Photodynamic Nanomedicine in the Treatment of Solid Tumors: Perspectives and Challenges

Abstract

Photodynamic therapy (PDT) is a promising treatment strategy where activation of photosensitizer drugs with specific wavelengths of light results in energy transfer cascades that ultimately yield cytotoxic reactive oxygen species which can render apoptotic and necrotic cell death. Without light the photosensitizer drugs are minimally toxic and the photoactivating light itself is non-ionizing. Therefore, harnessing this mechanism in tumors provides a safe and novel way to selectively eradicate tumor with reduced systemic toxicity and side effects on healthy tissues. For successful PDT of solid tumors, it is necessary to ensure tumor-selective delivery of the photosensitizers, as well as, the photoactivating light and to establish dosimetric correlation of light and drug parameters to PDT-induced tumor response. To this end, the nanomedicine approach provides a promising way towards enhanced control of photosensitizer biodistribution and tumor-selective delivery. In addition, refinement of nanoparticle designs can also allow incorporation of imaging agents, light delivery components and dosimetric components. This review aims at describing the current state-of-the-art regarding nanomedicine strategies in PDT, with a comprehensive narrative of the research that has been carried out in vitro and in vivo, with a discussion of the nanoformulation design aspects and a perspective on the promise and challenges of PDT regarding successful translation into clinical application.

1. Introduction

The global burden of cancer-related morbidity and mortality continues to be significant, with close to 8 million deaths annually [1]. Current clinical strategies for treating cancer include surgery, radiation therapy, chemotherapy, and more recently, immunotherapy and other small molecule-based therapies, along with combination of these strategies. Surgery is often the first attempted strategy though it may not be a stand-alone option if metastasis has occurred. Surgical removal of tumor is also not a viable method in cases where the tumor is located in sensitive areas, for example, in the vicinity of the spinal cord or near other vital organs. Radiation therapy damages the DNA of cells due to the ionizing nature of the radiation, which can prevent tumor cell replication and possibly shrink the tumor, if selectively focused on the tumor. However, radiotherapy frequently leads to long term side effects including damage to normal neighboring cells, formation of scar tissue, and immunosuppression. Chemotherapeutic agents are either cytostatic or cytotoxic molecules that can arrest or kill quickly dividing cell lines within the body. A chemotherapeutic agent is usually intravenously delivered yielding systemic effects. This approach may be advantageous when dealing with metastasized tumors but the concomitant challenge involves balancing the therapeutic effect of the drug with undesirable systemic toxic sideeffects [2]. In recent years the benefits of chemotherapy and other small molecule drugs have been significantly enhanced via the nanomedicine approach where the drug molecules are packaged within nanovehicles that keep the drug in circulation for longer periods of time (by preventing renal clearance and non-specific uptake) and allow increased uptake within the tumors via extravasation through the tumor-associated leaky vasculature (the enhanced permeation and retention or EPR effect) [3–5]. Immunotherapy works by enabling the patient's natural immune system to fight the cancer or reducing inherent signaling cascade mechanisms that promote cancer proliferation and aggression, by using cancer vaccines and cancer cell receptor-targeted monoclonal antibodies. However, cancer cells are prone to mutation leading to limited effectiveness of highly specific immunotherapies. Nevertheless, immunotherapy strategies in combination with chemo- or radiotherapy have recently shown clinical benefit in certain cancers [6–8].

