Mechanisms in photodynamic therapy: Part three—Photosensitizer pharmacokinetics, biodistribution,...

Photodiagnosis and Photodynamic Therapy (2005) 2, 91—106

REVIEW

Mechanisms in photodynamic therapy: Partthree—–Photosensitizer pharmacokinetics,biodistribution, tumor localization andmodes of tumor destruction

Ana P. Castanoa,b, Tatiana N. Demidovaa,c, Michael R. Hamblin PhDa,b,d,∗

a BAR414, Wellman Center for Photomedicine, Massachusetts General Hospital, 40 Blossom Street,Boston, MA 02114, USAb
Department of Dermatology, Harvard Medical School, USAc Cell, Molecular and Developmental Biology Program, Tufts University, USAd Harvard-MIT Division of Health Sciences and Technology, USA
KEYWORDSPhotodynamic therapy;Cancer;Pharmacokinetics;Biodistribution;Cellular destruction;Vascular shutdown;Anti-tumor immunity

Summary Photodynamic therapy (PDT) has been known for over a hundred years,but is only now becoming widely used. Originally developed as cancer therapy, someof its most successful applications are for non-malignant disease. The majority ofmechanistic research into PDT, however, is still directed towards anti-cancer applica-tions. In the final part of series of three reviews, we will cover the possible reasonsfor the well-known tumor localizing properties of photosensitizers (PS). When PSare injected into the bloodstream they bind to various serum proteins and this canaffect their phamacokinetics and biodistribution. Different PS can have very dif-ferent pharmacokinetics and this can directly affect the illumination parameters.Intravenously injected PS undergo a transition from being bound to serum proteins,then bound to endothelial cells, then bound to the adventitia of the vessels, thenbound either to the extracellular matrix or to the cells within the tumor, and finallyto being cleared from the tumor by lymphatics or blood vessels, and excreted eitherby the kidneys or the liver. The effect of PDT on the tumor largely depends at whichstage of this continuous process light is delivered.The anti-tumor effects of PDT are divided into three main mechanisms. Powerful

anti-vascular effects can lead to thrombosis and hemorrhage in tumor blood vesselsthat subsequently lead to tumor death via deprivation of oxygen and nutrients. Directtumor cell death by apoptosis or necrosis can occur if the PS has been allowed tobe taken up by tumor cells. Finally the acute inflammation and release of cytokinesand stress response proteins induced in the tumor by PDT can lead to an influx ofleukocytes that can both contribute to tumor destruction as well as to stimulate theimmune system to recognize and destroy tumor cells even at distant locations.© 2005 Elsevier B.V. All rights reserved.

∗ Corresponding author. Tel.: +1 617 726 6182; fax: +1 617 726 8566.E-mail address: [email protected] (M.R. Hamblin).

1572-1000/$ — see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/S1572-1000(05)00060-8

92 A.P. Castano et al.

Contents

Introduction ...................................................................................................... 92PS pharmacokinetics and biodistribution.......................................................................... 92PS biodistribution................................................................................................. 94PS structure, tumour localization and delivery vehicles........................................................... 94

PS structure.................................................................................................. 94Tumor localizing PS .......................................................................................... 95Non-covalent delivery vehicle complexes .................................................................... 96Enhanced PS accumulation in tumors as a means to deliver other molecules ................................. 96

Mechanisms of tumour destruction................................................................................ 97Direct tumor-cell death effects .............................................................................. 97Vascular effects .............................................................................................. 99Immune system effects ..................................................................................... 100

Conclusion....................................................................................................... 102Acknowledgements ............................................................................................ 102References .................................................................................................... 102

Introduction

The history of the development of photodynamictherapy (PDT) for cancer can be broadly dividedinto two areas. One area concerns studies on theability of certain porphyrins and related tetrapyr-

development of immunocompromisedmice and rats(nude, SCID and beige) has allowed many human

or alternatively can be implanted orthotopically inthe organ of origin. Thirdly, chemical and radiationinduced carcinogenesis has allowed tumors to beinduced in many species of animal with more or lessreliability, and can be thought to provide a morenatural tumor model than transplanted tumors.

tain series of events commences that can take verydifferent lengths of time to reach completion fordvtnPcnhc[wt

tumor cell lines to be easily grown as tumors invivo in laboratory animals. Secondly there is nowa large range of inbred mice and rat strains whereeach individual member of a strain has identicalmajor histocompatibility complex haplotypes. Thismeans that tumors that arise in any individual mem-ber of that strain and from which a stable cell linecan be isolated, can be freely transplanted amongstmembers of that strain in the presence of a fullyfunctioning immune system. These two laboratorymodels (human tumor xenografts and syngeneicrodent tumors) can be implanted subcutaneously,

ifferent PS. Firstly depending on the delivery sol-ent or vehicle that is used for the PS injection,he PS must come to equilibrium with the compo-ents of the circulating blood. This can involve theS disaggregating from itself or its delivery vehi-le and binding instead to various protein compo-ents of serum (see later). In addition the bloodas many circulating cells (erythrocytes and leuko-ytes) that would be available to bind injected PS3]. Secondly the circulating PS must bind to thealls of the blood vessels, and it is thought thathe nature of the various sizes and characteristics

role compounds to localize in cancers after injec-tion into the bloodstream. The second area is con-cerned with the tumoricidal effects when cancerswere illuminated with visible light after the patienthad received prior treatment with various photo-sensitizers (PS). This is the third part of a series ofthree reviews. In Part 1 [1] we covered the chemicalstructure, photochemistry, photophysics and sub-cellular localization of PS. In Part 2 [2] we discussedthe changes in cellular metabolism and intracellu-lar signalling and modes of cell death when cellstreated with PS are illuminated in culture. We nowcome to consider the mechanisms that operatewhen PDT is carried out in vivo.The vast explosion of biomedical research in

recent decades has led to the ready availabilityof multiple animal models of cancer. Firstly the

Fourthly the discovery of oncogenes and tumor sup-pressor genes has allowed the creation of trans-genic mice that either over express an oncogeneor lack a tumor suppressor gene, and more or lessreliably develop cancer in a manner that is proba-bly most similar to the human situation. The choiceof the most appropriate animal model of cancer totest PDT is complex and multifactorial and is oftennot rationally justified. However, the fact that prac-tical light delivery to the tumor is an integral partof PDT means that the majority of reported studieshave used subcutaneous (s.c.) tumors.

PS pharmacokinetics and biodistribution

When PS are injected into the bloodstream a cer-

Mechanisms in photodynamic therapy 93

of blood vessels in the tumor and normal tissues,and different physiological types of vessels in var-ious organs governs to a great extent where thePS localizes. Thirdly the PS will penetrate throughthe wall of the blood vessel at a rate that probablydepends on how strong the initial binding of the PSto the intimal surface was. PS that bind stronglyto the blood vessel wall will take a longer timeto cross the entire wall of the blood vessel, andthose that have an initial weaker binding will passthrough quicker. Fourthly after extravasating, thePS will diffuse throughout the parenchyma of theorgan or tumor to which it has been delivered. Ifthe organ happens to be the liver or other metaboli-cally active organ, the PS may be subject to changesby metabolic enzymes, but this is thought to beunlikely for most tetrapyrrole PS in common clin-ical use. Fifthly the PS will be eliminated from thetissue. Although this has not been much studied,if the tissue containing the PS is a tumor or othernon-metabolic organ, elimination will probably bylymphatic drainage and back into the circulation viathe thoracic duct. Sixthly the PS will be excretedfrom the body. For the majority of PS in clinicaluse, excretion is from the liver into the bile andte

stcfwayoasAtbPdomccamimapotb

fluorescent compound to be eliminated [6]. Insome cases the PS is quantified by a traditionalanalytical chemistry technique such as high perfor-mance liquid chromatography [7]. This method hasthe advantage of detecting possible metabolismproducts that have been formed from the originalpure PS after injection. Finally in some cases ofexperimental animal models, the PS has beenpresynthesized to include a radioisotope label (forinstance carbon-14 or tritium) or a chelator for ametal radioisotope and biological samples are thenquantified using a scintillation counter [8—11].An alternative approach to carrying analysis ofsamples of biological material removed sequen-tially, is to use a continuous monitoring methodto determine PS concentration in tissue based onfluorescence. Here a non-invasive or minimallyinvasive approach with fiber-based fluorescencedetection systems is used to measure the variationover time of a fluorescence signal linked to the PSof interest [12]. In some small animal models theskin can be removed from a subcutaneous tumorso the fiber that collects fluorescence can be incontact with the underlying tumor as well as theadjacent skin [13].

ncpuonwiPstPpoip

dcyRim0cscadT

hence to the intestine where it is lost via fecallimination.Measurement of pharmacokinetics requiresequential measurements of PS concentration inissue or other biological substance in order toonstruct temporal profiles that can be comparedor different PS. This is frequently done for bloodhere samples can be readily removed in smallmounts at frequent intervals and analysis thenields values of serum half-lives (first or secondrder) in minutes, hours or days. Alternatively,nalysis of PS concentrations in urine or fecalamples can give terminal elimination half-lives.ll these methods require a method of quantita-ively analyzing PS concentrations in samples ofiological material. The fact that the majority ofS used for PDT are also fluorescent, has led to theevelopment of assays involving homogenizationr dissolving the tissue or biological fluid andeasurement of fluorescence in solution afterentrifugation or extraction steps [4]. Calibrationurves are prepared with mixtures of knownmounts of PS and weighed amounts of biologicalaterial. Care must be taken when these exper-ments are carried out on small rodents such asice and rats because these animals accumulatered-fluorescent compound derived from chloro-hyll in their skin and digestive tract amongstther organs [5]. This can be avoided by feedinghe animals a chlorophyll-free diet for enough timeefore the experiment to allow the interfering

