Journal of Pharmaceutical and Biomedical Analysis 38 (2005) 781–784
Stable isotopic composition of the active pharmaceutical
A.M. Wokovich , J.A. Spencer , B.J. Westenberger , L.F. Buhse , J.P. Jasper
a Food and Drug Administration, Center for Drug Evaluation and Research, Division of Pharmaceutical Analysis, St. Louis, MO 63101, USA
b Molecular Isotope Technologies, LLC, 8 Old Oak Lane, Niantic, CT 06357-1815, USA
Received 12 November 2004; received in revised form 3 February 2005; accepted 11 February 2005
A survey of the multi-stable isotopic composition of an active pharmaceutical ingredient (API), naproxen, was performed to assess the
potential of Isotope Ratio Mass Spectrometry (IRMS) to distinguish the provenance of APIs. Twenty-six lots of naproxen from six man-ufacturers representing four countries (Italy, India, Ireland, and the USA) were analyzed for three isotope ratios (13C/12C, 18O/16O, andD/H). The samples were analyzed by either Elemental Analyzer/Isotope Ratio Mass Spectrometry (EA/IRMS: carbon (δ13C)) or by ThermalConversion-EA/IRMS (TCEA/IRMS: hydrogen (δD) and oxygen (δ18O)). Bivariate and trivariate isotope ratio graphs for naproxen showmarked clustering of the data for five out of the six naproxen manufacturers, suggesting that IRMS may be a plausible means to screen formanufacturer of given APIs. 2005 Elsevier B.V. All rights reserved. Keywords: IRMS; Isotope Ratio Mass Spectrometry; Naproxen; API; Active pharmaceutical ingredient; Isotope ratio; δ13C; δ18O; ␦D
ticity in food products Ratio Mass Spectrom-etry (IRMS) has been used to characterize different photosyn-
The geographic or lot-specific identity of botanicals, ac-
thetic pathways that impart distinctive isotopic compositions
tive pharmaceutical ingredients (APIs), and finished dosages
to various plant organic materials Many excipients
is important for the mitigation of counterfeiting, diversion,
used in drug manufacturing are derived from plant sources
reimportation, theft, vicarious liability, and potentially for
and so are also expected to show stable isotope ratios influ-
patent protection misidentification of drug
enced by their source and means of production.
products or pharmaceutical components (e.g. APIs, excipi-
The stable isotopic composition of a suite of naproxen API
ents) threatens the efficacy of and consumer confidence in
lots was examined via Isotope Ratio Mass Spectrometry to
these commodities, as well as the economic well-being of
determine whether this method can be used to distinguish the
pharmaceutical companies. Characterization of the natural
isotopic provenance. Naproxen, an anti-inflammatory drug,
lot-to-lot stable isotopic variation of such materials provides
was chosen because it is a widely used active pharmaceu-
a means to isotopically fingerprint individual lots
tical ingredient and manufactured worldwide. In this study,
Stable isotope ratios have been used as tracers of the source
three isotopic ratios (δ13C, δD, and δ18O) of naproxen were
or to provide the “isotopic provenance” of natural materials
examined and their results presented graphically.
ratio spectrometry is an accepted techniquefor the detection of adulteration or establishment of authen-
∗ Corresponding author. Tel.: +1 314 539 3874; fax: +1 314 539 2113.
The United States Food and Drug Administration, Center
E-mail address: [email protected] (A.M. Wokovich).
for Drug Evaluation and Research received 26 different lots
0731-7085/$ – see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jpba.2005.02.033
A.M. Wokovich et al. / Journal of Pharmaceutical and Biomedical Analysis 38 (2005) 781–784
around the analyte. Single-to-duplicate measurements of car-bon (δ13C) were performed on each API sample.
Individual samples of ∼0.2 mg were weighed and placed
into silver boats that had previously been dried in a vac-uum oven for 1.5 h. The boats were crimped tightly aroundthe analyte. Single-to-quadruplicate hydrogen (δD) and
Fig. 1. Structure of naproxen (C14H14O3).
duplicate-to-quadruplicate oxygen (δ18O) measurements ofeach naproxen sample were performed.
of naproxen API (C14H14O3; from 6 manufacturerslocated in 4 countries (India, Italy, Ireland, and the USA) that
2.4. Units of stable isotopic measurement
Carbon isotopic results are typically expressed in δ-values
India, Manufacturer A—five different lots.
