Journal of Chromatographic Science, 2016, Vol. 54, No.

10, 1813–1819
doi: 10.1093/chromsci/bmw162
Advance Access Publication Date: 7 October 2016
Article

Article

Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
Development and Validation of a Normal Phase
Chiral HPLC Method for Analysis of Afoxolaner
Using a Chiralpak® AD-3 Column
Jinyou Zhuang*, Satish Kumar, and Abu Rustum
Merial Inc., A Sanofi Company, 631 Rt 1 South, North Brunswick, NJ 08902, USA
*
Author to whom correspondence should be addressed. Email: jinyou.zhuang@merial.com
Received 20 April 2016; Revised 2 August 2016

Abstract
Afoxolaner is a new antiparasitic molecule from the isoxazoline family that acts on the insect acar-
ine gamma-aminobutyric acid and glutamate receptors. Isoxazoline family of compounds has
been employed as active pharmaceutical ingredient in drug products prescribed for control of
fleas and ticks in dogs. Afoxolaner with a chiral center at isoxazoline ring exists as a racemic mix-
ture. A normal phase chiral high performance liquid chromatography analytical method has been
developed to verify that afoxolaner is a racemic mixture as demonstrated by specific rotation, as
well as to determine enantiomeric purity of single enantiomer samples. A Chiralpak® AD-3 column
(150 × 4.6 mm I.D.) maintained at 35°C was used in the method. Analytes were analyzed with an
isocratic elution using n-Hexane/IPA/MeOH (89:10:1, v/v/v) as the mobile phase with a detection
wavelength of 312 nm. Desired separation of the two enantiomers was achieved in <10 minutes
with resolution and selectivity factors of 5.0 and 1.54, respectively. The analytical method was
appropriately validated according to ICH guidelines for its intended use. ® All marks are the prop-
erty of their respective owners.

Introduction Use of chiral chromatography has proven to be immensely valu-

Single enantiomers of biologically active chiral compounds have been
used in drug products by pharmaceutical industry for many decades.
Individual enantiomer of a chiral compound can potentially exhibit
different pharmacology, toxicology and metabolism activities in the
living system (1). Potentially only one of the two enantiomers may be
responsible for the desired therapeutic response. Whereas the other
enantiomer may be inactive or may even have deleterious effect.
Therefore, it is of a paramount importance to develop analytical
methods that can identify and determine enantiomeric purity of a chi-
ral active pharmaceutical ingredient (API) in bulk lots as well as in
drug products. To provide pharmaceutical industry further guidance
on this key topic US Food and Drug Administration (FDA) issued a
guidance entitled “Development of New Stereoisomeric Drugs” in
1992 (2). As stated in the FDA guidance document, specifications for
the final product should assure identity, strength, quality and purity
from a stereochemical viewpoint.
able for identification and quantitation of chiral compounds. high
performance liquid chromatography (HPLC) analysis for chiral
compounds can be conducted in two ways—indirect or direct. An
indirect way of chiral analysis involves derivatization of the analyte,
leading to two diastereomers followed by separation using tradi-
tional reversed-phase chromatography (3). Whereas, a direct method
utilizes a chiral selector on chiral stationary phases (CSPs) (4–10) or
in the mobile phase to distinguish the enantiomers (11). Direct meth-
ods are most widely used for chiral separations due to (i) availability
of a wide range of chiral selectors such as Pirkle type, cyclodextrins,
chirobiotic phases, protein and polysaccharide based (12, 13), and
(ii) no pretreatment of the analyte is needed.
Among the various types of CSP, modified cellulose and
amylose-based polysaccharides that were originally developed by
Okamoto and Yashima (14) have demonstrated very good chiral
recognition ability in a wide range of applications (15).

