1 / 9 �

A Novel Approach to Cleaning Validation for Pharmaceutical Manufacturing: Using Online Supercritical Fluid Extraction/Supercritical Fluid Chromatography �

Kenichiro Tanaka1, William Hedgepeth1, Daisuke Nakayama2, Hidetoshi Terada2, Minori Nakashima2, Tadayuki Yamaguchi2

1: Shimadzu Scientific Instruments, Inc., MD, USA, 2: Shimadzu Corporation, Kyoto, Japan�

2 / 9 �2 / 12 �

Introduction

Cleaning validation is necessary to establish the quality and safety of pharmaceutical drug products. In cleaning validation protocols, direct sampling is performed with swabs, which are sticks with textiles at one end. The sample on the swab after swabbing the surface of equipment is analyzed with a TOC analyzer and HPLC. Recently, HPLC has been more preferable because of the growing need for the individual analysis of products. Before the HPLC analysis, manual processes such as a sample extraction and a sample condensation are required. Such manual processes may affect to the quality of results. Thus, we evaluated the application of a novel on-line supercritical fluid extraction/chromatography system for the cleaning validation.�

3 / 9 �3 / 12 �

Experimental

Materials and Reagents A commercially available detergent containing alkylbenzene sulfonates was diluted with methanol and used as a standard sample. For the test of sample extraction, the sample was dropped onto a swab (ITW Texwipe, USA). System Nexera UC system (Shimadzu corporation, Japan) was used for both the screening of the method using supercritical fluid chromatography (SFC) (Figure 1A) and the supercritical fluid sample extraction (SFE) followed by SFC directly (SFE/SFC) (Figure 1B). The schematic diagram for static and dynamic extraction of supercritical fluid extraction unit is shown in Figure 1C.

4 / 9 �4 / 9 �

Figure 1: Flow diagrams of Nexera UC system

System

5 / 9 �5 / 12 �

Analytical Conditions

Column : Shim-pack UCX series columns (250 mm L. x 4.6 mm I.D., 5 µm) (I) UCX-RP (ODS with polar group), (II) UCX-GIS (ODS), (III) UCX-SIL, (IV) UCX-DIOL

Mobile Phase : A: CO2; B: Methanol Time program : Shown in the figure Flow Rate : 3.0 mL/min Column Temp. : 40°C Back pressure : 15 MPa Wavelength : 220 nm Injection Vol. : Shown in figure

Table 1: SFC Analytical conditions �

6 / 9 �6 / 12 �

[Sample Preparation] A total of 10 to 500 µg standard samples in methanol were dropped onto a total of three swabs. The swabs were enclosed into an extraction vessel and set to the SFE/SFC unit. [Static extraction] Extraction time : 3 min Mobile phase : A: CO2; B: 0.1% (w/v) ammonium formate in methanol B conc. : 10% Flow rate : 3.0 mL/min Back pressure : 15 MPa [Dynamic extraction] Extraction time : 3 min Mobile phase : A: CO2; B: Methanol B Conc. : 10% Flow rate : 3.0 mL/min Back pressure : 15 MPa [SFC] Column : Shim-pack UCX-SIL (250 mm L. x 4.6 mm I.D., 5 µm) Mobile Phase : A: CO2; B: Methanol Time program : 10%B (0-2 min), 10-60%B (2-7 min), 60%B (7-9 min), 10%B (9-13 min) Flow Rate : 3.0 mL/min Column Temp. : 40°C Back pressure : 15 MPa Wavelength : 220 nm

Table 2: On-line SFE/SFC Analytical conditions �

Analytical Conditions

7 / 9 �7 / 12 �

Results Analytical method development�The analytical method for SFC was developed by screening four columns and gradient conditions. The UCX-SIL column showed excellent peak shape (Figure 2) and was used for further analysis.

Figure 2: Comparison of columns for SFC analysis of standard detergent sample

0.0 � 1.0 � 2.0 � 3.0 � 4.0 � 5.0 � 6.0 � 7.0 � 8.0 � 9.0 � 10.0 � 11.0 � 12.0 � min �

-20�

-10�

0 �

10 �

20 �

30 �

40 �

50 �

60 �

70 �

80 �

90 �

100 �

110 �

120 �

130 �

mAU �

0.0 � 1.0 � 2.0 � 3.0 � 4.0 � 5.0 � 6.0 � 7.0 � 8.0 � 9.0 � 10.0 � 11.0 � 12.0 � min �

-10�

-5 �

0 �

5 �

10 �

15 �

20 �

25 �

30 �

35 �

40 �

45 �

50 �

55 �

60 �

mAU

0.0 � 1.0 � 2.0 � 3.0 � 4.0 � 5.0 � 6.0 � 7.0 � 8.0 � 9.0 � 10.0 � 11.0 � 12.0 � min �

-10�

0 �

10 �

20 �

30 �

40 �

50 �

60 �

mAU �

0.0 � 1.0 � 2.0 � 3.0 � 4.0 � 5.0 � 6.0 � 7.0 � 8.0 � 9.0 � 10.0 � 11.0 � 12.0 � min �

-10�

0 �

10 �

20 �

30 �

40 �

50 �

60 �

70 �

80 �

90 �

100 �

110 �

120 �

130 �

mAU �

(l) UCX-RP

0 10 20 30 40 50

0 5 10 15

B C

onc.

