TECHNICAL ASPECTS RELATED TO THE MANUFACTURE OF INJECTABLE PHARMACEUTICAL PRODUCTS – FROM R&D TO PRODUCTION Mafalda Filipa Machado Nunes Thesis to obtain the Master of Science Degree in Pharmaceutical Engineering Supervisors: Professor António José Leitão Neves Almeida Professor José Monteiro Cardoso de Menezes Examination Committee: Chairperson: Professor João Carlos Moura Bordado Supervisor: Professor António José Leitão Neves Almeida Members of the Committee: Professora Helena Maria Cabral Marques Dra. Ana Paula Gouveia Antunes Gageiro December 2015
85
Embed
TECHNICAL ASPECTS RELATED TO THE MANUFACTURE OF ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
TECHNICAL ASPECTS RELATED TO THE MANUFACTURE
OF INJECTABLE PHARMACEUTICAL PRODUCTS –
FROM R&D TO PRODUCTION
Mafalda Filipa Machado Nunes
Thesis to obtain the Master of Science Degree in
Pharmaceutical Engineering
Supervisors:
Professor António José Leitão Neves Almeida
Professor José Monteiro Cardoso de Menezes
Examination Committee:
Chairperson: Professor João Carlos Moura Bordado
Supervisor: Professor António José Leitão Neves Almeida
Members of the Committee: Professora Helena Maria Cabral Marques
Dra. Ana Paula Gouveia Antunes Gageiro
December 2015
2
TECHNICAL ASPECTS RELATED TO THE MANUFACTURE
OF INJECTABLE PHARMACEUTICAL PRODUCTS –
FROM R&D TO PRODUCTION
Mafalda Filipa Machado Nunes
Thesis to obtain the Master of Science Degree in
Pharmaceutical Engineering
December 2015
3
ABSTRACT
The master thesis subject falls within the scope of the work developed in the Technical Services
department of Hikma Farmacêutica S.A. and its aim is to assess the technical aspects and activities
related to the transfer of an injectable pharmaceutical product from development phase to production.
After development of a new pharmaceutical product, several technical aspects need to be evaluated
and numerous validation activities need to be performed prior to start routine production and
commercialization. Technology transfer involves transfer of product and process knowledge to achieve
product realization and includes all the activities required for successful progress from pharmaceutical
development (R&D) to production (for new products) or from one manufacturing site to another (for
marketed products).
Process validation is part of technology transfer and is used to demonstrate that the manufacturing
process developed, operated within established parameters, can consistently deliver the intended
product. A proper correlation between process inputs, their associated manufacturing controls and
process outputs is crucial to successful process validation. Since there are several inputs, outputs and
controls associated with each manufacturing operation, a systematic approach that emphasizes product
and process understanding, based on quality risk management, is crucial to identify and to evaluate the
process validation activities to be performed during technology transfer.
Keywords: injectable products, technology transfer, process validation
4
RESUMO
O assunto da dissertação de mestrado insere-se no âmbito do trabalho desenvolvido no departamento
Technical Services da Hikma Farmacêutica S.A. e o seu objetivo é avaliar os aspetos técnicos e as
atividades relacionadas com a transferência de um produto farmacêutico injetável da fase de
desenvolvimento para a fase de produção. Após o desenvolvimento de um novo produto farmacêutico,
diversos aspetos técnicos precisam de ser avaliados e numerosas atividades de validação têm que ser
realizadas antes de se iniciar a produção de rotina e a comercialização. A transferência de tecnologia
envolve a transferência de conhecimentos acerca do produto e do processo, de modo a permitir a
obtenção do produto e inclui todas as atividades necessárias para progredir da fase de desenvolvimento
para a fase de produção (no caso de novos produtos) ou de um local de fabrico para outro (no caso de
produtos já existentes).
A validação de processo faz parte da transferência de tecnologia e tem como intuito demonstrar que o
processo de fabrico desenvolvido, executado de acordo com os parâmetros estabelecidos, pode
originar consistentemente o produto pretendido. Uma correlação adequada entre inputs do processo,
controlos associados e outputs é essencial para uma validação de processo bem-sucedida. Uma vez
que existem diversos inputs, outputs e controlos associados a cada operação de fabrico, uma
abordagem sistemática que enfatize a compreensão do produto e do processo, com base na gestão de
risco, é essencial para identificar e avaliar as atividades de validação que têm que ser realizadas
durante a transferência de tecnologia.
Palavras-chave: injetáveis, produto acabado, transferência de tecnologia, validação de processo
5
TABLE OF CONTENTS
LIST OF ABBREVIATIONS ..................................................................................................................... 6
LIST OF FIGURES .................................................................................................................................. 8
LIST OF TABLES .................................................................................................................................... 9
Identification of all unit operations and their associated equipment is crucial when selecting those
parameters and attributes that are considered critical and, therefore, need to be controlled. All relevant
information about the product and the process obtained during the development phase should also be
properly reviewed and evaluated. [3] [23] [29] [30] [31] [25]
According to ICH Guideline Q8(R2), “A CPP is a process parameter whose variability has an impact on
a CQA and therefore should be monitored or controlled to ensure the process produces the desired
quality”. Therefore, it is crucial to identify and to determine the functional relationships that link CPPs
and CMAs to product CQAs, in order to establish a proper control strategy, which can be simplified and
improved by using a risk-based approach. The product quality attributes classified as CQAs usually
include product appearance, assay and impurities, which are critical because they have the potential to
be impacted by the formulation and/or manufacturing process variables. Bacterial endotoxins, product
sterility and particulate matter are also considered to be CQAs for injectable pharmaceutical products.
CQAs are generally ensured through a good pharmaceutical quality system and by implementing an
effective control strategy. [3] [23] [29] [30] [31] [25]
A control strategy may include control of input materials, process monitoring and controls, design space
around unit operations, in-process controls and final product specifications to ensure consistent quality.
The design space can be defined as the multidimensional combination and interaction of input variables
that have been demonstrated to provide assurance of quality (i.e., operating within the design space
leads to consistent product quality). PAT can be applied as part of a control strategy, by continuously
monitoring of critical process inputs, in order to maintain the process within an established design space.
39
The main steps for the selection of CPPs and establishment of a control strategy are as follows: [3] [23]
[29] [30] [31] [25]
Identify CQAs for the drug product;
Select API, excipients, container closure system components and other product contact materials;
Define all unit operations and process flowchart;
Define all product and process specification limits;
Characterize and validate all analytical methods;
Complete quality risk management for all critical unit operations and materials;
Explore the design space for all key factors identified during the risk assessment;
Identify and evaluate CPPs and CMAs that can have an effect on product CQAs.
Appendices 2 to 4 show an example of a schematic approach applied to facilitate and improve
technology transfer of a certain liquid injectable product (process mapping, process flowchart and
FMECA). Additionally, a risk assessment regarding the transfer of the same product from one
manufacturing site to another is presented in Appendix 5. The process map lists process parameters,
material attributes and quality attributes for each unit operation of a certain manufacturing process. The
process map provides a basis for the interdependencies between inputs, processing steps and desired
outputs and, therefore, this information can be used to define equipment/facility specific CPPs, CMAs
and CQAs. The process flowchart is a schematic view of the manufacturing process, which can also
simplify the identification of critical inputs and outputs. Risk management tools (e.g., FMEA and FMECA)
are extremely helpful to identify potential failure modes for the manufacturing process and their expected
effect on the outputs.
5.1. Review of the drug product information
5.1.1. Drug product formulation [3] [23]
All raw materials listed in the drug product formulation (i.e., drug substance(s) and excipients) should
be compatible with each other.
The properties of the drug substance that can influence the manufacturability and performance of the
drug product (for instance, water content and particle size) should be properly evaluated. An overage of
a drug substance can be used to compensate for losses during manufacture. Although the use of an
overage is not recommended, it is acceptable if properly justified considering the safety and efficacy of
the drug product.
Additionally, the excipients chosen, their concentration and the characteristics that can influence the
drug product manufacturability or performance should be assessed and it should be proven that the
excipients provide their intended functionality throughout the drug product shelf-life.
