International Journal of Pharmaceutics 483 (2015) 212–219

Pharmaceutical nanotechnology

Enhanced in vivo therapeutic efficacy of plitidepsin-loadednanocapsules decorated with a new poly-aminoacid-PEG derivative

Giovanna Lollo a,b,1, Pablo Hervella b, Pilar Calvo c, Pablo Avilés c, Maria Jose Guillén c,Marcos Garcia-Fuentes a,b, Maria José Alonso a,b, Dolores Torres b,*aCenter for Research in Molecular Medicine and Chronic Diseases (CIMUS), Campus Vida, University of Santiago de Compostela, 15782 Santiago deCompostela, SpainbDepartment of Pharmaceutics and Pharmaceutical Technology, School of Pharmacy and Heath Research Institute of Santiago de Compostela (IDIS), CampusVida, University of Santiago de Compostela, 15782 Santiago de Compostela, Spainc PharmaMar S.A, Colmenar Viejo, Madrid, Spain

A R T I C L E I N F O

Article history:Received 19 December 2014Received in revised form 8 February 2015Accepted 10 February 2015Available online 11 February 2015

Keywords:NanomedicineAntitumor drugsStealth propertiesPolyglutamic acidPlitidepsinCancer therapy

A B S T R A C T

The focus of this study is to disclose a new delivery carrier intended to improve the pharmacokineticcharacteristics of the anticancer drug plitidepsin and to favor its accumulation within the tumor. Thesenanocarriers named as nanocapsules, consist of an oily core surrounded by a highly PEGylatedpolyglutamic acid (PGA-PEG) shell loaded with plitidepsin. They showed a size of around 190 nm, a zetapotential of �24 mV and were able to encapsulate a high percentage (85%) of plitidepsin. In vivo studies,following intravenous injection in healthy mice, indicated that the encapsulation of the drug within PGA-PEG nanocapsules led to an important increase in its area under the curve (AUC) which is related to theimportant decrease of the clearance, as compared to the values observed for the drug dissolved in aCremophor1 EL solution. This improvement of the pharmacokinetic profile of the encapsulatedplitidepsin was accompanied by a high increase (2.5-fold) of the maximum tolerated dose (MTD) incomparison to that of plitidepsin Cremophor1 EL solution. The efficacy study performed in a xenografttumor mice model evidenced the capacity of PGA-PEG nanocapsules to significantly reduce tumorgrowth. These promising results highlight the potential of PGA-PEG nanocapsules as an effective drugdelivery system for cancer therapy.

ã 2015 Published by Elsevier B.V.

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

journa l home page : www.e l sev ier .com/ loca te / i jpharm

1. Introduction

Over the last decades, nanopharmaceuticals have a particularimpact in improving cancer therapeutics (Blanco et al., 2011; Wangand Thanou, 2010). Most anticancer drugs used in conventionalchemotherapy are rapidly cleared from the blood circulation and,because they do not differentiate between cancerous and normaltissues, their use generally leads to major systemic side effects (Shiet al., 2011). Nanocarriers with long-circulating times offer thepotential advantage of being accumulated and entrapped withintumors due to the high permeability of tumoral vasculature andfrequently poor lymphatic drainage (Fang et al., 2011; Bertrandet al., 2014). A large number of nanocarriers containing cytotoxicdrugs have been examined in clinical trials and been approved for

* Corresponding author. Tel.: +34 8814880; fax: +34 981547148.E-mail address: dolores.torres@usc.es (D. Torres).

1 Present address: LUNAM Université – INSERM U1066, IBS-CHU, F-49933 Angers,France.

http://dx.doi.org/10.1016/j.ijpharm.2015.02.0280378-5173/ã 2015 Published by Elsevier B.V.

use in humans. In particular, the FDA approval of Doxil1, PEGylatedliposomes containing doxorubicin, has opened the door for theclinical development of other long circulating nanocarriersincluding micelles, nanoparticles or conjugates (Shi et al., 2011).Despite these advances, there is a recognized need to furtherimprove the design and development of advanced nanocarriersthat should be able to improve the therapeutic benefits ofoncologicals (González-Aramundiz et al., 2012).

Among the biomaterials used in the development of nano-oncologicals, polyaminoacids and, in particular poly(L-glutamicacid) (PGA), have raised great expectancy because of theirbiodegradability and acceptable regulatory profile (Li and Wallace,2008). In fact, PGA conjugated with paclitaxel (Opaxio1, beforeXyotax1) has already reached phase III clinical trials (Singer, 2005),while other combinations, i.e., PGA conjugated with camptothecin,are in earlier clinical development (clinical phase I and II) (Singer,2005; Dipetrillo et al., 2012). An additional interesting property ofpolyaminoacids relies on the possibility to conjugate them withpoly(ethylene glycol) (PEG), thus, rendering their surface morehydrophilic and flexible to prevent the uptake by the mononuclear

G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219 213

phagocytic system (MPS). The PEG-modified PGA nanosystems willhave the opportunity to perform as long circulating carriers, andthen ideally to decrease the access to the MPS organs andmaximize their presence in the tumor (Huynh et al., 2010).

Recently, we reported the design of a novel nanocapsule-typecarrier made of PGA and PGA-PEG, with a lower PEG content, whichis particularly attractive for accommodating hydrophobic antican-cer drugs such as plitidepsin (Gonzalo et al., 2013). Plitidepsin is ahydrophobic antitumor drug originally isolated from the Mediter-ranean tunicate Aplidium albicans and currently produced bychemical synthesis (PharmaMar S.A., Spain). It is currently in phaseII clinical trials for solid and haematological malignant neoplasiaslike T cell lymphoma and in phase III clinical trials for multiplemyeloma (Geoerger et al., 2012; Ribrag et al., 2013). Because of itsextremely low aqueous solubility, instability in the biologicalenvironment, and non-selective distribution, we hypothesized thatthis drug could benefit from its formulation in the form ofnanocapsules. The results showed that encapsulation within thenanocapsules provided the drug with a prolonged blood circula-tion and a significantly reduced toxicity.