There has been a wide variation in pharmacoki-etics reported for various PS in clinical and pre-linical use. Bellnier and Dougherty [14] studiedharmacokinetics of Photofrin in patients sched-led to undergo PDT for the treatment of carcinomaf the lung or the skin. They found a triexpo-ential three-compartment pharmacokinetic modelith alpha, beta, and gamma half-lives of approx-mately 16 h, 7.5 days, and 155.5 days. Detectablehotofrin fluorescence was shown to persist in theerum for longer than one year. In clinical prac-ice light is usually delivered to patients 48 h afterhotofrin injection. In a Japanese study [15] skinhotosensitivity was found to persist for a monthr more after Photofrin injection and interest-ngly was significantly more pronounced for femaleatients.The pharmacokinetics of 2-[1-hexyloxyethyl]-2-evinyl pyropheophorbide-a (HPPH) was studied inancer patients [16]. A two-compartment modelielded alpha and beta half-lives of 7.77 and 596 h.adiolabeled Foscan pharmacokinetics was stud-ed in tumor-bearing rats yielding a tri-exponentialodel with alpha, beta and gamma half-lives of.46, 6.91 and 82.5 h, respectively [9]. Pharma-okinetics of the silicon phthalocyanine Pc4 weretudied in non-tumor-bearing mice giving a two-ompartment fit with alpha and beta half-lives ofpproximately 10min and 20 h with some variationepending on injected dose and solvent [17,18].he palladium bacteriopheophorbide PS known as


TOOKAD has very rapid pharmacokinetics with alphaand beta half-lives of approximately 2min and 1.3 hand in this case graphite furnace atomic absorptionspectroscopy was used to quantify the palladiumatom coordinated to the tetrapyrrole [17].

PS biodistribution

In experimental animals it has sometimes been pos-sible to establish the overall pattern of organ dis-tribution after intravenous injection of PS. A study[10] using radioisotope-labeled Photofrin in micewith subcutaneous fibrosarcoma tumors sacrificed24 h later found highest concentrations in liver,adrenal gland, urinary bladder, greater than pan-creas, kidney, spleen, greater than stomach, bone,lung, heart, greater than muscle, much greaterthan brain. Only skeletal muscle, brain, and skinlocated contralaterally to tumors had peak con-centrations lower than tumor tissue; skin overlyingtumors showed concentrations not significantly dif-ferent from tumor. The higher molecular weightcomponents of Photofrin were partially retained inliver and spleen at 75 days after injection.

The amount of PS that accumulates in spleenseems to vary widely with the structure of thePS (charge and hydrophobicity). Woodburn et al.[22] studied a range of porphyrins with vary-ing octanol/water partition coefficients and foundthe accumulation in the spleen varied the mostcompared to other organs. Egorin and co-workers[18,23] compared organ distribution of Pc4 admin-istered by i.v. injection of the same dose (10mg/kg)dissolved in several solvents and found that theaccumulation in the spleen was the most variable.The kidney and bladder frequently accumulate

high amounts of PS despite the clearance routebeing almost exclusively via liver and bile [24]The digestive organs (stomach, large and smallintestines) seem to take up middling amounts ofPS (i.e. more than liver but less than muscle) [11].Despite skin photosensitivity due to PS accumu-lating in patients’ skin being the main cause ofPDT-related side-effects, skin has not been foundto accumulate high amounts of PS in experimen-tal animals. The lowest concentrations of PS arefound in organs such as heart, skeletal muscle,bone, eye and brain. These are organs that areknown to have relatively impermeable blood sup-pfttfr

Pd

S5tbstucmTtai

P

Afc

In most cases where organ biodistribution datahave been reported for small animals such as miceand rats, the highest concentration of PS has beenfound in the liver [19,20]. Chan et al. studied a setof sulfonated phthalocyanines found that the accu-mulation in liver was inversely proportional to thedegree of sulfonation and hence to the lipophilic-ity of the PS molecules [21]. The liver is known tohave a highly permeable blood supply with fenes-trated endothelium that contains pores that allowmolecules to pass easily out of the vessels. This isneeded by the role of the liver in detoxifying thebody of extraneous molecules (particularly organicmolecules such as PS). These molecules are thenexcreted into the bile via the gall bladder in a sim-ilar fashion to the production of bile acids fromcholesterol. The bile is routed into the duodenumwhere it is possible for unchanged PS to be absorbedinto the circulation a second time from the smallintestine (a process known as entero-hepatic recy-cling) as well as being excreted via the feces. Itis known that soon after injection, large amountsof PS can accumulate in the lungs. This has beenreported [18] as the mode of dose-limiting tox-icity (in the dark) when PS-injected doses wereraised leading to lung hemorrhage and acute inter-stitial pneumonia after injection of 80mg/kg of Pc4into mice. The explanation is probably that largeinjected doses of PS can aggregate in the blood-stream and the particles so formed can easily col-lect in the fine capillary network of the lungs.

ly. The blood—brain barrier serves to exclude PSrom the parenchyma of the brain, and the fact thathe blood—brain barrier is frequently breached byhe growth of brain tumors is probably responsibleor the very large (over 100) tumor-to-normal brainatios reported for some PS [25].

S structure, tumour localization andelivery vehicles

ince the first observations [26] made more than0 years ago that porphyrins and their deriva-ives can localize in tumors when injected into theloodstream, the phenomenon has been intensivelytudied and although progress has been made,he mechanisms involved are still not completelynderstood [27]. A review by Boyle and Dolphin [25]ollected reported tumor—normal tissue ratios forany experimental studies of PS in animal models.he complexity of comparing different PS struc-ure, different tumor models, times after injectionnd doses meant that there were large variationsn these ratios even with the same PS.

S structure

s mentioned above different PS have very dif-erent pharmacokinetics and biodistribution. Theseonsiderations mean that the interval between the


PS administration and light exposure is another keyfactor in determining PDT efficacy. With the pos-sible exceptions of uroporphyrin and some of thelarger aggregates present in Photofrin, all of thetetrapyrrole PS, which have been suggested for useas PDT drugs are more or less firmly bound to serumproteins after i.v. injection. Three classes of thesecompounds, which have tumor localizing proper-ties can be delineated (although there is overlapbetween them).

(a) relatively hydrophilic compounds, which areprimarily bound to albumin (and possibly globu-lins) such as the tri and tetra-sulfonated deriva-tives of tetraphenylporphine (TPPS3, TPPS4)and chloroaluminum phthalocyanine (ClAlPCS3,ClAlPCS4);

(b) amphiphilic, asymmetric compounds, whichare thought to insert into the outer phos-pholipid and apoprotein layer of lipoproteinparticles, such as the adjacent disulfonates(TPPS2a, ClAlPCS2a), benzoporphyrin derivativemonoacid (BPD), lutetium texaphyrin (LuTex)and monoaspartyl chlorin(e6) (MACE) whichpartition between albumin and high-densitylipoprotein (HDL);

(

T

ShhosTrttptmcitpdt

time point after administration such as 3 h, whilewith another this can be at 7 days after injection.One of the particular properties of these tetrapyr-role compounds relevant to their tumor localizingability, is their tendency to bind strongly to serumproteins and to each other. This means that mostPS when injected into the bloodstream behave asmacromolecules either because they are more orless firmly bound to large protein molecules orbecause they have formed intermolecular aggre-gates of similar size. Many workers have reported[31—33] on the distribution of PS between the var-ious classes of serum proteins when mixed withserum in vitro. These proteins are usually dividedinto four classes: albumin and other heavy pro-teins, HDL, LDL, and VLDL. However, even this studyhas been complicated by the fact that the mostlipophilic PS are insoluble in aqueous media andneed to be delivered in a solvent mixture whichmay alter the serum protein distribution of the PS.It has been argued that PS that preferentially bindto LDL are better tumor localizers (see below) [30],but this is by no means always the case [34].It is useful to make a distinction between selec-

tive accumulation and selective retention. ThetpmPltitsTwttosoip(oibodOfOpev[

c) hydrophobic compounds, which require a solu-bilization vehicle such as liposomes, cremaphorEL or Tween 80. These are thought to localizein the inner lipid core of lipoproteins particu-larly low-density lipoprotein (LDL) (but also HDLand very low-density lipoprotein, VLDL). Exam-ples of these compounds are unsubstitutedphthalocyanines (ZnPC, ClAlPC) naphthalocya-nines (isoBOSINC), tin-etiopurpurin (SnET2).

