure, shear forces, and rates of addition. Rapid mixing of multiple incoming streams of components can be achieved using readily available mixers with little or no modification. Most are also capable of clean-in-place and steam-in-place, thereby greatly reducing the risk of contamination from previous batches, other products, and microbes. On the testing side, liquid chromatography has progressed where separations of complex mixtures can be achieved in minutes and yield precise and accurate quantitation. In-line, real-time methods for characteristics such as pH and particle size are in development that will allow for more continuous processes. These advances have allowed the formulation and production of very complex drug carrier systems. Regulatory agencies are working diligently to keep up with advances as demonstrated by the nanotechnology initiatives for both FDA and EMA. Each agency relies on the innovator/sponsor to understand, explain, and justify every aspect of the new drug product and manufacturing process, and each product is evaluated individually with a high level of stringency. The required CMC information is separated into two broad categories: (1) drug substance and (2) nanomedicine drug product, which are described next.
The regulatory requirements for the active pharmaceutical ingredient (API) in a medicine are the same whether the API is a totally novel compound (NME), a derivative of an existing API, or an unmodified existing API. Who is responsible for providing this information is dependent on which category applies to the API.
An active agent is considered an NME if it has never been approved by a regulatory agency. The sponsor is required to supply all information on the compound, including details on the synthesis, isolation, purification, raw materials, in-process controls, final product testing, specifications, and stability during storage. Full characterization of the molecule and its synthesis/purification is required. For example, for a small molecule, NMR, FTIR, mass spectrometry, and other such methods are used to elucidate the chemical structure. In the case of biological (protein) products, the amino acid sequence, glycosylation, and other modifications must be determined and secondary and tertiary structures provided as they are known. Beyond this characterization, each lot of API must be tested per preapproved specifications that list the attribute and its acceptance criteria. It is expected that the acceptance criteria will be wide at the beginning of development (e.g., at Phase 1) but will be refined as more experience with the manufacturing and thus data become available, with final acceptance criteria set when the marketing application is filed. Specifications should always include the following attributes: description (color, physical form), identification (often one of the characterization tests such as IR), assay, and impurities (inorganic, organic, solvents, and metal contaminates).
Physicochemical properties (e.g., melting point, refractive index), particle size, and polymorphic form should be included as appropriate. Manufacturing must be done in compliance with current GMP expectations, and the quality assurance unit must confirm, in writing, compliance with these standards. This confirmation, called a Certificate of Compliance or Certificate of Conformance, is in addition to full testing per preapproved specifications and acceptance criteria where results are tabulated in a Certificate of Analysis.
When using a drug delivery carrier, it is often necessary to modify a known drug substance in order to stably incorporate it into the carrier in the final drug product. Prodrugs are such a case where, for example, a water-soluble active agent is attached to a long-chain fatty acid through a labile linker, to make the API more hydrophobic. The prodrug would then reside in a hydrophobic core of a micellar-type nanoparticle. Alternatively, the active agent might be linked to a long-chain polymer that covers the outside of the nanoparticle. From a regulatory perspective, the prodrug is considered an NME and is subject to all of the same controls as described for a novel compound.
The active agent, however, may be purchased commercially, and information on the synthesis and stability of this now starting material may be obtained from the commercial manufacturer, who preferably will have a drug master file (DMF) established with the regulatory agencies. Characterization of the prodrug should include data to support a proposed pathway to generate the active agent. Questions to answer include as follows: What bond gets hydrolyzed first? Does the linker stay with the “pro” portion? Under what conditions does the hydrolysis take place (e.g., low pH, or is an enzyme needed)? What is the known safety profile of the “pro” portion, linker, and hydrolysis intermediates? The answers to these questions will determine what additional safety studies are required.
The most straightforward regulatory path is for a commercially available drug substance that is already approved for use in a generic drug product. A novel delivery system is usually designed in this case to improve safety or enhance efficacy, and the API can be incorporated without modification. The information on manufacturing scheme, controls, characterization, and validation may be provided in a DMF submitted by the vendor and referenced, with permission, by the drug product sponsor. The sponsor is still ultimately responsible for the quality of the API and is expected to have quality agreements in place, and conduct routine audits of the manufacturer, but does not have to duplicate the information in the DMF.