Based on this previous experience, this work was aimed atdeveloping a new version of these PGA-PEG nanocapsules, inwhich the shell is made of a di-block copolymer with a high PEGcontent (57% w/w). The purpose was to obtain further improve-ments in the biodistribution profile of plitidepsin, thus, achieving abetter toxicity-efficacy profile in comparison with plitidepsindissolved in a Cremophor1 EL solution as reference formulation.

2. Materials and Methods

2.1. Chemicals

Plitidepsin was provided by PharmaMar S.A. (Spain). Miglyol1

812, a neutral oil formed by esters of capric and caprilyc fatty acidand glycerol, was a gift sample from Sasol Germany GmbH(Germany). Epikuron1 170, a phosphatidylcholine enrichedfraction of soybean lecithin, was kindly provided by Cargill (Spain).Benzalkonium chloride, Poloxamer 188 (Pluronic1 F68) and D-(+)-trehalose dehydrate were purchased from Sigma–Aldrich(Spain). Poly-L-glutamic acid-polyethylene glycol (PGA-PEG Mw35 kDa) was synthetized and supplied by Alamanda Polymers(USA). PGA-PEG was a diblock copolymer with a PEG content of 57%w/w; PEG chains length was 20 kDa and the PGA chains length wasabout 15 kDa.

2.2. Preparation of PGA-PEG nanocapsules

The preparation of unloaded PGA-PEG nanocapsules was basedon a modification of the solvent displacement technique whichinvolved the ionic interaction between the PGA-PEG and a cationicsurfactant after the solvent diffusion (Gonzalo et al., 2013). Briefly,an organic phase made of 30 mg of Epikuron1 170, 0.125 mL ofMiglyol1, 9.5 mL of acetone, 0.5 mL of ethanol and 7 mg ofbenzalkonium chloride was poured over an aqueous phasecontaining 10 mg of the polymer PGA-PEG and 50 mg of Poloxamer188. Solvents were evaporated from the suspension under vacuumto a final volume of 10 mL. Plitidepsin-loaded PGA-PEG nano-capsules were obtained as previously described but dissolving20 mg of the hydrophobic drug in the 0.5 mL of ethanol and thenfollowing the mentioned procedure.

2.3. Characterization of PGA-PEG nanocapsules

Particle size and z potential of colloidal systems weredetermined, respectively, by photon correlation spectroscopyand laser Doppler anemometry, using a Zetasizer Nano ZS (Malvern

Instruments, UK). The morphology of the nanocapsules wasanalyzed by transmission electron microscopy using a PhilipsCM-12 microscope (FEI Company, Eindhoven). Samples werestained with phosphotungstic acid solution (2% w/v) and placedon a copper grid with Formvar1 films for analysis. Plitidepsinencapsulation efficiency in PGA-PEG nanocapsules was calculatedindirectly by the difference between the total amount ofplitidepsin in the system and the non-encapsulated drug measuredafter the isolation of loaded-nanocapsules:

- The total amount of drug was estimated from a fresh aliquot ofthe nanocapsules formulation by dissolving the colloidalsuspension in acetonitrile. These samples were centrifuged(4000 � g, 20 min, 20 �C) and the supernatant was analyzed byHPLC.

- The free drug was quantified by the HPLC following separation ofthe nanocapsules by ultracentrifugation (27,400 � g, 1 h, 15 �C).

Therefore, the encapsulation efficiency (E.E.) was calculated asfollows (Eq. (1)):

E:E:% ¼ A � BA

� �� 100 (1)

where A is the experimental total drug amount and B is theunloaded drug amount.

The HPLC system consisted of an Agilent 1100 series instrumentequipped with UV detector set at 225 nm. The analytic methodemployed was previously validated by PharmaMar S.A. (Spain)(Mohammad Shahin, 2014).

2.4. In vitro release study of plitidepsin from PGA-PEG nanocapsules

The release study of plitidepsin was performed by incubating analiquot of the nanocapsules formulation in PBS (pH 7.4) at anappropriate concentration (0.5 mg/mL) to assure sink conditions(drug concentration below 4.9 mg/mL). The vials were placed in anincubator at 37 �C with horizontal shaking. At different intervals(1 h, 3 h, 6 h and 24 h), 3 mL of the suspension diluted in PBS werecollected and ultracentrifuged in Herolab1 tubes (27,400 � g, 1 h,15 �C). The plitidepsin released at each time was calculatedindirectly by the difference between the total amount of drugpresent in the system and the free plitidepsin in the infranatantafter the ultracentrifugation, both determined by HPLC, asdescribed in the previous section.

2.5. Stability study of plitidepsin-loaded PGA-PEG nanocapsules uponstorage

The stability of plitidepsin-loaded nanocapsules was evaluatedupon storage in sealed tubes at 4 �C. Size, polydispersity index, zetapotential and leakage of the drug from the nanocapsules weremeasured for a period of 1 month.

2.6. Freeze-drying study of PGA-PEG nanocapsules

The selected variables for the lyophilisation study of blank PGA-PEG nanocapsules were the nanocapsules concentration (1,0.75 and 0.5% w/v) and the trehalose concentration (5 and 10%w/v). For this purpose, 1 mL of the diluted blank PGA-PEGnanocapsules containing trehalose was placed in glass vials andfrozen in liquid nitrogen. The freeze-drying program consisted ofan initial drying step at �35 �C, and a secondary drying where thetemperature was finally equilibrated at 20 �C over a period of 60 h(Labconco Corp., USA).

PGA-PEG nanocapsules were resuspended by adding 1 mL ofultrapure water to the freeze-dried powders followed by manual

214 G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219

stirring. Finally, their size distribution was measured afterresuspension. An additional freeze-drying study was done withdrug-loaded nanocapsules at a concentration of 1% w/v with a 10%w/v of trehalose.

