**3. The critical material attributes and critical process parameters: identification and linking to CQAs**

All the material attributes (MAs) and/or process parameters (PPs) that can affect the desired critical quality attributes (CQAs) of the final/intermediate product are identified through risk assessment. After risk identification, risk analysis evaluates the impact of the identified MAs and PPs on the CQAs. Further, through risk evaluation, a qualitative or quantitative scale is used for risk estimation of each identified factor on the desired CQA [35]. After the risk evaluation of potential MAs and PPs, only few of them will become potentially critical for quality attributes of the final drug product. In this case, identified MAs will become critical material attributes (CMAs), which must possess certain characteristics or should be chosen in an appropriate range to assure CQAs of the intermediate/final drug product [21]. Critical process parameters (CPPs) are those PPs that should be monitored and controlled in order to obtain the desired CQAs of the intermediate/final drug product [9]. Many tools are used for risk assessment, but the most used ones are Ishikawa or fish-bone diagram for risk identification and failure mode and effect analysis for risk evaluation [36]. Risk assessment should be done in the first step of the drug product development, and analysis of these risk factors is recommended to be reconsidered during different stages of the drug development [37].

To establish the CMAs, the main components of liposomes are evaluated, that is the active substance, the lipids, and others like buffer solution [38]. Each active substance has different physiochemical properties and can influence the desired CQAs [9, 39]. Depending on its solubility, it can be entrapped in the aqueous core or in the lipid bilayer [40]. In order to achieve a greater therapeutic effect, more than one active substance may be encapsulated in the same product, so different physiochemical properties of one may influence the other/others drug entrapment process [40, 41]. Besides this, their octanol/buffer partition must also be taken into consideration because a low partition will lead to a prolonged release of the drug and vice versa [42].

Regarding the lipids, their physicochemical characteristics are also important. For example, lipids that contain unsaturated fatty acids are predisposed to degradation reactions like oxidation or hydrolysis and those which contain saturated fatty acids have a higher transition temperature (Tm) [43]. Another specific characteristic of lipids is their chain length. Usually, a shorter chain length will contribute to a thinner lipid bilayer and a larger internal volume for drug encapsulation. However, comparing three lipids with a difference of two carbon atoms between them, the lipid bilayer thickness obtained did not differ with more than 1 nm and a very small difference in liposomes' size and EE was observed [30]. Lipid properties can have a great impact on liposomes' membrane fluidity, permeability, or charge [44]. In this regard, cholesterol increases liposomal stability, reduces membrane fluidity, and, consequently, contributes to an increased EE [45].

Different compounds used for surface modification, in order to obtain a prolonged blood circulation or modified drug release, must be evaluated during risk identification and analysis. For example, when chitosan was used for coating, its concentration, solubility, and molecular weight influenced liposomal size [46]. In the case of polyethylene glycols (PEGs), the molecular weight and the density on liposomal surface influenced the biodistribution and size [47].

Besides the nature of liposomal components, their concentration and ratios between them are also critical. Some of these ratios, identified in risk assessments of different studies, are: organic phase-to-aqueous phase ratio; cholesterol-tolecithin ratio; chloroform-to-methanol ratio; phospholipids-to-cholesterol molar ratio; and drug-to-lipid ratio [40, 46, 48, 49].

Regarding the preparation process, many techniques and methods like film hydration, emulsification, and reverse phase evaporation were developed but the most used one remains the film hydration method [37]. For this method to be efficient from the viewpoint of EE, parameters like temperature or rotation speed in the evaporation and hydration steps are critical and must be optimized [41, 49]. Film hydration method's great disadvantage is that the obtained liposomes are in the vast majority of micron size with a multilamellar structure, being characterized by a high percentage of lipids and a reduced internal volume. Thus, a reduction in particle size is mandatory, to increase their internal volume and to have a controlled size and narrow particle size distribution [30, 37]. For reduction of liposomal size, several techniques like sonication, freeze-thaw cycling, or extrusion may be used. Comparing these size-reduction methods through a screening experimental design, they were found to be critical for particle size distribution but also for EE [30]. When extrusion through membranes is used for size reduction, membrane pore size, temperature, and applied pressure are important CPPs because these factors define the final particle size and can seriously influence the final EE [30]. Sonication process also needs optimization as regards its time in order to achieve the desired liposomal size [46]. For the freeze-thaw cycling process, the number of the cycles might be taken into consideration for further optimization in order to get the desired particle size [30].

After the risk assessment, next step in QbD development is linking the identified CMAs and CPPs to CQAs by using Design of Experiments (DoE). The greatest advantage of using DoE is that it can track the interactions between the studied factors, CMAs and CPPs, and it can establish a quantitative relationship between the

**77**

*Pharmaceutical Development of Liposomes Using the QbD Approach*

identified variables and the results [36]. For a better understanding of this concept in liposomes development, the most important CQAs will be discussed further, and

The drug content is most frequently optimized in terms of EE than drug concentration as EE reflects better the preparation process performance and robustness. According to published data, the most influential factors for EE are: the amount of drug and phospholipids; cholesterol concentration; the nature of lipids and drug; the interactions between different components; the lipid-to-drug ratio; and several process parameters. Among these, several studies established, through DoE approaches, that lipid molar ratio and lipid-to-drug ratio are the most critical parameters for EE optimization. Using a great amount of lipids for liposome preparation favors the formation of many vesicles with a significant internal volume for drug encapsulation and, consequently, the EE of hydrophilic drugs increases [30, 45]. Including cholesterol in liposome formulations increases not only their stability but also the drug content, due to the so-called "pocket" theory, presuming that cholesterol can generate different size pockets inside the lipid bilayer where API can be entrapped [45]. The use of unsaturated lipids was shown to have a similar effect, the unsaturated lipids forming pockets inside the lipid bilayer where lipophilic drugs can be entrapped [50]. By increasing lipid-to-drug ratio from 1:1 to 10:1, the EE of ritonavir, a lipophilic drug, was doubled from 45 to 90% [43]. The ratio between phospholipid concentration and drug concentration was demonstrated to have a significant impact on EE in the case of simvastatin. Thus, at high simvastatin

how CMAs and CPPs can influence them according to different studies.

concentration and low concentration of phospholipids, EE decreased [51].

hydrating buffer and of external buffer being critical parameters [52].

cannot be further increased [41, 46].