umor localizing PS

ince the observation of tumor localizing ability ofematoporphyrin derivative (HPD), many workersave investigated the mechanism of this tendencyf PS to preferentially localize in tumors and otherpecific organs and anatomical sites [25,27—30].he precise definition of tumor to normal tissueatio has also been subject to debate. Some inves-igators working with s.c. tumors in experimen-al animals use the ratio between the tumor anderitumoral muscle or skin, while others use dis-ant muscle and skin. In addition, there has beenuch effort made to determine which factors in thehemical structures of the PS are optimal for max-mizing the selectivity for the tumor over normalissue and organs. This has proved to be quite com-licated because the pharmacokinetics can varyramatically. For instance, one PS can have its bestumor to normal tissue ratio at a relatively early

umor localizing ability of the PS with the fasterharmacokinetics is probably due to selective accu-ulation in the tumor, while the localization ofS with slower acting pharmacokinetics is moreikely due to selective retention. In the selec-ive accumulation model it is thought that thencreased vascular permeability to macromoleculesypical of tumor neovasculature is chiefly respon-ible for the preferential extravasation of the PS.hese quick acting PS frequently bind to albuminhich is of ideal size and Stokes radius to passhrough the ‘‘pores’’ in the endothelium of theumor microvessels [35]. The selective retentionf PS in tumors has been the subject of muchpeculation. As mentioned above, a popular the-ry maintains that the binding of the PS to LDLs of major importance [30]. In this theory, it isroposed that cancer cells overexpress the LDLapoB/E) receptor. Upregulation of the expressionf LDL receptors is one way that rapidly grow-ng malignant cells gain cholesterol needed for theiosynthesis of lipids needed for the rapid turnoverf cellular membranes. There is experimental evi-ence both for and against this theory [34,36].ther theories have been proposed to accountor the selective retention of PS in tumor tissue.ne is that tumors have poorly developed lym-hatic drainage, and that macromolecules, whichxtravasate from the hyperpermeable tumor neo-asculature, are retained in the extravascular space37]. Another involves the macrophages, which


infiltrate solid tumors to varying extents [27]. Thesetumor-associated macrophages have been shown toaccumulate up to 13 times the amount of some PScompared to cancer cells [38]. The explanation forthis has been proposed to be either the phagocy-tosis of aggregates of PS [39], or the preferentialuptake by macrophages of lipoproteins, which havebeen altered by the binding of porphyrins [27].Another theory proposes that the low pH commonlyfound in tumors has the effect of trapping some ofthe anionic PS, which are ionized at normal physio-logical pH. These PS then become neutrally chargedand hence more lipophilic, when they encounterthe lowered pH in the tumor environment [40]. Yetanother proposal is that PS accumulation in tissuesis related to the articular tissues’ propensity toaccumulate lipid droplets [41].It has been proposed that the peripheral benzo-

diazepine receptors located on the inner mitochon-drial membrane are responsible for binding someporphyrin-based PS that accumulate in mitochon-dria [42,43]. Alternatively it has been suggestedthat mitochondrially localizing dyes such as Rho-damine 123 take advantage of differences in mito-chondrial membrane potential between normal and

tein most frequently investigated has been LDL fortwo reasons. Firstly the hydrophobic core of theLDL particle can act as a solubilizing medium forhydrophobic PS in a similar way to liposomes or Cre-mophor, and secondly it was proposed to increasetumor targeting by taking advantage of receptor-mediated endocytosis of the complex by the LDLreceptor that has been reported to be overex-pressed on cancer cells (see above). Barel et al.[55], used HP precomplexed to lipoproteins to tar-get a murine fibrosarcoma. An increased deliveryof hematoporphyrin (HP) to the mouse tumor wasreported with the HP—LDL complex compared toHP complexes of HDL, VLDL, or free HP. Similarly,precomplexing of BPD with LDL led to a greateraccumulation of the PS in tumors as compared toBPD administration of an aqueous solution at 3 h[36]. Some authors have suggested that PS deliv-ery with various macromolecular systems may leadto differing mechanisms of tumor destruction asPS are delivered to different sites. For example,although albumin and globulins are believed todeliver PS mainly to the vascular stroma of tumors[29], HDL apparently delivers PS to cells via a non-specific exchange with the plasma membrane. LDLpa[lmnaLLw[Zar

Em

Tiegaatubmg[P

cancer cells [44,45].

Non-covalent delivery vehicle complexes

As mentioned above, many of the most effective PSare too hydrophobic to dissolve in aqueous solvents,and this necessitates the use of a delivery vehicle tokeep the molecules in a sufficiently disaggregatedstate to be able to travel in the blood vessels and toextravasate into tumors. It has been found that thechoice of delivery vehicle can influence the tumorselectivity of the PS [46]. The castor oil derivativeCremophor EL has been used as a delivery vehicle,but Kessel and co-workers have shown that its useactually changes the lipoprotein profile [47] andcan affect the biodistribution of PS and the efficacyof the tumor treatment [48]. An alternative methodof PS delivery is encapsulation in liposomes. Ithas been suggested that when liposomal PS areadministered to animals, the PS is more efficientlytransferred to LDL than an aqueous formulation[49]. In vivo liposomal delivery has been shownto give advantages in either biodistribution ortumor destruction compared to non-liposomaldelivery for Photofrin [50], BPD [51] and Zn-PC[52]. Other workers have used polyethylene-glycol-coated poly(lactic acid) nanoparticles[53], or albumin microspheres [54] to deliverPS.Some workers have investigated the pre-

complexing with various serum proteins. The pro-

robably delivers a large fraction of the PS via anctive receptor-mediated pathway [56]. Zhou et al.57] have suggested that aqueous solutions of HPead to predominantly vascular damage, while LDL-ediated PDT leads predominantly to damage ofeoplastic cells. However, this is not always true. Instudy of PDT of ocular melanoma in a rabbit modelDL complexed to BPD was used. Despite the use ofDL as a carrier, early damage to the vasculatureas demonstrated by light and electron microscopy58]. Larroque et al. [28] showed that the insolublen-PC could be formulated for i.v. administrations a complex with serum albumin, after which itedistributed primarily to HDL.

nhanced PS accumulation in tumors as aeans to deliver other molecules

he enhanced accumulation of tetrapyrrole PSn tumors has been proposed to act as a deliv-ry mechanism for other molecules. The sug-ested strategy includes using PS with covalentlyttached radioisotopes (or radioisotope chelators)s a tumor-targeting method for both detection andherapy of cancer. Imaging modalities that can besed if gamma-particle emitting radionuclides cane specifically targeted to cancers (such as earlyetastases) include planar gamma cameras and sin-le photon emission computed tomography (SPECT)59]. Technetium-labeled Photosan-3 (an analog ofhotofrin) [60] was proposed to be used in tumor


imaging as demonstrated by accumulation in s.c.mouse tumors.Radioiodine-labeled methylene blue has been

used to detect melanoma [61] and 211-astatinelabeled methylene blue (an alpha particle emit-ter) has been used to treat melanoma both in nudemouse models [62] and in a Phase I clinical trial[63]. It is thought that methylene blue shows partic-ular affinity for melanoma tumors due to a specificmolecular binding to melanin.Porphyrins have been proposed as a means to

deliver boron to tumors (particularly brain tumors)[64]. If a large accumulation of boron can betargeted to tumors then after application of aneutron beam the boron atoms capture neutrons(10iB(n,�)7Li) forming both a reactive alpha parti-cle (helium nucleus) and also a high energy lithiumatom both of which can damage tumor cells [65].In principle the same compound (a boronated PS)could be used as a binary therapy drug and couldinteract with both red light and with a neutronbeam to cause cell death. The boronated por-phyrin (BOPP) in combination with red light hasbeen tested in a Phase I clinical trial for high-gradegliomas [66].

ttdbbi[

tauateaaett[

mtoanwwai

illumination [74]. Notably texaphyrins substitutedwith the metals lutetium, europium, yttrium, andcadmium did not demonstrate enhancement ofradiation sensitivity [75]. Xcytrin has advancedto Phase III clinical trials for enhancing radiationtherapy of brain metastases of lung cancer and hasdemonstrated survival benefits [76].

Mechanisms of tumour destruction

Three distinct (but possibly interrelated) mecha-nisms have been identified which contribute to theobserved shrinkage (and frequent disappearance)of tumors when treated with PDT. These are graph-ically illustrated in Fig. 1. In the first case, theROS that is generated by PDT can kill tumor cellsdirectly by apoptosis and/or necrosis as discussedin Parts 1 and 2 of this review series [1,2]. PDT alsodamages the tumor-associated vasculature, lead-ing to tissue deprivation of oxygen and nutrientsand consequent tumor infarction. Finally, PDT canactivate an immune response against tumor cells.These three mechanisms can also influence eachother. The relative importance of each for the over-ahpcs

D

EntettPblfmtvalemtato[

Another strategy is to take advantage of theumor localizing properties of porphyrin-based PSo deliver paramagnetic metal atoms that can beetected by magnetic resonance imaging. This haseen demonstrated by the use of Mn-TPPS in ratrain tumors that gave a strong signal enhancementn T1 weighted images that lasted for three days67,68].It has been proposed that porphyrins and otheretrapyrrole PS can act as radiosensitizers in thebsence of visible light [69]. While the mechanismnderlying this effect is largely unknown, the ther-py takes advantage of the tumor-localizing proper-ies of the porphyrin to concentrate the cytotoxicffect of a beam of radiation in the tumor. In annimal model it was shown that only Photofrin haspositive therapeutic benefit, out of six differ-nt tetrapyrrole PS tested [70]. In a clinical trialhe combination allowed a previously inoperableumor to be shrunk enough to be surgically removed71].Gadolinium-substituted texaphyrin (Gd-Tex,otexafin gadolinium, Xcytrin) was originally syn-hesized as a contrast enhancement agent to carryut MRI visualization of tumors [72]. Its selectiveccumulation in tumors was demonstrated inormal and tumor-bearing rats and rabbits [73]. Itas subsequently discovered that this compoundas an effective radiation sensitizer due to itsbility to carry our redox cycling and producentracellular reactive oxygen species (ROS) without

ll tumor response is yet to be defined. It is clear,owever, that the combination of all these com-onents is required for optimum long-term tumorures, especially of tumors that may have metasta-ized.

