Docstoc

Quality by Design Application of NIR Spectroscopy for Formulation Development of Sustained Release Dosage Forms

Document Sample
Quality by Design Application of NIR Spectroscopy for Formulation Development of Sustained Release Dosage Forms Powered By Docstoc
					Quality by Design: Application of NIR Spectroscopy for Formulation Development of Sustained Release Dosage Forms

Raafat Fahmy, PH.D.

Food and Drug Administration
Center of Veterinary Medicine

Collaborative Research
• A collaborative research opportunity between FDA and the University of Maryland began as a result of the FDA’s new cGMP initiative for the 21st century. • The object of this initiative was to encourage the pharmaceutical industry to adopt more innovative and updated quality concepts (e.g., quality by design) when developing new products and to encourage innovation in the pharmaceutical manufacturing. • This collaboration is designed to help establish a scientific basis for understanding in-processes control using cutting edge technology, develop strategies to facilitate drug product development, and improve monitoring and control of product quality from real-time and upstream measurements.

Objective

The overall objective of this study is to investigate the application of NIR to predict the quality during the process including core tablets, coating and curing process. Investigated the influence of varying manufacturing parameters on the quality and processability to define critical quality parameters. In this project, the concept of using quality by design was applied through better understanding of manufacturing processes and determination of performance-based process controls and endpoints. Establish IVIVC based on NIR spectra.

 



Scope of Work
1- Manufacture orbifloxacin core tablets using

direct compression force and predict
manufacturing progress; • Compression force • Crushing strength • Content uniformity 2- Manufacture sustained release orbifloxacin coated tablets, predict coat thickness and release profile • Method for coating endpoint determination • Method for predicting in-vitro release profile 3- Curing process and predict curing stage

Chemometrics






Use of the statistical and mathematical techniques to analyze chemical data The entire process whereby data are transformed into information used for decision making Six steps for the Chemometric modeling

• • • • • •

Sample collection and scanning Feasibility study Sample selection for calibration  Calibration set  Validation set Construction of calibration model Assess the model (internal validation) Apply the model for prediction (external validation)

Tablet manufacturing process prediction

 Manufacturing orbifloxacin core tablets  Compression force
•
NIRS

 Crushing strength
• • • •
Hardness testing
NIRS

 Content uniformity
UV spectroscopy
NIRS

Chemical structure
F O COOH

CH3

H

CH3
H3C2

CH3

FH

C CH2 C CH2 C O C C O C O O O O CH3 C2H5 C2H4 N+(CH3)3 Cl-

CH C CH2 C NC O HO C HN O O CH CH3 2 5

N F

Eudrageit RS/RL
Number of Batches Calibration Set

Eudragit L Orbifloxacin
Validation Set 27

Prediction Set

Total

CH3 C CH2 Crushing strength O C O Content uniformity
Compression force

H CH3 2 C CH2 C 3 C O C O O O 8 C2H5

103 63 70
+

21

151 112 112

+
22 CH3 H CH2 C CH2 C 21
-

HCl
27 21

CH3

C2H4 N (CH3)3 O C O

C O O

Formulation Development

 Formulation of sustained release
orbifloxacin tablets

 Coat thickness prediction

• Fast release (T50 = 125 min) • Moderate release (T50 = 190 min) • Slow release (T50 = 260 min) • Optical microscopy • NIR • Dissolution testing • NIR

 Drug release prediction
 Curing process and predict curing stage

Components and composition
Ingredient
Orbifloxacin Avicel 102 Crosscarmellose Sodium Amorphous Fumed Silica Mg St Total Tablet weight Compression (lb) Hardness (kP)

Amount (g)
250 717.5
20

%
25 72
2

5 7.5 1000 0.3 g 1670 ± 75 10.9 ± 2.1

0.5 0.75 100

RL:RS
Materials
RL 30-D RS 30-D TEC PlassII (g) Water Total % DS

5:0
mL DS (g)
167 0 5 25 103 300 50 5 8 63 21%

4:1
mL
125 31 5 23 96 280

3:2
mL
100 67 5 25 103 300

2:3
mL
67 100 5 25 103 300

1:4
mL
33 134 5 25 103 300

0:5
mL
0 167 5 25 103 300

DS (g)
37 9 5 8 59 21%

DS (g)
30 20 5 8 63 21%

DS (g)
20 30 5 8 63 21%

DS (g)
10 40 5 8 63 21%

DS (g)
50 5 8 63 21%

Content Uniformity Determination
• A quantitative method using near infrared spectroscopy for the determination of the amount of active drug substance in a single tablet was developed. The linearity, specificity, of the method have been validated. • The development/validation of the method is based on: • Generation of tablet spectra • Creation of NIR calibration models • Validation of NIR models • Testing the method (comparing values of NIR vs. Reference analytical method using UV )

