HPLC Method

development for
 Conten
ts
1 Introduction

2 Regulatory Aspects

3 Implementation of QbD

4 Proposals of EFPIA/PhRMA Paper

5 QbD- Analytical methods

6 Benefits of QbD

7 Challenges
2
Pharmaceutical Quality in the 21st Century – A Integrated Systems Approach - Janet Woodcock
 What are some standard
method development
practices ….?
Follow preferred method development scheme and do “hand
method development – based on selectivity – changing para
e.g., pH, column or mobile phase types.
Use method development software – run a few predictive run
get prediction for best method.
Evaluate multiple columns/multiple mobile phases in a manu
automated fashion, determine best results (e.g. using value
switching to change between columns, software to track pe

actice #1 is used by many and incorporates “hands on” learn

requires nothing special to make it work, so we will work thro
method development scheme for this practice.
 Bio-Separations : An Overview

Analytical Separation &

methodologies Purification Methods

HPLC Mass Liquid-Liquid Membrane

Spectrometry Distribution Chromatography
Capillary Methodology Simulated
Electrophoresis Montage Displacement
Moving Bed
Chromatography
Chromatograph

Expanded-Bed
Adsorption
 Amino acids are the basic units of polypeptides and proteins.
 The primary amino acid sequence and tertiary structure (i.e.,
three-dimensional shape) are the most important features to
consider when planning an HPLC separation.
 Knowledge of the primary amino acid sequence allows for
consideration of which functional groups are present in what
quantity. Both amino acid side-chain and main-chain functional
groups can interact with HPLC phases.
 Knowledge of the protein's tertiary structure provides
consideration of which functional groups are on the exterior of
the protein. With a protein in its native conformation, surface
side chains will have the greatest interaction with the HPLC
support.
 During the course of an HPLC separation, unfolding or
denaturation of a protein or peptide may occur. Irreversible
unfolding will lead to a loss of the protein's biological activity.
 Unfolding can occur as the protein moves through the HPLC
system, and during this unfolding, interior amino acid side
chains will be exposed.
 Often, the interior amino acids of a protein are predominantly
hydrophobic and, during unfolding, additional interactions
with non-polar stationary phases will occur, leading to
increased retention and decreased separation efficiency.
Unfolding can occur in different degrees, and the native
conformation and unfolded conformations can have different
retention times.
 Various components of the chromatographic system can
contribute to protein unfolding, including stationary phase
hydrophobicity, mobile phase polarity, mobile phase pH, ion-
pairing agents, detergents, and oven temperature.
 The possibility of intermolecular interactions must be
considered when unexpected peaks or low recovery occurs
The hydrophobic amino acids are
shielded from the solvent,
providing stability to the structure.
Proteins and peptides can be
analyzed by several HPLC
modes, with reversed-phase
liquid chromatography (RPLC)
and hydrophobic interaction
chromatography (HIC) the most
common and the easiest for
method development.
However, ion-exchange
chromatography (IEC) and size-
exclusion chromatography (SEC)
are also used to successfully
separate and purify mixtures
containing proteins and/or
RPLC and HIC are popular means of separating or purifying proteins
and peptides.
The hydrophobic character of proteins and peptides enables elution.
Mixed hydrophobicity and charge characteristics of protein mixtures
lead to non-ideal chromatographic behavior.
HPLC methods tend to work well with small proteins and peptides, but
suffer with larger or more unstable protein samples.
Unfolding.
The stationary phase support, ligates, surface tension, mobile phase
polarity, temperature, pH, and mobile phase additives can all
contribute to unfolding and may need to be adjusted during method
optimization.
Conformation changes due to unfolding are minimized at low
temperatures and methods designed to prevent unfolding may be
more successful at 25-30~ r
Some protein unfolding is reversible. Partial unfolding that may occur
during RPLC or HIC analyses does not necessarily mean bioactivity is
lost.
For RPLC or HIC method, parameters to be optimized
include
column (stationary phase) choice,
type and concentration of ion-pairing agent,
organic solvent choice,
pH,
salt choice and concentration (HIC),
mobile phase gradient,
temperature.
 1
Column & S.P Choice

Particle size.
For RPLC, sample loading capacity is related to 2 the effective surface area available
for interaction, aMobile phase
property related to the particle size of the packing, among other
factors.
characteristics
Particle sizes of RPLC columns range from 1.5 to 25 µm. Band broadening can be
a problem with large samples.
3
HIC columns have particle sizes of 15-65 µm, and a larger sample can be
accommodated, Choice of column
assuming other Ion pairfactors including bonding density, type of
bonding, porosity and stationary phase
reagentvolume are equal.

