Health Canada Health Products and Food Branch Biologics and Genetic Therapies Directorate Update on Heparin Contamination: Assessment of Optimized Capillary Electrophoresis, Proton NMR and Ion-Exchange HPLC Methodologies Michel Girard*, Marc-André Joly, Barry Lorbetskie, Terry Cyr and Yves Aubin Centre for Biologics Research, Health Canada, Tunney s Pasture, Ottawa (ON), Canada K1A 0L2
CE Pharm 2008
Outline Heparin - structure Chronology of events Early events June-September 2008 Capillary Electrophoresis HPLC NMR Conclusion
What is Heparin? Highly sulfated polysaccharides ranging between 3 kda and 50 kda (Avg MW: 12-15 kda) Extracted from animals, mostly porcine origin Highest charge density of known biological molecules Widely used as anticoagulant in medical procedures, injectables, medical devices
Structures of major disaccharide repeat units of sulfated glycosaminoglycans (sgag) Chondroitin sulfate Dermatan sulfate (Chondroitin sulfate B) Heparin R = H, SO 3 H For heparin 1 in 5 disaccharide repeat unit bears an acetyl group: absorbance at λ 200 nm approx. 5 times less than DS or CS
Early Events (Nov 2007 Apr 2008) Nov 2007 Feb. 2008: dramatic increase of severe adverse drug reactions (ADR) due to heparin in the USA. Feb 2008: Baxter USA recalls its heparin products Mar 2008: - US FDA requires CE and NMR tests for detection of unknown impurity in unfractionated heparin; - Health Canada requests CE and NMR tests and provides assistance for analysis due to lack of necessary instrumentation at most manufacturers - CE and NMR analyses at HC confirm 1 contaminated sample in B. Braun products; product is recalled within 24 hours Apr 2008: - unknown identified as oversulfated chondroitin sulfate (OSCS)
FDA recommended CE method FAIL PASS Instrument: Agilent HP 3D-CE equipped with diode array detector or equivalent Capillary: Bare fused silica, 50μm i.d., 64.5cm-total length, 56cm-effective length Temp.: 25ºC Detection wavelength: 200nm (bandwidth 10nm) Polarity: Negative Voltage: 30 kv Injection: 50 mbar pressure for 10 seconds Filter: Cellulose acetate membrane filters (0.22μm) Separation Time: 15 minutes Electrolyte: 36mM Phosphate buffer (ph 3.5) Specification: The electropherogram of test solution does not exhibit a sharp distinguishable peak in front of the main heparin peak. The migration time of heparin in the test solution is about 5.7 min. See attached for examples. Reference: 1. Private communication, Baxter study number 41010 2. R.P. Patel, C. Narkowica, J.P. Hutchinson, E.F. Hilder, G.A. Jacobson, A simple CE method for the rapid separation and determination of intact low molecular weight and unfractionated heparins, J. Pharm. Biomed. Anal. 46 (2008) 30-35
OSCS-contaminated Heparin 0.0050 0.0050 0.0045 0.0045 Heparin 0.0040 OSCS Heparin 0.0040 DS 0.0035 0.0035 0.0030 0.0030 AU 0.0025 0.0025 AU 0.0020 0.0020 OSCS 0.0015 0.0015 0.0010 0.0010 0.0005 0.0005 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 Minutes
Summary of heparin analysis at Health Canada, Mar-Jun 2008 170 heparin samples received from 8 manufacturers between March and June 2008 Unfractionated heparin (120 samples) analyzed by CE and 1 H-NMR Low molecular weight heparin (50 samples) analyzed by 1 H-NMR Several samples showed variability in content of dermatan sulfate (DS)/chondroitin sulfate (CS) 1 sample with OSCS was detected and confirmed (CE + NMR) 1 sample contained another unidentified impurity
Events Jun-Sept 2008 Workshops between authorities and manufacturers are held EP and USP revise heparin monographs to include CE and NMR tests (Jun-Jul) with new revisions soon to follow On-going discussions in regulatory agencies on limits for DS and other sgag Methods to improve separation / quantification of OSCS are reported and discussed with regulatory agencies (Jun-Sept) Capillary electrophoresis Baxter, Utrecht Univ. SAX-HPLC NMR
Improved CE methods 2 new CE methods developed: Utrecht Univ. (UU) Baxter (BX) Both based on high molarity background electrolytes Lower limit of detection for OSCS Quantification of DS/CS
CE Conditions UU Method Bare-fused silica capillary 850 mm TRIS phosphate, ph 3.0, 35ºC -30 kv (reverse polarity) 60 cm total length, 25 μm i.d. Developed on PACE MDQ CE with regular capillaries Somsen G.et al., J Chromatogr A. 2009 May 1;1216(18):4107-4112. BX Method Bare-fused silica capillary 600 mm lithium phosphate, ph 2.5, 20ºC -15 kv (reverse polarity) 30 cm total length, 25 μm i.d. Developed on Agilent CE with bubble cell capillaries Wielgos, T., et al.,. J. Pharm. Biomed. Anal., 49 (2009) 319-326.
