ArticlePDF AvailableLiterature Review

Innovative Drying Technologies for Biopharmaceuticals

Authors:
  • SETU Waterford
Article

Innovative Drying Technologies for Biopharmaceuticals

Abstract and Figures

In the past two decades, biopharmaceuticals have been a breakthrough in improving the quality of lives of patients with various cancers, autoimmune, genetic disorders etc. With the growing demand of biopharmaceuticals, the need for reducing manufacturing costs is essential without compromising on the safety, quality, and efficacy of products. Batch Freeze-drying is the primary commercial means of manufacturing solid biopharmaceuticals. However, Freeze-drying is an economically unfriendly means of production with long production cycles, high energy consumption and heavy capital investment, resulting in high overall costs. This review compiles some potential, innovative drying technologies that have not gained popularity for manufacturing parenteral biopharmaceuticals. Some of these technologies such as Spin-freeze-drying, Spray-drying, Lynfinity® Technology etc. offer a paradigm shift towards continuous manufacturing, whereas PRINT® Technology and MicroglassificationTM allow controlled dry particle characteristics. Also, some of these drying technologies can be easily scaled-up with reduced requirement for different validation processes. The inclusion of Process Analytical Technology (PAT) and offline characterization techniques in tandem can provide additional information on the Critical Process Parameters (CPPs) and Critical Quality Attributes (CQAs) during biopharmaceutical processing. These processing technologies can be envisaged to increase the manufacturing capacity for biopharmaceutical products at reduced costs.
Content may be subject to copyright.
International Journal of Pharmaceutics 609 (2021) 121115
Available online 20 September 2021
0378-5173/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Review
Innovative Drying Technologies for Biopharmaceuticals
Ashutosh Sharma
a
,
*
, Dikshitkumar Khamar
b
, Sean Cullen
c
, Ambrose Hayden
a
, Helen Hughes
a
a
Pharmaceutical and Molecular Biotechnology Research Centre (PMBRC), Waterford Institute of Technology, Main Campus, Cork Road, Waterford X91K0EK, Ireland
b
Sano, Manufacturing Science, Analytics and Technology (MSAT), IDA Industrial Park, Waterford X91TP27, Ireland
c
Gilead Sciences, Commercial Manufacturing, IDA Business & Technology Park, Carrigtwohill, Co. Cork T45DP77, Ireland
ARTICLE INFO
Keywords:
Biopharmaceuticals
Drying Technologies
Characterization Techniques
Formulation
Process Analytical Technology
ABSTRACT
In the past two decades, biopharmaceuticals have been a breakthrough in improving the quality of lives of
patients with various cancers, autoimmune, genetic disorders etc. With the growing demand of bio-
pharmaceuticals, the need for reducing manufacturing costs is essential without compromising on the safety,
quality, and efcacy of products. Batch Freeze-drying is the primary commercial means of manufacturing solid
biopharmaceuticals. However, Freeze-drying is an economically unfriendly means of production with long
production cycles, high energy consumption and heavy capital investment, resulting in high overall costs. This
review compiles some potential, innovative drying technologies that have not gained popularity for
manufacturing parenteral biopharmaceuticals. Some of these technologies such as Spin-freeze-drying, Spray-
drying, Lynnity® Technology etc. offer a paradigm shift towards continuous manufacturing, whereas PRINT®
Technology and Microglassication
TM
allow controlled dry particle characteristics. Also, some of these drying
technologies can be easily scaled-up with reduced requirement for different validation processes. The inclusion of
Process Analytical Technology (PAT) and ofine characterization techniques in tandem can provide additional
information on the Critical Process Parameters (CPPs) and Critical Quality Attributes (CQAs) during biophar-
maceutical processing. These processing technologies can be envisaged to increase the manufacturing capacity
for biopharmaceutical products at reduced costs.
1. Introduction
Biopharmaceuticals, composed of either peptides, proteins, glyco-
proteins and nucleic acids or composite combinations of biomolecules,
include antibodies, antibody-drug conjugates, recombinant proteins,
nanobodies, enzymes, hormones, vaccines and gene therapy products
etc. (Gervasi et al., 2018; Walsh, 2010). The stability of protein-based
biopharmaceuticals can be negatively affected by temperature, mois-
ture, prolonged storage, denaturants, organic solvents, shear, oxygen,
changes in pH etc. (Declerck, 2012; Li et al., 2015). Protein aggregation
is one of the major challenges that impacts the CQAs of biopharma-
ceutical products (Wang and Roberts, 2018). This can be encountered at
any stage of the manufacturing process to the administration of drugs.
Parenteral biopharmaceuticals are marketed as either liquid or solid
products manufactured by various ll nish technologies (Martagan
et al., 2020). With about 34 % of biopharmaceuticals marketed as
Freeze-dried products in the European market (Gervasi et al., 2018),
dehydration via Freeze-drying is the gold standard for manufacturing
solid biopharmaceuticals. Several advantages of Freeze-drying have
been identied. Freeze-drying allows improved shelf-life of heat-
Abbreviations: ADH, Alcohol dehydrogenase; API, Active Pharmaceutical Ingredient; BSA, Bovine Serum Albumin; cGMP, current Good Manufacturing Practices;
CPP, Critical Process Parameters; CQA, Critical Quality Attributes; DNase, Deoxyribonuclease; DPPC, Dipalmitoyl phosphatidylcholine; DP, Drug Product; DSPC,
Distearoyl phosphatidylcholine; DS, Drug Substance; DLS, Dynamic Light Scattering; ELP, Elastin-like Polypeptide; FDA, Food and Drug Administration; FDKP,
Fumaryl diketopiperazine; FTIR, Fourier Transform Infrared Spectroscopy; IgE, Immunoglobulin E; IgG, Immunoglobulin G; IR, Infrared; IVIG, Intravenous
Immunoglobulin; LDH, Lactate dehydrogenase; mAb, Monoclonal Antibody; MS, Mass Spectrometry; PAT, Process Analytical Technology; PFPE, Peruoropolyether;
PCA, Polycyano acrylate, PVA, Polyvinyl alcohol; QbD, Quality by Design; QC, Quality Control; rhGH, recombinant Growth Hormone; RMC, Residual Moisture
Content; RTD, Resistance Temperature Detector; SEC, Size Exclusion Chromatography; ssHDX-MS, solid-state Hydrogen Deuterium Exchange Mass Spectrometry;
siRNA, small interfering Ribonucleic Acid; SSA, Specic Surface Area; ssPL, solid-state Photolytic Labelling; SRCD, Synchrotron Radiation Circular Dichroism
Spectroscopy; TOSAP, Temperature-controlled organic assisted precipitation.
* Corresponding author.
E-mail address: ashutosh.sharma@postgrad.wit.ie (A. Sharma).
Contents lists available at ScienceDirect
International Journal of Pharmaceutics
journal homepage: www.elsevier.com/locate/ijpharm
https://doi.org/10.1016/j.ijpharm.2021.121115
Received 13 April 2021; Received in revised form 24 August 2021; Accepted 15 September 2021
International Journal of Pharmaceutics 609 (2021) 121115
2
sensitive and labile biopharmaceuticals along with a reduced require-
ment for cold chain (Carpenter and Chang, 1996; Tang and Pikal, 2004).
Chemical decomposition is reduced at very low temperatures during the
process, with low residual moisture content (RMC) post-processing,
keeping the product stable (Matejtschuk, 2007). However, several
drawbacks associated with conventional Freeze-drying have been re-
ported. Firstly, processing times could vary from days to weeks (Nail and
Gatin, 1993; Tang and Pikal, 2004) with high capital costs (Stratta et al.,
2020) and high energy consumption leading to reduced efciency in the
overall process (Bando et al., 2017; Liapis and Bruttini, 2008; Liu et al.,
2008). Secondly, being a static process, it increases the risk of rejection
of the entire batch in case of process failure. A batch process is limited to
processing small volumes of unit doses which increases the requirement
for a large number of vials. Human intervention during the process can
increase the risk of contamination despite current Good Manufacturing
Practices (cGMP) (FDA, 2014a). Thirdly, concerns related to heat and
mass transfer and scaling up the process have been discussed which
depend on the dryer design (Barresi et al., 2010), container closure, load
condition and product formulation etc. (FDA, 2014a; Patel and Pikal,
2011). Vials situated on the edge of the shelves dry faster than vials
present in the center due to radiation from the chamber walls which
causes issues of vial to vial heterogeneity (Rambhatla and Pikal, 2003).
Furthermore, certain high concentration Freeze-dried protein cakes can
take very long to reconstitute compared to free-owing powders pro-
duced by alternative drying techniques (Cao et al., 2013; Dani et al.,
2007).
Though having not gained widespread acceptance in the biophar-
maceutical industry, some potential, continuous drying technologies
open avenues for research to combat challenges associated with con-
ventional Freeze-drying technology. Readers are referred to the cited
reviews herein for detailed information on other continuous Freeze-
drying technologies (Adali et al., 2020; Ishwarya et al., 2015; Pisano,
2020; Pisano et al., 2019). This review encompasses some innovative
drying technologies, namely, Spin-freeze-drying, continuous Freeze-
drying of suspended vials, Active-freeze-drying, Spray-freezing and
Dynamic Freeze-drying, Lynnity® Technology, Spray-drying, PRINT®
Technology and Microglassication
TM
for biopharmaceuticals along
with some considerations for the selection of a drying process. A variety
of biopharmaceutical characterization techniques (section 3) and PAT
(section 5) in tandem with drying processes can provide an in-depth
assessment of product CQAs and process CPPs, ensuring safety, quality
and efcacy of products before being delivered to patients. Moreover,
formulation components play an important role in conferring biophar-
maceutical stability and the rationale of choosing excipients specic to
the drying process has been discussed in section 4. In addition to the
above considerations, some of the alternative drying technologies can
reduce complexities associated with validation processes and also allow
successful scale-up with a Quality by Design (QbD) approach (section 6).
Some alternative drying technologies have been summarized in Table 1.
While some of the technologies are categorized as bulk drying tech-
nologies, these technologies can also be employed to produce single unit
doses via an integrated powder/product lling system (Dalton Pharma,
2021; Nova Laboratories, 2015).
2. Biopharmaceutical Drying Technologies
Some of the single dose and bulk drying technologies have been
discussed in sub-section 2.1 and sub-section 2.2, respectively.
2.1. Single Dose Drying Technologies
2.1.1. Conventional Batch Freeze-drying
A batch Freeze-drying process involves the removal of solvent,
typically water, from a solution based on the principle of sublimation. A
typical batch Freeze-dryer consists of a drying chamber with multiple
shelves, a condenser and a vacuum pump. The process of Freeze-drying
consists of 3 major steps: freezing, primary drying and secondary drying.
Fig. 1 depicts a laboratory-scale Freeze-drying cycle. Vials containing
the desired volume of liquid product are partially stoppered and loaded
into the drying chamber. Freezing of the product is carried out at very
low temperatures ranging between 40 to 60 C to ensure solidica-
tion below the eutectic point (T
m
) of crystalline components or below
the glass transition temperature of the frozen product (T
g
) of amor-
phous materials (Pisano, 2019). The average product temperature, in a
non-cGMP Freeze-drying cycle, can be monitored using thermocouples
and other wireless temperature probes (Wang and McCoy, 2015).
Annealing is an optional, additional step performed to crystallize bulk-
ing agents and to improve product homogeneity (Al-Hussein and Gies-
eler, 2012; Pisano, 2019). Post freezing, primary drying involves the
transformation of frozen water into water vapor below its triple point.
Partially stoppered vials allow the migration of water vapor from the
vials to the condenser. Increased shelf temperature at a lower chamber
pressure is ideal for sublimation, however, the product temperature is
maintained below its T
g
or T
m
(Tang and Pikal, 2004). The nal step,
secondary drying, is also known as the desorption phase. This phase is
carried out to reach the optimum residual moisture level (Wang et al.,
2015). While the bound water molecules are removed at much higher
temperatures, the product temperature during the secondary drying
phase is maintained below its solid-state glass transition temperature
(T
g
). Post secondary drying, all vials are typically backlled with sterile
nitrogen gas and stoppered and capped with silicone stoppers and
aluminium cover seals. All vials undergo inspection before they are
nally released and shipped.
2.1.2. Spin-freeze-drying of Unit doses
The method of Spin-freezing was rst patented by Becker in 1957,
patent no. DE967120 (Becker, 1957). This method was then employed
for Freeze-drying and patented by Broadwin in 1965, patent no.
US3203108A and Oughton et al. in 1999, patent no. US5964043
(Broadwin, 1965; Oughton et al., 1999). With modications to the
patents and to the conventional Freeze-drying process, authors invented
a novel continuous Freeze-drying process for unit doses (Corver, 2012;
De Meyer et al., 2015). A major characteristic of this continuous
Freeze-drying technology is the rotation of vials containing the liquid
product of interest along their longitudinal axis, therefore, this is known
as Spin-freeze-drying (Fig. 2).
The process begins with the continuous Spin-freezing step wherein
vials containing the liquid product are spun rapidly, typically at 2500
3000 rpm, along their longitudinal axis for a period of time. The axial
rotating motion results in the formation of a dispersion layer on the
inner walls of the vial with a relatively uniform thickness of 1 mm
(Corver, 2012). Subsequently, the rotating vials are exposed to a ow of
sterile cryogenic gas, such as nitrogen or carbon dioxide, which is
temperature controlled. As the frozen product is spread all over the inner
walls of the vial, this results in a large surface area, thereby, allowing
fast and homogenous freezing/heating of the dispersion layer (De Meyer
et al., 2015). The process of solidication takes about 1 2 min and the
product is typically subjected to a temperature between 40 C and
60 C for another 10 20 min (Corver, 2012). To achieve crystallization
and the desired morphology of the excipients, further modications are
made to the cooling process conditions in a temperature-controlled
chamber. Following the cooling step, the vials are transferred to the
primary drying chamber through a conveyor belt system. Each vial is
held in a heat conducting jacket or a pocket in the chamber with the
desired pressure and temperature conditions. The jacket surrounds the
outer surface of the vials to facilitate homogenous distribution of heat
through conduction or radiation. Subsequently, the vials are transferred
to the secondary drying chamber for the desorption of residual water.
Fig. 3 depicts a schematic for the continuous drying system. The drying
step lasts for about 30 min to 2 h (Corver, 2012). It was reported that the
total processing time is reduced by 10 40 times depending on the vial
dimensions and the product formulation (De Meyer et al., 2015).
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
3
Table 1
Potential drying technologies for biopharmaceutical application.
Drying Technology Key Features Molecule /
Biomolecule
Processed
Batch /
Continuous
processing
Product
Yield
Achievable
RMC (w/w)
Dry Product
Characteristics
Potential Process-
induced Stress
Aseptic
Equipment
Manufacturer
References
Single Dose Drying Technologies
Batch
Freeze-drying
Gentle drying,
improved shelf-
life, reduced cold
chain
requirement.
