Cancer Chemother Pharmacol 2015 76699712
`DOT 101007s0028001528335
`
`ORIGINAL ARTICLE
`
`IDCrossMark
`
`Albumin bound nanoparticle nab paclitaxel exhibits enhanced
`paclitaxel tissue distribution and tumor penetration
`
`Nianhang Chen Carrie Brachmann2 Xiping Liu2 Daniel W Pierce2 Joyoti Dey3
`William S Kerwin3 Yan Li Simon Zhoul
`Shihe Houl Michael Carleton3
`Richard A Klinghoffer3 Maria Palmisanol Rajesh Chopral
`
`Received 31 January 2015 Accepted 21 July 2015 Published online 1 August 2015
`C The Authors 2015 This article is published with open access at Springerlinkcom
`
`Abstract
`
`Purpose
`
`cal efficacy
`
`improved clini
`demonstrates
`nabpaclitaxel
`compared with conventional Cremophor EL
`in multiple tumor
`CrELpaclitaxel
`types This
`study
`the distinctions in drug distribution between
`and CrELpaclitaxel
`and the underlying
`
`explored
`
`nabpaclitaxel
`mechanisms
`
`Methods
`and transcytosis
`of paclitaxel were
`Uptake
`analyzed by vascular permeability assay across human
`cell monolayers The tissue penetration
`endothelial
`of
`paclitaxel within tumors was evaluated by local
`injections
`into tumor xenografts and quantitative image analysis The
`distribution profile of paclitaxel in solidtumor patients was
`assessed using pharmacokinetic modeling and simulation
`that albumin and
`Results
`Live imaging demonstrated
`
`in punctae in endothelial
`
`cells and
`
`paclitaxel were present
`could be observed
`close proximity suggesting
`in very
`cotransport Uptake and transport of albumin nabpacli
`taxel and paclitaxel were inhibited by clinically relevant
`CrEL concentrations Further nabpaclitaxel
`mitotic arrest
`in wider area within xenografted
`than CrEL or dimethyl sulfoxidepaclitaxel
`following
`enhanced paclitaxel
`local microinjection
`demonstrating
`
`causes greater
`
`tumors
`
`Electronic supplementary material
`article doi101007s0028001528335
`contains supplementary
`material which is available to authorized users
`
`The online version of this
`
`E Rajesh Chopra
`
`rachopracelgenecom
`
`penetration
`
`and uptake by albumin within tumors Mode
`ling of paclitaxel distribution in patients with solid tumors
`is more dependent upon trans
`indicated that nabpaclitaxel
`portermediated pathways for drug distribution into tissues
`than CrELpaclitaxel The percent dose delivered to tis
`to be
`sue via transporter mediated pathways
`is predicted
`constant with nabpaclitaxel but decrease with increasing
`CrELpaclitaxel dose
`Compared with CrELpaclitaxel nabpacli
`Conclusions
`taxel demonstrated more efficient transport across endothe
`and cytotoxic induction in
`lial cells greater penetration
`tumors and enhanced extravascular distribution
`xenograft
`in patients that are attributed to carriermediated transport
`These observations are consistent with the distinct clinical
`
`efficacy and toxicity profile of nabpaclitaxel
`
`Taxane Nanoparticle
`Keywords
`nabpaclitaxel Cremophor EL
`
`Albumin
`
`Introduction
`
`Paclitaxel
`agent with a broad
`is a potent antineoplastic
`spectrum of activity against solid tumors and is widely
`used clinically to treat breast ovarian lung prostate and
`other cancers It exerts its cytotoxic effects by interfering
`with microtubule function leading to altered mitosis and
`cell death 1 21 An obstacle to optimal efficacy of pacli
`nature which makes it difficult
`is its hydrophobic
`to formulate and deliver The conventional
`
`formulation
`
`taxel
`
`Celgene Corporation 86 Morris Avenue Summit NJ 07901
`USA
`
`2
`
`3
`
`Celgene Corporation San Francisco CA USA
`Presage Biosciences Seattle WA USA
`
`of paclitaxel requires the drug to be solubilized in the oil
`based solvent Cremophor EL CrEL and ethanol How
`tox
`is associated with significant
`ever CrELpaclitaxel
`
`icities including severe sometimes lethal hypersensitivity
`reactions HSRs and neuropathy in patients Slow infusion
`
`Springer
`
`Abraxis EX2060
`Actavis LLC v Abraxis Bioscience LLC
`1PR201701101 1PR201701103 1PR201701104
`
`

