1. VII. Special Case Compound Classes
Best Practices for OINDP Pharmaceutical Development Programs Leachables and Extractables
VII. Special Case Compound Classes
PQRI Leachables & Extractables Working Group
PQRI Training Course
September 20-21, 2006
Washington, DC

2. “Special Cases” PAHs - Polyaromatic Hydrocarbons N-Nitrosamines
Also referred to as PNAs (Polynuclear Aromatics)
N-Nitrosamines
2-Mercaptobenzothiazole

3. PAHs/PNAs as Leachables in OINDP
Historically, the primary source of PNAs is carbon black which is used as a filler in certain types of rubber (mostly sulfur cured).
There is some potential for other PNA sources (e.g. naphthalene contamination).
Some PNAs are known or suspect cancer causing agents (e.g. benzo(a)pyrene). FDA interest in MDIs traces back to the late 1980s.
Levels of PNAs in MDIs which employ “black rubber” seals are typically on the order of ng to low µg/canister.
The FDA historically requires that all elastomers in MDIs be evaluated and controlled for PNAs.
Analytical methods typically involve GC/MS.
September 2006
PQRI Training Course

4. PNAs Typically Analyzed and Controlled (EPA Method 610 list)
Naphthalene
Acenaphthylene
Acenaphthene
Fluorene
Phenanthrene
Anthracene
Fluoranthene
Pyrene
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(e)pyrene
Benzo(a)pyrene
Indeno(123-cd)pyrene
Dibenzo(ah)anthracene
Benzo(ghi)perylene
September 2006
PQRI Training Course

5. Structures of Some Typical PNAs
naphthalene
phenanthrene
pyrene
benzo(a)pyrene
benzo(ghi)perylene

6. Trace Organic Analysis
September 2006
PQRI Training Course

7. PNA Analysis in Rubber – Possible Method
Slice (or grind) a measured weight of critical rubber components.
Add prepared rubber to a boiling flask with a measured volume of organic solvent (e.g. toluene).
Extract via reflux for a pre-optimized time period (likely 24 hours or greater).
Remove solvent and reduce in volume.
Analyze by GC/MS (for example).
Note that internal standards can be added at various points in the overall process.
September 2006
PQRI Training Course

8. PNA Analysis in a Suspension Metered Dose Inhaler Drug Product - Possible Method
Cool sample MDI canisters (one or several for a composite sample) over dry ice.
Open canister(s) and filter contents to remove suspended drug particles.
Note that filter assembly and catch flask must be cold.
Wash filter contents with organic solvent.
Evaporate sample to dryness.
Dissolve residue in a measured quantity of a suitable organic solvent (e.g. toluene).
Analyze by GC/MS (for example).
Note that internal standards can be added at various points in the overall process.
September 2006
PQRI Training Course

9. September 2006
PQRI Training Course

10. A GC/MS System
September 2006
PQRI Training Course

11. Reference
The information summarized in the following slides related to PNA leachables studies is detailed in the following reference:
Norwood, D.L., Prime, D., Downey, B.P., Creasey, J., Sethi, S.K., Haywood, P., Analysis of polycyclic aromatic hydrocarbons in metereds dose inhaler drug formulations by isotope dilution gas chromatography/mass spectrometry, Journal of Pharmaceutical and Biomedical Analysis, 13(3), , 1995.
September 2006
PQRI Training Course

12. GC/MS Analysis of Target PNAs
September 2006
PQRI Training Course

13. EI Spectra of Pyrene and D10-Pyrene
Note stability of the molecular ions
Note the characteristic presence of doubly-charged molecular ions
September 2006
PQRI Training Course

14. Benzo(e)pyrene and Benzo(a) pyrene
September 2006
PQRI Training Course

15. Representation Linearity1 and Linearity of Recovery2 Results for a Drug Product Assay ( µg/inhaler)
Target PNA
Slope
Intercept
Correlation
Naphthalene
0.9731
0.045
0.9998
0.9932
0.033
Acenaphthene
0.658
0.021
0.663
0.014
0.9992
Phenanthrene
1.059
0.067
0.9997
1.061
0.080
0.9990
Fluoranthene
1.070
0.059
1.085
0.069
0.9996
Pyrene
1.034
1.040
0.072
Benzo(e)pyrene
1.485
0.056
1.487
0.087
0.9994
Benzo(a)pyrene
0.827
0.032
0.853
0.054
Benzo(ghi)perylene
1.320
-0.016
1.350
-0.028
September 2006
PQRI Training Course