A wide range of
enzymes,
antibodies,
hormones,
vaccines, gene
therapy products
etc.
Batch. 98 % 1 %
Intact, porous
cake.
Cake surface
area ~ 300
400 mm
2
.
Cold
denaturation.
Ice-liquid
interfacial
denaturation.
SP Scientic,
IMA Life, GEA
Lyophil,
Optima
Pharma,
MillRock
Technology.
(De Beer et al.,
2009; Ganguly
et al., 2018;
Gervasi et al.,
2018;
Harguindeguy
and Fissore, 2021;
Nail et al., 2017;
Tang and Pikal,
2004).
Spin-freeze-drying
Spinning on
longitudinal axis,
large surface area
and heat/mass
transfer, no shear
stress.
Alcohol
dehydrogenase,
IVIG.
Continuous. Not
available. 1 %
Intact dry
layer along the
inner surface of
vials.
Dry layer
surface area ~
2500 mm
2
.
Cold
denaturation.
Ice-liquid
interfacial
denaturation.
Prototype
available by
RheaVita.
(Corver, 2012; De
Meyer et al.,
2015; Lammens
et al., 2018;
Vanbillemont
et al., 2020a).
Continuous Freeze-
drying of
suspended vials
Controlled
nucleation via
VISF and
homogeneous heat
transfer,
continuous ow of
vials.
Aqueous solutions
of sucrose and
mannitol.
Continuous. Not
available. ~ 1 % Intact, porous
cake.
Cold
denaturation.
Ice-liquid
interfacial
denaturation.
Not available. (Capozzi et al.,
2019).
Foam drying in vials
Rapid
evaporation/
boiling at low
vapour pressure
and ambient
temperature, no
freezing required,
lower energy
consumption.
Vaccines, rhumAb,
bacteria. Batch. Not
available. >1 3 %
Dry foam
structure.
Lower specic
surface and
lower water
desorption rate.
Stress due to
surface tension
and cavitation.
Not available.
(Abdul-Fattah
et al., 2007;
Ohtake et al.,
2011a, 2011b)
Microwave vacuum
drying
Rapid
dehydration,
reduction in
freeze-drying
cycle time by 80
%, comparable
product
appearance and
enzyme activity.
Haemoglobin,
catalase, live virus
vaccine.
Semi-
continuous.
Not
available. ~ 1.8 %
Dried cake
with an average
SSA of 1.52 m
2
/
g comparable to
the SSA of
freeze-dried
(1.61 m
2
/g)
counterpart.
Cold
denaturation
during ash
freezing.
Thermal
denaturation due
to plasma
discharge at high
electromagnetic
eld intensity.
Heterogenous
heating and
arcing may cause
burning of
product.
EnWave.
(Bhambhani
et al., 2021;
Durance et al.,
2020; EnWave,
2021).
Bulk Drying Technologies
Active-freeze-drying
Bulk processing of
product, free
owing powder,
small particle size,
improved
properties with
continuous stirring
for certain
products.
Ketoconazole. Batch. 85 94 % <0.5 %
Free-owing
powder.
Particle size: 1
100 µm.
Stress due to
continuous
stirring by
impeller.
Ice-liquid
interfacial
denaturation.
Cold
denaturation.
Hosokawa
Micron B.V.
(Hosokawa
Micron, 2019;
Touzet et al.,
2018; Van Der
Wel, 2012).
Spray-drying
Continuous and
rapid drying,
particle
engineering, free-
owing powder,
low energy, and
equipment cost.
Raplixa®,
Exubera®,
Lysozyme, BSA,
mAbs, siRNA etc.
and some other
biologics listed in
Table 2.
Continuous. >50 95
% 1 2 %
Free-owing
powder.
Particle size:
300 nm 100
µm.
Shear due to
atomization.
Air-liquid
interfacial
denaturation.
Thermal
denaturation due
to residence time
at high outlet
temperatures.
SPX Flow
Technologies -
Anhydro, GEA
Niro, Fluid Air,
Ohkawara
Kakohi Co.
Ltd.
(Bowen et al.,
2013; FDA,
2015a; Silva
et al., 2013;
Uddin et al.,
2021; Vehring,
2008; Vehring
et al., 2020;
Walters et al.,
2014; Wang
et al., 2018;
White et al.,
(continued on next page)
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
4
De Meyer et al. showed that the freezing method signicantly
contributed to the sublimation kinetics of the Spin-frozen vials
compared to static batch freezing (De Meyer et al., 2015). This was due
to the fact that the presence of a thinner product layer and large surface
area in the inner walls of the vials contributed to much higher subli-
mation rates. Amongst the other factors studied, increasing the shelf
temperature and the chamber pressure increased the sublimation rates,
however, the freezing rate and the ll volume did not have a signicant
impact on the sublimation rates post 2 h of Spin-freeze-drying and batch
Freeze-drying (De Meyer et al., 2015). Furthermore, it was shown that
the shelf temperature and chamber pressure were observed to have
similar effects on the sublimation rates of Spin-frozen vials post 2 h of
Table 1 (continued )
Drying Technology Key Features Molecule /
Biomolecule
Processed
Batch /
Continuous
processing
Product
Yield
Achievable
RMC (w/w)
Dry Product
Characteristics
Potential Process-
induced Stress
Aseptic
Equipment
Manufacturer
References
2005; Wu et al.,
2019; Ziaee et al.,
2020).
Spray-freezing and
Dynamic Freeze-
drying
Uses frequency
driven prilling
nozzle. Primary,
secondary drying
occurs in rotary
freeze drying
chamber.
mAb. Continuous. >97 % <1 %
Free-owing
powder.
Particle size: <
300 µm 1000
µm.
Shear due to
frequency nozzle
and atomization.
Cold
denaturation.
Meridion
Technologies.
(Lowe et al.,
2018; Luy et al.,
2018; Struschka
et al., 2016).
Lynnity®
Technology
Uses piezoelectric
spray nozzle to
produce dried
spheres. Primary,
secondary drying
occurs on
cascading
vibratory shelves.
Not available. Continuous. Not
available. 1 %
Free-owing
powder.
Particle size: ~
600 µm.
Shear due to
piezoelectric
nozzle and
atomization.
Cold
denaturation.
IMA Life.
(IMA Life, 2019;
DeMarco and
Renzi, 2015).
PRINT®
Tunable shape,
size, and
morphology of
nano and
microparticles,
enhanced surface
properties and API
bioavailability,
large-scale
production is
possible.
Lysozyme, BSA,
DNase, IgG, siRNA,
ribavirin, vaccines.
Continuous. Not
available.
Not
available.
Free-owing
powder.
Particle size: 2
200 µm.
Customizable
particle shape
and
morphology.
Cold
denaturation.
Roller pressure
may induce shear
stress.
Liquidia
Corporation.
(DeSimone, 2016;
Galloway et al.,
2013; Garcia
et al., 2012; Kelly
and DeSimone,
2008; Wilson
et al., 2018; Xu
et al., 2013).
Microglassication
TM
Controlled particle
size, morphology,
release and
dissolution, no
excipients used.
BSA, lysozyme,
α
-chymotrypsin,
catalase,
horseradish
peroxidase, ELP.
Batch. Not
available.
Assumed
comparable
to Freeze-
drying by the
authors.
Particle beads.
Particle size: 1
- >10 µm.
Not available. Not available.
(Aniket et al.,
2015a, 2015b,
2014).
Electrospinning
Gentle, rapid
drying (~0.1 s)
and reconstitution
due to high surface
area. Ultra-ne
particles by
electrical
atomization.
Requires viscous
solutions.
Comparable
product stability.
Iniximab, zein,
siRNA, inulin,
β-galactosidase,
lysozyme.
Continuous. ~ 80 % ~ 6.5 % Dry particles
size: <10 µm.
Shear due to
electrical
atomization.
Bioinicia.
(Abraham et al.,
2019b; Domj´
an
et al., 2020; Jain
et al., 2014;
Karthikeyan
et al., 2015;
Wagner et al.,
2015).
Supercritical Fluid
drying
Rapid drying,
drying can occur
via Spray-drying
in the presence of
supercritical uid
(e.g., CO
2
) at low
temperatures
(>32C) and
pressure (~103
Bar) or via
supercritical
antisolvent
precipitation.
Particle
engineering
possible, no
freezing required.
Albumin, alkaline
phosphatase,
catalase,
chymotrypsin,
insulin, lactase,
rhDNase, trypsin,
urease, lysozyme,
myoglobin, IgG,
LDH.
Continuous. Not
available. 1 %
Free-owing
powder.
Micron-sized
particles (200
nm 50 µm).
Shear due to
atomization.
Fluid-uid
interfacial
denaturation.
High pressure
denaturation.
Extratex,
Natex,
Separex.
(Bouchard et al.,
2007; Jovanovi´
c
et al., 2008a,
2008b, 2004;
Long et al., 2019;
Sellers et al.,
2001).
API, Active Pharmaceutical Ingredient; BSA, Bovine Serum Albumin; DNase, Deoxyribonuclease; ELP, Elastin-like Polypeptide; IgG, Immunoglobulin G; LDH, Lactate
dehydrogenase; mAb, monoclonal antibody; rhDNase, recombinant human Deoxyribonuclease; siRNA small interfering Ribonucleic Acid; SSA, Specic Surface Area.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
5
primary drying but shelf temperature was reported to be the most sig-
nicant factor on the rate of sublimation of Spin-frozen vials when there
was an adequate contact between the vial and the shelf. The effect of
chamber pressure, compared to the shelf temperature, was much lower
for the batch frozen vials. This was again due to the fact that that the
product surface area on the inner walls of the vial was much greater for
Spin-frozen vials (2533 mm
2
) than the traditionally frozen vials (373
mm
2
) (De Meyer et al., 2015).
Through another study, the effects of shear, sedimentation and
diffusion velocity were studied on the stability of Alcohol dehydroge-
nase (ADH) during Spin-freeze-drying (Lammens et al., 2018). The
calculated shear rates at 2900 rpm, 800 rpm and 400 rpm for 2 4 min
were 2145 s
1
, 591 s
1
and 295 s
1
, respectively. These values were
comparatively lower than the shear rates (4000 s
1
20,000 s
1
)
generated during some processes such as cross-ow ltration and lobe
pumping (Bee et al., 2009; Gomme et al., 2006; GE, 2014), indicating
that Spin-freezing would not negatively impact ADH. It is worthwhile
understanding that the liquid in the vial experiences maximum shear
when the vial is accelerated from rest. As the vial attains the maximum
desired rotational velocity, the relative rotational velocity of the vial
with respect to the liquid reduces, thereby, reducing the shear rate. This
is analogous to a person sitting in a moving aircraft experiences negli-
gible force with respect to the aircraft. However, further product-
specic evaluation is required to study the effect of Spin-freezing on
labile biopharmaceuticals. Moreover, no signicant loss in the activity
of ADH and the absence of permanent aggregates in the Dynamic Light
Scattering (DLS) results conrmed that the shear rate experienced dur-
ing Spin-freezing did not affect the stability of ADH. Furthermore, in-
homogeneity associated with sedimentation velocity (6.59 ×10
-9
m s
1
)
and stress experienced due to diffusion velocity (diffusion coefcient =
6.10 ×10
-11
m
2
s
1
) during Spin-freezing were shown to have negligible
effects on proteins within 10 min of Spin-freezing (Lammens et al.,
2018). It was found that the sedimentation velocity of viruses was 5730
times higher and that of bacteria was up to 20,000 times higher
compared to the sedimentation velocity of proteins. This meant that
viruses and bacteria are more prone to inhomogeneity in the frozen
product layer due to sedimentation. More recently, the impact of Spin-
freeze-drying was studied on the stability of a commercial polyclonal
antibody, human intravenous immunoglobulin (IVIG), manufactured by
Baxter Healthcare Corporation (Vanbillemont et al., 2020a). The au-
thors concluded that the stability of the Spin-freeze-dried protein was
comparable to its conventionally Freeze-dried counterpart. Since low
shear rates and no major airliquid interfaces were generated, Spin-
freezing did not impact the stability of the protein. These results were
consistent with results shown previously (Lammens et al., 2018).
In terms of aseptic manufacturing, a GMP-like engineering prototype
for Spin-freeze-drying has been developed by RheaVita and Ghent
University (Corver et al., 2018). The authors propound that this tech-
nology can be scaled-up by adding 5 parallel lines within an area of 25
m
2
to produce 10,000 Spin freeze-dried vials per day. In comparison to
the throughput delivered by the Spin-freeze-drying prototype, a com-
mercial batch Freeze-dryer, within an area of 30 m
2
, can deliver 100,000
vials of a capacity of 2 mL over a 3-day cycle. In conclusion, Spin-freeze-
drying technology has shown to be a potential competitor to batch
Freeze-dying in terms of process associated stresses and PAT, though the
feasibility of implementing>5 parallel lines to generate a higher
throughput along with the associated costs in a cGMP environment
would be an interesting area of study.
2.1.3. Continuous Freeze-drying of suspended vials
More recently, a new concept of continuous Freeze-drying known as
continuous Freeze-drying of suspended vials has been developed based
on patent PCT no. WO2018204484 (Trout et al., 2018). As shown in
Fig. 4, the Freeze-drying setup consists of a sequence of modules for
different unit operations connected together to ensure a continuous ow
of vials. The vials are suspended over a track with multiple rows that
Fig. 1. Schematic evolution of a Freeze-drying cycle. Adapted from (LyophilizationWorld, 2020).
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
6
allow the transfer of vials through chambers with different temperature
and pressure conditions. Each chamber is separated by a load-lock sys-
tem that facilitates the transfer of vials between different modules. As
the vials are continuously being loaded into the Freeze-dryer, they are
lled and partially stoppered. After lling, the vials enter the freezing
zone where freezing occurs either through spontaneous or controlled
nucleation using Vacuum Induced Surface Freezing technique (VISF).
Heat transfer during freezing is achieved by cooling gas through forced
air convection or radiation. A signicant reduction in temperature
gradients within the product results in the formation of larger and uni-
form pores which is achieved by the suspended-vial conguration. This
further results in a faster sublimation rate, thereby, reducing the total
drying time compared to conventional Freeze-drying. In addition,
controlled nucleation reduces vial-to-vial ice crystal inhomogeneity
which results in a homogenous porous structure amongst vials.
Following the process of freezing, the vials subsequently move into the
primary and secondary drying chamber. The desired pressure is applied
to initiate sublimation. Controlled heat transfer is achieved through a
circulating heat transfer uid in the radiating surfaces. This feature
helps to overcome the drawback of heterogeneous heat and mass
transfer during conventional Freeze-drying. The authors asserted that
continuous Freeze-drying would require a cycle time of only about 6 h
compared to a 51 h hour batch Freeze-drying cycle (Pisano et al., 2019).