© The Author 2016. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com 1813
1814
This paper describes development of a normal phase chiral
HPLC method using amylose-based polysaccharides column for sep-
aration of a racemic mixture of afoxolaner (Scheme 1). Afoxolaner
is a new insecticide–acaricide molecule from isoxazoline family. This
compound is used as the API in veterinary drugs (e.g., NexGard®)
for fleas and ticks control by acting on the insect’s gamma-
aminobutyric acid and glutamate receptors. Afoxolaner is neutral
molecule without any ionizable groups. It has a logP of 5.15 indicat-
ing its strong hydrophobic character. As shown in Scheme 1 afoxo-
laner has one chiral center on the isoxazoline ring. For the purpose
of the method development two enantiomers of the compound were
named “Enantiomer 1” and “Enantiomer 2”.
As afoxolaner is a relatively new chemical entity, at present there
is no published report about the chiral separation of afoxolaner. For
the first time in this paper we report method development and vali-
dation of a chiral HPLC method to analyze afoxolaner and its indi-
vidual enantiomers. The major goals of the method development
were to develop a sensitive, rugged and QC-friendly normal phase
chiral HPLC method that can verify afoxolaner sample is racemic
mixture as well as can determine enantiomeric purity of the single
enantiomers. In this paper, “Enantiomer 1” is treated as the API
and “Enantiomer 2” as impurity or undesired enantiomer. As both
enantiomers are expected to behave similarly in chromatography
(except the retention time), method can be used as well when
“Enantiomer 1” is treated as impurity. Desired separation of the
two enantiomers was achieved in <10 minutes on Chiralpak® AD-3
column (150 × 4.6 mm I.D.) at 35°C with an isocratic elution using
n-Hexane/IPA/MeOH (89:10:1, v/v/v). Analytes were monitored
with a detection wavelength of 312 nm. The new analytical method
has been successfully validated for its intended purpose as per ICH
guidelines and demonstrated to be accurate, linear, precise, specific,
sensitive and robust.
Experimental
Materials
Reference standards and afoxolaner samples were obtained from
Merial, Inc. (North Brunswick, NJ). All solvents were HPLC grade.
Solvents/reagents were purchased from commercial sources and
used as is. n-Hexane was purchased from VWR (Radnor,
Pennsylvania). Isopropanol (IPA), n-propanol and 1-butanol were
purchased from EMD (Billerica, Massachusetts). Ethanol (200
proof) was purchased from Sigma-Aldrich (St. Louis, Missouri).
Cyclohexanol, 2-butanol and 3-methyl-1-butanol were purchased
from Alfa Aesar (Haverhill, Massachusetts). Methanol was pur-
chased from Honeywell (Morris Plain, NJ).
F O
F H
N mobile phases, the effect of a variety of organic modifiers such as
F N
H F F
O F
F
* F
N O F ditions. The final method was then used for analysis of both
Cl
Scheme 1. Molecular structure of afoxolaner.
Zhuang et al.
HPLC columns
Different polysaccharide based HPLC columns were used for
method development. Columns used were Chiralpak® AD-3 (150 ×
4.6 mm, 3 µm particle size), Chiralcel® OD-H (150 × 4.6 mm, 5 µm
particle size) and Chiralcel® OJ-3 columns (150 × 4.6 mm, 3 µm
particle size). Chiralpak® AD column has tris(3,5-dimethylphenyl-
carbamate) amylose derivative as the stationary phase which is
Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
coated on silica gel. While Chiralcel® OD and Chiralcel® OJ col-
umns contain cellulose modified with tris(3,5-dimethylphenylcarba-
mate) and tris(4-methylbenzoate), respectively on silica gel. All
columns were purchased from Chiral Technologies Co. (West
Chester, PA).
Chromatographic analysis
Agilent 1100 or 1200 series HPLC system (Agilent Technologies,
Santa Clara, CA) with UV or DAD detector equipped with
ChromSword® method development software (Merck KGaA,
Darmstadt, Germany) and LC Spiderling line HPLC Column
Selector (Chiralizer Services, L.L.C., Newtown, PA) were used for
method development and/or method validation. The major parts of
the mobile phase were n-Hexane and IPA. Several organic modifiers
(various alcohols) were evaluated as well. Chromatographic perfor-
mance was evaluated by retention time (Rt), selectivity (α), resolu-
tion (Rs) and peak efficiency (N) which were calculated by
ChemStation or Empower.
Afoxolaner (racemic mixture) reference standard and individual
enantiomer solutions were prepared in IPA. Each mobile phase solu-
tions were prepared by pre-mixing required amount of n-Hexane/
IPA and organic modifier (as needed).
Results
Optimization of chromatographic conditions
The main purpose of this work is to develop a fast Normal Phase
HPLC method to separate and quantitate the two enantiomers in
Afoxolaner (racemic mixture). During method development work,
several cellulose based chiral HPLC columns were selected on the
basis of the chemical structure and location of the chiral center of
the analyte and also on the characteristics of the chiral phase of the
analytical columns. Three columns namely Chiralcel® OJ-3,
Chiralcel® OD-H and Chiralpak® AD-3 were selected and screened
by using low polarity mobile phase system. Figure 1 shows the over-
laid chromatograms using these columns with mobile phase of n-
Hexane/IPA (95:5, v/v). Best chiral selectivity and resolution was ob-
tained on Chiralpak® AD-3 column. Therefore, Chiralpak® AD-3
column was selected for further optimization.
As indicated in Figure 2, the separation of the two enantiomers
was obtained using isocratic elution by use of n-Hexane and IPA
with the ratios of 90:10 and 95:5. Better and fast separation is
achieved with n-Hexane/IPA (90:10, v/v). For optimization of the
methanol, ethanol, 1-butanol, hexanol, t-butanol and cyclohexanol
were investigated. The final mobile phase is n-Hexane/IPA/MeOH
(89:10:1, v/v/v). Table I summarized the final chromatographic con-
Afoxolaner (racemic mixture) and Enantiomer 1 samples. Typical
chromatograms of both afoxolaner (racemate) and “Enantiomer 1”
that had a low level of “Enantiomer 2” as the impurity are shown in
Figures 3 and 4, respectively. Figure 3 is a chromatogram of
Afoxolaner (racemic mixture) which indicates the peak area ratio of
Development and Validation of a Normal Phase Chiral HPLC Method 1815

