Formulation by Design: An approach to designing

Article

Formulation by Design: An approach to designing better Drug Delivery Systems Subhashis Debnath*, M.N.L. Aishwarya, M.Niranajan Babu Department of Pharmaceutics, Seven Hills College of Pharmacy, Tirupati. Abstract: Formulation by design (FbD) is a holistic concept of formulation development aiming to design more efficacious, safe, economical and patient-compliant Drug Delivery System (DDS). With the recent regulatory quality initiatives, implementation of FbD has now become an integral part of drug industry and academic research. This review article describes these principles of design of experiment (DoE) and quality by design (QbD) as applicable to drug delivery development using a more apt expression, that is, ‘formulation by design’. It also involves the various applications of QbD, FbD methodology, Design of Experiments with its types along with optimization of factors and software designs respectively. Key Words: Patient-compliant, Formulation by Design (FbD), Quality by Design (QbD), Design of Experiment (DoE).

Introduction Over the past few decades, the domain of drug formulations has metamorphosed from the conventional tablets and capsules to advanced and intricate drug delivery systems (DDS), both temporal and spatial. Formulation development of the oral DDS cannot be adequately accomplished using the traditional ‘trial and error’ approaches of one variable at a time. This calls for the adoption of rational, systematized, efficient and cost-efficient strategies using ‘design of experiments (DoE)’. The recent regulatory guidelines issued by the key federal agencies to practice ‘quality by design (QbD)’ paradigms have coerced researchers in industrial milieu, in particular, to use experimental designs during drug product development [1]. The pharmaceutical Quality by Design (QbD) is a systematic approach to development that begins with predefined objectives and emphasizes product and process understanding and process control, based on sound science and quality risk management. Quality by Design (QbD) is emerging to enhance the assurance of safe, effective drug supply to the consumer and also offers promise to significantly improve manufacturing quality performance. Hence, QbD is related to product performance [1, 2]. The traditional approach of optimizing a formulation or process

essentially involves studying the influence of one variable at a time (OVAT), while keeping all others as constant. Using this OVAT approach, the solution of a specific challenging property can be achieved somehow, but attainment of the true optimal composition or process can never be guaranteed [2]. This may ostensibly be ascribed to the presence of interactions, that is, the influence of one or more variableson others. The final product though may be satisfactory, but mostly sub-optimal, as a better formulation might still prevail for the studied conditions. Design of experiments (DoE), on the other hand, is an optimization technique meant for products and/or processes, developed to evaluate all the potential factors simultaneously, systematically and speedily. Its implementation invariably encompasses the use of statistical experimental designs, generation of mathematical equations and graphic outcomes, portraying a complete picture of variation of the response(s) as a function of the factor(s), which can never be obtained using the traditional OVAT approach [3-7]. Table 1 succinctly enumerates the merits of FbD over the OVAT methodology. FbD Terminology: Specific terminology, both technical and otherwise, is usually used during FbD practice. To facilitate better clarity of precepts of FbD of oral DDS, important terms have been compiled in Table 2

Table 1: Comparison of OVAT and FbD methodology Attribute

OVAT

FbD

Choice of optimum formulation

May result only in sub-optimal solutions

Yields the best possible formulation

Interaction among the ingredients

Inept to reveal possible interactions

Estimates any synergistic or antagonistic interaction among constituents

Scale-up and postapproval changes

Very difficult to design formulation slightly differing from the desired formulation, especially beyond Level II

Changes in the optimized formulation can easily be incorporated, as all response variables are quantitatively governed by a set of input variables

Resource economics

Highly resource-intensive, as it leads to Economical, as it furnishes information on product/process unnecessary runs and batches performance using minimal trials

Time economics

Highly time-consuming, as each Can simulate the product or process behavior using model product is individually evaluated for its equations performance

*E-mail: [email protected]