irect tumor-cell death effects

xposure of tumors to PDT in vivo can reduce theumber of clonogenic tumor cells existing withinhe tumor through direct cellular damage; how-ver, this is thought to be probably insufficient forumor cure [77]. Studies [78] in rodent tumor sys-ems employing curative procedures with severalS showed direct photodynamic tumor cell kill toe less than 2 logs and in most cases less than 1og, i.e. far short of the 8-log reduction requiredor tumor cure, if tumor cell killing were the onlyechanism operating. The in vitro illumination ofumor cells isolated from photosensitized tumors inivo predicts that total eradication is feasible withsufficiently high light dose with some PS [77]. Butimitations appear to exist that do not allow theradication to be realized for in vivo PDT treat-ent. Inhomogeneous PS distribution within theumor might be one of these limitations. Korbeliknd Krosl [79] have shown that both PS accumula-ion and tumor cell kill decrease with the distancef tumor cells from the vascular supply. Lee et al.80] found that there were significant degrees of


Figure 1 Pathways of PDT-mediated tumor destruction illustrating possible contributions from direct tumor cell killing,vascular damage and host immune response.

variability in PS concentration (both intra-tumorand inter-tumor) depending on tumor and PS typeand delivery route. Zhou and colleagues [81] foundthat the PS (BPD) distribution differed significantlybetween the same rat prostate tumors whethergrown subcutaneously or orthotopically implantedin the rat prostate gland, probably due to differ-ent degrees of vascularity. Another parameter thatcan limit direct tumor cell kill is the availability ofoxygen within the tissue undergoing PDT treatment.Because the effects of almost all PDT drugs are oxy-gen dependent, PDT cell killing typically does notoccur efficiently in hypoxic areas of tissue. In vivostudies showed that induction of tissue hypoxia, byclamping, abolished the PDT effects of porphyrins[82]. The rates of singlet oxygen generation andtherefore tissue oxygen consumption and depletionwithin the tumor are significant when both tissuePS levels and the fluence rate of light are high [83].An important parameter influencing the rate of tis-

sue oxygen consumption is photobleaching of the PSbecause the reduction of PS levels also reduces therate of photochemical oxygen consumption [84].Since all these parameters can impose limits on thedirect photodestruction of tumor cells, other mech-anisms must operate to account for the demonstra-ble success of the treatment.Since the reductions in oxygen that occur dur-

ing PDT can limit the response, it may be possibleto lower the light fluence rate to reduce oxy-gen consumption rate [85]. Busch et al. [86] com-pared Photofrin-mediated PDT of RIF1 tumors witha total dose of 135 J/cm2, delivered at a fluencerate of either 75 or 38mW/cm2. The higher flu-ence rate generated significant regions of hypoxiaeven near to tumor blood vessels and a paral-lel decrease in tumor perfusion, which were notseen with the lower fluence rate. Snyder et al.[87] compared PDT regimens of two different flu-ences (48 and 128 J/cm2) using two fluence rates


(14 and 112mW/cm2) on Colo26 murine tumorswith the object of either conserving or deplet-ing tissue oxygen during PDT. Oxygen-conservinglow fluence rate PDT of 14mW/cm2 at a fluenceof 128 J/cm2 yielded approximately 70—80% tumorcures, whereas the same fluence at the oxygen-depleting fluence rate of 112mW/cm2 yieldedapproximately 10—15% tumor cures. Low fluencerate induced higher levels of apoptosis than highfluence rate PDT. There were PDT-protected tumorregions distant from vessels in the high fluencerate conditions, confirming regional tumor hypoxia.High fluence at a low fluence rate led to abla-tion of CD31-stained endothelium, whereas thesame fluence at a high fluence rate maintainedvessel endothelium. The optimally curative PDTregimen (128 J/cm2 at 14mW/cm2) produced min-imal inflammation. Depletion of neutrophils didnot significantly change the high cure rates ofthat regimen but abolished curability in the max-imally inflammatory regimen. This fluence rateeffect was also seen in human basal cell carci-nomas treated with PF-PDT [88]. A fluence rateof 150mW/cm2 significantly diminished tumor oxy-gen levels during initial light delivery in a majorityoic

etopoohos(nutampo

V

Tdvtci

severe and persistent post-PDT tumor hypoxia [93].The mechanisms underlying the vascular effectsof PDT differ greatly with different PS. Photofrin-PDT leads to vessel constriction, macromolecularvessel leakage, leukocyte adhesion and thrombusformation, all apparently linked to platelet acti-vation and release of thromboxane [94]. PDT withcertain phthalocyanine derivatives causes primarilyvascular leakage [95], and PDT with MACE resultsin blood flow stasis primarily because of plateletaggregation [96]. All of these effects may includecomponents related to damage of the vascularendothelium. PDT may also lead to vessel constric-tion via inhibition of the production or release ofnitric oxide by the endothelium [97]. In preclinicalexperiments, the microvascular PDT responses canbe partially or completely inhibited by the adminis-tration of agents that affect eicosanoid generation,such as indomethacin [98] and this inhibition canmarkedly diminish the tumor response. Much of theabove information was obtained from studies onnormal microvasculature. Damage to the tumor-supplying normal vasculature may greatly affecttumor curability by PDT as demonstrated by thelack of tumor cures when the normal tissue sur-r[

dmtbv(clooMTtigstmstsiiboP

a

f carcinomas, but a fluence rate of 30mW/cm2

ncreased tumor oxygenation in a majority ofarcinomas.Another method to overcome the inhibitingffect of oxygen consumption during PDT is to frac-ionate the light delivery to allow re-oxygenizationf the tissue. However, this modality can be com-licated as there is no consensus on either theptimum duration of the period when the light isn, or on the duration of the dark period. Thereave been examples of one or other of these peri-ds being as short as 50ms and as long as hours. Onetudy in normal rat colon using aminolevulanic acidALA)-PDT produced up to three times more tissueecrosis with fractionated as comparedwith contin-ous light [89]. However, Babilas et al. [90] failedo find any increased effect of either light fraction-tion or low fluence rate in a hamster amelanoticelanoma treated with i.v. ALA-induced protopor-hyrin IX (PPIX); in fact the best response wasbtained with non-fractionated high fluence rate.

ascular effects

he first additional mechanism shown to operateuring PDT of tumors is the vascular effect in whichascular damage, occurring after completion ofhe PDT treatment, contributes to long-term tumorontrol [91]. Microvascular collapse can be read-ly observed following PDT [92,93] and can lead to

ounding the tumor was shielded from PDT light98].It is becoming more and more clear that therug—light interval is a crucial parameter in opti-izing PDT. The rate at which i.v. injected PS leavehe flowing blood and bind to the blood vessel wallsefore leaking out into the tumor interstitium canary markedly depending on the chemical structuresize, charge and lipophilicity). Hamblin et al. [99]arried out a study using a fluorescence scanningaser microscope to follow the pharmacokineticsf fluorescent PS and PS-conjugates in vivo in anrthotopic rat prostate cancer model obtained withatLyLu cells injected into the prostate capsule.he tumor was surgically exposed via laparotomyo allow imaging to take place. Fig. 2A shows themage captured 10min after injection of unconju-ated ce6 into the rats. Although there has beenome leakage of fluorescence out of the very tor-uous vessels in the upper left of the image, theajority of the PS is still confined to the ves-els but is no longer circulating but rather boundo the vessel walls. In Fig. 2B which shows theame region of the tumor captured 75min post-njection, all of the PS has leaked out of the vesselsnto the tumor interstitium. This technology cane used to define the optimum times for carryingut ‘‘vascular’’ or ‘‘cellular’’ PDT using the sameS.Kurohane et al. [100] showed that PDT 15minfter bloodstream injection of BPD-MA caused


Figure 2 In vivo confocal fluorescence images of an orthotopic rat prostate tumor injected with the PS ce6. (A) Theimage captured 10min after i.v. injection and (B) same region of the tumor captured 75min post-injection. Scale baris 100�m.

strong suppression of tumor growth, perhapsthrough damaging endothelial cells in the tumorneovasculature rather than through a direct cyto-toxic effect on tumor cells. Another study foundthat BPD-PDT 15min post-injection gave signif-icantly better treatment response against RIF1tumors than a 3-h interval [101]. Chen et al. studiedhypericin-PDT of RIF1 tumors comparing drug—lightintervals of 0.5 and 6 h after 5mg/kg i.v. injection[102]. Complete cures were found at the short timeinterval while no cures were seen at the 6 h inter-val. Ferrario et al. found similar results in treat-ing a mouse mammary tumor model with mono-l-aspartyl chlorin e6 (Npe6) [103]. Maximal in vivoPDT effectiveness was achieved when light treat-ments were started within 2 h of drug injection.PDT effectiveness was decreased by 50% when lighttreatments were initiated 6 h after drug injectionand was abolished with a 12-h interval betweenNPe6 injection and light exposure. Responsivenessto NPe6-mediated PDT was correlated with photo-sensitizer levels in the plasma but not in tumor tis-sue. No less than three separate reports have shownthat tumor response to Foscan (mTHPC) does notcorrelate with tumor concentration of PS. Cramers

Dolmans et al. carried out a novel fractionationscheme in PDT [105]. Using themurine MCaIV tumor,grown in the mammary fat pad they found theplasma half-life of the pyropheophorbide derivativeMV6401 was approximately 20min, and the drugwas confined to the vascular compartment shortlyafter administration but accumulated in the inter-stitial compartment at 2—6h after administration.Two equal MV6401 doses injected 4 h and 15minbefore the light administration allowed the PS tolocalize in both vascular and tumor cell compart-ments. The fractionated drug dose PDT more effec-tively induced tumor growth delay than the sametotal dose given as a single dose either at 4 h or at15min before light administration.