(parts per thousand differences from international reference
India, Manufacturer B—five different lots.
Italy, Manufacturer C—three different lots.
• Italy, Manufacturer D—three different lots.
• Ireland, Manufacturer E—seven different lots.
USA, Manufacturer F—three different lots.
where Rsmpl is the 13C/12C ratio of the sample material and
These samples were sent to Isotech Laboratories, Inc. (Cham-
Rstd is the 13C/12C ratio of an International Atomic Energy
paign, IL, USA) as 26 unknown naproxen API samples.
Authority reference standard (known as “VPDB”, whose
Isotech Laboratories, Inc. analyzed the unknown samples to
13C/12C ratio has been defined as the official zero point of
determine their isotopic composition.
the carbon isotopic scale). Similarly, δD values and δ18O val-ues are reported relative to the international Vienna Standard
Mean Ocean Water (VSMOW) reference standard.
Carbon (δ13C) isotopic analyses were performed with a
Carlo Erba 1108 Elemental Analyzer (EA) interfaced via aConflo II interface to a Finnigan MAT Delta Plus XL iso-
The uncertainties (or precision) of the isotopic measure-
tope ratio mass spectrometer EA operated with an
ments in this study are presented in two ways. Pooled standard
oxidation furnace temperature of 1020 ◦C, reduction furnace
deviations (S.D.) of raw data were made to derive a represen-
temperature of 650 ◦C, and a packed-column temperature of
tative standard deviation from the entire raw data set for each
isotope ratio. In this calculation, a number of replicates arepooled to give a standard deviation that is representative of
2.2. Instrumentation: δD and δ18O
the entire sample suite. The standard errors (S.E.) expressedin the table represent the decreased uncertainty in the values
Hydrogen (δD) and oxygen (δ18O) stable isotopic analy-
due to the multiple determinations. These are calculated as
ses of each naproxen sample were performed on a Finnigan
S.E. = S.D./(square root of n), where n is the number of mea-
Thermal Conversion-Elemental Analyzer (TCEA) interfaced
surements performed on a given sample The standard
to Finnigan Delta Plus XL Isotope Ratio Mass Spectrometer.
errors are shown for the points plotted on the bivariate plots
Analogous to a standard Elemental Analyzer/Isotope Ratio
Mass Spectrometer (EA/IRMS) TCEA functions withsamples sequentially delivered into a furnace and the efflu-ent gases analyzed by an online IRMS but with pyrolysis
3. Results and discussion
(instead of oxidative combustion as in the EA/IRMS) per-formed at 1350 ◦C. The TCEA thermally converts analytes
the δD, δ13C, and δ18O values of the
to H2 and CO rather than combustion into H2O and CO2
26 lots of naproxen from 6 manufacturers. Three bivariate
as in the EAMS. The analyte gases, H2 and CO, are chro-
isotope plots encompass possible combinations of
matographically separated on a packed column at 85 ◦C. The
the three isotope ratios examined in this sample suite.
mass spectrometer measures H2 directly and 18O in the form
shows a plot of δD/δ13C, shows a plot of δ13C/δ18O,
and shows a plot of δ18O/δD. All three bivariate plotsand the trivariate plot (w general concentration (or
“clustering”) of sample points in relatively small spans of thetotal isotopic ranges ( δ), with one notable exception (one
Individual samples of ∼0.4 mg for δ13C analyses were
lot from Italian Manufacturer D). The clustering of the data
weighed and placed into tin boats that were crimped tightly
indicates manufacturer-based isotopic provenance. A.M. Wokovich et al. / Journal of Pharmaceutical and Biomedical Analysis 38 (2005) 781–784
Fig. 2. (a) Naproxen: δ13C vs. δD; (b) naproxen: δ13C vs. δ18O; (c) naproxen: δ18O vs. δD.