2.7. In vivo studies

2.7.1. AnimalsFemale athymic nu/nu mice and CD-1 male mice between

4 and 6 weeks of age and ranging in weight from 21 to 30 gwere purchased from Harlan Laboratories Models, S.L. (Barce-lona, Spain). Animals were housed in individually ventilatedcages (Sealsafe1 Plus, Techniplast S.P.A.), 10 mice per cage, on a12 h light-dark cycle at 21–23 �C and 40–60% relative humidity.Mice were allowed free access to irradiated standard rodentdiet (Tecklad 2914C) and sterilized water. Animals wereacclimated for five days prior to being individually tattoo-identified. Animal protocols were reviewed and approvedaccording to regional Institutional Animal Care and UseCommittees (Spain).

2.7.2. Pharmacokinetic evaluationThe pharmacokinetic study of plitidepsin was performed in CD-

1 healthy male mice (n = 4 animals each time point) afteradministration of a single i.v. dose of plitidepsin encapsulated inPGA-PEG nanocapsules and compared with i.v. administration ofplitidepsin dissolved in a Cremophor1 EL reference solution(Cremophor1 EL/ethanol/water 15/15/70% by weight); the pliti-depsin dose was 0.1 mg/kg of body weight and the volumeadministered was 150 mL.

On the day of dosing, blood samples were drawn via cardiacpuncture at 9 pre-established time points: 5, 15, 30 min and 1, 2, 4,8, 24, 48 h post-injection (two animals were used per time pointand formulation (n = 36)). Blood samples were transferred intotubes containing EDTA as anticoagulant. The blood was kept indarkness on ice until it was centrifuged at 3000 rpm for 15 min at5 �C. The plasma obtained was frozen at �20 �C and maintained indarkness until analysis.

Plitidepsin concentrations were determined in mouse plasmasamples using a modified HPLC/MS/MS method previouslydescribed (Brandon et al., 2005). Briefly, plitidepsin and theinternal standard (IS), PM91105, were extracted from plasma byliquid solid extraction. 200 mL of urea:glycine buffer 1:1 v/v wereadded to plasma samples (100 mL) containing IS (25 mL; 20 ng/mL)and this solution was extracted with 1500 mL of tert-butylmethy-lether (TBME)/hexane 1:1 v/v.

The analysis was carried out in a gradient reversed phasechromatography followed by positive ion electrospray tandemmass spectrometry (ESI/MS/MS) detection using multiple reactionmonitoring (MRM). The HPLC system consisted of a Shimadzu LC-10ADvp solvent delivery unit, on-line degasser, gradient mixer andsystem controller (Shimadzu Scientific, Columbia, MD, USA). ACTC-PAL Lcap autosampleer (LEAP Technologies, Carrboro, NC,USA) was used to inject samples. A PE Sciex API 4000 triplequadrupole mass spectrometer (Toronto, ON, Canada), equippedwith a Turbo Ion Spay interface, was used. The analytical columnwas a Waters Atlantis T3, 3 mm, 20 � 2.1 mm. The mobile phase wasacetonitrile (0.1% formic acid)/water (0.1% formic acid) with a flowrate of 600 mL/min. The column oven temperature was set at 50 �Cand the sample injection volume was 20 mL.

The assay was linear over the concentration range 0.05–50 ng/mL of plitidepsin. The calibration curve was defined by a slope of0.23 and an intercept of 0.006 (R = 0.9958). The coefficient ofvariations for low (0.1 ng/mL), mid (5 ng/mL) and high (30 ng/mL)quality control samples of plitidepsin were 12, 18.5 and 9.5%,respectively. The pharmacokinetic parameters of plitidepsin were

obtained using a non-compartmental pharmacokinetic methodwith WinNonlin 5.2 software (Brandon et al., 2005).

2.7.3. Toxicity studiesAcute toxicity of the formulations was determined by assessing

the MTD (maximum tolerated dose after single administration)and the MTMD (maximum tolerated dose after multiple adminis-tration) of plitidepsin-loaded PGA-PEG nanocapsules in healthyCD-1 male mice following i.v. injection (150 mL; n = 10/group).Toxicity of plitidepsin dissolved in the Cremophor1 EL referencesolution (Cremophor1 EL/ethanol/water 15/15/70% by weight) wasalso investigated. MTD and MTMD were defined as the highestdose not causing significant lethality (as death) or any prominentobservable changes during the experiment (14 days) according tothe standard ethical issues.

For the MTD evaluation, the formulations were administered asa single i.v. bolus in the lateral vein of the tail, whereas for theMTMD, they were intravenously administered following 2 cycles of5 consecutive treatment days spaced by 4 free days (Oliveira et al.,2014). Plitidepsin-loaded PGA-PEG and the reference plitidepsinsolution were tested from 1.5 to 0.1 mg/kg.

The MTD of blank systems could not be determined as the toxicdose was not reached with the tested concentrations.

2.8. Efficacy studies

2.8.1. Xenograft modelMRI-H-121 is a human renal carcinoma originally obtained

from the DCT Tumor Bank, developed by Dr. A.E. Bogden, MasonResearch Institute, MA, and maintained as a serial transplantedtumor line in athymic nude mice. The original tissue came from apatient at University of Massachusetts Medical Center (USA).

2.8.2. In vivo antitumor activityFor the study, 4–6 week-old athymic nu/nu mice were

subcutaneously implanted in their right flank with MRI-H-121tissue from serially transplanted donor mice using a 13G trocar.The tissue was debrided of membrane, haemorrhagic and necroticareas and 3 mm3 fragments were implanted. When tumorsreached 150–200 mm3, tumor-bearing animals (n = 10/group) wererandomly allocated into the following treatment groups: (i)plitidepsin dissolved in the Cremophor1 EL reference solution,(ii) plitidepsin-loaded PGA-PEG nanocapsules, and (iii) serumsaline as a control.

