Accepted Manuscript Title: analysis of Iobitridol in an injectable preparation by 1 H NMR spectroscopy Author: Anna Borioni Gianluca Gostoli Elena Boss`u Isabella Sestili PII: DOI: Reference:
S0731-7085(14)00047-8 http://dx.doi.org/doi:10.1016/j.jpba.2014.01.030 PBA 9435
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
14-11-2013 14-1-2014 21-1-2014
Please cite this article as: A. Borioni, G. Gostoli, E. Boss`u, I. Sestili, analysis of Iobitridol in an injectable preparation by 1 H NMR spectroscopy., Journal of Pharmaceutical and Biomedical Analysis (2014), http://dx.doi.org/10.1016/j.jpba.2014.01.030 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
[rel]
*Graphical Abstract
General ------DATE = 2013/04/23 TIME = 13:25 INSTRUM = spect PULPROG = zgpr
HO O O
I HN
OH
N
F1 (1H) ------------SI = 65536 SF = 400.06 SW_p = 8012.821
OH I
I N
0.2
O
ip t
OH
0.0
0.1
OH
3.5
3.0
[ppm ]
Ac
ce pt
ed
M
an
us
4.0
cr
0.3
HO
Page 1 of 15
*Highlights (for review)
A qNMR method was used for direct assay determination of Iobitridol in an injectable formulation.
Experiments were carried out in deuterated alkaline water with 3,5 dinitrobenzoic acid as internal reference standard.
Method was validated assessing specificity, accuracy, precision, linearity, stability of samples and robustness.
Method proved suitable for routine controls especially when high sample throughput is required.
Ac
ce pt
ed
M
an
us
cr
ip t
Page 2 of 15
*Revised Manuscript
1
1
Quantitative analysis of Iobitridol in an injectable preparation by H NMR spectroscopy.
2
Anna Borioni*, Gianluca Gostoli, Elena Bossù, Isabella Sestili
3
Istituto Superiore di Sanità, Dipartimento del Farmaco, Rome Italy
4
*Corresponding author. Tel.: +39 06 49902516; fax: +39 06 49902625. E-mail address: [email protected].
5 1.
INTRODUCTION
ip t
6 7
Nuclear Magnetic Resonance spectroscopy in quantitative analysis (qNMR) experienced a remarkable
9
increase since about 2005-2007. In the field of pharmaceutics, qNMR raised the interest of the Official
10
Medicine Control Laboratories on the occasion of a criminal adulteration of heparine with a component
11
detectable by NMR spectroscopy and capillary electrophoresis [1,2].
12
As it is nowadays well documented, quantitative analytical methods based on NMR spectroscopy proved to be
13
highly robust, accurate and precise. In general, qNMR allows quantifying compounds, even in mixtures, at
14
sub-micromolar range and at the same time provides information on the chemical structure of the molecules
15
involved. [3,4,5,6,7]. Moreover, quantification by NMR spectroscopy is truly straightforward in comparison to
16
the chromatographic and spectroscopic techniques which require specific standards or correction factors. In
17
fact the area of the NMR signals is directly proportional to the quantity of the nuclei evoking the signal with no
18
dependence on how the nuclei are combined in chemical structures. This implies that molecules with no
19
chemical similarity to the analyte are used as internal reference standard [8].
20
In the present case we report the determination of the content of iobitridol in an injectable aqueous solution
21
(Fig. 1). Iobitridol is an iodine based contrast medium approved in Europe and in the USA for radiographic
22
diagnostic imaging in adult and children [9]. No EP or USP monograph for iobitridol is available though
23
similar substances such as iodixanol, iohexol, iopamidol, iopromide, iotrolan, iopanoic acid, iotalamic acid
24
and ioxaglic acid are reported in EP. For the large majority of them the assay is estimated by potentiometric
25
titration of iodine after a work up consisting of three operational steps (reduction by Zn, boiling, filtration
26
[10,11,12,13,14]
27
To the best of our knowledge only one determination of iobitridol in biological fluids by HPLC is reported [15].
28
We approached the NMR spectroscopy with the aim of accomplishing the quantification of the content of
29
iobitridol directly in the final formulation with minimal work up, in short time and by a medium field instrument
30
which is of prevalent and established use in qNMR. The NMR experiments were carried out according to the
Ac
ce pt
ed
M
an
us
cr
8
1 Page 3 of 15
31
general principles laid down in the general EP monograph 01/2009:20233 Nuclear Magnetic Resonance
32
Spectrometry.
33 34
2.
Materials and methods
36
2.1.
Chemicals and NMR Instrumentation
37
NMR experiments were recorded on a Bruker Avance instrument (Bruker Italia, Milano, Italy) operating at 400
38
MHz equipped with a 5 mm H/ C BBI probe and some preliminary spectra were acquired on a Bruker AM
39
operating at 700 MHz. The software Bruker Xwinnmr 3.1 was used for acquisition, the software Bruker
40
Topspin 1.2 was used for processing.
41
D2O (99.97%) and DMSO-d6 (99.80%) from Euriso Top (Saint-Aubin, France) were used as solvents. NMR 5
42
mm diameter tubes were purchased from CortecNet (Voisins-Le-Bretonneux, France).
43
Sodium hydroxide, 3-(Trimethylsilyl)-1-propanesulfonic acid, Maleic acid TraceCERT
44
Dinitrobenzoic acid TraceCERT CRM were purchased from Sigma- Aldrich (Sigma- Aldrich Italia, Milano,
45
Italy).
46
2.2.
47
Explorative spectra were performed on a solution of iobitridol injectable solution (260 mg) and maleic acid
48
(10 mg) in D2O (1 ml) and on a solution of iobitridol injectable solution (260 mg) and 3,5-dinitrobenzoic acid
49
(10 mg) in DMSO-d6 (1 ml).
