Contents lists available at SciVerse ScienceDirect
Journal of Controlled Release
j ourna l homepage: www.e lsev ie r .com/ locate / jconre l
Plasmonic photothermal therapy increases the tumor mass penetration ofHPMA copolymers
Adam J. Gormley a,b,1, Nate Larson b,c, Afsheen Banisadr a,b, Ryan Robinson a,b, Nick Frazier a,b,Abhijit Ray b,c, Hamidreza Ghandehari a,b,c,⁎a Department of Bioengineering, University of Utah, Salt Lake City, UT 84112, USAb Center for Nanomedicine, Nano Institute of Utah, University of Utah, Salt Lake City, UT 84112, USAc Department of Pharmaceutics and Pharmaceutical Chemistry, University of Utah, Salt Lake City, UT 84112, USA
⁎ Corresponding author at: 36 S. Wasatch Dr., Salt LTel.: +1 801 587 1566; fax: +1 801 585 0575.
E-mail address: [email protected] Present address: Department of Materials, Imperial
Effective drug delivery to tumors requires both transport through the vasculature and tumor interstitium.Previously, it was shown that gold nanorod (GNR) mediated plasmonic photothermal therapy (PPTT) iscapable of increasing the overall accumulation of N-(2-hydroxypropyl)methacrylamide (HPMA) copolymersin prostate tumors. In the present study, it is demonstrated that PPTT is also capable of increasing the distri-bution of these conjugates in tumors. Gadolinium labeled HPMA copolymers were administered to micebearing prostate tumors immediately before treatment of the right tumor with PPTT. The left tumor servedas internal, untreated control. Magnetic resonance imaging (MRI) of both tumors showed that PPTT was ca-pable of improving the tumor mass penetration of HPMA copolymers. Thermal enhancement of delivery,roughly 1.5-fold, to both the tumor center and periphery was observed. Confocal microscopy of fluorescentlylabeled copolymers corroborates these findings in that PPTT is capable of delivering more HPMA copolymersto the tumor's center and periphery. These results further demonstrate that PPTT is a useful tool to improvethe delivery of polymer–drug conjugates.
The conjugation of hydrophobic anticancer drugs to water-solublepolymers represents an effective way of solubilizing them in blood plas-ma, prolonging blood circulation half-life, targeting biodistribution totumors and overcoming multidrug resistance [1]. In this way, drugs canbe retained in the blood and specifically delivered to the cancerous tissuewith dramatically reduced accumulation in healthy organs. While theadvantages of targeted delivery using polymer–drug conjugates arewell known, clinical translation has been slow. There are many reasonswhy this is the case, including poor drug release kinetics and carrier bio-compatibility. However, the major barrier to obtaining favorable clinicaloutcome remains limited tumor and cancer cell delivery [2].
There are many available techniques to improve the delivery ofpolymer–drug conjugates. The most obvious and widely used methodinvolves tailoring the size of the conjugates so that the therapeutictakes advantage of the increased vascular permeability of tumors tomacromolecules. Coined the ‘Enhanced Permeability and Retention’(EPR) effect, large intercellular and transcellular openings between
ake City, UT, 84112–5001.
(H. Ghandehari).College London, London, SW7
rights reserved.
endothelial cells that line the tumor vasculature allow macromoleculesup to roughly 1 μm in size to partition from the blood and enter thetumor interstitial space with limited lymphatic drainage [3,4]. Anothercommon approach involves the conjugation of biorecognizable motifssuch as peptides or antibodies for cancer cell receptor-mediated targeting[5]. Such active targeting then enables drug carriers to specifically bind tocancer cells which express the targeted receptor and trigger internaliza-tion and drug release. Finally, a number of other pharmacologic basedmethods for improving delivery have been shown including treatmentwith angiotensin to raise the patient's blood pressure [6–8], applicationof nitroglycerin [9] or heme oxygenase-1 [10], pre-treatment with vascu-lar disrupting [11,12] or anti-angiogenic agents [13,14], as well as directinjection of extracellularmatrix enzymes to reduce the interstitial density[15]. Each of these tools provides greater selectivity of nanocarrier deliv-ery to tumors.
Another technique which is shown to improve the delivery ofnanocarriers involves treating the tumors with hyperthermia. Recentfindings, for example, have shown that hyperthermia can increase therate of both endo- and phagocytosis which may then potentiate mac-romolecular uptake and intracellular delivery [16,17]. At the vascularlevel, when tumors are heated up to 43 °C, tumor blood flow canincrease roughly two-fold [18]. This change in blood flow then in-creases the overall availability of macromolecules to extravasate.The resulting increased vascular pressure and heat-induced cytoskel-etal injury then causes endothelial cell damage [19–21]. This causes
131A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
further expansion of the intercellular openings and therefore in-creased vascular permeability to macromolecules [22,23].
The observed changes in tumor vascular dynamics with heatinghave been leveraged to improve the delivery of various nanomedicines.In particular, tumor hyperthermia has been used to facilitate the deliv-ery of liposomes [24–28]. The application of heat is shown to enhancethe extravasation of liposomes in a thermal dose dependent mannerfor up to 6 hours after heat treatment [29]. Additionally, this effectwas also dependent on nanoparticle size where the larger systemsexhibited the greatest increase in overall deliverywith heat [30]. Precisecontrol over heating, however, is necessary as vascular collapse andblood flow stasis is probable when temperatures rise above 43 °C.
A major challenge with treating tumors with hyperthermia lies inthe ability to effectively deliver the appropriate thermal dose in a sitespecific manner. Evolving technologies such as radiofrequency abla-tion as well as high-intensity focused ultrasound (HIFU) have provenuseful in this regard [31], though these methods are not selective to-wards cancerous tissue and therefore rely on the physician to choosethe regions which should receive thermal therapy. As the margins oftumor and normal tissue are often unknown, this adds greater risk ofinjury to healthy tissue. A recent method of selectively delivering heatto tumors takes advantage of the plasmonic properties of colloidalgold. Of special interest in this regard is the unique capacity of thesecolloids to scatter and absorb light. Under conditions of surfaceplasmon resonance (SPR), strong light absorption results in particleheating [32]. When located within a tumor mass, direct tissue heatingcan occur with laser light excitation by plasmonic photothermal ther-apy (PPTT) [33–35]. This heating process can be used as a means toselectively induce tumor hyperthermia with therapeutic intentions[36–38].