0.180

0.135 AD-3
AU

0.090

Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
OD-H

0.045

OJ-3
0.000

0.00 12.00 24.00 36.00 48.00 60.00

Minutes

Figure 1. Chromatograms of afoxolaner (0.1 mg/mL) on various chiral HPLC columns with mobile phase n-Hexane/IPA (95:5, v/v); flow rate: 0.8 mL/min; wave-
length: 312 nm; column temperature: 35°C; injection volume: 10 μL.

0.340

0.255

0.170
AU

Hexane/IPA (90:10, v/v)

0.085

Hexane/IPA (95:5, v/v)

0.000

0.00 12.00 24.00 36.00 48.00 60.00

Minutes

Figure 2. Chromatograms of afoxolaner (0.8 mg/mL) on AD-3 column with different mobile phases; flow rate: 0.8 mL/min; wavelength: 312 nm; column tempera-
ture: 35°C; injection volume: 10 μL.

Table I. Final Chromatographic Conditions of the Method Results of these evaluations are summarized in sections below.
®
System suitability requirements [blank baseline: no interfering peaks
HPLC column Chiralpak AD-3 150 × 4.6 mm, 3 µm
above 0.1% level at the retention time of two afoxolaner enantio-
Mobile phase n-Hexane/IPA/MeOH (89:10:1, v/v/v)
mers, 0.2% standard solution have signal-to-noise ratio (S/N) ≥10
Temperature 35°C
Flow rate 0.8 mL/min for each enantiomer; standard solution preparation agreement
Wavelength 312 nm (≤2.0%), resolution factor (≥1.2), injector agreement (difference
Injection volume 10 μl between two injections ≤2.0%)] were met prior to performing the
Afoxolaner 0.8 mg/mL method validation experiments.

