Results from design of experiment and design space

4.7 Results and discussion

4.7.6 Results from design of experiment and design space

Based on visual observations, paracetamol (7.61 s/60 g) and sodium starch glycolate (14.44 s/60 g) exhibited a good and uniform flow rate. Microcrystalline cellulose formed a hole in the middle of sample and the flow stopped immediately. The type of microcrystalline cellulose was Vivapur®102 (bulk density: 0.30 g/ml, PSD: d50: 136 μm, d90: 255 μm according to the manufacturer). According to previous experiments, the flow rate of microcrystalline cellulose can be improved when the particle size is increased. In

A B

addition, increasing the relative humidity has been shown to decrease the flow rate of microcrystalline cellulose (Crouter and Briens, 2014). In this current study, relative humidity was not recorded. MgSt and StAc are both cohesive and sticky materials, thus the flow rate could not be measured with lubricants. One of the main functions of a lubricant is to increase the flow rate of powder blends.

The flow rate was rather high (2.81–7.51 s) with the powder blends containing StAc (N5–N7, N13–N19). One exception was N8 (StAc 0.5%, mixer speed 1200 rpm, feed rate 16 kg/h), where the flow rate was (11.59–14.72 s). The flow rates of the powder blends containing MgSt (N1–N4, N9–N11) were slower (8.05–17.82 s), compared to the powder blends with StAc. The flow rate of N12 (MgSt 0.5%, mixer speed 1200 rpm, feed rate 16 kg/h) was exceptionally faster (5.41–6.98 s) than other blends containing MgSt.

According to the flow rate investigation of lubricants conducted by Morin and Briens (2013), MgSt was reported to improve the flow rate of spray-dried lactose than StAc. This might be explained by MgSt's ability to fill the cavities (Perrault et al., 2011) between spray-dried lactose particles by creating more spherical and smoother particles (Roblot-Treupel and Puisieux, 1986). Thus, MgSt reduces surface irregularities of the excipients.

However, in this study, the opposite results could be explained by the different physical properties of the particles in the studied blends i.e., particle size and distribution, surface morphology, density and particle shape. However, there was rather extensive variation in the blends containing MgSt (the SD values of the blends with StAc varied between 0.27 and 1.73, whereas the blends with MgSt SD varied between 0.43 and 1.88), which might be evidence of a greater influence of input variables for the mixed blends. In summary, StAc has a higher impact than MgSt on achieving good flow rate properties for the mixed blends. The MLR model was developed to determine the effects of input variables on flowability of the mixed blends (see supplementary). The accuracy of the model was R2 (0.988), Q2 (0.843), model validity (> 0.25), and reproducibility (> 0.5), indicative of an excellent model. The 4D response contour plots of the flow rate properties are presented in Figure 20. From the coefficient plot (Supplementary data), the main significance terms were lubricant type and inlet port.

Figure 20. 4D Contour plot (MLR) of the effects of lubricant and process variables on flow rate. Plot A: lubricant inlet port A, Plot B: lubricant inlet port B.

From the 4D contour plots of flow rate, it can be seen that the flowability of powder blends with MgSt is lower than with StAc (Figure 20). With MgSt, flowability was relatively independent of process variables and concentration from inlet port A, whereas from inlet port B, process factors and concentration did exert some effect on flowability.

In the case of StAc, with feeding through the inlet port B, the flowability was fast and virtually independent of any variables whereas with feeding through the inlet port A, the flowability seemed to be dependent on the concentration and feed rate, even those variables were not statistically significant according to the coefficient plot (Supplementary data). Based on these observations, the most robust and best flowability could be achieved with StAc and inlet port B, a finding in agreement with the ejection force results (will be shown later). Based on the ejection force and flowability results, it seems that inlet port B achieved better results. In practice, when feeding the lubricant through the inlet port A, it is exposed to a more intense and longer blending time, which might cause over-lubrication.

4.7.8 Ejection force

One of the main functions of the lubricant (in addition to enhancing flowability) is to reduce the friction between the tablet and the die wall. The ejection force can be used as a measure of lubrication efficiency. The lower the ejection force needed to eject the tablet from the die, the more efficient the lubricant. The highest ejection force was needed with N3 (724.8 N), whereas the lowest ejection force was observed with N13 (369.8 N). An MLR model was developed to determine the effects of input variables on ejection force. The model for ejection force R2 (0.960), Q2 (0.833), model validity (> 0.25), and reproducibility (> 0.5), was excellent. As seen in the coefficients (Supplementary data), the main significance terms were lubricant type, concentration, inlet port and total feed rate.