40
5.1.2. Drug product specifications and characteristics [3] [23]
The physicochemical and microbiological attributes of the drug product should be verified. In the case
of injectable pharmaceutical products, particulate matter, sterility and bacterial endotoxins should
always be part of the finished product specifications. Different attributes can be identified for liquid or
solid (lyophilized) injectable products, as presented in tables 5 and 6.
Table 4 – Example of drug product release specifications for a liquid injectable product
Test Parameters Acceptance Criteria
Appearance Clear, yellow solution free from visible signs of contamination
Identification Retention time of the major peak in the sample chromatogram corresponds to that of the peak in the reference or house standard chromatogram
pH 3.5 to 4.5
Volume in Container ≥ 5.0 mL
Assay 95.0 % to 105.0 % of the labeled amount
Impurities – Impurity X ≤ 1.0 %
Impurities – Impurity Y ≤ 0.5 %
Impurities – Impurity Z ≤ 0.5 %
Impurities – Any other individual impurity ≤ 0.2 %
Impurities – Total impurities ≤ 2.2 %
Residual Solvents This product complies with the requirements, option 1, in the USP general chapter <467> Residual Solvents
Glass delamination is generally described as the detachment of thin layers from the inner surface of
glass containers, leading to the appearance of glass fragments (lamellae) in a drug product solution.
Although glass has many advantages over other packaging materials, glass delamination is a well-
known disadvantage. When glass delamination occurs, the glass fragments are released from the inner
surface of the glass container directly into the drug product, affecting its quality. This phenomenon is
particularly relevant for water-based liquid injectable products. Some conditions have the potential to
negatively influence the chemical durability of the inner surface of glass containers and, therefore, can
be considered as risk factors for glass delamination.
Manufacturing process of the glass containers:
Regarding the manufacturing process, two different types of glass containers are used for injectable
drug products: molded and tubular. Molded containers are formed in a single high heat cycle, where the
glass is melted, poured and then blown or pressed into a mold. Generally, the glass in molded containers
has a composition which is relatively low in silicon and high in alkali / alkaline elements, lowering the
working temperature and resulting in interior container surfaces with chemical homogeneity. On the
other hand, tubular containers are made from glass tubes and require two high heat cycles. The tubing
is made first and, then, it is segmented / converted in a second heating process into the final container
design (vials, ampoules or syringes). The converting process should be carefully controlled in order to
assure that the interior container surfaces maintain the resistance to chemical attack. Inadequately
converting processes can lead to evaporation of some glass components, changing the chemistry and
lowering the glass resistance. Tubular containers usually have higher amounts of silicon and lower
amounts of alkali / alkaline elements than molded containers. Although both molded and tubular glass
containers for injectable products have high chemical durability, tubular containers are generally less
resistant than molded containers and, therefore, more likely to release glass fragments into the products
that they contain. Nevertheless, proper control of the converting process results in tubular containers
with the equivalent non-delamination of molded containers.
Type of glass and glass composition:
Glass for pharmaceutical packaging can be classified as type I (borosilicate glass) (refer to table 6),
type II (treated soda-lime-silica glass) or type III (soda-lime-silica glass) based on its hydrolytic
resistance. Glass with a lower hydrolytic resistance is more prone to glass delamination and, therefore,
the type of glass usually used for injectable drug products is type I glass, which has high chemical
durability and low reactivity. For drug products formulated at low pH (pH < 7) or aqueous neutral
solutions, any kind of type I glass containers can be used. On the other hand, for drug products
formulated at high pH and/or with buffers (organic acids such as citrate, tartrate and glutarate), only
untreated borosilicate glass should be used. Ammonium sulfate treatment is used to remove the alkali
components from the container surface, which can be required to enhance drug stability since it reduces
the propensity to pH shift. However, this type of treatment is very aggressive, which can lead more easily
45
to the occurrence of delamination and, therefore, the use of sulfur treated containers should be restricted
to products with pH ≤ 7.
Glass in its pure form consists of silicon dioxide with a melting point of approximately 1700 ºC. However,
this is rarely used commercially because of the cost of working at these elevated temperatures. Added
network modifiers, such as sodium, potassium, or boron oxide, lower the melting point and lower the
chemical durability, whereas added network stabilizers, such as calcium and aluminum oxides, improve
the durability of the glass. Colored glass (e.g., amber glass) is produced by transition metal oxides such
as iron oxides. All additives to pure silicon dioxide, as well as silicon itself, can be viewed as potential
extractables from glass containers.
Table 6 – Typical borosilicate glass composition. Adapted from [41]
Main elements Molded containers (%) Tubular containers (%)
SiO2 70.5 74
Al2O3 5.5 6
Na2O 7.5 7
K2O 1.5 1
CaO 1 1
BaO 2.5 0-1
B2O3 11.5 11
Drug product formulation:
Basic solutions (formulated at high pH) and/or solutions with buffers, chelating agents and organic acids
are more aggressive to the glass and, therefore, more susceptible to this phenomenon.
Sterilization process:
Drug products sterilized by terminal sterilization are more likely to be affected by glass delamination
since this process can have a significant effect on the glass stability.
Storage time:
The time duration that the drug product is exposed to the inner surface of the container is directly related
to the potential for glass lamellae formation, which is the reason for this phenomenon to be usually only
detected during the product shelf-life. Chemical interactions might occur during the shelf-life of the
product which can lead to the presence of low concentrations of glass elements in the product.
Storage temperature:
Drug products stored at room temperature have a higher risk of glass delamination occurrence than
drug products stored at refrigerated or frozen conditions.
46
Glass delamination is the result of a complex chemical reaction between the drug product and the inner
surface of the glass container and the risk factors mentioned above influence the degree of this reaction.
This phenomenon can be minimized by proper selection of the type of glass (and glass composition),
appropriate selection and qualification of suppliers and proper quality control of the incoming vials.
The inner surface durability of glass containers should always be evaluated. Aggressive test solutions
(usually, basic solutions and/or solutions with buffers) can be used to predict the propensity of the
internal glass surface to delaminate. Nevertheless, for a proper risk evaluation, individual screening (i.e.,
glass delamination studies with the drug product itself) should be conducted when relevant, particularly
when one or more of the above conditions exist. However, these risk factors alone (or the lack of risk
factors) are not predictive of glass delamination, which is why individual testing is important. It can be
done by placing the drug product and the candidate container(s) together under accelerated conditions,
which allows to identify potential problems that might occur throughout the shelf-life of the product in a
few time (particularly since delamination is usually only detected after months or even years of product
storage). This evaluation should be done prior to the drug product commercialization in order to allow
the selection of the right container and to avoid future product recalls due to glass delamination.
The mechanisms that lead to glass attack by water-based liquid products are mainly related with ion
exchange and dissolution, depending on the pH value. The primary attack mechanism at acidic pH is
the exchange of hydrogen ions from the aqueous phase with the alkali ions from the glass, but the silica
network of the glass is not affected (refer to equation 1). On the other hand, the primary attack
mechanism at basic pH is the dissolution of the silica network of the glass by hydroxide ions (refer to
equations 2 and 3). A third mechanism may occur, which involves dissolution and reaction (particularly
in solutions with buffers). In this case, not only are the elements of the glass dissolving into the product
but some elements from the drug product interact with the glass, which might create a layer that can
detach easily from the glass surface.
𝑆𝑖𝑂𝑁𝑎 (𝑔𝑙𝑎𝑠𝑠) + 𝐻+ → 𝑆𝑖𝑂𝐻 + 𝑁𝑎+ Equation 1
𝑆𝑖𝑂2 (𝑔𝑙𝑎𝑠𝑠) + 2𝐻2𝑂 ↔ 𝐻4𝑆𝑖𝑂4 Equation 2
𝐻4𝑆𝑖𝑂4 + 𝑂𝐻− ↔ 𝐻3𝑆𝑖𝑂4− + 𝐻2𝑂 Equation 3
If glass delamination is predicted in an early phase, some approaches can be used to mitigate the risk,
as follows:
using a different type I glass composition to solve the problem of drug product / glass chemistry
incompatibility;
trying glass containers from different manufacturers due to differences in glass composition and
manufacturing processes;
using quartz-coated containers, since pure SiO2 of quartz is almost inert;
47
consider using plastic containers, which have their own issues but might solve a problem for a
specific drug product that is not compatible with any glass container;
as a last resort, consider modifying the formulation of the drug product.