**3.2 Particle size and size distribution**

In the case of drugs with pH-dependent solubility like doxorubicin, the EE can be improved by the use of active loading method at specific pH range, the pH of

Regarding the drug concentration, several studies concluded that the total amount of drug used for encapsulation has a breaking point from which the EE

EE may also be influenced by PPs. A fractional factorial design was used to establish the link between PPs like hydration temperature and the number of freeze-thaw cycles during preparation of liposomes loaded with FK50, an immunosuppressant. An interaction between these PPs was highlighted. It was observed that by using an increased number of cycles, and high hydration temperature, EE decreased, because repetition of the freeze-thaw cycle at high temperature might induce leakage of FK506 from the membrane of the liposomes [39]. In another study, it was observed that PPs like temperature during film hydration and rotation speed during solvent evaporation are critical. Using high temperature, above Tm, the concentration of the encapsulated drug increases because at high temperature, the lipid bilayer is more fluid and permits the entrapment of more drug. Also, using a high rotation speed at the film formation can have a great impact on EE because this CPP leads to formation of a thinner lipid bilayer which can easily be hydrated [53].

Maintaining a controlled particle size and low PDI is one of the main challenges

when preparing liposomes, and optimization of these parameters is improved through the use of DoE. There are many critical factors influencing these parameters, both formulation and process related. For example, the influence of phospholipid concentration, active substances (curcumin and doxorubicin) concentration,

*DOI: http://dx.doi.org/10.5772/intechopen.85374*

**3.1 Drug content**

identified variables and the results [36]. For a better understanding of this concept in liposomes development, the most important CQAs will be discussed further, and how CMAs and CPPs can influence them according to different studies.

## **3.1 Drug content**

*Liposomes - Advances and Perspectives*

consequently, contributes to an increased EE [45].

ratio; and drug-to-lipid ratio [40, 46, 48, 49].

liposomal surface influenced the biodistribution and size [47].

and vice versa [42].

or in the lipid bilayer [40]. In order to achieve a greater therapeutic effect, more than one active substance may be encapsulated in the same product, so different physiochemical properties of one may influence the other/others drug entrapment process [40, 41]. Besides this, their octanol/buffer partition must also be taken into consideration because a low partition will lead to a prolonged release of the drug

Regarding the lipids, their physicochemical characteristics are also important. For example, lipids that contain unsaturated fatty acids are predisposed to degradation reactions like oxidation or hydrolysis and those which contain saturated fatty acids have a higher transition temperature (Tm) [43]. Another specific characteristic of lipids is their chain length. Usually, a shorter chain length will contribute to a thinner lipid bilayer and a larger internal volume for drug encapsulation. However, comparing three lipids with a difference of two carbon atoms between them, the lipid bilayer thickness obtained did not differ with more than 1 nm and a very small difference in liposomes' size and EE was observed [30]. Lipid properties can have a great impact on liposomes' membrane fluidity, permeability, or charge [44]. In this regard, cholesterol increases liposomal stability, reduces membrane fluidity, and,

Different compounds used for surface modification, in order to obtain a prolonged blood circulation or modified drug release, must be evaluated during risk identification and analysis. For example, when chitosan was used for coating, its concentration, solubility, and molecular weight influenced liposomal size [46]. In the case of polyethylene glycols (PEGs), the molecular weight and the density on

Besides the nature of liposomal components, their concentration and ratios between them are also critical. Some of these ratios, identified in risk assessments of different studies, are: organic phase-to-aqueous phase ratio; cholesterol-tolecithin ratio; chloroform-to-methanol ratio; phospholipids-to-cholesterol molar

Regarding the preparation process, many techniques and methods like film hydration, emulsification, and reverse phase evaporation were developed but the most used one remains the film hydration method [37]. For this method to be efficient from the viewpoint of EE, parameters like temperature or rotation speed in the evaporation and hydration steps are critical and must be optimized [41, 49]. Film hydration method's great disadvantage is that the obtained liposomes are in the vast majority of micron size with a multilamellar structure, being characterized by a high percentage of lipids and a reduced internal volume. Thus, a reduction in particle size is mandatory, to increase their internal volume and to have a controlled size and narrow particle size distribution [30, 37]. For reduction of liposomal size, several techniques like sonication, freeze-thaw cycling, or extrusion may be used. Comparing these size-reduction methods through a screening experimental design, they were found to be critical for particle size distribution but also for EE [30]. When extrusion through membranes is used for size reduction, membrane pore size, temperature, and applied pressure are important CPPs because these factors define the final particle size and can seriously influence the final EE [30]. Sonication process also needs optimization as regards its time in order to achieve the desired liposomal size [46]. For the freeze-thaw cycling process, the number of the cycles might be taken into consideration for further optimization in order to get the desired

After the risk assessment, next step in QbD development is linking the identified CMAs and CPPs to CQAs by using Design of Experiments (DoE). The greatest advantage of using DoE is that it can track the interactions between the studied factors, CMAs and CPPs, and it can establish a quantitative relationship between the

**76**

particle size [30].

The drug content is most frequently optimized in terms of EE than drug concentration as EE reflects better the preparation process performance and robustness. According to published data, the most influential factors for EE are: the amount of drug and phospholipids; cholesterol concentration; the nature of lipids and drug; the interactions between different components; the lipid-to-drug ratio; and several process parameters. Among these, several studies established, through DoE approaches, that lipid molar ratio and lipid-to-drug ratio are the most critical parameters for EE optimization. Using a great amount of lipids for liposome preparation favors the formation of many vesicles with a significant internal volume for drug encapsulation and, consequently, the EE of hydrophilic drugs increases [30, 45]. Including cholesterol in liposome formulations increases not only their stability but also the drug content, due to the so-called "pocket" theory, presuming that cholesterol can generate different size pockets inside the lipid bilayer where API can be entrapped [45]. The use of unsaturated lipids was shown to have a similar effect, the unsaturated lipids forming pockets inside the lipid bilayer where lipophilic drugs can be entrapped [50]. By increasing lipid-to-drug ratio from 1:1 to 10:1, the EE of ritonavir, a lipophilic drug, was doubled from 45 to 90% [43]. The ratio between phospholipid concentration and drug concentration was demonstrated to have a significant impact on EE in the case of simvastatin. Thus, at high simvastatin concentration and low concentration of phospholipids, EE decreased [51].