Recent studies have shown the utility of PPTT to improve thedelivery of other nanomedicines [39–43] as well as radiotherapy[44]. In each of these studies, PPTT was applied to heat tumorsbetween 42 and 45 °C and a resulting increase in conjugate accu-mulation was observed. In a previous study, the tumor accumula-tion of heat shock targeted N-(2-hydroxypropyl)methacrylamide(HPMA) copolymers was evaluated in combination with PPTT[40]. A peptide which has known affinity for an extracellular heatshock protein was incorporated in the polymer design to specifical-ly target cancer cells treated with hyperthermia. PPTT for 10 min at43 °C caused a burst accumulation of the conjugates for up to4 hours. After 4 hours, while the untargeted conjugates diffusedback out of the tumor, the heat shock targeted conjugates wereretained for an extended period of time (up to 12 hours) due tocell specific targeting [40]. These results provided evidence forthe utility of this approach.
What remains unknown, however, is the tumor tissue distributionof HPMA copolymers after delivery enhancementwith PPTT. This infor-mation is important because drug delivery is not evenly distributed dueto tumor vascular heterogeneity, particularly for nanomedicines whichare larger in size [45,46]. The objective of this studywas to visualize thedistribution of HPMA copolymers in prostate tumors after treatmentwith PPTT.
2. Materials and Methods
2.1. Synthesis and characterization of PEGylated GNRs
Poly(ethylene glycol) (PEG) coated gold nanorods (GNRs)were syn-thesized as described previously [40]. GNR size and shape were charac-terized by transmission electron microscopy (TEM) and the lightabsorption profile was measured by UV spectrometry. Zeta potentialwas calculated in deionized (DI)water bymeasuring its electrophoreticmobility using laser Doppler velocimetry (Zetasizer Nano ZS, MalvernInstruments Ltd, Worcestershire, UK).
2.2. Synthesis and characterization of HPMA copolymers
HPMA [47], aminopropylmethacrylamide-1,4,7,10-tetraazacy-clododecane-1,4,7,10-tetraacetic acid (APMA-DOTA) [48], 5-[3-(methacryloylaminopropyl)thioureidyl] fluorescein (APMA-fluoresce-in) [49], and 3-[(N-methacryloylglycyl)glycyl]thiazolidine-2-thione(MA-GG-TT) [50] comonomers were synthesized as described previ-ously. The precursor copolymer conjugates contained the reactive car-boxyl groups (thiazolidine-2-thione) so that future studies withthe same copolymer could incorporate targeting peptides into their de-sign. In the present study, these groups were hydrolyzed to obtainuntargeted conjugates. Copolymerization was performed by reversibleaddition-fragmentation chain-transfer (RAFT) polymerization using2-cyano-2-propyl dodecyl trithiocarbonate as the chain transfer agentand VA-044 as the initiator in a DMF/MeOH (90:10) co-solvent at50 °C for 24 hours. The unpurified product was then dissolved in DIwater with gadolinium (Gd) (III) acetate hydrate (1.2 mol equivalent)and the pH was raised between 5.0 and 5.5. This solution was stirredovernight followed by addition of ethylenediaminetetraacetic acid(EDTA) to remove excess Gd (EDTA:GD, 1:1). The product was then fil-tered, dialyzed and lyophilized to obtain the final product.Mw,Mn, andMw/Mn were estimated by size exclusion chromatography (SEC) usingHPMAhomopolymer fractions of knownmolecularweight. The Gd con-tent was quantified by inductively coupled plasma mass spectrometry(ICP-MS) against a standard curve. Fluorescein labeled polymers weresynthesized as described previously [40].
To calculate the conjugate's longitudinal relaxivity, four different con-centrations of copolymer (0.1 to 0.015 mM polymer) were prepared inDI water and placed in a Bruker BioSpec 7.1 T horizontal-bore MRI. T1wasmeasured by an inversion recovery fast spin-echo imaging sequenceusing inversion times of 50, 100, 300, 500, 800, 1000, 2000, 4000, 7000and 8000 ms, echo time (TE) of 4.2 ms, and repetition time (TR) of12000 ms. T1 for each vial was calculated using Bruker software andthe relaxation rate (R1=1/T1) was plotted against Gd equivalent con-centration. Relaxivity was measured as the slope of this plot.
2.3. Prostate tumor model
Animal experimentswere performed in accordancewith the Institu-tional Animal Care and Use Committee (IACUC) of the University ofUtah. Four-to-six week old female athymic (nu/nu)micewere anesthe-tized using 2% isoflurane and bilaterally inoculated with 107 DU145prostate cancer cells in 200 μl phosphate buffered saline (PBS) on theflank of each animal. Animals were used in the study once the averagetumor volume reached 50–100 mm3 (usually 10–21 days).
2.4. MR imaging
Prior to the experiment, those animals which were ultimatelytreated with PPTT received an intravenous dose of PEGylated GNRs(9.6 mg/kg) roughly 48 hours before each experiment. This providedenough time for the GNRs to circulate and passively accumulate in thetumor tissue [51]. The animals in the laser only group did not receiveGNRs. Each animal was then anesthetized with 2% isoflurane, placedwithin a Bruker BioSpec 7.1 T horizontal-bore MRI and an axial T1flash pre-scan was taken. Each tumor was then swabbed with 50%propylene glycol to enhance laser penetration depth [52], and theGd labeled HPMA copolymers were then intravenously administered(0.03 mmol Gd/kg) in saline. Immediately after injection, the righttumor was radiated for 10 min using an 808 nm fiber coupled laserdiode (Oclaro Inc., San Jose, CA) with collimating lens (Thorlabs,Newton, NJ). Intratumoral temperature was monitored using a 33gauge needle thermocouple (Omega, Stamford, CT), and the laserpower (roughly 1.2 W/cm2) was directly controlled so that tumortemperature was maintained between 42 °C and 43 °C when treatedwith PPTT. Tumors on the left flank served as internal controls.
132 A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
Immediately following laser treatment, the bed was placed backinto the MRI and axial section T1 flash images of both tumors weretaken. Two hours after laser treatment, a series of multislice T1 flashimages were taken to provide anatomical information. Also, quantita-tive T1 MR images were obtained using an inversion recovery fastspin-echo pulse imaging sequence (slice thickness=1 mm, numberof slices=8). This was done using the same acquisition parameterswhen the conjugate's relaxivity was measured. In one animal, bothT1 flash and T1 inversion recovery images were acquired before treat-ment (pre-scan), and then every 42 min for 5 hours after treatment.This was done to obtain time-dependent information on conjugatedelivery.