Moreover, the size of this continuous Freeze-drier would be 6 8 times
smaller compared to a conventional batch Freeze-dryer (Capozzi et al.,
2019).
In conclusion, this concept of continuous Freeze-drying offers a
greater advantage in terms of reduced drying times, continuous
throughput and PAT for all individual vials, however, the feasibility of
setting up this continuous process in a cGMP environment with associ-
ated scale-up and validation processes need to be addressed.
2.2. Bulk Drying Technologies
2.2.1. Active-freeze-drying
In contrast to tray-based bulk Freeze-drying, Hosowaka Micron B.V.
developed stirred bulk Freeze-drying known as Active-freeze-drying
based on patent no. EP1601919A2 (Van Der Wel, 2012). The Active-
freeze-drying process allows drying of heat-sensitive bulk materials
ranging from solutions, suspensions and pastes to wet solids with min-
imal handling (Hosokawa Micron, 2019). The nal dried product is
obtained as free-owing powder, unlike Freeze-dried cakes. As an
additional feature, the characteristics of certain products can be
improved by stirring. Moreover, a higher rate of heat transfer and
reduced drying times can be achieved. The process ow includes a jac-
keted conical vacuum dryer, an impeller, a collection lter, a product
collector, a condenser and a vacuum pump as shown in Fig. 5.
The working principle involves dynamic freezing of the product in a
conical stirred drying chamber with the help of a freezing medium. The
chamber is surrounded with a controlled heating/cooling jacket. Frozen
granules of different sizes and shapes are obtained as a result of VISF and
stirring motion. Subsequently, sublimation takes place at a suitable
pressure and heat is distributed through the jacket along with the stir-
ring motion. Mixing provides a large surface area for sublimation
leading to a higher rate of heat transfer, thereby, reducing the drying
time (Touzet et al., 2018). Sublimation starts at the outer layers even-
tually moving towards the inner layers of the frozen granules. Due to the
stirring motion, the dried layer is continuously disintegrated into
Fig. 2. (a) Spin-freezing of a vial along its longitudinal axis and (b) Spin-freeze-dried vials. Reprinted from (De Meyer et al., 2017) with permission from Elsevier.
Fig. 3. A schematic of the continuous Spin-freeze-drying System.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
7
fragments which reduces resistance to vapour ow. These fragments are
driven into the collector by vacuum. The drying process is nished when
the product temperature is in equilibrium with the chamber wall
temperature.
A pilot-scale study conducted on Active-freeze-drying of nanocrystal-
based ketoconazole drug showed that the technique efciently produced
reconstitutable nanocrystal powder (Touzet et al., 2018). Out of the
process parameters evaluated, namely, freezing method, nanocrystal
concentration, jacket temperature and screw rotation, the jacket tem-
perature signicantly contributed to the rate of sublimation and the
yield. An increase in the jacket temperature approximately doubled the
rate of sublimation and resulted in the collection of signicantly larger
fragments due to the corresponding increase in the vapour ow rate and
allowing the transfer of large fragments from the chamber into the
collector, thereby, leading to a higher yield. Large fragments, as shown
by Scanning Electron Microscopy (SEM) (Fig. 6), were also obtained by
Fig. 4. Continuous Freeze-drying of Suspended Vials. Reprinted from (Capozzi et al., 2019) with permission from ACS.
Fig. 5. Pictorial representation of the Active-freeze-drying Process.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
8
increasing the nanocrystal concentration as a result of a stronger
nanocrystal network structure. Furthermore, particle size analyses of the
reconstituted Active-freeze-dried nanocrystal suspensions showed the
presence of nanocrystal aggregates between 1 and 100 µm along with
reduced presence of the original population by volume. The presence of
aggregates induced by Active-freeze-drying, reduced the surface area
available for rehydration by 45 50 %. A signicant reduction in the
fraction of aggregates in the formulation was observed by increasing the
concentration of D-
α
-tocopherol polyethylene glycol 100 succinate
(TPGS) by three-fold. This resulted in an increased availability of the
surface for re-dispersion, however, this did not reduce the reconstitution
time of the dried nanocrystals. Lastly, they also demonstrated that
lowering the freezing temperature from 19 C to 33 C led to reduced
aggregation. It was stated that no surface loss was observed over the
period of 2 weeks of storage post Active-freeze-drying and the nano-
crystal suspensions were physically stable (Touzet et al., 2018).
While this technology employs the principle of Freeze-drying to pro-
duce bulk powder, it is still a batch process. Continuous operation and
automated recovery of the dried product is yet to be addressed. Moreover,
the continuous stirring motion throughout the freezing and drying pro-
cess can be detrimental for the stability of proteins. Continuous stirring
can lead to increased foaming in the liquid product, thereby, exposing the
protein to interfacial denaturation and aggregation (Duerkop et al.,
2018). During the freezing step, breaking up of ice crystals by continuous
stirring can interfere with ice nucleation and the cooling rate which may
lead to instability at the ice-liquid interface (Authelin et al., 2020).
Furthermore, continuous stirring in an insulted vessel can lead to a rela-
tive increase in temperature and so, it is crucial to accurately monitor the
product temperature during sublimation. Sublimation above the T
g
may
lead to product collapse (Meister et al., 2009; Meister and Gieseler, 2009;
Ohori and Yamashita, 2017). Active-freeze-drying technology may be
more suited for small molecules and stable biopharmaceuticals, though
the feasibility of this technique needs investigation for commercially
manufactured parenteral biopharmaceuticals.
2.2.2. Spray-freezing and Dynamic Freeze-drying Technology by Meridion
Technologies
Meridion Technologies developed the SprayCon Lab® Spray-freeze-
dryer based on two patents by Sano Pasteur SA, patent no.
US10006706B2 (Luy et al., 2018) and US9347707B2 (Struschka et al.,
2016). A schematic of the production-scale Spray-freeze-drying process
and Spray-freeze-dried microspheres generated by Meridion Technolo-
gies are shown in Fig. 7. The transfer liquid vessel and the Spray-freezing
chamber are positioned over the rotary Freeze-dryer connected through
a cooled tube and a ap that tightly separates both process areas. The
dried product is transferred from the drying chamber through a transfer
tube into lling vessels.
This process is divided into two broad steps, namely, Spray-freezing
(SprayCon® Technology) and dynamic bulk Freeze-drying (LyoMotion®
Technology). The spraying process employs a frequency-driven prilling
nozzle. The liquid feed disintegrates into round droplets at the resonance
frequency and are guided by gravity into the freezing chamber. The ow
rate, nozzle frequency, viscosity and orice diameter inuence the
droplet size. It has been reported that approximately 1000 5000
droplets/s can be generated (Luy and Stamato, 2020). The freezing
chamber is a double-walled, cylindrical vessel into which the droplets
are frozen using a cryogenic medium (liquid N
2
and gas N
2
). The
droplets are not directly in contact with N
2
(l). Sterile N
2
(g), at an
operating temperature in the range of 80 C to 150 C is lled inside
the chamber where heat exchange occurs by convention. Additionally,
deector jets are installed along the trajectory of the falling droplets that
facilitate droplet dispersion and prevent agglomeration (Sebasti˜
ao et al.,
2019a). Several parameters such as gas temperature, droplet size, glass
transition temperature, total solid content and the height of the freezing
chamber impact the rate of freezing of the droplets. Typically, uniform
frozen spheres of a size of 300 1000 µm can be generated in 1 3 s of
travelling 1.5 m 3.5 m (Luy and Stamato, 2020).
Post Spray-freezing, the dynamic Freeze-drying process involves a
rotary Freeze-dryer made of a cylindrical drum located inside a double-
walled vacuum drum. For sterile operation, the freezing chamber and
rotary Freeze-dryer are connected through an isolation valve. The rate of
sublimation is increased as a result of the constant rotary motion of the
drum along its longitudinal axis and increase in temperature. Conduc-
tive heat transfer to the product is achieved by silicon oil that ows
through the double-walled surface and by infrared (IR) radiators
installed inside the drum and the pressure is set to 100 µbar or lower.
Moreover, the cross-sectional area for the ow of water vapour is
signicantly increased through the opening located at either ends of the
drum. By reverse rotation of the drum, the dried bulk product spheres
are taken up by discharge scoops that are directed downwards through a
funnel for nal collection or lling (Luy and Stamato, 2020).
Sebasti˜
ao et al. described mathematical models, including a derived
set of equations, for droplet kinetics (Sebasti˜
ao et al., 2021, 2019a,
2019b). These models could predict the temperature of a single sprayed
droplet at different positions in the freezing column in terms of mass,
velocity, sublimation rate etc. They found that the predicted droplet
temperatures were consistent with their experimental data within an
error of 10 %. Readers are referred to the cited paper herein for details
on the model parameters and their relationships (Sebasti˜
ao et al.,
2019a). In a complimentary parametric study, the authors described the
impact of spraying conditions on the freezability of droplets in the
freezing column (Sebasti˜
ao et al., 2019b). They elucidated that the
droplet diameter and the cryogenic gas temperature signicantly
inuenced the freezing distance of droplets compared to the nozzle ow
rate, initial droplet temperature, solute concentration, and the super-
cooling degree. The initial droplet temperature and the volumetric ow
Fig. 6. (A) SEM image of 10 % w/w ketoconazole drug crystals (B) SEM image of 20 % w/w ketoconazole drug crystals. Reprinted from (Touzet et al., 2018) with
permission from Elsevier.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
9
rate from the spray nozzle did not have a major impact on the freezing
distance. This meant that viscous solutions could be sprayed at higher
temperatures, thereby, preserving pumping efciency and performance.
Moreover, reduced freezing distances could be achieved by increasing
the initial solute concentration at higher droplet temperatures though
the impact of solute concentration was not signicant at lower tem-
peratures. Understanding the effect of these parameters is essential to
ensure proper freezing of droplets before entering the drying phase.
Though these results have not been generalised, they are crucial in
developing and optimizing a Spray-freeze-drying equipment and
process.
Reported benets of this technology are described here. The cryo-
genic gas, the rotating drum surface as well as the IR radiators and the
large surface area of the frozen microspheres confer increased heat and
mass transfer, thereby, reducing the total drying time. Reduction in the
water vapour diffusion length is achieved in a Spray-frozen microsphere
with a diameter of 1 mm and a maximum diffusion length of 500 µm
compared to a 10 mm thick Freeze-dried cake with a maximum diffusion
length of 10000 µm (Luy and Stamato, 2020). While a recommended
total solid content of 5 10 % w/w in the product feed allows strong
spherical structures, drying of a monoclonal antibody (mAb) formula-
tions with mAb concentrations in the range of 50 200 mg/mL and a
total solid content of 10 36 % w/w were achieved in about 24 30 h
(Lowe et al., 2018). Moreover, free-owing powder, rapid reconstitution
and minimal increase in protein aggregation was achieved at 25 C, over
a 9-month storage period for the formulation containing a ratio of 1:1
(mAb : sugar) (Lowe et al., 2018). Even though particle size of <300 µm
is achievable during Spray-freezing, the presence of high solid content
reduces the risk of loss of particles due to a high rate of sublimation. The
risk of particle loss for product feed with low solid content can be
reduced by generating larger particle sizes i.e., 2 3 mm. It has been
reported that the time required for Spray-freezing a 100 L bulk with 20
% solid content is ~ 10 to 20 h and the time required for dynamic
Freeze-drying is ~ 24 h with >97 % yield (Luy and Stamato, 2020).
2.2.3. Continuous Aseptic Spray-freeze-drying Technology by IMA Life
IMA Life America INC. invented and patented a bulk Freeze-drying
process design using a combination of Spray-freezing and Stirred Dry-
ing, patent no. US9052138B2 (DeMarco and Renzi, 2015). The process
ow begins with the freezing step involving spraying of the bulk product
along with an aseptic freezing medium into an aseptic freezing vessel.
This is followed by introduction of a vacuum to the frozen powder to
initiate sublimation. The frozen material is stirred using a spiral blade
agitator at a low speed. Subsequently, the frozen powder is heated to
increase the rate of sublimation. Lastly, the vacuum is released to obtain
the nal Freeze-dried product.
With modications to the patent, IMA Life developed the Lynnity®
Spray-freeze-dryer (IMA Life, 2019). A schematic diagram of the Lyn-
nity® process and microparticles generated by the process are depicted
in Fig. 8. The spraying process begins with the generation of uniform
droplets under the inuence of frequency vibrations as the product feed
is made to ow through a temperature-controlled droplet zone. The
disintegrated product feed passes through the nozzle into the freezing
column where the freezing process begins. The stainless-steel freezing
Fig. 7. (a) A schematic diagram of the Spray-freeze-drying process by Meridion Technologies and (b) Spray-freeze-dried microspheres. Reprinted (adapted) from
(Luy and Stamato, 2020) with permission from John Wiley and Sons.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
10
column is lined with a double walled jacket that utilizes liquid nitrogen
and silicone oil for controlling the temperature in the chamber. The
cooling gas is maintained at temperatures below 130 C. As the
droplets are sprayed into the column, they are instantaneously (in<5
feet from the point of ejection with an average volumetric diameter of
500 µm) frozen allowing them to maintain their shape (IMA Life, 2019).
The frozen spheres are collected at the base of the freezing chamber.
Before entering the drying module, an intermediate chamber allows the
product to be transferred from the freezing column at atmospheric
pressure to the drying chamber under vacuum conditions without dis-
rupting the continuous process. The movement of frozen spheres inside
the drying chamber is done at a controlled rate on cascading shelf stacks
using gentle vibratory agitation. Agitation through vibration along with
heat through heat transfer uid initiate rapid sublimation and prevent
agglomeration of the spheres inside the drying chamber. Post drying, the
dried spheres are collected as bulk in the collection chamber. Unlike a
conventional Freeze-dryer, the Lynnity® contains dual ice condensers
allowing continuous operation.
The Lynnity® production-scale Spray-freeze-dryer is show in Fig. 9.
Anticipated benets of this technique over conventional Freeze-drying
include bulk processing with minimal handling of trays without the
need for post-processing operations such as granulation and milling
(Siow et al., 2018) along with higher productivity and lower downtime
(IMA Life, 2019). Secondly, it confers increased efciency of heat and
mass transfer between the product, trays and shelves. Thirdly, it may
establish a continuous process in a sterile environment with greater
throughput exibility (DeMarco and Renzi, 2015). The specic surface
area of the dried spheres was measured using BrunauerEmmettTeller
(BET) theory and was found to be 7.03 m
2
/g, compared to that of Freeze-
dried cakes (0.47 m
2
/g), allowing greater interaction with solvent dur-
ing rehydration (IMA Life, 2019). Moreover, the dried spherical product
can be distributed into syringes, vials, inhalation systems etc. which is
not possible with Freeze-dried cakes. Despite the benets, a slight in-
crease in the turbidity of a rehydrated Spray-freeze-dried product,
compared to its Freeze-dried counterpart, was attributed to particle
agglomeration. They observed that increasing the concentration of
surfactant reduced turbidity due to aggregation.