Immune system effects

One of the key experiments to demonstrate thiseffect was carried out by Korbelik et al. [106] whogrew the same EMT6 mammary sarcoma tumors inimmunocompetent Balb/c mice and immunodefi-cient nude or scid mice of the same background.PF-PDT led to initial ablation of the tumors in allmice but long-term cures were obtained only intbtesaPacipalaa

et al. [104] studied nude mice bearing humanmesothelioma xenografts and found there was max-imal PDT response in both tumor and skin 1—3hpost-injection despite finding the highest levels ofdrug in the tumor at 48—120 h post-injection. Joneset al. [9] studied PDT of fibrosarcomas implantedinto BDIX rats and found two times at which theanti-tumor effect of Foscan was maximal; 2 h post-injection corresponding to maximal serum concen-tration, and 24 h post-injection corresponding tomaximal tumor concentration. Maugain et al. [3]found maximum anti-tumor activity with PDT at6 h post-injection of Foscan into nude mice bear-ing Colo26 tumors and observed lower activity atshorter and longer times after injection.

he immunocompetent mice. Adoptive transfer ofone marrow from Balb/c to SCID mice restoredhe ability of PDT to produce long-term cures. Thisstablished the necessity of a functioning immuneystem for complete tumor response to PDT. Therere now thought to be two aspects to the effect ofDT on the immune response against cancer: (1)nti-tumor activity of PDT-induced inflammatoryells and (2) generation of a long-term anti-tumormmune response. These effects can be elicited byhototoxic damage that is not necessarily lethal toll tumor cells and creates an inflammatory stimu-us. PDT-induced changes in the plasma membranend membranes of cellular organelles can promptrapid activation of membranous phospholipases


[107] leading to accelerated phospholipid degrada-tion with a massive release of powerful inflamma-tory mediators [108]. Other cytokines and growthfactors that are potent immunomodulators havebeen found to be strongly enhanced in PDT-treatedmouse tumors [109,110].The inflammatory signaling after PDT initiates

a massive regulated invasion of neutrophils, mastcells, and monocytes/macrophages [111] that mayoutnumber resident cancer cells. Most notable isa rapid accumulation of large numbers of neu-trophils, which have been shown to have a profoundimpact on PDT-mediated destruction of tumors, andit has been shown that depletion of neutrophils intumor-bearing mice decreased the PDT-mediatedtumor cure rate [106]. Another class of nonspecificimmune effector cells whose activation substan-tially contributes to the anti-tumor effects of PDTis monocytes/macrophages. The tumoricidal activ-ity of these cells was found to be increased by PDTin vivo and in vitro [112]. Adjuvant treatment witha selective macrophage-activating factor derivedfrom Vitamin D3-binding protein was shown topotentiate the cures of PDT-treated tumors [113].There have been substantial advances in the

uitiiacrbpdbapimacbbCitdciatc(l

Korbelik and his co-workers have shown thatcombination therapies designed to activateconstituent parts of the immune system candramatically increase the percentage of curesof various mouse tumors. Intratumoral injectionof the immunostimulant schizophyllan (SPG) (afungal preparation) in squamous cell carcinoma(SCCVII)-bearing mice raised the relative contentof Mac-1 positive host cells infiltrating the tumorand increased PF retention in these tumors. Thetumor cure rate increased approximately threetimes when PDT was preceded by the SPG therapy.In contrast, the administration of SPG after PDTwas of no benefit for tumor control [116]. Inthe same tumor model, local injections of killedtumor cells genetically engineered to producegranulocyte—macrophage colony-stimulating fac-tor, administered three times in 48-h intervals,starting 2 days before the light treatment, substan-tially improved the curative effect of PF-mediatedPDT [117]. Intratumoral injection of an extract ofmycobacterium cell walls potentiated the responseof EMT6 tumors to PDT mediated by PF, BPD,mTHPC, or ZnPc [118]. A study designed to investi-gate the role of complement activating agents topcricgcaweso[

Bgdli(cotnuwCtmmt

nderstanding of the PDT-induced tumor-specificmmune reaction. This effect may not be relevanto the initial tumor ablation, but may be decisiven attaining long-term tumor control. Anti-cancermmunity elicited by PDT has the attributes ofn inflammation primed immune development pro-ess [114] and bears similarities to the immuneeaction induced by tumor inflammation caused byacterial vaccines or some cytokines. Macrophageshagocytize large numbers of cancer cells killed oramaged by the cytotoxic effects of PDT. Directedy powerful inflammation-associated signaling, thentigen presenting cells will process tumor-specificeptides and present them on their membranesn the context of major histocompatibility class IIolecules. Presentation of tumor peptide antigens,ccompanied by intense accessory signals, createsonditions for the recognition of tumor antigensy CD4 helper T lymphocytes. These lymphocytesecome activated and in turn sensitize cytotoxicD8 T cells to tumor-specific epitopes. The activ-ty of tumor-sensitized lymphocytes is not limitedo the original PDT-treated site but can includeisseminated and metastatic lesions of the sameancer. Thus, although the PDT treatment is local-zed to the tumor site, its effect can have systemicttributes due to the induction of an immune reac-ion. PDT-generated tumor-sensitized lymphocytesan be recovered from distant lymphoid tissuesspleen, lymph nodes) at protracted times afteright treatment [115].

otentiate PDT looked at zymosan, an alternativeomplement pathway activator, that reduced theecurrence rate of PDT-treated tumors, markedlyncreasing the percentage of permanent cures. Inontrast, a similar treatment with heat-aggregatedamma globulin (complement activator via thelassical pathway) was of no significant benefit asPDT adjuvant. Systemic complement activationith streptokinase treatment had no detectableffect on complement deposition at the tumorite without PDT, but it augmented the extentf complement activity in PDT-treated tumors119].Work from our laboratory has shown that whilePD-mediated PDT of RIF1 tumors (poorly immuno-enic) in wild-type mice leads to initial tumorisappearance but no permanent cures due toocal recurrence; when the tumors were genet-cally engineered to express the foreign proteingreen fluorescent protein from jellyfish) 100%ures and long-term resistance to rechallenge werebtained [120]. The explanation for this observa-ion is likely to be the PDT-induced immune recog-ition of GFP. In another study we explored these of low-dose cyclophosphamide in combinationith PDT. This treatment selectively kills CD4+,D25+ T-suppressor cells and significantly increaseshe number of cures of the highly aggressive andetastatic J774 tumor in Balb/c mice. Again curedice were resistant to rechallenge with the sameumor [121].


An interesting observation was also made byGollnick et al., who reported that a tumor-celllysate that was isolated following PF-PDT in vitrowith could be used to vaccinate mice against thedevelopment of tumors from the same cell line[122]. PDT-induced lysates were more effectivethan those made from cells that were killed byultraviolet, ionizing irradiation or freeze—thaw andonly PDT-generated lysates were able to activatedendritic cells to express IL-12.

Conclusion

There may be a trend amongst second and thirdgeneration PS away from the formerly desired prop-erties of localizing well in tumors towards fast act-ing vascular PS. This shift of emphasis is explainedby several reasons. Firstly the achievement of hightumor-to-normal tissue ratios frequently involvedlong intervals between injection and illuminationthat were needed for the PS to clear from normaltissue, and that delay necessitated patients return-ing twice to the hospital for PS administration andfor illumination. Secondly these long-lasting PS also

Acknowledgements

Ana P. Castano was supported by a Departmentof Defense CDMRP Breast Cancer Research Grant(W81XWH-04-1-0676). Tatiana N. Demidova wassupported by a Wellman Center of Photomedicinegraduate student fellowship. Michael R. Hamblinwas supported by the US National Institutes ofHealth (R01-CA/AI838801). We are grateful toTayyaba Hasan for support.

References

[1] Castano AP, Demidova TN, Hamblin MR. Mechanisms inphotodynamic therapy: part one—–photosensitizers, pho-tochemistry and cellular localization. Photodiagn Photo-dyn Ther 2004;1:279—93.

[2] Castano AP, Demidova TN, Hamblin MR. Mechanisms inphotodynamic therapy: part two — cellular signaling, cellmetabolism andmodes of cell death. Photodiagn PhotodynTher 2005;2:1—23.

