N

L

ID

ARRAA

KPLST

(smboifsI(ffetaawcfpd2a

E

0h

International Journal of Pharmaceutics 453 (2013) 648– 650

Contents lists available at SciVerse ScienceDirect

International Journal of Pharmaceutics

journa l h o me pag e: www.elsev ier .com/ locate / i jpharm

ote

ong term stability of lyophilized plasmid DNA pDERMATT

ris van der Heijden ∗, Jos H. Beijnen, Bastiaan Nuijenepartment of Pharmacy and Pharmacology, Slotervaart Hospital/The Netherlands Cancer Institute, Amsterdam, The Netherlands

a r t i c l e i n f o

rticle history:eceived 7 May 2013eceived in revised form 10 June 2013ccepted 10 June 2013vailable online 19 June 2013

eywords:

a b s t r a c t

In this short note we report on the shelf-life stability of pDERMATT (plasmid DNA encoding recombi-nant MART-1 and tetanus toxin fragment-c) 2 mg lyophilized powder for reconstitution for intradermaladministration, used in an in-house, investigator-initiated clinical phase I study. pDERMATT was storedat 25 ◦C/60% relative humidity (6 months), 2–8 ◦C (24 months), and −20 ◦C (66 months) in the dark andanalyzed at several timepoints during the conduct of the clinical study for appearance, identity, purity(plasmid topology), content and residual water content. pDERMATT appeared stable at all storage con-

lasmid DNAyophilizationtabilityopology

ditions for the periods tested which, although patient inclusion in the study was significantly delayed,ensured the clinical supply needs. This study shows that lyophilization is an useful tool to preserve thequality of the pDNA and can prevent the need for costly and time-consuming additional manufactureof drug product in case of study delays, not uncommon at the early stage of drug development. To ourknowledge, this is the first study reporting shelf life stability of a pDNA formulation for more than 5 years.

An important aspect of investigational medicinal productsIMPs) used in clinical studies is their shelf-life. Often, at an earlytage of development (phases I and II), costs of (biological) IMP-anufacture are high and patient inclusion rates are prognosted

ut not guaranteed. In particular investigator-initiated studies,ften limited-funded, may run into problems as a result of extendednclusion periods and limited stability of the IMP of interest. There-ore, the development of an IMP drug product with a substantialhelf-life already at this early stage of development can be crucial.n this short note we report on the shelf-life stability of pDERMATTplasmid DNA Encoding Recombinant MART-1 and tetanus toxinragment-c), an in-house developed anti-cancer vaccine encodingor the MART-1 melanoma associated antigen (Fig. 1, Van Oijent al., 2004) and used in a phase I clinical trial investigating theoxicity and efficacy inducing tumor-specific T cell immunity indvanced-stage melanoma patients using intradermal tattooings route of administration (Verstrepen et al., 2008). Previously,e reported on the development of a GMP-manufacturing pro-

ess of pDERMATT and the development of a lyophilized dosageorm, using sucrose as lyoprotectant (Quaak et al., 2008). We storedDERMATT 2 mg lyophilized powder for reconstitution for intra-

ermal use at 25 ± 2 ◦C/60 ± 5% relative humidity (RH) (6 months),–8 ◦C (24 months), and −20 ◦C (66 months) in the dark, andnalyzed the drug product with our stability indicating assay

∗ Corresponding author at: Slotervaart Hospital, Pharmacy, Louwesweg 6, 1066C Amsterdam, The Netherlands. Tel.: +31 20 5124733; fax: +31 20 5124753.

E-mail address: Iris.vanderHeijden@slz.nl (I. van der Heijden).

378-5173/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.ijpharm.2013.06.010

© 2013 Elsevier B.V. All rights reserved.

method at several time points during the clinical study for appear-ance, identity, purity, content and residual water content asdescribed earlier (Quaak et al., 2008, 2009). Moreover, plasmidtopology (supercoiled (SC), open circular (OC), and linear isoforms)was assessed in time using anion exchange high performance liq-uid chromatography (AEX) and capillary gel electrophoresis (CGE)analysis (Quaak et al., 2009). To our knowledge, this is the first studyreporting shelf life stability of a pDNA formulation for more than 5years.