`

`700
`
`Cancer Chemother Pharmacol 2015 76699712
`
`and premedications with corticosteroids
`
`are therefore required for CrELpaclitaxel
`
`and antihistamines
`to prevent HSRs
`35 In addition CrEL alters the disposition of paclitaxel
`
`by forming micelles with highly hydrophobic interiors that
`
`entrap paclitaxel
`
`in circulation impeding drug delivery to
`the tumor exposure to
`
`reducing
`
`tissue and consequently
`
`paclitaxel 6
`
`nabPaclitaxel
`
`is a novel solvent free 130nm albumin
`stabilized nanoparticle formulation of paclitaxel
`In preclin
`ical studies nabpaclitaxel displayed a higher maximum
`tolerated dose increased antitumor efficacy and prolonged
`compared with solvent based taxanes paclitaxel
`survival
`in mice bearing human tumor xenograft
`and docetaxel
`models 17
`81 Clinically nabpaclitaxel
`demonstrated
`superior efficacy and safety to solvent based taxanes In a
`randomized phase 3 study in patients with metastatic breast
`
`showed greater efficacy with higher
`cancer nabpaclitaxel
`and
`response rates and longer time to tumor progression
`a favorable
`safety profile compared with CrELpaclitaxel
`
`9 In this study patients who had second line or greater
`therapy had significantly longer overall survival 9 In a
`
`randomized phase 3 study in patients with non small cell
`lung cancer NSCLC compared with CrELpaclitaxel
`first line administration of nabpaclitaxel
`tolerability and significantly
`
`carboplatin
`
`gemcitabine
`
`demonstrate
`
`infusion duration
`
`compared with
`
`pancreatic
`
`can
`
`carboplatin resulted in better
`improved overall response rates in patients with squamous
`histology 10 Furthermore combination of nabpaclitaxel
`and gemcitabine demonstrated
`significantly longer overall
`survival and improved clinical outcomes
`alone in patients with metastatic
`cancer 11 whereas solvent based taxanes have failed to
`clinically meaningful activity and adequate
`safety over a series of Phase 2 studies 1214 In addition
`being solvent free and devoid of HSRs nabpaclitaxel
`be administered to patients at higher doses during a shorter
`and without corticosteroid premedica
`tion Because of the improved benefitrisk ratio nabpacli
`taxel has been approved in the USA for the treatment of
`locally advanced or
`patients with metastatic breast cancer
`metastatic NSCLC and metastatic adenocarcinoma of the
`
`human breast tumor xenografts 7 Clinically the systemic
`
`proportional
`
`drug exposure of nabpaclitaxel was approximately dose
`from 80 to 300 mgm2 and was independent
`infusion duration 16 whereas CrEL
`of the intravenous
`paclitaxel displays a more than doseproportional
`increase
`in plasma drug exposure and infusion duration dependent
`clearance in a manner consistent with increased entrap
`ment in CrEL micelles in circulation with higher dose 17
`181 In a randomized crossover pharmacokinetic
`study in
`patients with solid tumors the mean fraction of unbound
`paclitaxel was 26 fold higher with nabpaclitaxel com
`pared with CrELpaclitaxel 19 suggesting
`that CrEL
`alters drug distribution in blood
`is hypothesized that nabpaclitaxel utilizes the endog
`enous transport pathways of albumin to achieve enhanced
`tissue distribution Albumin has
`drug delivery and tumor
`
`It
`
`high affinity for hydrophobic drugs including paclitaxel
`po and can be transported across the endothelial barrier
`of blood vessels through binding to gp60 albumin recep
`caveolaemediated
`tor and activating
`endothelial
`tosis 2123 Albumin is highly accumulated in tumors
`as tumor cells use albumin as a major energy and nitrogen
`source through endocytosis and lysosomal degradation 24
`251 In circulation nabpaclitaxel nanoparticles
`dissociate
`in a dynamic process into smaller nanoparticles and eventu
`ally to albuminbound paclitaxel complexes while distribut
`the precise mechanism and full
`ing into tissues However
`effect of albumin facilitated paclitaxel tumor delivery with
`
`transcy
`
`tigate
`
`to further character
`
`and inves
`
`nabpaclitaxel have yet to be completely elucidated
`The present study was conducted
`ize drug tissue distribution by nabpaclitaxel
`the underlying mechanisms The association
`of
`albumin with paclitaxel uptaketransport by human vascu
`cells was investigated
`lar endothelial
`using in vitro drug
`transport and imaging assays The effect of
`formulation
`on paclitaxel distribution within human tumor xenografts
`was measured by evaluating the area and fraction of intra
`tumor mitotic arrest
`
`following microinjection into living
`tumors Finally the distribution profile of nabpaclitaxel
`
`in patients with solid tumors was compared with that of
`using pharmacokinetic modeling and sim
`CrELpaclitaxel
`ulation Taken together these assessments explain the bio
`logical and clinical distinctions between nabpaclitaxel and
`CrELpaclitaxel
`
`Materials and methods
`
`Reagents and materials
`
`Cremophor EL was obtained from SigmaAldrich Buchs
`Switzerland and EMD Millipore Billerica MA USA
`CrELpaclitaxel was obtained from Teva Pharmaceuticals
`
`between
`
`pancreas
`These differences
`in clinical efficacysafety
`nabpaclitaxel and CrELpaclitaxel
`are paralleled by sig
`differences 15 with faster more
`nificant pharmacokinetic
`extensive distribution into the tissue compartments by nab
`the role of formulation in control
`paclitaxel emphasizing
`ling the disposition of hydrophobic drugs Previous stud
`ies have begun to elucidate the mechanistic
`basis of these
`showed that
`differences Preclinical
`
`results
`
`transcytosis
`of nabpaclitaxel across endothelial cell monolayers was
`increased compared with CrELpaclitaxel and nabpacli
`higher intratumoral paclitaxel concen
`taxel achieved 33
`in mice bearing
`
`tration than equal dose of CrELpaclitaxel
`
`Springer
`
`