5.1.4. Proposed manufacturing process [3] [23]
The critical formulation attributes should be considered together with the available manufacturing
process options in order to address the selection of the manufacturing process and confirm the
appropriateness of the components. Appropriateness of the equipment, filters and tubes to be used
should also be evaluated. Process development studies should provide the basis for process
improvement, process validation and any process control requirements. If possible, the critical process
parameters that must be monitored or controlled to ensure that the product is of the desired quality
should be identified. The selection, the control and any improvement of the manufacturing process
(intended for commercial production batches) should be explained.
In the case of injectable pharmaceutical products (and other products intended to be sterile), it should
be verified that an appropriate method of sterilization for the drug product and primary packaging
components was selected and the choice properly justified. The preferred sterilization method is moist
heat sterilization instead of aseptic filling. Any product that is not heat sensitive must be terminally
sterilized. Product approach sterilization parameters can be used for products that have some heat
sensitivity, providing that F0 achieved during sterilization is not lower than 8 minutes (refer to
Appendix 1).
The description of the manufacturing process development should be accompanied with the description
of measurement systems that allow monitoring of critical attributes and at least tentative control
strategies.
5.2. Validation of analytical methods and cleaning validation
5.2.1. Validation of analytical test methods [3] [24] [25]
Limit of detection, limit of quantification, precision and accuracy must be characterized for all analytical
test methods. All methods should be properly validated prior to product and process characterization
studies and the design and implementation of process controls. Any Microbiological method should be
validated to confirm that the product does not influence the recovery of microorganisms.
5.2.2. Cleaning validation [3] [24] [25]
Cleaning validation should be performed in order to confirm the effectiveness of any cleaning procedure
for all product contact equipment. Sufficient data from the verification should be available to support a
conclusion that the equipment is clean and can be released for further use. Adequate cleaning
48
procedures are essential to minimize the risk of contamination and cross-contamination (if the facility
manufactures multiple products), operator exposure and environmental effects.
Limits for the carryover of product residues should be based on a toxicological evaluation and the
justification for the selected limits should be properly documented in a risk assessment. Limits for the
removal of any cleaning agents used should also be established. Acceptance criteria should consider
the potential cumulative effect of multiple items of equipment in the process equipment train. The
influence of the time between manufacture and cleaning and the time between cleaning and use should
be considered to define dirty and clean hold times for the cleaning process.
A worst-case product approach may be used as a cleaning validation model. In this case, a scientific
rationale should be provided for the selection of the worst-case product and the impact of the
introduction of new products to the manufacturing site. The worst-case product may be determined
based on the following criteria: solubility, cleanability, toxicity and potency.
Cleaning validation protocols should specify the locations to be sampled, the rationale for the selection
of these locations and define the acceptance criteria. Sampling is usually carried out by swabbing and/or
rinsing however other methods may be used depending on the equipment. Analytical methods should
be challenged in combination with the sampling methods to demonstrate both the levels of recovery
from the equipment and the reproducibility of the results. Recovery should be shown to be possible from
all product contact materials sampled in the equipment with all the sampling methods used. The cleaning
procedure should be performed an appropriate number of times based on a risk assessment and meet
the acceptance criteria in order to prove that the cleaning method is validated.
5.3. Materials and equipment preparation
The preparation of all materials and equipment to be used for the manufacturing process of an injectable
product is critical in order to ensure the quality and quantity of the materials/components and to avoid
contaminations during production.
5.3.1. Washing and sterilization/depyrogenation of the container closure system components [3] [17]
[42]
The components of the container closure system have to be washed and sterilized with qualified
equipment according to validated procedures. These components can be prepared in the production
plant or can be received pre-sterilized by the manufacturer.
Glass containers (e.g., vials, ampoules) are usually washed and depyrogenated in the production plant
immediately prior to production. The washing is done using a washing machine and water for injection
at least for the last rinsing step. Dry heat depyrogenation consists in the thermal destruction of bacterial
49
endotoxins and is done following the washing process using a depyrogenation tunnel. A tunnel typically
includes three zones: a load / pre-heat zone to pre-warm the glassware, a hot zone where the glassware
is exposed to process temperature (usually between 250ºC and 400ºC) for sufficient time to achieve
sterilization / depyrogenation and a cool zone to bring the glassware to room temperature prior to exiting
the tunnel.
Stoppers can be provided ready-to-sterilize (RTS) or ready-to-use (RTU). RTS stoppers are packaged
in steam sterilizable bags and have to be sterilized in the production plant immediately prior to
production. On the other hand, RTU stoppers are pre-sterilized by the manufacturer and provided in
sterile double or triple bags, in order to maintain the closure integrity and sterility until filling in aseptic
conditions. The stoppers manufacturer has to demonstrate that validated washing cycles using water
for injection and, for the RTU stoppers, also validated sterilization cycles are applied. The manufacturer
have to test and provide acceptable results for particles, bioburden, bacterial endotoxins, sterility and,
in certain cases, residual moisture (a quality certificate should be provided). Moisture content is
particularly critical for lyophilization stoppers (i.e., stoppers used for lyophilized products). These
stoppers must have low moisture content (typically, residual moisture content should be equal or less
than 0.3 %), in order to prevent moisture release from the stoppers to the product throughout its shelf-
life. [43]
5.3.2. Selection and weighing of the raw materials (API and excipients) [3] [25]
Only raw materials from qualified manufacturers/suppliers should be used for the manufacture of the
drug product. All batches of each raw material should be accompanied with a certificate of analysis
(COA), which reflects the supplier’s test results for each specific batch being provided. If the raw material
has a monograph in the Pharmacopeia, the specifications should be in accordance with the
Pharmacopeia followed by the market the finished product is intended to be submitted to. For instance,
if the drug product is intended to be commercialized in the United States or in Europe, the raw materials
used for its production must comply with the specifications from the United States Pharmacopeia (USP)
or European Pharmacopoeia (EP) monograph, respectively. Usually, changing the
manufacturer/supplier of the excipients is not critical to the process provided that all comply with the
relevant Pharmacopeial monograph specifications. On the other hand, changing the manufacturer of
the API is considered critical and, in this case, the compounding process needs to be revalidated using
the new API source.
Each raw material should be stored and handled according to its sensitivity to light, heat, oxygen and
humidity. The information presented in the Pharmacopeial monograph and Material Safety Data Sheet
(MSDS) should always be consulted in order to determine the appropriate conditions and equipment to
be used during the raw material sampling and weighing and during the manufacture of the drug product.
It is crucial to prevent the raw material degradation and to ensure personnel protection. Different
approaches should be followed according to the raw material sensitivity, as follows:
50
Raw materials sensitive to light – The weighing of raw materials sensitive to light should be
performed under yellow lighting and the weighing container should be amber or wrapped with
aluminum foil, in order to avoid direct exposure of the raw material to white light.
Raw materials sensitive to heat – The weighing of raw materials stored in the fridge or in the
freezer should be performed in a short period of time after equilibration of the raw materials to
room temperature. The weighed amount should be immediately used for production or stored
at the same storage conditions as the original container (fridge or freezer), as appropriate.
Raw materials sensitive to oxygen – These raw materials should be weighed in containers with
the lower headspace possible. Immediately after weighing, the headspace of the original
container and of the weighed container should be overlaid with an inert gas (usually, Nitrogen),
in order to replace the oxygen and prevent the degradation of the raw material.
Raw materials sensitive to humidity (hygroscopic) – Excessive exposure of hygroscopic raw
materials to the environment should be avoided. These raw materials should be weighed in
containers with the lower headspace possible. In certain cases, these raw materials are weighed
in controlled humidity environments and the headspace of the original container and of the
weighed container are overlaid with an inert gas (e.g., Nitrogen), to avoid contact with water
molecules.