Analytical methods for the drug substance must be appropriate for the intended use and quantitative methods are preferred. If a test is described in a general chapter of a compendia (United States Pharmacopeia [USP]; European Pharmacopeia [PhEur]), it is recommended to use the procedure as described in the chapter, unless it is necessary to deviate. For example, loss on drying is a routine test performed on powders, crystals, and other solid APIs. The test can be done by placing the material in a sealed container with a desiccant, or drying a known amount in an oven or other means of drying. Both the USP, in general Chapter <731>, and the PhEur, in Chapter 2.2.32, provide procedures for each type of drying method. Using one of these compendial procedures, instead of a method developed in house, eliminates the need to fully validate the procedure. A simple verification run using a sample with a known loss-on-drying value would be sufficient.
For NME and prodrugs, novel methods for assay and impurities will need to be developed. A chromatography method is expected, where the amount of API is quantitated using a reference standard. The impurity method should be able to separate process-related impurities and degradation products. A full method development report should be written that provides details on conditions tried and showing the suitability of the final chromatography and sample preparation conditions. At the early clinical development stage (Phase 1 and 2), the chromatography methods can be used as long as they meet system suitability requirements for linearity of the standard curve and repeatability. Full method validation per ICH guidelines is expected prior to Phase 3 clinical trials.
Prior to manufacturing process validation, each lot of API should be placed on stability at the recommended storage condition, and at one or more accelerated stability conditions. Testing intervals are narrow at the beginning of the study (e.g., monthly) until some data are available and experience gained with the compound. Sensitivity to light and moisture should also be tested. Drug substances are assigned retest dates, not expiration dates, and the retest interval is based on the stability data. Once sufficient data are obtained and the manufacturing process validated, then normally one lot per year is monitored for stability through the retest date and beyond. If changes to the manufacturing process are made, a risk assessment should be done to determine if revalidation is required, although some type of comparability study will be needed. Changes made prior to commercialization of the drug product will usually require fewer of these studies.
21.4.2 NANOMEDICINE DRUG PRODUCT
Nanomedicines are a class of drug delivery systems receiving increased attention from regulatory agencies worldwide. There are no set regulatory definitions provided by the agencies but screening tools are available: the product itself, or a component, has at least one dimension of 100 nm or less, although this size limit is not absolute. Natural products of these dimensions, including proteins, are not considered nanotechnology drug products; there must be a specifically engineered component. Consequently, although the protein albumin is not considered to be a nanotechnology product in itself, Abraxane® (paclitaxel protein–bound particles for injection) is characterized as a nanomedicine, with albumin as a carrier for the anticancer API paclitaxel.
To date, most nanomedicines, either approved or at the clinical stage of development, are carrier types, which have been described extensively in Chapter 5. Liposomes, micelles, and other nanoparticles serve as a macromolecular delivery vehicle into which the active agents are incorporated; the i.v. route of administration is predominant. Chemical and physical characterization is considered a key aspect of any regulatory submission for these drug products.
21.4.2.1 Characterization of Nanomedicine Drug Product: Chemical/Content Specifications
Table 21.1 provides an example of the specifications required for a drug product that contains a delivery carrier in the nanoscale range and is administered intravenously.
The amount of drug substance (assay) must be measured and acceptance criteria proposed and justified. Degradation studies must be done with the drug product to determine if any new related substances arise during the incorporation of the API in the nanomedicine carrier during manufacturing, and limits of any such impurities should be proposed. Further, the relative amount of the API associated with the carrier (relative to unentrapped drug) must be determined and limits proposed in the final drug product at release and during stability testing (typical acceptance range is 90%–95%). This procedure will include a step that physically separates the carrier with bound drug substance from drug substance that is free in the drug product solution. Using a liposomal product as an example, the dispersion is first diluted 10-fold in an isotonic buffer and then placed into the sample portion of a spin column with a 10,000 molecular weight cutoff and spun for 10 minutes at a temperature below the phase transition temperature of the liposome membrane. The amount of API in the filtrate is quantitated and this amount relative to the total API in the sample reported as % unencapsulated API. Uniformity of dosage per USP <905> or PhEur 2.9.40 may also be required.