`

`Cancer Chemother Pharmacol 2015 76699712
`
`701
`
`10 min at room temperature permeabilized in 01
`sapo
`ninPBS for >1 h and blocked with Odyssey blocking buffer
`LiCor Primary antibodies
`used were human albumin
`H126 Santa Cruz Bio early endosome antigen 1 EEA1
`or lysosomalassociated membrane protein 1 LAMP1
`For live widefield microscopy of albumin and fluores
`early passage HUVEC
`labeled paclitaxel uptake
`cently
`monolayers on collagen coated coverslips were incubated
`for 60 min in 0005 iigiiL HSATRITC alone or with
`Flutax2albumin Flutax2 mixed with albumin and diluted
`to a final concentration of 00013 pLgmL Flutax2 Cells
`were pre incubated for 10 min in culture media containing
`CrEL prior to albumin or HSATRITC Albumin punctae
`cell were quantitated by counting 45 fields
`
`Intratumor paclitaxel pharmacodynamic assay
`
`for
`
`the
`
`USA Sellersville PA USA Paclitaxel
`and fluorescent
`labeled paclitaxel Flutax2 Oregon Green 488 conjugated
`paclitaxel were obtained
`from Molecular Probes Inc
`Eugene OR USA nabPaclitaxel Abraxane® and fluo
`rescent labeled nabpaclitaxel nabpaclitaxelFlutax con
`Oregon Green 488 conjugated nabpaclitaxel
`taining 2
`by Celgene Corporation Summit
`were manufactured
`NJ USA Cell
`lines of human pancreas carcinoma MIA
`PaCa2 CRL1420 melanoma A2058 CRL11147 and
`NSCLC H2122 CRL5985 were obtained from American
`Type Culture Collection ATCC Manassas VA USA All
`other reagents and materials were obtained from commer
`cial sources
`
`Cellular paclitaxel uptake assay
`
`To evaluate the effect of CrEL on paclitaxel cellular uptake
`in endothelial cells 20 rgmL nabpaclitaxelFlutax
`was
`incubated with monolayer HUVECs for 4 h at 37 °C with
`CO2 in the absence or presence of varying concentra
`5
`tions of CrEL Cells were trypsinized and washed with
`phosphate buffered saline PBS The cellular uptake of
`Flutax was analyzed by FACS
`
`In vitro vascular permeability assay
`
`taxel
`
`The effect of increasing concentrations of CrEL on pacli
`cells was evaluated in
`transport across endothelial
`an in vitro vascular permeability assay Briefly 20 rgmL
`of DMSOdissolved paclitaxel without CrEL control or
`spiked with increasing concentrations of CrEL within clini
`range 0001 001 003 01 and 03
`was
`cally relevant
`added to medium containing
`human serum albumin
`5
`HSA above a monolayer of HUVEC cells in a transwell
`CO2 The medium at the basolateral
`plate at 37 °C with 5
`side was quantitatively analyzed for amount of paclitaxel
`by LCMS at
`time points The IC50 value was
`indicated
`estimated by nonlinear fitting of data to a sigmoid model
`in which the paclitaxel
`for inhibitory effect
`transcytosis
`of control with no CrEL and 0
`was assumed to be 100
`of control at the infinitively high CrEL concentration
`
`Fluorescent microscopy analysis of albumin
`and paclitaxel uptake by HUVECs
`
`Imaging experiments were performed on early passage
`HUVEC monolayers in phenol red free endothelial
`basal
`FBS and supplements Lonza For con
`medium with 2
`human albumin was
`imaging of albumin uptake
`focal
`5
`incubated with HUVECs plated on collagen coated slides
`for 24 h Slides were moved to ice and were washed in
`HBSS then 2 min in PBS pH 26 and several more times
`in HBSS Cells were fixed in 4
`formaldehydePBS for
`
`Animal studies were conducted
`following all applicable
`international national and institutional guidelines
`care and use of animals To generate xenografts athymic
`NudeFoxnlnu mice Jackson
`laboratories were injected
`subcutaneously with 25 x 106 MIA PaCa2 cells in a 11
`ratio with BD Biosciences Matrigel Matrix Microinjec
`tions were performed using the CIVOTM arrayed microin
`jection device Presage Biosciences Seattle WA USA
`into flank tumors
`by inserting the device transcutaneously
`of anesthetized mice A minimum of three tumors per time
`point were used with 23 replicate injection sites per for
`mulation in each tumor An average drug volume of 3 lit
`was delivered via an extrusion method over an injection
`column length of 6 mm Inactivated
`near infrared dye
`VivoTag680S 50 rgmL was co injected with each drug
`At different
`time points postinjection animals were
`Tumors were harvested and resected fixed in
`euthanized
`buffered formalin for 48 h and scanned on a Xeno
`10
`gen IVIS in the near infrared spectrum excitation 680 nm
`emission 720 nm to confirm injection sites Each tumor
`was cut into 2mmthick cross sections perpendicular
`plane of injection to enable a threedimensional assessment
`the entire injection column Following IVIS imaging
`tumors were processed for standard paraffin embedding
`and histological analysis Four micron sections cut
`from
`2 mm gross level as described above were stained
`each
`with antiphosphohistone H3 pHH3 antibody and Alexa
`Fluor 488 secondary antibody to assess drug induced tumor
`responses mitotic arrest using custom software CIVOan
`alyzerTM Presage Biosciences Seattle Mean fraction val
`ues of pHH3 positive cells were plotted with standard error
`bars as a function of radial distance for each formulation
`and time point To assess the statistical
`significance of dif
`ferences between any pair of formulations a linear mixed
`model approach was used In the model the response to the
`formulation was assumed to be a random
`
`of
`
`CrELpaclitaxel
`
`to the
`
`Springer
`
`