The weighing of the raw materials has to be performed and verified by trained personnel and the scales
calibrated on a regular basis, to ensure that the exact amounts are weighed and used in the preparation
of the bulk solution.
5.4. Compounding (preparation of the bulk solution)
For liquid parenteral solutions, compounding consists in the preparation of the bulk solution, following
the defined formulation and compounding process. The API and the excipients are dissolved in a
vehicle, which could be aqueous, normally water for injection, or an oil. The equipment used for
compounding are jacketed compounding tanks equipped with mixers and temperature sensors. For
some products sensitive to oxygen, other tank accessories are required like sparging elements for inert
gas sparging and dissolved oxygen sensors. The sequence of addition of the raw materials, the mixing
speeds and the mixing times used can influence the dissolution of each raw material and the
homogeneity of the final bulk solution. The influence of the compounding process on the product quality
should be evaluated and the process parameters used properly validated for each batch size. Ranges
of mixing times and mixing speeds should be challenged to ensure that complete dissolution and
homogenization occur. Usually, a range of mixing times and mixing speeds is established in the PPQ
protocol for each compounding step based on the developmental experience as well as the current
experience of the site with similar products, batch size and compounding tanks. This range is used as
a reference and can be adjusted, if needed, during actual production of the PPQ batches to achieve
51
proper dissolution or homogenization. At the end of the compounding process, samples from the top
and bottom of the preparation tank are collected and tested to verify bulk solution homogeneity. [3]
The selection of the tank to be used for each batch should be based on the product characteristics and
process requirements.
5.4.1. Compatibility of the material of construction with the bulk solution [3] [44]
Usually, stainless steel 316L tanks (refer to figure 14) are used for preparation of solutions in the
pharmaceutical industry, since this material is considered chemically inert, has a high corrosion
resistance and is easily cleaned and sterilized. Glass lined tanks have to be used when the bulk solution
is not compatible with stainless steel. A compatibility study should be performed to ensure that the
product is compatible with the material of construction of the tank taking into consideration the
temperatures used during the compounding process, the maximum expected compounding time and
holding time after compounding. These studies are normally performed at the R&D stage.
Figure 14 – Stainless steel tank [44]
5.4.2. Batch size [3]
The tank operating capacity should be as close as possible to the final solution volume, in order to
decrease the tank headspace. This is particularly critical for products sensitive to oxygen since a lower
tank headspace leads to a lower contact of the bulk solution with oxygen and, therefore, minimizes
degradation.
5.4.3. Compounding process controls [3]
The presence of any heating or cooling step in the compounding process requires the use of a jacketed
tank to heat or cool down the solution until the desired temperature. The need to sparge the solution
52
with an inert gas to remove the dissolved oxygen requires the use of a sparging element. The controls
associated to the compounding process should be taken into consideration (e.g., temperature and
dissolved oxygen monitoring / control) for the tank selection, which should have sensors to measure the
desired parameters during the compounding process for real-time monitoring and control.
5.5. Holding times
Maximum holding times have to be defined for each product / process according to the product
sensitivity, the product compatibility with its contact materials during the manufacturing process and in
order to avoid Bioburden growth. Different holding times may be established according to the
characteristics of the product and the process. [3]
5.5.1. Maximum compounding time [3]
This holding time is calculated from the addition of the first raw material (which is considered as the
beginning of the compounding process) until the tank is hermetically closed (which is considered as the
end of the compounding process). Usually, a safety margin (of one or two hours, for instance) is added
to the maximum compounding time calculated during the manufacture of the PPQ batches.
5.5.2. Maximum bulk holding time [3]
A maximum bulk holding time is typically evaluated for one PPQ batch, in a portion of the bulk solution
left in the compounding or transference tank after the end of the compounding/transference process.
Samples are collected at different time points and evaluated in terms of physicochemical and
microbiological attributes. The results obtained at each time point should be compared with time zero
bulk results (obtained at the end of compounding) in order to understand the behavior of the product
and to verify if there was an increase in bioburden. This holding time can be calculated as follows,
according to the type of product and to the process characteristics:
Maximum holding time between the end of compounding and the end of filling – This holding
time is generally established for liquid products and is calculated from the time the tank is
hermetically closed until the end of filling (last unit is filled).
Maximum holding time between the beginning of API addition and the beginning of the
lyophilization cycle – This holding time is applicable to lyophilized products due to the instability
of the API in the liquid form. It is calculated from the beginning of the API addition to the
preparation tank and the beginning of the lyophilization cycle (freezing phase) in order to cover
the whole time of the manufacturing process that the API is in the liquid form.
5.5.3. Other holding times [3] [45]
A maximum holding time between the end of filling and the beginning of terminal sterilization is generally
evaluated for products terminally sterilized. The aim of this holding time is to establish a maximum time
53
that the product can stay in its final container until being subjected to the appropriate terminal sterilization
process without an increase in bioburden.
For cold chain products, a maximum total time at room temperature should be established. This holding
time is usually calculated from the beginning of API addition to the preparation tank until the end of
inspection/labelling/packaging (at room temperature), in order to evaluate the maximum time that the
product can be exposed above the refrigeration conditions without affecting its quality.
5.6. Filtration
Filtration is the process by which particles are removed from the bulk solution by passing it through a
porous material (filter). When the filtration process also removes microorganisms from the solution it is
called sterilizing filtration. A sterilizing-grade filter has a 0.2 µm or smaller pore size. However, the
classification of a filter by pore size has limited value and, therefore, this measurement has been
replaced by defining the filter in terms of its bacterial retention. Typically, a sterilizing-grade filter is a
filter that retains 107 CFU of a standard test organism (e.g., Brevundimonas diminuta ATCC® 19146™) 1
per cm2 of effective filtration area (EFA) 2 under process conditions. [3] [46] [47] [48] [49]
Pharmaceutical-grade filters are available in several sizes, membranes and configurations: [3] [46] [47]
[49]
Membranes – Filter membranes are made up of different materials of construction with specific
pore ratings and chemistries, such as, nylon, PVDF, polyetersulfone and PTFE.
Sizes – The filter size is usually determined in terms of EFA, which influences the flow rate 3
and total throughput 4 (the larger the EFA, the higher the flow rate and the throughput).
Configurations – The most widely filter configurations used in the pharmaceutical industry are
the following: capsule filter (a self-contained filter device – refer to figure 15) and cartridge filter
(a filter device requiring a housing for use – refer to figure 16).
1 Brevundimonas diminuta ATCC® 19146™ is the most commonly used microorganism for demonstrating a filter’s
bacterial retention capability. It can be obtained in lyophilized form from the American Type Culture Collection (ATCC) and, after reconstitution, stocks can be maintained either refrigerated or frozen on appropriate media. 2 Effective filtration area is the total surface area of the filter available to the process fluid. 3 Flow rate is the volumetric rate of flow of a solution expressed in units of volume per time (e.g., L/min). 4 Filter throughput (or capacity) is the amount of solution that can be filtered through the EFA and is expressed as volume per membrane area.
It is important to ensure that the filter does not adversely affect the product. Extractables are chemical
compounds that can be extracted from product contacting surfaces when exposed to an appropriate
solvent under exaggerated conditions (i.e., time and temperature). Pre-treatment of the product
contacting surface, such as gamma irradiation or steam sterilization, may also increase the levels of
extractables present. Assessment of extractables that may be introduced during pharmaceutical
6 It may not be possible to mimic pressure differential and flow rate simultaneously during validation and, therefore, the filter user should determine which parameter is more relevant to the process and provide proper rationale to support the decision.
