Op imising 3D p in ed medica ions o a e diseases: In-line mass
uni o mi y es ing in di ec powde ex usion 3D p in ing
Glo ia Mo a-Cas a˜
no
a,1
, Lucía Rod íguez-Pombo
b,1
, Paola Ca ou-Sen a
b
, Pa icija Januskai e
c
,
Ca los Rial
d,e
, Ca los Bendicho-La illa
d,e
, Ma ia L. Couce
, M´
onica Mill´
an-Jim´
enez
a
,
Isido o Ca aballo
a
, Abdul W. Basi
c,e,*
, Ca men Al a ez-Lo enzo
b
, Al a o Goyanes
b,c,d,e,*
a
Depa men o Pha macy and Pha maceu ical Technology, Facul y o Pha macy, Uni e sidad de Se illa, Se ille 41012, Spain
b
Depa amen o de Fa macología, Fa macia y Tecnología Fa mac´
eu ica, I+D Fa ma (GI-1645), Facul ad de Fa macia, Ins i u o de Ma e iales (iMATUS) and Heal h
Resea ch Ins i u e o San iago de Compos ela (IDIS), Uni e sidade de San iago de Compos ela, San iago de Compos ela 15782, Spain
c
Depa men o Pha maceu ics, UCL School o Pha macy, Uni e si y College London, 29-39 B unswick Squa e, London WC1N 1AX, Uni ed Kingdom
d
FABRX L d., Henwood House, Henwood, Ash o d, Ken TN24 8DH, Uni ed Kingdom
e
FABRX A i icial In elligence, Ca e e a de Escai ´
on, 14, Cu elos, O Sa i˜
nao CP 27543, Spain
Se icio de Neona ología, Unidad de Diagn´
os ico y T a amien o de En e medades Me ab´
olicas Cong´
eni as, Heal h Resea ch Ins i u e o San iago de Compos ela (IDIS),
Hospi al Clínico Uni e si a io de San iago de Compos ela, Uni e sidad de San iago de Compos ela, IDIS, RICORS, CIBERER, Me abERN, Spain
ARTICLE INFO
Keywo ds:
3D p in ed pha maceu icals
Addi i e manu ac u ing
D ug deli e y sys ems, modi ied elease
o mula ions
Ra e me abolic diso de s
Pedia ic p ecision ea men s
Pha ma-inks
ABSTRACT
Bio inidase de iciency is a a e inhe i ed diso de cha ac e ized by bio in me abolism issues, leading o neu o-
logical and cu aneous symp oms ha can be alle ia ed h ough bio in adminis a ion. Th ee-dimensional (3D)
p in ing (3DP) o e s po en ial o pe sonalized medicine p oduc ion o a e diseases, due o i s lexibili y in
designing dosage o ms and con olling elease p o iles. Fo such poin -o -ca e applica ions, igo ous quali y
con ol (QC) measu es a e essen ial o ensu e p ecise dosing, op imal pe o mance, and p oduc sa e y, especially
o low pe sonalized doses in p eclinical and clinical s udies. In his wo k, we add essed QC challenges by
in eg a ing a p ecision balance in o a di ec powde ex usion pha maceu ical 3D p in e (M3DIMAKER™) o
eal- ime, in-line mass uni o mi y es ing, a c i ical quali y con ol s ep. Small and la ge capsule-shaped bio in
p in le s (3D p in ed able s) o immedia e- and ex ended- elease we e p in ed. The in eg a ed balance moni-
o ed and egis e ed each p in le ’s weigh , iden i ying any de ia ions om accep able limi s. While all la ge
p in le ba ches me mass uni o mi y c i e ia, some small p in le ba ches exhibi ed weigh de ia ions. In i o
elease s udies showed la ge immedia e- elease p in le s eleasing 82% o bio in wi hin 45 min, compa ed o
100% o small immedia e- elease p in le s. Fo ex ended- elease o mula ions, 35% o he d ug was eleased
om small p in le s, whe eas 24% was eleased om la ge p in le s a he same ime poin . The in eg a ion o
p ocess analy ical echnology ools in 3DP shows p omise in enhancing QC and scalabili y o pe sonalized dosing
a he poin -o -ca e, demons a ing success ul in eg a ion o a balance in o a di ec powde ex usion 3D p in e
o in-line mass uni o mi y es ing ac oss di e en sizes o capsule-shaped p in le s.
1. In oduc ion
Bio inidase de iciency is an inhe i ed a e disease cha ac e ized by
an au osomal ecessi e diso de o bio in me abolism (Yılmaz, 2024).
Wi h a p e alence o 1 in 60,000, un ea ed pa ien s wi h bio inidase
de iciency may p esen wi h neu ological and/o cu aneous symp oms,
including de elopmen al delay, seizu es, hypo onia, senso ineu al
hea ing loss, op ic a ophy, alopecia, and skin ashes (Tankeu, 2023;
Cowan e al., 2010; Canda e al., 2020). Fo una ely, hese clinical ea-
u es can be imp o ed o p e en ed by adminis e ing pe sonalized doses
* Co esponding au ho s a : Depa men o Pha maceu ics, UCL School o Pha macy, Uni e si y College London, 29-39 B unswick Squa e, London WC1N 1AX,
Uni ed Kingdom (A.W. Basi ); Depa amen o de Fa macología, Fa macia y Tecnología Fa mac´
eu ica, I+D Fa ma (GI-1645), Facul ad de Fa macia, Ins i u o de
Ma e iales (iMATUS) and Heal h Resea ch Ins i u e o San iago de Compos ela (IDIS), Uni e sidade de San iago de Compos ela, San iago de Compos ela 15782,
Spain (A. Goyanes).
E-mail add esses: [email p o ec ed] (A.W. Basi ), [email p o ec ed] (A. Goyanes).
1
These au ho s con ibu ed equally o his wo k.