Enantiomer 1 and Enantiomer 2 is 50:50. Shown in Figure 4 is a Specificity

chromatogram of Enantiomer 1 obtained using the final chro-
matographic conditions. Enantiomer 1 and Enantiomer 2 were well
separated with Rs of 5.4. The %peak area of Enantiomer 1 and
Enantiomer 2 are 99.63% and 0.37%, respectively.
No significant interfering peaks (peak area >0.1%) was observed at
the retention time of the two enantiomers in blank solution. In addi-
tion, no evidence of co-elution was noted using peak purity analysis
(purity angle was less than the purity threshold using Empower
Software) for the two enantiomers.

Validation of method DL/QL evaluation

To ensure that method is suitable for the intended purpose key The DL and QL of method was evaluated using diluted afoxolaner
method characteristics, i.e., specificity, accuracy, linearity, precision, (racemate) standard solution. The estimated DL (S/N~3) was
DL (detection limit) and QL (quantitation limit) were evaluated. 0.06 μg/mL (0.02%) for each enantiomer. The QL for the method
1816 Zhuang et al.

0.260

0.195

AU 0.130

0.065

Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
0.000

0.00 3.00 6.00 9.00 15.00 12.00

Minutes

Figure 3. A typical chromatogram of 0.8 mg/mL afoxolaner racemate analyzed using the final chromatographic conditions (Chiralpak® AD-3 150 × 4.6 mm, 3 µm;
mobile phase = n-Hexane/IPA/MeOH (89:10:1, v/v/v); flow rate: 0.8 mL/min; wavelength: 312 nm; column temperature: 35°C; injection volume: 10 μL).

0.012

0.008 Impurity
Enantiomer 1 (Enantiomer 2)
AU

0.004

0.000

–0.004
0.00 3.00 6.00 9.00 12.00 15.00
Minutes

Figure 4. A typical chromatogram of 0.4 mg/mL “Enantiomer 1” lot analyzed using the final chromatographic conditions (Chiralpak® AD-3 150 × 4.6 mm, 3 µm;
mobile phase = n-Hexane/IPA/MeOH (89:10:1, v/v/v); flow rate: 0.8 mL/min; wavelength: 312 nm; column temperature: 35°C; injection volume: 10 μL).