The components of the drug delivery carrier are considered critical excipients and have to be controlled in the drug product much in the same way as the drug substance. The amount of each carrier component must be measured by a suitable assay and acceptance criteria listed in the drug product specifications. Uniformity of dosage should be included if it is performed for the API. An assay (or multiple assays) must be developed that can detect and separate possible degradation products for each carrier component, and which must also be able to separate and quantitate the respective degradation products for multiple components present in the final product. The method must be at least semiquantitative, with the limit of detection established and an upper limit proposed for each degradation product. For example, consider a liposomal product that has a membrane composed of the phospholipid palmitoyl-oleoyl-phosphatidylcholine (POPC) and cholesterol. The oleic acid chain contains a double bond that may oxidize, so a method that measures oxidation products is required. The oleic acyl chain may hydrolyze, requiring a method to measure either the oleic acid or the lyso-PC. Cholesterol can also oxidize to form 7-ketocholesterol and this degradation product must also be measured.
TABLE 21.1
Example of Drug Product Specifications for a Delivery Carrier Nanomedicine
Attribute |
Acceptance Criteria |
Appearance |
Physical state, color |
Identification |
Orthogonal method—corresponds to reference standard |
Active substance |
|
Identification |
HPLC retention time same as standard |
% Associated with carrier |
Lower limit, e.g., NLT 90% |
Assay |
Target ± x% (start high ±10%) |
Impurities |
Upper limit, e.g., |
Known 1 |
NMT 0.2% |
Uniformity |
Meets USP <905> |
Carrier component 1 |
|
Assay |
Target ± x% (start high ± 15%) |
Impurities |
Upper limit, NMT x% |
Uniformity |
Meets USP <905> |
Carrier component 2 |
|
Assay |
Target ± x% (start high ± 15%) |
Impurities |
Upper limit, NMT x% |
Uniformity |
Meets USP <905> |
Carrier component 3 |
|
Assay |
Target ± x% (start high ± 15%) |
Impurities |
Upper limit, NMT x% |
Uniformity |
Meets USP <905> |
Particle size distribution |
|
Mean diameter or D50 |
Target range, 80–100 nm |
D10 |
Lower limit, NLT 50 nm |
D90 |
Upper limit, NMT 150 nm |
D99 |
Upper limit for safety, NMT 500 nm |
In vitro release |
|
At incubation time 1 |
Range |
At incubation time 2 |
Range |
At incubation time 3 |
Range |
At 24-hour incubation |
Range but expected to be at least 70% |
At full release condition |
Lower limit, NLT 90% |
Assay for buffer or bulking agent |
|
pH |
Range |
Osmolality |
Range |
Viscosity |
Range |
Other |
|
Residual solvents |
|
Process solvent 1 |
All as upper limit based on ICH Q3 |
Process solvent 2 |
|
Etc. |
|
Residual metals (if not controlled completely in raw materials) |
All as upper limit based on PhEur 2.4.20 |
Particulate matter (for injectables) |
Meets USP <788> or PhEur 2.9.19 |
Endotoxin |
Limit based on USP <85> or PhEur 2.6.14 |
Sterility |
No growth. Test method per USP <71> or PhEur 2.6.1 |
Abbreviation: NMT, not more than; NLT, not less than. |
|
Particle size distribution is of particular interest to regulatory agencies because a change here may affect the safety or performance of the nanomedicine due to altered pharmacokinetic and tissue distribution properties. Mean or median (D50) diameter, as well as the size at the low end (e.g., D10) and high end of the distribution (e.g., D90), are reported. Acceptance criteria may be listed as a range for each measurement, but it is also acceptable to use limit specifications for the high and low ends of the distribution. The procedure and equipment used, e.g., dynamic light scattering or particle counting, must be capable of detecting particles of the full size range that exists in the product. The results will be slightly different between procedure types, and even using the same type of analysis, but with hardware and software from multiple vendors. Therefore, it is recommended to determine the most appropriate procedure while still at the preclinical stage of development, provide justification for the decision, and then use the same procedure and vendor through commercialization (with extensive measures incorporated for method/equipment calibration, to minimize method variability).