Isotopic ratio variation may result from thermodynamic
fractionation (e.g. differences in raw materials) and/orfrom kinetic fractionation (i.e. different synthetic (e.g. physical–chemical– or biochemical fractionation) pathways)is no apparent evidence for geographic identi-fication in these isotopic data; however, it is noted that phar-maceutical manufacturing is a complex process, which in-cludes raw materials and synthetic pathways plausibly de-rived from a variety of geographic regions, perhaps maskingisotopic-geographic origin in this study.
This relatively small study of the isotopic provenance of
naproxen API shows that the great majority of the ratios fromthe six manufacturers’ products fall within relatively well-defined areas, or “clusters”. IRMS in this instance does notprovide a definitive way to identify country of origin, butstable-isotopic analysis of pharmaceutical APIs appears to
Fig. 3. Naproxen isotope ratios. Plot made with raw isotope ration data
be a plausible means by which to screen for manufacturer-
normalize using Pt = (value − min)/range. A.M. Wokovich et al. / Journal of Pharmaceutical and Biomedical Analysis 38 (2005) 781–784
Table 1Stable isotopic data on 26 different lots of naproxen API from 6 manufacturers
 A. Longinelli, E. Selmo, J. Hydrol. 270 (2003) 75–88.  F. Palhol, C. Lamoureux, M. Chabrillat, N. Naulet, Anal. Chim. Acta
The authors thank the FDA—Forensic Chemistry Center
 I.T. Platzner, in: J.D. Winefordner (Ed.), Modern Isotope Ratio
for contributing some of the naproxen samples.
Mass Spectrometry, John Wiley & Sons, Chichester, 1997, pp. 403–
The views presented in this article do not necessarily re-
flect those of the Food and Drug Administration.
 E. Piasentier, R. Valusso, F. Camin, G. Versini, Meat Sci. 64 (2003)
 J.P. Renou, C. Deponge, P. Gachon, J.C. Bonnefoy, J.B. Coulon, J.P.
Garel, R. V´erit´e, P. Ritz, Food Chem. 85 (2004) 63–66. References
 J.P. Renou, G. Bielicki, C. Deponge, P. Gachon, D. Micol, P. Ritz,
 J.P. Jasper, B.J. Westenberger, J.A. Spencer, L.F. Buhse, M. Nasr, J.
 L. Pissinatto, L.A. Martinelli, R.L. Victoria, P.B. de Camargo, Food
Pharm. Biomed. Anal. 35 (2004) 21–30.
 X. del la Torre, J.C. Gonz´alez, S. Pichini, J.A. Pascual, J. Segura,
 J.P. Jasper, Pharma. Technol. 23 (August 1999) 106–114.
J. Pharm. Biomed. Anal. 24 (2001) 645–650.
 J.R. Ehleringer, J.F. Casale, M.J. Lott, V.L. Ford, Nature 408 (2000)
 J.F. Carter, E.L. Titterton, M. Murray, R. Sleeman, Analyst 127
 J.P. Jasper, Rapid Commun. Mass Spectrom. 15 (2001) 1554–1557.
 J.F. Carter, E.L. Titterton, H. Grant, R. Sleeman, Chem. Commun.
 J.P. Jasper, L.E. Weaner, B.J. Duffy, Forensic Isotope Ratio Mass
Chinese Remedies & Clinics ,August 2002 ,Vol 2 ,No. 4 A randomised clinical trial on nimesulide 2 methotrexate combined therapy of adult 2 onset Still s disease LIANG Liuqin , Xu Hanshi , ZHAN Zhongping , YE Yujing. Department of Rheumatology and clinical Immunlolgy , The First Affiliat 2 ed Hospital , Zhongshan University , Guangzhou 510080 ,China Abstract Objective To explore t
March 2005 Vol. 11 No. 5 The Development of Industrial Clusters Towards aThe aims of this study are to: (i) explore the factors contributing to thesuccessful formation of industrial clusters and the overall effects of industrialclustering on productivity; (ii) gain an understanding of the organization andnetworking of industrial clusters; (iii) examine the flow of human resources betweenclu