Pharma Times - Vol. 50 - No. 08 - August 2018

9

Table 2: Vital terminology used during FbD of drug delivery Optimize

Make as perfect, effective or functional as possible

Optimization

Implementation of systematic approaches to achieve ‘the best’ combination of product and/or process characteristics under a given set of conditions using FbD and computers

Independent variables

Input variables, which are directly under the control of the product development scientist

Quantitative variables

Variables that can take numeric values

Categorical variables

Qualitative variables which cannot be quantified

Runs or trials

Experiments conducted according to the selected experimental design

Factors

Independent variables, which tend to influence the product/process characteristics or output of the process

Design matrix

Layout of experimental runs in matrix form as per experimental design

Knowledge space

Scientific elements to be considered and explored on the basis of previous knowledge as product attributes and process parameters

Design space

Multidimensional combination and interaction of input variables and process parameters, demonstrated to provide quality assurance

Control space

Domain of design space selected for detailed controlled strategy

Levels

Values assigned to a factor

Constraints

Restrictions imposed on the factor levels

Response variables

Characteristics of the finished drug product or the in-process material

Critical quality attributes

Parameters ranging within appropriate limits, which ensure the desired product quality

Critical process parameters

Independent process parameters most likely to affect the quality attributes of a product or intermediates

Critical formulation attributes

Formulation parameters affecting critical quality attributes

Effect

The magnitude of the change in response caused by varying the factor level(s)

Main effect

The effect of a factor averaged over all the levels of other factors

Interaction

Lack of additivity of factor effects

Antagonism

Undesired negative change due to interaction among factors

Synergism

Desired positive change due to interaction between factors


10

Nuisance factors

Uncontrollable factors which complicate the estimation of main effect or interactions

Orthogonality

A condition where the estimated effects are due to the main factor of interest, but independent of interactions

Confounding

Lack of orthogonality

Resolution

The measure of the degree of confounding

Coding (or normalization)

Process of transforming a natural variable into a non-dimensional coded variable

Factor space

Dimensional space defined by the coded variables

Experimental domain

Part of the factor space, investigated experimentally for optimization

Blocks

A set of relatively homogenous experimental conditions, wherein every level of the primary factor occurs the same number of times with each level of nuisance factor

Response surface

Graphical depiction of the mathematical relationship

Empirical model

Mathematical model describing factorresponse relation using polynomial equations

Response surface plot

3D graphical representation of a response plotted between two independent variables and one response variable

Contour plot

Geometric illustration of a response obtained by plotting one independent variable against another, while holding the magnitude of response and other variables as constant

Application of QbD in Pharmaceutical industry: Quality refers to product free of contamination and delivers the therapeutic benefit promised in the label to the consumer. The quality of the pharmaceutical product can be evaluated by in vivo or in vitro performance tests “QbD” assures in vitro product performance and in vitro product performance provides assurance of in vivo product performance. Hence, QbD is related to Product Performance. Benefits for Industry • Better understanding of the process. • Less batch failure. • More efficient and effective control of change. • Return on investment/cost savings. • Provides opportunities for more flexible regulatory approaches. • Manufacturing changes within the approved design space without further regulatory review. • Reduction of post-approval submissions. • Better innovation due to the ability to improve processes without resubmission to the FDA when remaining in the Design Space [8,9]. Critical Quality attributes: a) It will be vital to recognize the quality characteristics that will be basic, i.e. those characterizing strength and surrogate for

Bioavailability Criticality and so forth. It is based on the effect of value property/parameter on the security, adequacy & quality (manufacturability) of the product. b) Establish a connection between CPP & CQAs: Identification of property or parameters that can be utilized as a surrogate for clinical security & adequacy (critical to patient) (Figure 3). c) Manufacturability will likewise be a credit (critical to business) that is basic to quality. d) The level of criticality might contrast for an API producing process relative to a drug product assembling methodology. e) API is one part of a medication item and one stage further away from the persistent continuum of Criticality. Few levels of criticality may be utilized to potray different levels of danger. f) As trait or parameter limits methodology edges of disappointment, the level of discriminatingly expanded with the risk. g) Certain Key Aspects of QBD h) The Target Product Quality Profile (TPQP) is a apparatus for setting the key establishment for drug improvement — “arranging with the end in personality.” More as of late, an extended utilization of the TPP being developed arranging, clinical also business choice making, administrative organization communications, and danger administration has begun to develop.