50
For quantitative analysis an accurate weighted quantity of 3,5-dinitrobenzoic acid TraceCERT CRM (9.5
51
mg) and an accurate weighted quantity of the injectable solution here named iobitridol FP (260 mg), were
52
added to 1 ml of D2O previously made alkaline by addition of 1.4% w/w of solid sodium hydroxide. After five
53
minutes sonication, 0.7 ml of the solution was transferred into an NMR tube. The longest longitudinal
54
relaxation time (T1max) was experimentally determined with the Bruker inversion recovery pulse program
55
and resulted about 6 s owing to the 3,5-dinitrobenzoic acid protons. Spectra were acquired in triplicate for
56
each independent weighing unless otherwise specified. Acquisition parameters were: pulse length 45°,
57
delay 30 s, temperature 298 K, 16 scans and TD = 64K. The signal at 2.63 ppm (1H, multiplet) was
58
integrated for iobitridol and the singlet signal at 8.88 ppm (2H, doublet) of the 2,6 protons of 3,5-
59
dinitrobenzoic acid (DNB) was integrated as quantitative reference. In the processing an exponential function
us
cr
13
an
1
ip t
35
ce pt
ed
Sample preparation and NMR method
CRM, 3,5-
M
®
®
Ac
®
2 Page 4 of 15
60
of 0.3 Hz was applied, baseline was corrected by the software fifth order polynomial algorithm and phase
61
was adjusted manually. S/N always resulted above 350.
62
Assignation of the proton spectra was based on 2D H- H COSY [16], 2D H- H TOCSY [16], 2D H- H
63
ROESY [16] and 2D H- C HSCQ [17] diagrams. Standard Bruker pulse sequences were employed for 2D
64
experiments. Typically, data were collected for 2K TD and 512 increments (omonuclear 2D) and 2K TD and
65
256 increments for HSQC. Number of scans varied from 48 to 112, TOCSY mixing time was 100 ms, a delay
66
for a one bond H- C coupling constant of 140 Hz was set for HSQC and full proton decoupling of
67
achieved with the waltz16 sequence.
1
1
1
13
13
68 69
1
13
C was
cr
1
1
ip t
1
1
Results and discussion
71
3.1.
Choice of internal standard and dilution solvent
72
Explorative proton spectra were required to select the suitable solvent/reference standard system as the
73
proton spectrum of iobitridol is not easily predictable. Indeed, iobitridol could present partially restricted
74
rotation of the amide bonds around the aromatic ring giving rise to multiple rotational isomers possibly
75
displaying different NMR signals [18].
76
Solutions of iobitridol injectable solution together with maleic acid in D 2O or together with 3,5-dinitrobenzoic
77
acid (DNB) in DMSO-d6 were first used. Results were not conclusive because in both systems no
78
satisfactorily integrable protons of iobitridol could be identified. In D2O, the multiplet at 4.10 ppm attributable
79
to the single proton of the 2,3-dihydroxypropyl substituent (proton 2 Fig. 1) was the most promising candidate
80
for integration. However, it was not finally selected because of its proximity to the large water signal and
81
above all because its relative integral value did not account for a neat number of protons. This fact suggested
82
that the multiplet could represent a mixture of the signals of different conformational isomers.
83
As iobitridol showed a more resolved spectrum in D2O, the aqueous system was further exploited with the
84
purpose of improving signals resolution. In alkaline solution, the proton spectrum of iobitridol FP displayed a
85
distinct signal at about 2.6 ppm accounting for one proton integral which was assigned to the single proton of
86
the propylcarboxamino substituent of the benzene ring (proton 1 Fig.1). The chemical shift of proton 1
87
showed dependence on the sodium hydroxide concentration: the more alkaline the solution the more
88
shielded and resolved the proton was (Fig. 2). This trend may be attributable to the partial deprotonation of
89
the adjacent amide group in alkali: the consequent local increase of electron density led to an upfield shift of
us
3.
Ac
ce pt
ed
M
an
70
3 Page 5 of 15
the proton 1 signal [19]. A 1.4% w/w NaOH concentration in D2O represented a good compromise between a
91
satisfactory separation of the selected proton from the neighboring signals of the N-methyl groups and the
92
drawbacks of high alkalinity (potential active standard degradation and lengthening of the NMR pulse). 3,5-
93
Dinitrobenzoic acid was selected as quantitative reference standard because of its deshielded NMR signals
94
which assured no interferences with the iobitridol signals and with the strong water signal.
95
With the aim of reducing the set up of the acquisition parameters and of standardizing the experiments, the
96
suppression of the water signal was skipped at first. Validation results proved that this approach was correct
97
since the high amount of water in the formulation (about 40:1 molar ratio in respect to active substance) did
98
not actually impair the correct evaluation of the content of iobitridol.
99
3.2.
us
Iobitridol FP assay
cr
ip t
90
The assay value of iobitridol FP was reported in the manufacturer’s certificate of analysis and it was
101
confirmed in our laboratory by repeating the analyses with the same manufacturer’s method. The method and
102
the relevant validation cannot be disclosed being confidential information reported in the Manufacturer’s
103
dossier for Marketing Authorization. Briefly, three different samples were assayed and found to have a
104
concentration of 34.3% (CV% = 0.31%; n=3). This value was in good agreement with the assay reported in
105
the manufacturer’s certificate of analysis (34.2%).
106
3.3.
107
The validation study was based on the ICH Q2 guidelines. The results obtained in the validation study were
108
considered satisfactory and in line with the expected performance already reported in literature. [20, 21].
109
3.3.1. Selectivity
110
Selectivity was demonstrated by showing that in the spectral zones where signals integrations were carried
111
out there was no interference of overlapping signals.