In the case of drugs with pH-dependent solubility like doxorubicin, the EE can be improved by the use of active loading method at specific pH range, the pH of hydrating buffer and of external buffer being critical parameters [52].

Regarding the drug concentration, several studies concluded that the total amount of drug used for encapsulation has a breaking point from which the EE cannot be further increased [41, 46].

EE may also be influenced by PPs. A fractional factorial design was used to establish the link between PPs like hydration temperature and the number of freeze-thaw cycles during preparation of liposomes loaded with FK50, an immunosuppressant. An interaction between these PPs was highlighted. It was observed that by using an increased number of cycles, and high hydration temperature, EE decreased, because repetition of the freeze-thaw cycle at high temperature might induce leakage of FK506 from the membrane of the liposomes [39]. In another study, it was observed that PPs like temperature during film hydration and rotation speed during solvent evaporation are critical. Using high temperature, above Tm, the concentration of the encapsulated drug increases because at high temperature, the lipid bilayer is more fluid and permits the entrapment of more drug. Also, using a high rotation speed at the film formation can have a great impact on EE because this CPP leads to formation of a thinner lipid bilayer which can easily be hydrated [53].

#### **3.2 Particle size and size distribution**

Maintaining a controlled particle size and low PDI is one of the main challenges when preparing liposomes, and optimization of these parameters is improved through the use of DoE. There are many critical factors influencing these parameters, both formulation and process related. For example, the influence of phospholipid concentration, active substances (curcumin and doxorubicin) concentration,

working temperature, buffer pH, and phospholipid-to-cholesterol ratio, on liposomal size, was examined through a screening experimental design. Out of the studied factors, only phospholipid concentration and phospholipid-to-cholesterol ratio significantly influenced the size, while the concentration of the drugs and the working conditions were not critical for particle size. Noteworthy, none of the studied parameters influenced particle size distribution [41].

Regarding the effect of lipid concentration, different studies showed contradictory results on the size in relation to this factor [43, 46]. Usually, on increasing lipid concentration, liposomal size increases, simultaneously with PDI values [54]. Another observation was that increasing lipid concentration over a certain value leads to smaller size, probably due to lipid bilayer rearrangement into a bigger number of liposomes with smaller size and better size distribution [55]. Depending on the lipid type, a different influence on liposomal size was observed [56].

The active substances influence liposomal size depending on their physicochemical properties. It was observed that quercetin, a lipophilic compound, might replace some lipids in the lipid bilayer causing a reduction in size when it is used in high concentrations [55]. On the other hand, high concentrations of pravastatin, a hydrosoluble compound, led to a small size of liposomes due to an interaction between the active substance and the lipids [53].

## **3.3 Zeta potential**

This parameter is influenced by formulation factors like ionic strength, pH, bilayer composition, or charged lipids and PPs like sonication time [48, 57]. In order to modulate the ZP values, different stabilizers, such as stearylamine and diacetyl phosphate, or modified lipids, such as poly(2-ethyl-2-oxazoline) (PEtOz) or PEG, can be incorporated in lipid bilayer. The concentration of these excipients may be optimized such as to obtain optimal stability [50, 57]. In conventional liposomes, cholesterol-to-lecithin ratio influences the zeta potential value [48]. Through DoE, several papers established the critical parameters influencing the zeta potential values. For example, a screening design study was used to determine which of the formulation factors (lipid concentration, cholesterol concentration, chitosan concentration, drug concentration, organic phase/aqueous phase ratio) and process parameters (temperature, stirring speed, sonication time) had a significant influence on zeta potential of chitosan-coated liposomes. As expected, chitosan concentration was a critical parameter, along with the temperature, which favored the coating process through reduction of vesicles size [46].

#### **3.4 In vivo stability and drug release**

By choosing a suitable lipid bilayer composition, a higher stability in vivo might be achieved. In order to sustain this, it was observed that using saturated lipids or cholesterol in formulations, lipid bilayer stability is increased and liposomal uptake by mononuclear phagocyte system will be reduced. Another used pathway is incorporation of different excipients, such as ganglioside GM1, phosphatidylinositol or PEG-lipids, creating a steric barrier which prevents their clearance from the system [42]. The organs in which liposomes accumulate for being eliminated are lungs, liver, spleen, and kidneys. It was observed that those liposomes with negative surface charge present a higher uptake in tumor cells and a slower uptake in liver cells [58]. In a study, four liposome formulations were compared regarding their *in vivo* stability, by tracking their accumulation in spleen. Results showed that the molecular weight of PEG attached to their surface as well as particle size were CMAs influencing the accumulation in the spleen [47].

**79**

*Pharmaceutical Development of Liposomes Using the QbD Approach*

**4. Defining the product and process design space**

Depending on the lipophilic or hydrophilic character of the active substance, the kinetics of release is different because the diffusion through liposomal membrane is influenced by its physiochemical properties [41]. This behavior was shown when the release of two different lipophilic drugs and a hydrophilic drug from the same liposomal system was studied in vitro*.* The lipophilic agent displayed prolonged release and a smaller total drug release in comparison with the hydrophilic one due

For pH-sensitive liposomes, the objective in terms of in vitro drug release is to have a very good stability of the encapsulated drug under physiological conditions and triggered drug release at certain pH values. To achieve this, the use of excipients having a membrane-destabilizing effect and their concentration are critical

The design space (DS) is a multidimensional combination and interaction of the input variables, such as material attributes, and process parameters that have been shown to assure quality [9]. Thus, the advantage of DS determination is that it establishes the operating region which ensures consistent product qualities between different batches. Working within the DS, the product will meet the specifications mentioned in the QTPP, while moving out of it is considered a change that would normally require a regulatory approval in the case of authorized products [9].