2.5. Image analysis
The quantitative T1 relaxation maps were calculated from the in-version recovery data sets using QuickVol II, a plugin for Image J[53]. The images were then prepared in the following way. Usingthe T1 flash images, regions of interest (ROI) were drawn aroundboth tumors and a mask was created for each slice. The mask wasthen used to isolate only the tumor data from the T1 relaxationmaps. Due to some noise in the data, pixel outliers were removedusing Image J. Each map was then overlayed onto its respective T1flash image and a montage was created to display the whole tumorvolume or the time dependent information as well as a 3D surfaceplot.
To compare the accumulation of polymers in both tumors, theaverage relaxation (R1=1/T1) was calculated in the skin aroundthe tumor as well as the tumor's center and periphery using ImageJ. This was done by first drawing ROIs on the T1 flash images (withoutthe T1 map overlay) for both tumors. These ROIs were then used tocalculate the average relaxation in each of these regions on the T1 re-laxation map. The TER for each of these regions was then calculatedby dividing R1 in the right tumor by R1 in the left tumor (TERPPTT=R1PPTT/R1control, or TERlaser only=R1laser only/R1control). This was donefor each mouse, tumor and slice. A normalized histogram of R1 valuesfor whole tumors (center plus periphery) was also obtained usingImage J software. This data was then graphically represented usingGraphPad Prism.
2.6. Histology
After MR imaging, the animals were euthanized by CO2 inhalationand the tumors were removed and fixed in neutral buffered formalin.Samples were then dehydrated, paraffin-embedded and cut into4-micron thick sections. Immunohistochemical (IHC) analysis of Fac-tor VIII expression, an endothelial cell marker, was then performed onsome sections using a Factor VIII rabbit polyclonal antibody (Dako,Carpinteria, CA) and a biotinylated rabbit IgG secondary antibody.Positive signal was visualized using a streptavidin-HRP system, utiliz-ing DAB (3,3′-diaminobenzidine) as the chromogen. All sections werethen counterstained with hematoxylin. Washing with an iodine solu-tion, followed by sodium thiosulfate, removed any precipitates. Final-ly, sections were dehydrated in alcohol, rinsed in xylene, coverslippedand imaged.
2.7. Fluorescence imaging
Prior to the experiment, animals received an intravenous dose ofPEGylated GNRs (48 hours before) and both tumors were swabbedwith 50% propylene glycol (10 min before). Each animal (N=4)was then intravenously administered 7.0 mg of FITC labeled HPMAcopolymers and the right tumor was lased in the same way as theMRI experiment. Two hours after treatment with PPTT, each animalwas then administered 5 mg of rhodamine labeled Concanavalin A(Vector Laboratories, Burlingame, CA) and euthanized 5 min later to
visualize the vasculature. Both tumors were then collected, immedi-ately placed in Tissue-Tek® OCT™ compound, frozen and cryo-sectioned into 12 μm thick sections (Leica CM3050, Wetzlar, Germa-ny). Immediately before imaging, slides were dried and a cover slipwas mounted using Cytoseal 60 diluted 1:5 in toluene. Large fluores-cent imaging mosaics were acquired using a Nikon A1 confocal lasermicroscope system with a 10x objective.
2.8. Statistics
Statistical analyses were performed in GraphPad Prism. Compari-sons between two groups (left vs. right tumors), were performedusing a two-tailed, Welch-corrected unpaired t-test. p-Values lessthan 0.05 were considered statistically significant. Data reported asmean±SEM.
3. Results
3.1. GNR and HPMA copolymer synthesis and characterization
The GNRs were synthesized to be 58.6±5.7×15.4±0.8 nm in sizewhich corresponds to an aspect ratio of 3.8 and a SPR peak at 800 nm(Fig. 1A–B, Table 1). After PEGylation, the GNRs had a slightly nega-tive zeta potential of −10 mV. These GNRs were found to be stablein a wide variety of buffers and solvents due to steric protectionfrom aggregation.
The HPMA copolymers were synthesized by RAFT copolymerizationto be roughly 65 kDa so that theywere slightly above renal threshold totake advantage of the EPR effect (Fig. 1C, Table 1). In order for the copol-ymers to be imaged by MRI, they contained APMA-DOTA comonomerswhich chelate Gd. For copolymers used for fluorescent imaging,APMA-FITCwas used instead. An additional comonomerwith a reactivecarboxyl group, MA-GG-TT, was also incorporated so that future studiescould incorporate receptor-mediated active targeting using the samecopolymer if needed. In the present study, the TT groupwas hydrolyzedto obtain an untargeted conjugate. The copolymer Gd content wasfound to be similar to HPMA copolymer conjugates synthesized previ-ously [48].
3.2. HPMA copolymer tumor delivery 2 hours after treatment by MRI
In this experiment, half the animals were administered PEGylatedGNRs (N=3) and the other half saline (N=3) 48 hours before MRimaging. On the day of the experiment, each animal was administeredHPMA copolymer-Gd conjugates prior to laser radiation of the righttumor for 10 min. The tumors in animals which were previouslygiven GNRs exhibited rapid heating which was maintained near43 °C. Animals without GNRs (laser alone) were slightly heatedwhich is consistent with previous results [39]. Two hours after lasertreatment, quantitative MR imaging shows significantly enhanced co-polymer delivery to tumors treated with PPTT (Fig. 2). While the lefttumor (control) had some polymer accumulation, the right tumor(PPTT) displayed signs of greater copolymer accumulation. As expected,in both tumors the deliverywas not evenly distributed likely due to vas-cular heterogeneity. However, the extent of distribution does appeargreater in tumors treated with PPTT. In animals not previously givenGNRs, differences between the laser radiated (right) and control (left)tumors were not apparent (Fig. 2).
When the relaxation rate (R1) was quantified for the skin around thetumors as well as the tumor's center and periphery and expressed as athermal enhancement ratio (TER, R1PPTT/R1control, or R1laser only/R1control),a clear trendwas observed (Fig. 3A). Treatmentwith PPTT resulted in in-creased copolymer delivery to each of these regions. PPTT caused a1.36-fold increase in delivery to the skin, though this difference wasnot statistically significant relative to the laser only group due to largevariability in the data and some enhancement to the skin with laser
Fig. 1. GNR characterization and HPMA copolymer schematic. GNRs were synthesized to be 58.6 x 15.4 nm in size (A) with an SPR peak at 800 nm (B). Scale bar, 100 nm. TwoHPMA copolymers were synthesized for this study (C). The first was copolymerized with HPMA (i), DOTA to chelate Gd to provide MRI contrast (ii), and a hydrolyzed reactivecarboxyl group to enable targeting in future studies using the same copolymer (iv). The other was copolymerized with HPMA (i), APMA-FITC for fluorescent imaging (iii), and ahydrolyzed reactive carboxyl group for the same reasons (iv).