112
In general NMR 2D experiments can support selectivity by smoothly resolving in two dimensions signals
113
which are not separated in monodimensional spectra [20]. 2D H, H TOCSY and H, C HSQC were acquired
114
on iobitridol FP samples, the outcomes confirmed that no interfering signals affected quantification.
115
Furthermore, selectivity can be demonstrated by assessing the capability of the analytical method to provide
116
an exact result for the content of the analyte in a sample. This was confirmed by the results here obtained for
117
accuracy.
118
3.3.2 Accuracy
ed
M
an
100
Ac
ce pt
Validation
1
1
1
13
4 Page 6 of 15
Accuracy was carried out by analyzing three samples of iobitridol FP at about 80%, 100% and 120% by
120
weighing adequate amount of the drug product solution (between about 200 mg and 320 mg). Three
121
independent replicates for each level were prepared and each of these was analyzed in duplicate. The results
122
are reported in Table 1. Note that one of the samples (45%) referring to the 120% level is actually 131%. The
123
difference of +11% was not considered critical and it was included in the calculations.
124
The recoveries (ranging between 99.6% and 100.7%) were considered satisfactory also considering that the
125
analyzed sample is not a pure substance.
126
3.3.3 Precision
127
Repeatability was assessed by performing seven analyses of the same sample carried out by one analyst in
128
the same day. The CV% for the repeatability was 0.25%.
129
Intermediate precision was assessed by analyzing six independently prepared samples. The analyses and
130
sample preparations were carried out by two different analysts in two different days. The total CV% for the
131
intermediate precision was 0.38% (n=12).
132
Averages (and variances) obtained by two analysts on six independent samples were: 34.4% (0.0184) and
133
34.4% (0.0167). ANOVA analysis did not show any statistical difference: F=0.4308011 (Fcrit= 4,964603 -
134
p=005).
135
3.3.4 Linearity
136
Linearity was assessed by varying the amount of the tested samples in the range from 70% to 120% of the
137
target weight (260 mg) of the analytical procedure.
138
The relevant equation was y=0.1346x-0.763 (R =0.9985).
139
3.3.5 Robustness
140
The considered parameters and their variations were reported in Table 2.
141
The experiments were aimed to evaluate the recovery by changing one parameter in each experiment.
142
Robustness of the method was also evaluated on test solutions containing different sample/RS ratios (w/w).
143
In this case the robustness was checked by assessing the linearity of the integral of the iobitridol proton vs
144
the sample/RS ratios in the range +/-50% of the target ratio (Table 2).
145
The relevant equation was y=0.1391x-0.05321 (R =0.9994).
146
Data showed that the method is robust respect to sample/RS ratio variations.
147
3.3.6 Stability of the solution
ce pt
ed
M
an
us
cr
ip t
119
Ac
2
2
5 Page 7 of 15
148
Iobitridol is reported to be quite stable in alkaline solutions. In particular less than 2% degradation of iobitridol
149
in alkaline solution at pH 9 after three months at 50°C is reported [18].
150
Our experiments showed that the assay value of iobitridol is not affected by a 24 hours storage at 25°C when
151
dissolved in 1.4% w/w NaOH/D2O.
152 4.
Conclusion
ip t
153 154
1
In this report a quantitative H-NMR method for quantification of iobitridol in an injectable preparation was
156
developed and successfully validated with good results of accuracy and precision. It is expected that the
157
method is adequate also to quantify iobitridol as drug substance. The method does not require NMR
158
instrumentation equipped with high magnetic fields as validation proved successful on a magnet operating at
159
400 MHz which is nowadays quite common in pharmaceutical laboratories.
160
Furthermore the NMR method is more rapid and simple in comparison to assay determination based on
161
titrimetric and HPLC methodologies; it requires no specific reference standard, there is no need for sample
162
pretreatment and, because of short analysis times, it can be recommended for routine controls especially
163
when high sample throughput is required.
ed
M
an
us
cr
155
164 Acknowledgments
166
The authors acknowledge Mr. Tonino Puccio for technical assistance.
167
169
References: [1]
170 171
I. McEwen, A. Amini, I.M. Olofsson, Identification and purity test of heparin by NMR - a summary of
Ac
168
ce pt
165
two years' experience at the Medical Products Agency, Pharmeur. Bio. Sci. Notes 1 (2010) 65-72. [2]
,
T. Beyer, B. Diehl, G. Randel, E. Humpfer, H. Schäfer, M. Spraul, C. Schollmayer, U. Holzgrabe, 1
172
Quality assessment of unfractionated heparin using H nuclear magnetic resonance spectroscopy, J.
173
Pharm. Biomed. Anal. 48 (2008) 13–19.
174 175
[3]
U. Holzgrabe, Quantitative NMR spectroscopy in pharmaceutical applications, Prog. Nucl. Magn. Reson. Spectrosc. 57 (2010) 229-240.
6 Page 8 of 15
176
[4]
177 178
M. Malet-Martino, U. Holzgrabe, NMR techniques in biomedical and pharmaceutical analysis, 55 (2011) J. Pharm. Biomed. Anal. 1-15.
[5]
1
G F. Pauli, T. Gödecke,. B.U. Jaki, D.C. Lankin, Quantitative H NMR: Development and Potential of an Analytical Method – an Update, J. Nat. Prod. 75 (2012) 834–851.
179 [6]
S.K. Bharti, R. Roy, Quantitative 1H NMR spectroscopy, Trends Anal. Chem. 35 (2012) 5-26.
181
[7]
U. Holzgrabe, R. Deubner, C. Schollmayer, B. Waibel, Quantitative NMR spectroscopy—Applications
182
in drug analysis, J. Pharm. Biomed. Anal. 38 (2005) 806–812. [8]
cr
183
ip t
180
T. Rundlöf, I. McEwen, M. Johansson, T. Arvidsson, Use and qualification of primary and secondary standards employed in quantitative 1H NMR spectroscopy of pharmaceuticals, J. Pharm. Biomed.