Feasibility study for Content Uniformity
A: 2nd derivative spectra of pure orbifloxacin and mixture of excipients.
0.0220 0.0156

A

0.0092

0.0028

Intensity

-0.0036

-0.0100

-0.0164

-0.0228

-0.0292

Orbifloxacin Excipients

-0.0356

-0.0420 1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400

Wavelength

B: overlaid spectra of orbifloxacin tablets with varying drug content (60-90 mg) is showing rank order of intensity with orbafloxacin concentration.

0 .0 2 7 0 0 .0 2 1 7 0 .0 1 6 4 0 .0 1 1 1

B

In te n s ity

0 .0 0 5 7 0 .0 0 0 4 -0 .0 0 4 9 -0 .0 1 0 2 -0 .0 1 5 6 -0 .0 2 0 9 -0 .0 2 6 2 1574 1587 1600 1613 1626 1639 1652 1665 1678 1691 1704 1717 1730 1743 1756 1769

W a v e le n g th

Content uniformity

Release Regression and Prediction
Factors 6
Loadings
1.5020 1.0918 0.6815

SEC (mg) 1.13

SECV (mg) 2.06

R2 0.983

SEP (mg) 1.36

Corr 0.988

Calibration Set : Calculated vs Lab Data
91.0

Validation Set : Calculated vs Lab Data

Predicted value (calculated)

Predicted value (calculated)

A

88.0 85.0 82.0 79.0 76.0 73.0 70.0 67.0 64.0 61.0 58.0 55.0

B

87.8 84.6 81.4 78.2 75.0 71.8 68.6 65.4 62.2 59.0

Intensity X 10^3

0.2713 -0.1390 -0.5492 -0.9595 -1.3697 -1.7800 -2.1902 -2.6005 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 1/0.0010 2200

C

55.0

59.0

63.0

67.0

71.0

75.0

79.0

83.0

87.0

91.0

59.0

63.0

67.0

71.0

75.0

79.0

83.0

87.0

91.0

Wavelength

Reference value (lab data)

Reference value (lab data)

PLS Loadings (A), calibration (B) and prediction (C) regressions for content uniformity

PCA for content uniformity
x 10
-3

Samples/Scores Plot of datacont

6 5 4

75Nmg

Scores on PC 2 (34.59%)

3
60mg

2 1 0
82.5mg 90mg 75mg 67.5mg

-1
78.75mg

71.25mg

-2 -5

-4

-3

-2 -1 0 1 Scores on PC 1 (52.52%)

2

3 x 10

4
-3

Score plot of first and second principal components for content uniformity study. PC1 shows the biggest variable which is the drug substance, PC2 shows other variabilities with less intensity. Manual compressed tablets versus rotary machine tablets.

Feasibility Study for Compression Force
0.0270 0.0215 0.0160

B

•Overlaid raw (A)
Intensity

0.0106 0.0051 -0.0004 -0.0058 -0.0113 -0.0168

• 2nd derivative (B) spectra of orbifloxacin tablet with compression force ranging from 750-2000 ponds. Higher the compression force, more the absorbance.

-0.0222 -0.0277 1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400

2500

Wavelength

0.7033 0.6117 0.5201 0.4285

A

Absorbance

0.3369 0.2454 0.1538 0.0622 -0.0294 -0.1210 -0.2126 1100

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400

2500

Wavelength

Compression force

Compression force regression and prediction
Factors 5 SEC (lb) 69.86 SECV (lb) 88.38 R2 0.984 SEP (lb) 59.48 Corr 0.994

Calibration Set : Calculated vs Lab Data
2391.0

Validation Set : Calculated vs Lab Data

Predicted value (calculated)

Predicted value (calculated)

2410.0 2210.0 2010.0 1810.0 1610.0 1410.0 1210.0 1010.0 810.0 610.0 410.0 410.0

A

2200.4 2009.8 1819.2 1628.6 1438.0 1247.4 1056.8 866.2 675.6 485.0

B

763.5

1117.0

1470.5

1824.0

2177.5

2531.0

Reference value (lab data)