subsequently generated to guide the drug delivery scientist. The drug delivery formulations are experimentally prepared according to the chosen experimental design, and the chosen response variables are evaluated meticulously. • In Step IV, a suitable numeric model is proposed on the basis of experimental data thus generated, and its statistical significance is discerned. Response surface methodology (RSM) is used to relate a response variableto the levels of input variables. Optimum formulation compositions are searched within the experimental domain, using graphical or numerical techniques. • Step V, is the ultimate phase of the FbD exercise, involving validation of response predictive ability of the proposed design model. Drug delivery performance of some studies, taken as the confirmatory runs, is assessed in relation to that predicted using RSM, and the results are critically compared. The optimum formulation is scaled-up and set forth ultimately for the production cycle.

i) To consistently achieve the drug-product quality specified in the label, the drug substance needs to be thoroughly characterized with respect to its physical, chemical, biological, and mechanical properties such as solubility, polymorphism, stability, particle size and flow properties. [7-9] FbD Methodology: FbD hits the bull’s eye using five key strengths, that is, apt choice of experimental designs, accurate computer aided optimization, meticulous drug product development, precise definition of design and control space and identification of critical quality attributes (CQAs), critical formulation attributes(CFAs) and critical process parameters(CPPs). Figure 1 illustrates the concept. The theme of DoE optimization methodology provides complete information on diverse DoE aspects organized in a five-step sequence.

Figure 1: Five cardinal elements of FbD

Experiment) : • The FbD study begins with Step I, where an endeavor is made DoE (Design ofFigure 1: Five cardinal elements of FbD to explicitly ascertain the drug delivery objective(s). Various It is a mathematical tool for systematically planning and CQAs or response variables, which pragmatically epitomize the conducting scientific studies that change experimental variables objective(s), are earmarked for the purpose. All the independent DoE (Design together of Experiment) : to determine their effect on a given response [14]. in order product/process variables are also listed. It is a mathematical tool for systematically planning and conducting It makes controlled changes to input variables scientific in order studies to gainthat change experimental maximum variables together in oforder to determine theirand effect on arelationships given response [14]. It amounts information on cause effect • In Step II, the response variables which directly represent makes controlled to input variables to gain maximum amounts of information withchanges a minimum sample sizeinfororder optimizing the formulation. the product quality (e.g., particle size for nanoparticles, cause and effect relationships with a minimum sample size for optimizing the formulation emulsification time for self-emulsifying systems) are on selected. There are mainly four steps associated with DOE: There are four steps associated with DOE: Also, selection of a ‘prominent few’ influential factors among themainly 1. The design of the experiment (by using various models) 1. The design of the experiment (By using various models) ‘possible many’ input variables is conducted using experimental 2. The collection of the data [9] 2. The collection of the data designs through a process, popularly termed as ‘screening’ . 3. The statistical analysis 3. The statistical analysis of the data and of the data and The formulators, at times, can even by pass the rigors of screening process to choose these factors, that is, CFAs and/ 4. The conclusions reached and recommendations as a result of the experiment. 4. The conclusions reached andmade recommendations made as a or CPPs by virtue of their experience, wisdom and previous result of the experiment. In Optimization Method various types of Model used from preliminary screening of factors to knowledge. Factor influence studies are usually conducted later select their level and of forExperimental finally study ofDesign: their effect so it’s depend upon the formulator to choose Types to quantify the effect of factors and determine the interactions, [9] . a suitable study and help in minimizing the experimenting timemethods if any. Experimental studies are also undertaken to define themodel for There are various type of Experimental design are broad range of factor levels. available out of which method we have to use depends upon the • During Step III, a suitable experimental design is worked out resources we have and what we want to study. to map the responses on the basis of the study objective(s), responses being explored, number and the type of factors, 1. Screening Designs: These are used to identify the important factor and their level and factor levels, that is, high, medium or low. The important experimental designs along with their pros and cons are which affect the quality of formulation. Screening Designs generally discussed in subsequent sections. A design matrix is support only the linear responses. Pharma Times - Vol. 50 - No. 08 - August 2018