133A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
alone (ns, p=0.434). For the tumor's center, PPTT significantly raised thecopolymer concentration by 1.42-fold relative to those animals treatedwith laser only (*p=0.049). The periphery of the tumors, regions ofthe tumors not including their center, saw the greatest thermal enhance-ment of 1.54-fold and was the most significant (*p=0.016). When theright tumors were treated with laser alone, i.e. no GNRs present, thesame effect was not observed (Fig. 3A). Treatment with laser alone didnot cause any appreciable increase in copolymer delivery to the skin,tumor center or periphery.
Plotting a histogram of R1 values for control and PPTT treatedtumors provides some additional information (Fig. 3B). In controltumors, the majority of its volume comprised of relatively low R1and therefore polymer concentration values. This distribution wasGaussian (R2=0.83) and centered at roughly R1mean=1.02 s−1
with a narrow standard deviation of σ=0.28. For tumors treatedwith PPTT, the distribution was less Gaussian (R2=0.67), centeredat R1mean=2.31 s−1 and exhibited a much broader distribution ofσ=1.03. This indicates that the majority of the tumor volume re-ceived a greater andmore variable distribution of copolymer delivery,though overall its entire volume received more copolymers. All
Table 1Physicochemical characteristics of GNRs and HPMA copolymers.
regions in the PPTT histogram which do not overlap with the controlhistogram received some benefit of delivery due to therapy.
3.3. HPMA copolymer tumor delivery over time
To better understand the kinetics of delivery, one animal was im-aged for 5 hours after treatment with PPTT (Fig. 4). Similar to previousfindings [40], the majority of the dose was delivered to tumors withinthe first hour of treatment. Two to 3 hours after treatment accumula-tion increased with time. In the control tumor, copolymer deliverywas observed as expected, though the increase in R1 was slight. In theright tumor treated with PPTT, large differences in R1 were observedmostly within the first hour.
3.4. Fluorescence imaging of HPMA copolymer delivery
When the tumor delivery 2 hours after treatment was visualizedby fluorescent imaging, a similar trend was observed (Fig. 5). Treat-ment with PPTT facilitated more copolymer delivery overall. Deliv-ery enhancement occurred mostly in the outer rim of the tumor,
MA-GG-TT(mol %)
Gd content(mmol Gd/g polymer)
Apparent Mw
(kDa)Mw/Mn Relaxivity
(s−1 mM Gd−1)
5 0.37 64.9 1.3 7.15 – 62.4 1.4 –
Fig. 2. HPMA copolymer delivery 2 hours after treatment. Laser treatment of the right tumor in animals previously administered GNRs (PPTT) facilitated significant enhancement ofHPMA copolymer delivery in terms of both accumulation and overall tumor distribution (A, top row). Laser alone did not cause any increased delivery (A, bottom row). A 3D surfaceplot provides better visualization of this effect at a single slice (B).
134 A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
though increased delivery in the tumor's center was also observed.By this method, overall enhancement of delivery was less pro-nounced than those results obtained by MRI. In both tumors, vascular-ization density appeared the same. Interestingly, HPMA copolymerdelivery did not directly correlate with location of blood vessels.
3.5. Histology of tumors after treatment with PPTT
After imaging, the right and left tumors were evaluated for dam-age by histology (Fig. 6). In both the control and PPTT treated tumors,many blood and lymphatic vessels were observed throughout andappeared to be intact (Fig. 6A–D). There was, however, one major dif-ference between the groups. In all PPTT treated tumors, areas of tissuedamage and cell death were observed (Fig. 6E–F). These regions ofdamage were typically less than 20% of the tumor's total volumeand were usually confined to the tumor's center. In some cases, intactblood vessels were found in these areas indicating that vessel injurywas not necessarily the reason for damage (Fig. 6F). Also, tumorstreated with PPTT did not appear to have as many dividing cells fur-ther suggesting that some damage had occurred.
4. Discussion
It was shown previously that PPTT can be used to effectively deliv-er greater numbers of nanocarriers to solid tumors [39–43]. But sim-ply delivering more drugs to tumors may not necessarily improveoverall delivery to cancerous cells. For example, excessive deliveryof drug to only perivascular regions and not areas which are distantfrom viable vasculature may not improve overall treatment outcome.For this reason, delivery strategies which also increase tumor masspenetration are equally important [45].
There are many tumor tissue abnormalities that resist efficienttransport of small and macromolecules to all cells. Most importantly,unregulated angiogenesis causes blood vessels to formwith an abnor-mal and disorganized architecture [54,55]. The spatial distribution ofblood vessels lacks order and continuity which ultimately generates aheterogenous distribution of tissue which is poorly perfused [56].Upon entering the tumor microenvironment, efficient transportthrough the interstitium is further restricted due to several other ab-normalities [57]. High extracellular matrix (ECM) densities in thetumor interstitium prevent large objects such as nanocarriers fromdiffusing freely [58]. Also, poor lymphatic drainage and related high
Fig. 3. Image analysis of HPMA copolymer delivery. Treatment of tumors with PPTT was capable of significantly enhancing the delivery of HPMA copolymers to the tumor's centerand periphery 2 hours after treatment (A, left). Treatment with laser alone, absence of GNRs, did not increase delivery. Representative ROIs for this analysis is also shown (A, right).A histogram of R1 values of both control and PPTT treated tumors is shown (B). These data show the capability of PPTT to increase tumor mass distribution. *Indicates a statisticallysignificant difference (pb0.05) by t-test. Error bars represented as±standard error of the mean.
135A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
interstitial fluid pressure (IFP) restricts fluid transport, particularly to thetumor's center where pressure is the highest [59]. Finally, nonspecificbinding to ECM or cellular components due to charge-charge or Vander Waals interactions restricts motion.
Using hyperthermia to improve the delivery of polymer–drug conju-gates may offer distinct advantages towards maximizing nanocarrierdelivery in the context of interstitial transport. To describe interstitialtransport or flux, Ji, one must consider the contributions to convective,Jc, and diffusive, Jd, molecule transport [57,60]:
Ji ¼ Jd þ Jc ¼ −D∂c∂x−CRFK
∂p∂x ð1Þ
whereD is the diffusion coefficient of the nanocarrier, C and ∂C/∂x is theconcentration and concentration gradient respectively, RF is the retar-dation factor, K is the tissue's hydraulic conductivity, and ∂p/∂x is thepressure gradient in the tissue. During mild hyperthermia (T≤43 °C),it is known that tumor blood flow increasesmost likely due to increasesin blood flow from the host vessels or increased cardiac output [18]. Theresult is increased microvascular pressure and therefore passive
Fig. 4. HPMA copolymer delivery over time. The majority of
dilation of tumor blood vessels. Also during hyperthermia it has beenfound that IFP decreases [61]. The result of these two phenomenon isan increase in ∂p/∂x and therefore interstitial convective transport.