Con en s lis s a ailable a ScienceDi ec
In e na ional Jou nal o Pha maceu ics
jou nal homepage: www.else ie .com/loca e/ijpha m
h ps://doi.o g/10.1016/j.ijpha m.2024.124964
Recei ed 6 Sep embe 2024; Recei ed in e ised o m 14 No embe 2024; Accep ed 15 No embe 2024
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
A ailable online 16 No embe 2024
0378-5173/© 2024 The Au ho s. Published by Else ie B.V. This is an open access a icle unde he CC BY license (
h p://c ea i ecommons.o g/licenses/by/4.0/ ).
o he wa e -soluble i amin bio in. T ea men is ailo ed o he pa ien ’s
me abolic ac i i y and he e o e equi es con inuous adjus men s: o al
bio in doses o 5–10 mg/day o hose wi h <10 % o a e age no mal
se um enzyme ac i i y, and 2.5–10 mg/day o hose wi h 10 % −30 %
o a e age no mal enzyme ac i i y (Tankeu, 2023; Canda e al., 2020;
Sayegh, 2020).
Th ee-dimensional (3D) p in ing (3DP) is an inno a i e echnology
ha combines compu e -aided design and manu ac u ing o c ea e an
objec h ough successi e laye s o ma e ial (Milliken, 2024; Ca doso
and E.s., 2024). Being able o c ea e objec s wi h complex s uc u es and
geome ies, his echnique is well sui ed o mee he needs o de eloping
pe sonalized dosing o a ious pa ien g oups, including paedia ics
and ge ia ics, polypha macy pa ien s, and indi iduals wi h a e dis-
eases (Pa el, 2024; Huanbu a, 2023; Fas ø, 2019; Januskai e, 2020;
Ca ou-Sen a, 2023; Maza u a e al., 2022; Ahola, 2024; Funk, 2024).
3DP allows o he lexible p oduc ion o small ba ches wi h cus om-
isa ions in dose, shape, d ug elease kine ics, composi ion, and he
inco po a ion o mul iple d ugs in o a single dosage o m (Bonia i,
2021; Mo a-Cas a˜
no e al., 2023; Shojaie e al., 2023; Muhindo, 2023;
Denis, 2024; Windol , 2022). Pha maceu ical 3DP has al eady shown
ad an ages o e con en ional o mula ions in he p oduc ion o
pe sonalized medicines o bioequi alence (Lyousou i, 2023) and clin-
ical s udies a ge ing speci ic popula ions and a e diseases (Goyanes,
2019a,b; Rod íguez-Pombo, 2024a,b; Liu, 2023).
Among he a ious 3DP echnologies, ma e ial ex usion is he mos
commonly used in he pha maceu ical sec o , which includes semisolid
ex usion (SSE), used deposi ion modelling (FDM), and di ec powde
ex usion (DPE), depending on he ype o d ug-loaded ink (pha ma-ink)
used (Algah ani e al., 2018). SSE 3DP, which employs a gel o pas e as
he pha ma-ink and ope a es a low empe a u es, has been in es iga ed
ex ensi ely in biop in ing, pe sonalized medicine, and no el dosage
o ms, such as chewable p in le s (Ca ou-Sen a, 2023; Wang, 2023a,b;
Awad, 2023a,b; Cha zi aki, 2023; U omo, 2023). Addi ionally, he use o
disposable p e- illed sy inges makes SSE close o mee ing he quali y
con ol (QC) equi emen s manda ed by egula o y bodies (Vi hani,
2019).
FDM, in con as , uses a d ug-loaded he moplas ic ilamen as he
pha ma-ink, equi ing a ho mel ex ude (HME) o ilamen manu-
ac u e (Ghanizadeh Tab iz, 2023). Howe e , his wo-s ep he mal
p ocess can deg ade he ac i e pha maceu ical ing edien (API) and
inc ease manu ac u ing ime. The need o ilamen s wi h speci ic
heological and mechanical p ope ies u he limi s i s b oade appli-
ca ion. The e o e, i is necessa y o e alua e he s abili y, quali y, and
beha iou o he ilamen s used in his echnique (Banda i, 2021; Mo a-
Cas a˜
no, 2022; Oladeji, 2022; Xu, 2020; Okwuosa, 2021; Ayyoubi,
2023; Yang, 2023).
DPE has eme ged as an al e na i e o FDM, by di ec ly p in ing
powde o g anula e blends as pha ma-inks wi hou ilamen p epa a-
ion (Zheng, 2021; Aguila -de-Ley a, 2023). The powde blend is ed
di ec ly in o a hoppe , anspo ed o a hea ed nozzle by a single sc ew,
and ex uded in a laye -by-laye manne (Goyanes, 2019a,b). This p o-
cess o e comes some limi a ions associa ed wi h FDM, as i allows o
single-s ep p oduc ion, educing he mal s ess, and is mo e e icien due
o he small quan i y o aw ma e ials needed (Goyanes, 2019a,b; Rosch,
2023; Fanous, 2020). DPE also enables he o ma ion o amo phous solid
dispe sions (ASDs) wi h high d ug loading, known o enhance d ug
solubili y (Bonia i, 2021; Goyanes, 2019a,b; Wang, 2023a,b; Pis one,
2023; Mo a-Cas a˜
no, 2024). Despi e i s ad an ages, DPE p esen s i s
own challenges in he p oduc ion o pha maceu ical dosage o ms, such
as he in luence o powde low, heological p ope ies, and elec os a ic
o ces. Poo powde can hinde con inuous and homogeneous ma e ial
eed h ough he sc ew, causing a iabili y in dosing and limi ing he
uni o mi y o weigh and API con en in he inal p oduc (Bonia i,
2021; Rosch, 2023; Pis one, 2023).