`

`702
`
`Cancer Chemother Pharmacol 2015 76699712
`
`effect and the differential
`response due to nabpaclitaxel
`or DMSOpaclitaxel was assumed to be fixed effects A p
`value <005 adjusted for multiple comparisons indicates
`significant differences
`
`statistically
`
`Pharmacokinetic simulations of nabpaclitaxel
`and CrELpaclitaxel
`
`The population pharmacokinetic
`model of nabpacli
`taxel was described previously 26 The analysis dataset
`150 patients enrolled in 8 clinical studies The
`
`included
`
`studies were conducted in accordance with the ethical prin
`ciples originating in the Declaration of Helsinki and ICH
`Good Clinical Practice guidelines applicable
`regulatory
`requirements and in compliance with the protocols All
`
`patients provided written informed consent
`
`therapy over
`
`solid tumors
`All patients had advanced or metastatic
`nabPaclitaxel was administered intravenously as mono
`the dose range of 80375 mgm2 Paclitaxel
`in whole blood or plasma were measured at
`concentrations
`specified time points for up to 72 h postdose The popula
`tion pharmacokinetic model of CrELpaclitaxel was devel
`oped by Joerger et al 27 using similar methodology The
`included 168 solid tumor
`analysis dataset
`CrELpaclitaxel
`
`patients enrolled in five clinical studies CrELpaclitaxel
`
`was administered intravenously
`the dose range of
`over
`100250 mgm2 with plasma concentration of paclitaxel
`measured up to 48 h postdose The distribution of patient
`demographics age gender body surface area and baseline
`parameters associated with hepatic or renal function were
`comparable between the two analysis datasets 20 27
`The paclitaxel exposure in plasma the central compart
`the first and second
`ment and peripheral
`tissuesorgans
`between nab
`peripheral compartments was compared
`using simulations The con
`paclitaxel and CrELpaclitaxel
`centrationtime profile in the central and peripheral com
`partments for a typical patient
`in the respective analysis
`
`dataset median values for each of the covariates included
`in the final model was simulated using the published typi
`cal model parameters population estimates for the fol
`lowing three scenarios 1 at the approved maximum dose
`for the once every 3 week Q3W dosing schedule which
`is 260 mgm2 over 05h infusion for nabpaclitaxel
`and
`175 mgm2 over 3h infusion for CrELpaclitaxel 2 at
`the commonly used dose for the once weekly QW dosing
`schedule which is 100 mgm2 over 05h infusion for nab
`paclitaxel and 80 mgm2 over 1h infusion for CrELpacli
`taxel and 3 at the same infusion duration of 1h over
`the
`clinical dose range of 80 to 300 mgm2 for both formula
`tions The amount of drug in each peripheral compartment
`was estimated by multiplying the simulated drug concen
`tration in the peripheral compartment with the volume of
`to the compartment
`
`distribution corresponding
`
`Springer
`
`Results
`
`Uptake and transport of paclitaxel
`by albumin and inhibited by Cremophor EL
`
`is facilitated
`
`increasing
`
`In a previous study it has been shown that
`of CrEL can
`concentrations
`inhibit binding of pacli
`taxel to human serum albumin and human umbilical vein
`endothelial cells HUVECs in a dose dependent manner
`7 In this study the effect of CrEL on uptake of nab
`into endothelial cells was evaluated in a fluo
`paclitaxel
`rescence activated cell sorting FACS assay using fluo
`rescent labeled
`nabpaclitaxelFlutax
`nabpaclitaxel
`
`showed
`uptake by
`that nabpaclitaxelFlutax
`Results
`HUVECs was strongly inhibited with increasing concen
`trations of CrEL Fig la HUVECs incubated with nab
`intensity dem
`paclitaxelFlutax had high fluorescence
`onstrating efficient cellular uptake In contrast CrEL at
`28 almost
`of 03
`relevant concentration
`completely inhibited cellular uptake of nabpaclitaxel
`Flutax and reduced fluorescence
`
`levels close to those of
`
`a clinically
`
`unstained cells
`To model the effect of CrEL on extravasation of pacli
`taxel drug transport across intact endothelial cell monolay
`and
`ers was determined
`using unlabeled
`nabpaclitaxel
`CrELpaclitaxel Drugs were added to the media above a
`monolayer of HUVECs in a transwell plate and the level
`across the HUVEC monolayer
`of paclitaxel
`transported
`was determined by liquid chromatographymass spectrom
`etry LCMS The amount of paclitaxel
`transport across
`cells was significantly higher with nabpacli
`endothelial
`than CrELpaclitaxel Fig lb In addition CrEL
`across endothelial
`strongly inhibited paclitaxel
`transport
`required for 50
`cells with fit determined concentrations
`IC50 of 019 012
`inhibition of paclitaxel
`transcytosis
`at 1 2 4 and 24 h respectively Fig lc
`016 and 022
`
`taxel
`
`Localization of albumin and paclitaxel
`vesicles of endothelial cells
`
`in endosomal
`
`endothelial
`
`In a previous study albumin internalized by monolayer
`cells was found in plasmalemmal vesicles but
`not in lysosomes 23 To further determine the mechanism
`and the role of albumin in paclitaxel uptake by endothelial
`cells monolayer HUVECs were incubated with human
`albumin Consistent with the previous
`study internal
`ized albumin was observed in endocytic vesicles some of
`which were early endosomes as indicated by the presence
`of EEA1 protein Fig 2a Additionally very little albu
`min was found in lysosomes as indicated by the staining
`of LAMP1 Fig 2b These findings are consistent with
`endocytic uptake of albumin and transendothelial
`traffick
`than breakdown of the protein
`ing of the molecule rather
`
`