Regarding how such a combination might be beneficial from a dif-fusive standpoint, one must look at the polymer's diffusion coefficientrelative to other nanocarriers and the temperature of the environ-ment. Using the Stokes-Einstein equation, D is calculated using thefollowing relationship:
D ¼ kT6πnR
ð2Þ
where k is Boltzmann's constant, T is the absolute temperature, n isviscosity and R is the carrier's hydrodynamic radius. Here, it isshown that diffusive transport is higher for smaller nanocarriers(lower R) and at elevated temperatures (higher T, lower n). Also,as shown previously [40], hyperthermia is able to increase the over-all amount of copolymer delivery which therefore causes an in-crease in ∂C/∂x.
delivery occurs within the first hour of PPTT treatment.
Fig. 5. Fluorescent imaging of HPMA copolymer delivery. Two hours after treatmentwith PPTT, HPMA copolymer delivery (green) enhancement in the tumor's peripheryand center is observed relative to untreated controls. This effect is most pronouncedin the tumor's periphery. Blood vessel (red) density does not appear to be affectedby treating with PPTT. Scale bar=1 mm.
136 A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
The above highlights a) the importance of nanocarrier size on in-terstitial diffusive transport, and b) the advantages that mild hyper-thermia provides for improving both convective and diffusivetransport. In this context, using heat to drive the distribution ofpolymer–drug conjugates which are typically less than 15 nm in hy-drodynamic diameter may offer the best opportunity to increasetumor tissue delivery. This may also be true as flexible, linear poly-mers such as HPMA copolymers have been shown to have greatertransport properties than branched or rigid systems [62,63]. Of course,a balance between favorable mass penetration and unfavorable renal
Fig. 6. Histology of control and PPTT treated tumors. No differences between the tumorperiphery of both control and PPTT treated tumors were observed (A–B, 10× objective,scale bar=100 μm). IHC staining of blood vessels (BV) in the periphery did not provideevidence of damage in either group (C–D, 20× objective, scale bar=50 μm). The centerof tumors treated with PPTT had evidence of cell and tissue damage most likely due toexcessive heating in this region (E, 10× objective, scale bar=100 μm). A higher resolutionview of this area shows the presence of capillary blood vessels which appear viabledespite surrounding damage (F, 20× objective, scale bar=50 μm).
clearance for these polymers is a necessary consideration in therapydesign.
In this study, PPTT was used to selectively heat prostate tumorsbetween 42 and 43 °C to facilitate the delivery of HPMA copolymers.Using PPTT in this way decreases the chances of heating healthy tis-sue due to tumor specific delivery of GNRs by EPR and provides ahigh degree of control over heating. It is also possible that using tissueembedded antennas for energy generation, similar to brachytherapy,provides an advantage in terms of heat distribution. Also, PPTT mayhave other unknown benefits as it has clearly been shown to improvethe delivery of both albumin [39] and HPMA copolymers [40], where-as previous studies using other methods to induce hyperthermia havenot shown greater delivery of each of these [30,64].
When PPTT was used to direct the delivery of Gd labeled HPMAcopolymers to the right tumor of the animal and imaged by MRI, aclear difference between the tumors was observed (Fig. 2). Theright tumor exhibited greater T1 contrast and therefore polymer con-centration than the left, untreated tumor. In these images, it is ob-served that these treated tumors did not just receive higher amountof copolymer accumulation. Rather, more of the tumor volume overallreceived greater delivery suggesting greater tumor mass penetrationof these copolymers due to PPTT. This is in stark contrast to tumorstreated with laser but not previously administered GNRs, showingthat PPTT is in fact responsible and not just laser radiation itself.Though it would be ideal to have an even distribution of high copoly-mer concentrations throughout the whole tumor volume, the imagesshow that this is not the case. There still exist regions in the tumorthat did receive less copolymer. These regions are likely to be the ne-crotic core of the tumor and are thus difficult to reach.
When these images were analyzed for thermal enhancement ofdelivery based on region (skin, tumor center and periphery), ineach case thermal enhancement was observed (Fig. 3A). For theskin, this difference was not statistically significant from laser alonethough enhanced delivery to the skin is expected during heating.Both the tumor's center and periphery had a roughly 1.5-fold increasein delivery, where delivery to the center was barely significant andthe periphery highly significant. This analysis further suggests thatgreater mass penetration to the tumor's center was achieved. This in-terpretation should be taken with a degree of caution however. UsingMRI, it is very difficult to delineate the viable from the necrotic re-gions of the tumor. The ROIs were drawn based on what appearedto be the tumor's center and periphery, but this does not necessarilyreflect the true location of the tumor's necrotic core. Vascular hetero-geneity often makes the location of the tumor's core highly variablewhich explains why there is large variation in the data and limitedsignificance in the tumor's center.
A better representation of mass penetration is found is Fig. 3B. Inthis histogram, it is easy to observe that PPTT caused more of thetumor to have higher concentrations of copolymer than the controltumor. For example, let us consider the total percent area of thetumor (area under the curve) less than and greater than R1=1.34 s−1 for both tumors. In the control (untreated) tumors, 83.3%of its area had R1 values less than 1.34 s−1. In PPTT treated tumors,46.5% of its area was less than 1.34 s−1 and 53.5% of its area had R1values over this threshold. This data further suggests that higher con-centrations of copolymers were able to more readily be transportedthrough the tumor's interstitium when treated with PPTT.
The fact that the majority of delivery occurred within the first hourof laser radiation is not surprising (Fig. 4). In the previous studywhere HPMA copolymer delivery was quantified, roughly 60-70% ofthe copolymers were delivered within the first 15 min after laser radi-ation [40]. This is likely due to two main reasons. First, clearance ofthe HPMA copolymers from the blood via the kidneys or reticuloendo-thelial system (RES) is likely due to the conjugate's size and negativecharge which makes the window of opportunity for delivery withinthe first few hours. Second, the differential increase in pressure, ∂p/∂x,
137A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
whichdrives delivery during hyperthermia occurs over a short period oftime (10–15 min). For this reason, future work aims to investigate lon-ger periods of laser radiation to further improve delivery.