While refinement of the above described conventional cancer treatment modalities to improve their tumorselectivity is a significant part of current cancer research, investigations have also focused on developing alternate treatment modalities that may be safer or more beneficial in cases where extensive surgery, chemotherapy or radiotherapy may be limited due to risk of cosmetic and functional tissue damage. To this end, a promising strategy is photodynamic therapy (PDT). PDT involves three components: a photosensitizer (PS) drug, a specific wavelength of drug-activating light and oxygen. Light activation of a photosensitizer results in energy transfer cascades that ultimately yield cytotoxic reactive oxygen species (Figure 1) which can then render apoptotic and necrotic cell death. Since the first modern demonstration of harnessing this mechanism to kill tumor cells by Dougherty et al. in 1978, PDT has undergone extensive investigations in cancer treatment [9] (Figure 2). A major advantage of PDT over conventional chemotherapy is that both the PS itself is minimally toxic in the absence of light and hence PS accumulation in non-specific tissues confers minimal systemic toxicity. Furthermore, compared to radiotherapy, the activating light is non-ionizing and hence its effect on tissues without the PS drug is not harmful. As a result, PDT has the potential to be repeated safely if needed without the risk of harming neighboring healthy tissue [10]. This makes PDT very attractive for tumors where loco-regionally recurrent treatment may be needed. By virtue of these characteristics, PDT can offer the possibility of dualselectivity in cancer therapy by developing technologies that ensure PS accumulation selectively in the tumor and light irradiation selectively on the tumor [11]. The clinical interest in PDT is exhibited by health agency approvals with the PS formulations Photofrin®, Levulan®, and Visudyne® in the US, various other formulations (Foscan®, Photosense®, Metvix®) in other countries, and ongoing clinical trials [12, 13]. Table 1 provides a listing of some of the current clinically approved uses of PDT. As evident from Table 1, PDT is currently being used mainly to treat skin diseases and easily accessible malignant and pre-malignant lesions. One reason for this limited clinical repertoire is the challenge associated with PS delivery. Most PS molecules are highly hydrophobic and hence difficult to be incorporated into intravenously deliverable formulations. In most current clinical applications the PS is formulated in lipidic (e.g. Cremophor) or organic (e.g. dextrose, propylene glycol etc.) excipients which are reasonable for topical and local administration, but can lead to unpredictable biodistribution profiles, allergy, hypersensitivity and toxicity issues if used intravenously, especially in multiple doses. The unpredictable biodistribution can lead to PS accumulation in easily light-exposed uninvolved tissues like the eyes and skin, which in turn result in prolonged phototoxicity and photosensitivity in patients when exposed to ambient light. One typical example is Foscan®, an ethanol/propylene glycol formulation of the PS temoporfin, which had failed to achieve FDA approval for treatment of head-and-neck cancers due to poor tumor selectivity, high plasma retention, and serious burns arising from skin photosensitivity [14–17]. As for excipients like Cremophor, the issues with allergy, hypersensitivity and toxicity are well known [18,19]. The other challenge in enhancing the clinical repertoire of PDT is appropriate delivery of the drug-activating light. Easily accessible tissues can undergo convenient illumination for PDT, but deep tissue tumors may present limitations to light penetration. For example, blue light (400-450 nm) penetrates < 1 mm into tissue while orange light (590-620 nm) penetrates approximately 1.5 mm and red light (620-750 nm) penetrates even further up to 3 mm [20].

Figure 1.

Jablonski diagram depicting the electronic transition states and energy transfer phenomena between the photosensitizer molecule and oxygen in photodynamic therapy that ultimately leads to oxidative cell damage.

Figure 2. The history, development and current state-of-the-art for PDT.

Table 1. Current clinically-approved photosensitizer formulations and their corresponding applications.

Trade Name	Drug		Company		Approval Information		Application		Excipient Formulation	Activation Wavelength	Chemical Structure
Photofrin®	Hematoporphyrin derivative		Axcan Pharm		Canada, 1993 Japan, 1994 USA, 1995		Bladder cancer, lung cancer, esophageal cancer, Barrett'sesophagus		Dextrose or NaCl injection	630 nm
Metvixia®	Methyl aminolevulinate		Photocure/Galderma		USA,2004 EU, 2001 New Zealand, 2002 Australia, 2003		Actinic keratosis, basal cell carcinoma, non-melanoma skin cancer		Topical cream (includes cetostearyl alcohol and peanut oil)	630 nm
Foscan®	Temeporfin/m-THPC		Biolitec Pharma		EU, 2001		Head and neck cancer		Ethanol/Propylene Glycol	652 nm
Visudyne®	Verteporfin/benzoporphyrin derivative monoacid		Novartis		USA, 2000 Canada, 2000 EU, 2000 Japan, 2003		Wet age-related macular degeneration, pathologic myopia, histoplasmosis		Liposomal	690 nm
Levulan®	Aminolevulinic acid		DUSA Pharma		USA, 1999		Actinic keratosis, esophageal dysplasia		Topical cream	417 nm
Laserphyrin®	Talaporfin/Mono-L-aspartyl chlorin e6		Meiji Seika Pharma		Japan, 2004		Lung cancer		Saline injection	664 nm
Photosens®	Sulfonated aluminum phthalocyanine		NIOPIK		Russia, 2001		Various cancers		NaCl injection	675 nm