In conclusion, continuous bulk processing can allow the use of more
PATs for real-time process and product monitoring, though further
consideration is required in terms of product yield, stability of Spray-
freeze-dried parenteral biopharmaceuticals, footprint associated with
the commercial-scale equipment and suitable powder lling options in a
cGMP environment.
2.2.4. Spray-drying
Spray-drying is a technique that has potential applications in the
biopharmaceutical industry. It is one of the few techniques used to
produce dried powder formulation from liquid, slurry or low-viscosity
paste (Celik and Wendell, 2010). Several advantages of Spray-drying
have been reported. Firstly, it eliminates the need for a large number
of unit operations which makes it cost-effective and improves produc-
tion efciency. The manufacturing cost associated with Spray-drying is
Fig. 8. (a) A schematic diagram of the LYnnity® Spray-freeze-drying Process and (b) Spray-freeze-dried microparticles generated by LYnnity® Technology (IMA
Life, 2019).
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
11
only 20 % of the manufacturing cost associated with Freeze-drying
(Roser, 1991; Santivarangkna et al., 2007). Secondly, it is a rapid one
step process, allows continuous production of owable powders and can
be implemented across a range of biopharmaceuticals (Walters et al.,
2014). Properties such as particle size and shape can be controlled and
engineered by Spray-drying (Vehring, 2008). Furthermore, this tech-
nique can take thermolabile products into consideration. The evapora-
tion process takes about milliseconds to a few seconds and the process is
very instantaneous, thereby, minimizing exposure to high inlet air
temperatures (Celik and Wendell, 2010). The production of millions of
small droplets provides a large surface area for heat and mass transfer
allowing rapid evaporation. SPX Flow Inc. has had aseptic Spray-dryers
that have been inspected by the FDA and produced clinical supplies for
phase 3 pivotal studies. The authors have demonstrated the use of
Anhydro MS-35 Spray-dryer to successfully produce dry powder-based
mAbs (Bowen et al., 2013; Gikanga et al., 2015). However, some of
the factors that may induce instability in proteins have been depicted in
Fig. 10.
Fig. 10 represents a schematic diagram of the Spray-drying process
using a cyclone-based separator. The liquid feed is drawn towards the
spray nozzle at a ow rate using a peristaltic pump. With the help of an
atomizing gas ow and desired size of spray nozzle orice, the liquid is
atomized and sprayed into the upper drying chamber. The temperature
of the sprayed droplets in the upper chamber is lower or equal to the wet
bulb temperature. The droplets experience evaporative cooling and are
pulled towards the lower chamber under the inuence of gravity and
drying gas. The outlet gas temperature is measured at the bottom of the
lower drying chamber. With the help of a cyclone gas ow rate, the
particles are guided towards the cyclone where particles are separated
based on their densities. The dry product is collected at the bottom as
dry powder and the low density ne particles can be recovered using a
bag lter (GEA, 2020). Possible sites for protein denaturation in the
upper drying chamber include shear at the nozzle, airliquid interfacial
denaturation and dehydration due to evaporative cooling. As the
particles exit the lower drying chamber, they experience temperatures
greater than the wet bulb temperature but lower or equal to the outlet
gas temperature. Outlet gas temperatures greater than the T
g
of the
product and the residence time of the particles in the cyclone collector
may affect the stability of proteins.
In comparison to Spray-drying, drug administration methods such as
nebulization and nasal spray expose proteins to airliquid interfaces,
shear and temperature, which may cause deterioration in the CQAs of
biopharmaceuticals (Albasarah et al., 2010; Bodier-Montagutelli et al.,
2020; Fr¨
ohlich and Salar-Behzadi, 2021; Hertel et al., 2015; Niven et al.,
1996, 1995). Examples of some commercial and clinical dry powder-
based biopharmaceuticals administered via the parenteral route, inha-
lation or nebulization are collated in Table 2. The stability and
composition of such products can provide more insight while studying
the impact of the atomization process during Spray-drying or Spray-
freeze-drying.
More recently, authors have demonstrated approaches to identify
CPPs and formulation components for Spray-drying proteins (Batens
et al., 2018; Grasmeijer et al., 2019; Ziaee et al., 2019). The outlet
temperature was found to be the most critical factor that affected the
enzymatic activity of lysozyme. Along with high outlet temperatures,
ultrasonic vibrations and mechanical stress produced from ultrasonic
nozzles had a negative impact on the activity of lysozyme (Ziaee et al.,
2020).
While the impact of Spray-drying on proteins has been studied in
terms of temperature, the effect of shear and interfacial denaturation
during atomization and spraying is also crucial as some proteins are
susceptible to such stresses (Broadhead et al., 1993; Grasmeijer et al.,
2019; Koshari et al., 2017; Maa et al., 1998; Maa and Hsu, 1997;
Mumenthaler et al., 1994; Wilson et al., 2019; Ziaee et al., 2020). Un-
derstanding the impact of spraying conditions prior to dehydration, can
provide more insight while developing and choosing excipients for
Spray-drying proteins. Typically, the liquid feed is drawn into a two-
uid nozzle at velocity,
ν
liq
and exits the nozzle tip with a diameter, d
i
Fig. 9. LYnnity® Production-Scale Spray-freeze-dryer (IMA Life, 2019).
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
12
(Fig. 11). A resultant velocity,
ν
av
is generated at the mixing zone with
the help of an atomizing gas ow rate,
ν
gas
. The shear rate generated
from a two-uid spray nozzle has been estimated using Equation (1)
(Ghandi et al., 2012; Hede et al., 2008).
γ=[2(
ν
av
ν
liq) ]
di
(1)
Where γ is the shear rate in s
1
,
ν
av
is the average velocity at mixing
point in m/s,
ν
liq
is the velocity of liquid in m/s and d
i
is the inner
diameter of the nozzle tip in mm.
ν
av
is a function of the mass ow rates
(kg/s) of the gas and liquid and must not be confused with the arithmetic
mean of their velocities.
Using Equation (1), authors have shown that the shear rates gener-
ated from a two-uid nozzle at different atomizing ow rates and liquid
densities can range from 97,000 s
1
to 992,000 s
1
(Ghandi et al., 2012;
Morgan et al., 2020). Shear-induced inactivation for some proteins can
occur at <2000 s
1
for 20 min (Ashton et al., 2009; Charm and Wong,
1981), while some proteins may remain stable up to a shear rate of >
250,000 s
1
for >30 min (Bee et al., 2009; Duerkop et al., 2018). Bekard
et al. elucidated that the
α
-helical content in poly-L-lysine was inversely
proportional to the square root of shear strain and the extent of
unfolding decreased with increasing molecular weight due to greater
cohesive forces (Bekard et al., 2011). Such shear levels can be experi-
enced during similar drying methods such as Spray-freeze-drying and so,
it is crucial to study the stability of biopharmaceuticals as a function of
shear and atomization.
In conclusion, Spray-drying is one of the most popular industrial
drying technology and has been studied over a wide range of products
over the past few decades. The challenges associated with the CPPs of
Spray-drying of labile parenteral biopharmaceuticals, choice of formu-
lation excipients and their molecular mechanism of interactions with
biopharmaceuticals during Spray-drying require further product-
specic study.
2.2.5. PRINT® Technology
Particle Replication in Non-Wetting Templates, also known as
PRINT® technology, originated from lithographic techniques applied in
the microelectronics and semiconductor industry. PRINT® is a micro-
moulding based particle design / engineering technology employed to
generate monodisperse, uniquely shaped (i.e. laments, rods, spheres,
discs, toroids) micro and nano-particles of hydrogels, polymers, APIs etc.
with tunable size and morphology (Galloway et al., 2013; Garcia et al.,
2012; Kelly and DeSimone, 2008). The fabrication of PRINT® particles
was rst demonstrated by Rolland et al. (Rolland et al., 2005) and the
pharmacokinetic characteristics of these particles as delivery vectors
were rst studied by Euliss et al. and Gratton et al. (Euliss et al., 2006;
Gratton et al., 2007). Kelly and DeSimone demonstrated the generation
of protein particles, namely, insulin and albumin using PRINT® tech-
nology (Kelly and DeSimone, 2008). With the incorporation of cGMP
practices, this technique has been scaled-up with continuous roll-to-roll
system which allows continuous particle production for pre-clinical and
clinical study of pharmaceutical inhalation powders by Liquidia Cor-
poration (DeSimone, 2016; Liquidia Corporation, 2021).
The process ow for generating PRINT® particles reported by au-
thors is discussed here (Garcia et al., 2012; Gratton et al., 2007; Kelly
and DeSimone, 2008). Peruoropolyether (PFPE) was poured onto a
prepared silicon master template containing the desired etching patterns
of 2 µm, 5 µm and 200 nm sized shapes to produce a mould containing
the same sized cavities (Fig. 12 (a)). Following the preparation of the
PFPE mould, aqueous protein samples containing insulin, albumin and
albumin mixtures with siRNA or paclitaxel were sandwiched between
the cavities present in the mould and a high surface energy polyethylene
lm (Fig. 12 (b)). A pressure of 50 psi was applied through a roller to
prevent the formation of layers between the lled cavities and to
Fig. 10. A schematic diagram of the Spray-drying process with a cyclone separator and possible sites for protein denaturation. Blue dotted lines represent the sprayed
droplets. The red dotted lines represent the trajectory of the particles.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
13
laminate the samples present between the mould and the lm (Fig. 12
(c)). Subsequently, the polyethylene lm was removed and the mould
containing the samples was Freeze-dried (Fig. 12 (d)). The dehydration
process can also occur through either photocuring, vitrication or
evaporation (Xu et al., 2013). A liquid harvesting layer, made of either
polycyano acrylate (PCA) or polyvinyl pyrrolidinone (PVP), was casted
onto a glass slide (Fig. 12 (e)). Post Freeze-drying, the PFPE mould was
placed over the adhesive harvesting lm (Fig. 12 (f)). Once the har-
vesting layer was dried, the PFPE mould was removed yielding dried
protein particles onto the adhesive lm (Fig. 12 (g)). Finally, free-
owing protein powder was recovered by dissolving the adhesive lm
(Fig. 12 (h and i)). SEM images of the uniquely shape powders are shown
in Fig. 13.
It was shown that no aggregation was observed in albumin and in-
sulin particles generated via PRINT® technology. Additionally, Pulmo-
zyme (DNase) and siRNA therapeutic molecules were processed using
PRINT® technology (Garcia et al., 2012). Minimal aggregation was
observed in the size exclusion chromatography (SEC) prole of DNase
PRINT® microparticles with comparable enzyme activity to native
DNase and the chemical structure of PRINT® generated siRNA particles
Table 2
Some commercial/clinical dry powder-based biopharmaceuticals and some biopharmaceuticals administered via spraying or nebulization.
Biopharmaceutical and
Manufacturer
Biomolecule API Formulation Excipients Manufacturing Process
(Administration Method)
Reference
Exubera® by Pzer /
Nektar (Discontinued). Insulin (Hormone) Sodium citrate dihydrate, mannitol, glycine,
sodium hydroxide.
Spray-dried powder.
(Inhaled for diabetes) (FDA, 2006; White et al., 2005)
Raplixa® by Probrix, The
Medicines Company
(Commercial).
Fibrin and Thrombin.
Trehalose, calcium chloride, human albumin,
sodium chloride, sodium citrate, L-arginine
hydrochloride.
Spray-dried powder.
(Powder applied on surface of
bleeding tissue for uncontrolled
bleeding)
(FDA, 2015a; Manufacturing
Chemist, 2015)
Afrezza® / Technosphere
insulin (TI) by Mannkind
(Commercial).
Recombinant Human
Insulin. FDKP, Polysorbate 80.
Technospheres® by
precipitation, adsorption and
Freeze-drying.
(Inhaled for diabetes)
(McElroy et al., 2013; Sarala
et al., 2012; Tsai-Turton, 2014)
Inbrija® by Acorda
Therapeutics
(Commercial).
Levodopa (aromatic amino
acid). DPPC and sodium chloride.
Arcus® Technology Spray-
dried powder.
(Inhaled for off episodes in
patients with Parkinsons
disease).
(Acorda Therapeutics, 2021;
FDA, 2018)
Somatuline® LA by Ipsen
(Commercial).
Lanreotide acetate
(octapeptide analogue of
somatostatin hormone).
PLGA, mannitol, carmellose sodium,
polysorbate 80.
Phase separation and spray-
dried microspheres.
(Powder and solvent for
prolonged-release suspension
for injection against multiple
conditions)
(EMA, 2013; HPRA, 2019; Pinto
et al., 2021)
Trelstar® LA by Verity
Pharmaceuticals
(Commercial).
Triptorelin pamoate
(synthetic decapeptide
analogue of GnRH
hormone).
PLGA, mannitol, carboxymethycellulose
sodium, polysorbate 80.
Phase separation and spray-
dried microspheres.
(Powder and solvent for
prolonged-release suspension
for injection for the treatment of
prostate cancer)
(Pinto et al., 2021; Vhora et al.,
2019)
Sandostatin® by Novartis
(Commercial).
Octreotide acetate (cyclic
octapeptide).
Mannitol, D,L-lactic and glycolic acids
copolymer, carboxymethylcellulose sodium.
Dry powder prepared by Phase
separation and Spray-drying.
(Injectable suspension for the
treatment of acromegaly etc.)
FDA, 2008; Hou et al., 2018;
Vhora et al., 2019)
TOBI® Podhalerby
Novartis
(Commercial).
Tobramycin (Antibacterial
aminoglycoside). DSPC, calcium chloride, and sulfuric acid.
PulmoSphereby Spray-
drying.
(Orally inhaled for cystic
brosis against Pseudomonas
aeruginosa).
(FDA, 2015b; Weers and Tarara,
2014)
Fludase® by Ansun
Biopharma (NexBio)
(Clinical Trial Phase 2).
DAS181 sialidase
(recombinant
neuraminidase).
Histidine, trehalose, citric acid, magnesium
sulphate, acetate buffer.
TOSAP.
(Dry powder for oral inhalation
against inuenza like illness)
(Bodier-Montagutelli et al.,
2018; ECRI, 2011; Mack et al.,
2012; Moss and Li, 2015)
CSJ117 by Novartis
(Clinical Trial Phase 2)
Anti-TSLP antibody
fragment. Leucine, trileucine, mannitol and trehalose.
PulmoSol
TM
engineered
powder.
(Dry powder inhaled for
asthma).