185
Anal. (2013) available on line at
186
http://www.sciencedirect.com/science/article/pii/S0731708513004354.
188
an
[9]
P.L. McCormack, Iobitridol: a review of its use as a contrast medium in diagnostic imaging, Clin. Drug. Investig. 33 (2013) 155-166.
M
187
us
184
[10]
Iodixanol European Pharmacopoeia 7.0 07/2010:2215.
190
[11]
Iohexol European Pharmacopoeia 7.0 01/2008:1114.
191
[12]
Iopamidol European Pharmacopoeia 7.0 01/2008:1115.
192
[13]
Iopanoic acid European Pharmacopoeia 7.0 01/2008:0700.
193
[14]
Iotalamic acid European Pharmacopoeia 7.0 01/2008:0751.
194
[15]
P. Bourrinet, H. Feldman, A. Dencausse, C. Chambon, B. Bonnemain, High-performance liquid
ce pt
ed
189
chromatographic determination of iobitridol in plasma, urine and bile, J. Chromatogr. B 670 (1995)
196
369–372.
197
[16]
198 199
M. F. Summers, L. G. Marzilli, A. Bax, Complete 1H and I3C Assignments of Coenzyme B12 through the Use of New Two-Dimensional NMR Experiments, J. Am. Chem. Soc. 108 (1986) 4285-4294.
[17]
200 201
Ac
195
N. E. Jacobsen, NMR Spectroscopy Explained: Simplified Theory, Applications and Examples for Organic Chemistry and Structural Biology, John Wiley & Sons, West Sussex, 2007, pp. 489-550.
[18]
D. Meyer , M. Petta, M.H. Fouchet, M. Vadel, M. Schaefer, M. Dugast-Zrihen, M. Guillemot,
202
Stabilization of the hydrophilic sphere of iobitridol, a new nonionic iodinated contrast agent, Acta
203
Radiol. 37 (1996) 8-16.
7 Page 9 of 15
204
[19]
205
E. Gaidamauskas, E. Norkus, E. Butkus, D.C. Crans, G.˙Grinciené,˙ Deprotonation of β-cyclodextrin in alkaline solutions, Carbohydr. Res. 344 (2009) 250–254.
206
[20]
F. Malz, H. Jancke, Validation of quantitative NMR, J. Pharm. Biomed. Anal. 38 (2005) 813–823.
207
[21]
G. Maniara, K. Rajamoorthi, S. Rajan, G.W. Stockton
Method performance and validation for
quantitative analysis by (1)H and (31)P NMR spectroscopy. Applications to analytical standards and
209
agricultural chemicals, Anal. Chem. 70 (1998) 4921-4928.
Ac
ce pt
ed
M
an
us
cr
ip t
208
8 Page 10 of 15
Table 1
Table 1
26.3%
26.2%
99.6%
80%
28.2%
28.4%
100.7%
80%
26.7%
26.6%
99.8%
100%
35.3%
35.2%
99.7%
100%
34.3%
34.4%
100.5%
100%
33.4%
33.5%
100.6%
120%
39.8%
39.9%
100.3%
120%
39.7%
39.9%
100.3%
131% a
45.0%
44.9%
99.9%
a
The difference of +11% between 120% and 131% was not considered critical and it was included in the calculations.
Mean = 100.1%
0.41%
Ac
ce pt
ed
M
CV% =
ip t
80%
a
Recovery (%)
cr
Found (n=2)
us
Theoretical
an
Level
Page 11 of 15
Table 2
Table 2 Parameter target valuea
Modified parameters
Recovery
P1 =4 0°
100.3%
P1 = 50°
100.3%
Number of scans = 12
99.7%
Number of scans = 20
99.8%
D1 = 25 sec
100.0%
D1 = 35 sec
100.2%
O1p = 4.2
100.9%
O1p = 5.2
100.4%
Receiver gain = 3
99.5%
Receiver gain = 6
100.6%
Temperature = 296 K
99.8%
Temperature = 300 K
100.6%
P1= 45°
ip t
Number of scans = 16
cr
D1 = 30 s
an
Receiver gain = 5
M
Temperature = 298 K
NaOD = 1.2%
100.9%
NaOD 1.4% NaOD % = 1.6%
100.6 %
ce pt
ed
P1 = applied pulse; D1 = delay time, O1p = center of the spectral window.
Ac
a
us
O1p = 4.7 ppm
Page 12 of 15
h
us
g
cr
ip t
Figure 2
f
an
e d
M
c
b
Ac c
ep te
d
a
Page 13 of 15
Figure 1
HO
1
O O
I HN
OH
N OH I
I N
ip t
O
OH
2
ce pt
ed
M
an
us
cr
OH
Ac
HO
Page 14 of 15
Captions to figures
ip t
Figure 1 Chemical structure and IUPAC name of Iobitridol: 1-N,3-N-bis(2,3-dihydroxypropyl)-5-[3-hydroxy-2(hydroxymethyl)-1-oxopropylamino)]-2,4,6-triiodo-1-N,3-N-dimethylbenzene-1,3-dicarboxamide. Proton 1 was integrated for quantification.
Ac ce p
te
d
M
an
us
cr
Figure 2 Preliminary studies on the solvent/ reference standard system: Effect of NaOH concentration on the proton spectra Iobitridol and maleic acid in D2O. a: NaOH 0.1%; b: NaOH 0.3%; c: NaOH 0.7%; d: NaOH 1.5%; e: NaOH 2.0%; f: NaOH 2.5%; g: NaOH 5.0 %; h: From bottom to top NaOH 1.2%, 1.3% and 1.4%. NMR were acquired on a Bruker AV 700 MHz The integral of the multiplet under the arrow was selected for quantification.