While advances in fiber-optic based laser technologies can provide novel avenues to deliver the drug-activating light interstitially and intra-tumorally, the nanomedicine approach can provide an effective way to resolve the challenges associated with intravenous administration and selective delivery of the PS to deep tissue tumors. Several recent reviews have highlighted the various nanoparticle designs that are being investigated for the encapsulation and controlled release of photosensitizers [21–25]. Some other review articles have summarized the progress of PDT as such, in the clinical arena of cancer treatment [26–28]. However, a state-of-the-art review of in vitro and in vivo PDT studies using nanovehicle-based photosensitizer delivery, along with a critical discussion of design parameters for efficient nanoformulation of photosensitizers, perspectives and challenges regarding in vivo drug distribution and tumor-selective light delivery is not currently available. Therefore, we have attempted to provide this comprehensive information, to aid future research in photodynamic nanomedicine.

2. In vitro and in vivo investigations in nanoparticle-based PDT

2.1. In vitro studies in photodynamic nanomedicine

Nanoparticle-mediated PDT of cancer has been studied using a wide variety of photosensitizers and nanovehicles. The majority of published reports in this area focus on in vitro studies, most likely due to several in vivo challenges such as ensuring site-selectivity of drug delivery, difficulties in light delivery and variabilities in oxygen levels due to tumor hypoxia. These challenges and ongoing research approaches to resolve them are discussed later in this article after reviewing the in vivo studies. Table 2 provides a comprehensive list of reported in vitro studies, including the various photosensitizers, the corresponding nanovehicles used for their formulation and the corresponding cancer cell lines that these formulations have been tested on, along with appropriate references. As evident from Table 2, most photosensitizers tested fall under a few main categories, namely, porphyrins, chlorins and phthalocyanines. In addition, some other dyes like hypericin, hypocrellin, indocyanine and methylene blue have been studied as photosensitizers. In the following sections we describe the various in vitro nanomedicine studies that have been carried out under these categories of photosensitizers.

Table 2. In vitro PDT studies reported on nanoparticle-based photosensitizer formulations.