57
manufacturing is an important consideration in evaluating the suitability of a process and its equipment
(including filters) for a particular application. The presence of extractables may be related to degradation
of the filter components ultimately affecting its ability to perform as intended. The quantity and
composition of extractables should be considered when determining the suitability of the filter for the
intended application. Flushing the filter prior to use may reduce the level of extractables potentially
entering the process stream. Extractables studies are usually conducted by the filter manufacturer using
model solvents that bracket the properties of pH, ionic strength and/or level of organic components of
the actual drug product. These studies should be performed with the entire filter device under specific
laboratory conditions that simulate worst-case process conditions of contact time, temperature and pre-
treatment (e.g., sterilization of the filter). The level of extractables is proportional to the effective filtration
area and, therefore, the test filter can have the same effective filtration area or higher than the filter to
be used during actual production. Leachables are compounds that migrate from the filter material in the
presence of the actual product formulation under normal process operating conditions. The need for
leachables testing should be assessed on a case-by-case basis by the filter user and, if applicable,
potential leachables are identified and evaluated to ensure they do not compromise the product quality.
5.6.4. Compatibility [3] [46] [47] [48] [49]
Chemical compatibility between the filter and the product can be qualified by the filter manufacturer.
However, filter user testing is required to confirm the compatibility of the product with the filter under
process conditions. Chemical compatibility should include the entire device and depends on the fluid,
filtration temperature and contact time. After product exposure, integrity testing must be performed on
the filter to verify if its integrity was compromised. Additionally, the filter should be visually inspected for
any signs of discoloration, distortion or damage to ensure that no observable physical change occurred.
5.6.5. Binding [3] [46] [47] [48] [49]
Adsorption is a mechanism of product binding to the filter materials (i.e., filter membrane and/or support
materials). It may lead to the loss of API and/or certain excipients that have an affinity for the filter
materials, having an impact on the product composition and concentration. The level of adsorption can
be affected by several factors, such as, product concentration, contact time, flow rate, temperature and
pH. This issue can be solved by selecting a filter with an appropriate composition and compatible with
the product formulation.
5.6.6. Sterilization 7 [3] [46] [47] [48] [49]
The sterility of the filtration assembly is one of the main elements of a successful sterilizing filtration
process. However, the sterilization method used may lead to damage if filters are not properly sterilized.
Therefore, the capability of the filter to be sterilized and sterilization conditions must be qualified by the
7 In a sterilization process, microbiological death or reduction is described by an exponential function. Therefore, the number of microorganisms that withstand a sterilization process can be expressed in terms of probability (which can be reduced to a very low number but can never be reduced to zero).
58
filter manufacturer. The filter user should sterilize the filters according to the manufacturer’s
recommendation and is responsible for validating the sterilization method selected (except gamma-
irradiated filters, which are generally sterilized by the filter manufacturer using validated conditions).
Sterilization of filters is typically performed by one of the following methods: steam sterilization or
irradiation sterilization. The most common sterilization method is steam under pressure, which is usually
performed in an autoclave, at a temperature of 121ºC. Steam sterilization validation should demonstrate
that the sterilization cycle leads to a probability of non-sterility equal to or lower than 10-6. Irradiation
sterilization is a method that can be used as an alternative to steam sterilization. It has several
advantages, such as, high sterility assurance level, no residual sterilization components (water) and,
consequently, dry filters. Since gamma-irradiated filters do not contain any residual water, they are
particularly advantageous for the filtration of non-aqueous products.
Generally, a maximum filtration time is established for each product/process based on the contact time
between the filter and the product evaluated during validation studies (bacterial retention, extractables
and compatibility studies).
Once a specific filter is validated for use in a certain process, further validation (i.e., revalidation) is
required only when some changes are made, such as:
Modifications regarding the drug product formulation, including product concentration, pH,
conductivity or viscosity;
Increase in the amount (volume) of bulk solution to be filtered through a given effective filtration
area;
Filtration temperature;
Sterilization method modifications;
Flow rate and/or pressure used during filtration.
The filtration process may also have some influence on the assay and pH of the bulk product, which is
typically evaluated during process performance qualification, as part of dead volume evaluation or filter
conditioning. Dead volume or filter conditioning is translated in the amount of bulk solution that needs
to be discarded prior to the start of filling in order to obtain a product within specifications. Typically, the
evaluation involves the testing of the first units filled (dead volume), which should be sequentially
numbered when they are collected, or the testing of a sample of bulk product collected immediately after
the filter at determined contacted times, in order to determine the amount of bulk solution to be rejected
prior to the start of filling, if any. Dead volume and filter conditioning time evaluation is typically performed
in the first PPQ batch manufactured, in order to establish an appropriate amount to be discarded in the
subsequent produced batches.
The assay and pH of the first units filled may be affected by the filtration process, particularly in the case
of online filtration (i.e., when filtration and filling occur simultaneously instead of the whole bulk solution
being previously filtered to a holding tank). If steam sterilized filters are not properly dried prior to use,
59
the first units filled may have lower assay results and/or higher pH due to the residual water that
remained in the filtration assembly. The use of an inappropriate drying procedure is particularly critical
for filters with a higher EFA, since a larger amount of water may remain in the filter. In this case, a high
amount of product has to be discarded as part of initial set up activities (i.e., prior to start actually filling),
which leads to excessive product losses. In order to avoid this situation, adequate drying procedures
should be qualified and validated for each filter or, as an alternative, the use of gamma-irradiated filters
may be considered.
5.7. Filling
Filling is the process of bringing the product in its final container. The effects of the filling process on the
product quality should be evaluated by analyzing samples collected at different time points (usually,
beginning, middle and end) considered representative of the whole filling process and after machine
stoppages. The filling uniformity can be affected by the characteristics of the product, which may
influence the dosing system (for instance, more viscous products can be more difficult to fill leading to
fill volume discrepancies). Therefore, the filling process has to be validated to assure that the filling
machine is accurate and that the pumps and needles used are adequate for each product. Usually, the
filling pumps and needles are selected based on the fill volume and the physical characteristics of the
product (particularly, viscosity). [3]
The filling process is directly related to the filling line. If a product is produced in more than one filling
line, the filling process should be validated in all of them. The effect of line stoppages on the product
quality should be evaluated in order to assess eventual unintentional stoppages that might occur during
the filling process (which may be particularly relevant in the case of products sensitive to oxygen and
viscous products). A line stoppage with appropriate duration (e.g., one or two hours) is usually
incorporated into the filling of one PPQ batch and the first units collected after the line stoppage are
analyzed to evaluate its impact on the product quality. The duration of the line stoppage should be
enough to allow proper intervention and resolution of eventual mechanical problems but without having
an impact on the product quality. [3]
5.8. Lyophilization (if applicable)
When a lyophilized product is to be manufactured, an appropriate lyophilization cycle should be
developed. Cycle development is done during the R&D phase however adjustments in cycle times and
parameters may occur when transferring from an R&D lyophilizer to an industrial lyophilizer. Therefore,
manufacturing of engineering / test batches is advisable before production of the PPQ batches. [3] [11]
60
Lyophilization involves heat and mass transfer, which must be taken into consideration for process
design and optimization, since these phenomena vary according to lyophilizer load condition (partial or
full load), lyophilizer design and container closure system. Monitoring and control of CPPs, such as,
shelf temperature and chamber pressure, is essential to achieve proper process control and to obtain a
cake with the desired appearance and quality (refer to figure 18). [3] [11]
Figure 18 – Appearance of a parenteral product before and after lyophilization [50]
During primary drying, it is crucial to maintain the temperature below the critical temperature, which is
the collapse temperature for amorphous solutes or the eutectic temperature for crystalline solutes.
Drying above the critical temperature leads to loss of cake structure (collapse or meltback), which can
ultimately results in rejection of the entire batch. [3] [11]
An example of a lyophilization cycle for an injectable pharmaceutical product is presented in table 7.