16. Limit of Detection/Quantitation Results for Selected Target PNAs
(ng/inhaler)
Limit of Quantitation
Naphthalene
0.7 4
Acenaphthylene
Fluorene
0.9 5
Phenanthrene
Flouranthene
Pyrene
Benzo(ghi)perylene 6 30
September 2006
PQRI Training Course

17. PNA Profile of an MDI Drug Product
September 2006
PQRI Training Course

18. PNAs as Leachables in Metered Dose Inhalers
Target PNA
Product A
(µg/inhaler)
Product B
Product C
Naphthalene
0.29
0.15
0.57
Acenaphthylene
0.43
0.22
0.45
Acenaphthene ND
Fluorene
<0.05
Phenanthrene
1.96
0.88
2.14
Anthracene
0.10
0.12
Fluoranthene
1.20
0.53
1.37
Pyrene
1.26
0.61
2.13
Benzo(a)anthracene
Chrysene
Benzo(b)fluoranthene
Benzo(k)fluoranthene
Benzo(e)pyrene
0.08
<0.025
Benzo(a)pyrene
Dibenzo(ah)anthracene
Indeno(123-cd)pyrene
Benzo(ghi)perylene
0.03
0.06
Total
5.50
2.45
7.02
September 2006
PQRI Training Course

19. N-Nitrosamines as Leachables in OINDP
Historically, the formation of “nitrosamines” in rubber involves sulfur curing agents (e.g. thiurams).
The issue of N-nitrosamines in rubber goes back to late 1970s/early 1980s with concern over their presence in baby bottle rubber nipples. FDA became involved in the issue. Official analytical methods for rubber developed and validated.
FDA interest in MDIs (and other OINDP) traces to the early 1990s.
Levels of nitrosamines in MDIs which employ “black rubber” seals are typically on the order of ng/canister.
The FDA historically requires that all elastomers in MDIs be evaluated and controlled for nitrosamines.
Analytical methods typically involve GC with “Thermal Energy Analysis” detection (GC/TEA).
September 2006
PQRI Training Course

20. Target N-nitrosamines
N-nitrosodimethylamine
N-nitrosodiethylamine
N-nitrosodi-n-butylamine
N-nitrosomorpholine
N-nitrosopiperidine
N-nitrosopyrrolidine

21. N-nitrosamine Formation
X = NO+, N2O3, N2O4, NOZ (Z = Cl, Br, I, thiocyanate)
September 2006
PQRI Training Course

22. N-nitrosamine Analysis in Rubber (AOAC Method 987.05)
Place 5g cut rubber sample in 250mL flask with 100mL methylene chloride and 100mg propyl gallate, and hold for 17-18h.
Transfer solvent and rubber sample to a Soxhlet extractor.
Spike in internal standard.
Extract for 1 hour.
Add 100mL 5N NaOH and 2g Ba(OH)2 to flask and carefully distill methylene chloride (discard). Continue distilling 70mL of aqueous distillate into a sepratory funnel.
Add 300mg anhydrous Na2CO3 to funnel, followed by 50mL methylene chloride. Extract (repeat twice more). Combine extracts in sepratory funnel.
Pass through anhydrous Na2SO4 (to dry), into a Kuderna_Danish apparatus (with appropriate washes).
Concentrate to approximately 4mL.
Remove from KD and further concentrate to 1.0mL with a nitrogen stream.
Analyze by GC/TEA.
September 2006
PQRI Training Course

23. N-nitrosamine Analysis in Rubber (AOAC Method 987.05)
Steam distillation
Soxhlet extraction
Extract concentration
Image provided by Rubber Consultants
September 2006
PQRI Training Course

24. Principles of Thermal Energy Analysis Detection
N-nitrosamines elute from a GC column into a pyrolyzer, where they undergo pyrolysis and release nitrosyl radicals (NO.). The pyrolysis temperature is set low enough so that nitro-compounds will not pyrolyze.
Nitrosyl radicals are then oxidized with ozone in a reaction chamber to give electronically excited NO2*.
The NO2* decays back to ground state releasing a photon at a characteristic wavelength.
This process is known as “chemiluminescence”.
Sensitivity is further increased through use of a filter-photometer for detection.
September 2006
PQRI Training Course