Page 15 of 15
S0731-7085(14)00047-8 http://dx.doi.org/doi:10.1016/j.jpba.2014.01.030 PBA 9435
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
14-11-2013 14-1-2014 21-1-2014
Please cite this article as: A. Borioni, G. Gostoli, E. Boss`u, I. Sestili, analysis of Iobitridol in an injectable preparation by 1 H NMR spectroscopy., Journal of Pharmaceutical and Biomedical Analysis (2014), http://dx.doi.org/10.1016/j.jpba.2014.01.030 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
[rel]
*Graphical Abstract
General ------DATE = 2013/04/23 TIME = 13:25 INSTRUM = spect PULPROG = zgpr
HO O O
I HN
OH
N
F1 (1H) ------------SI = 65536 SF = 400.06 SW_p = 8012.821
OH I
I N
0.2
O
ip t
OH
0.0
0.1
OH
3.5
3.0
[ppm ]
Ac
ce pt
ed
M
an
us
4.0
cr
0.3
HO
Page 1 of 15
*Highlights (for review)
A qNMR method was used for direct assay determination of Iobitridol in an injectable formulation.
Experiments were carried out in deuterated alkaline water with 3,5 dinitrobenzoic acid as internal reference standard.
Method was validated assessing specificity, accuracy, precision, linearity, stability of samples and robustness.
Method proved suitable for routine controls especially when high sample throughput is required.
Ac
ce pt
ed
M
an
us
cr
ip t
Page 2 of 15
*Revised Manuscript
1
1
Quantitative analysis of Iobitridol in an injectable preparation by H NMR spectroscopy.
2
Anna Borioni*, Gianluca Gostoli, Elena Bossù, Isabella Sestili
3
Istituto Superiore di Sanità, Dipartimento del Farmaco, Rome Italy
4
*Corresponding author. Tel.: +39 06 49902516; fax: +39 06 49902625. E-mail address: [email protected].
5 1.
INTRODUCTION
ip t
6 7
Nuclear Magnetic Resonance spectroscopy in quantitative analysis (qNMR) experienced a remarkable
9
increase since about 2005-2007. In the field of pharmaceutics, qNMR raised the interest of the Official
10
Medicine Control Laboratories on the occasion of a criminal adulteration of heparine with a component
11
detectable by NMR spectroscopy and capillary electrophoresis [1,2].
12
As it is nowadays well documented, quantitative analytical methods based on NMR spectroscopy proved to be
13
highly robust, accurate and precise. In general, qNMR allows quantifying compounds, even in mixtures, at
14
sub-micromolar range and at the same time provides information on the chemical structure of the molecules
15
involved. [3,4,5,6,7]. Moreover, quantification by NMR spectroscopy is truly straightforward in comparison to
16
the chromatographic and spectroscopic techniques which require specific standards or correction factors. In
17
fact the area of the NMR signals is directly proportional to the quantity of the nuclei evoking the signal with no
18
dependence on how the nuclei are combined in chemical structures. This implies that molecules with no
19
chemical similarity to the analyte are used as internal reference standard [8].
20
In the present case we report the determination of the content of iobitridol in an injectable aqueous solution
21
(Fig. 1). Iobitridol is an iodine based contrast medium approved in Europe and in the USA for radiographic
22
diagnostic imaging in adult and children [9]. No EP or USP monograph for iobitridol is available though
23
similar substances such as iodixanol, iohexol, iopamidol, iopromide, iotrolan, iopanoic acid, iotalamic acid
24
and ioxaglic acid are reported in EP. For the large majority of them the assay is estimated by potentiometric
25
titration of iodine after a work up consisting of three operational steps (reduction by Zn, boiling, filtration
26
[10,11,12,13,14]
27
To the best of our knowledge only one determination of iobitridol in biological fluids by HPLC is reported [15].
28
We approached the NMR spectroscopy with the aim of accomplishing the quantification of the content of
29
iobitridol directly in the final formulation with minimal work up, in short time and by a medium field instrument
30
which is of prevalent and established use in qNMR. The NMR experiments were carried out according to the
Ac
ce pt
ed
M
an
us
cr
8
1 Page 3 of 15
31
general principles laid down in the general EP monograph 01/2009:20233 Nuclear Magnetic Resonance
32
Spectrometry.
33 34
2.
Materials and methods
36
2.1.
Chemicals and NMR Instrumentation
37
NMR experiments were recorded on a Bruker Avance instrument (Bruker Italia, Milano, Italy) operating at 400
38
MHz equipped with a 5 mm H/ C BBI probe and some preliminary spectra were acquired on a Bruker AM
39
operating at 700 MHz. The software Bruker Xwinnmr 3.1 was used for acquisition, the software Bruker
40
Topspin 1.2 was used for processing.
41
D2O (99.97%) and DMSO-d6 (99.80%) from Euriso Top (Saint-Aubin, France) were used as solvents. NMR 5
42
mm diameter tubes were purchased from CortecNet (Voisins-Le-Bretonneux, France).
43
Sodium hydroxide, 3-(Trimethylsilyl)-1-propanesulfonic acid, Maleic acid TraceCERT
44
Dinitrobenzoic acid TraceCERT CRM were purchased from Sigma- Aldrich (Sigma- Aldrich Italia, Milano,
45
Italy).
46
2.2.
47
Explorative spectra were performed on a solution of iobitridol injectable solution (260 mg) and maleic acid
48
(10 mg) in D2O (1 ml) and on a solution of iobitridol injectable solution (260 mg) and 3,5-dinitrobenzoic acid
49
(10 mg) in DMSO-d6 (1 ml).
