Issue link: https://www.e-digitaleditions.com/i/1331851
44 January/February 2021 Tablets & Capsules The batches showed low vari- ability, and therefore, good con- sistency in terms of tablet tensile strength, friability, and disintegra- tion. Variation can be expected to be even lower in non-placebo pharmaceutical formulations that would contain lower amounts of anhydrous lactose. This implies that batch-to-batch and production line-to-line variation is unlikely to cause any variation in pharma- ceutical formulations containing SuperTab 21AN. Conclusions Like APIs, excipients have been shown to introduce variability into formulations, although this variabil- ity differs between excipients and between suppliers. An intelligent use of QbD and MVA can help in the characterization and quantifica- tion of excipient variability and its impact on formulation robustness. This study demonstrates how to use PCA to de-risk the use of an excipient from two different produc- tion lines. PCA revealed its power by deviation. The maximum friability for all batches was 0.1 percent w/w. Overall, product variability from roller dryer 2 was slightly lower than product variability from roller dryer 1. This was to be expected due to the smaller knowledge space that roller dryer 2 covered in the PCA score plot, as result of the shorter production period for that roller dryer. Hausner ratio was calculated from the bulk and tapped density. Figure 7 shows a comparison of two flow- ability parameters for both roller dry- ers. Flowability did not differ signifi- cantly between the two roller dryers. The table below the figure shows the average values of FFC and Hausner ratio with a standard deviation. Figure 8 compares the tablet- ability of batches from each roller dryer. This data was obtained by evaluating tableting behavior in an extreme case to showcase max- imum difference. Blends of 95.5 percent w/w anhydrous lactose with 4 percent w/w croscarmel- lose sodium and 0.5 percent w/w magnesium stearate were produced and tableted. The disintegration time was normalized to tablets of 2.5-megapascal tensile strength by interpolation. No significant dif- ferences in tablet tensile strength, friability, or normalized disintegra- tion time were observed between batches from the two roller dryers. The table below the figure shows the average values with a standard Figure 8 Tableting comparison for the six selected batches from each roller dryer that represent historical variation on tensile strength (left) and disintegration (right) Roller dryer Average value Tablet tensile strength @10kN Tablet tensile strength @15kN Friability @10kN Friability @15kN Disintegration time @2.5MPa 1 2.2±0.4 3.4±0.7 0.10±0.02% 0.06±0.04% 227±35 2 2.1±0.2 3.2±0.3 0.09±0.02% 0.06±0.01% 207±18 Roller dryer 1 Roller dryer 2 8 10 12 14 16 Tablet tensile strength (MPa) 6 5 4 3 2 1 0 Compaction force (kN) Roller dryer 1 Roller dryer 2 Disintegration time (min) 7 6 5 4 3 2 1 0 Tensile strength (MPa) 0 1 2 3 4 5 6 No significant differences in tablet tensile strength, friability, or normalized disintegration time were observed between batches from the two roller dryers.