The implemen a ion o 3DP in pha maceu icals mus add ess chal-
lenges ela ed o Good Manu ac u ing P ac ice (GMP), and he quali y o
3D p in ed p oduc s mus be gua an eed h ough igo ous QC measu es
o op imal pe o mance and sa e y (Muhindo, 2023; Rosch, 2023;
Bendicho-La illa, 2024). Mass uni o mi y es ing du ing he p in ing
p ocess is essen ial o e alua ing ep oducibili y and ensu ing accu a e
dosing in each pha maceu ical o m, a c ucial poin in he QC o 3DP
(Rosch, 2023; Deon, 2022). This becomes pa icula ly impo an in he
case o low pe sonalized doses o a e diseases, whe e ea men cus-
omiza ion and p eclinical and clinical s udies demand accu a e weigh
de e mina ion o ensu e e icacy and sa e y (Díaz-To es, 2023; Johan-
nesson, 2023). P ocess analy ical echnology (PAT) ools o e a solu ion
o ensu ing ba ch-wide mass uni o mi y and can be ins alled in-line, a -
line, o -line, o on-line (Seoane-Via˜
no, 2023). Fo example, a p e ious
s udy in eg a ed a balance in o an SSE pha maceu ical 3D p in e , o
non-des uc i e and in-line weigh uni o mi y es ing o hyd oco isone
p in le s (Bendicho-La illa, 2024).
The aim o his wo k was o implemen , o he i s ime, a balance
on a mul i-p in head 3D p in e o moni o he weigh and ep oduc-
ibili y o medicine manu ac u ed by DPE 3DP. Two di e en excipien s
(polye hylene oxide 100,000 and hyd oxyp opyl cellulose) we e
selec ed o manu ac u e capsule-shaped p in le s o immedia e- and
ex ended- elease sys ems, espec i ely. Two p in le ba ch sizes, con-
aining 2.5 and 10 mg o bio in, we e manu ac u ed o po en ial use in
clinical and p eclinical s udies using he heal hca e M3DIMAKER™
S udio so wa e, which con olled bo h he 3D pha maceu ical p in e
and he in-line mass uni o mi y es ing. Finally, physicochemical cha -
ac e iza ion echniques we e used o e alua e he p ope ies o he
de eloped capsule-shaped p in le s, assessing bio in con en and elease
p o iles o he di e en p in le s.
2. Ma e ials and me hods
2.1. Ma e ials
Bio in (MW 244.31 g/mol, Aco a ma, Ba celona, Spain) was used as
he model d ug. Polye hylene oxide 100,000 (PEO, 100,000 Da, Sigma-
Ald ich, S . Louis, USA) and hyd oxyp opyl cellulose (HPC Klucel ELF,
40,000 Da, Ashland, Wilming on, USA) we e he selec ed polyme s o
manu ac u e he pha ma-inks. Pea li ol lash-manni ol (Roque e, Les-
em, F ance) was added o he immedia e- elease pha ma-ink o
imp o e p in le esolu ion and dissolu ion.
Hyd ochlo ic acid (37 %, Ph. Eu , Scha lau, Ba celona, Spain), so-
dium phospha e ibasic dodecahyd a e (≥98.0 %, Honeywell Fluka,
Buchs, Swi ze land), sodium hyd oxide (Ph. Eu . pelle s, VWR In e na-
ional, Radno , Pennsyl ania, USA), and phospha e bu e solu ion (pH
=3.9) we e used o he p epa a ion o he dissolu ion es ing media.
Ace oni ile (≥99.9 % / , HPLC g ade, Me ck, Da ms ad , Ge many)
and i luo ace ic acid (≥99.0 % / , HPLC g ade, Fishe Scien i ic,
Hamp on, USA) we e used as he mobile phase o he d ug con en
assay. All ma e ials we e used as ecei ed.
2.2. P epa a ion o he pha ma-ink
The pha ma-inks used o di ec powde ex usion (DPE) we e made
by weighing ou each componen , con aining 5 % w/w o he model
d ug, bio in. The aw ma e ials we e manually mixed using a mo a and
Table 1
Composi ion o he pha ma-inks o bo h immedia e- and ex ended- elease
sys ems.
Pha ma-ink HPC ELF
(% w/w)
PEO 100,000
(% w/w)
Manni ol (%
w/w)
Bio in (%
w/w)
Immedia e-
elease (IR)
−60 35 5
Ex ended-
elease (ER)
95 − − 5
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
2
pes le. Table 1 shows he composi ion o he pha ma-inks de eloped in
his wo k.
2.3. DPE 3D p in ing p ocess alida ion
The p in ing p ocess was ca ied ou using he M3DIMAKER™ 2
mul i-p in head 3D p in e (FABRX L d., London, UK) wi h he inco -
po a ed balance (Fig. 1). The build pla e consis ed o a glass pla o m
co e ed by blue ape o aid in ilamen adhesion (3 M, Sain Paul,
Minneso a, USA). This was hen placed on he op o he balance o ac as
he p in ing pla o m.
Ini ially, he capsule 3D model was selec ed om he so wa e’s e-
posi o y. The p in ing pa ame e s we e hen selec ed using he
M3DIMAKER S udio™ so wa e. The nex s ep was o choose he 3DP
echnology (in his case, DPE), and he d ug loading o he pha ma-ink
(5 % w/w). Subsequen ly, based on he selec ed p in ing pa ame e s,
he capsule shape model (2.7 mm diame e , 8.6 mm leng h, Fig. 2) was
scaled and p in ed in 4 a ying sizes. The so wa e sen ins uc ions o
he 3D p in e o p in a ba ch o 20 p in le s a anged in 4 ows o 5
p in le s each, wi h a ying weigh s pe ow. The weigh o each p in le
was hen au oma ically measu ed by he in eg a ed balance inside he
3D p in e . The so wa e hen used his in o ma ion o es ablish a “dose
lib a y”, an algo i hm ha c ea es ela ionships be ween he size,
weigh , dose, and p in ing pa ame e s. This “dose lib a y” allows he
use o selec any dose in a simple manne , wi h he so wa e adap ing
he size o he 3D model o manu ac u e he selec ed dose (Rod íguez-
Pombo, 2024a,b).