The doses of plitidepsin and schedules were selected based onMTMD determination (i.e., 0.15 and 0.3 mg/kg for plitidepsin-loaded PGA-PEG nanocapsules and Cremophor1 EL referencesolution, respectively), having a final dose of 3 mg/kg for bothformulations.

The formulations were injected intravenously in the tail veins ofthe mice:

(i) Plitidepsin dissolved in Cremophor1 EL was injected at a doseof 0.3 mg/kg during 2 cycles of 5 consecutive treatment daysand then 4 days free.

(ii) Plitidepsin-loaded PGA-PEG nanocapsules were injected dailyat a dose of 0.15 mg/kg during 20 days.

The lower dose used for the nanocapsules formulation wasfixed according to the MTMD value, which clearly is related withtheir long circulating properties and their accumulation in bloodcirculation for extended periods of time.

Tumor volume and mice body weight were measured 2–3 timesper week starting from the first day of treatment (day 0).Treatments that produced 20% lethality and/or 20% of net bodyweight loss were considered toxic.

Table 1Physicochemical characteristics of blank and plitidepsin-loaded PGA-PEG nano-capsules (mean � S.D.; n = 3). NCs: nanocapsules; P.I.: polydispersity index; E.E.:encapsulation efficiency.

Formulation Size (nm) P.I. z potential (mV) E.E. (%)

Blank PGA-PEG NCs NCs 180 � 4 0.1 �20 � 4 –

Plitidepsin-loaded PGA-PEG NCs 190 � 15 0.1 �24 � 5 85 � 4

G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219 215

Tumor volume was calculated using the equation (Eq. (2)):

V ¼ a � b2

2(2)

where a and b were the longest and shortest diameters,respectively.

Animals were euthanized when their tumors reached a volumeof 2000 mm3 and/or severe necrosis was seen.

The antitumor effect was calculated by using DT/DC (%),defined as the percent change in tumor volume for each treated(T) and placebo (C) group. DT/DC was calculated on days 7,14 and 21. The data are presented as medians and interquartilerange (IQR).

Treatment tolerability was assessed by monitoring body weightevolution and clinical signs as well as evidence of local damage atthe injection site. All placebo-treated animals died or weresacrificed for ethical reasons from day 0 to 21.

2.8.3. Experimental design and statistical analysisDesign, randomization and monitoring of body weight and

tumor measurements were performed using NewLab OncologySoftware (version 2.25.06.00). Tumor volume data are presented asmedians and interquartile range (IQR). Tumor volumes of thetreated groups on day 0 and day X (T0� TX) and those of the controlgroup (C0� CX) were used to determine the activity rating asfollows (Eq. (3)):

DTDC

% ¼ T0 � TX

C0 � CX

� �� 100 (3)

Activity rating:

� DT/DC > 50% inactive (�).� DT/DC > 25–50% tumor inhibition (+).� DT/DC < 25% and �TX/T0 > 75–125% tumor stasis (++) or �TX/T0 > 10–75% partial regression (+++).

TX/T0: tumor volume of the compound-treated group on day Xand on day 0.

Tumor volume data from groups following the 1st, 2nd and 3rdstudy weeks were compared using a two-tailed Mann–Whitney Utest. In all cases, p < 0.05 was accepted as denoting a statisticaldifference.

3. Results and discussion

We have recently developed a new type of nanocapsules madeof PGA and PGA-PEG (PEG-grafted copolymer; 24% w/w of PEG)that were able to provide the anticancer drug plitidepsin with half-lives appreciably prolonged, and increased AUC values incomparison with the drug-loaded uncoated cores. In addition,we could also appreciate the favorable effect of the polymerPEGylation on plitidepsin’s pK profile. Based on those results, themain goal of the present work has been to develop a moreadvanced prototype of PGA-PEG nanocapsules by increasing theirPEG content, with the final aim of maximizing its passiveaccumulation at the tumor site. This advanced prototype consistsof an oily core and a polymeric shell made of a highly PEGylatedPGA-PEG diblock copolymer (57% w/w of PEG). By selecting thisdiblock copolymer we could easily double the PEG content incomparison with the previous grafted copolymer. Taking intoaccount the extreme low solubility of plitidepsin and the necessityof using solvents to make its i.v. administration feasible, this drugshould clearly benefit from this new technology. The pharmacoki-netic parameters and therapeutic efficacy of the loaded systems

were evaluated and compared for first time with those ofplitidepsin dissolved in a Cremophor1 EL solution, used asreference.

3.1. Preparation and characterization of plitidepsin-loaded PGA-PEGnanocapsules

PGA-PEG nanocapsules were prepared by the solvent displace-ment technique (Gonzalo et al., 2013). By using this method, thedeposition of a polymer coating onto the oily core is produced oncethe organic solvent diffuses immediately into the polymer aqueoussolution. In our case, the attachment of the PGA-PEG to the oilycore was driven by the inclusion of the positively chargedsurfactant, benzalkonium chloride, in the oily phase. Benzalko-nium chloride was selected on the basis of its acceptabletoxicological profile and it was used in the minimum amountthat allowed the formation of stable systems.

The polymer chosen was a block copolymer that contains a highpercentage, around 57% w/w, of high Mw PEG (20 kDa). It isexpected that its disposition on the external layer create acomplete PEG-hydrated corona around the particles. This disposi-tion is expected to endow the system with improved stealthproperties (Mosqueira et al., 2001) which could also to contributeto a better access to the tumor area.

The physicochemical properties of blank and plitidepsin-loadedPGA-PEG nanocapsules are summarized in Table 1. It can be notedthat the use of adequate concentrations of polymer and cationicsurfactant results in the formation of homogenous populations ofnanocapsules around 180–190 nm. The results showed that theincorporation of the drug into PGA-PEG nanocapsules did not affectthe size and the z potential of the systems. Both empty and loadedsystems showed negative zeta potential values, which indicatesthe inversion from the positive values of the uncoatednanoemulsion (+38 mV), confirming the success in the formationof the PGA-PEG coating. Furthermore, an encapsulation efficiencyvalue around of 85% indicated that the oily core could easilyallocate the drug plitidepsin.