485.0

866.2

1247.4

1628.6

2009.8

2391.0

Reference value (lab data)

(A) calibration (B) prediction regressions for compression force

Crushing strength
Crushing strength regression and prediction
Factors 5 SEC (KP) 0.55 SECV (KP) 0.78 R2 0.969 SEP (KP) 0.57 Corr 0.986

Calibration Set : Calculated vs Lab Data
13.0
13.0

Validation Set : Calculated vs Lab Data

Predicted value (calculated)

Predicted value (calculated)

12.0 11.0 10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 2.0

A

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

B

Reference value (lab data)

4.2

6.4

8.6

10.8

13.0

2.5

Reference value (lab data)

4.0

5.5

7.0

8.5

10.0

11.5

13.0

PLS (A) calibration (B) prediction regressions for crushing strength

Results and discussion

PLS was the best model to be used. Three PLS models with the best statistical parameters.


 

The SEC and SEP for content uniformity were 1.13 and 1.36 mg.
PLS models present SEC and SEP equal to 69.86 and 59.48 lbs for compression force, and 0.55 and 0.57 KP for crushing strength of the tablets. NIR spectroscopy in combination with multivariate modeling is a powerful, rapid and nondestructive technique that could reliably predict content uniformity, compression force and crushing strength for orbifloxacin tablets during production.

Experimental
Final coating process

Three different Coat (RL:RS) to the core tablets. 5:0, 4:1, and 3:2 Coating conditions
Used Pan Coater

Inlet temperature: 45-50 °C

Outlet temperature: 30-32 °C
Atomized air pressure: 2 Bar, 1.478 kg/cm2 RPM: 11 Flow rate of polymer solution: 0.8 - 2 ml/min Batch size: 0.5 kg

Final formulations


Formulation of sustained release orbifloxacin tablets

• • •

Fast release (T50 = 125 min) Moderate release (T50 = 190 min) Slow release (T50 = 260 min)
Orbifloxacin tablets for in vivo study 100.0 90.0 80.0 70.0

100 90 80

Release profile for orbifloxacin tablets coated with various ratios of Eudragit blends

% Drug release

70

% Release

60 50 40 30 20 10 0 0 100 200 Time (min) 300
Orbiflo xacin IR RL:RS 5:0 RL:RS 4:1 RL:RS 3:2 RL:RS 2:3 RL:RS 1 :4 RL:RS 0:5

60.0 50.0 40.0 30.0 20.0 10.0 0.0 0 50 100 150 200 250 300 350 Time (min)
Orbifloxacin IR % Release RL:RS 5:0 % Release RL:RS 2:3 % Releae RL:RS 1:4

Regression and prediction Dissolution profile
Calibrations (A) PLS (1 hr)
2nd D (segment: 10 nm) 2nd D (segment: 20 nm) SG (2nd D, quadratic) BC (1800 nm)
103.0

Pretreatment
NPS (segment: 20 nm) NPS (segment: 20 nm) NPS (segment: 15 nm) SG nd D, quintic) (2
Calibration value (calculated)
102.0 96.5 91.0 85.5 80.0 74.5 69.0 63.5 58.0 52.5 47.0 41.5 36.0 30.5 25.0 25.0 36.0 47.0 58.0 69.0 80.0 91.0

Region (nm)
SNV 1150-2200

(B) PLS (2 hrs)
(C) PLS (3 hrs) (D) PLS (4 hrs)
calibration value (calculated)
103.0 96.0 89.0 82.0 75.0 68.0 61.0 54.0 47.0 40.0 33.0 26.0 19.0 12.0 5.0 5.0

SNV
SNV SNV

1150-2350
1150-2350 1150-2350

A

Calibration value (calculated)

95.0 87.0 79.0 71.0 63.0 55.0 47.0 39.0 31.0 23.0 15.0 15.0

B

C

Calibration value (calculated)

103.0 95.9 88.8 81.7 74.6 67.5 60.4 53.3 46.2 39.1 32.0

D

Reference value (lab data)

19.0

33.0

47.0

61.0

75.0

89.0

103.0

Reference value (lab data)

32.6

50.2

67.8

85.4

103.0

Reference value (lab data)
102.0

102.0

Reference value (lab data)
105.0

32.0

46.2

60.4

74.6

88.8

Validation Set : Calculated vs Lab Data

102.0

103.0

Predicted value (calculated)