A B

As Figure 21 shows, the main effects on ejection force were lubricant type and concentration, with StAc being a slightly more efficient lubricant than MgSt. In addition, the inlet ports A and B have also a significant effect on ejection force. The influence of feed rate was observed clearly with MgSt 0.5%. The ejection force was decreased when increasing the feed rate up to 16 kg/h. With inlet port B, the ejection force was found to be lower than when using inlet port A.

Figure 21. The 4D contour plots (MLR) of the effects of lubricants and process variables on ejection force. Plot A: lubricant inlet port A, Plot B: lubricant inlet port B.

4.7.9 Tablet strength

The level of tablet strength was lower with tablets containing MgSt (N1–N4, N9–N12), compared to tablets containing StAc (N5–N8, N13–N19). The same trend towards higher tablet strengths with StAc has been reported previously (Bastos et al., 2008). Bastos et al.

reported that an increase in the concentration of MgSt, leads to a reduction in the strength of the tablet. This decrease in tablet strength with MgSt is also related to an increase in tablet friability (Perrault et al., 2011). During the process from the beginning to the end of the process, the tablet strength varied between 50.0 N and 67.7 N with tablets containing MgSt, the corresponding range was between 104.0 N and 153.0 N with tablets containing StAc (Supplementary data). N6 varied most at the beginning and in middle of the process, but reached the same level with other tablets of StAc at the end of the process (at 20 min).

In addition, it could be seen that the variation in tablet strength was higher with tablets containing StAc. The accuracy of the model was R2 (0.988), Q2 (0.974), model validity (> 0.25), and reproducibility (> 0.5). The 4D response contour plots of tablet strength at the tableting time point of 20 min are presented in Figure 22.

A B

Figure 22. 4D Contour plot (MLR) of the effects of lubricant and process variables on tablet strength. Plot A: lubricant inlet port A, Plot B: lubricant inlet port B.

As seen in Figure 22, in contrast to the lubricant type, the process variables have no effect on tablet strength. The tablet strength was high with StAc and low with MgSt.

However, a slight effect was observed also with lubricant concentration, as both lubricant types reduced the tablet strength at a concentration of 1.5% compared to 0.5%. To summarize, lubricant quality exerts a significant influence on tablet strength. However, there was no effect of inlet port on tablet strength which would indicate that no severe over-lubrication occurred.

4.7.10 Dissolution of tablets

In general, dissolution was slightly slower with tablets containing MgSt than with StAc.

The effect of several commercial grades of MgSt on the in vitro dissolution behavior of direct compressible paracetamol has been studied e.g. Hussain et al., 1992. All the various grades retarded dissolution; however, there were major differences in the dissolution between the commercial grades. According to the results of this study, the dissolution of StAc was faster than with MgSt. Bastos et al. (2008), have also shown that the formulation prepared with StAc has a higher dissolution rate.

The each individual tablet results show that the dissolution of tablets containing MgSt (N1–N4, N9–N12) varied between 68.8 and 93.1% and tablets containing StAc (N5–N8, N13–N19) varied between 77.4 and 101.6% with a dissolution time point of 2.5 min. With tablets containing MgSt, the fastest dissolution was achieved with N9 (92.4%, mean), whereas the lowest dissolution was with N1 (71.5%, mean). For tablets containing StAc, the fastest dissolution was obtained with N8 (93.5%, mean) and the lowest with N13 (80.7%, mean). The lubricant concentration influenced the dissolution at the time point of 2.5 min, as with N1 and N13, the concentration of lubricant was 1.5%, whereas with N8 and N13, it was 0.5%. The goodness of model was R2 (0.782), Q2 (0.425), model validity

A B

(> 0.25), and reproducibility (> 0.5). The 4D response contour plots of the dissolution at the time point of 2.5 min are presented in Figure 23.

Figure 23. 4D Contour plot (MLR) of the effects of lubricant and process variables on dissolution at time point 2.5 min. Plot A: lubricant inlet port A, Plot B: lubricant inlet port B.

The main influencing factors were the lubricant type and lubricant concentration. If the concentration of either one of the lubricant types is increased, the dissolution becomes retarded. There was no difference between the effects of the inlet port on dissolution.

Thus, it can be concluded that over-lubrication had not occurred or at least it did not influence the final quality attributes of the tablets. Other input variables were not statistically significant in the model. It could be summarized that the selection of lubricant quality and concentration exerts an impact on dissolution profile only at the beginning of the dissolution process (at the dissolution time point of 2.5 min). However, there is no influence with respect to the quality of the final product. The dissolution profiles of the tablet are presented in Figure 24.

A B

Figure 24. The dissolution profiles of dissolved amount of paracetamol of the finished product (the error bars are not presented for the sake of clarity).

The shape of the dissolution profile was similar in all tablets. At the dissolution time point of 5 min, the dissolved amount of paracetamol was within the range of 95–105%.