To corroborate the MRI results, FITC labeled HPMA copolymers wereapplied using the same experimental design to image distribution byfluorescence imaging. Because the location of copolymer delivery rela-tive to vascularization is important, animals were administered rhoda-mine labeled lectin prior to euthanasia to stain the tumor's vasculature.These results provide similar information in that PPTT increased deliveryto both the tumor's periphery and center. This observationwas howeverless pronounced than the MRI results which may be due to the relativesmall z-direction cross-section; 0.012 mm vs. 1.0 mm for fluorescenceand MR imaging respectively. It was anticipated that the location ofblood vesselswould directly correlatewith the location of copolymer de-livery.While thiswas indeed the case in the tumor's outer rimwhere theblood vessels were of the greatest diameter (Fig. 6C–D), the inner capil-laries did not permit much delivery. This is likely because of their smalland constricted nature due to high cellular density and interstitial pres-sure in these regions.
Histological evaluation of the tumors provides some insight intothe impact of mild PPTT on tumor viability. In both tumors, controland PPTT treated, the tumor's periphery appeared undamaged witha high degree of vascularization (Fig. 6A-D). No signs of vascular dam-age was observed which is a concern when tumors are heated nearthe 43 °C vascular damage threshold [65]. There was a differencethough in the tumor's center. In each of the PPTT treated tumors,areas of tissue damage were observed (Fig. 6E-F). This is likely to bebecause of the inability of the tumor's center to dissipate heat effec-tively during hyperthermic treatment. Residual increases in tempera-ture in these regions then cause direct cell death and tissue damage.What was interesting, however, was the presence of viable blood ves-sels found in these regions (Fig. 6F). These capillary blood vesselsappeared to be intact, though further analysis of blood vessel viabilityin these regions is required to confirm this observation.
Effective transport of nanocarriers through the tumor's interstitiumremains a major challenge in nanomedicine. Large nanoparticles suchas liposomes, micelles and inorganic nanoparticles are severely limitedby this biological barrier, though strategies such as the one described inthis studymay help overcome this. An evolving and interesting conceptto overcome the problem involves the design of multistage drug deliv-ery platforms [66]. In these nanoparticle systems, large nanoparticlescarry drugs to the site of the tumor followed by release of free drug inthe interstitium due to external stimuli such as light and heat. In thisway, such systems take full advantage of EPR and the high diffusivityof free drugs through the tumor's interstitium.
5. Conclusions
This study shows byMR and fluorescence imaging that PPTT is capa-ble of improving the tumor distribution of HPMA copolymers. Duringlaser radiation of the tumor, heating may facilitate both convectiveand diffusive interstitial transport of these conjugates. PPTTwas capableof not only providing greater amounts of copolymer delivery, but alsomore pervasive distribution throughout thewhole tumormass. This ob-servation is important for more effective drug delivery to cancerouscells. Necrotic and unavailable regions of the tumor were still presenteven after treatment with PPTT, and future studies improving deliveryto these regions remains a significant challenge.
Acknowledgements
The authors thank Osama Abdullah for his help with the MRI imag-ing, Jacob Hinkle for his help with the image analysis, as well as SherylTripp and Dr. Mohamed Salama at ARUP Laboratories for their helpwith histology preparation and interpretation. This research was
supported by a Department of Defense Prostate Cancer PredoctoralTraining Award (PC094496) as well as the National Institutes of Health(EB-R01EB7171) and the Utah Science, Technology, and Research(USTAR) Initiative.
References
[1] R. Duncan, Polymer conjugates as anticancer nanomedicines, Nat. Rev. Cancer 6(2006) 688–701.
[2] R. Duncan, R. Gaspar, Nanomedicine (s) under the microscope, Mol. Pharm.8 (2011) 2101–2141.
[3] H. Maeda, J. Wu, T. Sawa, Y. Matsumura, K. Hori, Tumor vascular permeability andthe EPR effect in macromolecular therapeutics: a review, J. Control. Release 65(2000) 271–284.
[4] H. Hashizume, P. Baluk, S. Morikawa, J.W. McLean, G. Thurston, S. Roberge, R.K.Jain, D.M. McDonald, Openings between defective endothelial cells explaintumor vessel leakiness, Am. J. Pathol. 156 (2000) 1363–1380.
[5] D. Peer, J.M. Karp, S. Hong, O.C. Farokhzad, R. Margalit, R. Langer, Nanocarriers asan emerging platform for cancer therapy, Nat. Nanotechnol. 2 (2007) 751–760.
[6] M. Suzuki, K. Hori, I. Abe, S. Saito, H. Sato, A new approach to cancer chemotherapy:selective enhancement of tumor blood flow with angiotensin II, J. Natl. Cancer Inst.67 (1981) 663–669.
[7] C. Li, Y. Miyamoto, Y. Kojima, H. Maeda, Augmentation of tumour delivery ofmacromolecular drugs with reduced bone marrow delivery by elevating bloodpressure, Br. J. Cancer 67 (1993) 975–980.
[8] A. Noguchi, T. Takahashi, T. Yamaguchi, K. Kitamura, H. Tsurumi, K. Takashina,H. Maeda, Enhanced tumor localization of monoclonal antibody by treatmentwith kininase II inhibitor and angiotensin II, Jpn. J. Cancer Res. 83 (1992)240–243.
[9] T. Seki, J. Fang, H. Maeda, Enhanced delivery of macromolecular antitumor drugsto tumors by nitroglycerin application, Cancer Sci. 100 (2009) 2426–2430.
[10] J. Fang, H. Qin, H. Nakamura, K. Tsukigawa, T. Shin, H. Maeda, Carbon monoxide,generated by heme oxygenase-1, mediates the enhanced permeability and reten-tion effect in solid tumors, Cancer Sci. 103 (2012) 535–541.
[11] K.P. Aicher, J.W. Dupon, D.L. White, S.L. Aukerman, M.E. Moseley, R. Juster, W.Rosenau, J.L. Winkelhake, R.C. Brasch, Contrast-enhanced magnetic resonance im-aging of tumor-bearing mice treated with human recombinant tumor necrosisfactor α, Cancer Res. 50 (1990) 7376–7381.
[12] L. Zhao, L.M. Ching, P. Kestell, L.R. Kelland, B.C. Baguley,Mechanisms of tumor vascu-lar shutdown induced by 5, 6-dimethylxanthenone-4-acetic acid (DMXAA): In-creased tumor vascular permeability, Int. J. Cancer 116 (2005) 322–326.