Table 7 – Example of a lyophilization cycle for an injectable product
Process phase Time (h:min) Shelf temperature Chamber pressure
(vacuum)
Loading 00:10 5 ºC -
Freezing 06:00 - 45 ºC -
Vacuum preparation 01:00 - 45 ºC 0.1 mbar
Primary drying 30:30 - 5 ºC 0.1 mbar
Secondary drying 02:00 40 ºC 0.1 mbar
Total cycle time 39:40 (h:min)
Freeze drying is typically an expensive and time-consuming process. Therefore, it is usual to try to
improve the process by reducing the cycle time, focusing particularly on optimization of the primary
drying phase, which is the longest of all the three phases. Once a lyophilization cycle is developed and
optimized, process performance qualification is performed as part of process validation in order to
demonstrate that the lyophilization process allows obtaining a product within specifications. The
uniformity and efficiency of the lyophilization process is evaluated by collecting samples from several
positions of the lyophilizer, which are usually tested for cake appearance, reconstitution time and water
content. Evaluation of the lyophilization process is important not only to assure uniform product quality
within the batch but also from batch to batch. [3] [11]
61
5.9. Terminal sterilization (if applicable)
Terminal sterilization is a process whereby the product is sterilized within its final container. This process
is typically used for heat-stable products and is usually accomplished by moist heat sterilization (in an
autoclave). The efficacy of the sterilization process is dependent on heat exposure, number of
microorganisms present in the load (bioburden) and heat resistance of those microorganisms.
Sterilization must lead to a SAL 8 of at least 10-6 (less than one non-sterile unit per one million units).
The determination of the sterilization method to be used for product sterilization (moist heat sterilization
or aseptic filling) is done during R&D stage. Sterilization in the final container is the preferred sterilization
method which is considered to provide a great assurance of sterility. [3] [33] [51] [52]
5.9.1. Terminal sterilization process development and qualification [3] [33] [51] [52]
Two main design approaches are used for the development of moist heat sterilization cycles to be
applied in pharmaceutical manufacturing: overkill approach or product-specific approach. Usually, the
design approach is selected based on the thermal stability of the product and the materials to be
sterilized. The overkill design approach requires less information on the bioburden of the items to be
sterilized than the product-specific design approach. A greater heat input is required, which has a greater
potential to degrade the product and materials subject to sterilization. Therefore, this approach is
normally employed to products and materials that can withstand high heat without affecting their quality.
On the other hand, the product-specific design approach requires a greater amount of information
regarding the items to be sterilized, the indicator organisms (the test organisms shown to be most
resistant to the sterilization process) and the bioburden levels than the overkill approach. Gathering all
this information provides confidence in the values determined in development to use a lower thermal
input than required for the overkill design approach. This is advantageous for the terminal sterilization
of products that cannot withstand the higher temperatures required for the overkill approach, which
provides greater stability and potentially increases the shelf-life of these products.
A survivor curve is a graphical representation of the inactivation or death of a population of
microorganisms exposed to certain lethal conditions and is described using the following semi-
logarithmic equation:
log 𝑁𝐹 =−𝐹(𝑇,𝑧)
𝐷𝑇+ log 𝑁0 Equation 4
Where,
NF – Number of microorganisms after exposure of F equivalent minutes
F(T,z) (Lethality factor: F-value) – equivalent lethality of a cycle, calculated as minutes at a
reference temperature (T), using a defined temperature coefficient (z); it is a measurement of
8 Sterility Assurance Level expresses the probability of occurrence of a non-sterile unit after exposure to a sterilization process.
62
the sterilization cycle effectiveness. z-value is the number of degrees of temperature change
necessary to change the D-value by a factor of 10.
DT (Thermal resistance value: D-value) – Time, in minutes, required for a one-logarithm (or
90%) reduction of the population of microorganisms used as a biological indicator at a specified
temperature (T).
N0 – Number of microorganisms prior to exposure
The terminal sterilization process should be properly qualified in order to ensure that it consistently
meets the design criteria determined for the cycle. Qualification must include both physical and biological
qualification. Biological qualification demonstrates, by use of biological indicators, that the required
lethality (FBIO) is achieved consistently through the load. FBIO is a term used to describe the delivered
lethality measured in terms of actual kill of microorganisms. The most common microorganism used as
a biological indicator is Geobacillus stearothermophilus due to its high heat resistance but other resistant
bacteria may be acceptable. Nevertheless, the biological indicator selected should contain a higher
population and resistance than the expected product bioburden and only spores should be used as
microbiological challenges. Physical qualification demonstrates that predetermined physical
requirements, including sterilization temperature and time, minimum F0 9 and minimum exposure time,
are achieved consistently from load to load. Three maximum loads and three minimum loads have to
be validated per product. FPHYS is a term used to describe the delivered lethality calculated based on the
physical operational parameters of the cycle, which is the integration of the lethal rate over time.
Physical attributes of the drug product, such as, container size, fill volume, mass and physical
configuration may affect temperature distribution 10, heat penetration 11 and microbiological inactivation.
Therefore, it is crucial that temperature distribution and heat penetration studies are performed for each
sterilization process and load pattern (number of containers per tray and number of trays in the load).
Adequate operational parameters must be established for the sterilization process to ensure that the
required physical and biological lethality are achieved while maintaining integrity of the container closure
system and product quality. These parameters consist of a set point and an operating range, which
should be carefully evaluated since the lower end of the range can affect the sterilization process
efficacy and the upper end may affect the product stability.
5.9.2. Relationship between process validation and parametric release of pharmaceutical products (only
applicable to terminally sterilized products) [3] [33] [53]
Process performance qualification is performed as part of process validation in order to demonstrate
that the terminal sterilization process allow obtaining a product within specifications, not only in terms of
9 F0 is the number of equivalent minutes of moist heat sterilization at a temperature of 121ºC delivered to a unit of product. This is calculated using a z-value of 10ºC. If, for instance, a cycle has a stated F0 of 8 minutes, then the sterilization effectiveness of that cycle is equivalent to 8 minutes at 121ºC. 10 Temperature measurement of the heating medium across the autoclave chamber load zone. 11 Temperature measurement that is used to evaluate the amount of thermal energy that has been transferred to the materials that are to be sterilized within the load.
63
sterility but also regarding its physicochemical attributes. Usually, for terminally sterilized products, it is
recommended to evaluate the effects of double terminal sterilization as a worst-case scenario, in order
to determine if the product can withstand two sterilization cycles without affecting its quality.
Parametric release is a system of release that provides sterility assurance based on effective control,
monitoring and documentation as well as a thorough understanding gained during the manufacturing
process validation rather than finished product sterility testing. CPPs that are considered critical for
sterility assurance should be identified for the sterile product manufacturing process. Once identified,
these parameters are properly monitored and controlled, leading to a predictable and reproducible
process. Time, temperature and pressure are examples of critical parameters that must be closely
monitored and controlled during the sterilization cycle, in order to ensure the sterility of the finished
product.
Sterility test is limited in its sensitivity and lacks statistical significance for the evaluation of sterility for
terminally sterilized products due to the low probability of detection of contaminated units. Like all
destructive testing, it is not possible to prove that a batch of product is sterile unless the entire lot is
tested. Additionally, microbiological results are not available in real-time and physical measurements
are used to provide verification of the achievement of operational parameters after cycle completion.
Therefore, sterility assurance should be established through execution of a well-designed and validated
sterilization process instead of being tested into the product.
A parametric release program is usually applicable to pharmaceutical products terminally sterilized by
moist heat due to the following reasons:
It is a well understood and reliable process;
It can be easily controlled and validated;
It is effective against several types of microorganisms (such as, molds, yeasts and
bacteria/spores);
Lethality can be mathematically modeled.
A parametric release program can be used for new and existing products or processes but in both cases
a well justified risk assessment should be developed to preclude the potential to manufacture and
release a non-sterile product.
Proper monitoring and control of pre-sterilization bioburden should be conducted to support sterility
assurance of products that are parametrically released. Limits for product bioburden should be
established, which allow adopting the proper corrective actions whenever negative trends are detected
and ensure that the sterilization process efficacy is not compromised by an unacceptable level and/or
resistance of pre-sterilization bioburden. The load is considered non-sterile if the pre-sterilization
bioburden and/or resistance are greater than the biological indicator challenge used in validation (which
is particularly relevant for sterilization processes designed based on product-specific design approach).
64
Once a sterilization cycle is qualified and validated, load release should be based on meeting
sterilization specifications, as follows:
Product bioburden is within specifications;
The autoclave and all instrumentation used was properly qualified and calibrated;
The validated load pattern was used;
CPPs have been achieved (failure to meet CPPs leads to rejection of the load).