Pore size can be chosen based on knowledge of molecular diameter.

The minimum pore diameter that is acceptable for protein-ligate interaction is four
times the protein's diameter.
For beginning method development for a protein separation, choosing a pore size
of greater than 30 nm will ensure acceptable diffusion of the large analyte into the
pore. For peptide separation, a pore size of 10 nm is generally sufficient.
In RPLC, this is a water-organic eluent. Typically, aqueous-based mobile phases
are used since non-aqueous phases contribute to protein denaturation.
Often the mobile phase is at a low pH to stabilize ionizable side chains.
A gradient mobile phase program that increases the concentration of organic is
utilized for samples requiring separation of several components. Gradient programs
running to 100% organic are common for pharmaceutical products that have
complex matrices in order to ensure elution of compounds that associate strongly
with the stationary phase.
The gradient used in HIC is an increasing salt concentration in a buffered aqueous
mobile phase.
 IEC has found a home as the
common technique for the
separation of charged species.
 IEC achieves separation of
proteins and peptides by analyte-
ligate electrostatic interactions.
 Multiple charged functional
groups.
 Electrostatic interactions between
ions in the solute, mobile phase,
and stationary phase are the
primary contributors to retention.
 Elution of these biomolecules is
complicated by the presence of
multiple charged regions on a
molecule.
In singly charged molecules,
elution is
 Method development parameters in IEC include
stationary phase choice
mobile phase pH
ionic strength The IEC
Elution incolumn consists
IEC occurs of the
through
choice of M.P ionsstationary
The degree
increasing phase
the of resin withofof
ionization
ionic strength weak
the
temperature. immobilized
[Link]
exchangers
mobile changes
Thus, a orgradient
weak cation
as mobileof phase
Small
orpH amounts
anion of organic
exchangers.
approaches the pK a of is the
increasing salt concentration the
modifier
The may
degree
exchanger's of functional
ionization
minimum requirement forgroup.
an IEC
bedifferentiates
added to minimize
phase. a mechanism
The retention
mobile strong exchanger
for IEC of
hydrophobic interactions with
Afrom a weak
proteins
small and
change exchanger.
peptides becomes very
in ionic strength
the stationary
Strong
can haveanion
complicated
a largeexchangers,
with usefor
the on
effect of weak
phase.
example, quaternary
exchangers. As amine
a basic guide
retention since many groups on for:
a
functional
protein maygroups,
choosing abecolumn, or strong
affected acidic cation are
proteins
at one
exchangers,
analyzed
time. onsuch
The choice as
anofanionsulfonated
salt exchange
affects
functional
column, and
selectivity, groups,
basicare
although completely
proteins
predicting are
the
ionized.
analyzed
exact effect on a cation
of any exchange
particular salt column.
Separations of proteins and
. SEC tends to be used for
separation of high molecular weight
species from biological matrices .
. It does not play an extensive role
in identification or quantitation of
proteins or peptides mainly
because of the low resolution.
. Molecular hydrodynamic volume.
. Hydrophobic interactions.
. Protein unfolding.
Ex. protein aggregation can lead
to
decreased SEC retention
time.
. Method development parameters
for SEC include buffer choice, salt
concentration, organic modifier and
flow rate.
 Mobile phase optimization is the most difficult portion of SEC
method development.

 SEC mobile phases are generally neutral buffers with a salt

concentration near 0.1 M. However, method development will
likely require adjustment of pH and salt concentration.

 The wide variety of possible protein and peptide combinations

makes it difficult to suggest particular changes.