50
For quantitative analysis an accurate weighted quantity of 3,5-dinitrobenzoic acid TraceCERT CRM (9.5
51
mg) and an accurate weighted quantity of the injectable solution here named iobitridol FP (260 mg), were
52
added to 1 ml of D2O previously made alkaline by addition of 1.4% w/w of solid sodium hydroxide. After five
53
minutes sonication, 0.7 ml of the solution was transferred into an NMR tube. The longest longitudinal
54
relaxation time (T1max) was experimentally determined with the Bruker inversion recovery pulse program
55
and resulted about 6 s owing to the 3,5-dinitrobenzoic acid protons. Spectra were acquired in triplicate for
56
each independent weighing unless otherwise specified. Acquisition parameters were: pulse length 45°,
57
delay 30 s, temperature 298 K, 16 scans and TD = 64K. The signal at 2.63 ppm (1H, multiplet) was
58
integrated for iobitridol and the singlet signal at 8.88 ppm (2H, doublet) of the 2,6 protons of 3,5-
59
dinitrobenzoic acid (DNB) was integrated as quantitative reference. In the processing an exponential function
us
cr
13
an
1
ip t
35
ce pt
ed
Sample preparation and NMR method
CRM, 3,5-
M
®
®
Ac
®
2 Page 4 of 15
60
of 0.3 Hz was applied, baseline was corrected by the software fifth order polynomial algorithm and phase
61
was adjusted manually. S/N always resulted above 350.
62
Assignation of the proton spectra was based on 2D H- H COSY [16], 2D H- H TOCSY [16], 2D H- H
63
ROESY [16] and 2D H- C HSCQ [17] diagrams. Standard Bruker pulse sequences were employed for 2D
64
experiments. Typically, data were collected for 2K TD and 512 increments (omonuclear 2D) and 2K TD and
65
256 increments for HSQC. Number of scans varied from 48 to 112, TOCSY mixing time was 100 ms, a delay
66
for a one bond H- C coupling constant of 140 Hz was set for HSQC and full proton decoupling of
67
achieved with the waltz16 sequence.
1
1
1
13
13
68 69
1
13
C was
cr
1
1
ip t
1
1
Results and discussion
71
3.1.
Choice of internal standard and dilution solvent
72
Explorative proton spectra were required to select the suitable solvent/reference standard system as the
73
proton spectrum of iobitridol is not easily predictable. Indeed, iobitridol could present partially restricted
74
rotation of the amide bonds around the aromatic ring giving rise to multiple rotational isomers possibly
75
displaying different NMR signals [18].
76
Solutions of iobitridol injectable solution together with maleic acid in D 2O or together with 3,5-dinitrobenzoic
77
acid (DNB) in DMSO-d6 were first used. Results were not conclusive because in both systems no
78
satisfactorily integrable protons of iobitridol could be identified. In D2O, the multiplet at 4.10 ppm attributable
79
to the single proton of the 2,3-dihydroxypropyl substituent (proton 2 Fig. 1) was the most promising candidate
80
for integration. However, it was not finally selected because of its proximity to the large water signal and
81
above all because its relative integral value did not account for a neat number of protons. This fact suggested
82
that the multiplet could represent a mixture of the signals of different conformational isomers.
83
As iobitridol showed a more resolved spectrum in D2O, the aqueous system was further exploited with the
84
purpose of improving signals resolution. In alkaline solution, the proton spectrum of iobitridol FP displayed a
85
distinct signal at about 2.6 ppm accounting for one proton integral which was assigned to the single proton of
86
the propylcarboxamino substituent of the benzene ring (proton 1 Fig.1). The chemical shift of proton 1
87
showed dependence on the sodium hydroxide concentration: the more alkaline the solution the more
88
shielded and resolved the proton was (Fig. 2). This trend may be attributable to the partial deprotonation of
89
the adjacent amide group in alkali: the consequent local increase of electron density led to an upfield shift of
us
3.
Ac
ce pt
ed
M
an
70
3 Page 5 of 15
the proton 1 signal [19]. A 1.4% w/w NaOH concentration in D2O represented a good compromise between a
91
satisfactory separation of the selected proton from the neighboring signals of the N-methyl groups and the
92
drawbacks of high alkalinity (potential active standard degradation and lengthening of the NMR pulse). 3,5-
93
Dinitrobenzoic acid was selected as quantitative reference standard because of its deshielded NMR signals
94
which assured no interferences with the iobitridol signals and with the strong water signal.
95
With the aim of reducing the set up of the acquisition parameters and of standardizing the experiments, the
96
suppression of the water signal was skipped at first. Validation results proved that this approach was correct
97
since the high amount of water in the formulation (about 40:1 molar ratio in respect to active substance) did
98
not actually impair the correct evaluation of the content of iobitridol.
99
3.2.
us
Iobitridol FP assay
cr
ip t
90
The assay value of iobitridol FP was reported in the manufacturer’s certificate of analysis and it was
101
confirmed in our laboratory by repeating the analyses with the same manufacturer’s method. The method and
102
the relevant validation cannot be disclosed being confidential information reported in the Manufacturer’s
103
dossier for Marketing Authorization. Briefly, three different samples were assayed and found to have a
104
concentration of 34.3% (CV% = 0.31%; n=3). This value was in good agreement with the assay reported in
105
the manufacturer’s certificate of analysis (34.2%).
106
3.3.
107
The validation study was based on the ICH Q2 guidelines. The results obtained in the validation study were
108
considered satisfactory and in line with the expected performance already reported in literature. [20, 21].
109
3.3.1. Selectivity
110
Selectivity was demonstrated by showing that in the spectral zones where signals integrations were carried
111
out there was no interference of overlapping signals.
112
In general NMR 2D experiments can support selectivity by smoothly resolving in two dimensions signals
113
which are not separated in monodimensional spectra [20]. 2D H, H TOCSY and H, C HSQC were acquired
114
on iobitridol FP samples, the outcomes confirmed that no interfering signals affected quantification.