UU CE method 0.005 UV- 200nm Braun 60910839 heparin braun 6091083950mg-ml 850mmtrisphosphate ph3 30kv 35c 50cm-rep1 UU method (60 cm cap), -30 kv UV- 200nm Heparin Organon L00027442 heparin organon l00027442 50mg-ml 850mmtrisphosphate ph3 30kv 35c 50cm-rep2 0.0175 UU method (30 cm cap), -15 kv 0.0150 0.004 0.0125 0.0100 0.003 0.0075 AU AU 0.002 0.0050 0.0025 0.001 0.0000-0.0025 0.000-0.0050-0.0075 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Minutes 0 1 2 3 4 5 6 7 8 9 10 Minutes
Comparison of UU and BX methods on MDQ 0.020 0.0175 UU method 0.018 0.016 0.0150 UU method 0.0175 0.0150 BX method 0.0125 0.014 0.0125 AU 0.0100 0.0075 AU 0.012 0.010 BX method AU 0.0100 0.0075 0.0050 0.0050 0.008 0.0025 0.0025 0.0000 0.006 0.0000-0.0025 0.004-0.0025-0.0050-0.0050 0.002-0.0075-0.0075 0 1 2 3 4 5 6 7 8 9 10 0.000 Minutes 0 1 2 3 4 5 6 7 8 9 10 Minutes 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 Minutes
Selectivity of BX and UU methods for reference samples BX method UU method 0.05 CS shark 1 0.0 6 CS shark CS bovine 0.04 CS bovine 0.0 5 0.0 4 DS 0.03 AU DS AU 0.0 3 Heparin 0.02 Heparin 0.0 2 0.01 0.0 1 0.00 0.0 0 0 2 4 6 8 10 Minutes 0 2 4 6 8 1 0 Minut
Separation of DS - UU method UU CE method: Dermatan Sulfate Linearity AU 0.004 0.003 Dermatan Reference solution 1.25mg/ml : Heparin Sodium BRP solution 8.75mg/ml Heparin DS Corrected Area Units 70000 60000 50000 40000 30000 20000 10000 0 y = 12770x - 267.04 R 2 = 0.9855 0 1 2 3 4 5 6 Concentration (mg/ml) 0.002 N=3 0.00250 0.00225 LOQ = 0.3 mg/ml 0.001 AU 1 0.00200 0.00175 0.00150 0.00125 0 2 4 6 8 10 Minutes 0.00100 0.00075 5.0 5.5 6.0 6.5 7.0 7.5 Minutes
Comparative Evaluation of UU and BX Methods Both methods perform similarly Linear, precise and sensitive Provide better separation and quantification of OSCS (LOD < 0.1%) than original method Provide quantification of DS/CS (LOQ < 0.5%)
SAX-HPLC (Initial EP method) Strong-anion exchange HPLC (SAX) Developed by heparin manufacturer Quantification of OSCS and DS/CS DS and CS co-elute
SAX-HPLC / Initial EP method Column: Dionex IonPac AS11 2 x 250 mm with pre-column Dionex IonPac AG11 2 x 50 mm or equivalent Mobile phases: (A) 2.5mM NaH2PO4, ph 3.0; (B) 2.5mM NaH2PO4 + 1M NaClO4 ph 3.0 Detection at 202 nm Column temperature 40 C Flow rate 0.22ml/min Equilibration + Run time 75 min Gradient composition Time (min) 0 20 60 90 61 20 75 20 % mobile phase B DS-CS Heparin CS marine OSCS
Comparison of SAX and UU CE for Sample J 0.020 OSCS 1 0.015 Hep 2 AU 0.010 0.005 DS 3 0.000 0 2 4 6 8 10 Minutes