2.4. 3D p in ing p ocess and mass uni o mi y es ing
The p epa ed pha ma-inks we e added o he hoppe o a pha ma-
ceu ical mul i-p in head 3D p in e , M3DIMAKER™ 2 (FABRX L d.,
London, UK), using he di ec single-sc ew powde ex ude p in head,
wi h a nozzle diame e o 0.4 mm. Two sizes we e p in ed, small and
la ge capsule-shaped p in le s, o ob ain he desi ed doses o bio in (2.5
mg and 10 mg), o be used o p eclinical and clinical s udies,
espec i ely.
The p in ing pa ame e s gene a ed using M3DIMAKER S udio™
so wa e a e highligh ed in Table 2 o bo h bio in o mula ions.
Ba ches o 10 capsule-shaped p in le s we e p in ed and weighed by
he in eg a ed balance o each size and o mula ion, and he so wa e
collec ed all he eco ded weigh s o each ba ch. The in eg a ed balance
weighed each p in le immedia ely a e i was p in ed as epo ed in
p e ious wo k (Bendicho-La illa, 2024). The M3DIMAKER™ 2 mul i-
p in head 3D p in e manu ac u ed he p in le s indi idually. A e
each p in le was comple e, he p in head mo ed up and emained
mo ionless o some seconds o allow he balance o s abilise. The bal-
ance epo ed he weigh measu emen o each p in le o he so wa e,
which displayed he alue in he espec i e posi ion o he p in le . The
balance was hen au oma ically a ed and he nex p in le was p in ed.
2.5. Cha ac e iza ion o he capsule-shaped p in le s
The dimensions (leng h and diame e ) o he p in le s (n =10) in
each ba ch we e measu ed using a digi al callipe (Me ce Type 130,
G&M Tools and Equipmen L d, Ca lisle, UK).
2.5.1. X- ay powde di ac ion (XRPD) analysis
The di ac og ams o pu e powde samples (bio in, HPC ELF, PEO
100,000, and manni ol), he pha ma-ink, and he p in le s we e ob-
ained in a D8 Ad ance di ac ome e (B uke , Bille ica, MA, USA)
equipped wi h a sealed X- ay ube ((CuK
α
1 (λ =1.5406 Å)) and a
LYNXEYE- ype de ec o . The in ensi y and ol age applied du ing he
assay we e 40 mA and 40 kV, espec i ely. The di ac og ams we e
ob ained in he 2θ angula ange o 3 – 50◦wi h a s ep o 0.04◦and a
coun ing ime o 4 s pe s ep. Samples we e deposi ed on an o ien ed Se
(511) pla e o a oid sca e ing noise and we e spun o minimize he
e ec o he p e e en ial o ien a ion.
2.5.2. The mal analysis
Di e en ial scanning calo ime y (DSC) was u ilized o unde s and
he he mal beha iou o he pu e excipien s (HPC ELF, PEO 100,000,
and manni ol) and d ug (bio in), he pha ma-inks, and p in le s. The -
mog ams we e ob ained using a DSC Q100 (TA Ins umen s, New Cas le,
DE, USA) wi h a e ige a ed cooling accesso y a a hea ing a e o
10 ◦C/min. The empe a u e ange was 0 – 300 ◦C and ni ogen was used
as he pu ge gas ( low a e =50 mL/min). All measu emen s we e
pe o med using non-he me ic aluminium pans, in which 2 – 5 mg o
sample was accu a ely weighed using he p ecision balance (0.0001 mg)
o TGA55 Disco e y se ies equipmen (TA Ins umen s, New Cas le, DE,
Fig. 1. Image o he balance-3D p in e M3DIMAKER™ 2 sys em.
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
3
USA).
The mal G a ime y Analysis (TGA) was used o in es iga e he
he mal s abili y o all o mula ion componen s. The he mal beha iou
o he pu e HPC Klucel ELF and bio in, he pha ma-inks, and he p in le
we e e alua ed. TGA measu emen s we e ca ied ou using a TGA55
Disco e y se ies (TA Ins umen s, New Cas le, DE, USA) wi h a e ig-
e a ed cooling accesso y a a hea ing a e o 10 ◦C/min. The empe a u e
ange was 25–350 ◦C and ni ogen was used as he pu ge gas, a a low
a e o 50 mL/min. All expe imen s we e pe o med using non-he me ic
aluminium pans, in which 3 – 9 mg o sample was accu a ely weighed
using he p ecision balance (0.0001 mg) o he same equipmen .
All da a om DSC and TGA expe imen s was collec ed wi h he TA
Ad an age so wa e o Q se ies ( e sion 2.8) and p ocessed using TA
Ins umen s Uni e sal Analysis 2000 ( e sion 4.5.0.5).
2.5.3. Fou ie - ans o m in a ed spec oscopy (FTIR)
The a enua ed o al e lec ance Fou ie ans o m in a ed (ATR-
FTIR) spec a o pu e excipien s and bio in, he pha ma-inks, and
p in le s we e collec ed using a Va ian 670 FTIR spec ome e (Va ian
Inc., Palo Al o, USA). All samples we e scanned be ween 4000 and 400
cm
−1
a a esolu ion o 4 cm
−1
o 32 scans.
2.5.4. Scanning elec on mic oscopy (SEM) imaging
SEM imaging was pe o med o isualize he p in le su ace and he
in e nal s uc u e a e cu ing. Bo h IR and ER p in le s we e cu in hal
and placed on me al suppo s, be o e being analysed using a Zeiss EVO
LS15 scanning elec on mic oscope equipped wi h a backsca e ed
elec on de ec o (Obe kochen, Ge many), ope a ing a 20 kV.
2.6. D ug loading de e mina ion
Samples o he di e en p in le s (la ge and small) we e placed in a
beake (50 mL o small p in le s and 100 mL o la ge p in le s) wi h
Milli-Q® wa e and we e subjec ed o magne ic s i ing o e nigh .