This indicates that the tablets had dissolved quickly and met the requirements for immediate release tablets. The coefficient model and summary of fit (MLR) of the dissolved amount of paracetamol at the time point of 2.5 min are presented in supplementary data.

4.7.11 Creation of design space

Design Space (DS) is the multidimensional-space of formulative and process settings where the predefined product quality attributes remain within the specified limits when formulative and/or process settings are being changed. MLR models created from the formulative and process factors to model responses are the basis for the establishment of DS. The first element is the determination of the QTPP as shown in section 2.2. The MLR models were created from the responses described in QTPP (flowability, tablet strength, dissolution, ejection force), see supplementary data. Table 5 presents the target values of responses, the optimizer set points with factor settings and predicted response values, and the robust set point of factors and corresponding responses.

Table 5. The target values of responses, the optimizer set points with factor settings and predicted response values % outside of range, and the robust set point of factors and corresponding responses.

Target Optimizer initial set point

Robust set point (± SD)

% outside of range Responses

Flow rate (60 g/s) 2.5 2.6 5.0 (1.0) 0

Tablet strength (N) 147.1 149.1 145.1(4.0) 0

Dissolution (%) (at 2.5 min) 90 90.5 88.7 (2.4) 0.14

Ejection force (N) 425 451 435.5 (25.6) 0.98

Factors

Feed rate (kg/h) 5.0 9.9 (3.1)

Mixer (RPM) 500.2 816.1 (206.6)

Lubricant (magnesium

stearate or stearic acid) Stearic acid Stearic acid

Lubricant (%) 0.56 1.02 (0.29)

Inlet port (A or B) B B

In Fig 25, DS is presented as a function of the feed rate, mixer rpm and lubricant concentration that fulfilled the QTPP. Only the lubricant type StAc and the inlet port B satisfied the QTPP with respect to the qualitative variables. In the case of multi-dimensional DS, the alternative representation of DS is to describe a hypercube which defines the edges of the DS. The hypercube edges of DS are represented in Table 6.

Figure 25. Design Spaces for the continuous direct compression tableting process. Color scale represents the probability of failure to satisfy Quality Target Product Profile.

Table 6. The hypercube edges of the Design Space.

Factors Robust

setpoint

Hypercube low edge

Hypercube high edge

Fed rate (kg/h) 9.9 5.4 12.8

Mixer RPM 816.1 612.9 1132.3

Lubricant concentration

(%) 1.02 0.76 1.40

As a conclusion of DS, only qualitative factors of lubricant type StAc and mixer inlet port B satisfied the QTPP. With respect to the quantitative factors, relatively large ranges of factor settings fulfilled the QTPP. Thus, it can be said that robust factor settings can be easily found for the product which fulfills the QTPP.

4.8 BENEFITS OF USING THE CONTINUOUS MANUFACTURING PROCESS

One dimension in this study was to vary the batch size using the same equipment. The benefit of this set up is its capability to process different quantities of materials, i.e.

different feed rates. The duration of each run (= defined as a batch in this study) was approximately 20 min (the samples were collected only during 20 min runs). The total material consumption in the DoE study was 66.502 kg, varying for each run between 1.667 kg (feed rate 5 kg/h), 3.500 kg (feed rate 10.5 kg/h) and 5.333 kg (feed rate 16 kg/h) (presented in Supplementary data). Compared to the traditional batch manufacturing methods, the material consumption could probably be the same, if a batch of same size (production rate and time for tableting) would have been used. Usually in batch processing, to increase the batch size from 1.667 kg to 5.333 kg (i.e. increase of 3.2-fold), larger equipment are needed, and equipment would vary by design. Blenders of different size are available in the market and this DoE study could be possible to perform by using traditional batch processing. Still, creating the design space for different sizes of equipment needs more runs compared to continuous manufacturing (mixing parameters should be studied for different size of mixers). This means larger material consumption if batch processing is used. As shown in the results, the process parameters studied had only a very minor effect on the quality of the final product and the design space. Thus, to increase the batch size by varying the feed rate or process time with the same continuous manufacturing set-up is justified. For process and product development studies, the clear benefit is also the use of continuous manufacturing with lower production rate, i.e. lower material consumption. For industry, this kind of flexibility is an advantage when the batch size can be altered based on development phase or market demands by using continuous manufacturing process.