Comparison of SAX and CE Sample SAX EP area ratio OSCS/hep CE UU area ratio CE Baxter area ratio B 1.37 0.93 0.91 E 0.58 0.55 0.55 F G 1.37 1.05 0.94 H 1.21 1.05 1.15 J 0.46 0.44 0.45 Sample SAX EP area ratio DS CS/hep CE UU area ratio CE Baxter area ratio B 0.02 0.07 0.07 D 0.37 0.29 0.29 F 0.45 0.41 0.41 G 0.02 0.09 0.09 H 0.03 0.08 0.08 J 0.02 0.05 0.05
Comparative Evaluation of SAX and CE Methods SAX and CE provided results that were consistent for a series of samples containing various levels of OSCS and DS/CS CE provides better peak shapes for OSCS and heparin but not for DS/CS SAX method was not used with depolymerization as proposed
Outcome of 1 st EP coll. Study SAX-HPLC difficult to adequately quantify DS Concentration sensitivity of CE not as good as HPLC Next collaborative study to require 2-steps procedure for optimized SAX-HPLC: identification quantification after depolymerization
SAX-HPLC / Optimized EP method Heparin Heparin DS-CS OSCS CS marine DS/CS OSCS CS marine
SAX-HPLC / Optimized EP method: reference sample
SAX-HPLC / Optimized EP method: sample 2 After depolymerization
SAX-HPLC / Optimized EP method: sample 5 After depolymerization
Optimized SAX-HPLC Shorter analysis time for identification step Better peak shape for DS Resolution between DS and heparin not as good Depolymerization worked well but increases overall analysis time (2 steps) potential for artefacts
CE analysis of Test samples 1-6 from 2 nd EP collaborative study 0.016 OSCS Heparin 0.016 0.014 0.014 0.012 0.012 0.010 0.010 AU 0.008 DS 0.008 AU 0.006 0.006 0.004 DS 0.004 0.002 0.002 0.000 0.000 0 2 4 6 8 10 Minutes
Optimized 1 H-NMR method At 600 MHz, the low field signal from the carbon-13 satellites of the methyl from the acetyl group overlaps with the signal of OSCS. To ensure the absence of OSCS, (<0.2 %w/v), a weak carbon decoupling field was applied at the methyl carbon chemical shift (17 ppm) to collapse the proton signals arising from the carbon satellites Heparin + 1% OSCS without carbon decoupling Heparin with carbon decoupling Heparin without carbon decoupling
Conclusion Rapid setup of CE and NMR methods provided results in timely fashion and prevented potential problems from the presence of contaminated product. Improved CE methods have been developed and performed well CE and SAX-HPLC results were compatible Optimized SAX-HPLC has been retained by EP for inclusion in revised heparin monograph
Team and Collaborators CE - Marc-André Joly HPLC - Barry Lorbetskie 1 H-NMR Dr. Yves Aubin Coordination Dr. Terry Cyr Collaborators Dr. G.Rautmann (EP) Dr. G. Somsen (Utrecht Univ.) Dr. R. Weinberger (CE Technol.) Dr. T.Wielgos (Baxter) Collaborative study team from CE Pharm