Table 1 presents the stability endpoints of the different stor-age conditions with test items and release specifications. Apartfrom the residual water content, for all three storage conditionsno significant changes in appearance, identity, purity, and con-tent were observed during the respective test periods and stillconformed to the initial release specifications. Dependent on thestorage condition, the residual water content increases in time,with the highest content reached at the highest temperature condi-tion (+25 ± 2 ◦C/60 ± 5% RH). This limited increase in water contentis common for lyophilized products and is caused by water sorptionfrom the rubber closures of the vials, followed by moisture uptake ofthe hygroscopic freeze-dried cake (Vromans et al., 1992; Corveleynet al., 1997). More important, however, is that this increase in watercontent does not affect product quality during the periods testedat the respective storage conditions (Table 1). No aggregation as aresult of the residual moisture increase was seen as earlier reported

by Sharma and Klibanov (2007).

As for the plasmid topology, AEX-analysis showed only minor,non-significant variation in supercoiled and open circular (SC/OC)contents during the period and storage conditions tested, with no

I. van der Heijden et al. / International Journal

F1

amp(OistOeedc

atcpfaer

F2o

ig. 1. Plasmid DNA map of pDERMATT (plasmid DNA encoding recombinant MART- and tetanus toxin fragment-c) including selected restriction sites.

ppearance of linear isoforms (Fig. 2A). Additional CGE measure-ents gave comparable results with identical topology-profiles for

DERMATT at time zero compared to the stability study endpointsFig. 2B). Also AGE measurements indicate presence of both SC andC species at all time points (see Table 1 and Fig. 3.). This is an

mportant stability aspect, as degradation of pDNA can result in ahift in topoisoform distribution by disruption of the tertiary struc-ure of the molecule, resulting in a conversion of SC pDNA into theC or linear topoisoform. As SC pDNA is considered to be mostfficient and results in the highest transfection efficiency (Remautt al., 2006; Urthaler et al., 2005; Walther, 2003; Quaak et al., 2009),egradation to other isoforms can potentially influence clinical out-ome.

Indeed, several authors have reported on pDNA stability andlthough quite stable during production, storage and administra-ion (Schleef, 2005), it undergoes various types of stress whichan lead to degradation (Lengsfeld and Anchordoquy, 2002). Nakedlasmid DNA in solution stored at 2–8 ◦C undergoes a conversion

rom SC topology into OC topology within a week. Additionally,fter prolonged storage the OC pDNA was converted into lin-ar pDNA. This was prevented by storage at −20 ◦C, howeverepeated freeze thaw cycles did also affect the topology (Quaak

Storage time (months)0 20 40 60

Am

ount

/ %

0

20

40

60

80

100

SC OSC OSC O

A

ig. 2. pDERMATT topology in time at different storage conditions: (A) AEX was used5 ± 2 ◦C/60% RH, 2–8 ◦C, and −20 ◦C, all in the dark over a period of 6, 24, and 66 months rf pDERMATT plasmid DNA at t = 0 months (a) and t = 66 months (stored at −20 ◦C) (b).