The morphological appearance of the nanocapsules wasobserved by transmission electron microscopy (Fig. 1). Themicrographs indicated that PGA-PEG nanocapsules have a roundshape and a size below than 200 nm, similar to that obtained byphoton correlation spectroscopy.

In a second step, we evaluated the release pattern ofplitidepsin-loaded PGA-PEG nanocapsules under “sink condi-tions”. The in vitro release profile of loaded nanocapsules(Fig. 2) indicated that the system follows a biphasic profilecharacterized by an initial burst of about 80% of the drug payload,followed by a second phase in which no further drug release wasobserved. This profile, typical of other nanocapsules and nano-emulsions, indicates that the release process is highly dependenton the partition of the drug between the oily cores and the greatvolume of the external aqueous phase (Oliveira et al., 2014). Eventhough these results cannot be extrapolated to the in vivo situation,the fact that a fraction of the drug remained encapsulated despitethe “sink conditions” is an indication of the affinity of plitidepsinfor the oily core and/or the shell of the nanocapsules.

Fig. 3. Particle size of the reconstituted freeze-dried blank PGA-PEG nanocapsules

Fig. 1. TEM images of plitidepsin-loaded PGA-PEG nanocapsules.

216 G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219

3.2. Stability and freeze-drying studies of PGA-PEG nanocapsules

The stability of the plitidepsin-loaded nanocapsules understorage at 4 �C in terms of size, zeta potential and leakage of thedrug, was assessed during a period of 1 month. There was nomodification on the particle size neither on the zeta potential ofthe nanocapsules, which maintained their original values through-out the study. Additionally, no leakage of plitidepsin could beobserved, which demonstrates the effectiveness of PGA-PEGnanocapsules as carriers for the hydrophobic anticancer drugplitidepsin.

In a second step, we explored the optimal freeze-dryingconditions for the conversion of the aqueous suspension ofnanocapsules into a powder. The blank nanocapsules were freeze-dried at two different trehalose concentrations (5 and 10% w/v)and the concentration of blank nanocapsules was tested at threelevels (0.5–0.75–1% w/v). The results indicated that the recoveryof the initial properties of PGA-PEG nanocapsules upon freeze-drying and reconstitution was dependent on the cryoprotectantand nanosystem concentration. PGA-PEG nanocapsules could befreeze-dried using a trehalose concentration of 10% at any of thetested concentrations tested, remaining the size close to theinitial values (Fig. 3). The lower concentration of trehalose led tosizes slightly higher after reconstitution. When drug-loadednanocapsules at a 1% w/v concentration were freeze-dried with a10% w/v of trehalose, no changes in size or drug integrity werefound.

Fig. 2. Plitidepsin release in PBS (pH 7.4, 37 �C) from PGA-PEG nanocapsules(mean � SD; n = 3).

3.3. In vivo studies

3.3.1. Pharmacokinetic evaluationThe plasma disposition characteristics of plitidepsin formulated

in PGA-PEG nanocapsules and in the Cremophor1 EL referencesolution were evaluated in health CD-1 mice. The formulationswere administered i.v. through a single bolus injection (0.1 mg/kgplitidepsin). As shown in Fig. 4, encapsulating the drug withinPGA-PEG nanocapsules led to a great prolongation of its presencein the blood circulation. In fact, approximately a 10% of the injected

(NCs). Different concentrations (w/v) of nanocapsules were lyophilized usingtrehalose at 5% (&) or 10% (&) w/v (mean � S.D.; n = 3).

Fig. 4. Plasma concentration–time profiles of plitidepsin following i.v. injection inmice of plitidepsin-loaded PGA-PEG nanocapsules (^) and plitidepsin Cremophor1

EL solution (&). Data represent mean � S.D.

Table 4Tumor volumes (TV) and mortality in mice bearing MRI-H-121 xenografts treatedwith multiple i.v. plitidepsin-loaded PGA-PEG nanocapsules and the plitidepsinCremophor1 EL reference solution.

3

Table 2Pharmacokinetic parameters of plitidepsin-loaded PGA-PEG nanocapsules and plitidepsin Cremophor1 EL reference solution after single i.v. injection in mice. PK parametersare average values.

Formulation t1/2 (h) AUC0!48h/dose (ng h/mL/mg) CL (mL/min/kg) V (L/kg) MRT (h)

Cremophor1 EL solution 8.2 57.9 157.0 106.9 10.5PGA-PEG NCs 17.0 274.7 52.7 77.7 20.8

Fig. 5. Evolution of tumor volume median following i.v. multiple administration ofplitidepsin-loaded PGA-PEG nanocapsules (~), Cremophor1 EL reference solution(^) and saline serum (&) in a subcutaneously implanted MRI-H-121 human renalxenograft mouse model (total plitidepsin dose was 0.3 mg/kg).

G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219 217

dose of plitidepsin-loaded nanocapsules remained in circulation at48 h after injection, whereas plitidepsin formulated in theCremophor1 EL solution was below the limit of detection afterthe sampling time of 8 h.

Pharmacokinetic analysis of the plitidepsin levels uponadministration of the nanocapsules or the reference formulationis shown in Table 2. Plitidepsin in PGA-PEG nanocapsules exhibiteda 2-fold higher half-life (17.0 h) and MRT (20.8 h) compared toplitidepsin dissolved in the Cremophor1 EL solution. The plasmaAUC0–24h of plitidepsin-loaded nanocapsules was about 5-foldgreater than that obtained with the reference formulation.Moreover, plitidepsin in PGA-PEG nanocapsules showed a 3-foldlower plasmatic cleareance (52.7 mL/min/kg) as compared to theplitidepsin reference solution (157.0 mL/min/kg).