[13] R.K. Jain, Normalization of tumor vasculature: an emerging concept in antiangiogenictherapy, Science 307 (2005) 58–62.
[14] V.P. Chauhan, T. Stylianopoulos, J.D. Martin, Z. Popović, O. Chen, W.S. Kamoun,M.G. Bawendi, D. Fukumura, R.K. Jain, Normalization of tumour blood vessels im-proves the delivery of nanomedicines in a size-dependent manner, Nat.Nanotechnol. 7 (2012) 383–388.
[15] M. Magzoub, S. Jin, A. Verkman, Enhanced macromolecule diffusion deep in tu-mors after enzymatic digestion of extracellular matrix collagen and its associatedproteoglycan decorin, FASEB J. 22 (2008) 276–284.
[16] V.L. Vega, W. Charles, A. De Maio, A new feature of the stress response: increase inendocytosis mediated by Hsp70, Cell Stress Chaperones 15 (2009) 517–527.
[17] V.L. Vega, A. De Maio, Increase in phagocytosis after geldanamycin treatment orheat shock: role of heat shock proteins, J. Immunol. 175 (2005) 5280–5287.
[18] C.W. Song, Effect of local hyperthermia on blood flow and microenvironment: areview, Cancer Res. 44 (1984) 4721s–4730s.
[19] B. Hildebrandt, P. Wust, O. Ahlers, A. Dieing, G. Sreenivasa, T. Kerner, R. Felix, H.Riess, The cellular and molecular basis of hyperthermia, Crit. Rev. Oncol. Hematol.43 (2002) 33–56.
[20] L. Fajardo, A. Schreiber, N. Kelly, G. Hahn, Thermal sensitivity of endothelial cells,Radiat. Res. 103 (1985) 276–285.
[21] B. Chen, M. Zhou, L. Xu, Study of vascular endothelial cell morphology duringhyperthermia, J. Therm. Biol. 30 (2005) 111–117.
[22] A.T. Lefor, S. Makohon, N.B. Ackerman, The effects of hyperthermia on vascularpermeability in experimental liver metastasis, J. Surg. Oncol. 28 (1985) 297–300.
[23] K. Fujiwara, T. Watanabe, Effects of hyperthermia, radiotherapy and thermora-diotherapy on tumor microvascular permeability, Pathol. Int. 40 (2008) 79–84.
[24] G. Kong, M.W. Dewhirst, Hyperthermia and liposomes, Int. J. Hyperthermia 15(1999) 345–370.
[25] S.K. Huang, P.R. Stauffer, K. Hong, J.W.H. Guo, T.L. Phillips, A. Huang, D.Papahadjopoulos, Liposomes and hyperthermia in mice: increased tumor uptakeand therapeutic efficacy of doxorubicin in sterically stabilized liposomes, CancerRes. 54 (1994) 2186–2191.
[26] J. Weinstein, R. Magin, M. Yatvin, D. Zaharko, Liposomes and local hyperthermia:selective delivery of methotrexate to heated tumors, Science 204 (1979)188–191.
[27] M.H. Gaber, N.Z. Wu, K. Hong, S.K. Huang, M.W. Dewhirst, D. Papahadjopoulos,Thermosensitive liposomes: extravasation and release of contents in tumor mi-crovascular networks, Int. J. Radiat. Oncol. Biol. Phys. 36 (1996) 1177–1187.
[28] M.L. Matteucci, G. Anyarambhatla, G. Rosner, C. Azuma, P.E. Fisher, M.W.Dewhirst, D. Needham, D.E. Thrall, Hyperthermia increases accumulation oftechnetium-99m-labeled liposomes in feline sarcomas, Clin. Cancer Res. 6(2000) 3748–3755.
138 A.J. Gormley et al. / Journal of Controlled Release 166 (2013) 130–138
[29] G. Kong, R.D. Braun, M.W. Dewhirst, Characterization of the effect of hyper-thermia on nanoparticle extravasation from tumor vasculature, Cancer Res. 61(2001) 3027–3032.
[30] G. Kong, R.D. Braun, M.W. Dewhirst, Hyperthermia enables tumor-specific nano-particle delivery: effect of particle size, Cancer Res. 60 (2000) 4440–4445.
[31] G. Ma, G. Jiang, Review of tumor hyperthermia technique in biomedical engineer-ing frontier, in: 2010 3rd International Conference on Biomedical Engineeringand Informatics (BMEI), IEEE, 2010, pp. 1357–1359.
[32] S. Link, M.A. El-Sayed, Shape and size dependence of radiative, non-radiative andphotothermal properties of gold nanocrystals, Int. Rev. Phys. Chem. 19 (2000)409–453.
[33] L.R. Hirsch, R.J. Stafford, J.A. Bankson, S.R. Sershen, B. Rivera, R.E. Price, J.D. Hazle,N.J. Halas, J.L. West, Nanoshell-mediated near-infrared thermal therapy of tumorsunder magnetic resonance guidance, Proc. Natl. Acad. Sci. U. S. A. 100 (2003)13549–13554.
[34] D.P. O'Neal, L.R. Hirsch, N.J. Halas, J.D. Payne, J.L. West, Photo-thermal tumor ab-lation in mice using near infrared-absorbing nanoparticles, Cancer Lett. 209(2004) 171–176.
[35] J.M. Stern, J. Stanfield, W. Kabbani, J.T. Hsieh, J.A. Cadeddu, Selective prostatecancer thermal ablation with laser activated gold nanoshells, J. Urol. 179 (2008)748–753.
[36] X. Huang, P.K. Jain, I.H. El-Sayed, M.A. El-Sayed, Gold nanoparticles: interestingoptical properties and recent applications in cancer diagnostics and therapy,Nanomedicine 2 (2007) 681–693.
[37] P.K. Jain, I.H. El-Sayed, M.A. El-Sayed, Au nanoparticles target cancer, Nano Today2 (2007) 18–29.
[39] A.J. Gormley, K. Greish, A. Ray, R. Robinson, J.A. Gustafson, H. Ghandehari, Goldnanorod mediated plasmonic photothermal therapy: a tool to enhance macromo-lecular delivery, Int. J. Pharm. 415 (2011) 315–318.
[40] A.J. Gormley, N. Larson, S. Sadekar, R. Robinson, A. Ray, H. Ghandehari, Guided de-livery of polymer therapeutics using plasmonic photothermal therapy, NanoToday 7 (2012) 158–167.
[41] J.H. Park, G. von Maltzahn, M.J. Xu, V. Fogal, V.R. Kotamraju, E. Ruoslahti, S.N.Bhatia, M.J. Sailor, Cooperative nanomaterial system to sensitize, target, andtreat tumors, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 981–986.
