321 Microsponge Loaded Topical Gel The actual values PLGA and PVA for −1, 0, and +1 is 0.5, 1, and 1.5 and for span 80 −1, 0, +1 is 0.1, 0.2, and 0.3. Determined in vitro drug release and entrapment efficiency values of all 14 runs generated with central composite response surface in design of experiments (RS DOE) are presented in Table III as response 1 and 2 accordingly. A mathematical model describing the relations between the followed responses and the studied variables was established. The in vitro drug release of the prepared formulation was found to be in a range from 50.42% to 98.74%. One-way ANOVA (p =0.05) indicated that the formulation parameters A-PLGA, B-span 80, C-PVA, AB, AC, BC, A2, B2 and C2 have statistically significant effect, as shown in Table IV. A graphical illustration of the dependence from the corresponding variables is given in Figure 3 (34). In Table IV, factor coding is coded. Sum of squares is type III– partial. The model F value of 4.32 implies the model is significant relative to the noise. There is only a 3.50% chance that an F value this large could occur due to noise. The p values 0.0500 indicate model terms are significant. In this case there are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve your model. Table II Calibration Valuea S. No. Concentration (µg/ml) Absorbance Standard deviation 1. 0 0 0 2. 0.5 0.1023 ±0.001 3. 1 0.2012 ±0.001 4. 1.5 0.288 ±0.012 5. 2 0.386 ±0.0011 6. 2.5 0.491 ±0.001 7. 3 0.586 ±0.012 8. 3.5 0.681 ±0.0012 9. 4 0.765 ±0.001 10. 4.5 0.855 ±0.012 11. 5 0.999 ±0.0011 a n =3 (average of three determinations). Figure 2. Calibration graph.

322 JOURNAL OF COSMETIC SCIENCE ANOVA for Quadratic model. The entrapment efficiency was in a range from 52.74% to 98.07%. The influence of the examined variables on 1.0/square root. Actual factors were described by a reduced square model. One-way ANOVA (p =0.05) indicated that the formulation parameters A-PLGA, B-span 80, C-PVA, AB, AC, BC, A2, B2, and C2 have a statistically significant effect on entrapment efficiency. Table V and Figure 4 represent the influence of studied variables on 1.0/Sqrt (35). In Table V, factor coding is coded. Sum of squares is type III–partial. The model F value of 4.40 implies the model is significant. There is only a 3.74% chance that an F value this large could occur due to noise. The p values 0.0500 indicate model terms are significant. In this case there are significant model terms. Values greater than 0.1000 indicate the Table III Experimental Design Points for the Studied Variables Values Factor 1 Factor 2 Factor 3 Response 1 Response 2 Std Run A:PLGA B:span 80 C:PVA In vitro drug release EE g ml g %%11 1 0 −1 0 50.42 98.07 12 2 0 1 0 54.63 94.67 7 3 −1 1 1 58.79 89.46 8 4 1 1 1 62.48 87.04 14 5 0 0 1 66.89 82.52 2 6 1 −1 −1 70.77 79.87 3 7 −1 1 −1 73.89 75.38 9 8 −1 0 0 76.07 72.49 6 9 1 −1 1 80.99 68.39 4 10 1 1 −1 83.58 64.97 10 11 1 0 0 87.99 60.57 13 12 0 0 −1 91.83 58.77 1 13 −1 −1 −1 94.77 55.99 5 14 −1 −1 1 98.74 52.74 EE: Entrapment efficiency PLGA: poly lactic-co-glycolic acid PVA: polyvinyl alcohol Std: standard. Table IV Response 1: In Vitro Drug Release Source Sum of squares df Mean square F value p value Model 2,552.12 9 283.57 4.32 0.0216 A-PLGA 27.06 1 27.06 4.22 0.0462 B-span 80 388.38 1 388.38 3.97 0.0149 C-PVA 220.43 1 220.43 4.81 0.0250 AB 379.91 1 379.91 4.11 0.0152 AC 0.0078 1 0.0078 4.01 0.0494 BC 317.39 1 317.39 2.60 0.0182 A2 487.51 1 487.51 3.95 0.0116 B2 586.20 1 586.20 4.80 0.0935 C2 320.08 1 320.08 4.62 0.0180 Residual 488.24 4 122.06 Cor Total 3,040.35 13 AB: PLGA &SPAN 80 AC: PLGA &PVA BC: SPAN 80 &PVA df: degrees of freedom PLGA: poly lactic-co-glycolic acid PVA: polyvinyl alcohol.