11

2. Response Surface Designs: These are used when we required exact image of response, estimating interaction and even quadratic effects. Response surface designs generally support non linear and quadratic response and capable of detecting curvatures

much on the quantity of each substance present but on their proportions. The sum total of the proportions of all the excipients is unity, and none of the fractions can be negative. Therefore, the levels of different components can be varied with the restriction that the sum total should not exceed one.

3. Factorial Designs : Factorial designs (FDs) are very frequently used response surface designs. A factorial experiment is one in which all levels of a given factor are combined with all levels of every other factor in the experiment. These are generally based upon first-degree mathematical models. Full FDs involve studying the effect of all the factors (k) at various levels (x), including the interactions among them, with the total number of experiments being xk. If the number of levels is the same for each factor in the optimization study, the FDs are said to be symmetric, whereas in cases of a different number of levels for different factors, FDs are termed asymmetric.’’ When we study three factors at two level 23 the total Number of run will be=8 & When we study two factors at three level 32 the total Number of run will be=9. 4. Fractional Factorial Design (FFD) : Fractional factorial design is generally used for screening of factor. This design has low resolution due to less number of run. Although, these designs are economical in terms of number of experiments, the ability to distinguish some of the factor effects is partly sacrificed by reduction in the number of experiments. 5. Plackett-Burman Designs (Hadamard designs) : Plackett-Burman designs (PBD) are special two-level FFDs used Fig 2: Various screening and Response surface designs generally for screening of factors. This design is generally used Fig 2: Various screening and Response surface designs [15] when we want to screen high number of factors if we want to Optimization of important factors: study the effect of 7 factors then we have to show four dummy of important factors: Model Development: factors. The interpretations of results in FFD, Plackett-Burman Optimization Designs & Taguchi design are drawn with the help of Pareto chart Model Development: A model is an expression defining the quantitative dependence A model is an expression defining the quantitative dependence of a response variable on the and Half normal plot. of a response variable on the independent variables. Usually, it is a independent variables. Usually, it is a set of polynomials of a given order or Degree. From this set of polynomials of a given order or Degree. From this polynomial polynomial equation we calculate the coefficient with the help of Principal of MLRA (Multiple 6. Central Composite Design (Box-Wilson design) : equation we calculate the coefficient with the help of Principal of Linear Regression Analysis). By the help of software we can also study here the effect of For non linear responses requiring second-order models, central MLRA (Multiple Linear Regression Analysis). By the help of software excipients, their interaction study, 3D Response plot, Contour Plot etc. composite designs (CCDs) are the most frequently employed. A two- we can also study here the effect of excipients, their interaction In screening design with the help of half normal plot and Pareto chart we can find out easily the 2 factor CCD is identical to a 3 FD with rectangular experimental mainstudy, 3Dtheir Response plot, Contour Plot etc. factor and level domain at α = ±1, On the other hand, the experimental domain In screening, design with the help of half normal plot and Pareto is spherical in shape for α = √2= 1.414. The CCD is quite popular chart we can find out easily the main factor and their level in response surface optimization during pharmaceutical product development. From the models thus selected, optimization of one response or the simultaneous optimization of multiple responses needs to 7. Box-Behnken Designs : be optimized graphically, numerically and by using Brute force A specially made design, the Box-Behnken design (BBD), search technology [9]. requires only three levels for each factor -l, 0 and +1. It employs 15 experiments run with three factors at three levels. It is economical (a) Graphical Optimization: then CCD because it requires less number of Trial. 8. Taguchi Design : Taguchi refers to experimental design as “off-line quality control” because it is a method of ensuring good performance in the development of products or processes.” It is also used for screening of factors and it provides 8 experimental run for 7 factors. 9. Mixture Design: Mixture designs are used when the characteristics of the finished product (Drug delivery system) usually depend not so Pharma Times - Vol. 50 - No. 08 - August 2018