For other excipients in the final product, a direct test to quantitate them, or an indirect test where the result is dependent on the amount of excipient present, must be established. For example, if maltose is used as a stabilizing agent, osmolality could be measured instead of the actual maltose content. Tests such as appearance and pH are common, and if the product is lyophilized, the reconstitution time must be tested and acceptance limits proposed. Residual amounts of processing agents must be quantitated. These include solvents, chelation agents, and water for a lyophilized product. All drug products for i.v. administration must conform to the USP and PhEur requirements for these types of drugs, and tests for endotoxin, sterility, and particulate matter are required.
21.4.2.2 Physicochemical Characterization of Nanomedicines
In addition to the proposed product specifications, the biophysical properties of nanomedicines are considered critical to their safety and efficacy. This additional characterization is initiated during the early stages of drug development. It is necessary because differences in product performance can occur in vivo due to alterations in the physical organization of the drug and/or excipients in a nanomedicine, even if the ingredient content and purity specifications are unchanged. Biophysical characterization is particularly useful when changes to the manufacturing process, or composition of the drug product, are made. For liposomal products, the core properties include vesicle morphology and lamellarity, zeta potential, trapped volume, and membrane phase transition temperature, as well as the physical state of the API inside the liposome. Each liposomal product may have specific testing based on the formulation. For example, the intrinsic fluorescence of an anthracycline may be altered/quenched when encapsulated into a liposome. Product-specific characterization focuses on drug–excipient interaction, drug-loading mechanisms, drug–drug interactions, and physical disposition of the drug in the carrier.
A specific example is the characterization of CPX-1 (irinotecan/floxuridine) liposome injection, a clinical stage liposomal nanomedicine intended for treatment of advanced colorectal cancer (Dicko et al. 2008). The manufacturing process for CPX-1 liposomes involves the coencapsulation of the APIs irinotecan HCl and floxuridine, at a 1:1 molar ratio, into preformed liposomes. As presented in Figure 21.1, data from proton NMR spectroscopy, as well as an in vitro release (IVR) assay, were used to help characterize and determine the final liposomal drug product formulation. The aromatic region of the 1H NMR spectrum of the drug product shows three peaks from irinotecan. When the buffer inside the liposomes is sodium gluconate–triethanolamine (NaGluc-TEA), the 1H peaks are quite broad, suggesting the self-association of irinotecan into large aggregates (Figure 21.1). However, the presence of entrapped copper gluconate–triethanolamine (CuGluc-TEA) buffer inside the liposomes results in 1H peaks that are much sharper, suggesting an interaction between irinotecan and copper, which prevents irinotecan self-aggregation (Figure 21.1). Large aggregates of irinotecan reduce membrane permeability and liposomal drug release. This was borne out by the results of the IVR assay, which showed that coordinated liposomal drug release was disrupted by removal of the copper gluconate (Figure 21.1). The liposomes with NaGluc-TEA as buffer demonstrated a slower and uncoordinated release of irinotecan relative to floxuridine and loss of the synergistic drug/drug ratio. In contrast, for liposomes with CuGluc-TEA as the entrapped buffer, the two drug substances were released from the liposomes at the same rate and at a fixed, synergistic 1:1 molar ratio. Taken together, the results strongly suggest an interaction between the API (irinotecan) and the excipient (copper), in the drug–liposome product.