Determination of the DS is based on multivariate analysis, considering the main effects of factors as well as their interaction, which helps in determining an operational region based on a predefined confidence level. The DS includes the product design space and process design space. The product DS is established with product's CQAs as dimensions, while the process design space is exhibited as CQAs with

One approach to establish the DS for liposomal products is to take simultaneously into consideration both formulation factors and CPPs. This method was used for the determination of the DS for lyophilized liposomes with simvastatin. Thus, the cholesterol concentration, the PEG proportion, the cryoprotectant-tophospholipids molar ratio, and the number of extrusions through polycarbonate membranes were selected as the most influential factors for lyophilized liposome CQAs. Their variation range was determined, in which the established quality requirements of the product are met: reduced particle size, maximized drug retention during lyophilization, reduced change in phospholipid transition temperature, low residual moisture content. The validity of the DS was confirmed by determining the CQAs of a formulation corresponding to the robust set point, that is, the formulation for which the prediction errors are lowest. Thus, defining the DS was found to be a useful strategy for the development of stable lyophilized liposomes having predictable quality [60]. The same approach was used to generate the DS for preparation of prednisolon-loaded long-circulating liposomes at laboratory scale. In this case, the selected formulation parameters were prednisolon concentration and the PEG proportion in lipidic membrane, and the process parameters were the extrusion temperature and the rotation speed at the hydration of the lipid film. The DS was developed such as to ensure high liposomal drug concentration, high EE,

When process parameters are not found as critical for the quality of the product, or their impact on quality is easily controlled by fixing a certain operating level, the DS is proposed as a function of formulation variables. This approach has been used in several studies, the most studied formulation variables being the active substance

*DOI: http://dx.doi.org/10.5772/intechopen.85374*

to their different characteristics [40].

parameters [52, 56].

respect to CPPs [59].

and controlled size [49].

*Pharmaceutical Development of Liposomes Using the QbD Approach DOI: http://dx.doi.org/10.5772/intechopen.85374*

*Liposomes - Advances and Perspectives*

working temperature, buffer pH, and phospholipid-to-cholesterol ratio, on liposomal size, was examined through a screening experimental design. Out of the studied factors, only phospholipid concentration and phospholipid-to-cholesterol ratio significantly influenced the size, while the concentration of the drugs and the working conditions were not critical for particle size. Noteworthy, none of the

Regarding the effect of lipid concentration, different studies showed contradic-

tory results on the size in relation to this factor [43, 46]. Usually, on increasing lipid concentration, liposomal size increases, simultaneously with PDI values [54]. Another observation was that increasing lipid concentration over a certain value leads to smaller size, probably due to lipid bilayer rearrangement into a bigger number of liposomes with smaller size and better size distribution [55]. Depending

on the lipid type, a different influence on liposomal size was observed [56]. The active substances influence liposomal size depending on their physicochemical properties. It was observed that quercetin, a lipophilic compound, might replace some lipids in the lipid bilayer causing a reduction in size when it is used in high concentrations [55]. On the other hand, high concentrations of pravastatin, a hydrosoluble compound, led to a small size of liposomes due to an interaction

This parameter is influenced by formulation factors like ionic strength, pH, bilayer composition, or charged lipids and PPs like sonication time [48, 57]. In order to modulate the ZP values, different stabilizers, such as stearylamine and diacetyl phosphate, or modified lipids, such as poly(2-ethyl-2-oxazoline) (PEtOz) or PEG, can be incorporated in lipid bilayer. The concentration of these excipients may be optimized such as to obtain optimal stability [50, 57]. In conventional liposomes, cholesterol-to-lecithin ratio influences the zeta potential value [48]. Through DoE, several papers established the critical parameters influencing the zeta potential values. For example, a screening design study was used to determine which of the formulation factors (lipid concentration, cholesterol concentration, chitosan concentration, drug concentration, organic phase/aqueous phase ratio) and process parameters (temperature, stirring speed, sonication time) had a significant influence on zeta potential of chitosan-coated liposomes. As expected, chitosan concentration was a critical parameter, along with the temperature, which favored the

By choosing a suitable lipid bilayer composition, a higher stability in vivo might

be achieved. In order to sustain this, it was observed that using saturated lipids or cholesterol in formulations, lipid bilayer stability is increased and liposomal uptake by mononuclear phagocyte system will be reduced. Another used pathway is incorporation of different excipients, such as ganglioside GM1, phosphatidylinositol or PEG-lipids, creating a steric barrier which prevents their clearance from the system [42]. The organs in which liposomes accumulate for being eliminated are lungs, liver, spleen, and kidneys. It was observed that those liposomes with negative surface charge present a higher uptake in tumor cells and a slower uptake in liver cells [58]. In a study, four liposome formulations were compared regarding their *in vivo* stability, by tracking their accumulation in spleen. Results showed that the molecular weight of PEG attached to their surface as well as particle size were CMAs

studied parameters influenced particle size distribution [41].

between the active substance and the lipids [53].

coating process through reduction of vesicles size [46].

influencing the accumulation in the spleen [47].

**3.4 In vivo stability and drug release**

**3.3 Zeta potential**

**78**

Depending on the lipophilic or hydrophilic character of the active substance, the kinetics of release is different because the diffusion through liposomal membrane is influenced by its physiochemical properties [41]. This behavior was shown when the release of two different lipophilic drugs and a hydrophilic drug from the same liposomal system was studied in vitro*.* The lipophilic agent displayed prolonged release and a smaller total drug release in comparison with the hydrophilic one due to their different characteristics [40].

For pH-sensitive liposomes, the objective in terms of in vitro drug release is to have a very good stability of the encapsulated drug under physiological conditions and triggered drug release at certain pH values. To achieve this, the use of excipients having a membrane-destabilizing effect and their concentration are critical parameters [52, 56].

#### **4. Defining the product and process design space**

The design space (DS) is a multidimensional combination and interaction of the input variables, such as material attributes, and process parameters that have been shown to assure quality [9]. Thus, the advantage of DS determination is that it establishes the operating region which ensures consistent product qualities between different batches. Working within the DS, the product will meet the specifications mentioned in the QTPP, while moving out of it is considered a change that would normally require a regulatory approval in the case of authorized products [9].

Determination of the DS is based on multivariate analysis, considering the main effects of factors as well as their interaction, which helps in determining an operational region based on a predefined confidence level. The DS includes the product design space and process design space. The product DS is established with product's CQAs as dimensions, while the process design space is exhibited as CQAs with respect to CPPs [59].