(Fr¨
ohlich and Salar-Behzadi,
2021; Liang et al., 2020;
NCT04410523, 2021)
Aerovant® by Aerovance /
Bayer
(Clinical Trial Phase 2).
Cytokine Pitrakinra.
Different formulations including sucrose,
mannitol or trehalose, leucine or poly (amino
acid) and citrate, acetate or lactate buffer
were evaluated.
Spray-dried powder.
(Inhaled for asthma)
(Bodier-Montagutelli et al.,
2018; Liang et al., 2020;
Otulana, 2011; Vehring et al.,
2020; Wenzel et al., 2007)
Abrezekimab (VR942) by
UCB Pharma, Vectura.
(Clinical Trial Phase 2).
CDP7766 (IL-13 mAb
fragment).
Trehalose dihydrate, L-leucine and
phosphate buffer.
Spray-dried powder.
(Inhaled for asthma)
(Burgess et al., 2018; Giles
Morgan et al., 2017; Liang et al.,
2020; Vectura, 2015; Vehring
et al., 2020)
Pulmozyme® by
Genentech / Roche
(Commercial).
Deoxyribonuclease I. Calcium chloride dihydrate, sodium chloride.
Liquid formulation.
(Inhaled through jet nebulizer
against cystic brosis to
improve pulmonary function).
(Bodier-Montagutelli et al.,
2018; FDA, 2014b)
Miacalcin® by Novartis /
Mylan
(Commercial).
Polypeptide hormone
(Calcitonin).
Sodium chloride, benzalkonium chloride,
hydrochloric acid.
Liquid formulation.
(Nasal spray for the treatment of
postmenopausal osteoporosis).
(FDA, 2017; Ozsoy et al., 2009)
DPPC, Dipalmitoyl phosphatidylcholine; DSPC, Distearoyl phosphatidylcholine; FDKP, Fumaryl diketopiperazine; GnRH, Gonadotropin releasing hormone; IL-13,
Interleukine-13, PLGA, Poly(lactic-co-glycolic acid); TOSAP, Temperature-controlled organic assisted precipitation.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
14
was preserved without any denaturation.
Furthermore, PRINT® technology was combined with a scalable
Spray-Assisted layer-by-layer (LbL) technique to enhance the charac-
teristics of the fabricated nanoparticles (Morton et al., 2013). The
combination of these techniques offer improved stability, sustained drug
release and improved physicochemical properties of nanoparticles
(Morton et al., 2013; Poon et al., 2011b, 2011a). In this case, the
polyvinyl alcohol (PVA) harvesting layer was crosslinked using 50 %
glutaraldehyde and 10 % hydrochloric acid to reduce its solubility in
water. This was done to prevent any loss of particles during the spraying
process. An aqueous solution of cationic polyelectrolyte was sprayed at a
concentration of 1 mg/mL onto the nanoparticles for 3 s. To remove
excess cationic polyelectrolyte, a wash step with water was included for
another 3 s. This was followed by another spray of anionic poly-
electrolyte at a concentration of 1 mg/mL for 3 s and a nal wash with
water. The sprayed-LbL particles were nally recovered by sonication,
0.45 µm ltration and ultracentrifugation. DLS analysis showed
consistent monodisperse particles with the polydispersity ranging be-
tween 0.01 and 0.1 and hydrodynamic diameter between 190 nm and
246 nm for uncoated and coated nanoparticles (Morton et al., 2013). A
decrease in the hydrodynamic diameter of coated particles was observed
due to contraction forces on the PVA layer. The shape and integrity of
the recovered particles was conrmed by electron microscopy images
wherein particles were entirely coated by polyelectrolytes. Moreover,
the biological functionality of these particles was retained and could be
ne-tuned by altering the lm thickness as per its application.
Furthermore, researchers developed cylindrical nanoparticles of
commercially available vaccines, such as Fluzone® (Sano-Pasteur),
Fluvirin® and AgriFlu® (Novartis) and Auria® (Merck), using this
technology (Galloway et al., 2013). PRINT® fabricated vaccine particles
showed 200-fold improved antigen binding along with enhanced im-
mune response. It was elucidated that the shape, size along with other
surface properties play an important role in the interaction of these
nanoparticles with other biomolecules in their surroundings (Galloway
et al., 2013). Xu et al. demonstrated the fabrication of transiently
insoluble BSA particles after being cross-linked by a disulphide-based
cross-linker using PRINT® with new opportunities for drug and gene
delivery (Xu et al., 2012). More recently, results of a phase 1 clinical trial
study reported for a PRINT® fabricated dry-powder ribavirin
Fig. 11. A schematic representation of external mixing in a two-uid nozzle.
Fig. 12. The process ow diagram of PRINT® Technology. Reprinted (adapted) from (Garcia et al., 2012; Hofmann et al., 2019; Kelly and DeSimone, 2008) with
permission from ACS and Hindawi Publishing Corporation.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
15
formulation by GSK showed improved physicochemical properties,
efcient and convenient delivery of API to the lungs (Dumont et al.,
2020).
Overall, this technology has shown promising results for some
inhaled proteins, gene therapy products and vaccines with the ability for
continuous production in a large-scale cGMP facility, though the sta-
bility of large parenteral mAbs and enzymes via PRINT® will be an
interesting area of study.
2.2.6. Microglassication
TM
In therapeutic protein formulations water substitution by excipients,
generally saccharides and polyalcohols, stabilize therapeutic proteins in
their dehydrated state (Allison et al., 1999; Liao et al., 2002; Mensink
et al., 2017). With modications to previously described micropipette
manipulation techniques in literature (Duncan and Needham, 2006,
2004; Rickard et al., 2010), Aniket et al. developed a new technique
called Microglassication
TM
for the preservation of proteins by dehy-
drating protein microdroplets in an immiscible drying solvent (Aniket
et al., 2014). The technique was performed at room temperature and
produced stable, excipient-free protein microglassied beads. The
technique was performed using two chambers; one containing BSA so-
lution and the other lled with an organic solvent (Aniket et al., 2014;
Su et al., 2010). A small plug of organic solvent was withdrawn into a
micropipette. The micropipette was then positioned into the chamber
containing BSA solution and the desired amount of protein was pulled
into it. The micropipette was then positioned back into the organic
chamber, releasing a single droplet of the protein solution which was
held rmly at the tip of the micropipette in the organic chamber. By
doing so, water from the single droplet was extracted into the organic
chamber, thereby, leading to the formation of a Microglassied
TM
bead
(Fig. 14).
The authors demonstrated this technique on BSA and showed that
the protein exhibited full recovery and restoration of its secondary
structure upon rehydration of the Microglassied
TM
beads (Aniket et al.,
2014). The uorescence spectra of native and rehydrated Micro-
glassied
TM
BSA were comparable, showing that the tertiary structure of
BSA was preserved. Moreover, the Microglassied
TM
beads were purely
amorphous which was shown through X-ray diffraction spectroscopy.
This was due to rapid dehydration of the microdroplet which did not
give enough time for crystallization. A signicantly lower percentage of
aggregates was reported in the rehydrated Microglassied
TM
(3.3 %)
sample compared to rehydrated Freeze-dried (10 %) sample. Further-
more, accelerated storage results for Microglassied
TM
BSA showed that
by increasing the saturation factor of water in alcohol from 0.4 to 0.5, a
signicant decrease in the soluble monomeric content of BSA was
observed just after the rst 10 min of storage at 65 C. This meant that by
increasing the availability of water to the Microglassied
TM
bead, an
increase in the formation of protein aggregates was observed for a total
of 60 min of storage at 65 C.
Through another study, this technique was performed on lysozyme,
α
-chymotrypsin, catalase and horseradish peroxidase (Aniket et al.,
2015a). Five different organic solvents, namely, n-pentanol, n-octanol,
n-decanol, triacetin and butyl lactate were used as an immiscible drying
solvent. It was reported that butyl lactate showed higher water solubility
compared to other solvents. Fourier transform infrared (FTIR) spectro-
scopic analyses for dried and rehydrated enzyme microspheres showed
that even though distortions in the secondary structural conformation of
Microglassied
TM
enzymes were observed, these changes reverted to
Fig. 13. SEM images of (a) 200 ×200 nm cylindrical BSA-Lactose, (b) 10 µm pollen IgG-Lactose, (c) 1.5 µm torus DNase and (d) 1.5 µm torus siRNA PRINT®
fabricated particles. Reprinted (adapted) from (Garcia et al., 2012) with permission from Hindawi Publishing Corporation.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
16
their native-like conformation upon reconstitution. Moreover, the
Microglassied
TM
enzymes exhibited signicantly higher activity in n-
pentanol compared to triacetin and butyl lactate. Accelerated storage
results for lysozyme showed a signicant reduction in enzyme activity of
the Microglassied
TM
and Freeze-dried samples after 1 month of storage
at 40 C. Bioactivities of both Microglassied
TM
and Freeze-dried
enzyme were comparable. Post 1 month and up to 3 months of stor-
age, minimal changes in the enzyme activity and secondary structure
content was observed. The authors assumed similar low levels of re-
sidual moisture, as both Microglassied
TM
and Freeze-dried lysozyme
absorbed the same amount of water as a function of water activity.
Furthermore, the potential to Microglassify
TM
a recombinant
biopolymer elastin-like polypeptide (ELP) with controlled size and
morphology for chemotherapy has been shown (Aniket et al., 2015b).
More recently, this technique has seen application in the fabrication of
biolasers for biosensing and optical device implantation purposes
(Nguyen et al., 2019; Nguyen and Ta, 2020). In summary, this technique
has demonstrated the potential as a novel drying technique on a wide
range of enzymes without the incorporation of any excipients at room
temperature. Further study is required to elucidate its application to
large-scale biopharmaceutical manufacturing along with the time
associated with the formation of Microglassied
TM
beads and their
reconstitution. Moreover, the evaluation of Microglassication
TM
on
commercial enzymes, mAbs, vaccines and the development of high
concentration biopharmaceutical parenteral formulations are areas that
can increase the scope of this technique.
2.3. Other Drying Technologies
In addition to some of the potential alternative drying technologies
described in this review, other drying techniques such as Microwave
drying, Foam drying, Vacuum drying, Supercritical Fluid drying, Elec-
trospinning, Fluidized bed drying, Hybrid drying etc. are gradually
gaining popularity as alternatives to Freeze-drying of bio-
pharmaceuticals as well but are beyond the scope of this review.
Recently, authors have demonstrated Microwave vacuum drying via
REV
TM
technology to produce efcacious and stable bio-
pharmaceuticals, including a live virus vaccine, with an 80 % reduction
in the time associated with batch Freeze-drying (Bhambhani et al.,
2021). Similar studies exploiting microwaves for the drying of mAbs,
have been found in literature as Microwave-assisted Freeze-drying
(Gitter et al., 2019, 2018). Moreover, various other microsphere gen-
eration technologies under evaluation have been listed in Table 3. For
further reading on other drying technologies, readers are referred to the
cited reviews herein (Emami et al., 2018; Lovalenti and Truong-Le,
2020; Pardeshi et al., 2021; Thorat et al., 2020; Vass et al., 2019; Walters
et al., 2014; Durance et al., 2020).
3. Biopharmaceutical Characterization
As described in the ICH Q5E guidelines, product comparability
subject to any changes in the manufacturers manufacturing process
requires the evaluation of the impact of an alternative process on the
safety, quality and efcacy of biopharmaceutical products (ICH, 2004).
Any aberrations in the CQAs of biopharmaceutical products post drying
can be assessed using various analytical and characterization tech-
niques. Fig. 15 shows a comprehensive list of techniques currently
employed to study some of the product CQAs in the solid and liquid-
state. Some of these techniques are not employed for routine analyses
but can provide additional information in understanding the impact of
CPPs on product CQAs.
While some chromatographic and spectroscopic techniques are
employed as QC release tests, they fail to provide intricate and high-
resolution information in understanding protein stability and their in-
teractions with excipients in the solid-state. ssHDX-MS (Kammari and
Topp, 2020; Moorthy et al., 2014; Wilson et al., 2019) and ssPL-MS (Iyer
et al., 2016, 2013) are novel, high resolution mass spectrometric tech-
niques that have provided further insights in understanding and eluci-
dating protein stability. HDX-MS has been previously demonstrated to
study the conformational stability of proteins in liquid solutions (Houde
et al., 2011; Tsutsui and Wintrode, 2007; Wales and Engen, 2006), in
frozen solutions (Zhang et al., 2012, 2011) and protein adsorption onto
solid surfaces (Buijs et al., 2003, 2000, 1999; Zhang and Smith, 1993).
One of the potential applications of this technique is predictive stability.
Moreover, this technique offers an insight to study protein degradation
in the solid-state which is a poorly understood area. In addition to the
application of NMR in studying protein-excipient interactions (Mensink
et al., 2016; Tian et al., 2007; Yoshioka et al., 2011), time-domain Nu-
clear Magnetic Resonance Spectroscopy (TD-NMR) has been demon-
strated to determine the RMC in Freeze-dried biopharmaceuticals
(Abraham et al., 2019a). Also, NMR coupled with Magnetic Resonance
Imaging (MRI) has been studied to determine complete reconstitution of
Freeze-dried products (Partridge et al., 2019). More recently, the
determination of reconstitution time of Freeze-dried BSA has been
demonstrated using uorescence spectroscopy (ElKassas et al., 2021).
Furthermore, advancements in CD spectroscopy have allowed re-
searchers to study intricate information in the lower vacuum ultraviolet
(VUV) region (<190 nm) using SRCD spectroscopy (Miles et al., 2008;
Miles and Wallace, 2020, 2006; Wallace, 2019, 2009; Wallace et al.,
2004). SRCD has also seen application in protein photostability,
Fig. 14. (a) Optical image of Microglassied
TM
BSA in decanol. (b) SEM image of Microglassied
TM
beads. Reprinted (adapted) from (Aniket et al., 2014) with
permission from Elsevier.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
17
photoisomerization, proteinligand interactions and RNA characteriza-
tion (Auvray et al., 2019; Hussain et al., 2018; Nasser et al., 2018; Wien
et al., 2021).
Overall, the characterization techniques listed in Fig. 15 may have
certain merits and demerits on their own but if employed as compli-
mentary techniques, they can provide further insights on biopharma-
ceutical stability. These characterization techniques combined with
different drying technologies can help the biopharmaceutical industry in
choosing appropriate methods for manufacturing and testing their
products.
4. Formulation aspects for Drying Technologies
Majority of the Freeze-dried biopharmaceutical formulations contain
buffers, salts, amino acids, sugars, bulking agents, surfactants, tonici-
ers, preservatives etc. (Bjeloˇ
sevi´
c et al., 2020; Gervasi et al., 2018).