Sample solu ions we e hen il e ed h ough 0.22 µm il e s (Millipo e
L d., Dublin, I eland) and he concen a ion o bio in was de e mined
wi h a HPLC-UV LC-1100 Se ies HPLC sys em (Agilen Technologies,
San a Cla a, USA). The assay en ailed injec ing 30 µL samples o anal-
ysis using a mobile phase o ace oni ile and wa e (25:75 / ), con-
aining 0.05 % / i luo o ace ic acid h ough a Symme y 5 µm C18
column, 4.6 mm ×250 mm (Wa e s, Mil o d, Massachuse s), main-
ained a 30 ◦C. The mobile phase was pumped a a low a e o 1.0 mL/
min and he eluen was sc eened a a wa eleng h o 210 nm. The
e en ion ime was 3.8 min, and he concen a ion ange was 0.50–120
µg/mL.
2.7. In i o elease s udies
Bio in elease p o iles o he la ge capsule-shaped p in le s we e
e alua ed using an SR8-Plus Dissolu ion Tes S a ion (Hanson Resea ch,
Cha swo h, CA, USA) wi h USP-II appa a us. The bio in elease p o iles
o he small capsule-shaped p in le s we e e alua ed using an ERWEKA
DT 126 ligh se ies (ERWEKA GmbH, Langen, Ge many) wi h USP-II
appa a us (moun ed wi h mini paddles). The speed o bo h paddles
was se a 100 pm wi h a empe a u e o 37±0.5 ◦C.
Fo he in i o immedia e- elease s udies, he la ge p in le s we e
d opped in 1000 mL o 0.1 M HCl (pH =1.2) o 2 h o simula e human
gas ic condi ions o clinical s udies. The small p in le s we e d opped
in 100 mL o phospha e bu e pH 3.9 o 2 h o simula e a gas ic
condi ions o p eclinical s udies.
Fo he in i o ex ended- elease s udies, he la ge p in le s we e
d opped in 750 mL o 0.1 M HCl o 2 h o simula e gas ic condi ions.
A e 2 h, 250 mL o isodium phospha e solu ion (0.2 M) was added
in o each essel and he pH was adjus ed o 6.8 using 1 M NaOH o 1 M
HCl solu ions o simula e human in es inal condi ions. The small
p in le s we e d opped in 100 mL o phospha e bu e pH 3.9 o 2 h o
simula e a gas ic condi ions, ollowed by adjus men o he pH o 6.5
wi h 5 M NaOH, simula ing a in es inal condi ions.
5 mL o sample was manually wi hd awn om each essel and
immedia ely eplaced wi h esh medium, il e ed h ough 0.22
μ
m il-
e s (Millipo e L d., Dublin, I eland), and analysed using HPLC o
de e mine he amoun o d ug eleased (me hod desc ibed p e iously).
Tes s we e conduc ed in iplica e unde sink condi ions. Da a we e e-
po ed h oughou as mean ±s anda d de ia ion (n =3).
3. Resul s and discussion
3.1. 3D p in ing p ocess and in-line mass uni o mi y es ing
Two di e en sizes o bo h IR and ER p in le s we e selec ed o
e alua e, o he i s ime, he easibili y o an in eg a ed balance and
mul i-p in head 3D p in e sys em o ca y ou he in-line mass uni o -
mi y assay o DPE 3DP. The mul i-p in head 3D p in e sys em (Fig. 1)
Fig. 2. Images o he design o he capsule-shaped p in le model (2.7 mm diame e , 8.6 mm leng h).
Table 2
P in ing pa ame e s used du ing he p in ing p ocess o bo h IR and ER
p in le s.
P in ing
pa ame e
IR small size
p in le
ER small size
p in le
IR la ge size
p in le
ER la ge size
p in le
Fill pa e n Rec ilinea Rec ilinea Rec ilinea Rec ilinea
In ill (%) 50 50 50 50
Ra and b im No No No No
Ex usion speed
(mm/s)
10.5 5.25 16.5 8.25
T a el speed
(mm/s)
10 10 10 10
Fi s laye
heigh (mm)
0.30 0.15 0.30 0.15
Laye heigh
(mm)
0.30 0.30 0.30 0.30
Ex usion
empe a u e
(◦C)
100 170 100 170
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
4
was insula ed using wo polyca bona e co e doo s o mi iga e any
weigh measu emen luc ua ions esul ing om ai mo emen . Addi-
ionally, he balance was placed benea h he p in ing pla o m, which
was i ed wi h an an i- ib a ion pla e o p e en weigh a ia ions due
o ib a ions p oduced du ing he p in ing p ocess. The M3DIMAKER
S udio™ so wa e was con igu ed o con ol bo h he 3D p in e and he
in-line mass uni o mi y es . Fi s , a alida ion p ocess was conduc ed o
op imise he p in ing pa ame e s o each pha ma-ink, allowing he
so wa e o c ea e a dose lib a y ha linked he 3D scale models o
di e en weigh s and doses. The ini ial capsule-shaped 3D model (Fig. 2)
was hen scaled o p oduce ou di e en sizes. The mean weigh pe ow
(n =5) was eco ded, and he so wa e es ablished co ela ions be ween
he model size, weigh , dose and p in ing pa ame e s. This alida ion
s ep allows use s o selec any weigh o dose wi hin he alida ed ange,
as M3DIMAKER S udio™ will au oma ically selec he co esponding
ile om he dose lib a y.
The so wa e-con olled balance was success ully in eg a ed in o he
pha maceu ical mul i-p in head 3D p in e , enabling he p epa a ion o
up o h ee dosage o ms simul aneously and conduc ing in-line mass
uni o mi y es ing in he same p ocess. This s udy ex ends ou p e ious
wo k ela ed o he in-line mass uni o mi y es o SSE 3DP by allowing
up o h ee DPE p in heads o be used a he same ime and e alua ing a
di e en 3DP echnology (Bendicho-La illa, 2024). The capsule shape
was chosen o i s a ou able swallowabili y and ease o adminis a ion
in bo h p eclinical and clinical s udies (Goyanes, 2017). Howe e , his
shape poses a challenge o DPE, because he i s laye s a e composed o
a single line ha suppo s he en i e s uc u e. This may cause he objec
o mo e du ing p in ing due o he ins abili y o he s uc u e i he
pha ma-ink does no adhe e well o he p in ing pla o m.