 Not only does the hydrodynamic volume of proteins change

with pH, mobile phase additives, or organic modifiers, but for
peptide separation, a significantly different mobile phase
composition may be required to achieve elution of different
classes (e.g., acids, neutrals, etc.).
 UV detection

 PDA detection
 It is a very popular detection method
foroffer
 It can pharmaceuticals.
increased sensitivity and
 Large peak around 210nm, and a
 Derivative
multiple wavelength analysis.
spectroscopy
smaller peakpurity
around 280nm.
 Allows
 for peak to be
 The prominence
210-220nm due of
to computer-assisted
the absorbance by
assessed.
analysis in pharmaceutical labs has
the peptide bond.
 Fluorescence
 led and electrochemical
to derivative
Aromatic amino spectroscopy as a
acid side chains
detection detection
absorb attechnique.
280 nm.
 First, second, third, even fourth
derivatives of the UV spectra can be
 The evaporative
plotted. lightto scattering
 It is likely find usefulness for
 The resulting new minima or maxima
detector (ELSD).methods with highly volatile mobile
offer additional information regarding
phases and lower-volatility high
themelting
peak. point analytes.
 Columns and the Separation of Proteins and Peptides

An investigation of HPLC protein and peptide separations using micro-bore HPLC

columns was carried out by Issaq et al.
The group looked at the effect of changes in column parameters and mobile phase
composition on separation efficiency and selectivity.
The group found improved sensitivity using smaller diameter columns and
the smallest pore size that is appropriate for the size of protein or peptide
being analyzed.
A tryptic digest of myoglobin showed maximum sensitivity
with a 0.18 mm i.d. column and silica pore size of 90 A. A 300 ~i
pore size would be appropriate for analysis of larger proteins and
polypeptides.
Several bonded phases, C18 , C8, C3, and CN, were tested
for their effect on a polypeptide separation.
The polypeptides separated include Rnase, insulin, cytochrome C, lysozyme,
pavalbumin, CDR, myoglobin, carbonic anhydrase, S-100b, and S-100A.
Increasing the stationary phase alkyl chain length from C3-C 8 to C18 increased
retention and resolution for all of the polypeptides except CDR and myoglobin.
Those two analytes nearly co-eluted on the C18 column, co-eluted on the
C 8 column, then reversed elution order and eluted well resolved on both
the C 3 and CN columns.
Thus, the increased hydrophobicity of a longer alkyl chain can lead to changes in
the three-dimensional shape of a protein that may account for the change in
selectivity.
 Mobile Phase Conditions and the Separation of Proteins and Peptides

For the investigation of mobile phase conditions in the same study, a reversed-phase
gradient was used since the majority of protein and peptide separations are carried
out using RPLC.
The basic gradient system was made of aqueous +0.1% TFA mobile phase A and
organic +0.1% TFA mobile phase B with a linear gradient increasing in organic
content.
There was little difference in separation when acetonitrile and methanol
were compared as the organic modifier.
The only difference in the separation of a mixture of five peptides (Gly-Tyr, Val-Tyr-
Val, methionine enkephalin, leucine enkephalin, and angiotensin II) was improved
resolution of the leucine enkephalin and angiotensin II peaks in the chromatogram
with methanol as the modifier.
As the final concentration of organic in the linear gradient was increased (5-35%, 5-
50%, 5-75%), retention decreased and efficiency increased. Selectivity changes
occurred between leucine enkephalin and angiotensin II.
At 5-35% acetonitrile, the peaks were resolved; at 5-50%, the peaks co-eluted; and
at 5-75%, the peaks were resolved but the elution order was reversed. Several
ionpairing agents were compared including TFA, heptafluorobutyric acid
(HFBA), and formic acid.
In the separation of a cytochrome C digest, TFA gave the best sensitivity, peak
shape, and resolution.
The effect of ion-pair agent concentration was also studied using the five-peptide
mixture.
A change in concentration of TFA affected peak height, selectivity, and resolution.
An increase in TFA concentration improved peak symmetry and
peak width. The changes in resolution and selectivity did not follow a
trend
 Ion-Pairing Agents and the Separation of Peptides

A detailed series of studies on ion-pairing agents and HPLC peptide

separations was carried out by Hodges and coworkers.

They looked at the concentration of TFA and determined that increasing

the
concentration of TFA to 0.20-0.25% increased sensitivity and separation
of a mixture of peptides with similar charges.

The group also found that increasing the temperature from 25 to 70~
gave additional
improvement in separation.