Minimizing pH shifts, protein mobility in the solid-state and denatur-
ation at airliquid interfaces, increasing colloidal stability and providing
increased solubility etc. are some of the key roles played by formulation
components. Buffer salts such as sodium phosphate show a tendency to
crystallize, thereby, causing pH shifts during freezing and depress crit-
ical temperatures of formulations crucial for Freeze-drying (Kolhe et al.,
2009; Wu et al., 2015). These challenges may be eliminated for Spray-
drying of formulations containing such buffers.
Some of the amorphous saccharides, provide improved stability for
Freeze-dried and Spray-dried proteins (Carpenter et al., 1994; Chang
et al., 2005a; Green and Angell, 1989; Kreilgaard et al., 1999). Treha-
lose, compared to sucrose, is less frequently used in Freeze-dried for-
mulations but is the most preferred disaccharide for Spray-drying (Pinto
et al., 2021). Trehalose can signicantly protect biopharmaceuticals at
higher temperatures during Spray-drying due to its high glass transition
temperature (>100 C) (Liao et al., 2004; Massant et al., 2020; Simperler
et al., 2006). Authors have demonstrated that while sucrose preserves
the proteins secondary structure during dehydration, trehalose pro-
vides protection during long-term storage of Freeze-dried and Spray-
dried lysozyme (Starciuc et al., 2019). Apart from disaccharides,
cyclodextrin is widely used in Spray-dried protein formulations (Pinto
et al., 2021). A good stabilizing effect was observed at a 2:1 (protein :
sugar) ratio, respectively and an optimized inclusion of both trehalose
and sucrose could improve the overall stability of lysozyme (Liao et al.,
2003, 2002). A 9:1:10 blend of mannitol, trehalose and lysozyme,
respectively, exhibited higher bioactivity and stability post Spray-drying
(Hulse et al., 2008). Moreover, the inclusion of ethanol as a co-solvent
improved the aerosol performance of Spray-dried lysozyme by exhibit-
ing a higher percentage of ne particle fraction (FPF) compared to the
water-based lysozyme formulation (Ji et al., 2016). In spite of their
benets, it is important to note that trehalose and sucrose may crystalize
in their frozen state as well as in their dried state during storage at
accelerated temperature and moisture and cause damage to protein
structure and stability (Singh, 2018; Singh et al., 2011). Therefore,
biopharmaceutical products are processed and stored below their glass
Table 3
Microsphere Technologies under evaluation for pharmaceuticals.
Microsphere
Technology
Manufacturer Reference
Q-Sphera
TM
Technology MidaTech Pharma (MidaTech Pharma, 2021;
Seaman et al., 2019)
iSPHERE
TM
Technology Pulmatrix (Pulmatrix, 2021)
Kureha Microsphere
Technology
Kureha (Kureha, 2019; Vhora et al.,
2019)
Plexis® Technology Auritec
Pharmaceuticals
(Auritec Pharmaceuticals, 2016;
NCT03626714, 2019)
Stratum
TM
Technology Orbis Biosciences (Dormer and Berkland, 2016;
Vhora et al., 2019)
FormEZE
TM
Microparticle
Technology
Evonik Industries (Evonik, 2015; Vhora et al.,
2019)
Fig. 15. Ofine analytical and characterization techniques for biopharmaceutical products. AUC, Analytical Ultracentrifugation; BCA, Bicinchoninic acid Assay; CD,
Circular Dichroism Spectroscopy; DLS, Dynamic Light Scattering; FTIR, Fourier Transform Infrared Spectroscopy; HIAC, High Accuracy Fluid Particle Counting; NIR,
Near Infrared Spectroscopy; NMR, Nuclear Magnetic Resonance Spectroscopy; NTA, Nanoparticle Tracking Analysis; RMM, Residual Mass Measurement; RP-HPLC,
Reverse Phase High Performance Liquid Chromatography; SDS-PAGE, Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis; SEC, Size Exclusion Chro-
matography; SEM, Scanning Electron Microscopy; ssHDX-MS, solid-state Hydrogen Deuterium Exchange Mass Spectrometry; ssPL, solid-state Photolytic Labelling;
SRCD, Synchrotron Radiation Circular Dichroism Spectroscopy; UVVis, UltravioletVisible Spectroscopy.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
18
transition or eutectic temperature.
On the other hand, crystalline bulking agents such as mannitol and
glycine may not necessarily confer protein stability but provide robust
and elegance Freeze-dried cakes structures (Johnson et al., 2002; Peters
et al., 2016; Varshney et al., 2007). Mannitol has also been reported to
reduce the reconstitution times in high concentration Freeze-dried cakes
(Kulkarni et al., 2018) and improve the aerosol performance of Spray-
dried anti-IgE formulations and salbutamol (Costantino et al., 1998;
Kaialy et al., 2010; Molina et al., 2019). Mannitol is the most popular
monosaccharide in dried biopharmaceuticals (Gervasi et al., 2018; Pinto
et al., 2021), although, crystallization of mannitol in the absence of
amorphous stabilizers can negatively impact protein structure and sta-
bility. Crystallization of mannitol leads to phase separation that reduces
the possible number of interactions between the protein and excipient,
thereby, rendering the protein unstable (Wilson et al., 2019). Approxi-
mately, 20 40 % of sugar content is sufcient to safe guard the anti-
body in terms of its stability (Dani et al., 2007; Maury et al., 2005). In
spite of studies carried out on the stability of antibodies, the effect of
elevated temperature stress (>180
C) on mAbs due to high inlet tem-
peratures of large-scale Spray-dryers is still a concern. Bowen et al.
conducted studies by testing different commercial mAb : trehalose for-
mulations (2:1 and 1:2 ratio by weight, respectively) and stated that the
percentage of monomers was lower in the Freeze-dried product even
though the residual water content in the Freeze-dried product was quite
low compared to the Spray-dried product (Bowen et al., 2013). A yield
of >95 % and improved storage stability was reported for commercial
mAbs post Spray-drying (Gikanga et al., 2015). More recently, authors
have shown comparable stability of sucrose-containing myoglobin and
lysozyme post Freeze-drying and Spray-freeze-drying (Mutukuri et al.,
2021). Along with trehalose, sucrose and mannitol, excipients such as
polyethylenimine, hyaluronic acid, leucine, phenylalanine, arginine,
cysteine, glycine etc. have been used for Spray-freeze-drying of some
biopharmaceuticals (Adali et al., 2020; Chaurasiya and Zhao, 2020).
Amongst amino acids, L-arginine and L-arginine hydrochloride have
been reported to increase protein stability and solubility and reduce the
viscosity of protein solutions (Inoue et al., 2014; Shah et al., 2012;
St¨
artzel, 2018; St¨
artzel et al., 2015). Interestingly, it has been reported
that arginine along with other excipients and by itself in protein for-
mulations is capable of acting as the main stabilizer (Baynes et al., 2005;
Reslan et al., 2017; Shukla and Trout, 2011; Tsumoto et al., 2004).
Moreover, the stabilizing effect of a large number of amino acids and
their combinations were studied on Spray-dried catalase, lysozyme and
Pandemrix inuenza vaccine containing haemagglutinin (Ajmera and
Scherließ, 2014). A combination of arginine, glycine, and protein in the
ratio of [(1 +1) +1] resulted in a very good stabilizing effect post Spray-
drying. Amongst the commonly used excipients for proteins, a combi-
nation of trehalose, arginine and protein in a ratio of 1:1 (excipient :
protein) by weight improved the properties of Spray-dried mAbs in
terms of reconstitution time and stability (Massant et al., 2020).
Furthermore, leucine, isoleucine and trileucine have been reported to
effectively protect the protein from sheer stress caused during atomi-
zation and improve powder owability, dispersibility and aerosolization
(Ganderton et al., 1999; Lechuga-Ballesteros et al., 2008; Schüle et al.,
2008; Staniforth et al., 2001). While histidine is the most popular amino
acid used in liquid and Freeze-dried protein formulation (Gervasi et al.,
2018), leucine is the most preferred in Spray-dried protein formulations
(Pinto et al., 2021).
In addition, surfactants play a major role in reducing aggregation due
to protein exposure at airliquid, solidliquid and liquidliquid in-
terfaces (Chen et al., 2021; Chernysheva et al., 2018; Maa et al., 1998).
These interfaces may be generated during freezing, atomization,
reconstitution etc. Therefore, consideration to select appropriate ex-
cipients specic to the product and process right at the formulation
development stage is crucial to ensure product stability. A summary of
some of the commonly used excipients for Freeze-drying and Spray-
drying has been described in Table 4, though further investigations
are required to elucidate the mechanism of stabilization by excipients
for other drying technologies.
5. Feasibility of PAT for Drying Technologies
As per the ‘Pharmaceutical DevelopmentICH Q8(R2) guidelines,
PAT is a QbD approach to design, analyse and control manufacturing
(ICH, 2009). Several PATs have been explored in literature, however,
some drawbacks associated with their feasibility during batch Freeze-
drying have been identied. Most of the PATs provide an average
result of the batch and cannot be implemented in-line, invasively or non-
invasively for all individual product vials during processing. The risk of
damage due to sterilization in the drying chamber makes these tools
unt for commercial cGMP cycles. On the contrary, most of these PATs
can be potentially employable, in-line or at-line for unit doses or bulk
product for some of the alternative drying technologies.
Typically, pressure and temperature sensors are used to monitor
Freeze-drying cycles (Fissore et al., 2018; Nail et al., 2017; Nail and
Johnson, 1992). This is essential for the development and optimization
of Freeze-drying cycles and to account for the RMC in Freeze-dried
cakes. Optical ber sensors (OFS) (Kasper et al., 2013) and wireless
data loggers such as temperature remote interrogation system (TEMP-
RIS) (Schneid and Gieseler, 2008) and TrackSense® (Ellab, 2020) are
other available options for product temperature monitoring during
Freeze-drying. Comparative measurements between the pirani gauge
and the capacitance manometer, and between the temperature probes
and the shelf temperature are used to determine the primary drying end-
point (Fissore et al., 2018; Nail et al., 2017). In comparison, Tunable
Diode Laser Absorption Spectroscopy (TDLAS) and Mass Spectrometry
(MS) have been used as better alternatives (Ganguly et al., 2018; Gies-
eler et al., 2007; Patel et al., 2010a, 2010b). Along with estimating the
primary drying end-point, vapour ow rate, average product tempera-
ture, heat transfer coefcient and mass transfer resistance, these tools
can also be used to monitor the secondary drying end-point and RMC
with high sensitivity and accuracy (Ganguly et al., 2018; Gieseler et al.,
2007; Patel et al., 2010a; Schneid et al., 2009). Ganguly et al. showed
that MS was highly sensitive to an average cake moisture of <3 % in the
late secondary drying phase (Ganguly et al., 2018). Additionally, MS can
be used to detect silicon oil and helium gas leaks in a Freeze-dryer. In
contrast to some of the drawbacks associated with these tools during
batch Freeze-drying, TDLAS and MS can be potentially congured with
most of the continuous drying technologies to estimate the RMC for all
individual vials or bulk product.
More interestingly, authors explored potential applications of Near
Infrared Chemical Imaging (NIR-CI) and 4D Micro-Computed X-ray
tomography and imaging for Spin-freeze-drying (Brouckaert et al.,
2018; Goethals et al., 2020; Vanbillemont et al., 2020b). NIR-CI was able
to capture different polymorphs of mannitol and the distribution of re-
sidual moisture in mannitol and mannitol-sucrose containing Spin-
freeze-dried vials whereas 4D Micro-Computed X-ray tomography and
imaging were able to detect intra-vial differences in the mass transfer
resistance and primary drying end-point. Also, a NIR probe coupled to a
FT-NIR analyser was connected to the vial holder to demonstrate the
primary and secondary drying end-point (De Meyer et al., 2015).
However, concerns relating to heat generation from NIR radiating
halogen bulbs, mechanical challenges, and the feasibility of imple-
menting this PAT in a cGMP environment need to be addressed. More-
over, Near Infrared Frequency Modulated Spectroscopy (NIR-FMS) is a
non-invasive, easy and a quick method for determining headspace ox-
ygen and moisture levels in Freeze-dried vials (Cook and Ward, 2011a,
2011b; Lin and Hsu, 2002; Victor et al., 2017). Correlation observed
between Karl Fisher analysis and NIR spectroscopy can make it easier to
predict the RMC post drying (Afeck et al., 2021; Brouckaert et al.,
2018; Carfagna et al., 2020). This technique can be employed in-line or
at-line to analyze all dried vials or bulk product generated via some of
the alternative drying technologies. Furthermore, NIR, MIR and Raman
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
19
spectroscopy have been used as potential PATs for process control and
quality assurance of infant formula and dairy ingredients powders
(Wang et al., 2018). NIR and Raman spectroscopy have also been
employed in-line to study the protein conformational stability and ag-
gregation (Nitika et al., 2021; Pieters et al., 2013, 2012). A NIR or a
Raman probe can be positioned to analyze vials as they move across
different drying chambers during continuous Freeze-drying of sus-
pended vials. Similarly, a probe can be placed in-line at the different
stages of Active-freeze-drying, Spray-drying, and Spray-freeze-drying or
at-line during powder lling into unit doses. These PATs can also reduce
batch release test time from ll nish to the market.
Particle size is one of the major CQAs for free-owing powder-based
products. Laser diffraction has been used both in-line and at-line to
measure the particle size distributions during Spray-drying (Chan et al.,
2008). Moreover, a variety of different PATs such as Spatial Filtering
Velocimetry (SFV), Focused Beam Reectance Measurements (FBRM),
Photometric Stereo Imaging (PSI) and Eyecon® technology have been
explored for analysing dry powder particle size (Silva et al., 2013).
These PATs can be congured in-line or at-line for Active-freeze-drying,
Spray-freeze-drying and PRINT® Technology to measure the particle
size of dry powder. However, discrepancies observed in the particle size
measurements of the PATs have been discussed by the authors (Silva
et al., 2013) and so, users must consider pre-requisite knowledge on the
theory, mechanism and equipment for appropriate results.
In summary, RMC, protein structural conformation and aggregation,
particle size and polymorphism of excipients are some of the major
CQAs for biopharmaceutical products that can be monitored using PATs.
Monitoring CPPs such as product temperature, drying rate etc. are as
important as product CQAs and so, some of the continuous drying
technologies offer a greater advantage from PAT perspective. Table 5
summarizes the application of some PATs based on authorsassessments
for only some of the drying technologies. The potential compatibility of
these PATs with other drying technologies listed in Table 5 has not been
found in literature and is based on opinion.
6. Scale-up, Packaging and Validation aspects for drying
Technologies
Scale-up and technology transfer involve moving a pharmaceutical
manufacturing process from one facility to another i.e. from a devel-
opment / pilot-scale to a commercial-scale or an intra-site / inter-site ll
nish line to line transfer. As per ICH Q12 guidelines, technology
transfer may also be required for lifecycle changes across different
commercial facilities (ICH, 2019).