12

Graphical optimization deals with selecting the best possible formulation out of a feasible factor space region. To do this, the desirable limits of response variables are set, and the factor levels are screened accordingly by the help of overlay plot. (b) Brute-force search (Feasibility and Grid search) :

Brute-force search technique is the simple and exhaustive search optimization technique. It checks each and every single point in the function space. Herein, the formulations that can be prepared by almost every possible combination of independent factors and screened for their response variables. Subsequently, the acceptable limits are set for these responses, and an exhaustive

Table 3: A few of recent literature instances of FbD optimization of various oral drug delivery systems, process optimization and designs used in screening studies. DDS

Drug

Nanocapsules Solid lipid nanoparticles

Benzocaine Buspirone HCl

Orodispersible tablet

Ondansetron HCl

Nanosuspension Nanostructured liquid

Simvastatin Valproic acid

carrier Colon-targeted systems

Mesalamine

Floating-bioadhesive tablets

Tramadol

Pellets Osmotic pump SR tablets

Lithium carbonate Propranolol HCl Ketoprofen

Factors Screening studies Size, polydispersion index, zeta potential, drug loading Lipid type, surfactant percentage, speed of homogenizer, acetone:DCM ratio Concentrations of glycine, chitosan and drug and tablet crushing strength Product optimization Amounts of polymers and solvents Concentrations of aqueous and organic phases, and relative ratios of solvents Amounts of polymers in compression coating, coating mass and coating force Amounts of constituents polymers Process optimization Rotor speed, slit air flow rate, spray air rate Rotation speed, ionic strength, pH pH, dissolution medium volume, stirring speed

Design FFD Taguchi PBD

CCD Taguchi BBD CCD

FD SSD PBD

search is again conducted by further narrowing down the feasible region. The optimized formulation is searched from the final feasible space (termed as grid search), which fulfills the maximum criteria set during experimentation. (c) Numerical Optimization : It deals with selecting the best possible formulation out of a suitable factor. To do this, the desirable limits of response variables are set, and the factor levels are displayed by the software. Other techniques used for optimizing multiple responses are canonical analysis, ANNs and mathematical optimization. Softwares for design and optimization: Many commercial software packages are available which are either dedicated to experimental design alone or are of a more general statistical type. Software’s dedicated to experimental designs • Design Expert • ECHIP • Multi-Simplex • NEMRODW • Software for general statistical nature • SAS • Minitab • SYSTAT • Graphpad Prism Overall FbD strategy for Drug Delivery Development: The overall approach for conduct of an FbD study in oral DDS can be described by a holistic plan [6,8]. The salient steps involved in this FbD strategy include: Problem definition: The FbD problem is clearly comprehended and defined. Selection of factors and factor levels: The independent factors are identified amongst the quantifiable and easily controllable variables. Design of experimental protocol: Based on the choice of independent factors and the response variables, a suitable

Fig 3: Overall FbD strategy during drug delivery development. experimental design is selected and the number of experimental runs calculated. Formulating and evaluating the dosage form: Various drug delivery formulations are prepared as per the chosen design and evaluated for the desired response(s). Prediction of optimum formulation: The experimental data are used for generation of a mathematical model and an optimum formulation is located using graphical and/or numeric methods. Pharma Times - Vol. 50 - No. 08 - August 2018