[3] Maugain E, Sasnouski S, Zorin V, Merlin JL, GuilleminF, Bezdetnaya L. Foscan-based photodynamic treatmentin vivo: correlation between efficacy and Foscan accu-mulation in tumor, plasma and leukocytes. Oncol Rep2004;12:639—45.

tended to have more pronounced skin photosensi-tivity problems that could last for an inconvenientlylong time. Thirdly when PS are compared on thebasis of moles injected, absorption spectrum andlight delivered, it is likely that fast acting vascularPS are actually more effective in destroying tumorsthan first generation tumor-localizing dyes. Some-what paradoxically there are also ongoing effortsto take advantage of the tumor-localizing proper-ties of tetrapyrrole PS to deliver other imaging andtherapeutically active molecules more specificallyto tumors.While the immune stimulating effects of PDT

are well established in animal models these effectshave not been much studied in clinical trials. Thereare scattered anecdotal reports of patients withterminal cancer receiving palliative PDT who sur-vived much longer than expected. While theseoccurrences might be attributed to stimulation ofthe immune system, and while the laboratory assaysnecessary to confirm the generation of an anti-tumor immune response exist, they have neverbeen applied to patients in a systematic manner. Itis known that the three commonest cancer thera-pies; surgery, ionizing radiation and chemotherapyall tend to be immunosuppressive. It is thereforepossible that PDT could emerge as the ideal cancertreatment: a local therapy capable of efficientlydestroying tumors while at the same time sensi-tizing the immune system to seek out and destroymetatstases.

[4] Bellnier DA, Greco WR, Parsons JC, Oseroff AR, Kuebler A,Dougherty TJ. An assay for the quantitation of Photofrinin tissues and fluids. Photochem Photobiol 1997;66:237—44.

[5] Gudgin Dickson EF, Holmes H, Jori G, et al. On the sourceof the oscillations observed during in vivo zinc phthalo-cyanine fluorescence pharmacokinetic measurements inmice. Photochem Photobiol 1995;61:506—9.

[6] Holmes H, Kennedy JC, Pottier R, Rossi R, Weagle G. Arecipe for the preparation of a rodent food that elimi-nates chlorophyll-based tissue fluorescence. J PhotochemPhotobiol B 1995;29:199.

[7] Cai H, Lim CK. Comparison of HPLC, capillary elec-trophoretic and direct spectrofluorimetric methods forthe determination of temoporfin-poly(ethylene glycol)conjugates in plasma. Analyst 1998;123:2243—5.

[8] Little FM, Gomer CJ, Hyman S, Apuzzo ML. Observa-tions in studies of quantitative kinetics of tritium labelledhematoporphyrin derivatives (HpDI and HpDII) in thenormal and neoplastic rat brain model. J Neurooncol1984;2:361—70.

[9] Jones HJ, Vernon DI, Brown SB. Photodynamic therapyeffect of m-THPC (Foscan) in vivo: correlation with phar-macokinetics. Br J Cancer 2003;89:398—404.

[10] Bellnier DA, Ho YK, Pandey RK, Missert JR, Dougherty TJ.Distribution and elimination of Photofrin II in mice. Pho-tochem Photobiol 1989;50:221—8.

[11] Schuitmaker JJ, Feitsma RI, Journee-De Korver JG,Dubbelman TM, Pauwels EK. Tissue distribution of bacteri-ochlorin a labelled with 99mTc-pertechnetate in hamsterGreene melanoma. Int J Radiat Biol 1993;64:451—8.

[12] Frisoli JK, Tudor EG, Flotte TJ, Hasan T, DeutschTF, Schomacker KT. Pharmacokinetics of a fluores-cent drug using laser-induced fluorescence. Cancer Res1993;53:5954—61.

[13] Sheng C, Pogue BW, Wang E, Hutchins JE, HoopesPJ. Assessment of photosensitizer dosimetry and tissue


damage assay for photodynamic therapy in advanced-stage tumors. Photochem Photobiol 2004;79:520—5.

[14] Bellnier DA, Dougherty TJ. A preliminary pharmacokineticstudy of intravenous Photofrin in patients. J Clin Laser MedSurg 1996;14:311—4.

[15] Moriwaki SI, Misawa J, Yoshinari Y, Yamada I, TakigawaM, Tokura Y. Analysis of photosensitivity in Japanesecancer-bearing patients receiving photodynamic therapywith porfimer sodium (Photofrin). Photodermatol Pho-toimmunol Photomed 2001;17:241—3.

[16] Bellnier DA, Greco WR, Loewen GM, et al. Populationpharmacokinetics of the photodynamic therapy agent 2-[1-hexyloxyethyl]-2-devinyl pyropheophorbide-a in can-cer patients. Cancer Res 2003;63:1806—13.

[17] Brun PH, DeGroot JL, Dickson EF, Farahani M, Pot-tier RH. Determination of the in vivo pharmacoki-netics of palladium-bacteriopheophorbide (WST09) inEMT6 tumour-bearing Balb/c mice using graphite furnaceatomic absorption spectroscopy. Photochem Photobiol Sci2004;3:1006—10.

[18] Egorin MJ, Zuhowski EG, Sentz DL, Dobson JM, CalleryPS, Eiseman JL. Plasma pharmacokinetics and tissue dis-tribution in CD2F1 mice of Pc4 (NSC 676418), a siliconephthalocyanine photodynamic sensitizing agent. CancerChemother Pharmacol 1999;44:283—94.

[19] Ismail MS, Dressler C, Koeppe P, et al. Pharmacokineticanalysis of octa-alpha-butyloxy-zinc phthalocyanine inmice bearing Lewis lung carcinoma. J Clin Laser Med Surg1997;15:157—61.

[20] Zuk MM, Rihter BD, Kenney ME, Rodgers MA, Kreimer-

[30] Jori G, Reddi E. The role of lipoproteins in the deliv-ery of tumour-targeting photosensitizers. Int J Biochem1993;25:1369—75.

[31] Kessel D, Morgan A, Garbo GM. Sites and efficacy of photo-damage by tin etiopurpurin in vitro using different deliv-ery systems. Photochem Photobiol 1991;54:193—6.

[32] Kongshaug M, Moan J, Brown SB. The distribution of por-phyrins with different tumour localising ability amonghuman plasma proteins. Br J Cancer 1989;59:184—8.

[33] Maziere JC, Santus R, Morliere P, et al. Cellular uptakeand photosensitizing properties of anticancer porphyrinsin cell membranes and low and high density lipoproteins.J Photochem Photobiol B 1990;6:61—8.

[34] Korbelik M. Low density lipoprotein receptor pathway inthe delivery of Photofrin: how much is it relevant forselective accumulation of the photosensitizer in tumors?J Photochem Photobiol B 1992;12:107—9.

[35] Yuan F, Leunig M, Berk DA, Jain RK. Microvascularpermeability of albumin, vascular surface area, andvascular volume measured in human adenocarcinomaLS174T using dorsal chamber in SCID mice. Microvasc Res1993;45:269—89.

[36] Allison BA, Pritchard PH, Levy JG. Evidence for low-density lipoprotein receptor-mediated uptake of benzo-porphyrin derivative. Br J Cancer 1994;69:833—9.

[37] Roberts WG, Hasan T. Role of neovasculature and vascu-lar permeability on the tumor retention of photodynamicagents. Cancer Res 1992;52:924—30.

[38] Korbelik M, Krosl G. Photofrin accumulation in malignantand host cell populations of a murine fibrosarcoma. Pho-

Birnbaum M. Pharmacokinetic and tissue distributionstudies of the photosensitizer bis(di-isobutyl octadecyl-siloxy)silicon 2,3-naphthalocyanine (isoBOSINC) in nor-mal and tumor-bearing rats. Photochem Photobiol1994;59:66—72.

[21] Chan WS, Marshall JF, Svensen R, Bedwell J, Hart IR.Effect of sulfonation on the cell and tissue distributionof the photosensitizer aluminum phthalocyanine. CancerRes 1990;50:4533—8.

[22] Woodburn KW, Stylli S, Hill JS, Kaye AH, Reiss JA, PhillipsDR. Evaluation of tumour and tissue distribution of por-phyrins for use in photodynamic therapy. Br J Cancer1992;65:321—8.

[23] Alvarez MG, Moran F, Yslas EI, et al. Pharmacoki-netic and tumour-photosensitizing properties ofmethoxyphenyl porphyrin derivative. Biomed Pharma-cother 2003;57:163—8.

[24] Richter AM, Cerruti-Sola S, Sternberg ED, Dolphin D, LevyJG. Biodistribution of tritiated benzoporphyrin deriva-tive (3H-BPD-MA), a new potent photosensitizer, in nor-mal and tumor-bearing mice. J Photochem Photobiol B1990;5:231—44.

[25] Boyle RW, Dolphin D. Structure and biodistribution rela-tionships of photodynamic sensitizers. Photochem Photo-biol 1996;64:469—85.

[26] Figge FH, Weiland GS, Manganiello LO. Affinity of neoplas-tic, embryonic and traumatized tissues for porphyrins andmetalloporphyrins. Proc Soc Exp Biol Med 1948;68:640.

[27] Hamblin MR, Newman EL. On the mechanism of thetumour-localising effect in photodynamic therapy. J Pho-tochem Photobiol B 1994;23:3—8.

[28] Larroque C, Pelegrin A, Van Lier JE. Serum albumin as avehicle for zinc phthalocyanine: photodynamic activitiesin solid tumour models. Br J Cancer 1996;74:1886—90.

[29] Jori G. In vivo transport and pharmacokinetic behav-ior of tumour photosensitizers. Ciba Found Symp1989;146:78—86.

tochem Photobiol 1995;62:162—8.[39] Korbelik M, Krosl G, Chaplin DJ. Photofrin uptake by

murine macrophages. Cancer Res 1991;51:2251—5.[40] Pottier R, Kennedy JC. The possible role of ionic species

in selective biodistribution of photochemotherapeuticagents toward neoplastic tissue. J Photochem PhotobiolB 1990;8:1—16.