4.9 CONCLUSIONS

The aim was to determine robust and stable continuous manufacturing process settings, by devising a design space based on the investigation of lubrication dependent attributes which would influence continuous direct compression tableting of high dose paracetamol tablets. The obtained data verifies that a continuous manufacturing process is feasible for high dose direct compressible paracetamol tablets. The material properties of the main components have the most important influence on the process performance of a continuous manufacturing process. In this current study, the accuracy of feed rate of pre-blends was excellent. Interestingly, a high variation in the feed rates of lubricants did not exert a negative effect on the final product quality. It can be concluded that high variation in feed rates of lubricants could be eliminated with the good flow rate of main components and efficient blending unit operation. The type and concentration of lubricant had the greatest statistically significant effect on the studied responses. Stearic 67

acid was better than magnesium stearate. The selection of lubricant type and its concentration have an effect on the dissolution profile only at the beginning of the dissolution (at the dissolution time point of 2.5 min), the ejection force, tablet strength and flowability. It can be concluded that continuous direct compression tableting process is rather insensitive and robust with respect to variations in lubrication. The process parameters have only a slight influence on the tablet properties. Adjusting the total feed rate and selection of inlet port might alter the flowability properties and maximum ejection force. Based on the tableting results, even a high variation in the feed rate of lubricants had no influence on the tableting process or the quality of the final product.

However, the process parameters did not affect the quality of the tablet properties; in contrast, formulation parameters did exert a major influence on the properties of the end product. The created design space showed that this type of continuous manufacturing process line is suitable for producing high dose direct compressible paracetamol tablet which possessed a predetermined acceptable quality. However, in the future, longer runs should be completed to observe whether the quality of the product can be preserved as a function of time.

Acknowledgements

Authors gratefully acknowledge Vitabalans Oy for providing materials to this study. The authors also thank the PROMIS Centre consortium, funded by Tekes (A31473) (ERDF), the North Savo Centre for Economic Development, Transport and the Environment, for providing excellent research facilities. Jari Leskinen is acknowledged for his assistance in SEM images. Göran Salminen and Pasi Tapanainen are acknowledged for their assistance in setting up the dissolution and analysis method.

5 THE EFFECTS OF UNINTENTIONAL AND INTENTIONAL PROCESS DISTURBANCES ON TABLET QUALITY DURING LONG

CONTINUOUS MANUFACTURING RUNS

ABSTRACT

Several kinds of process disturbances can occur during (continuous) tablet manufacturing, i.e. unintentional or intentional disturbances. Long run-time continuous manufacturing studies are used to investigate the effects of intentional and unintentional deviations. In this study, the horizontal continuous manufacturing line included a double mixing - direct compression set up. The study consisted of two long duration test runs.

In the first run, the API (paracetamol) was fed in during the first feeding and blending stage with lubricant (Mg.Stear.) added during the second feeding and blending stage. In the second run, the API and lubricant feeding stages were reversed. The run protocol included a long run with several feeder re-fills and an overnight hold-time, continuing with the same API concentration followed by a change to a higher API concentration on the fly (without cleaning). The objectives of this pilot study were to determine the intentional (e.g. overnight hold time, product concentration change) and unintentional (e.g. equipment or software failures) deviations, which could affect the critical quality attributes (CQA's) of the final product and to create a deviation document which would reveal the changes that had occurred in the product concentration during the runs.

Another goal was to study the effect of feeding location of lubricant and API feeding. The CQA's were the assayed values of API, tablet strength, friability, tablet weight and its dissolution profile. The vacuum conveyors, which were needed to transfer materials in the horizontal set-up, were observed to introduce variation into the mass flow rates and feeding. Thus, there were significant challenges to ensuring a constant mass flow rate during the runs. One expected effect was that over-lubrication was evident when the lubricant was fed during the first feeding and mixing stage, resulting in a significantly reduced tablet strength and a slower dissolution of API. There were no observable trends over time in the process parameters or CQAs i.e. evidence of a stable process. The overnight hold-time did not affect the CQAs of tablets. Moreover, the variation in all CQAs was smaller after the overnight hold-time which was particularly unexpected. In conclusion, the results reveal that the process itself was able to produce a quality end product, but the set-up needs to be better designed and controlled to ensure a constant mass flow and prevent over-lubrication.

____________________________

2 Adapted with permission of Elsevier from: Taipale-Kovalainen K, Karttunen A-P, Niinikoski H, Ketolainen J, Korhonen O. The effects of unintentional and intentional process disturbances on tablet quality during long continuous manufacturing runs. European Journal of Pharmaceutical Sciences. 129: 10-20, 2019

5.1 INTRODUCTION

In recent years, there has been increasing interest towards the continuous manufacturing of drug products and this is being already utilized by some pharmaceutical companies.

Marketing Authorization (MA) has been approved for a few continuously manufactured products, i.e. Orkambi, Symdeko, Prezista (Continuous Manufacturing Symposium, 2014). There are several approaches to integrating different unit operations into a continuous manufacturing line (Badman and Trout, 2015; Singh et al., 2013). The manufacturing line can consist of many units such as feeders, mixers, vacuum conveyors, roller compactors and a tablet press. Most of the direct compression lines are constructed

In document Impact of formulation and process design on tablet quality in continuous manufacturing (sivua 60-0)