The same group compared four ion-pairing reagents (phosphoric acid,

TFA, pentafluoropropionic acid (PFPA), and HFBA) at various
concentrations for the effect on separating two peptide mixtures, one
mixture containing peptides with the same
net charge, the other mixture containing peptides with varying net
The researchers found that for both peptide mixtures, retention time
increased with increasing concentration and hydrophobicity of the
reagent.

Hydrophobicity had a larger affect on change in retention than

concentration change. Increasing hydrophobicity of ion-pairing reagent
also improved peak shape.

For peptides of identical charge, resolution was not affected by choice of

ion-pairing reagent and was affected only slightly by an increase in
concentration.

For peptides of varying charge, resolution increased with increased

reagent concentration, although improvement in resolution diminished
between 20 and 40mM for all reagents.

At 20mM, the best overall resolution was achieved with phosphoric acid,
although phosphoric acid also showed the least change in resolution with
an increase in reagent concentration.
 The Separation and Evaporative Light Scattering Detection of
Phospholipids

Two research groups recently published methods to separate and quantitate

phospholipids using HPLC with ELSD.
Both methods separated a variety of phospholipids including
phosphatidylethanolamine (PE), phosphatidylcholine (PC), phosphatidylinositol (PI),
lysophosphatidylcholine (LPC), sphingomylelin (SM), and phosphatidic acid (PA).