HPC was selec ed o ex ended- elease p in le s due o i s good me-
chanical s eng h, s abili y ac oss a wide pH ange, and i s es ablished
use as a ma ix in con olled- elease d ug deli e y sys ems (Rowe e al.,
2009). HPC was also chosen since i s a ou able p in abili y has been
success ully demons a ed in o he s udies wi h DPE 3DP (Zheng, 2021;
S´
anchez-Gui ales, 2021; Mendibil, 2021). PEO 100,000 was selec ed o
immedia e- elease p in le s due o i s high wa e solubili y and p io use
in ex usion based 3DP o p oduce o odispe sible ilms and immedia e-
elease able s (Jenno e, 2023; Racaniello, 2023; Cho, 2020).
Each size and o mula ion ba ch (n =10) was success ully p in ed
using he M3DIMAKER™ 2 (Fig. 3). The pha ma-inks lowed smoo hly
h ough he single-sc ew DPE sys em and adhe ed well o he build pla e
wi hou he need o hea ing he p in ing pla o m. Mo eo e , he
p in le s we e easily emo able om he build pla e, e aining hei
shape and lea ing no esidual ma e ial.
All p in le s we e indi idually weighed in-line and epo ed by he
M3DIMAKER S udio™ so wa e du ing he p in ing p ocess (Fig. 4).
Addi ionally, dimensional (diame e and leng h) measu emen s o
he capsule-shaped p in le s o each ba ch we e aken (Table 3). The
su ace a ea/ olume (SA/V) a io was also calcula ed.
Acco ding o he Eu opean Pha macopeia (Ph. Eu .) (Eu ope, C.o.
2.9.5, 2024), no mo e han 2 o he 20 indi idual masses in a ba ch
should de ia e om he a e age mass by mo e han ±10 % o small
p in le s o ±7.5 % o la ge p in le s, wi h none de ia ing by mo e han
wice hese pe cen ages. The indi idual weigh s o each p in le om
each ba ch a e shown in Fig. 5.
As obse ed in Fig. 5B and D, all la ge p in le ba ches me he
accep ed mass uni o mi y limi s, acco ding o he c i e ia ha no mo e
han 2 o he 20 indi idual masses in a ba ch may de ia e om he
a e age mass by mo e han ±10 % o small p in le s o ±7.5 % o la ge
p in le s, wi h none de ia ing by mo e han wice hese pe cen ages.
Howe e , some o he small p in le s (Fig. 5A and C) ell ou side he
accep ed weigh limi s, ende ing hem non-complian wi h he mass
uni o mi y c i e ia. This is he case o he ER small p in le ba ch, whe e
ou p in le s de ia ed om he a e age heo e ical weigh . This is
consis en wi h he highe coe icien o a ia ion (CV) in weigh s o he
small p in le s, which was double ha o he la ge p in le s. This could
be a ibu ed o he smalle dimensions o he small capsule-shaped
p in le s and he nozzle diame e used (Bendicho-La illa, 2024). Using
a smalle nozzle diame e could po en ially esol e his issue. I should
be highligh ed ha he na owes accep ed weigh ange applies o he
small p in le ba ches, indica ing ha e en mino de ia ions in weigh
a e de ec ed as ou o limi s by he M3DIMAKER S udio™ so wa e.
These indings highligh he po en ial o in eg a ing a balance as a
weigh con ol sys em wi hin he DPE 3D p in e . Using he
M3DIMAKER S udio™ so wa e, weigh de ia ions om he a e age can
be iden i ied in eal- ime du ing he 3DP p ocess, which can accele a e
he adi ionally ime-in ensi e QC assessmen s essen ial o solid
dosage o ms. In a ela ed s udy, Bendicho-La illa e al. success ully
in eg a ed an online analy ical balance in o an SSE pha maceu ical
p in e , au oma ing mass uni o mi y es ing (Bendicho-La illa, 2024).
Building on his, we aimed o u he ad ance QC by implemen ing an
analy ical balance in a mul i-p in head DPE pha maceu ical p in e .
This mul i-p in head capabili y allows simul aneous use o mul iple DPE
p in heads, wi h each p in le weighed in eal- ime.
Fig. 3. Images o he capsule-shaped IR p in le s: (A) small and la ge size; (B) comple e la ge ba ch p in ed on o he p in ing pla o m. Images o he capsule-shaped
ER p in le s: (C) small and la ge size; (D) comple e la ge ba ch p in ed on o he p in ing pla o m. Scale is in cm.
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
5
The p esen wo k o e s signi ican scien i ic con ibu ions by
add essing he echnical challenges o in eg a ing a mul i-p in head DPE
p in e wi h an analy ical balance. Unlike SSE, which is easie o pai
wi h a balance due o gen le ope a ing condi ions, such as minimal
ib a ions, no su ace con ac , and no an equi emen , DPE p esen s a
mo e demanding se up. Speci ically, DPE equi es smalle laye heigh s,
leading o equen con ac wi h he balance su ace and addi ional i-
b a ions. Addi ionally, he an on he DPE p in head mus be in e mi -
en ly u ned o du ing balance a ing and weighing, while a oiding
p olonged down ime o p e en ma e ial clumping in he hoppe . This
s udy ep esen s a meaning ul ad ance in QC o medicines p oduced
using DPE echnology.
3.2. Cha ac e iza ion o he capsule-shaped p in le s
3.2.1. The mal beha iou and XRPD analysis
DSC and XRPD analyses we e pe o med o de e mine he solid-s a e
s uc u e o bio in in he pha ma-inks be o e p in ing and in he inal
p in le s. Pu e bio in exhibi ed a mel ing poin o 232 ◦C in he i s
hea ing cycle, as obse ed in Fig. 6A. The absence o his endo he mic
peak in he DSC he mog ams o he pha ma-inks and p in le s sugges s
ha bio in may be amo phous and solubilised wi hin he 3D p in ed
capsule-shaped p in le s du ing he excipien mel ing p ocess. Al e na-
i ely, bio in may ha e emained c ys alline and was only solubilized
du ing he DSC assay. These indings a e in conjunc ion wi h he XRPD
esul s (Fig. 7), which showed c ys alline peaks o bio in, in which no
hea was applied du ing he analysis, sugges ing solubilisa ion o bio in
wi hin he excipien s du ing he hea ing in he DSC pan. PEO and
manni ol exhibi ed endo he mic peaks a 65 ◦C and 170 ◦C, espec i ely.