One approach to establish the DS for liposomal products is to take simultaneously into consideration both formulation factors and CPPs. This method was used for the determination of the DS for lyophilized liposomes with simvastatin. Thus, the cholesterol concentration, the PEG proportion, the cryoprotectant-tophospholipids molar ratio, and the number of extrusions through polycarbonate membranes were selected as the most influential factors for lyophilized liposome CQAs. Their variation range was determined, in which the established quality requirements of the product are met: reduced particle size, maximized drug retention during lyophilization, reduced change in phospholipid transition temperature, low residual moisture content. The validity of the DS was confirmed by determining the CQAs of a formulation corresponding to the robust set point, that is, the formulation for which the prediction errors are lowest. Thus, defining the DS was found to be a useful strategy for the development of stable lyophilized liposomes having predictable quality [60]. The same approach was used to generate the DS for preparation of prednisolon-loaded long-circulating liposomes at laboratory scale. In this case, the selected formulation parameters were prednisolon concentration and the PEG proportion in lipidic membrane, and the process parameters were the extrusion temperature and the rotation speed at the hydration of the lipid film. The DS was developed such as to ensure high liposomal drug concentration, high EE, and controlled size [49].

When process parameters are not found as critical for the quality of the product, or their impact on quality is easily controlled by fixing a certain operating level, the DS is proposed as a function of formulation variables. This approach has been used in several studies, the most studied formulation variables being the active substance

concentration and the composition of the lipid membrane. For example, the DS for liposomes with tenofovir was constructed with respect to phospholipid, concentration, cholesterol concentration, and drug concentration, with a focus to obtain high drug encapsulation efficiency, as this was considered the most difficult property to predict and control for liposomes containing hydrophilic drugs. Other quality attributes considered critical in the study, particle size and stability, were controlled by the pore size of the extrusion membrane and by storing the samples at low temperature, respectively [61]. In another paper, the DS for chitosan-coated liposomes was established as a function of drug concentration, chitosan concentration, and the organic phase-to-aqueous phase ratio during liposome preparation by ethanol injection technique. These variables were found to be the only significant factors affecting the CQAs of the product, although other formulation and process parameters were evaluated through a screening study. The composite desirability function based on constraints was used to determine the conditions that would result in an optimal formulation design, in terms of particle size, encapsulation efficiency, and coating efficiency. Target values were selected for the mentioned quality attributes, and, on the basis of these target values, the optimum values for each variable or processing parameter were obtained. In an additional step, the robustness of the DS was analyzed and the results showed that the selected CPPs may help minimize the variations in QAs that might arise due to the variability of the raw materials [46].

The development of liposomal systems entrapping more than one active substances is more complex, because the properties of each will influence the CQAs of the product and their stability in the processing conditions will impact the process parameters. In this regard, a group of authors established and evaluated the DS for long-circulating liposomes co-encapsulated with curcumin and doxorubicin. DS development was based on a previous screening study, which revealed the critical parameters, that is, phospholipid concentration, the phospholipid-to-cholesterol molar ratio, doxorubicin concentration, and curcumin concentration. The purpose was to obtain the variation range of these factors for which the size of the liposomes is minimized and the encapsulation efficiencies of both drugs are maximized. The DS was established as the region within which the prediction of the CQAs is made with a probability of failure of less than 1%. Moreover, a DS hypercube was set out as a restricted zone in the DS where factors' values can vary independent of each other, without influencing the quality of the product [41].

The incomplete understanding of the manufacturing process is a major barrier in liposomal products' industrial production and clinical application. The destabilization of their structure during long-time storage as aqueous dispersions revealed the need for complex fabrication processes, involving drying steps such as lyophilization and spray drying. The key to the successful design and preparation of optimal liposomal dry powder formulations is an understanding of the significance of the drying process parameters [62]. This aim was achieved by several authors by determining the DS for lyophilization or spray drying process.

The DS for the freeze-drying process of pravastatin-loaded long-circulating liposomes was established as a function of the freezing rate and shelf temperature during primary drying. The two process parameters were found to have a great impact on product's CQAs, along with the presence of an annealing step. The conditions to obtain freeze-dried liposomes with the desired characteristics were generated using the combination trehalose-mannitol as cryoprotectant and by including an annealing step. A series of limitations and target values were applied for the critical quality attributes of the lyophilized product. Thus, the DS was constructed such as to ensure high drug retention after lyophilization, particle size below a certain value, low zeta potential, low residual moisture content, and a short duration of the primary drying. Out of the DS, an optimal formulation was selected and testing this

**81**

*Pharmaceutical Development of Liposomes Using the QbD Approach*

formulation confirmed that the process delivers the desired quality of the product,

A process design space for spray drying of liposomes was developed such as to get a product that met the criteria for all CQAs of an inhalable powder formulation. When developing an inhalable product, the mass median aerodynamic diameter (MMAD) is the most important CQA, as this characteristic influences the deposition in the conducting airway. Besides this, the size stabilization, relative moisture content, and process yield were identified as CQAs. The DS was plotted by imposing restrictions especially for MMAD, which should be within the range 4.5–5.5 μm, but also for liposome size ratio (before/after drying) and the yield of the process, while for moisture content, no restrictions were set, as acceptably low values were obtained for all the experiments. First, an optimal operating space (OOS) was identified with a high feed flow rate, a low outlet temperature, a medium aspirator rate and in the area of low feedstock concentration and high atomizing airflow. The MMAD was the QA restricting the entire ODS to the low feedstock concentration and high atomizing airflow, whereas the other CQAs met the imposed criteria in a

Due to the great success of liposomal systems, not only in pharmaceutical formulation but also in cosmetics and food industry, there is a huge demand on developing and standardizing analytical and bioanalytical methods for liposome complete characterization, as well as for their detection in blood and tissues. Official guidance regarding the manufacturing and controls recommend several methods for lipid components as well as for the drug products [65]. According to FDA guideline, "liposome structures and integrity are important physicochemical properties and they reflect the ability of the liposome drug formulation to contain the drug substance and to retain the drug substance within the appropriate lipo-