Predicted value (calculated)

92.5 83.0 73.5 64.0 54.5 45.0 35.5 26.0 16.5 7.0 7.0

Predicted value (calculated)

Predicted value (calculated)

A

94.4 85.8 77.2 68.6 60.0 51.4 42.8 34.2 25.6 17.0

B

94.4 86.8 79.2 71.6 64.0 56.4 48.8 41.2 33.6 26.0

C

100.0 95.0 90.0 85.0 80.0 75.0 70.0 65.0 60.0 55.0 50.0 45.0 40.0 35.0 35.0

D

Reference value (lab data)

26.0

45.0

64.0

83.0

102.0

17.0

Reference value (lab data)

34.2

51.4

68.6

85.8

26.0

Reference value (lab data)

41.2

56.4

71.6

86.8

Reference value (lab data)

45.0

55.0

65.0

75.0

85.0

95.0

105.0

Dissolution profile prediction

Release PLS (1 hr) PLS (2 hrs) PLS (3 hrs) PLS (4 hrs)

Factors % SEC 7 6 6 7 2.39 2.86 2.82 2.85

R2 0.993 0.985 0.980 0.973

% SECV % SEP Bias Adj SEP Corr 2.54 3.06 2.96 3.83 2.05 2.21 2.46 3.01 -0.39 -0.36 -0.33 0.44 0.998 0.996 0.992 0.986

Slope Adj

Bias Adj

SEP Pssbl 2.01 2.18 2.41 2.98

0.995 ± 0.011 -0.23 ± 0.48 0.9995 ± 0.015 -0.34 ± 0.76 0.983 ± 0.020 0.996 ± 0.028 0.62 ± 1.19 0.71 ± 1.89

100
Orbiflo xa c in R L:R S 5:0 R L:R S 4:1 R L:R S 3:2 R L:R S 2:3 R L:R S 1:4 R L:R S 0:5

100
Orbiflo xa c in R L:R S 5:0 R L:R S 4:1 R L:R S 3:2 R L:R S 2:3 R L:R S 1:4 R L:R S 0:5

80
Measured Release (%)

80

60

Predicted Release (%)
2 3 4

60

40

40

20

20

0 0 1 Time (hr)

0 0 1 2 Time (hr) 3 4

NIRS and chemometrics modeling for Coating
 Reference method
•Optical microscopy

 Tablets analyzed by Foss NIRSystems 6500
(RCA Diffuse Reflectance)

 Wavelength range: 400 - 2500 nm  Samples collected every 2 nm  Each tablet was scanned on both sides  For thickness study both sides of each
independent sample

90 μm

tablet were used as

 For

curing and dissolution study the spectra of both sides of each tablet averaged into one spectra regression models was used for both coat thickness and dissolution

 PLS

Feasibility study
 
Feasibility study Presence of non interference absorbance for both coat and core tablet
0.0738 0.0552




Core tablet

Absorbance at 1600-1700 nm and 22002400 nm corresponding to C-H+C-H first overtone and combination Increase in coating thickness results in increase in absorbance in the region
Intensity

0.0366 0.0180 -0.0007 -0.0193 -0.0379 -0.0565 -0.0751 -0.0937 -0.1123 1100

Coat

corresponding to C-H+C-H first overtone
and combination
0.8982 0.7907

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300

2400

2500

Wavelength

0.0385

0.0235
0.6831 0.5756

Absorbance

0.4681 0.3606 0.2531

Intensity

80 μm 60 μm 40 μm 20 μm 0 μm

0.0085

-0.0065

-0.0215
0.1456 0.0380 -0.0695 -0.1770 1100

-0.0365

0 μm 20 μm 40 μm 60 μm 80 μm

1300

1500

1700

1900

2100

2300

-0.0515 1580

1680

1780

1880

1980

2080

2180

Wavelength

Wavelength

Film Coated
• Incomplete film formation affects the drug dissolution rate. Thus, film coat curing is critical to coating quality, product aging (drug release stability) and overall product performance. • Presently, the process of film coat curing is still not well understood, and this lack of knowledge leads to empirical formulation techniques and process method development. This trial-and-error approach is not the best way to develop a formulation and may lead to a formulation and process that are optimal for a narrow range of materials and processing conditions; as result, these formulations are not very robust and may not reliably produce a quality tablet.