Rombaut et al. TM utilized a 3.2 x 150mm, 3 gm silica column and a linear mobile
phase gradient.
The gradient began with 87.[Link].5 choloroform : methanol : buffer (1 M formic acid
with triethylamine to pH 3) adjusting to [Link] in 16 min. The buffer pH was found to
affect peak shape and resolution of PS, PI, and PC.
The acidic pH also increased column life. Several ELSD conditions were critical for
detection of phospholipids.
The optimal flow rate of the nebulizing gas (N2) was 1.4 mL/min and optimal
nebulizing temperature was 85~ Increasing flow rate led to decreased response of all
phospholipids. Increasing temperature had a similar but less pronounced effect.
Detection limits for the phospholipids ranged from 11 to 40 ng.
Zhang et al. is used a 4.6 • 250 mm, 5 gm silica column with a gradient flow
using a three-part mobile phase. An initial, isocratic composition of [Link]
isopropanol: water: hexane at flow rate of 0.8 mL/min was held for 8 min, then
programmed to a final mobile phase composition of [Link] from 8 to 25min at
1.2mL/min. Finally, an isocratic mobile
phase composition of [Link] was set from 25 to 35 min at 0.8 mL/min. The
percent water in the mobile phase was the critical factor leading to elution and
resolution of the phospholipids. ELSD conditions were 2 mL/min carrier gas
(N2), drift tube temperature of 63.5~ and pressure of 0.4 MPa. This group
determined that the drift tube temperature was the most important ELSD
parameter during the analysis. Temperatures lower than 63.5~ led to baseline
drift, and temperatures above 63.5~ led to decreased signal-to-noise ratio.
 As a new type of chromatographic stationary phase, monoliths have
been subjected to intensive study in the last years. They differ from
other supports mainly in their characteristic structure, which results in
the improved chromatographic properties.
 Packed (a) and monolithic (b) chromatographic column.
 While most of the chromatographic supports are particle shaped,
monoliths consist of a single piece of highly porous material. In
contrast to porous particle, the pores inside the monolith are open,
forming a highly interconnected network of channels. Monoliths can
be prepared in various ways and can have an inorganic or an organic
based skeleton.
 Silica-based monoliths
 The first being silica-based monolithic columns, generally prepared
using sol‑gel technology. This technology can be applied to create a
continuous sol‑gel network throughout the column former by gelation
of a sol solution within the capillary. Alternatively, it can be used to
glue LC silica-based particles, once the capillary has been packed
conventionally, producing a continuously bonded bed.
Organic polymer-based monoliths
The second category is rigid organic polymer-based monolithic columns, and these
include acrylamide-based, methacrylate-based and styrene‑based polymers.
The polymer network is generally formed inside the capillary by a step-wise chain
polymerization reaction.
Polymerization reaction mixtures usually consist of a combination of monomers
and cross-linker, initiator and a porogenic mixture of solvents.
A variety of monomers can be employed to fabricate the final monolith, being
both charged and hydrophilic, to generate electroosmotic flow for capillary
electrochromatography, or uncharged and hydrophobic, to allow reversed-phase
interactions used in HPLC.
The cross-linker concentration can be adjusted to change the degree of cross-
linking which influences the overall porosity. An initiator is needed to begin the
step-wise chain reaction, and it is often 2,2’‑azo‑bis‑isobutyronitrile (AIBN).
The polymerization can be initiated using UV light or thermal treatment.
Precipitation of the polymer occurs after it becomes insoluble in the reaction
medium.
Solubility is influence both the cross-linking and choice of porogen (a poor solvent
for the polymer), which is commonly a mixture of alcohols.
The formation of the monolith can be achieved in-situ within either untreated
or pre-treated capillaries.
The pre-treatment of the capillary often involves surface preparation for the
introduction of double-bond functionality, allowing covalent bonding of the
monolith to the capillary wall, which is of particular importance for HPLC
application where the monoliths needs to withstand high pressures.
Preparation of organic polymer monoliths
The polymerization mixture is forced into the capillary and generally initiated
 Historic Review
Hjerten introduced
term continuous polymer bed
for his compressed
polyacrylamide gel. The popularity of these
Hjerten
terms
For example, BIA
1980’s p o l y (glycidy
l
Separations is using their
ro u s
macropo crylate-co- CIM (Convective
metha ts
h a c r y l a te) shee Interaction Media) disks
e di m e t merizati
on
-ethylen u l k p o l y for separation processes
d by a b
materials prepare " m a c r oporous they call "high-
having simila
r 1990’s proces s s"
me mbrane performance monolithic
macroporous polymer chromatography" since
d
structure an 1998.
cylindrical Recently, Merck K.G.
sh a pe commercialized their silica
Continuous rod columns under the
polymer 1993 trade name Chromolith |
rods in which the
Later expression monolith is
First appearance of the
also hidden.
term "monolith" to
describe a single piece of Expression "monolithic"
functionalized related to rigid macroporous
cellulose sponge used for polymers prepared by bulk
the protein separation polymerization in a closed 7
mold has also been
 Development of monolithic
materials
Early attempts
Hydrogels
 Researchers published the preparation of a continuous
polymer matrix and its use for the chromatographic
separation in size-exclusion mode.
 The column packed with the hydrogel particles afforded
slightly better HETP of 0.88 mm but the selectivity was also
rather poor.
 The failure of this early approach most likely results from
both unsuitable porous structure of the gel and column
clogging by the action of hydrostatic pressure used to drive
the flow.
 Only Hjerten revisited the homogeneous swollen gels, used
them in capillary electrochromatographic mode with
electroosmosis as the flow driving force and achieved
excellent separations.
 Foams
 In 1970s, two groups
experimented independently with
polyurethane foams prepared in
situ within the confines of large
chromatographic columns.
 The Monsanto group adopted idea
of shaping polymer foams and
inserting them into a column. They
extended the concept and
prepared open pore polyurethane
foam in situ.
 They demonstrated that a decent
gas chromatographic separations
of various hydrocarbon mixtures GC separation of C6-C9 alkanes on
could be achieved. open pore polyurethane column.
 The German group explored the polyurethane
foams at about the same time demonstrated two
techniques:
(i) production of a foam in situ and
(ii) packing columns with the powdered foam.
 They expanded the arsenal of polymers and, in
addition to polyurethane, also used other foamed
polymers such as styrene copolymers, natural
rubber, and polyethylene.
Recent developments
Compressed beds
 Prof. Hjerten knew from his early experiments
that beds of soft beads packed in
chromatographic columns could be deformed to
make the distances between the beads smaller
thus achieving an enhanced resolution.
 The major difficulty was to design such beds that
simultaneously fulfilled two contradictory
requirements in order to obtain the desired high
resolution: a relatively low backpressure upon a
strong compression of the beads.
 Continuing in these efforts, his workers proposed
an alternative approach, the gel was cut to small
This
compression
of the bed to
only about 17
% of its
original length
resulted in a
column
enabling an
excellent
chromatograp
hic separation HPLC of model proteins in cation-exchange mode
using compressed continuous gel at different
of proteins in indicated flow rates
ion-exchange
 Monolithic disks
 Since objects in the bead shape were not
well suited for the measurements of mass
transfer through the pores.
 This led to the successful development of
chromatographic media in a novel shape-
flat disks.
 However, soon thereafter, a small
company BIA (recently renamed to BIA
Separations) also noticed the advantages
of the monolithic disk technology and got
rapidly involved.
 Quality by design and
operations