of Pharmaceutics 453 (2013) 648– 650 649

et al., 2008). Similar findings have also been described by Walther(2003) who showed a rapid conversion of SC forms into OC formsafter 6 months of storage of pDNA in solution at 4 ◦C and theappearance of the linear form after 9 months, which is a strongindicator of pDNA instability. These conversions did not occurwhen stored at −80 ◦C. No extensive studies on the long termstability of lyophilized naked pDNA are reported other than forpDNA formulated into nanomedicines (e.g., pDNA-polycation/lipidcomplexes, gold particles) and during limited periods of time,mostly as a part of accelerated stability studies (Molina et al.,2008; Anchordoquy et al., 2005; Allison et al., 2000; Cherng et al.,1997; Zelphati et al., 1998). Reports on freeze-dried pDNA mostlyrefer to continued degradation in the dried state and damage thatincreasingly accumulates even during storage at lower tempera-tures, albeit al slower rates. Stability data of storage studies ofnaked pDNA and pDNA-based therapeutics range from 3 weeksto about a year at several storage conditions varying from −20 ◦Cto 75 ◦C (Costantino and Pikal, 2004). Therefore, the results of ourstudy are of added value and, given the similar molecular struc-ture of DNA plasmids, might be extrapolated to other pDNA drugproducts.

During the conduct of the study, pDERMATT lyophilized productwas stored at −20 ◦C. Indeed, also in this trial patient inclusion wasdelayed and the strategy of stabilization of pDNA by dehydrationthrough lyophilization in combination with storage at low temper-ature ensured a stable pharmaceutical pDNA product of sufficientquality throughout the clinical study and prevented costly andtime-consuming additional manufacture of drug product. Also, thestability results generated are indicative for significant stability oflyophilized pDNA at a higher storage temperature (i.e. >24 monthsat +2–8 ◦C). Therefore, for an upcoming clinical trial at our insti-tute examining the toxicity and efficacy of a plasmid DNA directed

toward human papilloma virus (HPV) induced cancers, the sameformulation composition (Quaak et al., 2008) and storage condition(−20 ◦C) will be used. In parallel, stability testing will be performedat higher storage temperature conditions.

Tim e (minute s)25 30 35 40 45 50

SC

OCSC dimerab

SC

OCSC di mer

forms -20ºCC forms -20ºC

forms 5ºCC forms 5ºC

forms 25ºCC forms 25ºC

B

for quantitative analysis of plasmid DNA isoforms. Formulations were stored atespectively. (B) Electropherigram from CGE analyzing different plasmid topologies

650 I. van der Heijden et al. / International Journal of Pharmaceutics 453 (2013) 648– 650

Table 1pDERMATT quality control in time at different storage conditions. Abbreviations: AEX-HPLC, anion exchange high performance liquid chromatography; Rt , retention time;UV, ultra violet; AGE, agarose gel electrophoresis.

Storage condition 25 ◦C/60%RH 2–8 ◦C −20 ◦C

Test item Release specification T = 0 months T = 6 months T = 24 months T = 66 months

Visual inspection White, freeze-dried cake Conforms Conforms Conforms ConformsReconstitution Complete, leaving no visible

residue as undissolved matterand resulting in a clearcolorless solution

Conforms Conforms Conforms Conforms

AEX-HPLC analysis1. Identification Rt reference standard = Rt

powder for ID injectionConforms Conforms Conforms Conforms

2. Purity ≥90% SC 95.7% 95.1% 99.0% 98.8%

UV analysis1. Content 90.0–110.0% of labeled content 99.6% 97.5% 97.4% 98.8%2. Purity A260/280 = 1.80–1.95 1.80 1.88 1.85 1.9

AGE analysis1. Identification 3395–4149 bp 3774 bp 3680 bp 3809 bp 3762 bp2. Purity Report size of other visible

bands1. 7595 bp2. 3959 bp

1. 7430 bp2. 3607 bp

1. 7729 bp2. 4105 bp

1. 7538 bp2. 3810 bp

Residual water content (%w/w) ≤2.0% 0.7%

Fig. 3. Analysis of pDERMATT in time at different storage conditions by 1% agarosegel electrophoresis: (A) t = 0 months, lane 1 Lambda DNA BstEII digest; lane 2 pDER-MATT linear standard; lane 3 pDERMATT open circular standard; lane 4 supercoiledDNA ladder; lane 5 pDERMATT final product. (B) t = 6 months, lane 1 lambda DNABstEII digest; lane 2 pDERMATT linear standard; lane 3 pDERMATT open circularstandard; lane 4 supercoiled DNA ladder; lane 5 pDERMATT final product 25 ◦C. (C)t = 24 months, lane 1 lambda DNA BstEII digest; lane 2 supercoiled DNA ladder; lane3 pDERMATT final product 2–8 ◦C. (D) t = 66 months, lane 1 supercoiled DNA ladder;lane 2 lambda DNA BstEII digest; lane 3 pDERMATT final product −20 ◦C.