13

Validation of optimization: The predicted optimal formulation is prepared and the responses evaluated. Results, if validated, are carried further to the production cycle via pilot plant operations and scale-up techniques. Overall, Flow chart 2 depicts the various salient steps involved during an FbD strategy as a whole. Conclusion: FbD is a thirst area of Research now a day in every industry. Experimental design is a tried and tested approach to formulation optimisation that really works. Experimentation is cost effective. Hence, Experimental design can be used to explore the potential for formulation optimisation against desired criteria before committing a modified formulation to production. In addition, the process is conducive to creative thinking as preconceptions inevitably get challenged when the scope for formulation is considered in mapping the possible formulation. Mostly, orally administered DDS are optimized by FbD. Not only oral DDS but FbD have also been

ideal for development of ‘optimized’ DDS of all other kinds. In this article, an overview of FbD was given. References:

1. Bhupinder Singh, Rishi K, Mousumi N & Naveen A. Available at: https:// www.researchgate.net/publication/51522336. 2. Singh B, Kumar R, Ahuja N.Crit Rev TherDrugCarrierSyst 2005,22:27106. 3. Dhawan S, Kapil R, Singh B.J Pharm Pharmacol 2011,63:342-51. 4. Singh B, Mehta G, Kumar R.Curr Drug Deliv2005,2:143-53. 5. Huang J, Goolcharran C, Ghosh K. Eur J Pharm Biopharm. 2011,78:14150. 6. Rahul Kumar Garg, Indrajeet Singhvi. Asian journal Of Pharmaceutical research.2015,5:217-212. 7. Hemangi Pandit Bendale, Akshada A. Bakliwal, Swati G. World Journal of Pharmaceutical Research.2015,4:402-422, 2015. 8. Verma S, Lan Y, Gokhale R. Int J Pharm. 2009,377:185-98 9. Patel MM, Amin AF. J Pharm Sci,2011,100:1760-1772.

IRF Life Time Achievement Award 2018 IPA Fellowship Award 2018 IPA Eminent Pharmacist Award 2018 The Indian Pharmaceutical Association and Shri Ramanbhai B. Patel Foundation (IRF) established in the year 2004, is dedicated to the memory of Late Shri Ramanbhai Patel, Former President of IPA and Founder Chairman of Zydus Cadila. The Foundation every year recognizes Commitment and Excellence made by a person throughout his / her life in the field of Pharmacy Profession by giving Life Time Achievement Award. Any member of IPA can nominate a person for this award who has made outstanding contributions in the field of pharmacy profession and healthcare / pharmaceutical sciences and powered its growth through vision and passion. Please forward the nomination of an eligible candidate in the prescribed format so as to reach the Hon. Gen. Secretary, IPA, Kalina, Santacruz (East), Mumbai – 400098, positively on or before 30th September, 2018. You may add justification for the nomination on not more than two A4 size papers using font size not less than 12. The Fellowship Award of the Indian Pharmaceutical Association is given every year to IPA Member, who has distinguished himself or herself by his or her magnanimous work in the field of Pharmacy Profession and service to the Association. Any Member of IPA can nominate the eligible candidate for this award supported by two members of IPA. (Out of the three, one should be Fellow of IPA) Please forward the nomination of an eligible candidate in the prescribed format so as to reach the Hon. Gen. Secretary, IPA, Kalina, Santacruz (East), Mumbai – 400098, positively on or before 30th September, 2018. You may add justification for the nomination on not more than two A4 size papers using font size not less than 12. The Eminent Pharmacist Award is the highest award of IPA. The Eminent Pharmacist Award is given to a person who has distinguished him/herself by his/her significant contributions in the field of pharmacy profession. All members of the Central Executive Council and Presidents of State and Local Branches of IPA are requested to send nomination of eligible candidates for the award. Forward your nomination in the prescribed format up to 30th September, 2018. For detail guidelines and prescribed format, please refer to IPA Website www.ipapharma.org or e.mail – [email protected]

Suresh Khanna Hon. Gen. Secretary, IPA


14