`

`Cancer Chemother Pharmacol 2015 76699712
`
`703
`
`nabPaclitaxel
`
`CrELpaclitaxel
`
`2
`
`Time hr
`
`3
`
`4
`
`0 1 hour
`
`Ai
`
`2 hours
`
`4 hours
`
`IF
`
`24 hours
`
`600
`
`4001
`
`2001
`
`0 L
`1
`
`100 1
`
`50 1
`
`PaclitaxelngmL
`
`Saline
`nabP Flutax
`nabP Flutax + 003 CrEL
`nabP Flutax + 03 CrEL
`
`Fluorescent Intensity
`
`t0
`
`oCrELa
`
`Relativet
`
`7 eTQ C00
`
`100 7
`
`80 1
`
`60 H
`
`40 H
`
`20 1
`
`a
`
`4c
`
`0 a
`
`100
`
`80
`
`60
`
`40
`
`20
`
`Li
`
`a P c
`
`e
`
`1nabPFlutax
`
`0
`
`CrEL
`
`003
`
`03
`
`96
`
`0
`
`0
`
`0001
`
`001
`
`CrEL Concentration
`
`01
`volume
`
`and
`Fig 1 CremophorELinhibited
`cellular
`paclitaxel
`uptake
`cell monolayer a The effect of CrEL
`transport across endothelial
`on uptake of nabpaclitaxelFlutax by HUVECs was evaluated in a
`assay The upper panel shows the FACS inten
`flow cytometrybased
`and the lower panel shows
`in the Flutax channel
`sity distributions
`the derived mean fluorescence intensity MFI values b Paclitaxel
`intact monolayers of HUVEC cells with nabpacli
`transport across
`taxel and CrELpaclitaxel
`c Inhibition of paclitaxel
`transport across
`
`intact monolayers of HUVEC cells by CrEL at 1 2 4 and 24 h The
`at different CrEL concentrations
`amount of paclitaxel
`transcytosis
`in the absence of CrEL is
`calculated relative to paclitaxel
`transport
`to a sigmoid model
`plotted The dashed lines show fit
`for inhibitory
`in which the paclitaxel
`transcytosis was assumed to be 100
`effect
`of control with no CrEL and the baseline maximally inhibited value
`was fixed at 0
`of control
`
`in lysosomes To determine whether albumin associated
`paclitaxel could also be visualized in vesicles monolayer
`HUVECs were incubated with fluorescently labeled pacli
`taxel Flutax2 mixed with albumin and fluorescently
`labeled albumin HSATRITC Live imaging demon
`strated that both fluorescent molecules were present
`in
`punctae and could be observed in very close proximity
`Fig 2c Consistent with vesicle trafficking
`no fluores
`cent paclitaxel was found in lysosomes
`as visualized by
`data not shown The combined
`LysotrackerRed
`that paclitaxel can be found
`in punctae in
`cells and that
`their pattern and proximity to
`suggests that paclitaxel uti
`albumincontaining vesicles
`lizes the same endocytosis and transcytosis mechanism as
`albumin To investigate whether CrEL influences albumin
`uptake endothelial cells were incubated with albumin in
`the presence of increasing concentrations of CrEL and the
`number of albumin punctae was quantified by microscopic
`imaging Consistent with FACS assay results of paclitaxel
`uptake
`a concentration dependent
`decrease in albumin
`
`demonstrate
`
`endothelial
`
`results
`
`of CrEL was observed with both anti albumin
`up to 03
`immunofluorescent staining and HSATRITC Fig 2d
`
`Paclitaxel penetration within tumors facilitated
`by albumin but limited by solvents
`
`To
`
`assess whether
`
`impact drug penetration
`
`formulations measurably
`paclitaxel
`through tumor tissue and uptake
`
`into target cells a novel and highly precise instrument was
`
`facilitating subsequent
`
`quantitative
`
`formulations into different
`
`regions of
`
`used to simultaneously microinject multiple paclitaxel
`the same tumor
`comparisons Equal
`amounts of nabpaclitaxel CrELpaclitaxel and DMSO
`paclitaxel as verified by LCMS data not shown were
`delivered through direct intratumoral microinjection into
`flank human pancreatic MIA PaCa2 tumor xenografts
`Tumors were analyzed 24 48 or 72 h postdrug microin
`jection for mitotic arrest by immunofluorescent staining of
`phosphohistone H3 pHH3 which was used as a phar
`macodynamic
`indicator of paclitaxel activity to monitor
`
`uptake in the presence of clinically relevant concentrations
`
`drug penetration
`
`and tumor cell uptake at defined radial
`
`Springer
`
`