In a recent work, plitidepsin-loaded micelles were developedand their pharmacokinetics was also compared with that of thedrug dissolved in the Cremophor1 EL solution (Oliveira et al.,2014). These micellar structures, especially those made of poly(trimethylene carbonate)-block-poly(glutamic acid) were able toincrease the AUC0–24h of plitidepsin, in an extent almost attainingthe double of the reference value. However the half-life was hardlymodified (from 8 to 8.6 h), being notably shorter than that showedin the present work.

Overall, the pharmacokinetic parameters of plitidepsin-loadedPGA-PEG nanocapsules highlight the long-circulating properties ofPGA-PEG in comparison with the reference formulation. It seemsclear, as with other PEGylated nanocarriers containing hydropho-bic antitumor drugs, that the long circulating properties and highAUC values are related to an important decrease in the clearance ofthe antitumor drug. This was observed, for example, for PEGylatedlipid nanocapsules containing docetaxel, when compared with thecommercial formulation Taxotere1 (Khalid et al., 2006), or forPEGylated polymeric micellar and liposomal nanoformulationscontaining paclitaxel when compared with Taxol1 (Yang et al.,2007; Xiao et al., 2012; Zhang et al., 2012).

Moreover, if we examine our results and compare them withthose previously reported by us for plitidepsin-loaded PGA-PEGnanocapsules with a low PEG content, it is important to note thatthe plitdepsin’s AUC was considerably increased and the clearanceconsiderably decreased (a 3-fold change for each parameter)(Gonzalo et al., 2013), demonstrating the effect of the higherPEGylation on the pharmacokinetic behavior. These changes werenot reflected, however, in an increase in the half-life or MRT values,as might be expected. In fact, these values remained almostunchanged (t1/2 �18 h and MRT �20 h for both high and lowPEGylated nanocapsules). The different values of the plitidepsindistribution volumes observed for the two nanoformulations(106.9 L for high-PEGylated and 265.8 L for low-PEGylated nano-capsules) would explain these results. Reports in the literature

Table 3MTD (maximum tolerated dose after single administration) and the MTMD(maximum tolerated dose after multiple administration) of plitidepsin-loaded PGA-PEG nanocapsules and plitidepsin dissolved in the Cremophor1 EL referencesolution, in healthy CD-1 male mice following i.v. injection (n = 10/group).

Formulation MTD (mg/kg) MTMD (mg/kg)

PGA-PEG nanocapsules 0.75 0.15Cremophor1 EL reference solution 0.30 0.30

suggest that long circulating times and a smaller distributionvolume of taxanes are consistent with less extensive antitumordrug distribution in tissues such as liver, spleen or kidneys (Khalidet al., 2006; Yoshizawa et al., 2011). Interestingly, this increasedplasmatic residence and lower tissue concentrations were alsoaccompanied by a significant increase in the tumor accumulationmeasured in tumor-bearing mice (Yoshizawa et al., 2011). On theother hand, these results are in agreement with extensive data inthe literature that correlates the long residence time in plasma oftaxanes after being administered in a PEGylated nanostructurewith an increased deposition of the drug in the tumor via passivetargeting (Yang et al., 2007; Senthilkumar et al., 2008; Jing et al.,2014).

3.3.2. Toxicological evaluationThe toxicological study was aimed at establishing the MTD and

the MTMD of plitidepsin-loaded PGA-PEG nanocapsules followingi.v. injection. The results were compared with those of plitidepsinin Cremophor1 EL reference solution.

As observed in Table 3, the MTD of PGA-PEG nanocapsules was2.5 times higher than that of Cremophor1 EL solution, and higher

Formulation Day TV mm median (IQR) p Mortality

Cremophor1 EL solution 7 149 (133.9–268.6) 0.0001 0/1014 80.1 (52.7–127.7) <0.0001 0/1021 181.3 (144.0–264.9) <0.0001 1/10

PGA-PEG NCs 7 426.1 (376.5–561.6) NS 0/1014 358.8 (275.8–486.0) <0.0001 0/1021 154.5 (80.5–236.8) <0.0001 0/10

Data are presented as median and interquartile range (IQR).p value for Mann–Whitney U test (control group compared against the rest).NS: not statistically significance; NCs: nanocapsules.

Table 5Antitumor effect parameters and activity ranking of plitidepsin-loaded PGA-PEG nanocapsules and plitidepsin Cremophor1 EL formulation after i.v. multiple injection in micebearing MRI-H121 xenograft during a period of 20 days. NCs: nanocapsules.

Formulation Dose (mg/kg) DT/DC% Activity rating

Day 7 Day 14 Day 21 Day 7 Day 14 Day 21

Cremophor1 EL solution 0.30 �5.8 �9.1 0.3 ++ +++ ++PGA-PEG NCs 0.15 55.0 17.2 �1.3 – + ++

Tumor inhibition (+); tumor stasis (++); partial regression (+++).DT/DC: difference of volumes of the treated groups on day 0 and day X (T0� TX) and those of control group (C0� CX) as reported in Section 2.DT/DC > 50% inactive (�).DT/DC > 25–50% tumor inhibition (+).DT/DC < 25% and �TX/T0 > 75–125% tumor stasis (++) or �TX/T0 > 10–75% partial regression (+++).TX/T0: tumor volume of the compound-treated group on day X and on day 0.

218 G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219

than their MTMD value. Besides, the MTMD of drug-loadedparticles is lower than the value obtained for the Cremophor1 ELformulation (0.15 vs 0.30 mg/kg).

The MTD results indicate that nanocapsules are better toleratedthan the reference formulation and could be administered athigher unique doses. However, in the case of MTMD values, theopposite occurs, this being in accordance with the pharmacoki-netic evaluation above reported. The long circulation properties ofplitidepsin-encapsulated into PGA-PEG nanocapsules help us toexplain this MTMD reduction. The drug exposure is higher,plitidepsin remains for longer time into the blood circulationand it is eliminated slowly. When plitidepsin is dissolved inCremophor1 EL, the pharmacokinetic behavior is different, leadingto a rapid elimination of the drug.