Related
Related Regulatory
Regulatory Guidance
Guidance

Facility Operation
QbD based

Risk Pharm.
Pharm. Managemt.
Develop. Quality
Q9 Systems
Q8
Q10

8
 Linkages – Q8, Q9 and Q10
TEXT
Risk Pharm.
Pharm.
Managemt. Quality
Develop.
TEXT Q9
TEXT TEXT
Systems
Q8
Q10

Pharmaceutic Quality Risk Pharmaceutica

al Management l
Development Q9 Quality
Q8 System
Application Q10
Collecting to our need
Maintenance of
the using the our product,
knowledge knowledge process
needed throughout the 9
product lifecycle
 DOE APPROACH
• Select the study
variables

• Define the study

1 ranges
Define the experimental
design region Conduct formal
experimental
design
2 Analyze results-
Develop the built
equations
knowledge base
Define best
3 conditions
Define process
Establish the design
space robustness
19
The QbD Based methodology for development of robust analytical methods, [Link] Verseput
 ATP

Zidovudine

Zidovudine is available as a
Formula :
300-mg tablet, a 100- or
C10H13N5O4
250-mg capsule, a 50-mg/5
Mol. Mass : 267.242
mL syrup and a 200-mg/20
g/mol
mL injection solution.
CAS number : 30516-
The following impurities are
87-1
limited by the British and
Description : White to
European pharmacopoeias:
off-white crystals or  1-[(2R,5S)-5-
needles.
hydroxymethyl-2,5-
Solubility : Soluble in
dihydro-2-furyl]-5-
water (25 mg/mL at 25 °C)
methylpyrimidine-
and ethanol
2,4(1H,3H)-dione.
(67 mg/mL)  1-(3-chloro-2,3-dideoxy-β-
Optical rotation : +99°
D-ribofuranosyl)-5-
(c = 0.5 in water)
methylpyrimidine-
Dissociation constant : 24
2,4(1H,3H)-dione.
 Method Development
System suitability criteria can be defined Define analytical method
to manage risk and ensure the method performance control strategy
delivers the desirable method attributes.
Perform risk ass.
Structured methodologies such as with rugged. & robust.
Fishbone diagram is implemented
Evaluate exp. Results &
level- tailing, symmetry, fronting
select final method
I level- tailing less than 1.5

Perform the experimental

Comprehensive Exp. Design
design
Based on syst. scouting

To separate & Define the method intent

quantify

QbD approach to analytical RP-HPLC method development and its validation, ijpps-journal, Vol
253,
 Highly complex
Protein mixtures
Text

Tertiary structure

Purification Textfor
HPLC
processes Biomol. Is
Txt
critical
Txt

Degradation
Functional products
Text
groups
Knowledge and Experience Beats Blind
Experimentation
 Diagram

Title
Add your text

ThemeGallery ThemeGallery
is a Design Digital is a Design Digital
Content & Contents Content & Contents
mall developed by mall developed by
Guild Design Inc. Guild Design Inc.
 Diagram

Text

Add Your Add Your

Title Text Title Text
• Text 1 Text • Text 1
• Text 2 • Text 2
• Text 3 Text • Text 3
• Text 4 • Text 4
• Text 5 • Text 5
Text

Text
 Diagram

Text Text

Text Concept Text

Text Text

Add Your Text

 Diagram

Add Your Text

Add Your Text Add Your

Title

Add Your Text

 Diagram

Text Text Text

Add Your Text Add Your Text Add Your Text

 Diagram

Text

Text

Text Add Your Title

Text
 Progress Diagram

Phase
Phase 11 Phase
Phase 22 Phase
Phase 33
 3-D Pie Chart

Text2
Text3

Text1
Text4

Text5
 Marketing Diagram

Add
Add Your
Your Text
Text

Add Your Title here

Text1 Text1 Text1 Text1