5.10. Inspection
All finished product batches should be 100% inspected in order to verify the level of rejects due to
particles and/or defects (liquid products are inspected for particles and defects while lyophilized products
are only inspected for defects). The inspection results can be used to specify a tentative rejection rate
and they are extremely useful if properly evaluated since the rejected units might be related to the
process or to the product itself. When a particular type of particle or defect appears often, an
investigation should be conducted in order to understand if it is associated with any issue related to the
process or to the product formulation. Once the root cause is identified, proper corrective measures can
be established in order to decrease the number of defective units. [3]
Finished product units should be also subjected to non-destructive leak testing, in order to confirm the
integrity of the container closure system. One of the most commonly used methods is high voltage leak
detection, which ensures product seal integrity by identifying small pinholes, cracks and seal
imperfections that cannot be detected by visual inspection. [3]
5.11. Scale-up and scale-down considerations
When there is the intention to change the validated batch size of a certain product, several aspects need
to be taken into consideration to evaluate its feasibility. Both scale-up or scale-down can be considered
for existing products, usually due to changes in market demand or line transfer. [3]
For new products, scale-up can be done based on submission batches of a specific product
manufactured in a specific line. Adopting a systematic approach might be useful to determine the
proposed scale-up batch size based on pilot-scale batches previously manufactured for submission or
process validation purposes, for instance, by schematizing all data available in a table (refer to table 8
as an example). This is particularly useful for products intended to be commercialized in the US market,
since a proposed batch size is selected based on the data obtained from the batches manufactured for
submission purposes. In this case, any batch size between the batch size of the submission batches
manufactured and the approved proposed batch size can be considered for commercial production. The
batch size of the submission batches should be at least 10% of the proposed maximum size commercial
65
batch, 50 L (per batch if the fill volume per container is larger than 2.0 mL) or 30 L (per batch if the fill
volume per container is up to 2.0 mL), whichever is larger. For products intended to be commercialized
in the European market, the approved batch size is equivalent to the batch size of the batches
manufactured for submission purposes. [3]
5.11.1. Compounding [3] [25]
The availability of tanks of a higher capacity adequate for the new batch size should be ensured. The
amount of API and excipients required for the new batch size should be taken into consideration. For
instance, the possibility of adding large amounts of a raw material to the tank should be properly
evaluated before the scale-up is done.
For products that require pH adjustment, it might be important to calculate the estimated amount of pH
adjustment solutions needed for the new batch size.
5.11.2. Holding times, filtration and filling [3]
The holding times are evaluated and established for the product regardless of the batch size and,
therefore, should not be changed when a scale-up is done. They may be confirmed however during
validation of the new batch size.
It should be evaluated if a filter with an effective filtration area adequate to the new batch size is
available. If not, multiple filters may be used however the impact of this change should be evaluate in
terms of process and filter validation studies. The effective filtration area required to filter a certain
volume is usually determined by small-scale filterability studies. Either constant flow or constant
pressure filterability tests can be used to predict manufacturing performance. The majority of the
filterability tests are performed by keeping pressure constant and measuring the decline in flow rate as
a function of the filtered volume. In this case, the pressure used should mimic production pressure
conditions. Regarding constant flow testing, the flow rate is controlled while the pressure gradually
increases until a maximum pressure is reached.
The output of the filling line should be considered to calculate the expected filling duration for the new
batch size. The expected filling duration should be covered by the maximum filling time qualified by
Media Fill. In case of online filtration, the expected filling duration should also be covered by the
maximum contact time of the filter with the product, since the filtration and filling occur simultaneously
and, therefore, their duration is almost the same (available filter validation studies should be evaluated).
The product should also be compatible with the remaining product contact materials (for instance, tubing
and gaskets), during the expected filling time. If there are no studies available to cover such duration, a
decision should be made to determine if a surfaces compatibility study is required to be done prior to
production or if compatibility will be assessed during production of the first scale-up batch (applicable
when the risk is considered to be low).
66
5.11.3. Process validation requirements [3]
New process performance qualification studies need to be performed due to the increase in the batch
size. The following parameters usually have to be re-evaluated during process validation:
evaluation of the quality of the bulk solution by taking samples from the top and bottom of the
compounding tank for physicochemical testing and bottom sample for bioburden testing –
although the drug product formulation is the same, the batch size and the tank to be used are
different;
the bulk holding time should be confirmed in the new tank – although the material of construction
is the same, a tank with a higher capacity is considered a worst-case for bioburden growth and
for products sensitive to oxygen if the tank headspace is higher;
evaluation of the effects of filtration and filling on the quality of the compounded solution by
taking samples in the beginning, middle and end of filling (although the drug product formulation
and the fill volume are the same, the filtration and filling duration will be longer for the scale-up
batches);
evaluation of finished product results by taking representative samples of the batch after
lyophilization or terminal sterilization, if applicable.
5.11.4. Other considerations [3] [25]
Before the scale-up protocol is designed, the available data for the current batch size needs to be
evaluated, particularly the existence of previous deviations or non-conformities, in order to understand
if there is any issue related to the process or to the product itself that needs to be solved or improved
before the scale-up is done.
The capacity of all equipment must be considered when evaluating scale-up feasibility. For instance, for
lyophilized products, the scale-up batch size should be adjusted to the lyophilizer capacity (minimum
and maximum load) and condenser capacity.
67
Table 8 – Example of a schematic approach for presenting a rationale for submission of a proposed scale-up
batch size of a specific product in a specific filling line
Product presentation 5 mL presentation 10 mL presentation 25 mL presentation
Batch number * XXXXX1 XXXXX2 XXXXX3
Batch size * 50 L 100 L 250 L
Theoretical number of vials * 9259 vials 9345 vials 9652 vials
Viable vials (after 100 % inspection, including samples for quality control) *
8610 vials 8410 vials 9169 vials
Batch average volume * 5.41 mL 10.69 mL 25.92 mL
Yield *
[Yield (%) = (Viable vials / Theoretical number of vials) x 100]
93 % 90 % 95 %
Theoretical proposed maximum batch size
(Theoretical proposed batch size for US Market = Viable vials x Batch average
volume x 10)
465 L 899 L 2376 L
Proposed batch size 465 L
500 L
(Although theoretically it is possible to have bigger batch sizes, due to the available
equipment, the proposed batch size cannot exceed 500 L.)
Target fill volume 5.40 mL 10.70 mL 25.90 mL
Theoretical output
(according to the vial size)
15000 vials/hour for 6 mL vials
10000 vials/hour for 10 mL vials
6000 vials/hour for 25 mL vials
Number of vials of the proposed batch size
(based on the target fill volume)
86111 vials 46728 vials 19305 vials
Approximately filling duration for the proposed batch size
05 h 44 min 04 h 40 min 03 h 13 min
Maximum filling time qualified by Media Fill
24 h 30 min
* Data from the submission batches already manufactured for the product.
68
5.12. Stability studies
After manufacturing, samples from all process validation batches should be placed in stability chambers.
The purpose of stability studies is to provide evidence on how the quality of a drug product varies with
time under the influence of different environmental factors (e.g., temperature and humidity) and to
establish a proposed shelf-life or confirm the shelf-life for the drug product and the recommended
storage conditions. [3] [54] [55]
The design of the stability studies for the drug product should be based on knowledge of the
characteristics of the drug substance, from stability studies on the drug substance and on experience
gained from earlier phases of pharmaceutical development. Generally, a drug product should be
evaluated under storage conditions that test its thermal stability and its sensitivity to moisture or potential
for solvent loss, if applicable. Data from stability studies should be provided on at least three batches of
the drug product, which should have exactly the same formulation, packaged in the same container
closure system and produced using the same manufacturing process as proposed for commercial
batches. Preferably, all batches used for process validation purposes should enter stability studies. [3]
[54] [55]
Specific types of stability studies may be performed on at least one batch of the drug product, such as,
photostability and freeze thaw studies. The photostability characteristics of new drug products should
be evaluated to demonstrate that light exposure does not result in unacceptable change. This evaluation
should allow to clearly define if the product is photostable or photolabile. If the results of the study are
equivocal, testing of one or two additional batches should be conducted. Usually, photostability studies
are carried out in a sequential manner starting with testing the directly exposed drug product and,
afterwards, continuing as necessary to the product in the primary packaging and then in the secondary
packaging. This evaluation should progress until the results demonstrate that the drug product is
adequately protected from light exposure. Freeze thaw studies are performed to predict the impact of
temperature excursions 12 on the drug product quality during the transportation / distribution process.