FIGURE 21.1 Identifying physicochemical features that relate to performance in CPX-1 (irinotecan/floxuridine) liposome injection. The 1H NMR spectra of the aromatic region for irinotecan show broadening for the formulation without copper in the buffer inside the liposome (left top), compared to the same region in a formulation that includes copper in the entrapped buffer (left bottom). The results from an in vitro release assay (right) show that release of irinotecan is dependent on the presence/absence of copper. (With kind permission from Springer Science+Business Media: Pharm. Res., Intra and inter-molecular interactions dictate the aggregation state of irinotecan co-encapsulated with floxuridine inside liposomes, 25, 2008, 1702–1713, Dicko, A., Frazier, A.A., Liboiron, B.D. et al.)
An example of the full biophysical characterization for a liposomal product has been reported for CPX-351 (cytarabine/daunorubicin) liposome injection, a late clinical stage liposomal drug combination product in which two drug substances are encapsulated in a single liposome (Dicko et al. 2010). These studies demonstrate the importance of extensive biophysical study of liposomal drug products, to elucidate the key physicochemical properties that may impact their in vivo performance.
It should be noted that similar principles apply to synthetic polymer-based nanomedicines. For block copolymer micelle drug products, the biophysical characterization list in the EMA reflection paper includes morphology, zeta potential, association number, critical micelle concentration, surface properties, and physical state of the active substance. If the carrier has a surface coating, e.g., PEGylation, then (1) surface coverage heterogeneity, (2) conformational state of the coating molecules, and (3) stability of the coating under conditions of storage and use, should all be characterized. Biophysical characteristics are particularly important during scale-up from preclinical to Phase 2 and later to Phase 3/commercial where changes in equipment and manufacturing sites can introduce subtle alterations in the physical properties of a nanomedicine formulation, despite providing the desired content-related specifications. This is highlighted in FDA and EMA documents on nanomedicines where thorough biophysical characterization is advised for such manufacturing transitions to ensure reliability in correlations of preclinical/clinical results during the course of product development toward commercialization.
As alluded to earlier, a key assay for any drug delivery carrier product is a reliable and discriminatory IVR assay. Focus on this assay has significantly increased within the regulatory agencies over recent years, largely due to the fact that drug release kinetics of a nanomedicine may be one of the most important features dictating its efficacy/safety profile. This awareness has been heightened with the emergence of generic versions of liposomal anticancer agents such as Doxil®, where concerns about bioequivalence impacting in vivo performance and patient outcomes require additional characterization diligence. The IVR test is modeled after the requirements for modified-/extended-release solid oral dosage forms and is based on the premise that the API is released from the delivery carrier over time in the bloodstream. For the IVR test, the drug product is incubated in suitable media, the samples are taken at selected intervals, and the free/released API is physically separated from the drug delivery carrier. A quantitative method, usually HPLC, is used to assay the free/released API.
Method development can be extensive in order to identify conditions where the drug dissociates from the carrier, but the carrier does not completely disintegrate, in a manner that is unrepresentative of the expected in vivo situation. It is usually necessary to develop incubation media that either perturbs the carrier (e.g., an alcohol or detergent) such that the API is not stably entrapped, or allows for hydrolysis to occur (e.g., low pH, added enzyme) so that a covalently bound API is released. The method must be discriminatory, and the regulatory agencies expect to see data demonstrating that the IVR assay is capable of identifying a change in the drug product that may affect safety or efficacy in patients. Often, formulation variants are generated to demonstrate the predictive nature of the assay. For example, a liposome product can be prepared where 10% of the phospholipid degradation product, lyso-PC, is added to the normal POPC/cholesterol composition when the liposomes are made. The lyso-PC disrupts the packing of the membrane making it leaky such that an encapsulated API will diffuse out of the entrapped volume into the incubation media. A suitable IVR assay will show continuous release of the API into the incubation media where at least 70% of the API is in free/released form after 24 hours and eventually more than 90% of the drug substance becomes free in the media. An ideal IVR assay will be one able to show an in vitro–in vivo discriminatory comparability. Whereas most biophysical characterization is used for development purposes only, the IVR assay is included in the drug product specifications and routinely tested at batch release and during the stability program.