`

`704
`
`El
`
`11
`
`Cancer Chemother Pharmacol 2015 76699712
`
`nalbuminiETRITCHSA
`
`n=21
`
`59
`
`ND
`
`I
`
`0003
`
`9
`
`0300
`0030
`CremophorEL
`
`3000
`
`l00
`
`80
`
`60
`
`40
`
`20
`
`0
`
`fcontrol
`
`cello
`
`punctaep
`
`er
`
`cells was visualized in
`Fig 2 Endocytosed albumin in endothelial
`in close proximity to vesicular paclitaxel a Albu
`endocytic vesicles
`min green was
`some of which were
`found in endocytic
`vesicles
`early endosomes as indicated by colocalization with EEA1 red b
`albumin green colocalized with the lysosomal marker
`Very little
`red a b Are digital closeups of a representative confocal
`LAMP1
`image with nuclei shown in blue c Merged image of live visualiza
`tion of albumin HSATRITC in red and Flutax2albumin green
`
`uptake White arrows indicate vesicles containing both fluorescent
`image of unusually large HUVEC cell with
`molecules
`Inset shows
`a large number of fluorescent punctae d Increasing concentrations of
`cells were incu
`CrELinhibited endocytosis of albumin Endothelial
`bated with albumin or HSATRITC and CrEL for 2 h and the number
`of punctae per cell are graphed The number of cells counted for each
`condition is indicated
`
`distances extending
`
`to all
`
`from the site of injection Exposure
`three formulations of paclitaxel
`induced an increase
`in the number of pHH3positive
`from the site of
`with further
`radial distance
`
`cells which diminished
`
`injection
`
`NSCLC xenografts n = 3 tumors p < 0001 Supplement
`Fig 1
`
`Enhanced paclitaxel distribution to tissues
`in patients with solid tumors mainly
`by nabpaclitaxel
`attributable to a saturable transport process
`
`Fig 3ad Importantly at all
`three time points the area
`of response and the total fraction of pHH3positive
`at a specific radial distance were significantly greater
`
`cells
`
`for
`
`compared with either of the
`microinjected nabpaclitaxel
`solvent based CrELpaclitaxel
`and DMSOpaclitaxel
`p < 001 Fig 3ad
`
`induced a larger
`Similarly microinjected nabpaclitaxel
`increase in both the area of response and total fraction of
`cells arrested in mitosis at 24 h postinjection when com
`A2058 melanoma n = 5
`pared to CrELpaclitaxelinjected
`tumors p < 0001 and DMSOpaclitaxelinjected
`H2122
`
`in
`
`The plasma concentration
`versus time data of paclitaxel
`solid tumor patients treated with nabpaclitaxel or CrEL
`paclitaxel were best described by a three compartment
`pharmacokinetic model 26 271 the central compartment
`plasma and well perfused organs the first peripheral
`tissuesorgans to which the drug was dis
`compartment
`tributed through a saturable transportermediated mecha
`nism and the second peripheral compartment
`tissue
`
`Springer
`
`