Film Coated
• Currently, there are no on-line methods that can determine the curing end-point based upon the actual formation of a fully dense coat. • In addition, being able to determine in realtime when a fully dense coat has formed would prevent over curing of the coat, which could lead to unwanted drug migration into the coat and to drug stability problems because curing occurs at elevated temperatures.

Film Coated
• To study film coat curing, the film coats (either cast or on tablets) were cured at 40, 50 or 60 °C for 2 to 48 hrs. At each time point. • samples were taken and scanned using NIR, and in parallel differential scanning calorimetry (DSC) and hot stage optical microscopy measurements were taken. • As the extent of curing progressed, the 2nd derivative of the spectra showed a reduction in the peak at 1908 nm. As expected, an increase in curing temperature results in faster reduction in the peak at 1908 nm.

DSC results for cured films

-0.2

––––––– ––––––– ––––––– ––––––– –––––––

37RSRLL151x0test1 50RSRLL151x30mintest1 60RSRLL151x30mintest1 40RSRLL151x0test1 RTDSC1512

-0.4

Heat Flow (W/g)

-0.6

-0.8

60 °C 50 °C 40 °C 37 °C RT

-1.0 -20
Exo Up

0

20

40

60

80
Universal V2.6D TA Instruments

Temperature (°C)

DSC results for cured films; Eudragit RL:RS 1:4 cured for 30 min at RT, 37, 40, 50 and 60 C after 2 hours.

RL:RS 1:4 cured at 40 °C

0.1342

0.0957

0.0572

A
Intensity

0.0166

0.0086
0.0187

B
Heat Flow (W/g)

-0.2

-0.4

C
17 hr 1 hr RT
0 20 40 60

––––––– 40DSC41x11 ––––––– RTdsc411 ––––––– 40DSC41x41

-0.0197

Intensity

0.0006
-0.6

-0.0582

-0.0967

-0.0074

-0.1352

-0.1737

-0.0154

40 hr 4 hr 3 hr 2 hr 1 hr RT
1841 1860 1879 1898 1917 1936 1955 1974

-0.8

-1.0

-0.2122

-0.2507 1100

-0.0234 1822
1300 1500 1700 1900 2100 2300 2500

-1.2 80
Universal V2.6D TA Instruments Exo Up

Wavelength

Wavelength

Temperature (°C)

Eudragit RL:RS 30-D 1:4. Panel A: 2nd derivative full NIR spectra during curing at 40 oC, Panel B: magnified NIR spectra for peak 1908nm, Panel C: DSC overlay profile at different curing time for the same film

RL:RS 1:4 cured at 50 °C

0.1297

0.0950

0.0604

A
Intensity

0.0187

0.0258

0.0087

B
24 hr 2 hr 1 hr 0.5 hr 0.5hr (37oC) RT
Heat Flow (W/g)

-0.4

C
-0.6 -0.8

––––––– RTdsc411 ––––––– 50DSC41x11 ––––––– 50DSC41x21 ––––––– 50DSC41x41

-0.0088

Intensity

-0.0013

-0.0434

-0.0780

-0.0113

-0.1126

-0.0213
-0.1472 -0.1818

-1.0

17 hr 2 hr 1 hr RT

-0.2164 1100

-0.0313 1826 1839 1852 1865 1878 1891 1904 1917 1930 1943 1956 1969
1300 1500 1700 1900 2100 2300 2500

-1.2 -20
Exo Up

Wavelength

Wavelength

0

20

40

60

80
Universal V2.6D TA Instruments

Temperature (°C)

Eudragit RL:RS 30-D 1:4. Panel A: 2nd derivative full NIR spectra during curing at 50 oC, Panel B: magnified NIR spectra for peak 1908nm, Panel C: DSC overlay profile at different curing time for the same film.

RL:RS 1:4 cured at 60 °C

At 0 min

30 min

60 min

24 hr

0.1232

0.0919

0.0606

A
Intensity

0.0127 0.0027

0.0292

B
20 hr 1 hr 0.5 hr 0.5 hr (37oC) RT
Heat Flow (W/g)

-0.2

-0.4

C

––––––– RTdsc411 ––––––– 60DSC41x302 ––––––– 60DSC41x12 ––––––– 60DSC41x41

-0.0021

Intensity

-0.0334

-0.0073 -0.0173

-0.6

-0.0648

-0.0961

-0.1274

-0.0273 -0.0373 1818
1300 1500 1700 1900 2100 2300 2500

-0.8

-0.1588

-0.1901 1100

1844

1870

1896

1922

1948

1974

2000

-1.0 -40 -20 0 20
Exo Up

17 hr 1 hr RT
40 60 80
Universal V2.6D TA Instruments

Wavelength

Wavelength

Temperature (°C)

Eudragit RL:RS 30-D 1:4. Panel A: 2nd derivative full NIR spectra during curing at 60 oC, Panel B: magnified NIR spectra for peak 1908nm, Panel C: DSC overlay profile at different curing time for the same film.