25. A GC/TEA System – Schematic Diagram
ozone
pyrolysis
vacuum
cold trap
450oC
detector GC
-130oC
electronics
September 2006
PQRI Training Course

26. A GC/TEA System Image provided by Rubber Consultants September 2006
PQRI Training Course

27. A GC/TEA System Image provided by Cardinal Health September 2006
PQRI Training Course

28. GC/TEA N-nitrosamines - Separation
September 2006
PQRI Training Course

29. GC/TEA N-nitrosamines - Sensitivity
10 ng/mL
September 2006
PQRI Training Course

30. Some Typical Limit of Detection/Quantitation Results for Target N-nitrosamines
AOAC Method LOQs target acceptance criteria of NMT 10ppb (ng/g) for an individual N-nitrosamine.
Based on the LOQs for rubber, MDI methods should target LOQs around 1 ng/canister.
September 2006
PQRI Training Course

31. N-nitrosamines in OINDP – Points to Consider
N-nitrosamines are usually associated with sulfur-cured black rubber.
Even with the sensitivity and selectivity of the GC/TEA, other peaks are often noted in OINDP leachables profiles.
N-nitrosamines are very light sensitive, which suggests a possible procedure for identifying “non-nitrosamine” GC/TEA peaks.
September 2006
PQRI Training Course

32. Analysis of Mercaptobenzothiazole (MBT) Compounds from Sulfur Cured Rubber by a Liquid Chromatography – Tandem Mass Spectrometry (LC-MS-MS) Method
Tianjing Deng*, Shuang Li, Xiaoya Ding and Song Klapoetke
PPD
8551 Research Way
Middleton, WI 53562   
* Corresponding author
September 2006
PQRI Training Course

33.   
Mercaptobenzothiazole (MBT) and other benzothiazoles are common vulcanization accelerators for rubber materials that are used in pharmaceutical container/systems, such as the gaskets in the pressurized Metered-Dose Inhaler (pMDI). MBT is of particular concern since it is considered a potential carcinogen and has been shown to migrate into drug formulations.
Due to the toxicological concern and leachability of MBT and other benzothiazoles, analytical methods have been developed to study these types of compounds in the fields of food additives and contaminants (1), contact dermatitis caused by the rubbers (2), as well as pharmaceutical packaging systems (3). MBT can be analyzed by gas chromatography (4) but many other benzothiazoles are thermally-labile and readily decomposed in the GC inlet. HPLC methods are commonly used to study
September 2006
PQRI Training Course

34. MBT
MBTS
  In this study, a method using liquid chromatography with tandem mass spectrometer (LC-MS-MS) was developed to analyze MBT in the sulfur cured rubber. The method is capable of detecting ng level of MBT in the rubber extracts. This study demonstrates the feasibility of using detector with high selectivity, such as LC-MS-MS method, for extractable/leachable with special toxicological concern that requires greater sensitivity and specificity.
September 2006
PQRI Training Course

35. RReferences:
1Barnes, K.A., Castle L., Damant, A. P., Read, W. A., and Speck, D. R., Food Additives and Contaminants, Vol. 20, No. 2, (2003).
2Hansson, C., Bergendorff, O., Ezzelarab, M., and Sterner, O., Contact Dermatitis, 36, , (1997).
3Gaind, V. S., and Jedrzejczak, K., Journal of Analytical Toxicology, Vol. 17, 34-37, (1993).
4Niessen, W. M. A., McCarney, C. C., Moult, P. E.G., Tjaden, U. R., and Van der Greef, J., Journal of Chromatography, 647, , (1993).
5Mathieu, C., Herbreteau, B., Lafosse, M., Morin, Ph., Renaud, M., Cardinet, C., and Dreux, M., J. High Resol. Chromatogr., 23, (9), , (2000). 6
September 2006
PQRI Training Course

36. Mobile Phase: Water:Methanol:Formic acid 20:80:0.05 (v/v/v)
Method Conditions:
HPLC Parameters
Mobile Phase: Water:Methanol:Formic acid 20:80:0.05 (v/v/v)
Flow Rate: 0.2 mL/min
Column: Waters Symmetry C18, 3.5 m, 2 x 100 mm
Column Temperature: 40°C
Autosampler Temperature: Ambient
Injector Volume: 5 L
September 2006
PQRI Training Course

37. “Bench-top” LC/MS Systems
Time-of-flight
Linear ion trap
Triple Quadrupole
September 2006
PQRI Training Course