[42] J.H. Park, G.v. Maltzahn, L.L. Ong, A. Centrone, T.A. Hatton, E. Ruoslahti, S.N. Bhatia,M.J. Sailor, Cooperative nanoparticles for tumor detection and photothermallytriggered drug delivery, Adv. Mater. 22 (2010) 880–885.
[43] G. Von Maltzahn, J.H. Park, K.Y. Lin, N. Singh, C. Schwöppe, R. Mesters, W.E.Berdel, E. Ruoslahti, M.J. Sailor, S.N. Bhatia, Nanoparticles that communicate invivo to amplify tumour targeting, Nat. Mater. 10 (2011) 545–552.
[44] P. Diagaradjane, A. Shetty, J.C. Wang, A.M. Elliott, J. Schwartz, S. Shentu, H.C. Park,A. Deorukhkar, R.J. Stafford, S.H. Cho, Modulation of in vivo tumor radiationresponse via gold nanoshell-mediated vascular-focused hyperthermia: character-izing an integrated antihypoxic and localized vascular disrupting targeting strat-egy, Nano Lett. 8 (2008) 1492–1500.
[45] R.K. Jain, T. Stylianopoulos, Delivering nanomedicine to solid tumors, Nat. Rev.Clin. Oncol. 7 (2010) 653–664.
[46] F. Yuan, M. Leunig, S.K. Huang, D.A. Berk, D. Papahadjopoulos, R.K. Jain,Mirovascular permeability and interstitial penetration of sterically stabilized(stealth) liposomes in a human tumor xenograft, Cancer Res. 54 (1994)3352–3356.
[47] J. Strohalm, J. Kopecek, Poly N-(2-hydroxypropyl) methacrylamide. 4. Heteroge-neous polymerization, Angew. Makromol. Chem. 70 (1978) 109–118.
[48] B. Zarabi, M.P. Borgman, J. Zhuo, R. Gullapalli, H. Ghandehari, Noninvasive moni-toring of HPMA copolymer–RGDfK conjugates by magnetic resonance imaging,Pharm. Res. 26 (2009) 1121–1129.
[49] V. Omelyanenko, P. Kopecková, C. Gentry, J. Kopecek, Targetable HPMA copoly-mer–adriamycin conjugates. Recognition, internalization, and subcellular fate,J. Control. Release 53 (1998) 25–37.
[50] V. Šubr, K. Ulbrich, Synthesis and properties of new N-(2-hydroxypropyl)methacrylamide copolymers containing thiazolidine-2-thione reactive groups,React. Funct. Polym. 66 (2006) 1525–1538.
[51] A.J. Gormley, A. Malugin, A. Ray, R. Robinson, H. Ghandehari, Biological evaluationof RGDfK-gold nanorod conjugates for prostate cancer treatment, J. Drug Target.19 (2011) 915–924.
[52] R.K. Wang, V.V. Tuchin, Enhance light penetration in tissue for high resolution op-tical imaging techniques by the use of biocompatible chemical agents, J. Xray Sci.Technol. 10 (2002) 167–176.
[53] K.F. Schmidt, M. Ziu, N. Ole Schmidt, P. Vaghasia, T.G. Cargioli, S. Doshi, M.S.Albert, P.M.L. Black, R.S. Carroll, Y. Sun, Volume reconstruction techniques im-prove the correlation between histological and in vivo tumor volume measure-ments in mouse models of human gliomas, J. Neurooncol 68 (2004) 207–215.
[54] M. Konerdingi, E. Fait, A. Gaumann, C. Dimitropouloui, W. Malkusch, Scanningelectron microscopy of corrosion casts in the study of tumor, in: M.E.Maragoudakis (Ed.), Angiogenesis: Models, Modulators, and Clinical Applications,Plenum Press, New York, USA, 1998, pp. 429–447.
[55] J. Folkman, Angiogenesis in cancer, vascular, rheumatoid and other disease, Nat.Med. 1 (1995) 27–30.
[56] J.W. Baish, R.K. Jain, Fractals and cancer, Cancer Res. 60 (2000) 3683–3688.[57] R.K. Jain, Transport of molecules in the tumor interstitium: a review, Cancer Res.
47 (1987) 3039–3051.[58] S. Ramanujan, A. Pluen, T.D. McKee, E.B. Brown, Y. Boucher, R.K. Jain, Diffusion
and convection in collagen gels: implications for transport in the tumorinterstitium, Biophys. J. 83 (2002) 1650–1660.
[59] L.T. Baxter, R.K. Jain, Transport of fluid and macromolecules in tumors. I. Role ofinterstitial pressure and convection, Microvasc. Res. 37 (1989) 77–104.
[60] E.A. Swabb, J. Wei, P.M. Gullino, Diffusion and convection in normal and neoplas-tic tissues, Cancer Res. 34 (1974) 2814–2822.
[61] M. Leunig, A.E. Goetz, M. Dellian, G. Zetterer, F. Gamarra, R.K. Jain, K. Messmer,Interstitial fluid pressure in solid tumors following hyperthermia: possible corre-lation with therapeutic response, Cancer Res. 52 (1992) 487–490.
[62] W. Deen, M. Bohrer, N. Epstein, Effects of molecular size and configuration on dif-fusion in microporous membranes, AIChE J. 27 (1981) 952–959.
[63] S. Sadekar, A. Ray, M. Janat-Amsbury, C. Peterson, H. Ghandehari, Comparativebiodistribution of PAMAM dendrimers and HPMA copolymers in ovarian-tumor-bearing mice, Biomacromolecules 12 (2011) 88–96.
[64] T. Lammers, P. Peschke, R. Kühnlein, V. Subr, K. Ulbrich, J. Debus, P. Huber, W.Hennink, G. Storm, Effect of radiotherapy and hyperthermia on the tumor accu-mulation of HPMA copolymer-based drug delivery systems, J. Control. Release117 (2007) 333–341.
[65] C.W. Song, M.S. Kang, J.G. Rhee, S.H. Levitt, Effect of hyperthermia on vascularfunction in normal and neoplastic tissues, Ann. N. Y. Acad. Sci. 335 (1980) 35–47.
[66] C. Wong, T. Stylianopoulos, J. Cui, J. Martin, V.P. Chauhan, W. Jiang, Z. Popović, R.K.Jain, M.G. Bawendi, D. Fukumura, Multistage nanoparticle delivery system fordeep penetration into tumor tissue, Proc. Natl. Acad. Sci. U. S. A. 108 (2011)2426–2431.