rubber closures of injection vials. Int. J. Pharm. 79, 301–308.Walther, W., 2003. Stability analysis for long-term storage of naked DNA: impact on

nonviral in vivo gene transfer. Anal. Biochem. 318, 230–235.Zelphati, O., et al., 1998. Stable and monodisperse lipoplex formulations for gene

delivery. Gene Ther. 5, 1272–1282.

2.41% 2.39% 1.49%

References

Allison, S.D., Molina, M.C., Anchordoquy, T.J., 2000. Stabilization of lipid/DNA com-plexes during the freezing step of the lyophilization process: the particleisolation hypothesis. Biochim Biophys Acta 1468, 127–138.

Anchordoquy, Thomas, J., Armstrong, T.K., Molina, M.D.C., 2005. Low molecularweight dextrans stabilize nonviral vectors during lyophilization at low osmolal-ities: concentrating suspensions by rehydration to reduced volumes. J. Pharm.Sci. 94, 1226–1236.

Cherng, J.Y., et al., 1997. Freeze-drying of poly((2-dimethylamino)ethylmethacrylate)-based gene delivery systems. Pharm. Res. 14, 1838–1841.

Corveleyn, S., De Smedt, S., Remon, J.P., 1997. Moisture absorption anddesorption of different rubber lyophilisation closures. Int. J. Pharm. 159,57–65.

Costantino, H.R., Pikal, M.J., 2004. Lyophilization of Biopharmaceuticals. AAPS press.Lengsfeld, C.S., Anchordoquy, T.J., 2002. Shear-induced degradation of plasmid DNA.

J. Pharm. Sci. 91, 1581–1589.Molina, M.D.C., Anchordoquy, Thomas, J., 2008. Formulation strategies to minimize

oxidative damage in lyophilized lipid/DNA complexes during storage. J. Pharm.Sci. 97, 5089–5105.

Van Oijen, M., et al., 2004. On the role of melanoma-specific CD8+ T-cell immunityin disease progression of advanced-stage melanoma patients. Clin. Cancer Res.10, 4754–4760.

Quaak, S.G.L., et al., 2009. Development and validation of an anion-exchange LC–UVmethod for the quantification and purity determination of the DNA plasmidpDERMATT. J. Pharmaceut. Biomed. 49, 282–288.

Quaak, S.G.L., et al., 2008. GMP production of pDERMATT for vaccinationagainst melanoma in a phase I clinical trial. Eur. J. Pharm. Biopharm. 70,429–438.

Remaut, K., et al., 2006. Influence of plasmid DNA topology on the transfectionproperties of DOTAP/DOPE lipoplexes. J. Control. Release 115, 335–343.

Schleef, M., 2005. DNA-pharmaceuticals: formulation and delivery in gene therapy.In: Schleef, M. (Ed.), DNA Vaccination and Immunotherapy. Wiley-VCH VerlagGmbH & Co. KGaA, Weinheim, FRG.

Sharma, V.K., Klibanov, A.M., 2007. Moisture-induced aggregation of lyophilizedDNA and its prevention. Pharm. Res. 24, 168–175.

Urthaler, J., Buchinger, W., Necina, R., 2005. Improved downstream process forthe production of plasmid DNA for gene therapy. Acta Biochim. Pol. 52,703–711.

Verstrepen, B.E., et al., 2008. Improved HIV-1 specific T-cell responses by short-interval DNA tattooing as compared to intramuscular immunization in non-human primates. Vaccine 26, 3346–3351.

Vromans, H., Van Laarhoven, J.A.H., 1992. A study on water permeation through