As expec ed, he immedia e- elease pha ma-ink and p in le showed
bo h endo he mic peaks a ibu ed o PEO and manni ol.
To assess bio in s abili y a he p in ing empe a u e, TGA was pe -
o med. The highes p in ing empe a u e used in his wo k was o he
ER pha ma-ink (170 ◦C), and acco ding o he TGA esul s (Fig. 6B), pu e
bio in had a deg ada ion onse o 250 ◦C, well abo e he maximum
p in ing empe a u e. The obse ed weigh loss in he pha ma-ink
be o e p in ing and in he p in le could be a ibu ed o some esidual
mois u e loss. The e o e, he p in ing empe a u es used in his wo k
(100 ◦C and 170 ◦C o IR and ER pha ma-inks, espec i ely) did no lead
o any bio in deg ada ion, which was la e con i med wi h HPLC
analysis.
As shown in he x- ay di ac og ams (Fig. 7), o bo h ER and IR
pha ma-inks, he c ys allini y peaks o bio in can be obse ed. Peaks a
16◦and 17◦could be obse ed in p in le IR and he in ensi y was highe
compa ed o pha ma-ink IR. A peak a 22◦is obse ed in bo h he
pha ma-ink ER and p in le ER. These indings sugges ha pa o he
bio in emains in a c ys alline s a e wi hin he p in le s. Al hough
c ys alline bio in was no obse ed in he DSC analysis, i was con i med
by he x- ay di ac og ams. Toge he , he DSC and XRPD esul s indi-
ca e ha a po ion o he bio in may be solubilized in he polyme ma-
ix, while some emains in he c ys alline o m. In addi ion, PEO
cons i u ed he majo excipien in he IR pha ma-ink (60 % w/w), which
may o e lap wi h o he bio in peaks in he IR p in le di ac og am,
po en ially obscu ing some o he bio in signals.
3.2.2. FTIR analysis
FTIR was pe o med o elucida e any possible chemical in e ac ions
be ween he d ug and he excipien s du ing he p in ing p ocess (Fig. 8).
The spec um sugges s ha he e we e no chemical in e ac ions be-
ween he d ug and he excipien s. All he bands p esen in he pha ma-
ink and p in le s can be a ibu ed o he p esence o HPC Klucel ELF and
PEO due o he highe pe cen age in he o mula ion (95 % w/w o HPC
Klucel ELF in ER samples and 60 % w/w o PEO in IR samples). I is
possible ha he d ug peaks ha e o e lapped wi h he cha ac e is ic
bands o he HPC Klucel ELF and PEO.
3.2.3. SEM imaging
The images ob ained om SEM mic oscopy allowed obse a ion o
he su ace and he c oss-sec ions o he p in le s (Fig. 9).
Fig. 9B and D gi e isual ep esen a ions o he di e en laye s in he
Fig. 4. Sc eensho o he M3DIMAKER S udio™ so wa e, showing he weigh s ob ained in-line du ing he 3DP p ocess using he in eg a ed balance o (A) small IR
p in le s and (B) la ge IR p in le s. P in le s wi hin he accep ed limi s o he mass uni o mi y assay a e shown in g een. P in le s ou side o he accep ed limi s would
be shown in ed.
Table 3
Dimensions and weigh esul s o each p in ed ba ch. Resul s a e shown as mean ±s anda d de ia ion (n =10). CV is e e ed o as he Coe icien o Va ia ion.
Ba ch Diame e (mm) Leng h (mm) SA/V a io (mm
−1
) In eg a ed balance weigh (mg) CV in eg a ed balance weigh (%)
IR small 3.11±0.06 7.97±0.17 1.62±0.03 57.00±4.00 6.21
IR la ge 4.60±0.10 12.89±0.10 1.07±0.02 197.00±7.00 3.74
ER small 2.71±0.07 8.71±0.13 1.79±0.04 62.00±6.00 10.04
ER la ge 4.51±0.11 13.93±0.09 1.08±0.02 194.00±11.00 5.87
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
6
p in le s. In Fig. 9D, he laye s o p in le ER a e less de ined and his can
be a ibu ed o he i s p in ed laye which adhe es o he p in ing
pla o m and is squashed unde he weigh o he p in le .
As depic ed in Fig. 9A and C, he c oss-sec ions o bo h p in le s a e
homogeneous and smoo h, bu some whi e poin s can be obse ed.
Those poin s a e likely a ibu ed o bio in, as indica ed by XRPD esul s,
which sugges ed he p esence o bio in in i s c ys alline s a e.
3.3. D ug loading in he p in le s
The d ug loading o he di e en ba ches was close o he heo e ical
d ug loading o 5 % w/w (Table 4). The d ug loading o he ER p in le s
was sligh ly lowe han he heo e ical alue. This educ ion in d ug
loading may be a ibu ed o incomple e ex ac ion be o e HPLC quan-
i ica ion. This ensu es ha he e was no deg ada ion du ing he p in ing
p ocess, as expec ed by he TGA esul s.
3.4. In i o elease s udies
Bio in elease om ER and IR p in le s is shown in Fig. 10. I was
obse ed ha he p in le s e oded du ing he dissolu ion s udies and he
d ug was p og essi ely eleased om he dosage o ms as hey e oded. I
was obse ed ha he elease om he small p in le s was as e han he
la ge p in le s due o he dosage o m size, independen o he compo-
si ion. Each dosage o m has a di e en su ace a ea o olume a io
(SA/V) because o he di e en dimensions (Table 3), esul ing in a
sligh ly di e en d ug elease a e.