[41] Freitas I. Lipid accumulation: the common feature tophotosensitizer-retaining normal and malignant tissues[news]. J Photochem Photobiol B 1990;7:359—61.

[42] Graham A, Li G, Chen Y, et al. Structure-activity rela-tionship of new octaethylporphyrin-based benzochlorinsas photosensitizers for photodynamic therapy. PhotochemPhotobiol 2003;77:561—6.

[43] Wendler G, Lindemann P, Lacapere JJ, PapadopoulosV. Protoporphyrin IX binding and transport by recom-binant mouse PBR. Biochem Biophys Res Commun2003;311:847—52.

[44] Lampidis TJ, Bernal SD, Summerhayes IC, Chen LB. Selec-tive toxicity of rhodamine 123 in carcinoma cells in vitro.Cancer Res 1983;43:716—20.

[45] Castro DJ, Saxton RE, Rodgerson DO, Fu YS, Bhuta SM,Fetterman HR, et al. Rhodamine-123 as a new laser dye: invivo study of dye effects on murine metabolism, histologyand ultrastructure. Laryngoscope 1989;99:1057—62.

[46] Reddi E. Role of delivery vehicles for photosensitizers inthe photodynamic therapy of tumours. J Photochem Pho-tobiol B 1997;37:189—95.

[47] Woodburn K, Sykes E, Kessel D. Interactions of SolutolHS 15 and Cremophor EL with plasma lipoproteins. Int JBiochem Cell Biol 1995;27:693—9.

[48] Woodburn K, Chang CK, Lee S, Henderson B, Kessel D.Biodistribution and PDT efficacy of a ketochlorin photo-sensitizer as a function of the delivery vehicle. PhotochemPhotobiol 1994;60:154—9.

[49] Ginevra F, Biffanti S, Pagnan A, Biolo R, Reddi E,Jori G. Delivery of the tumour photosensitizer zinc(II)-


phthalocyanine to serum proteins by different liposomes:studies in vitro and in vivo. Cancer Lett 1990;49:59—65.

[50] Jiang F, Lilge L, Logie B, Li Y, Chopp M. Photodynamic ther-apy of 9L gliosarcoma with liposome-delivered photofrin.Photochem Photobiol 1997;65:701—6.

[51] Richter AM, Waterfield E, Jain AK, Canaan AJ, Allison BA,Levy JG. Liposomal delivery of a photosensitizer, ben-zoporphyrin derivative monoacid ring A (BPD), to tumortissue in a mouse tumor model. Photochem Photobiol1993;57:1000—6.

[52] Polo L, Segalla A, Jori G, et al. Liposome-delivered 131I-labelled Zn(II)-phthalocyanine as a radiodiagnostic agentfor tumours. Cancer Lett 1996;109:57—61.

[53] Allemann E, Rousseau J, Brasseur N, Kudrevich SV, LewisK, van Lier JE. Photodynamic therapy of tumours withhexadecafluoro zinc phthalocyanine formulated in PEG-coated poly(lactic acid) nanoparticles. Int J Cancer1996;66:821—4.

[54] Margalit R, Silbiger E. Albumin microspheres as deliverysystems for photodynamic drugs: physico-chemical studiesand their implications for in vivo situations. J Microencap-sul 1985;2:183—96.

[55] Barel A, Jori G, Perin A, Romandini P, Pagnan A, BiffantiS. Role of high-, low- and very low-density lipoproteins inthe transport and tumor-delivery of hematoporphyrin invivo. Cancer Lett 1986;32:145—50.

[56] Morliere P, Kohen E, Reyftmann JP, et al. Photosensitiza-tion by porphyrins delivered to L cell fibroblasts by humanserum low density lipoproteins. A microspectrofluoromet-ric study. Photochem Photobiol 1987;46:183—91.

[67] Wilmes LJ, Hoehn-Berlage M, Els T, et al. In vivo relaxom-etry of three brain tumors in the rat: effect of Mn-TPPS,a tumor-selective contrast agent. J Magn Reson Imaging1993;3:5—12.

[68] Ikezaki K, Nomura T, Takahashi M, Fritz-Zieroth B, Ina-mura T, Fukui M. Selective and prolonged MRI enhance-ment by Mn-TPPS in an experimental rat brain tumourwith peripheral benzodiazepine receptors. Neurol Res1994;16:393—7.

[69] Schaffer M, Ertl-Wagner B, Schaffer PM, et al. The appli-cation of Photofrin II as a sensitizing agent for ionizingradiation—–a new approach in tumor therapy? Curr MedChem 2005;12:1209—15.

[70] Schaffer M, Schaffer PM, Corti L, et al. Photofrin as aspecific radiosensitizing agent for tumors: studies in com-parison to other porphyrins, in an experimental in vivomodel. J Photochem Photobiol B 2002;66:157—64.

[71] Schaffer M, Schaffer PM, Vogesser M, et al. Application ofPhotofrin II as a specific radiosensitising agent in patientswith bladder cancer—–a report of two cases. PhotochemPhotobiol Sci 2002;1:686—9.

[72] Young SW, Qing F, Harriman A, et al. Gadolin-ium(III) texaphyrin: a tumor selective radiation sensi-tizer that is detectable by MRI. Proc Natl Acad Sci USA1996;93:6610—5.

[73] Young SW, Sidhu MK, Qing F, et al. Preclinical evaluation ofgadolinium (III) texaphyrin complex. A new paramagneticcontrast agent for magnetic resonance imaging. InvestRadiol 1994;29:330—8.

[74] Magda D, Lepp C, Gerasimchuk N, et al. Redox cycling by
[57] Zhou CN, Milanesi C, Jori G. An ultrastructural compar-
ative evaluation of tumors photosensitized by porphyrinsadministered in aqueous solution, bound to liposomes orto lipoproteins. Photochem Photobiol 1988;48:487—92.

[58] Schmidt-Erfurth U, Bauman W, Gragoudas E, et al. Pho-todynamic therapy of experimental choroidal melanomausing lipoprotein-delivered benzoporphyrin. Ophthalmol-ogy 1994;101:89—99.

[59] Pani R, Pellegrini R, Cinti MN, et al. New devices forimaging in nuclear medicine. Cancer Biother Radiopharm2004;19:121—8.

[60] Babbar AK, Singh AK, Goel HC, Chauhan UP, SharmaRK. Evaluation of (99m)Tc-labeled photosan-3, a hemato-porphyrin derivative, as a potential radiopharmaceuti-cal for tumor scintigraphy. Nucl Med Biol 2000;27:587—92.

[61] Link EM, Costa DC, Lane D, Blower PJ, Spittle MF. Radioio-dinated methylene blue for diagnosing early melanomametastases. Lancet 1996;348:753.

[62] Link EM, Carpenter RN. 211At-methylene blue for targetedradiotherapy of human melanoma xenografts: treatmentof cutaneous tumors and lymph node metastases. CancerRes 1992;52:4385—90.

[63] Link EM. Targeting melanoma with 211At/131I-methyleneblue: preclinical and clinical experience. Hybridoma1999;18:77—82.

[64] Evstigneeva RP, Zaitsev AV, Luzgina VN, Ol’Shevskaya VA,Shtil AA. Carboranylporphyrins for boron neutron capturetherapy of cancer. Curr Med Chem Anti-Cancer Agents2003;3:383—92.

[65] Hill JS, Kahl SB, Kaye AH, et al. Selective tumor uptakeof a boronated porphyrin in an animal model of cerebralglioma. Proc Natl Acad Sci USA 1992;89:1785—9.

[66] Rosenthal MA, Kavar B, Hill JS, et al. Phase I and pharma-cokinetic study of photodynamic therapy for high-gradegliomas using a novel boronated porphyrin. J Clin Oncol2001;19:519—24.

motexafin gadolinium enhances cellular response to ion-izing radiation by forming reactive oxygen species. Int JRadiat Oncol Biol Phys 2001;51:1025—36.

[75] Miller RA, Woodburn K, Fan Q, Renschler MF, Sessler JL,Koutcher JA. In vivo animal studies with gadolinium (III)texaphyrin as a radiation enhancer. Int J Radiat Oncol BiolPhys 1999;45:981—9.

[76] Evens AM. Motexafin gadolinium: a redox-active tumorselective agent for the treatment of cancer. Curr OpinOncol 2004;16:576—80.

[77] Henderson BW, Waldow SM, Mang TS, Potter WR, MalonePB, Dougherty TJ. Tumor destruction and kinetics of tumorcell death in two experimental mouse tumors followingphotodynamic therapy. Cancer Res 1985;45:572—6.

[78] Chan WS, Brasseur N, La Madeleine C, van Lier JE. Evi-dence for different mechanisms of EMT-6 tumor necro-sis by photodynamic therapy with disulfonated aluminumphthalocyanine or photofrin: tumor cell survival and bloodflow. Anticancer Res 1996;16:1887—92.

[79] Korbelik M, Krosl G. Cellular levels of photosensitisers intumours: the role of proximity to the blood supply. Br JCancer 1994;70:604—10.

[80] Lee CC, Pogue BW, O’Hara JA, et al. Spatial heterogeneityand temporal kinetics of photosensitizer (AlPcS2) concen-tration in murine tumors RIF-1 and MTG-B. PhotochemPhotobiol Sci 2003;2:145—50.