These studies are done by placing samples of the drug product at extreme temperatures (i.e., samples
are exposed at freezing temperatures followed by exposure at accelerated storage conditions) to
evaluate if the product is stable after cycles of temperature excursions. [3] [56] [45]
A stability protocol needs to be issued for each drug product. The stability protocol defines the number
of samples needed, the storage conditions to be followed, the sample storage orientation (upright or
inverted), the testing points, the tests to be performed and the drug product shelf-life specifications.
Stability studies should include testing of those attributes of the drug product that are susceptible to
change during storage and are expected to influence quality, safety and/or efficacy. The testing should
cover, as appropriate, the physical, chemical and microbiological attributes, preservative content
12 Any event in which the product is exposed to temperatures outside of the recommended storage and/or transportation temperature range.
69
(antioxidant or antimicrobial preservative) and functionality tests (for instance, for a dose delivery system
or multidose vial). [3] [54] [55]
Usually, a stability study should start not more than 30 days after the end of manufacturing. When this
is not possible and samples only enter stability after 30 days, a new time zero testing will be performed.
In case samples enter stability within the defined period, time zero corresponds to release testing date.
Several stability chambers with different storage conditions of temperature and relative humidity (RH)
can be used to perform these studies (refer to table 9), according to the container closure system,
product storage conditions and the climatic conditions of the target markets (refer to table 10). [3] [54]
[55]
Table 9 – Stability storage conditions and study duration per type of stability study
Stability study Storage conditions Study duration
Long-term
25ºC ± 2ºC / 60% ± 5% RH
Until the end of shelf-life
25ºC ± 2ºC / 40% ± 5% RH
30°C ± 2°C / 35% ± 5% RH
30ºC ± 2ºC / 65% ± 5% RH
30ºC ± 2ºC / 75% ± 5% RH
5ºC ± 3ºC
- 20°C ± 5°C
Intermediate 30ºC ± 2ºC / 65% ± 5% RH 12 months
Accelerated
40ºC ± 2ºC / 75% ± 5% RH
6 months 40ºC ± 2ºC / ≤ 25% RH
25ºC ± 2ºC / 60% ± 5% RH
Table 10 – Climatic zones. Adapted from [55]
Climatic zone Definition Mean annual temperature / mean annual partial water
Qualified cycles are used for each component sterilization. Probes
calibrated on annual basis
O
Physicochemical and microbiological
finished product testing
Finished product specifications not met.
Human error. Use of non-validated test methods.
OOS results leading to batch rejected.
5 3 2 30
Low risk.
Validated test methods are used. All results are documented and verified.
S – Severity: Level 1 (low) to 5 (high); O – Occurrence / Probability: Level 1 (rare) to 5 (frequent); D – Detectability: Level 1 (high) to 5 (low) RPN (SxOxD): 0 – 19: No risk 20 – 39: Low risk 40 – 59: Medium risk 60 – 100: High risk
84
Appendix 5 – Risk assessment regarding transfer of a liquid injectable pharmaceutical product between manufacturing sites (FMEA)
Process Step Item Current manufacturing site New manufacturing site Rationale for the change proposed Risk evaluation
Raw materials/
Components
API API from manufacturer “x” is used. API from manufacturer “x” will be used. N/A No risk.
API from the same manufacturer will be used.
Excipients Excipients from manufacturer “y” are
used. Excipients from manufacturer “z” will be
used.
As part of the Technology Transfer to the new manufacturing site, the excipients already available will be used in order to decrease the number of new item codes.
Low risk.
All excipients comply with USP and EP specifications.
Tubing Platinum cured Silicone tubing is used. Platinum cured Silicone tubing available
will be used.
As part of the Technology Transfer to the new manufacturing site, the tubing, already available will be used in order to decrease the number of new item codes.
Low risk.
The tubing material is the same (Platinum cured Silicone) and, therefore, incompatibility issues are not predicted.
Nevertheless, product compatibility with the tubing will be assessed during the manufacture of the PPQ batches.
Container Closure System
Vial 20 mm neck, type I glass, tubular vials
from manufacturer “a” 20 mm neck, type I glass, tubular vials
from manufacturer “b”
As part of the Technology Transfer to the new manufacturing site, container closure system components already available will be used in order to decrease the number of new item codes and to run properly in the equipment available.
Low risk.
The type of glass and the capacity of the vials are the same.
Nevertheless, the impact of this change will be assessed during the manufacture of the PPQ batches.
Stopper RTU bromobutyl stoppers, 20 mm RTS bromobutyl stoppers, 20 mm
Low risk.
The elastomer is the same and, therefore, incompatibility issues are not predicted. RTS stoppers will undergo a validated sterilization cycle.
Nevertheless, the impact of this change will be assessed during the manufacture of the PPQ batches.
Seal Dark blue flip-off caps, 20 mm Dark blue flip-off caps, 20 mm N/A No risk.
The same type of seals will be used.
85
Process Step Item Current manufacturing site New manufacturing site Rationale for the change proposed Risk evaluation
Compounding
Batch size The batch size is 100 L (compounded in
a 100 L tank). The batch size will be 200 L (to be
compounded in a 250 L tank).
The batch size was selected based on the equipment available and market demand.
Low risk.
Compounding conditions will be evaluated for the new batch size (200 L batch size compounded in a 250 L tank). The product is not sensitive to oxygen and, therefore, the headspace in the preparation tank is not expected to have any impact on the product quality.
Initial WFI amount
The initial amount of WFI added to the tank is 70 % of final Q.S. weight.
The initial amount of WFI to be added to the tank will be 90 % of final Q.S.
volume.
During evaluation of the process, some API dissolution issues were noticed and it was concluded that increasing the initial WFI amount would enhance the API dissolution.
Medium risk.
The suggested amount of initial WFI is intended to allow proper API dissolution, in order to improve the compounding process and to solve the dissolution issues noticed during evaluation of the process.
The impact of this change will be assessed during the manufacture of the PPQ batches.
Compounding temperature
15 – 25 ºC 20 – 25 ºC During evaluation of the process, it was noticed that the API solubility is compromised below 20 ºC.
Low risk.
The temperature upper limit will remain the same and, therefore, there is no risk of product degradation. Tightening the temperature range will avoid incomplete API dissolution at low temperatures.
Preparation of pH
adjustment solutions
0.5 N pH adjustment solutions are used. 1 N pH adjustment solutions will be
used.
Since a higher amount of WFI will be initially added to the tank, it is recommended to use a more
concentrated pH adjustment solution in order to leave more room for Q.S. and
to ease the pH adjustment.
Low risk.
The IPC pH range remains the same, regardless of the pH adjustment solution concentration.
Nevertheless, the impact of this change will be assessed during the manufacture of the PPQ batches.
Filtration Filters Filter cartridges are used. Filter capsules will be used.
Due to the unavailability of housings in the new manufacturing site, filter
capsules with the same filter membrane will be used.
Low risk.
The same filter membrane will be used. Sterile filtration validation studies are available and include both cartridge and capsule filters.
Risk evaluation:
No risk – No change and, therefore, there is no impact on the product quality.
Low risk – Low risk, which is not expected to have any impact on the product quality. Nevertheless, the impact of the proposed changes will be assessed during process validation.
Medium risk – Acceptable risk, which might have an impact on the product quality. Therefore, the impact of the proposed changes will be assessed during process validation.
High risk – High risk, which is expected to have an impact on the product quality. Therefore, the impact of the proposed changes will be assessed with additional studies.