Freeze-drying in vials requires the qualication of not only the
drying process, but several other critical ll nish operations such as
compounding, ltration and vial lling. This brings in further technical
and compliance requirements such as mixing studies, lter bacterial
retention, ll volume cycle development, media ll qualication and
environmental monitoring (FDA, 2014a). Moreover, large loading times
(4 12 h) for vial lling impacts process efciency, may lead to issues
such as product splashing and/or foaming and may also impact product
Table 4
Key roles of some commonly used excipients for Freeze-drying and Spray-drying of biopharmaceuticals.
Excipients Examples Freeze-drying Spray-drying
Amorphous
saccharides
Sucrose Preservation of protein secondary structure by glassy-state stabilization, H-bonding (Liao et al., 2003, 2002; Starciuc et al., 2020).
Trehalose Protection during long-term storage by glassy-state stabilization, H-bonding and high glass transition temperature (>100 C) (Liao
et al., 2004; Massant et al., 2020; Simperler et al., 2006; Starciuc et al., 2020).
Rafnose n/a
Glassy-state stabilization, high glass temperature (114 C), increased
ne particle fraction and aerosolization (Alhajj et al., 2021; Amaro
et al., 2011; Zhao et al., 2018).
Glucose
Reducing sugars not preferred due to pH shifts and
Maillard reaction (Mensink et al., 2017).
Glassy-state stabilization (Ying et al., 2012).
Lactose
Improved particle dispersibility (Horn et al., 2020; Pilcer et al., 2012;
Seville et al., 2007) but high hygroscopicity (Hebbink and Dickhoff,
2019).
Polyols
Mannitol Bulking agent and reduced reconstitution time (Kulkarni
et al., 2018; Mehta et al., 2013).
Improved aerosol performance (Costantino et al., 1998; Kaialy et al.,
2010; Molina et al., 2019).
Sorbitol Plasticize
α
-motions but antiplasticize β-motions in combination with non-reducing amorphous sugars (Chang et al., 2005b; Cicerone
and Soles, 2004). Glycerol
Amino Acids
Leucine, isoleucine,
trileucine n/a
Protection from atomization stress, improved powder owability,
dispersibility and aerosolization (Alhajj et al., 2021; Lechuga-
Ballesteros et al., 2008; Seville et al., 2007).
Arginine
Increased protein stability, solubility and reduced
viscosity (Inoue et al., 2014; Shah et al., 2012; St¨
artzel,
2018; St¨
artzel et al., 2015).
Improved protein stability, reduced turbidity and reconstitution time (
Ajmera and Scherließ, 2014; Massant et al., 2020).
Glycine Bulking agent (Varshney et al., 2007). Improved protein activity and stability (Ajmera and Scherließ, 2014).
Histidine Amino-acid buffer (Al-hussein and Gieseler, 2013; Liao
et al., 2013). Improved protein activity and stability (Ajmera and Scherließ, 2014).
Surfactants
Non-ionic
(Tween 80, 20)
Reduced airliquid interfacial protein adsorption, reduced aggregation, improved protein refolding (Arsiccio and Pisano, 2018;
Chernysheva et al., 2018; Maa et al., 1998).
Anionic
(Sodium stearate,
magnesium stearate)
n/a Moisture protectant, improved pore formation and aerosolization (
Parlati et al., 2009; Tewes et al., 2014; Yu et al., 2018).
Pulmonary
(DPPC, DSPC) n/a Improved aerosolization and surface enrichment properties (Cuvelier
et al., 2015; Miller et al., 2015; Weers and Tarara, 2014).
Other
polysaccharides
Cyclodextrin Improved protein stability, elegant cake appearance (
Haeuser et al., 2020).
Glassy-state stabilization, improved powder owability, anti-
hygroscopicity (Branchu et al., 1999; Serno et al., 2010; Zhao et al.,
2018).
Inulin Improved protein stability (Hinrichs et al., 2001; Ke et al., 2020).
DPPC, Dipalmitoyl phosphatidylcholine; DSPC, Distearoyl phosphatidylcholine; n/a, not applicable or not available.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
20
stability upon increased validated time out of refrigeration (Patel et al.,
2017; Rathore and Rajan, 2008). Along with large loading times, large
unloading times and vial inspection also impact the efciency of com-
mercial operations and slows the inventory turnover time in the most
expensive footprint area of a commercial site i.e., controlled areas for
aseptic ll nish. Such additional steps are eliminated for Active-freeze-
drying, Spray-drying and Spray-freeze-drying technologies. These dry-
ing technologies can also provide the option for drug substance (DS)
drug product (DP) validation as continuous processes and eliminate the
requirement for additional validation steps, thereby, minimizing the
complexities associated with the regulatory ling and qualication of an
end-to-end ll nish process (Pisano, 2020). Also, with the bulk product
stored in the dried state as opposed to the liquid or frozen state, cold
chain shipping validation of the bulk can be reduced. On the negative
side, a more comprehensive cleaning validation for alternative drying
technologies may be a requirement as per cGMP.
Furthermore, many of the Freeze-dried biopharmaceuticals such as
Eloctate® (FDA, 2014c), Alprolix® (FDA, 2014d), Fabrazyme® (FDA,
2010) etc. are manufactured in multiple dose strengths. Batch Freeze-
drying for multiple dose strengths require completely different ll n-
ish processes with additional qualication and validation for primary
packaging components supply chain, ll nish line equipment, sterili-
zation methods, ll volume, Freeze-drying cycles, capping, inspection,
container closure integrity etc. In contrast, alternative drying technol-
ogies such as Active-freeze-drying, Spray-drying, Spray-freeze-drying
etc. make it easier to ll and pack free-owing product into different
container types such as vials, ampules, syringes, sachets etc. at multiple
dose strengths. This also helps in simplifying infusion requirements at
Table 5
Potential / compatible PATs for drying technologies.
PAT Application Batch Freeze-
drying
Active-
freeze-
drying
Spin-
freeze
drying
Spray-
freeze-
drying
Continuous
Freeze-drying
of suspended
vials
Spray-
drying
PRINT®
Technology
Temperature Probes:
Thermocouples and RTDs
(Fissore et al., 2018; Nail et al.,
2017).
Wireless Probes:
TrackSense®, TEMPRIS
(Schneid and Gieseler, 2008;
Ellab, 2020).
Optical Fibers (Kasper et al.,
2013).
Average product
temperature mapping. Yes
Yes
(RTDs can
be
installed
At-line)
No
Yes
(RTDs can
be
installed
At-line)
Yes
(Wireless
probes can be
installed in-
line)
Yes
(RTDs can
be
installed
At-line)
n/a
Product temperature
mapping for all individual
vials or bulk product.
No Yes
(At-line) No Yes
(At-line) No Yes
(At-line) n/a
IR Thermography
(Harguindeguy and Fissore,
2021).
Product temperature
mapping for individual
vials or bulk product only
within the cameras eld
of view.
Yes, but the center
vials are
calculated based
on average.
Yes Yes Yes Yes Yes
(At-line)
Yes
(At-line)
NIR Spectroscopy (De Beer
et al., 2009; Mensink et al.,
2015; Pieters et al., 2012;
Wang et al., 2018).
In-line protein structure
analysis, protein
aggregation and
distribution of excipient
polymorphs for all
individual vials or bulk
product.
No, but can
provide an
average
measurement.
Yes Yes Yes
(At-line) Yes Yes
(At-line)
Yes
(At-line) Raman Spectroscopy (De Beer
et al., 2009; Nitika et al., 2021;
Pieters et al., 2013; Wang et al.,
2018).
NIR-FMS (Carfagna et al.,
2020; Cook and Ward, 2011a,
2011b; Lin and Hsu, 2002;
Victor et al., 2017).
In-situ vial headspace
oxygen and moisture
measurements for all
individual vials or bulk
product.
No, but can
provide an
average
measurement.
Yes No Yes
(At-line) Yes Yes
(At-line) n/a
NIR-CI (Brouckaert et al.,
2018).
In-situ RMC estimation
and distribution of
excipient polymorphs in
all individual vials or bulk
product.
Yes Yes Yes Yes Yes Yes
(At-line) Yes
TDLAS (Gieseler et al., 2007;
Kessler et al., 2006; Kuu et al.,
2011, 2009; Schneid et al.,
2011, 2009; Sharma et al.,
2019).
MS (Fissore et al., 2018;
Ganguly et al., 2018; Nail et al.,
2017).
In-situ RMC, drying end
point, vapor ow rate
estimation for all
individual vials or bulk
product.
No, but can
provide an
average
measurement.
Yes Yes Yes Yes Yes Yes
MS (Barfuss, 2014; Connelly
and Welch, 1993; Ganguly
et al., 2018).
In-line silicon oil or gas
leak detection. Yes Yes Yes Yes Yes Yes Yes
Laser
Diffraction (Chan et al.,
2008; Dos
Reis et al.,
2021; Petrak
et al., 2018;
Silva et al.,
2013).
In-line / at-line dry powder
particle sizing. n/a Yes n/a Yes n/a Yes Yes
Light
Microscopy
SFV
PSI
FBRM
Eyecon®
RTDs, Resistance Temperature Detectors; n/a, not applicable or not available.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
21
clinics.
To enable successful scale-up and technology transfer, drying pro-
cesses require a QbD approach where the process boundaries are well
dened and provide adequate robustness for commercial operations
(Nail and Searles, 2008). Small scale process modelling for Freeze-
drying is complex as each individual vial behaves as its own drying
system, subject to variability, as a function of vial heat transfer coef-
cient, freezing temperature and location on the shelf. Active-freeze-
drying, Spray-drying and Spray-freeze-drying, for example, are not as
hindered by the challenges of drying in individual containers as they
provide a more predictive and consistent manufacturing performance.
In terms of scalability, moving an existing commercial product from
a batch Freeze-drying process to an alternative drying process is regar-
ded as a major regulatory change and comprehensive comparability
data would be required as part of implementation and approval (Pisano,
2020). The signicant cost of biopharmaceutical DS, commercial line
time for engineering, validation batches, long term stability data and the
requirement for ling a change mean any efciency gain obtained by
alternative drying processes may be offset by such costs. Overall, the
most viable route to introduce alternative drying processes may very
well be on the back of the development and industrialization of new
products.
7. Conclusion and Future Directions
At present, batch Freeze-drying is a well-established drying tech-
nology for the majority of biopharmaceutical products. Many of the
alternative drying technologies have increasingly shown promising
prospects for manufacturing solid biopharmaceuticals without
compromising on the safety, quality and efcacy of biopharmaceutical
products. These potential drying technologies are signicant to the
biopharmaceutical industry as they will not only reduce time, energy
consumption and associated costs with the manufacturing of life-saving
drugs but also help in mitigating any risks with the supply of drugs
during pandemics such as Covid-19. While some of the alternative
methods offer continuous manufacturing at reduced operational costs,
the impact of CPPs such as temperature, shear, etc. on product CQAs is
the fundamental requirement for the selection of drying technologies.
Although drying technologies, namely, Spin-freeze-drying, Spray-
freeze-drying, Spray-drying, PRINT® and Microglassication
TM
have
shown positive results on the stability of some proteins and inhaled
biopharmaceuticals, their impact on a wide range of parenteral bio-
pharmaceuticals is yet to be studied. Through product-specic research,
sufcient stability data is required to move from conventional Freeze-
drying to continuous manufacturing. Along with CPPs, the choice of
formulation components with respect to the drying process as well as the
product is crucial to ensure product stability. Moreover, the molecular
mechanism of interaction of biopharmaceuticals with specic excipients
in the solid-state is poorly understood. Some of the advanced charac-
terization techniques and PATs in tandem can offer faster and in-depth
analysis in understanding and evaluating the product-process relation-
ship. While most of the alternative drying methods can offer signicant
benets with the usage of PATs, their feasibility at commercial scale
requires further exploration. In terms of scale-up, packaging and vali-
dation aspects, some of the alternative drying processes offer a greater
advantage in reducing the complexities associated with the validation of
multiple ll nish unit operations. The commercial scale operation for
some alternative drying technologies has been demonstrated with
proven potential in the biopharmaceutical industry though some scale-
up challenges are yet to be addressed.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgement
The authors acknowledge funding received from Waterford Institute
of Technology Sano Waterford Co-fund PhD Scholarship Program
(Ashutosh Sharma) and the Irish Research Council Enterprise Part-
nership Scheme (Project ID: EPSPG/2020/56). The authors also
acknowledge Robert Turok at SPX Flow Inc. for his valuable suggestions
on spray-drying in the manuscript.
References
Abdul-Fattah, A.M., Truong-Le, V., Yee, L., Nguyen, L., Kalonia, D.S., Cicerone, M.T.,
Pikal, M.J., 2007. Drying-induced variations in physico-chemical properties of
amorphous pharmaceuticals and their impact on stability (I): Stability of a
monoclonal antibody. J. Pharm. Sci. 96, 19832008. https://doi.org/10.1002/
jps.20859.
Abraham, A., Elkassabany, O., Krause, M.E., Ott, A., 2019a. A nondestructive and
noninvasive method to determine water content in lyophilized proteins using low-
eld time-domain NMR. Magn. Reson. Chem. 57, 873877. https://doi.org/
10.1002/mrc.4864.
Abraham, I., Ali Elkordy, E., Haj Ahmad, R., Ahmad, Z., Ali Elkordy, A., 2019. Effect of
Spray-Drying and Electrospraying as Drying Techniques on Lysozyme
Characterisation, in: Electrospinning and Electrospraying - Techniques and
Applications. IntechOpen. https://doi.org/10.5772/intechopen.86237.
Acorda Therapeutics, 2021. INBRIJA® (levodopa inhalation powder). URL: htt
ps://www.acorda.com/products/products/inbrija (accessed 8.4.21).
Adali, M.B., Barresi, A.A., Boccardo, G., Pisano, R., 2020. Spray freeze-drying as a
solution to continuous manufacturing of pharmaceutical products in bulk. Processes.
https://doi.org/10.3390/PR8060709.
Afeck, Richard P., Khamar, Dikshitkumar, Lowerre, Katie M., Adler, Nikoletta,
Cullen, Sean, Yang, Mark, McCoy, Timothy R., 2021. Near infrared and frequency
modulated spectroscopy as non-invasive methods for moisture assessment of freeze-
dried biologics. J. Pharm. Sci. 110 https://doi.org/10.1016/j.xphs.2021.06.016.
Ajmera, A., Scherließ, R., 2014. Stabilisation of proteins via mixtures of amino acids
during spray drying. Int. J. Pharm. 463, 98107. https://doi.org/10.1016/j.
ijpharm.2014.01.002.
Al-hussein, A., Gieseler, H., 2013. Investigation of histidine stabilizing effects on LDH
during freeze-drying. J. Pharm. Sci. 102, 813826. https://doi.org/10.1002/
jps.23427.