The manufacture of nanomedicine drug delivery carrier products is typically a multistep process, and each step needs to be fully understood, characterized, and controlled. The expectations for how this control is demonstrated have changed with the introduction of process analytical technology initiatives and the issuance of the ICH guidance documents on Pharmaceutical Development (Q8), Quality Risk Management (Q9), and Quality Management System (Q10). A science-based and risk-based approach is used to evaluate changes and results throughout the life cycle of the drug product.
Critical quality attributes (CQA) are the features of the drug product that have an impact on the product performance—safety, efficacy, and stability. A proposed list of CQAs should be defined early (ideally at the initial IND or CTA submission) and revised as more experience with both the manufacturing process and product use is obtained. The CQAs are linked to critical process parameters (CPP): conditions during manufacturing that, if changed, would likely result in not achieving a CQA and hence resulting in a drug product that has altered characteristics and performance. A CQA specific for drug carriers is the attribute(s) that keeps the API stably associated with the carrier until it reaches the site in vivo where, by design, the API is released and becomes bioavailable. General examples include particle size distribution, surface coating density, and API-to-lipid/polymer/colloid ratio. The CQAs of the product would be assured by determining the CPPs for liposome preparation, drug substance encapsulation, lyophilization, and feasibility of reworking.
Let us use the POPC/cholesterol liposome with encapsulated API to provide a more specific example. We define a CQA to be particle size distribution, where the mean diameter of the liposome is 100 nm, D10 is ≥60 nm, and D90 is ≤200 nm, based on the following:
• Liposomes with diameters of about 100 nm have extended plasma circulation times in vivo and can be taken up by cells with a leaky vasculature, as occurs in fast growing tumors.
• The size range of 40–50 nm is near the limit at which liposomes are stable. These smaller-sized particles have different plasma circulation times, and, because of the decreased trapped volume, they encapsulate less API. Therefore, D10 assures that there are not too many of these very small liposomes in the drug product.
• Liposomes much above 200 nm also have altered plasma circulation times (typically more rapid elimination from the plasma) and may be too large to be taken up by a cell by passive mechanisms; consequently, less API gets delivered to the site of action. D90 assures limited presence of such very large liposomes in the drug product.
The next step is to determine where in the manufacturing process the liposome size becomes defined. For our example product, the liposomes form during an emulsion process, and then the size is refined by extrusion through membranes with a defined and homogeneous pore size of 100 nm. Process parameters include shear, temperature, and time. A series of experiments are designed to determine the effect of altering these three process parameters. A classical approach may be used where one parameter is changed at a time, or a design of experiments (DOE) approach may be used where multiple parameters are altered at the same time. The latter falls into the Quality by Design (QbD) described in the ICH guidelines. Both approaches are acceptable to the agencies with justification. The results of the experiments indicate that high shear can cause small liposomes to form and hence the D10 specification would not be met. The results also show that changes to time and temperature have little effect on particle size distribution. We then address the specific process settings that control shear, i.e., mixing rate, mixer configuration, sparge rate, and determine, for a specific batch size and defined equipment type, which two marine-type impellers fixed to a single central shaft, stirring at 20–30 rpm with nitrogen sparge set at 40–80 L/min, provide shear forces that ultimately result in the desired particle size. We now have CPPs and defined operating ranges with data to support them.
The numerous components of liposome products (multiple membrane lipids, additional excipients, possibly multiple active agents), as well as the multistep, multiday manufacturing process, pose challenges to performing development and validation studies using the QbD approach when the process is viewed as a whole; the design space would be extremely complex. It should be possible, however, to obtain a good understanding by studying each step in the manufacturing campaign separately, using enhanced science-based and process-based protocol designs.
These enhanced studies fall into the general QbD principles, which use a systematic approach to product development. Objectives are predefined and are based on a thorough understanding of the product and manufacturing process that relate raw material attributes and process controls to the CQAs. The process begins at the very early stages of product development by creating a target product profile. The formulation, manufacturing process, and controls are designed to meet this product profile, and these parameters are continually monitored throughout the product life cycle, including after commercialization. Trends are identified and the process improved as needed, to achieve product quality standards and performance.