To allow an easier comparison between the two formulationsfor the in vivo efficacy study, plitidepsin was used around theirMTMD values (0.15 and 0.3 mg/kg).

3.3.3. Antitumor activityThe in vivo antitumor efficacy of plitidepsin-loaded PGA-PEG

nanocapsules was evaluated in a human renal xenograft mousemodel (MRI-H-121). A comparative study was performed bydividing animals into 3 groups according to the treatmentreceived (plitidepsin-loaded PGA-PEG nanocapsules, referenceformulation and serum) and the schedule of administrationestablished. A strong antitumor activity was seen after thetreatment with both plitidepsin formulated in Cremophor1 ELand PGA-PEG nanocapsules (Fig. 5). Moreover, there is asignificant difference (p < 0.001) in tumor volume at the day14 and 21 for plitidepsin-loaded PGA-PEG nanocapsules and thereference Cremophor1 EL solution with the respect to thecontrol, indicating that both formulations had a similarantitumor activity (Table 4). In fact, tumor stasis was recoveredat the end of the experiment after the treatment withplitidepsin-loaded PGA-PEG nanocapsules and the referencesolution (Table 5). However, neither mortality nor significantchanges in body weight were observed throughout the studywhen the nanocapsules were administered, whereas one animalwas found dead in the reference formulation group on day 15.

The activity rating and the antitumor effect calculated as DT/DC, and reported in Table 5, highlight the differences in themechanism of action of the two formulations. Plitidepsin dissolvedin the Cremophor1 EL formulation showed a rapid onset of action(delayed tumor growth), but this effect decreased once thetreatment was finished, at day 21. The trend of the plitidepsin-loaded nanocapsules was slightly different, showing a tendencynot only to achieve the stasis of the tumor at the end of thetreatment (day 21; Table 4) but also to stop the tumor growthtoward the end of the study (day 26; Fig. 5). These differentbehaviors could be clearly related with the different pharmacoki-netic profiles of both formulations, showing the nanocapsules a

more prolonged plasma residence and hence a delayed butapparently more efficient response. Considering the antitumoractivity, we suggest that the improved biodistribution profileobserved for PGA-PEG nanocapsules as compared to that of theCremophor1 EL formulation, might be responsible for the passiveaccumulation of the drug in the pathological site. These results arein agreement with those described for PEGylated nanocarrierswhich have shown enhancements of the therapeutic index ofdifferent antitumor drugs (Hureaux et al., 2010; Kim et al., 2001).

4. Conclusions

High PEGylated PGA-PEG nanocapsules were developed asnovel carriers for the antitumor hydrophobic drug plitidepsin.Results from the present study demonstrated that the coating ofnanocapsules with a high PEGylated PGA-PEG diblock copolymerimproved the pharmacokinetic profile of the drug as compared tothe reference formulation consisting of a Cremophor1 EL solution.Moreover, plitidepsin-loaded PGA-PEG nanocapsules were bettertolerated than plitidepsin formulated in the Cremophor1 ELsolution. In vivo antitumor activity in a xenograft tumor model inmice also revealed an important suppression of tumor growthupon multiple i.v. administration, an effect that was comparable tothat of the reference formulation. All these data indicate theinterest of PGA-PEG nanocapsules as a novel delivery platform forcancer chemotherapy.

Acknowledgements

The authors would like to acknowledge financial support fromCENIT-NANOFAR XS53 project, PharmaMar, Spain; the Xunta deGalicia (Competitive Reference Groups-FEDER Funds Ref 2014/043) and the European Commission FP7 EraNet-EuroNanoMedProgram-Instituto Carlos III (Lymphotarg proyect, Ref. PS09/02670). Giovanna Lollo was a recipient of a FPU fellowship fromthe Ministry of Education of Spain. Marcos Garcia-Fuentes was arecipient Isidro Parga Pondal Fellowship from Xunta de Galicia.

References

Bertrand, N., Wu, J., Xu, X., Kamaly, N., Farokhzad, O.C., 2014. Cancernanotechnology: the impact of passive and active targeting in the era of moderncancer biology. Adv. Drug Deliv. Rev. 66, 2–25.

Blanco, E., Hsiao, A., Mann, A.P., Landry, M.G., Meric-Bernstam, F., Ferrari, M., 2011.Nanomedicine in cancer therapy: innovative trends and prospects. Cancer Sci.102, 1247–1252.

Brandon, E.F.A., van Ooijen, R.D., Sparidans, R.W., Lázaro, L.L., Heck, A.J.R., Beijnen, J.H., Schellens, J.H.M., 2005. Structure elucidation of aplidine metabolites formedin vitro by human liver microsomes using triple quadrupole mass spectrometry.J. Mass Spectrom. 40, 821–831.

Dipetrillo, T., Suntharalingam, M., Ng, T., Fontaine, J., Horiba, N., Oldenburg, N.,Perez, K., Birnbaum, A., Battafarano, R., Burrows, W., Safran, H., 2012.Neoadjuvant paclitaxel poliglumex, cisplatin, and radiation for esophagealcancer: a phase 2 trial. Am. J. Clin. Oncol. 35, 64–67.

G. Lollo et al. / International Journal of Pharmaceutics 483 (2015) 212–219 219

Fang, J., Nakamura, H., Maeda, H., 2011. The EPR effect: unique features of tumorblood vessels for drug delivery, factors involved, and limitations andaugmentation of the effect. Adv. Drug Deliv. Rev. 63, 136–151.

