optimization techniques in pharmaceutical product development

optimization techniques and factorial designs in pharmaceutical product development.its examples in pharmaceutics and formulation for both b.pharm, m.pharm and pharm.d students

1. UNIT 4 Optimization techniques, statistical Designs in pharmaceutical product development Dr. K. L.DEEPTHI , M.PHARM. PH.D ASSOCIATE PROFESSOR DEPARTMENT OF PHARMACEUTICS 1 Dr.K.L.DEEPTHI

2. CONTENTS Concept of optimization Parameters of optimization Types of experimental designs . Factorial designs 2 Dr.K.L.DEEPTHI

3. 3 Dr.K.L.DEEPTHI

4. 4 Dr.K.L.DEEPTHI

5. Why optimization is necessary 5 Dr.K.L.DEEPTHI

6. 6 Dr.K.L.DEEPTHI

7. 7 Dr.K.L.DEEPTHI

8. 8 Dr.K.L.DEEPTHI

9. Examples of dependent and independent variables Independent variables X1 Diluent ratio X2 compressional force X3 disintegrant level X4 biner level X5 lubricant level Dependent variables Y1 disintegration time Y2 hardness Y3 dissolution Y4 friability Y5 weight uniformity 9 Dr.K.L.DEEPTHI

10. Statistical design: techniques used are divided into 2 types 1. experimentation continues as optimization proceeds., represented by evolutionary operations(EVOP),simplex methods. 2. experimentation is completed before optimization takes place.it is represented by classical mathematical & search methods. 10 Dr.K.L.DEEPTHI

11. Classic optimization : it involves application of calculus to basic problem for maximum/minimum function. Limited applications: 1. problems that are not too complex. 2. they do not involve more than 2 variables. For more than 2 variables graphical representation is not possible. 11 Dr.K.L.DEEPTHI

12. 12 Dr.K.L.DEEPTHI

13. The design of experiments is determines the relation between the factors affecting a process and the output of the process. Statistical DOE refers to the process of planning the experiment where appropriate data can be collected and analysed statistically. 13 Dr.K.L.DEEPTHI

14. 14 Dr.K.L.DEEPTHI

15. Considering the changes in input and effect on output, the optimization techniques are categorized into 5 types: 1. Evolutionary operations 2. simplex method 3. Lagrangian method 4. search method 5. canonical analysis. 15 Dr.K.L.DEEPTHI

16. Evolutionary operations Widely used method(mostly used for tablets). Technique is well suited for production.(formulation process) Small changes in formulation/process are made(i.e. repeated experimentation & statistically analysed for improvement.) It continues until no further changes takes place i.e.it has reached optimum the peak. Drawbacks : EVOP is not a substitute for good laboratory – scale investigation, because of necessarily small in the EVOP. It is not suitable for the lab, therefore it’s impractical & expensive . More repetition is required hence time consuming. 16 Dr.K.L.DEEPTHI

17. Applications: 1. It was applied to tablets by Rubinstein. 2. It has also been applied to an inspection system for parenteral products. 17 Dr.K.L.DEEPTHI

18. Simplex method It is the most widely applied technique. It was proposed by Spendley et,al . This technique has even wider appeal in areas other than formulation & processing. A good example to explain it’s principle is the application to the development of an analytical method i.e. A continuous flow analyser , it was predicted by Deming and king. Simplex method is a geometric figure that has one or more point than the number of factors. If two factors or any independent variables are there, then simplex is represented as triangle. Once the shape of the simplex has been determined , the method can employ a simplex of fixed size or of variable sizes that are determined by comparing the magnitude of the responses after each successive calculation. Triangle for 2 vriable Line for 1variable 18 Dr.K.L.DEEPTHI

19. Applications: This method was used to search for an capsule formula. This was applied to study the solubility problem involving butoconazole nitrate in a multicompartment system. Applied successfully to a direct compression tablet of acetaminophen. Applied the approach to a liquid system i.e. a pharmaceutical solution and was able to optimize physical stability. 19 Dr.K.L.DEEPTHI

20. Basic simplex method and modified simplex method 20 Dr.K.L.DEEPTHI

21. Lagrangian method This optimization method represents the mathematical technique, is an extension of the classic method. Fonner.et,al gave ideas of understanding this technique by applying it to a tablet formulation and by considering 2 independent variables. 21 Dr.K.L.DEEPTHI