[%]

Time [min]

(lll) UCX-SIL

(ll) UCX-GIS (lV) UCX-DIOL

5-50%B (0-7 min)

Time program:

5-40%B (0-7 min)

5-30%B (0-7 min)

RT : 7.87 min N : 16454 Tf : 1.767

RT : ND N : ND Tf : ND

RT : 8.56 min N : 59056 Tf : 0.693

RT : 7.96 min N : 48053 Tf : 0.692

Abbreviations: RT : Retention time N : Theoretical plate Tf : Tailing factor (5%)

Injected sample: 1 µL of 1% Standard sample

8 / 9 �8 / 12 �

Results The time program was developed in consideration of on-line SFE/SFC (data not shown). The optimized time program and the result of performance evaluations were shown in Figure 3.

Figure 3: Performance evaluations of SFC analytical method

0.0� 1.0� 2.0� 3.0� 4.0� 5.0� 6.0� 7.0� 8.0� 9.0� min �

-20�

-10�

0�

10�

20�

30�

40�

50�

60�

mAU �

0.0 � 1.0 � 2.0 � 3.0 � 4.0 � 5.0 � 6.0 � 7.0 � 8.0 � 9.0 � min �-10�0 �

10 �20 �30 �40 �50 �60 �70 �80 �90 �

100 �110 �

mAU �(l) Linearity test

(ll) Reproducibility test

1% Sample concentration:

0.50% 0.10% 0.05% 0.01% 0.005% 0%

0.00� 0.25� 0.50� 0.75� 1.00�0�

2.0�

4.0�

6.0�

8.0�

10.0�

12.0�

Peak Area (x100,000)�

Conc. (%)�

R2 = 0.998

Sample

Reproducibilities (n=6, %RSD)

Retention time Area

0.10% standard sample

0.17 2.46

Time program: 10%B (0-2 min), 10-60%B (2-7 min), 10%B (9-13 min)

Injected sample: 2 µL of 0 to 1% Standard samples

0 20 40 60

0 2 4 6 8 10 12 14

B C

onc.

[%]

Time [min]

9 / 9 �9 / 12 �

Results

On-line SFE/SFC analysis of detergent in swab �Based on the optimized SFC analytical condition, an on-line SFE/SFC analytical condition was developed and its performance was evaluated (Figure 4).

Figure 4: On-line SFE/SFC analysis of standard detergent sample-spiked swabs

(l) Sequential diagram of on-line SFE/SFC

0 10 20 30 40 50 60

-6 -3 0 3 6 9 12

B C

onc.

[%]

Time [min]

SFC Analysis (13 min)

Dynamic Extraction

(3 min)

Static Extraction

(3 min)

10 / 9 �10 / 12 �

Results

Figure 4: On-line SFE/SFC analysis of standard detergent sample-spiked swabs

50 µg

100 µg

200 µg

500 µg

Spiked sample amount:

(ll) Linearity test

0 100 200 300 400 µg/Swab 0.00

0.25

0.50

0.75

1.00

1.25

1.50

Peak Area (x10,000,000)

500

R2 = 0.996

SFC Analysis

Dynamic Extraction

Static Extraction

0 �

0.5 �

1.0 �

1.5 �

2.0 �

2.5 �

3.0�

mAU �

-5.0 � -2.5 � 0.0 � 2.5 � 5.0 � 7.5� min �

Alkylbenzenesulfonate

10 µg

20 µg

(lll) Reproducibility test

SFC Analysis

Dynamic Extraction

Static Extraction

0 �

0.5 �

1.0 �

1.5 �

2.0 �

mAU �

-5.5 � -2.5 � 0.0 � 2.5 � 5.0 � 7.5� min �

Sample

Reproducibilities (n=5, %RSD)

Retention time Area

100 µg standard sample

0.19 5.76

11 / 9 �11 / 9 �

Conclusion

The evaluation results showed that the on-line SFE/SFC can provide reliable data for the cleaning validation for pharmaceutical manufacturing. The benefit of on-line SFE/SFC analysis is not only data reliability but also the convenience and safety gained by eliminating the manual extraction process. It may streamline laboratory processes and improve productivity and safety of pharmaceutical manufacturing facilities.

The sample was provided by DAIICHI SANKYO COMPANY, LIMITED. Acknowledgement�

12 / 9 �

Thank you for viewing this presentation. Should you have any questions or require additional information about our research, products or services, please visit our support page: www.ssi.shimadzu.com/support/

@shimadzussi

Follow us on Twitter

Need More Info?