`

`Cancer Chemother Pharmacol 2015 76699712
`
`705
`
`24 Hours
`
`48 Hours
`
`72 Hours
`
`d 006
`
`or 005
`004O
`c 003
`0
`47 002
`
`001
`
`0
`
`24 Hours
`
`48 Hours
`
`006
`
`005
`
`004
`
`003
`
`002
`
`001
`
`0
`
`72 Hours
`nabpaclitaxel
`DMSOpaclitaxel
`CrELpaclitaxel
`
`006
`
`005
`
`004
`
`003
`
`002
`
`001
`
`0
`
`200
`
`400
`
`600
`
`800
`
`1000
`
`1200
`
`Radial distance urn
`
`200
`
`400
`600
`800
`1000
`Radial distance urn
`
`1200
`
`200
`
`400
`
`600
`
`800
`
`1000
`
`1200
`
`Radial distance um
`
`Fig 3 nabPaclitaxel induced increased mitotic arrest in a larger area
`within MIA PaCa2 tumor xenografts compared with solvent based
`paclitaxel ac Representative immunohistochemical
`images of mitot
`ically arrested cells in pancreatic MIA PaCa2 xenograft
`tumors at
`24 48 and 72 h following microinjection with equal amounts 12 Rg
`of nabpaclitaxel a DMSOpaclitaxel b and CrELpaclitaxel c
`
`co injected with an inert near infrared dye green
`Each drug was
`sites Mitotically arrested cells were stained
`to delineate injection
`with antipHH3 antibody white and nuclei were stained with DAPI
`
`blue Representative images from a single injection site are shown
`Immunohistochemical
`analysis for pHH3 shows
`to each
`responses
`formulation extending radially from the center of the injection site
`Scale bars 500 Rm d Fraction of arrested cells as a function of dis
`tance from the center of the injection site in pancreatic MIA PaCa2
`tumors n = 8 p <001 at 24 48 and 72 h postinjection
`In each row ad the left middle and right panels show
`xenograft
`
`injection respectively Data are
`
`respectively
`the results at 24 48 and 72 h after
`expressed as mean ± standard error
`
`Springer
`
`

`

`706
`
`a
`
`Cancer Chemother Pharmacol 2015 76699712
`
`Distribution rate
`
`C
`
`Distribution volume
`
`Transporter
`
`driven
`
`Transporter
`
`driven
`
`2000
`
`1500
`
`000
`
`500
`
`nabpaclitaxel
`
`nabpaclitaxel
`
`CrELpaclitaxel
`
`CrELpaclitaxel
`
`Diffusion driven
`
`400
`
`300
`
`100
`
`fdistributionL
`
`Volumeo
`
`Diffusion driven
`
`nabpaclitaxel
`
`nabpaclitaxel
`
`CrELpaclitaxel
`
`CrELpaclitaxel
`
`400
`
`300
`
`200
`
`100
`
`0
`
`60
`
`45
`
`30
`
`µgh
`
`Maximumdistribtionrate
`
`DistribtionrateL
`
`h
`
`Peripheral
`
`compartment
`
`1
`
`saturable
`
`VMTR
`
`MVITR
`
`Central
`
`compartment
`
`Peripheral
`
`compartment 2
`non saturable
`
`VMEL
`
`KM> Elimination
`
`Paclitaxel dose
`
`R1V
`
`and
`
`Fig 4 Pharmacokinetic model
`parameters describing
`pacli
`and CrELpaclitaxel
`taxel
`distribution
`with nabpaclitaxel
`patients with solid tumors a Three compartment PK model
`and
`parameters KMEL = paclitaxel plasma concentration at half VMEL
`KMTR = paclitaxel
`at half VMTR Q = inter
`plasma concentration
`clearance between the central and second peripheral
`compartmental
`
`in
`
`organs to which the drug was distributed through a non
`saturable passive diffusion Fig 4a The rate of both sat
`urable transporterdriven and passive distribution is more
`than doubled with nabpaclitaxel
`versus CrELpaclitaxel
`Fig 4b consistent with previous data showing faster dis
`tribution of drug into tissues with nabpaclitaxel 26 The
`volume of the first peripheral compartment involving satu
`rable distribution of paclitaxel Fig 4c was approximately
`larger when
`ninefold
`administered as nabpaclitaxel
`1650 L versus CrELpaclitaxel 177 L consistent with
`
`In contrast
`
`the
`
`concentrations
`
`deeper penetration of the drug into tissues via transporter
`mediated pathways with nabpaclitaxel
`volume of distribution of
`the second peripheral compart
`ment involving passive diffusion Fig 4c was approxi
`smaller for nabpaclitaxel 754 L versus
`mately 70
`CrELpaclitaxel 252 L
`The modelpredicted drug
`in plasma
`between nabpaclitaxel
`were compared
`and CrELpacli
`the therapeutic dose levels for the Q3W and the
`QW dosing schedules
`respectively Fig 5a The differ
`profiles was
`in the predicted plasma concentration
`relatively small between the two formulations especially
`for the QW dosing regimens The area under
`the plasma
`concentrationtime curve AUC was estimated to be simi
`100 mgm2 over 05h infusion
`lar between nabpaclitaxel
`
`taxel at
`
`ence
`
`Springer
`
`rate VMEL = maximum elimination
`Rw = infusion
`compartment
`rate VMTR = maximum distribution rate from the central
`to the first
`b Distribute rate through transporter driven
`peripheral compartment
`and passive diffusion driven distribution for nabpaclitaxel and CrEL
`paclitaxel c Distribute volume through transporter driven and passive
`diffusion driven distribution for nabpaclitaxel and CrELpaclitaxel
`
`at
`
`80 mgm2 over 1h
`
`3875 h ngmL and CrELpaclitaxel
`infusion 4120 h ngmL
`The model predicted drug exposure in the two periph
`between nab
`eral compartments were also compared
`and CrELpaclitaxel
`paclitaxel
`therapeutic
`dosing
`regimens At the approved maximum Q3W dose for the
`the model predicted that nab
`treatment of breast cancer
`paclitaxel 260 mgm2 over 05h infusion would deliver
`considerably more drugs into tissues via transporter medi
`ated pathways than CrELpaclitaxel 175 mgm2 over 3h
`infusion Fig 5b In contrast CrELpaclitaxel was pre
`dicted to deliver drugs into tissues mainly via passive dif
`fusion Fig Sc In recent years weekly CrELpaclitaxel
`with a lower dose and shorter infusion duration has
`and less toxic dos
`been considered as a more effective
`ing regimen than the Q3W dosing regimen 29 301 For
`the weekly dosing regimen nabpaclitaxel 100 mgm2
`over 05h infusion was predicted to deliver consider
`ably more drugs into the tissues via transportermediated
`than CrELpaclitaxel 80 mgm2 over 1h infu
`pathways
`sion Fig 5b even though their plasma AUC was simi
`lar Fig 5a
`The relationship between dose and distribution in tis
`sues was further assessed by assuming the same infusion
`duration 1 h for both formulations with the exposure in a
`
`