was set to (0.2 μg/mL; 0.05%) as S/N for “Enantiomer 1” and
“Enantiomer 2” was 21 and 15, respectively. These results demon-
strate the method is sensitive for detection and estimation of both
afoxolaner enantiomers.
Linearity/accuracy
In this study, “Enantiomer 1” is treated as API and “Enantiomer 2”
is treated as impurity. The linearity of both enantiomers was evalu-
ated. Since “Enantiomer 1” was treated as API, its linearity was
evaluated from 0.2% to 125% of the nominal concentration of
0.4 mg/mL. A linear least square analysis of the data gave correla-
tion coefficient (r) of 1.000. The y-intercept obtained was ≤0.5% of
the 100% level indicating that there is no significant bias for quanti-
tation for “Enantiomer 1”. Method is considered accurate for
“Enantiomer 1” because as per ICH Q2B method’s precision (see
section precision), linearity and specificity have been established.
“Enantiomer 2” is treated as impurity and the linearity and
recovery were studied together by preparing a series of “Enantiomer
2” solution with concentrations corresponded to 0.2%, 0.5%,
1.0%, 2.5% and 5.0% in “Enantiomer 1” (0.4 mg/mL). The solu-
tions were prepared by spiking Enantiomer 2 into the “Enantiomer
1” (0.4 mg/mL) solution at calculated ratios. At each level, duplicate
injections were performed.
A linear least square analysis of the data gave correlation coeffi-
cient (r) of 0.999. The y-intercept obtained was ≤3.0% of the 100%
level indicating that there is no significant bias for quantitation for
“Enantiomer 2”. The method is proven to be linear within the inves-
tigational range.
The recoveries of “Enantiomer 2” at each level were also calcu-
lated and the results ranged from 98% to 104% meeting the accep-
tance criteria of 70–130%. The %RSD of the recovery from five
levels in the linearity/recovery study was calculated to be 2% meet-
ing the acceptance criteria of ≤10%. These results demonstrate the
method is linear and accurate for “Enantiomer 1” (as API) and
“Enantiomer 2” (as an impurity).
Precision
The method precision was evaluated by analyzing six solutions of
Enantiomer 1 at 100% level (0.4 mg/mL). The enantiomer purity
value was calculated for each preparation. The average enantiomer
purity was 99.6% and %RSD = 0.0% (n = 6) thus met the criteria
of %RSD ≤2%. These results demonstrate that the method is pre-
cise for determination enantiomer purity of the API samples.
Discussion
Column and mobile phase screening
Polysaccharide-based HPLC columns were selected for the method
development. As polysaccharide-based CSPs lack ionic sites, they
are suitable for chromatography of a neutral compound such as
afoxolaner (16, 17). In addition, polysaccharide-based columns
have been demonstrated to be very useful in separation of a wide
Development and Validation of a Normal Phase Chiral HPLC Method
spectrum of racemate in pharmaceutical industry due to their versa-
tility and durability (18).
Normal phase chromatographic conditions were selected for
the method. This is because hydrogen bonding, dipole–dipole
and π–π interactions which are considered critical for chiral
recognition are known to be more effective under normal pha-
se conditions (14). Among the large number of available
polysaccharide-based columns, Chiralcel® OD, Chiralpak ® AD
and Chiralcel® OJ columns are most commonly used for pharma-
ceutical analysis; therefore, these were selected for the method
development. A detection wavelength of 312 nm was selected for
analysis since afoxolaner has λmax around this wavelength
(Figure 5). In addition, selection of higher wavelength minimizes
potential interference from common solvent/reagents contami-
nants. At the initial stage of method development 0.1 mg/mL
afoxolaner reference standard (a racemic mixture) solution was
used. To increase sensitivity of the method, the concentration
was increased to 0.8 mg/mL later during method development.
The HPLC columns utilized were Chiralpak® AD-3 (150 ×
4.6 mm, 3 µm particle size), Chiralcel® OD-H (150 × 4.6 mm,
5 µm particle size) and Chiralcel® OJ-3 columns (150 × 4.6 mm,
3 µm particle size) columns.
For the method isocratic mobile phase was preferred over a gra-
dient mobile phase due to slow equilibrium of mobile phase in nor-
mal phase chromatography. N-Hexane and IPA were selected as
non-polar and polar components, respectively as they are the most
commonly used solvent combination in normal phase chromatogra-
phy and were found to suitable for afoxolaner separation.
Chromatographic analysis with n-Hexane/IPA (95:5, v/v) mobile
phase and Chiralcel® OJ-3 column showed two broad peaks around
6 and 37 minutes. With same mobile phase there was no separation
of two enantiomers on Chiralcel® OD-H column. Using this mobile
phase combination, the best separation was achieved on Chiralpak®
AD-3 column with a resolution factor (Rs) of 6.0 and selectivity (α)
of 1.8 between the two enantiomers (Figure 1). However, the reten-
tion times of “Enantiomer 1” and “Enantiomer 2” peaks were ~17
and 28 minutes, respectively.
To reduce analysis time, the mobile phase polarity was changed
to n-Hexane/IPA (90:10, v/v). On Chiralpak® AD-3 column, the
two enantiomer peaks eluted at 6 and 8 minutes, respectively with
resolution of 5.6 (see Figure 2). On Chiralcel® OJ-3 column, two
0.80
0.60
0.40
AU
0.20
0.00
228.00 266.00
Figure 5. Typical UV spectrum of afoxolaner in n-Hexane/IPA/MeOH (89:10:/1, v/v/v).
1817
peaks eluted around 22 and 30 minutes, respectively with Rs of 1.6
and selectivity of 1.4. Both peaks showed peak tailing factor great
than 1.5. No separation was achieved on Chiralcel® OD-H column
under this mobile phase condition.
As it has been reported in literature (19) changing from one alco-
hol to another can affect chiral separation (including elution order),
alcohols other than IPA were studied as well in combination with
Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
IPA. The other alcohols evaluated were methanol, ethanol, n-propa-
nol, 1-butanol, 2-butanol, t-butanol, 3-Methyl-1-butanol, Hexanol
and cyclohexanol. These alcohols are different in terms of their
polarity, chain length and steric effects. The different butanols stud-
ied have varied steric effect. All analyses were performed on
Chiralpak® AD-3 column with afoxolaner standard solution
(0.8 mg/mL) with n-Hexane/alcohol (95:5, v/v) as mobile phase.
Partial or no separation was obtained for all alcohols except 3-
Methyl-1-Butanol. However, selectivity factor for 3-Methyl-1-
Butanol (α = 1.23) was much lower than IPA (α = 1.60). Based on
these experiments only IPA was pursed further as the polar compo-
nent in the mobile phase.
Effect of organic modifiers
In normal phase chiral chromatography, it is common to use organic
modifiers such as acids or alcohol to improve HPLC column’s selec-
tivity. The addition of organic acid in the mobile phase is to mini-
mize interactions with residuals silanols for better peak shape and
resolution (20). Also, as reported in literature (5, 21) addition of
small amounts of alcohols does affect column selectivity as well.
Various alcohols were evaluated as organic modifier to the mobile
phase. First, the n-Hexane/IPA (90:10, v/v) was modified to include
1% methanol. Addition of 1% methanol improved both resolution
and selectivity compared to n-Hexane/IPA (90:10, v/v). It appears
that small amount of methanol alters the steric environment around
the chiral cavities thus favoring further differentiation of the two en-
antiomers. However, addition of more methanol (up to 3%) did not
improve resolution or selectivity (Table II).
Besides methanol several other alcohols were also evaluated as the
organic modifiers at 1% level. These were longer linear alcohols such
as ethanol, 1-butanol and hexanol, branched alcohols such as t-buta-
nol and cyclohexanol and diol (propylene glycol). Results of this eval-
uation are summarized in Table III. As shown in Table III, methanol
312 nm
304.00 342.00 380.00
nm
1818 Zhuang et al.