Al-Hussein, A., Gieseler, H., 2012. The effect of mannitol crystallization in mannitol-
sucrose systems on LDH stability during freeze-drying. J. Pharm. Sci. https://doi.
org/10.1002/jps.23173.
Albasarah, Y.Y., Somavarapu, S., Taylor, K.M.G., 2010. Stabilizing protein formulations
during air-jet nebulization. Int. J. Pharm. 402, 140145. https://doi.org/10.1016/j.
ijpharm.2010.09.042.
Alhajj, N., OReilly, N.J., Cathcart, H., 2021. Designing enhanced spray dried particles
for inhalation: A review of the impact of excipients and processing parameters on
particle properties. Powder Technol. 384, 313331. https://doi.org/10.1016/j.
powtec.2021.02.031.
Allison, S.D., Chang, B., Randolph, T.W., Carpenter, J.F., 1999. Hydrogen bonding
between sugar and protein is responsible for inhibition of dehydration-induced
protein unfolding. Arch. Biochem. Biophys. 365, 289298. https://doi.org/10.1006/
abbi.1999.1175.
Amaro, M.I., Tajber, L., Corrigan, O.I., Healy, A.M., 2011. Optimisation of spray drying
process conditions for sugar nanoporous microparticles (NPMPs) intended for
inhalation. Int. J. Pharm. 421, 99109. https://doi.org/10.1016/j.
ijpharm.2011.09.021.
Aniket, Gaul, D.A., Bittereld, D.L., Su, J.T., Li, V.M., Singh, I., Morton, J., Needham, D.,
2015a. Enzyme dehydration using microglassicationTM preserves the Proteins
structure and function. J. Pharm. Sci. 104, 640651. https://doi.org/10.1002/
jps.24279.
Aniket, Gaul, D.A., Rickard, D.L., Needham, D., 2014. MicroglassicationTM: A novel
technique for protein dehydration. J. Pharm. Sci. 103, 810820. https://doi.org/
10.1002/jps.23847.
Aniket, Tang, N., Rosenberg, J., Chilkoti, A., Needham, D., 2015b. Elastin-like
Polypeptide Microspheres: A Stimuli-Responsive Vehicle for Controlled Drug
Delivery, in: Society for Biomaterials. Charlotte, North Caroline.
Arsiccio, A., Pisano, R., 2018. Surfactants as stabilizers for biopharmaceuticals: An
insight into the molecular mechanisms for inhibition of protein aggregation. Eur. J.
Pharm. Biopharm. 128 https://doi.org/10.1016/j.ejpb.2018.04.005.
Ashton, L., Dusting, J., Imomoh, E., Balabani, S., Blanch, E.W., 2009. Shear-induced
unfolding of lysozyme monitored in situ. Biophys. J . 96, 42314236. https://doi.
org/10.1016/j.bpj.2009.02.024.
Auritec Pharmaceuticals, 2016. Plexis [WWW Document]. accessed 8.2.21. http://www.
auritecpharma.com/plexis/.
Authelin, J.R., Rodrigues, M.A., Tchessalov, S., Singh, S.K., McCoy, T., Wang, S.,
Shalaev, E., 2020. Freezing of Biologicals Revisited: Scale, Stability, Excipients, and
Degradation Stresses. J. Pharm. Sci. 109, 4461. https://doi.org/10.1016/j.
xphs.2019.10.062.
Auvray, F., Dennetiere, D., Giuliani, A., Jamme, F., Wien, F., Nay, B., Zirah, S., Polack, F.,
Menneglier, C., Lagarde, B., Hirst, J.D., R´
efr´
egiers, M., 2019. Time resolved transient
circular dichroism spectroscopy using synchrotron natural polarization. Struct. Dyn.
6, 054307 https://doi.org/10.1063/1.5120346.
A. Sharma et al.
International Journal of Pharmaceutics 609 (2021) 121115
22
Bando, K., Kansha, Y., Ishizuka, M., Tsutsumi, A., 2017. Innovative freeze-drying process
based on self-heat recuperation technology. J. Clean. Prod. 168, 12441250. https://
doi.org/10.1016/j.jclepro.2017.09.088.
Barfuss, H., 2014. Qualit¨
atsüberwachung in pharmazeutischen
Gefriertrocknungsanlagen. Vak. Forsch. und Prax. 26, 3538. https://doi.org/
10.1002/vipr.201400543.
Barresi, A.A., Pisano, R., Rasetto, V., Fissore, D., Marchisio, D.L., 2010. Model-Based
Monitoring and Control of Industrial Freeze-Drying Processes: Effect of Batch
Nonuniformity. Dry. Technol. 28, 577590. https://doi.org/10.1080/
07373931003787934.
Batens, M., Massant, J., Teodorescu, B., Van den Mooter, G., 2018. Formulating
monoclonal antibodies as powders for reconstitution at high concentration using
spray drying: Models and pitfalls. Eur. J. Pharm. Biopharm. 127, 407422. https://
doi.org/10.1016/j.ejpb.2018.02.002.
Baynes, B.M., Wang, D.I.C., Trout, B.L., 2005. Role of arginine in the stabilization of
proteins against aggregation. Biochemistry 44, 49194925. https://doi.org/
10.1021/bi047528r.
Becker, W., 1957. Gefriertrocknungsverfahren. DE967120.
Bee, J.S., Stevenson, J.L., Mehta, B., Svitel, J., Pollastrini, J., Platz, R., Freund, E.,
Carpenter, J.F., Randolph, T.W., 2009. Response of a concentrated monoclonal
antibody formulation to high shear. Biotechnol. Bioeng. 103, 936943. https://doi.
org/10.1002/bit.22336.
Bekard, I.B., Barnham, K.J., White, L.R., Dunstan, D.E., 2011.
α
-Helix unfolding in simple
shear ow, in: Soft Matter. Royal Society of Chemistry, pp. 203210. https://doi.
org/10.1039/c0sm00692k.
Bhambhani, A., Stanbro, J., Roth, D., Sullivan, E., Jones, M., Evans, R., Blue, J., 2021.
Evaluation of Microwave Vacuum Drying as an Alternative to Freeze-Drying of
Biologics and Vaccines: the Power of Simple Modeling to Identify a Mechanism for
Faster Drying Times Achieved with Microwave. AAPS PharmSciTech 22, 116.
https://doi.org/10.1208/s12249-020-01912-9.
Bjeloˇ
sevi´
c, M., Zvonar Pobirk, A., Planinˇ
sek, O., Ahlin Grabnar, P., 2020. Excipients in
freeze-dried biopharmaceuticals: Contributions toward formulation stability and
lyophilisation cycle optimisation. Int. J. Pharm. https://doi.org/10.1016/j.
ijpharm.2020.119029.
Bodier-Montagutelli, E., Mayor, A., Vecellio, L., Respaud, R., Heuz´
e-Vourc, N, 2018.
Designing inhaled protein therapeutics for topical lung delivery: what are the next
steps? Expert Opin. Drug Deliv. 15, 729736. https://doi.org/10.1080/
17425247.2018.1503251.
Bodier-Montagutelli, E., Respaud, R., Perret, G., Baptista, L., Duquenne, P., Heuz´
e-
Vourch, N., Vecellio, L, 2020. Protein stability during nebulization: Mind the
collection step! Eur. J. Pharm. Biopharm. 152, 2334. https://doi.org/10.1016/j.
ejpb.2020.04.006.
Bouchard, A., Jovanovi´
c, N., Jiskoot, W., Mendes, E., Witkamp, G.J., Crommelin, D.J.A.,
Hoand, G.W., 2007. Lysozyme particle formation during supercritical uid drying:
Particle morphology and molecular integrity. J. Supercrit. Fluids 40, 293307.
https://doi.org/10.1016/j.supu.2006.07.005.
Bowen, M., Turok, R., Maa, Y.F., 2013. Spray Drying of Monoclonal Antibodies:
Investigating Powder-Based Biologic Drug Substance Bulk Storage. Dry. Technol. 31,
14411450. https://doi.org/10.1080/07373937.2013.796968.
Branchu, S., Forbes, R.T., York, P., Petr´
en, S., Nyqvist, H., Camber, O., 1999.
Hydroxypropyl-β-cyclodextrin inhibits spray-drying-induced inactivation of
β-galactosidase. J. Pharm. Sci. 88, 905911. https://doi.org/10.1021/js9804819.
Broadhead, J., Rouan, S.K.E., Hau, I., Rhodes, C.T., 1993. The Effect of Process and
Formulation Variables on the Properties of Spray-dried β-Galactosidase. J. Pharm.
Pharmacol. 46, 458467. https://doi.org/10.1111/j.2042-7158.1994.tb03828.x.
Broadwin, S.M., 1965. Centrifugal freeze drying apparatus. US3203108A.
Brouckaert, D., De Meyer, L., Vanbillemont, B., Van Bockstal, P.J., Lammens, J.,
Mortier, S., Corver, J., Vervaet, C., Nopens, I., De Beer, T., 2018. Potential of Near-
Infrared Chemical Imaging as Process Analytical Technology Tool for Continuous
Freeze-Drying. Anal. Chem. 90, 43544362. https://doi.org/10.1021/acs.
analchem.7b03647.
Buijs, J., Ramstr¨
om, M., Danfelter, M., Larsericsdotter, H., Håkansson, P., Oscarsson, S.,
2003. Localized changes in the structural stability of myoglobin upon adsorption
onto silica particles, as studied with hydrogen/deuterium exchange mass
spectrometry. J. Colloid Interface Sci. 263, 441448. https://doi.org/10.1016/
S0021-9797(03)00401-6.
Buijs, J., Speidel, M., Oscarsson, S., 2000. The stability of lysozyme adsorbed on silica
and gallium arsenide surfaces: Preferential destabilization of part of the lysozyme
structure by gallium arsenide. J. Colloid Interface Sci. 226, 237245. https://doi.
org/10.1006/jcis.2000.6818.
Buijs, J., Vera, C.C., Ayala, E., Steensma, E., Håkansson, P., Oscarsson, S., 1999.
Conformational stability of adsorbed insulin studied with mass spectrometry and
hydrogen exchange. Anal. Chem. 71, 32193225. https://doi.org/10.1021/
ac9809433.
Burgess, G., Boyce, M., Jones, M., Larsson, L., Main, M.J., Morgan, F., Phillips, P.,
Scrimgeour, A., Strimenopoulou, F., Vajjah, P., Zamacona, M., Palframan, R., 2018.
Randomized study of the safety and pharmacodynamics of inhaled interleukin-13
monoclonal antibody fragment VR942. EBioMedicine 35, 6775. https://doi.org/
10.1016/j.ebiom.2018.07.035.
Cao, W., Krishnan, S., Ricci, M.S. pee., Shih, L.Y., Liu, D., Gu, J.H. u., Jameel, F., 2013.
Rational design of lyophilized high concentration protein formulations-mitigating
the challenge of slow reconstitution with multidisciplinary strategies. Eur. J. Pharm.
Biopharm. 85, 287293. https://doi.org/10.1016/j.ejpb.2013.05.001.
Capozzi, L.C., Trout, B.L., Pisano, R., 2019. From Batch to Continuous: Freeze-Drying of
Suspended Vials for Pharmaceuticals in Unit-Doses. Ind. Eng. Chem. Res. 58,
16351649. https://doi.org/10.1021/acs.iecr.8b02886.
Carfagna, M., Rosa, M., Lucke, M., Hawe, A., Frieß, W., 2020. Heat ux sensor to create a
design space for freeze-drying development. Eur. J. Pharm. Biopharm. 153, 8494.
https://doi.org/10.1016/j.ejpb.2020.05.028.
Carpenter, J., Chang, B., 1996. Lyophilization of Protein Pharmaceuticals, in: Avisand,
K., Wu, V. (Eds.), Biotechnology and Biopharmaceutical Manufacturing, Processing,
and Preservation. pp. 199263.
Carpenter, J.F., Prestrelski, S.J., Anchordoguy, T.J., Arakawa, T., 1994. Interactions of
Stabilizers with Proteins During Freezing and Drying, in: Cleland, J.L., Langer, R.
(Eds.), Formulation and Delivery of Proteins and Peptides. American Chemical
Society (ACS), pp. 134147. https://doi.org/10.1021/bk-1994-0567.ch009.
Celik, M., Wendell, S.C., 2010. Spray Drying and Pharmaceutical Applications, in:
Parikh, D.M. (Ed.), Handbook of Pharmaceutical Granulation Technology. Taylor
and Francis, pp. 98125. https://doi.org/10.1201/9780849354953.ch5.
Chan, L.W., Tan, L.H., Heng, P.W.S., 2008. Process Analytical Technology: Application to
Particle Sizing in Spray Drying. AAPS PharmSciTech 9, 259266. https://doi.org/
10.1208/s12249-007-9011-y.
Chang, L., Shepherd, D., Sun, J., Ouellette, D., Grant, K.L., Tang, X., Pikal, M.J., 2005a.
Mechanism of protein stabilization by sugars during freeze-drying and storage:
Native structure preservation, specic interaction, and/or immobilization in a glassy
matrix? J. Pharm. Sci. 94, 14271444. https://doi.org/10.1002/jps.20364.
Chang, L., Shepherd, D., Sun, J., Tang, X., Pikal, M.J., 2005b. Effect of sorbitol and
residual moisture on the stability of lyophilized antibodies: Implications for the
mechanism of protein stabilization in the solid state. J. Pharm. Sci. 94, 14451455.
https://doi.org/10.1002/jps.20363.
Charm, S.E., Wong, B.L., 1981. Shear effects on enzymes. Enzyme Microb. Technol. 3,
111118. https://doi.org/10.1016/0141-0229(81)90068-5.
Chaurasiya, B., Zhao, Y.-Y., 2020. Dry Powder for Pulmonary Delivery: A Comprehensive
Review. Pharmaceutics 13, 31. https://doi.org/10.3390/pharmaceutics13010031.
Chen, Y., Mutukuri, T.T., Wilson, N.E., Zhou, Q. (Tony), 2021. Pharmaceutical protein
solids: Drying technology, solid-state characterization and stability. Adv. Drug Deliv.
Rev. 172, 211233. https://doi.org/10.1016/j.addr.2021.02.016.
Chernysheva, M.G., Badun, G.A., Shnitko, A.V., Petrova, V.I., Ksenofontov, A.L., 2018.
Lysozyme-surfactant adsorption at the aqueous-air and aqueous-organic liquid
interfaces as studied by tritium probe. Colloids Surfaces A Physicochem. Eng. Asp.
537, 351360. https://doi.org/10.1016/j.colsurfa.2017.10.048.
Cicerone, M.T., Soles, C.L., 2004. Fast dynamics and stabilization of proteins: Binary
glasses of trehalose and glycerol. Biophys. J . 86, 38363845. https://doi.org/