Fo he IR p in le s, 82 % and 100 % o he bio in was eleased wi hin
45 min o he la ge and small p in le s, espec i ely (Fig. 10A). Ac-
co ding o he Ph. Eu ., con en ional IR dosage o ms should elease a
leas 75 % o he ac i e subs ance wi hin a speci ied ime, ypically 45
min o less (Eu ope, E.C.o. 5.17.1, 2024). The e o e, i was demon-
s a ed ha PEO-based p in le s me he c i e ia o IR dosage o ms,
ega dless o he bio in’s solid-s a e. As shown in he XRPD esul s, pa
o he bio in emained in he c ys alline o m; howe e his did no a ec
Fig. 5. In-line egis e ed weigh s by he in eg a ed balance du ing he p in ing p ocess: (A) small IR p in le ba ch, (B) la ge IR p in le ba ch, (C) small ER p in le
ba ch, and (D) la ge ER p in le ba ch (n =10). The ed lines show he a e age weigh o each ba ch (57 mg and 197 mg o he small and la ge size IR p in le s,
espec i ely; and 62 mg and 194 mg o small and la ge size ER p in le s, espec i ely). The g een dashed lines ep esen he accep ed pe cen age o de ia ion o he
a e age mass (±10 % and ±7.5 % o de ia ion o small and la ge p in le s, espec i ely), pe cen ages a e di e en depending on a e age mass acco ding o he
co esponding Ph.Eu monog aph (Eu ope, C.o. 2.9.5. Uni o mi y o mass os single-dose p epa a ions. Eu opean Pha macopoeia, 2024). The o ange dashed lines
ep esen wice he accep ed pe cen age de ia ion om he a e age mass.
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
7
he dissolu ion o he IR o mula ion, as he p in le s complied wi h Ph.
Eu . Speci ica ions o IR dosage o ms (a leas 75 % o he d ug eleased
wi hin 45 min).
Fo he ER o mula ions, 35 % and 24 % o he bio in was eleased a
45 min o he small and la ge p in le s, espec i ely (Fig. 10B). A 120
min, 100 % o he bio in was eleased om he small p in le s, while 91
% was eleased om he la ge p in le s by 210 min. Based on he ob-
ained esul s, he HPC p in le s complied wi h he ER dosage o m
equi emen s.
A 120 min, he pH was modi ied o he ER p in le s. As obse ed in
Fig. 10B, he d ug elease a e o he la ge p in le s was no a ec ed by
he pH inc ease, wi h 59 %, 78 %, 83 % and 91 % o he bio in eleased
a 120, 150, 180, and 210 min, espec i ely. The e o e, he pH change
did no impac he ex ended- elease p o ile. In con as , he small
p in le s eleased 100 % o he bio in a 120 min, so he pH change was
no signi ican .
Fig. 6. (A) DSC he mog ams o pu e bio in and excipien s, pha ma-inks, and p in le s; (B) TGA he mog ams o pu e bio in and HPC Klucel ELF, pha ma-ink be o e
p in ing and p in le ER.
Fig. 7. X- ay powde di ac og ams o pu e bio in and excipien s, pha ma-inks be o e p in ing and p in le s.
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
8
The balance-3D p in e sys em accele a ed, in e ms o ime and
human esou ces, he mass uni o mi y assay which is manda o y o
solid dosage o ms acco ding o a ious pha macopoeias. This wo k
success ully demons a ed he easibili y o a poin -o -ca e
manu ac u ing model, whe e pha ma-inks a e p epa ed by a GMP-
complian indus y pa ne , and QC is ensu ed h ough an in-line PAT
ool, speci ically he in eg a ed balance.
Bo h he 3DP p ocess and he p in le weigh measu emen s we e
con olled by a single so wa e sys em, acili a ing inc eased
h oughpu . Fu he in eg a ion o addi ional PAT ools could enable a
mo e comp ehensi e analysis o p in ed d ug p oduc s, ensu ing hey
mee necessa y C i ical Quali y A ibu es (CQAs). Fo example, he
in eg a ion o a p essu e senso in o an SSE 3D p in e enabled he
moni o ing o he p in ing p ocess whils simul aneously cha ac e izing
he heological p ope ies o he pha ma-inks (Díaz-To es, 2022).
Addi ionally, an in-line nea -in a ed (NIR) spec oscopy sys em se ed
as a PAT ool o quan i y he d ug load o he p in le s (Seoane-Via˜
no,
2023).
4. Conclusions
This s udy in es iga ed, o he i s ime, he implemen a ion o an
in-line and so wa e-con olled analy ical balance wi hin a pha ma-
ceu ical mul i-p in head DPE p in e . This ep esen s a signi ican
ad ancemen in QC o he pha maceu ical DPE 3D p in ing p ocess,
pa icula ly o low dose d ug deli e y sys ems, ea men pe sonaliza-
ion, and p eclinical and clinical s udies o a e diseases. Mo eo e , wo
di e en pha ma-inks based on PEO (immedia e- elease excipien ) and
HPC (ex ended- elease excipien ) we e de eloped o p in ing capsule-
shaped p in le s using DPE o he i s ime. Au oma ed mass uni o -
mi y es ing was conduc ed on di e en ba ches (n =10) o DPE 3D
p in ed bio in o mula ions in wo di e en sizes. The specialised in e-
g a ed so wa e success ully egis e ed he weigh o each p in le and
de ec ed any de ia ions om he accep able limi s. The mass uni o mi y
esul s indica ed ha he p in ing p ocess was mo e eliable o
manu ac u ing la ge capsule-shaped p in le s compa ed o small ones.
Howe e , u he s udies and imp o emen s a e necessa y o ensu e
consis en DPE p in ing o small capsule-shaped p in le s as well.
Fig. 8. FTIR spec a o he excipien s, pu e bio in, pha ma-inks, and p in le s.
G. Mo a-Cas a˜
no e al.
In e na ional Jou nal o Pha maceu ics 668 (2025) 124964
9