Table II. Effect of MeOH as Organic Modifier in Mobile Phase using AD-3 Column

n-Hexane (%) IPA (%) MeOH (%) RT (min) Resolution (Rs) Selectivity (α)

Enantiomer 1 Enantiomer 2

90 10 6.1 8.3 5.9 1.36

89 10 1 6.3 9.0 8.2 1.45
87 10 3 5.6 7.4 7.0 1.3

Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
(Afoxolaner concentration: 0.1 mg/mL; flow rate: 0.8 mL/min; Wavelength: 312 nm; Column temperature: 35°C; Injection volume: 10 μL).

Table III. Effect of Various Alcohols as Organic Modifiers in Mobile Phase with AD-3 Column

n-Hexane (%) IPA (%) Organic modifier RT (min) Resolution (Rs) Selectivity (α)

Enantiomer 1 Enantiomer 2

89 10 1% MeOH 5.9 8.5 5.0 1.54

89 10 1% Ethanol 6.2 8.7 4.5 1.48
89 10 1% 1-Butanol 6.0 8.2 4.2 1.46
89 10 1% 2-Butanol 6.1 8.6 4.5 1.50
89 10 1% t-Butanol 6.3 9.0 4.5 1.50
89 10 1% Hexanol 6.0 8.3 4.1 1.45
89 10 1% Cyclohexanol 5.8 7.8 4.1 1.42
89 10 1% Propylene Glycol 5.8 8.0 4.3 1.47

(Afoxolaner concentration: 0.8 mg/mL; flow rate: 0.8 mL/min; Wavelength: 312 nm; Column temperature: 35°C; Injection volume: 10 μL).