38. Triple Quadrupole Mass Spectrometer
September 2006
PQRI Training Course

39. MS-MS Spectrum of MBT Mercaptobenzothiazole (MBT) September 2006
PQRI Training Course

40. MS-MS Spectrum of MBTS Dibenzothiazyl Disulfide (MBTS) September 2006
PQRI Training Course

41. PE Sciex API 2000/API365 Triple Quadruple Mass Spectrometer
PE Sciex API 2000/API365 Triple Quadruple Mass Spectrometer
September 2006
PQRI Training Course

42. Selectivity/Specificity
MRM Chromatogram of Extraction Blank:
MBT (blue trace) and MBTS (red trace)
September 2006
PQRI Training Course

43. MRM Chromatograms of MBT (blue) and MBTS (red) in the 500 ng/mL standard solution.
September 2006
PQRI Training Course

44. MRM Chromatograms of MBT (blue) and MBTS (red) in the 30 min TBME Extract.
September 2006
PQRI Training Course

45. Extraction Method
Hansson et al. studied the extraction of MBT/MBTS using different solvents. They found out that Methyl tert-Butyl Ether (MTBE) is a good solvent for MBT/MBTS due to its:
Powerful extraction medium
Low toxicity
Inertness to MBT/MBTS
High volatility
In this study, the rubber was cut into 3 x 3 mm squares. One gram of the rubber was extracted with 10 mL MTBE for 30 minutes by sonication. After extraction, the extract was diluted to different volume using Methanol: Water 50: 50 diluent to give varying MBT concentrations and filtered using glass fiber syringe filters for LC-MS study.
September 2006
PQRI Training Course

46. The Extraction Study of MBT by MTBE from Sulfur-Cured Rubber
September 2006
PQRI Training Course

47. Linearity Plot of MBT (50 – 1000 ng/mL)
September 2006
PQRI Training Course

48. Calculated Concentration (PPM)
Repeatability
Calculated MBT Concentration (PPM) in Three Replicates of Extract.
Calculated Concentration (PPM)
Mean
%RSD (n=3)
Replicate 1
Replicate 2
Replicate 3
153.4
157.1
149.6
2.4
September 2006
PQRI Training Course

49. MRM Chromatogram of MBT Standard (50 ng/mL)
LOQ/LOD
The DL of MBT was calculated using S/N ratio = 3. DL = 6 ng/mL in the solution or 12 pg on column.
MRM Chromatogram of MBT Standard (50 ng/mL)
September 2006
PQRI Training Course

50. Accuracy - Filter Study
A Filter study was conducted to verify that the syringe filter used in the sample preparation did not reduce the recovery of MBT and MBTS. Three 500 ng/mL standards were analyzed before and after the filtration and the area responses of MBT and MBTS were compared. The percent differences between the filter and non-filtered samples are less than 2.5% indicating that filtration does not affect the method accuracy.
September 2006
PQRI Training Course

51. Accuracy- MBT Recovery
Approximately 360 ng/mL of MBT was spiked into the extract. The sample was prepared using the sample preparation procedure and analyzed. Three replicates of spiking samples were prepared and analyzed. The mean recovery of MBT was 87.3% 
Recovery Results of MBT
Extract
(ng/mL)
Replicate 1
Replicate 2
Replicate 3
Calculated
219.5
567.8
552.6
553.2
%RSD
3.5 (n=7)
2.3
2.4
3.0
%Recovery NA
89.9
86.0
86.1
September 2006
PQRI Training Course

52. Summary Points
The developed LC/MS/MS method looks good for MBT, and potentially MBTS.
MBTS was demonstrated to hydrolyze under the extraction conditions selected, forming more MBT.
This method requires full optimization and validation.
September 2006
PQRI Training Course

53. Analytical Method Validation
System suitability
Chromatographic parameters (e.g. resolution, tailing)
Injection precision
Precision
Repeatability
Intermediate Precision
Selectivity
Accuracy (three spiking levels)
Linearity/Range
LOD/LOQ
Robustness (e.g. column, mobile phase, temperatures, MS parameters)
September 2006
PQRI Training Course

54. Concluding Points
“Special Case” compounds require dedicated and highly specific analytical methods.
Dedicated and highly specific analytical methods have been developed for all “special case” compounds and compound classes.
September 2006
PQRI Training Course