Geoerger, B., Estlin, E.J., Aerts, I., Kearns, P., Gibson, B., Corradini, N., Doz, F., Lardelli,P., Miguel, B.D., Soto, A., Prados, R., Vassal, G., 2012. A phase I andpharmacokinetic study of plitidepsin in children with advanced solid tumours:an innovative therapies for children with cancer (ITCC) study. Eur. J. Cancer 48,289–296.

González-Aramundiz, J.V., Lozano, M.V., Sousa-Herves, A., Fernandez-Megia, E.,Csaba, N., 2012. Polypeptides and polyaminoacids in drug delivery. Expert Opin.Drug Deliv. 9, 183–201.

Gonzalo, T., Lollo, G., Garcia-Fuentes, M., Torres, D., Correa, J., Riguera, R., Fernandez-Megia, E., Calvo, P., Avilés, P., Guillén, M.J., Alonso, M.J., 2013. A new potentialnano-oncological therapy based on polyamino acid nanocapsules. J. Control.Release 169, 10–16.

Hureaux, J., Lagarce, F., Gagnadoux, F., Rousselet, M.-C., Moal, V., Urban, T., Benoit, J.-P., 2010. Toxicological study and efficacy of blank and paclitaxel-loaded lipidnanocapsules after i.v. administration in mice. Pharm. Res. 27, 421–430.

Huynh, N.T., Roger, E., Lautram, N., Benoît, J.-P., Passirani, C., 2010. The rise and rise ofstealth nanocarriers for cancer therapy: passive versus active targeting.Nanomedicine 5, 1415–1433.

Jing, X., Deng, L., Gao, B., Xiao, L., Zhang, Y., Ke, X., Lian, J., Zhao, Q., Ma, L., Yao, J., Chen,J., 2014. A novel polyethylene glycol mediated lipid nanoemulsion as drugdelivery carrier for paclitaxel. Nanomed. Nanotechnol. Biol. Med. 10, 371–380.

Khalid, M., Simard, P., Hoarau, D., Dragomir, A., Leroux, J.-C., 2006. Long circulatingpoly(ethylene glycol)-decorated lipid nanocapsules deliver docetaxel to solidtumors. Pharm. Res. 23, 752–758.

Kim, S.C., Kim, D.W., Shim, Y.H., Bang, J.S., Oh, H.S., Kim, S.W., Seo, M.H., 2001. In vivoevaluation of polymeric micellar paclitaxel formulation: toxicity and efficacy. J.Control. Release 72, 191–202.

Li, C., Wallace, S., 2008. Polymer-drug conjugates: recent development in clinicaloncology. Adv. Drug Deliv. Rev. 60, 886–898.

Mohammad Shahin, M.S., 2014. Analytical methods for determination of anticancerdrugs from marine sources. Int. J. Res. Pharm. Chem. 4, 829–836.

Mosqueira, V.C.F., Legrand, P., Morgat, J.-L., Vert, M., Mysiakine, E., Gref, R.,Devissaguet, J.-P., Barratt, G., 2001. Biodistribution of long-circulating

PEG-grafted nanocapsules in mice: effects of PEG chain length and density.Pharm. Res. 18, 1411–1419.

Oliveira, H., Thevenot, J., Garanger, E., Ibarboure, E., Calvo, P., Aviles, P., Guillen, M.,Lecommandoux, S., 2014. Nano-encapsulation of plitidepsin: in vivopharmacokinetics, biodistribution, and efficacy in a renal xenograft tumormodel. Pharm. Res. 31, 983–991.

Ribrag, V., Caballero, D., Fermé, C., Zucca, E., Arranz, R., Briones, J., Gisselbrecht, C.,Salles, G., Gianni, A.M., Gomez, H., Kahatt, C., Corrado, C., Szyldergemajn, S.,Extremera, S., de Miguel, B., Cullell-Young, M., Cavalli, F., 2013. Multicenterphase II study of plitidepsin in patients with relapsed/refractory non-Hodgkin’slymphoma. Haematologica 98, 357–363.

Senthilkumar, M., Mishra, P., Jain, N.K., 2008. Long circulating PEGylated poly(D,L-lactide-co-glycolide) nanoparticulate delivery of Docetaxel to solid tumors. J.Drug Target. 16, 424–435.

Shi, J., Xiao, Z., Kamaly, N., Farokhzad, O.C., 2011. Self-assembled targetednanoparticles: evolution of technologies and bench to bedside translation. Acc.Chem. Res. 44, 1123–1134.

Singer, J.W., 2005. Paclitaxel poliglumex (XYOTAXTM, CT-2103): a macromoleculartaxane. J. Control. Release 109, 120–126.

Wang, M., Thanou, M., 2010. Targeting nanoparticles to cancer. Pharmacol. Res. 62,90–99.

Xiao, W., Luo, J., Jain, T., Riggs, J.W., Tseng, H.P., Henderson, P.T., Cherry, S.R., Rowland,D., Lam, K.S., 2012. Biodistribution and pharmacokinetics of a telodendrimermicellar paclitaxel nanoformulation in a mouse xenograft model of ovariancancer. Int. J. Nanomed. 7, 1587–1597.

Yang, T., Cui, F.-D., Choi, M.-K., Cho, J.-W., Chung, S.-J., Shim, C.-K., Kim, D.-D., 2007.Enhanced solubility and stability of PEGylated liposomal paclitaxel: in vitro andin vivo evaluation. Int. J. Pharm. 338, 317–326.

Yoshizawa, Y., Kono, Y., Ogawara, K.-i., Kimura, T., Higaki, K., 2011. PEGliposomalization of paclitaxel improved its in vivo disposition and anti-tumorefficacy. Int. J. Pharm. 412, 132–141.

Zhang, L., He, Y., Ma, G., Song, C., Sun, H., 2012. Paclitaxel-loaded polymeric micellesbased on poly(e-caprolactone)-poly(ethylene glycol)-poly(e-caprolactone)triblock copolymers: in vitro and in vivo evaluation. Nanomed. Nanotechnol.Biol. Med. 8, 925–934.