Table IV. Results Summary for Method Robustness Study

Parameters Variation Retention time (min) Resolution (Rs)

Enantiomer 1 Enantiomer 2

Normal None 6.2 8.6 4.4

Column Temperature −2°C 6.3 8.8 4.7
+2°C 6.1 8.3 4.2
Flow rate −0.1 (mL/min) 7.1 9.8 4.5
+0.1 (mL/min) 5.5 7.6 4.4
Injection volume −2 µl 6.2 8.6 5.0
+2 µl 6.2 8.6 4.0
Wavelength −2 nm 6.3 8.8 4.7
+2 nm 6.3 8.8 4.7
%IPA n-Hexane/IPA/MeOH (91:8:1, v/v/v) 6.8 9.3 4.5
n-Hexane/IPA/MeOH (87:12:1, v/v/v) 4.6 6.0 3.5

(Afoxolaner concentration: 0.8 mg/mL; Chiralpak® AD-3 150 × 4.6 mm, 3 µm).

gave the best selectivity. So based on the data n-Hexane/IPA/MeOH
(89:10:1, v/v/v) was selected as the mobile phase for the final method.
Column temperature
The column temperature effect on the separation of the two enantio-
mers was studied at 25, 30, 35 and 40°C. At lower temperature (25°C)
better separation was achieved (Rs: 6.6) compared to higher tempera-
tures (35°C) (Rs: 5.0) and 40°C (Rs: 4.2). For rugged HPLC method,
a well-controlled column temperature is crucial. Ideally, column tem-
perature should be at least 10°C above or below the ambient temper-
ature (~25°C) for better temperature control. Column temperature set
around ambient is difficult to tightly control due to environmental
fluctuations. Furthermore, 40°C is the highest recommended tempera-
ture for this type of column. Based on these rationales the column
temperature of 35°C was selected for the final method.
Robustness study
The robustness of the method was demonstrated by showing the
capacity of the method remained unaffected while deliberately
changing HPLC parameters. Several key parameters, including %
IPA in mobile phase, flow rate, detector wavelength, temperature
and injection volume were varied around the procedural values.
Resolution of the two enantiomers under each HPLC parameter var-
iation was assessed against the procedural parameters. Except varia-
tion of %IPA in mobile phase, all other method robustness
parameters were evaluated using ChromSword AutoRobust, a com-
puter software that tests method’s robustness with minimal analyst
intervention. Method robustness results are summarized in
Table IV. In all varied conditions, the two enantiomers were well
separated with a minimum resolution factor of 4 and runtime of
<10 minutes. This shows method is robust as well as efficient and
suitable for high throughput analysis in a QC laboratory.
Development and Validation of a Normal Phase Chiral HPLC Method
Conclusions
A new chiral HPLC method was successfully developed and vali-
dated for afoxolaner API. Method has been shown to be linear,
accurate, precise, robust and sensitive. The method can be used to
verify that afoxolaner is a racemic mixture as demonstrated by
specific rotation, as well as to determine enantiomeric purity of
single enantiomer samples. The method is also considered QC
friendly as it is robust, uses isocratic mobile phase with both enan-
tiomers eluting in <10 minutes, and employs commonly used sol-
vents as mobile phase.
Acknowledgments
The authors would like to thank all analytical scientists in Analytical R&D
Merial for their support during this study.
References
1. Nguyen, L.A., He, H., Chuong, P.H.; Chiral drugs: an overview;
International Journal of Biomedical Science, (2006); 2: 85–100.
2. FDA. (1992) Development of New Stereoisomeric Drugs. http://www.fda.
gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm1
22883.htm (accessed April, 2016).
3. Toyo’oka, T.; Resolution of chiral drugs by liquid chromatography based
upon diastereomer formation with chiral derivatization reagents; Journal
of Biochemical and Biophysical Methods, (2002); 54: 25–56.