[81] Zhou X, Pogue BW, Chen B, Hasan T. Analysis of effectivemolecular diffusion rates for verteporfin in subcutaneousversus orthotopic Dunning prostate tumors. PhotochemPhotobiol 2004;79:323—31.

[82] Gomer CJ, Razum NJ. Acute skin response in albino micefollowing porphyrin photosensitization under oxic andanoxic conditions. Photochem Photobiol 1984;40:435—9.

[83] Foster TH, Murant RS, Bryant RG, Knox RS, Gibson SL, HilfR. Oxygen consumption and diffusion effects in photody-namic therapy. Radiat Res 1991;126:296—303.


[84] Georgakoudi I, Nichols MG, Foster TH. The mech-anism of Photofrin photobleaching and its conse-quences for photodynamic dosimetry. Photochem Photo-biol 1997;65:135—44.

[85] Xu T, Li Y, Wu X. Application of lower fluence rate for lessmicrovasculature damage and greater cell-killing duringphotodynamic therapy. Lasers Med Sci 2004;19:150—4.

[86] Busch TM, Wileyto EP, Emanuele MJ, et al. Photodynamictherapy creates fluence rate-dependent gradients in theintratumoral spatial distribution of oxygen. Cancer Res2002;62:7273—9.

[87] Snyder JW, Greco WR, Bellnier DA, Vaughan L, Hender-son BW. Photodynamic therapy: a means to enhanced drugdelivery to tumors. Cancer Res 2003;63:8126—31.

[88] Henderson BW, Busch TM, Vaughan LA, et al. Photofrinphotodynamic therapy can significantly deplete or pre-serve oxygenation in human basal cell carcinomas dur-ing treatment, depending on fluence rate. Cancer Res2000;60:525—9.

[89] Curnow A, Haller JC, Bown SG. Oxygen monitoring dur-ing 5-aminolaevulinic acid induced photodynamic ther-apy in normal rat colon. Comparison of continuous andfractionated light regimes. J Photochem Photobiol B2000;58:149—55.

[90] Babilas P, Schacht V, Liebsch G, et al. Effects of lightfractionation and different fluence rates on photodynamictherapy with 5-aminolaevulinic acid in vivo. Br J Cancer2003;88:1462—9.

[91] Abels C. Targeting of the vascular system of solid tumoursby photodynamic therapy (PDT). Photochem Photobiol Sci

[101] Chen B, Pogue BW, Goodwin IA, et al. Blood flow dynamicsafter photodynamic therapy with verteporfin in the RIF-1tumor. Radiat Res 2003;160:452—9.

[102] Chen B, Roskams T, de Witte PA. Antivascular tumor eradi-cation by hypericin-mediated photodynamic therapy. Pho-tochem Photobiol 2002;76:509—13.

[103] Ferrario A, Kessel D, Gomer CJ. Metabolic propertiesand photosensitizing responsiveness of mono-l-aspartylchlorin e6 in a mouse tumor model. Cancer Res1992;52:2890—3.

[104] Cramers P, Ruevekamp M, Oppelaar H, Dalesio O,Baas P, Stewart FA. Foscan uptake and tissue distribu-tion in relation to photodynamic efficacy. Br J Cancer2003;88:283—90.

[105] Dolmans DE, Kadambi A, Hill JS, et al. Targeting tumor vas-culature and cancer cells in orthotopic breast tumor byfractionated photosensitizer dosing photodynamic ther-apy. Cancer Res 2002;62:4289—94.

[106] Korbelik M, Krosl G, Krosl J, Dougherty GJ. The roleof host lymphoid populations in the response of mouseEMT6 tumor to photodynamic therapy. Cancer Res1996;56:5647—52.

[107] Agarwal ML, Larkin HE, Zaidi SI, Mukhtar H, OleinickNL. Phospholipase activation triggers apoptosis inphotosensitized mouse lymphoma cells. Cancer Res1993;53:5897—902.

[108] Yamamoto N, Homma S, Sery TW, Donoso LA, HooberJK. Photodynamic immunopotentiation: in vitro activationof macrophages by treatment of mouse peritoneal cellswith haematoporphyrin derivative and light. Eur J Cancer

2004;3:765—71.[92] Star WM, Marijnissen HP, van den Berg-Blok AE, Versteeg

JA, Franken KA, Reinhold HS. Destruction of rat mammarytumor and normal tissue microcirculation by hematopor-phyrin derivative photoradiation observed in vivo in sand-wich observation chambers. Cancer Res 1986;46:2532—40.

[93] Henderson BW, Fingar VH. Relationship of tumor hypoxiaand response to photodynamic treatment in an experi-mental mouse tumor. Cancer Res 1987;47:3110—4.

[94] Fingar VH, Siegel KA, Wieman TJ, Doak KW. The effects ofthromboxane inhibitors on the microvascular and tumorresponse to photodynamic therapy. Photochem Photobiol1993;58:393—9.

[95] Fingar VH, Wieman TJ, Karavolos PS, Doak KW, Ouel-let R, van Lier JE. The effects of photodynamic therapyusing differently substituted zinc phthalocyanines on ves-sel constriction, vessel leakage and tumor response. Pho-tochem Photobiol 1993;58:251—8.

[96] McMahon KS, Wieman TJ, Moore PH, Fingar VH. Effects ofphotodynamic therapy using mono-l-aspartyl chlorin e6 onvessel constriction, vessel leakage, and tumor response.Cancer Res 1994;54:5374—9.

[97] Gilissen MJ, van de Merbel-de Wit LE, Star WM, KosterJF, Sluiter W. Effect of photodynamic therapy on theendothelium-dependent relaxation of isolated rat aortas.Cancer Res 1993;53:2548—52.

[98] Fingar VH, Wieman TJ, Doak KW. Role of thrombox-ane and prostacyclin release on photodynamic therapy-induced tumor destruction. Cancer Res 1990;50:2599—603.

[99] Hamblin MR, Rajadhyaksha M, Momma T, Soukos NS, HasanT. In vivo fluorescence imaging of the transport of chargedchlorin e6 conjugates in a rat orthotopic prostate tumour.Br J Cancer 1999;81:261—8.

[100] Kurohane K, Tominaga A, Sato K, North JR, Namba Y,Oku N. Photodynamic therapy targeted to tumor-inducedangiogenic vessels. Cancer Lett 2001;167:49—56.

1991;27:467—71.[109] Herman S, Kalechman Y, Gafter U, Sredni B, Malik

Z. Photofrin II induces cytokine secretion by mousespleen cells and human peripheral mononuclear cells.Immunopharmacology 1996;31:195—204.

[110] Gollnick SO, Liu X, Owczarczak B, Musser DA, HendersonBW. Altered expression of interleukin 6 and interleukin 10as a result of photodynamic therapy in vivo. Cancer Res1997;57:3904—9.

[111] Krosl G, Korbelik M, Dougherty GJ. Induction ofimmune cell infiltration into murine SCCVII tumourby photofrin-based photodynamic therapy. Br J Cancer1995;71:549—55.

[112] Yamamoto N, Homma S, Nakagawa Y, et al. Activationof mouse macrophages by in vivo and in vitro treat-ment with a cyanine dye, lumin. J Photochem Photobiol B1992;13:295—306.

[113] Korbelik M, Naraparaju VR, Yamamoto N. Macrophage-directed immunotherapy as adjuvant to photodynamictherapy of cancer. Br J Cancer 1997;75:202—7.

[114] Korbelik M. Induction of tumor immunity by photody-namic therapy. J Clin Laser Med Surg 1996;14:329—34.

[115] Korbelik M, Dougherty GJ. Photodynamic therapy-mediated immune response against subcutaneous mousetumors. Cancer Res 1999;59:1941—6.

[116] Krosl G, Korbelik M. Potentiation of photodynamic ther-apy by immunotherapy: the effect of schizophyllan (SPG).Cancer Lett 1994;84:43—9.

[117] Krosl G, Korbelik M, Krosl J, Dougherty GJ. Potentiationof photodynamic therapy-elicited antitumor response bylocalized treatment with granulocyte-macrophage colony-stimulating factor. Cancer Res 1996;56:3281—6.

[118] Korbelik M, Cecic I. Enhancement of tumour response tophotodynamic therapy by adjuvant mycobacterium cell-wall treatment. J Photochem Photobiol B 1998;44:151—8.


[119] Korbelik M, Sun J, Cecic I, Serrano K. Adjuvant treatmentfor complement activation increases the effectiveness ofphotodynamic therapy of solid tumors. Photochem Photo-biol Sci 2004;3:812—6.

[120] Castano AP, Liu Q, Hamblin MR. Photodynamic therapycures green fluorescent protein expressing RIF1 tumors inmice. Proc SPIE 2004;5319:50—9.

[121] Castano AP, Hamblin MR. Combination of photodynamictherapy and low-dose cyclophosphamide as a treatmentfor metastatic murine tumors. Proc Am Assoc Cancer Res2004;45:997.

[122] Gollnick SO, Vaughan L, Henderson BW. Generation ofeffective antitumor vaccines using photodynamic therapy.Cancer Res 2002;62:1604—8.

Mechanisms in photodynamic therapy: Part three—Photosensitizer pharmacokinetics, biodistribution,...

Documents

Transcript of Mechanisms in photodynamic therapy: Part three—Photosensitizer pharmacokinetics, biodistribution,...