115
Furthermore, selectivity can be demonstrated by assessing the capability of the analytical method to provide
116
an exact result for the content of the analyte in a sample. This was confirmed by the results here obtained for
117
accuracy.
118
3.3.2 Accuracy
ed
M
an
100
Ac
ce pt
Validation
1
1
1
13
4 Page 6 of 15
Accuracy was carried out by analyzing three samples of iobitridol FP at about 80%, 100% and 120% by
120
weighing adequate amount of the drug product solution (between about 200 mg and 320 mg). Three
121
independent replicates for each level were prepared and each of these was analyzed in duplicate. The results
122
are reported in Table 1. Note that one of the samples (45%) referring to the 120% level is actually 131%. The
123
difference of +11% was not considered critical and it was included in the calculations.
124
The recoveries (ranging between 99.6% and 100.7%) were considered satisfactory also considering that the
125
analyzed sample is not a pure substance.
126
3.3.3 Precision
127
Repeatability was assessed by performing seven analyses of the same sample carried out by one analyst in
128
the same day. The CV% for the repeatability was 0.25%.
129
Intermediate precision was assessed by analyzing six independently prepared samples. The analyses and
130
sample preparations were carried out by two different analysts in two different days. The total CV% for the
131
intermediate precision was 0.38% (n=12).
132
Averages (and variances) obtained by two analysts on six independent samples were: 34.4% (0.0184) and
133
34.4% (0.0167). ANOVA analysis did not show any statistical difference: F=0.4308011 (Fcrit= 4,964603 -
134
p=005).
135
3.3.4 Linearity
136
Linearity was assessed by varying the amount of the tested samples in the range from 70% to 120% of the
137
target weight (260 mg) of the analytical procedure.
138
The relevant equation was y=0.1346x-0.763 (R =0.9985).
139
3.3.5 Robustness
140
The considered parameters and their variations were reported in Table 2.
141
The experiments were aimed to evaluate the recovery by changing one parameter in each experiment.
142
Robustness of the method was also evaluated on test solutions containing different sample/RS ratios (w/w).
143
In this case the robustness was checked by assessing the linearity of the integral of the iobitridol proton vs
144
the sample/RS ratios in the range +/-50% of the target ratio (Table 2).
145
The relevant equation was y=0.1391x-0.05321 (R =0.9994).
146
Data showed that the method is robust respect to sample/RS ratio variations.
147
3.3.6 Stability of the solution
ce pt
ed
M
an
us
cr
ip t
119
Ac
2
2
5 Page 7 of 15
148
Iobitridol is reported to be quite stable in alkaline solutions. In particular less than 2% degradation of iobitridol
149
in alkaline solution at pH 9 after three months at 50°C is reported [18].
150
Our experiments showed that the assay value of iobitridol is not affected by a 24 hours storage at 25°C when
151
dissolved in 1.4% w/w NaOH/D2O.
152 4.
Conclusion
ip t
153 154
1
In this report a quantitative H-NMR method for quantification of iobitridol in an injectable preparation was
156
developed and successfully validated with good results of accuracy and precision. It is expected that the
157
method is adequate also to quantify iobitridol as drug substance. The method does not require NMR
158
instrumentation equipped with high magnetic fields as validation proved successful on a magnet operating at
159
400 MHz which is nowadays quite common in pharmaceutical laboratories.
160
Furthermore the NMR method is more rapid and simple in comparison to assay determination based on
161
titrimetric and HPLC methodologies; it requires no specific reference standard, there is no need for sample
162
pretreatment and, because of short analysis times, it can be recommended for routine controls especially
163
when high sample throughput is required.
ed
M
an
us
cr
155
164 Acknowledgments
166
The authors acknowledge Mr. Tonino Puccio for technical assistance.
167
169
References: [1]
170 171
I. McEwen, A. Amini, I.M. Olofsson, Identification and purity test of heparin by NMR - a summary of
Ac
168
ce pt
165
two years' experience at the Medical Products Agency, Pharmeur. Bio. Sci. Notes 1 (2010) 65-72. [2]
,
T. Beyer, B. Diehl, G. Randel, E. Humpfer, H. Schäfer, M. Spraul, C. Schollmayer, U. Holzgrabe, 1
172
Quality assessment of unfractionated heparin using H nuclear magnetic resonance spectroscopy, J.
173
Pharm. Biomed. Anal. 48 (2008) 13–19.
174 175
[3]
U. Holzgrabe, Quantitative NMR spectroscopy in pharmaceutical applications, Prog. Nucl. Magn. Reson. Spectrosc. 57 (2010) 229-240.
6 Page 8 of 15
176
[4]
177 178
M. Malet-Martino, U. Holzgrabe, NMR techniques in biomedical and pharmaceutical analysis, 55 (2011) J. Pharm. Biomed. Anal. 1-15.
[5]
1
G F. Pauli, T. Gödecke,. B.U. Jaki, D.C. Lankin, Quantitative H NMR: Development and Potential of an Analytical Method – an Update, J. Nat. Prod. 75 (2012) 834–851.
179 [6]
S.K. Bharti, R. Roy, Quantitative 1H NMR spectroscopy, Trends Anal. Chem. 35 (2012) 5-26.
181
[7]
U. Holzgrabe, R. Deubner, C. Schollmayer, B. Waibel, Quantitative NMR spectroscopy—Applications
182
in drug analysis, J. Pharm. Biomed. Anal. 38 (2005) 806–812. [8]
cr
183
ip t
180
T. Rundlöf, I. McEwen, M. Johansson, T. Arvidsson, Use and qualification of primary and secondary standards employed in quantitative 1H NMR spectroscopy of pharmaceuticals, J. Pharm. Biomed.