Multivariate experiments using DOE and statistical analysis of results are conducted to understand the interactions of input variables on the overall process and product attributes. Examples of ways to present results from multivariate studies are provided in ICH guidance Q8 (R2).
Ideally, a design space is defined that describes the multidimensional interactions where product quality is assured. Changes within the design space are not considered critical and can be made without approval from the regulatory agencies. The key feature here is the use of multivariate testing; acceptable ranges that are found by testing only one parameter at a time may not provide information on the interactions between process parameters. If using only univariate studies, a design space cannot be described, and changes to the manufacturing process will have to be submitted to the appropriate regulatory agency.
21.4.2.4 Changes to Manufacturing
It is understood that the manufacturing process will undergo changes during development, with the final process not defined until validation studies are completed, prior to the filing of the marketing application. Even then, the QbD life cycle approach to development provides for continual process improvements after commercialization. The more that is understood about the manufacturing process, especially the CQAs and related CPPs, the more straightforward the pathway is to getting the regulatory agencies to agree that the change can be made.
A key aspect here is the ability to clearly identify and establish quantitative/semiquantitative assays that will detect alterations in biophysical properties of nanomedicine products that could lead to altered in vivo performance despite exhibiting identical content-related specifications. Consequently, establishing biophysical parameters and ranges that ensure reproducible blood circulation times of the delivery vehicle, and drug release rates from the vehicle after injection, are critical to the development of a nanomedicine product. The reason for this is that while content of components in the formulation typically can be readily controlled (and assayed) going from lab scale to toxicology/Phase 1 scale and ultimately to Phase 3/commercial scale, the physical disposition of the delivery vehicle and encapsulated drug(s) require more complicated and product-specific biophysical assays, which may require more than 6 months to develop and qualify. The utility of the biophysical assays is the most important during transitions in production scale, where modifications in the process, and most certainly in the equipment used, can result in altered physicochemical properties. Alterations in the physicochemical properties may occur despite generating a final drug product with no changes in content and purity of ingredients. This understanding is especially important for drug delivery carrier products, and the regulatory agencies are requesting information earlier in development. One example is the IVR assay; previously, this assay was developed and used starting in Phase 3, whereas currently FDA has suggested that the assay be used starting in Phase 1.
21.4.2.5 In Vivo Characterization of Nanomedicine Performance
While the IVR assay provides an in vitro QC-compatible method to assess drug bioavailability for a nanomedicine, ultimately this characteristic should be evaluated in patients. This requires the establishment of an assay procedure that is capable of handling plasma/serum from treated patients and separating nanomedicine carrier-associated drug from drug that has been released from the carrier. The regulatory agencies have placed increasing importance on this analysis, with the desire to differentiate the pharmacodynamics relationships for toxicity and efficacy responses as they may relate to encapsulated vs. free/released drug.
In this regard, many different assays have been developed, based on solid-phase extraction columns or ultrafiltration cartridges. The common observation arising from such analyses in patients treated with nanomedicines is that the free drug concentration is typically so small (due to the rapid tissue distribution of most free drugs) that plasma drug exposure of total drug is equivalent to encapsulated drug. Nonetheless, the development of more sensitive analysis procedures may be important in refining the elucidation of pharmacodynamics relationships for nanomedicines and warrants additional refinement efforts. For example, separation methods utilizing equilibrium conditions (e.g., ultrafiltration cartridges) appear to minimize in-process drug release and overestimations of free drug plasma concentrations that can arise with nonequilibrium methods such as solid-phase extraction columns (Mayer and St. Onge 1995).