4. Neue, U.D.; HPLC columns: theory, technology and practices. Wiley-
VCH, New York, NY, (1997).
5. Lu, J., Rustum, A.M.; Separation of the two enantiomers of T-3811ME
by normal-phase HPLC using modified amylose as chiral stationary
phase; Journal of Chromatographic Science, (2008); 46: 466–471.
6. Zhou, L., Welch, C., Lee, C., Gong, X., Antonucci, V., Ge, Z.;
Development of LC chiral methods for neutral pharmaceutical related
compounds using reversed phase and normal phase liquid chromatogra-
phy with different types of polysaccharide stationary phases; Journal of
Pharmaceutical and Biomedical Analysis, (2009); 49: 964–969.
7. Rustum, A.M.; Separation of the R(−)- and S(+)-enantiomers of the ethyl
ester of tiagabine HCl using a Chiralcel-OG column; Journal of
Chromatography. A, (1994); 684: 29–36.
8. Rustum, A.M., Menon, G., Patel, S.B.; Separation of the S(+) and R(−)-
enantiomers of tiagabine HCl and its two chiral precursors by chiral
1819
chromatography: application to chiral inversion studies; Journal of
Pharmaceutical and Biomedical Analysis, (1998); 17: 1439–1447.
9. Rustum, A.M., Estrada, V.; Separation and quantitation of the
S-(+)-enantiomer in the bulk drug tiagabine·HCl by chiral high-
performance-liquid chromatography using a Chiralcel-OD column;
Journal of Chromatography B: Biomedical Sciences and Applications,
(1998); 705: 111–117.
10. Rustum, A.M.; Determination of chiral purity of ethyl nipecotate using a
Downloaded from https://academic.oup.com/chromsci/article/54/10/1813/2527514 by National Science & Technology Library user on 23 May 2023
Chiralcel-OG column; Journal of Chromatography A, (1995); 696:
75–81.
11. Zheng, C., Huang, W.X., Chen, Z., Rustum, A.M.; Separation of chiral
primary amino compounds by forming a sandwiched complex in
reversed-phase high performance liquid chromatography; Journal of
Chromatography A, (2010); 1217: 4965–4970.
12. Beesley, T.E., Scott, R.P.W.; Chiral chromatography. John Wiley and
Sons Ltd, Chichester, West Sussex, (1998).
13. Gübitz, G., Schmid, M.G.; Chiral separation by chromatographic and
electromigration techniques. A review; Biopharmaceutics and Drug
Disposition, (2001); 22: 291–336.
14. Okamoto, Y., Yashima, E.; Polysaccharide derivatives for chro-
matographic separation of enantiomers; Angewandte Chemie
International Edition, (1998); 37: 1020–1043.
15. Bargann-Leyer, N., Tambute, A., Caude, M.; A comparison of LC and
SFC for cellulose- and amylose-derived chiral stationary phases; Chirality,
(1995); 7: 311–325.
16. Yashima, E.; Polysaccharide-based chiral stationary phases for high-
performance liquid chromatographic enantioseparation; Journal of
Chromatography A, (2001); 906: 105–125.
17. Perrin, C., Vu, V.A., Matthijs, N., Maftouh, M., Massart, D.L., Heyden,
Y.V.; Screening approach for chiral separation of pharmaceuticals: Part I.
Normal-phase liquid chromatography; Journal of Chromatography A,
(2002); 947: 69–83.
18. Tachibana, K., Ohnishi, A.; Reversed-phase liquid chromatographic sepa-
ration of enantiomers on polysaccharide type chiral stationary phases;
Journal of Chromatography A, (2001); 906: 127–154.
19. Persson, B., Andersson, S.; Unusual effects of separation conditions on
chiral separations; Journal of Chromatography A, (2001); 906: 195–203.
20. Stringham, R.W., Ye, Y.K.; Chiral separation of amines by high-
performance liquid chromatography using polysaccharide stationary
phases and acidic additives; Journal of Chromatography A, (2006); 1101:
86–93.
21. Kunath, A., Theil, F., Jähnisch, K.; Influence of the kind of the alcoholic
modifier on chiral separation on a Chiralpak AD column; Journal of
Chromatography A, (1996); 728: 249–257.