Regression and prediction Combined
Combined Regression and Prediction
Factors 7 SEC (μm) 2.91 SECV (μm) 3.16 R2 0.989 SEP (μm) 3.05 Corr 0.994
Validation Set : Calculated vs Lab Data

Loadings
2.4221 1.5268 0.6315

Calibration Set : Calculated vs Lab Data

Predicted value (calculated)

96.3 85.6 74.9 64.2 53.5 42.8 32.1 21.4 10.7 0.0 0.0

-0.2638 -1.1591 -2.0544 -2.9497 -3.8450 -4.7403 -5.6356 -6.5309 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 2100 1/0.0010 2200

Predicted value (calculated)

A

B

98.1 87.2 76.3 65.4 54.5 43.6 32.7 21.8 10.9 0.0

Intensity X 10^3

C

Wavelength

Reference value (lab data)

21.4

42.8

64.2

85.6

107.0

0.0

Reference value (lab data)

21.8

43.6

65.4

87.2

109.0

PLS Loadings (A), calibration (B) and prediction (C) regressions for coating thickness

Curing coated tablets Orbifloxacin
Samples/Scores Plot of Orb 1:4 cured PCA.xls

0.15

60C 2hr

0.1

Scores on PC 3 (10.09%)

50C 2hr 0.05
40C 2hr

40C 17hr

0 50C 4hr -0.05 RT -0.1 0.2 0 0.1 0.05 0 -0.05 -0.1 -0.15 0.2 Scores on PC 2 (22.15%) -0.2

0.15

Scores on PC 1 (36.79%)

3-D score plot for coated orbifloxacin tablets (1:4) cured at 40 °C for 0, 2 and 17 hrs, at 50 ° C for 2 and 4 hrs, and at 60 °C for 2 hrs

Conclusions
• Principal component analysis on free films, uncoated tablets (used to account for any tablet matrix changes) and coated tablets provide valuable information. Although the NIR spectral variation and shifts for uncoated and coated tablets are not as large as cast films, the score plots show that the curing process significantly affects the coated tablet formulations. Possible factors responsible for the spectral variation include the removal of residual water, further inclusion of additives such as plastiziers, polymer particle coalescence and polymer chain diffusion across the latex particle interface . • Principal component analysis was able to successfully identify and qualify the stage of curing for film coated tablets based on curing temperature and duration. • Partial least square (PLS) models were developed and validated for tablets coated with varying ratios of Eudragit RL:RS these models were able to reliably predict the extend of curing and the curing endpoint.

Conclusions
• The curing process and its impact on NIR spectra is case specific. Factors such as coating polymer, plasticizer, glidant, coat aging and drug polymer interactions are all examples of things that can affect the calibration model and consequently model predictability and robustness. • In addition, the presence of micro-fractures and coat thickness inhomogeneity can result in unwanted swelling that can affect drug dissolution, which is independent of the coat composition and the extent of curing and thus won’t be detected by the current PLS models. • Despite these limitations, NIR has been successfully used to monitor Eudragit polymer coat curing which illustrates the potential of NIR as tool for monitoring the coating process.

Summary

Various immediate and sustained release dosage forms were manufactured and the process was monitored using NIRS The best-fit models for compression force, crushing strength, content uniformity, coat thickness and drug release were developed. This study shows the potential of NIRS as an alternative method to conventional methods
Partial least square (PLS) models were developed and validated for tablets coated with varying ratios of Eudragit RL:RS these models were
able to reliably predict the extend of curing and the curing endpoint. The results of this study expand the application NIR in pharmaceutical sustained release products. This study opens new research opportunities and applications for controlling the process & and open some of the black boxes.






Acknowledgments

 Dr S. Hoag, PhD


e-mail: shoag@rx.umaryland.edu

 Simin Tabasi
e-mail: shass001@umaryland.edu


				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:10
posted:9/14/2009
language:English
pages:37
pptfiles pptfiles
About