185
Anal. (2013) available on line at
186
http://www.sciencedirect.com/science/article/pii/S0731708513004354.
188
an
[9]
P.L. McCormack, Iobitridol: a review of its use as a contrast medium in diagnostic imaging, Clin. Drug. Investig. 33 (2013) 155-166.
M
187
us
184
[10]
Iodixanol European Pharmacopoeia 7.0 07/2010:2215.
190
[11]
Iohexol European Pharmacopoeia 7.0 01/2008:1114.
191
[12]
Iopamidol European Pharmacopoeia 7.0 01/2008:1115.
192
[13]
Iopanoic acid European Pharmacopoeia 7.0 01/2008:0700.
193
[14]
Iotalamic acid European Pharmacopoeia 7.0 01/2008:0751.
194
[15]
P. Bourrinet, H. Feldman, A. Dencausse, C. Chambon, B. Bonnemain, High-performance liquid
ce pt
ed
189
chromatographic determination of iobitridol in plasma, urine and bile, J. Chromatogr. B 670 (1995)
196
369–372.
197
[16]
198 199
M. F. Summers, L. G. Marzilli, A. Bax, Complete 1H and I3C Assignments of Coenzyme B12 through the Use of New Two-Dimensional NMR Experiments, J. Am. Chem. Soc. 108 (1986) 4285-4294.
[17]
200 201
Ac
195
N. E. Jacobsen, NMR Spectroscopy Explained: Simplified Theory, Applications and Examples for Organic Chemistry and Structural Biology, John Wiley & Sons, West Sussex, 2007, pp. 489-550.
[18]
D. Meyer , M. Petta, M.H. Fouchet, M. Vadel, M. Schaefer, M. Dugast-Zrihen, M. Guillemot,
202
Stabilization of the hydrophilic sphere of iobitridol, a new nonionic iodinated contrast agent, Acta
203
Radiol. 37 (1996) 8-16.
7 Page 9 of 15
204
[19]
205
E. Gaidamauskas, E. Norkus, E. Butkus, D.C. Crans, G.˙Grinciené,˙ Deprotonation of β-cyclodextrin in alkaline solutions, Carbohydr. Res. 344 (2009) 250–254.
206
[20]
F. Malz, H. Jancke, Validation of quantitative NMR, J. Pharm. Biomed. Anal. 38 (2005) 813–823.
207
[21]
G. Maniara, K. Rajamoorthi, S. Rajan, G.W. Stockton
Method performance and validation for
quantitative analysis by (1)H and (31)P NMR spectroscopy. Applications to analytical standards and
209
agricultural chemicals, Anal. Chem. 70 (1998) 4921-4928.
Ac
ce pt
ed
M
an
us
cr
ip t
208
8 Page 10 of 15
Table 1
Table 1
26.3%
26.2%
99.6%
80%
28.2%
28.4%
100.7%
80%
26.7%
26.6%
99.8%
100%
35.3%
35.2%
99.7%
100%
34.3%
34.4%
100.5%
100%
33.4%
33.5%
100.6%
120%
39.8%
39.9%
100.3%
120%
39.7%
39.9%
100.3%
131% a
45.0%
44.9%
99.9%
a
The difference of +11% between 120% and 131% was not considered critical and it was included in the calculations.
Mean = 100.1%
0.41%
Ac
ce pt
ed
M
CV% =
ip t
80%
a
Recovery (%)
cr
Found (n=2)
us
Theoretical
an
Level
Page 11 of 15
Table 2
Table 2 Parameter target valuea
Modified parameters
Recovery
P1 =4 0°
100.3%
P1 = 50°
100.3%
Number of scans = 12
99.7%
Number of scans = 20
99.8%
D1 = 25 sec
100.0%
D1 = 35 sec
100.2%
O1p = 4.2
100.9%
O1p = 5.2
100.4%
Receiver gain = 3
99.5%
Receiver gain = 6
100.6%
Temperature = 296 K
99.8%
Temperature = 300 K
100.6%
P1= 45°
ip t
Number of scans = 16
cr
D1 = 30 s
an
Receiver gain = 5
M
Temperature = 298 K
NaOD = 1.2%
100.9%
NaOD 1.4% NaOD % = 1.6%
100.6 %
ce pt
ed
P1 = applied pulse; D1 = delay time, O1p = center of the spectral window.
Ac
a
us
O1p = 4.7 ppm
Page 12 of 15
h
us
g
cr
ip t
Figure 2
f
an
e d
M
c
b
Ac c
ep te
d
a
Page 13 of 15
Figure 1
HO
1
O O
I HN
OH
N OH I
I N
ip t
O
OH
2
ce pt
ed
M
an
us
cr
OH
Ac
HO
Page 14 of 15
Captions to figures
ip t
Figure 1 Chemical structure and IUPAC name of Iobitridol: 1-N,3-N-bis(2,3-dihydroxypropyl)-5-[3-hydroxy-2(hydroxymethyl)-1-oxopropylamino)]-2,4,6-triiodo-1-N,3-N-dimethylbenzene-1,3-dicarboxamide. Proton 1 was integrated for quantification.
Ac ce p
te
d
M
an
us
cr
Figure 2 Preliminary studies on the solvent/ reference standard system: Effect of NaOH concentration on the proton spectra Iobitridol and maleic acid in D2O. a: NaOH 0.1%; b: NaOH 0.3%; c: NaOH 0.7%; d: NaOH 1.5%; e: NaOH 2.0%; f: NaOH 2.5%; g: NaOH 5.0 %; h: From bottom to top NaOH 1.2%, 1.3% and 1.4%. NMR were acquired on a Bruker AV 700 MHz The integral of the multiplet under the arrow was selected for quantification.
Page 15 of 15