`

`Cancer Chemother Pharmacol 2015 76699712
`
`707
`
`Plasma
`
`Tissues via transporter
`
`C
`
`Tissues via diffusion
`
`nabPTX 260 mgm2 05 h IV Q3W
`CrELPTX 175 mgm2 3 h IV Q3W
`
`400
`
`320
`
`I
`
`240
`
`172
`
`160
`
`80
`
`8
`
`nabPTX 260 mgm2 05
`IV Q3W
`Cr ELPD 175 mgm2 3 h IV Q3W
`
`400
`
`320
`
`a 240
`
`160
`
`80
`80
`
`173
`
`CL
`
`S
`
`nabPD 260 mgm2 05 h IV Q3W
`
`CrELPTX 175 mgm2 3 h IV Q3W
`
`100000
`
`10000
`
`1000
`
`100
`
`plasamngmL
`
`in
`
`Drugconcentration
`
`a
`
`Every3weeks
`
`0
`
`12
`
`24
`
`36
`
`48
`
`60
`
`72
`
`2
`
`24
`
`36
`
`48
`
`60
`
`72
`
`12
`
`24
`
`36
`
`48
`
`60
`
`72
`
`Time after dosing hour
`
`Time after dosing hour
`
`Time after dosing hour
`
`10000
`
`nabPTX 100 mgm2 05 h IV QW
`
`160
`
`CrELPTX 80 mgm2 1 h IV QW
`
`nabPTX 100 mgm2 05 h
`
`IV QW
`
`CrELPTX 80 mgm2 1 h IV QW
`
`200
`
`nabPTX 100 mgm2 05 h IV QW
`
`CrELPTX 80 mgm2 1 h IV QW
`
`160
`
`120
`
`80
`
`40
`
`peripheralcompartmentm
`
`g
`
`in
`
`Drug
`
`120
`
`80
`
`peripheralcompartmentm
`
`g
`
`in
`
`Drug
`
`1000
`
`100
`
`plasamngmL
`
`in
`
`Drugconcentration
`
`GJ
`
`GJ
`
`GJ
`
`0
`
`0
`
`12
`
`24
`
`36
`
`48
`
`60
`
`72
`
`0
`
`12
`
`24
`
`36
`
`48
`
`60
`
`72
`
`12
`
`24
`
`36
`
`48
`
`60
`
`72
`
`Time after dosing hour
`
`Time after dosing hour
`
`Time after dosing hour
`
`Fig 5 Model predicted drug exposure in