21.5 TIMING FOR DEVELOPMENT ACTIVITIES
In the development of a drug product, the amount of information available will be limited in the early stage. The manufacturing processes, as well as analytical methods, can be revised and improved as the size of the database increases. During the clinical stage of development, CMC changes are reported in the United States to FDA as part of the annual reports for each active IND application. Substantive changes that require review and consent prior to implementation are filed as Quality Amendments to each regulatory agency where a clinical trial is taking place. As a reference, Figure 21.2 provides a general timeline for the progress of CMC development.
The identification and control of key features that dictate drug pharmacokinetics and tissue distribution for a nanomedicine drug delivery vehicle must be initiated as soon as evidence is generated that supports development as a potential product candidate. This is due to the fact that the efficacy/toxicity profile of nanomedicines can be impacted by physicochemical properties of the delivery vehicle to the same extent, or greater, than content and purity of the drugs and excipients in the final drug product. Therefore, key performance-related biophysical assays should be in place by the time that batches are being produced at the GLP toxicology/Phase 1 scale. Timing is an important consideration for assay methods such as IVR and free vs. encapsulated drug, where it can take up to 12 months to refine assay conditions to where they provide acceptable accuracy, precision, reproducibility, and relevance to product in vivo performance. Validation studies for the analytical methods, the manufacturing process, and sterility assurance are spaced throughout development. Timed well, the CMC studies will be done shortly before completion of the clinical trials and may be submitted prior to, or at the time of, the safety and efficacy results in the marketing application.
FIGURE 21.2 Overview of timeline for chemistry, manufacturing, and controls development studies from preclinical stage to marketing application.
Dicko, A., A.A. Frazier, B.D. Liboiron et al. 2008. Intra and inter-molecular interactions dictate the aggregation state of Irinotecan co-encapsulated with floxuridine inside liposomes. Pharm. Res. 25:1702–1713.
Dicko, A., S. Kwak, A.A. Frazier et al. 2010. Biophysical characterization of a liposomal formulation of cytarabine and daunorubicin. Int. J. Pharm. 391:248–259.
EMA. January 2013a. Joint MHLW/EMA reflection paper on the development of block copolymer micelle medicinal products. EMA/CHMP/13099/2013. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2013/02/WC500138390.pdf.
EMA. February 2013b. Reflection paper on the data requirements for intravenous liposomal products developed with reference to an innovator liposomal product. EMA/CHMP/806058/2009/Rev.02. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2013/03/WC500140351.pdf.
EMA. June 2013c. Reflection paper on surface coatings: General issues for consideration regarding parenteral administration of coated nanomedicine products. EMA/325027/2013. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2013/08/WC500147874.pdf.
EMA. March 2015. Reflection paper on the data requirements for intravenous iron-based nano-colloidal products developed with reference to an innovator medicinal product. EMA/CHMP/SWP/620008/2012. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2015/03/WC500184922.pdf.
EMEA. June 2006. Reflection paper on nanotechnology-based medicinal products for human use. EMEA/CHMP/79769/2006. http://www.ema.europa.eu/docs/en_GB/document_library/Regulatory_and_procedural_guideline/2010/01/WC500069728.pdf.
FDA. October 2015. Guidance for industry. Liposome drug products. Chemistry, manufacturing and controls; human pharmacokinetics and bioavailability; and labeling documentation. www.fda.gov/ucm/groups/fdagov-public/@fdagov-drugs-gen/documents/document/ucm070570.pdf.
FDA. July 2007. Nanotechnology. A report of the U.S. Food and Drug Administration Nanotechnology Task Force. http://www.fda.gov/downloads/ScienceResearch/SpecialTopics/Nanotechnology/ucm110856.pdf.
Hamburg, M. 2012. FDA’s approach to regulation of products of nanotechnology. Science 336:299–300.
Mayer, L.D. and G. St. Onge. 1995. Determination of free and liposome-associated doxorubicin and vincristine levels in plasma under equilibrium conditions employing ultrafiltration techniques. Anal. Biochem. 232:149–157.
Zamboni, W.C., V. Torchilin, A.K. Patri et al. 2012. Best practices in cancer nanotechnology: Perspective form NCI nanotechnology alliance. Clin. Cancer Res. 18:3229–3241.