22. 22 Dr.K.L.DEEPTHI

23. Y- response, a- regression coefficient, X- dependent variable 23 Dr.K.L.DEEPTHI

24. 24 Dr.K.L.DEEPTHI

25. 25 Dr.K.L.DEEPTHI

26. 26 Dr.K.L.DEEPTHI

27. 27 Dr.K.L.DEEPTHI

28. 28 Dr.K.L.DEEPTHI

29. 29 Dr.K.L.DEEPTHI

30. 30 Dr.K.L.DEEPTHI

31. Advantages of search method: It takes 5 independent variables into account. Persons unfamiliar with mathematics of optimization & with no previous computer experience could carry out an optimization study. 31 Dr.K.L.DEEPTHI

32. 32 Dr.K.L.DEEPTHI

33. 33 Dr.K.L.DEEPTHI

34. Factorial Design: These are designs of choice for simultaneous determination of the effects of several factors and interactions. Symbols to denote levels are: When both the variables are in low concentration. one low variable and second high variable one high variable and second is low variable. Both variables are high. Factorial designs are optimal to determine the effect of pressure and lubricant on the hardness of a tablet. Effect of disintegrant and lubricant concentration on tablet dissolution. It is based on theory of probability & test of significance. Formulation Standard symbols Effect  (%drug release) Low drug- low excipient 1 10 % Low drug- high excipient a 10% High drug- low excipients b 20% High drug- high excipient ab 30% Experiment f1 f2 f3 Interpretation 1 -1 -1 -1 Zero level interaction 2 -1 +1 -1 Main factor effect f2 3 +1 -1 -1 Main factor effect f1 4 -1 -1 +1 Main factor effect f3 5 +1 +1 +1 Interaction between f1, f2,f3. 34 Dr.K.L.DEEPTHI

35. It defines the chance variation and the assignable variations. 1. Factorial designs are helpful to deduce IVIVC. 2. IVIVC are helpful to serve a surrogate measure of rate and extent of oral absorption. 3. BCS classification is based on solubility & permeability issue of drugs which are predictive of IVIVC. 4. sound IVIVC omits the need of bioequivalence study. 5. IVIVC is predicted at 3 levels. LEVEL-A : point to point relationship of invitro dissolution & in vivo performance. LEVEL-B : mean invitro & mean in vivo dissolution is compared & correlated. LEVEL-C : correlation between amount of drug dissolved at one time & one pharmacokinetic parameter is deduced. 35 Dr.K.L.DEEPTHI

36. Factorial design again categorized as 1. full: used for small set of factors 2 . fractional: used to examine multiple factors efficiently with fewer runs. Types of fractional factorial designs: 1. Homogenous fractional 2. Mixed level fractional 3. Box-Hunter 4. Plackett -Burman 5. Taguchi 6. Latin square 36 Dr.K.L.DEEPTHI

37. Full factorial designs Factorial design factors levels No.of runs 2 2 2 2 4 2 3 3 2 8 3 2 2 3 9 3 3 3 3 27 Types: Two level: Three level Two level 37 Dr.K.L.DEEPTHI

38. Three level: levels 3 or more advantage: allows independent estimation of main effect an dinteractions limitation: incresesd no.of experiments require increse no of factors 38 Dr.K.L.DEEPTHI

39. Fractional factorial design(1/x): 39 Dr.K.L.DEEPTHI

40. 40 Dr.K.L.DEEPTHI

41. 1.Homogenous fractional: useful when large no. of factors are screened. 2 . Mixed level fractional: useful when variety of factors are needed to be evaluated for main effects and higher level interactions can be assumed to be negligible. Ex-objective is to generate a design for one variable A at 2 levels and another X at three levels mixed & evaluated. 3 . Box-Hunter: fractional designs with factors of more than 2 levels can be specified as homogenous fractional or mixed level fractional. 41 Dr.K.L.DEEPTHI

42. 4.Plackett-Burman: 1. it is a popular class of screening design. 2. these designs are very efficient screening designs when only main effects are of interest. 3. useful for detecting large main effects economically, assuming all interactions are negligible. 4. used to investigate n-1 variables in n experiments proposing experimental designs for more than seven factors. 42 Dr.K.L.DEEPTHI

43. 5.Taguchi: 1. It is similar to PBD’s. 2. It allows estimation of main effects while minimizing variance 3. It treats optimization problems into 2 categories: static & dynamic. 6. Latin square: These are special case of fractional factorial design where there is one treatment factor of interest and two or more blocking factors. 43 Dr.K.L.DEEPTHI

44. Thank you 44 Dr.K.L.DEEPTHI