A key issue in the control of liposomes is closely related with the quality control of the lipid components, including modified lipids, which could dramatically influence the properties and performances of liposome drug product [65]. The quality of the final product is influenced by the source of lipids and also by the type of the

All types of separation methods, such as gas chromatography (GC) or liquid chromatography (LC), capillary electrophoresis (CE), gel electrophoresis, or electrochromatography have proved their value in the analysis of lipid components of liposomes [66–68]. In order to evaluate the chemical stability of the liposomes, it is mandatory to assess the chemical stability of the lipid components in the final drug formulation, taking into account that some lipids could be degraded by oxidation or hydrolysis. From the practical point of view, most of the separation methods offer important information on this matter. The broad versatility, high selectivity,

While the analysis of lipids using liquid chromatography does not need a long stage of pretreatment, the lipids analyzed using gas chromatography have to be derivatized in order to obtain more volatile compounds which are not turning into degradation products at their boiling point [69]. For fatty acids, the most used derivatization method consists in the esterification of the acids and their transformation in methyl esters. The esterification takes place under various conditions of temperature, mixing process, using different catalysts and for various periods of time. There are plenty of studies among scientific literature about derivatization

efficiency, and low time of analysis are making them a good choice.

*DOI: http://dx.doi.org/10.5772/intechopen.85374*

as long as it is operated within the DS [63].

larger space [64].

**5. The control strategy**

some structure" [65].

lipids: synthetic, semi-synthetic, or natural.

#### *Pharmaceutical Development of Liposomes Using the QbD Approach DOI: http://dx.doi.org/10.5772/intechopen.85374*

formulation confirmed that the process delivers the desired quality of the product, as long as it is operated within the DS [63].

A process design space for spray drying of liposomes was developed such as to get a product that met the criteria for all CQAs of an inhalable powder formulation. When developing an inhalable product, the mass median aerodynamic diameter (MMAD) is the most important CQA, as this characteristic influences the deposition in the conducting airway. Besides this, the size stabilization, relative moisture content, and process yield were identified as CQAs. The DS was plotted by imposing restrictions especially for MMAD, which should be within the range 4.5–5.5 μm, but also for liposome size ratio (before/after drying) and the yield of the process, while for moisture content, no restrictions were set, as acceptably low values were obtained for all the experiments. First, an optimal operating space (OOS) was identified with a high feed flow rate, a low outlet temperature, a medium aspirator rate and in the area of low feedstock concentration and high atomizing airflow. The MMAD was the QA restricting the entire ODS to the low feedstock concentration and high atomizing airflow, whereas the other CQAs met the imposed criteria in a larger space [64].

### **5. The control strategy**

*Liposomes - Advances and Perspectives*

concentration and the composition of the lipid membrane. For example, the DS for liposomes with tenofovir was constructed with respect to phospholipid, concentration, cholesterol concentration, and drug concentration, with a focus to obtain high drug encapsulation efficiency, as this was considered the most difficult property to predict and control for liposomes containing hydrophilic drugs. Other quality attributes considered critical in the study, particle size and stability, were controlled by the pore size of the extrusion membrane and by storing the samples at low temperature, respectively [61]. In another paper, the DS for chitosan-coated liposomes was established as a function of drug concentration, chitosan concentration, and the organic phase-to-aqueous phase ratio during liposome preparation by ethanol injection technique. These variables were found to be the only significant factors affecting the CQAs of the product, although other formulation and process parameters were evaluated through a screening study. The composite desirability function based on constraints was used to determine the conditions that would result in an optimal formulation design, in terms of particle size, encapsulation efficiency, and coating efficiency. Target values were selected for the mentioned quality attributes, and, on the basis of these target values, the optimum values for each variable or processing parameter were obtained. In an additional step, the robustness of the DS was analyzed and the results showed that the selected CPPs may help minimize the variations in QAs that might arise due to the variability of the raw materials [46]. The development of liposomal systems entrapping more than one active substances is more complex, because the properties of each will influence the CQAs of the product and their stability in the processing conditions will impact the process parameters. In this regard, a group of authors established and evaluated the DS for long-circulating liposomes co-encapsulated with curcumin and doxorubicin. DS development was based on a previous screening study, which revealed the critical parameters, that is, phospholipid concentration, the phospholipid-to-cholesterol molar ratio, doxorubicin concentration, and curcumin concentration. The purpose was to obtain the variation range of these factors for which the size of the liposomes is minimized and the encapsulation efficiencies of both drugs are maximized. The DS was established as the region within which the prediction of the CQAs is made with a probability of failure of less than 1%. Moreover, a DS hypercube was set out as a restricted zone in the DS where factors' values can vary independent of each

other, without influencing the quality of the product [41].

determining the DS for lyophilization or spray drying process.

The incomplete understanding of the manufacturing process is a major barrier in liposomal products' industrial production and clinical application. The destabilization of their structure during long-time storage as aqueous dispersions revealed the need for complex fabrication processes, involving drying steps such as lyophilization and spray drying. The key to the successful design and preparation of optimal liposomal dry powder formulations is an understanding of the significance of the drying process parameters [62]. This aim was achieved by several authors by

The DS for the freeze-drying process of pravastatin-loaded long-circulating liposomes was established as a function of the freezing rate and shelf temperature during primary drying. The two process parameters were found to have a great impact on product's CQAs, along with the presence of an annealing step. The conditions to obtain freeze-dried liposomes with the desired characteristics were generated using the combination trehalose-mannitol as cryoprotectant and by including an annealing step. A series of limitations and target values were applied for the critical quality attributes of the lyophilized product. Thus, the DS was constructed such as to ensure high drug retention after lyophilization, particle size below a certain value, low zeta potential, low residual moisture content, and a short duration of the primary drying. Out of the DS, an optimal formulation was selected and testing this

**80**

Due to the great success of liposomal systems, not only in pharmaceutical formulation but also in cosmetics and food industry, there is a huge demand on developing and standardizing analytical and bioanalytical methods for liposome complete characterization, as well as for their detection in blood and tissues. Official guidance regarding the manufacturing and controls recommend several methods for lipid components as well as for the drug products [65]. According to FDA guideline, "liposome structures and integrity are important physicochemical properties and they reflect the ability of the liposome drug formulation to contain the drug substance and to retain the drug substance within the appropriate liposome structure" [65].

A key issue in the control of liposomes is closely related with the quality control of the lipid components, including modified lipids, which could dramatically influence the properties and performances of liposome drug product [65]. The quality of the final product is influenced by the source of lipids and also by the type of the lipids: synthetic, semi-synthetic, or natural.

All types of separation methods, such as gas chromatography (GC) or liquid chromatography (LC), capillary electrophoresis (CE), gel electrophoresis, or electrochromatography have proved their value in the analysis of lipid components of liposomes [66–68]. In order to evaluate the chemical stability of the liposomes, it is mandatory to assess the chemical stability of the lipid components in the final drug formulation, taking into account that some lipids could be degraded by oxidation or hydrolysis. From the practical point of view, most of the separation methods offer important information on this matter. The broad versatility, high selectivity, efficiency, and low time of analysis are making them a good choice.

While the analysis of lipids using liquid chromatography does not need a long stage of pretreatment, the lipids analyzed using gas chromatography have to be derivatized in order to obtain more volatile compounds which are not turning into degradation products at their boiling point [69]. For fatty acids, the most used derivatization method consists in the esterification of the acids and their transformation in methyl esters. The esterification takes place under various conditions of temperature, mixing process, using different catalysts and for various periods of time. There are plenty of studies among scientific literature about derivatization

methods of fatty acids for their GC analysis, researchers being still interested in improving the pretreatment of these compounds to obtain an efficient and rapid process [70].

Even though the literature abounds in examples of separation techniques applied for lipid detection and quantification, some drawbacks are obvious: poor solubility, poor absorbance properties, the need for derivatization, and laborious sample preparations. Nevertheless, the hyphenation between techniques could overcome the drawbacks and offers promising results.

Regarding the control of the final product, the following properties are generally determined to characterize a liposome drug formulation: morphology; surface charge (expressed as zeta potential); particle size (average diameter) and polydispersity index; encapsulation efficiency; the amount of drug relative to the amount of lipids; phase transition; residual solvents; in vitro and in vivo drug release [71]. Variability in these properties may lead to changes in the quality of the liposomal drug products, including leakage of the drug from the liposomes. The QAs monitored for liposomes are presented in **Table 2**, together with the currently employed methods of analysis.

Particle size is one of the crucial parameters for further in vivo application of liposomes. The required size is usually in the range 20–250 nm. When using microscopy techniques, one will obtain a high-resolution three-dimensional profile of the vesicle surface under study. For instance, AFM permits liposomes' visualization without alteration of their native form, given that the requisite surface immobilization does not adversely affect the sample and that the force of the probe itself does not have deleterious effects on the vesicles [75]. AFM is a rapid, powerful, and relatively non invasive technique and compared to TEM, does not requiers complicated sample preparation and removal of liposomes from their native environment. Additionally, TEM provides information on the size distribution and shape of vesicles. Unfortunately, liposomes can suffer from structure perturbations triggered by the high vacuum conditions and the staining process.


*AFM, atomic force microscopy; TEM, transmission electron microscopy; SEM, scanning electron microscopy; ELS, electrophoretic light scattering; LC, liquid chromatography; SEC, size-exclusion chromatography; DLS, dynamic light scattering; MRI, magnetic resonance imaging; CT, computed tomography; MS, mass spectrometry.*

**83**

*Pharmaceutical Development of Liposomes Using the QbD Approach*

To measure particle size and size distribution, three types of techniques could be used: all together, separation, and particle-by-particle counting. In the first case, multi-angle (static) and dynamic light scattering (MALS and DLS, respectively) techniques permit the calculation of the average particle size and charge from the signal generated by multiple particles within the sample. Even though these techniques are fast, they have low resolution and do not provide the particle concentration. Separation techniques, such as disk centrifuge and field-flow fractionation, have the advantage of improving size analysis resolution by using differences in the sample properties, typically sedimentation rates, to pre-separate the sample prior to light-based (absorbance or scattering) analysis. The separation techniques do not measure particle concentration or charge, and they often suffer from separation-based issues. More effective are particle-byparticle counting techniques, such as tunable resistive pulse sensors (TRPSs). The main advantage is the possibility to measure the properties of individual liposomes, offering a direct measurement of the particle concentration as well as high resolution and more accurate analysis of the particle size and charge (zeta potential) distribution. This ability to simultaneously measure the distribution of both the size and zeta potential represents a new and effective means of analyzing

To measure the zeta potential, ELS technique is currently used. It consist in using heterodyne scattering methods in which a fraction of the laser beam is split away by a mirror before reaching the sample, and is directed to the detector where it is combined with scattered light from particles diffusing in the sample. The fraction of redirected light is referred to as the "local oscillator" and, unlike the light scattered by the sample, does not fluctuate. It is used as a reference beam and must be much larger than the average intensity of the scattered light produced by the diffusing particles. To determine the zeta potential, the electrophoretic mobility must

The surface modification of liposomes is sometimes performed in order to increase their in vivo drug-delivery performances. As mentioned earlier, several ways of modifying the surfaces were reported, like the addition of polyethylene glycol (PEG) chains, or the attachment of antibodies and cellular receptor recognition molecules (e.g., the RGD peptide) as molecular targeting probes. Tracking the successful modification of liposomes is made by measuring the change in their electrophoretic mobility, in fact the modification of their zeta potential arising from the change in the number of charged surface groups. Another method for measuring the zeta potential of liposome particles is via ensemble light scattering techniques,

Within the control strategy, process analytical technology (PAT) has been evidenced lately as a significant tool for measuring parameters and attributes related to the active substance, the finished product as well as the processing conditions. Among the PAT tools, in the field of liposomes, near-infrared spectroscopy (NIRs) has been shown to be useful for the chemical characterization of liposomes in terms of the composition of the lipid membrane, as well as for the simultaneous quantifi-

Besides the large-scale production, which is a major challenge, the standardization of analysis procedures easily scalable is another important goal in practical application of liposomes. The liposome production is done in small size batches compared with other pharmaceutical products which are produced in large batches [71]. However, the possible application of liposomes and the increasing number of clinical trials involving liposomes prove the fact that this field of research is very dynamic and the synthesis and analysis methods become more effective, encourag-

*DOI: http://dx.doi.org/10.5772/intechopen.85374*

liposome properties [76].

first be ascertained [77].

which use a similar principle as DLS [78].

cation of excipients and active substance [79, 80].

ing their application in the development of new drug carriers.

#### **Table 2.**

*The quality attributes monitored for liposomal products.*

#### *Pharmaceutical Development of Liposomes Using the QbD Approach DOI: http://dx.doi.org/10.5772/intechopen.85374*

*Liposomes - Advances and Perspectives*

the drawbacks and offers promising results.

by the high vacuum conditions and the staining process.

Morphology UV-vis spectroscopy

Drug encapsulation efficiency Spectroscopy

In vitro drug release Spectroscopy

*The quality attributes monitored for liposomal products.*

process [70].

methods of analysis.

methods of fatty acids for their GC analysis, researchers being still interested in improving the pretreatment of these compounds to obtain an efficient and rapid

Even though the literature abounds in examples of separation techniques applied for lipid detection and quantification, some drawbacks are obvious: poor solubility, poor absorbance properties, the need for derivatization, and laborious sample preparations. Nevertheless, the hyphenation between techniques could overcome

Regarding the control of the final product, the following properties are generally determined to characterize a liposome drug formulation: morphology; surface charge (expressed as zeta potential); particle size (average diameter) and polydispersity index; encapsulation efficiency; the amount of drug relative to the amount of lipids; phase transition; residual solvents; in vitro and in vivo drug release [71]. Variability in these properties may lead to changes in the quality of the liposomal drug products, including leakage of the drug from the liposomes. The QAs monitored for liposomes are presented in **Table 2**, together with the currently employed

Particle size is one of the crucial parameters for further in vivo application of liposomes. The required size is usually in the range 20–250 nm. When using microscopy techniques, one will obtain a high-resolution three-dimensional profile of the vesicle surface under study. For instance, AFM permits liposomes' visualization without alteration of their native form, given that the requisite surface immobilization does not adversely affect the sample and that the force of the probe itself does not have deleterious effects on the vesicles [75]. AFM is a rapid, powerful, and relatively non invasive technique and compared to TEM, does not requiers complicated sample preparation and removal of liposomes from their native environment. Additionally, TEM provides information on the size distribution and shape of vesicles. Unfortunately, liposomes can suffer from structure perturbations triggered

**The investigated property Methods Ref.**

Angle X-ray scattering (SAXS)

Net charge (zeta potential) ELS [74]

In vivo drug release Radiolabeling, fluorescence labeling, MRI, CT, MS [75] *AFM, atomic force microscopy; TEM, transmission electron microscopy; SEM, scanning electron microscopy; ELS, electrophoretic light scattering; LC, liquid chromatography; SEC, size-exclusion chromatography; DLS, dynamic light scattering; MRI, magnetic resonance imaging; CT, computed tomography; MS, mass spectrometry.*

Field-flow fractionation and static

Freeze fracture technique with subsequent transmission

[71–73]

[74, 75]

[75]

[75]

Spectrofluorimetry

electron microscopy

RMN

Particle size Microscopy techniques (TEM, AFM, SEM) SEC

DLS

LC

LC

**82**

**Table 2.**

To measure particle size and size distribution, three types of techniques could be used: all together, separation, and particle-by-particle counting. In the first case, multi-angle (static) and dynamic light scattering (MALS and DLS, respectively) techniques permit the calculation of the average particle size and charge from the signal generated by multiple particles within the sample. Even though these techniques are fast, they have low resolution and do not provide the particle concentration. Separation techniques, such as disk centrifuge and field-flow fractionation, have the advantage of improving size analysis resolution by using differences in the sample properties, typically sedimentation rates, to pre-separate the sample prior to light-based (absorbance or scattering) analysis. The separation techniques do not measure particle concentration or charge, and they often suffer from separation-based issues. More effective are particle-byparticle counting techniques, such as tunable resistive pulse sensors (TRPSs). The main advantage is the possibility to measure the properties of individual liposomes, offering a direct measurement of the particle concentration as well as high resolution and more accurate analysis of the particle size and charge (zeta potential) distribution. This ability to simultaneously measure the distribution of both the size and zeta potential represents a new and effective means of analyzing liposome properties [76].

To measure the zeta potential, ELS technique is currently used. It consist in using heterodyne scattering methods in which a fraction of the laser beam is split away by a mirror before reaching the sample, and is directed to the detector where it is combined with scattered light from particles diffusing in the sample. The fraction of redirected light is referred to as the "local oscillator" and, unlike the light scattered by the sample, does not fluctuate. It is used as a reference beam and must be much larger than the average intensity of the scattered light produced by the diffusing particles. To determine the zeta potential, the electrophoretic mobility must first be ascertained [77].

The surface modification of liposomes is sometimes performed in order to increase their in vivo drug-delivery performances. As mentioned earlier, several ways of modifying the surfaces were reported, like the addition of polyethylene glycol (PEG) chains, or the attachment of antibodies and cellular receptor recognition molecules (e.g., the RGD peptide) as molecular targeting probes. Tracking the successful modification of liposomes is made by measuring the change in their electrophoretic mobility, in fact the modification of their zeta potential arising from the change in the number of charged surface groups. Another method for measuring the zeta potential of liposome particles is via ensemble light scattering techniques, which use a similar principle as DLS [78].

Within the control strategy, process analytical technology (PAT) has been evidenced lately as a significant tool for measuring parameters and attributes related to the active substance, the finished product as well as the processing conditions. Among the PAT tools, in the field of liposomes, near-infrared spectroscopy (NIRs) has been shown to be useful for the chemical characterization of liposomes in terms of the composition of the lipid membrane, as well as for the simultaneous quantification of excipients and active substance [79, 80].

Besides the large-scale production, which is a major challenge, the standardization of analysis procedures easily scalable is another important goal in practical application of liposomes. The liposome production is done in small size batches compared with other pharmaceutical products which are produced in large batches [71]. However, the possible application of liposomes and the increasing number of clinical trials involving liposomes prove the fact that this field of research is very dynamic and the synthesis and analysis methods become more effective, encouraging their application in the development of new drug carriers.