Photosensitizer	NP Vehicle	Targeting Motif	Cell Type	Ref.
Porphyrin derivatives (e.g. hematoporphyrin, porfimer sodium etc.)	Human serum albumin nanoparticles	N/A (Passive)	Jurkat (human T-cell leukemia)	[29–31]
	Metal oxide nanopartices	N/A (Passive)	A549 (human lung) NIH:OVCAR-3 (human ovarian)	[32, 33]
		AS1411 aptamer	MCF-7 (human breast)	[34]
	Magnetic chitosan nanoparticles	N/A (Passive)	SW480 (human colon)	[35]
	Silica nanoparticles	N/A (Passive)	SW480 (human colon) HCT 116 (human colon) HT-29 (human colon) MCF-7 (human breast) MDA-MB-231(human breast) A431 (human epidermoid) LLBC37 (lymphoblastoid) HeLa (human cervical)	[36–38]
		Mannose	MDA-MB-231 (human breast)	[39, 40]
	Virus-like particles	Glycan-targeting of CD22	CHO (Chinese hamster ovary)	[54]
	Polymeric nanoparticles	N/A (Passive)	SW480 (human colon) H2009 (human lung) HeLa (human cervical) HepG2 (human liver) C6 (rat glioma)	[41–43, 45, 50]
		Folic acid	HeLa (human cervical)	[44]
		F3 peptide	MDA-MB-435 (human breast) F98 (rat glioma) MCF-7 (human breast) 9L(rat gliosarcoma)	[46–49]
	Liposomes	N/A (Passive)	HCT-116 (human colorectal) DU 145 (human prostate)	[93]
	Fluorescence organic nanoparticles	N/A (Passive)	CL1-0 (human lung) HeLa (human cervical)	[94]
	Marine atelocollagen nanocapsules	N/A (Passive)	HeLa (human cervical)	[95]
	Gold nanoparticles	N/A (Passive)	Jurkat (human T-cell leukemia) K562 (human leukemia) MDA-MB-231 (human breast)	[51–53]
	Chemiluminescence nanoparticles	N/A (Passive)	C6 (rat glioma) LoVo (human colon)	[96]
Chlorin derivatives(chlorin e6, mTHPC, pheophorbide etc)	Chitosan nanoparticles	N/A (Passive)	MDA-MB-231 (human breast) HeLa (human cervical) SCC-7 (murine squamous cell carcinoma)	[97, 98]
	Human serum albumin nanoparticles	N/A (Passive)	HeLa (human cervical)	[99]
		N/A (Passive)	Jurkat (human T-cell leukemia)	[29–31]
	NaYF4:Yb,Er/NaGdF4 nanoparticles	N/A (Passive)	U87MG (human glioma)	[64]
		N/A (Passive)	HeLa (human cervical) 4T1 (murine breast)	[65]
	Iron oxide nanoparticles	N/A (Passive)	MGC-803 (human gastric)	[63]
	Mesoporous silica nanoparticles and Hyaluronic acid nanoparticles	Hyaluronic acid targeting of CD44	HT-29 (human colon) HCT-116 (human colorectal)	[61, 100]
	Polymeric nanoparticles	N/A (Passive)	EMT-6 (murine breast) Colon-26 (murine colon) C6 (rat glioma)	[101–103]
		Folate	HeLa (human cervical)	[62]
		Neuropilin-I targeting peptide	MDA-MB-231 (human breast)	[60]
	Silica nanoparticles	N/A (Passive)	RIF-1 (murine fibrosarcoma)	[104, 105]
	Liposomes	Cetuximab (C225)Ki-67 antibody	OVCAR-5 (human ovarian) A431 (human epidermoid)	[106, 107]
Phthalocyanine derivatives(zinc (II) phthalocyanine, silicon phthalocyanine 4 etc.)	NaGdF4:Yb,Er/NaGdF4 nanoparticles	N/A (Passive)	MEAR (murine hepatoma) HeLa (human cervical)	[77, 78]
			MB49 (murine bladder)	[55]
	Polymeric nanoparticles	GE11 peptide	MCF-7 (human breast) A431 (human epidermoid)	[56, 69, 70]
	Gold nanoparticles	EGF	HeLa (human cervical) 9L (rat glioma)	[71, 72]
		T antigen-specific lectin	HT-29 (human colon)	[74]
		Anti-HER2 antibody	SK-BR-3 (human breast) MDA-MB-231 (human breast)	[75]
		N/A (Passive)	MCF-7 (human breast)	[76]
	Mesoporous silica nanoparticles	N/A (Passive)	H22 (murine hepatoma)	[108]
	PLGA nanoparticles	N/A (Passive)	P388-D1 (murine lymphoma)	[68]
	Dendrimers	N/A (Passive)	A549 (human lung)	[109]
	Liposomes	N/A (Passive)	MCF-7 (human breast)	[76]
	LDL nanoparticles	N/A (Passive)	HepG2 (human liver)	[110]
Cyanine IR-768	Poly(n-butyl cyanoacrylate) nanocapsules	N/A (Passive)	MCF-7 (human breast)	[79]
Hypericin	PLA and PLGA nanoparticles	N/A (Passive)	NuTu-19 (rat ovarian)	[59]
Hypocrellin A	Silica nanoparticles	N/A (Passive)	HeLa (human cervical)	[85]
Indocyanine green	Poly(propargyl acrylate)	N/A (Passive)	HepG2 (human liver) P388-D1 (murine lymphoma)	[80, 81]
	ORMOSIL	Anti-HER-2 antibody	LNCaP (human prostate)	[82]
	Gold nanoparticles	Anti-EGFR antibody	A549 (human lung)	[83]
	Calcium phosphosilicate	Anti-CD117 antibody (anti-SCFR)	32D-p210-GFP (murine CML) Human patient AML cells	[84]
C60 and C70 (fullerene cages)	Viral nanoparticles	N/A (Passive)	PC-3 (human prostate)	[86]
	Cyclodextrin nanoparticles	N/A (Passive)	HeLa (human cervical)	[87]
	Fullerene nanocages	N/A (Passive)	HeLa (human cervical)	[88]
Methylene blue	Apoferritin nanocages	N/A (Passive)	MCF-7 (human breast)	[89]
	Dioctyl sodium sulfosuccinate (aerosol OT; AOT) alginate nanoparticles	N/A (Passive)	MCF-7 (human breast) and 4T1 (murine metastatic breast) NCI/ADR-RES (human ovarian)	[57, 92]
	Phosphonate-terminated silica nanoparticles	N/A (Passive)	HeLa (human cervical)	[91]
	Polyacrylamide, sol-gel silica and ORMOSIL nanoparticles	N/A (Passive)	C6 (rat glioma)	[58, 90]

Abbreviations

AML: Acute myeloid leukemia

CML: Chronic myeloid leukemia

EGF: Epidermal Growth Factor

LDL: low density lipoprotein

ORMOSIL: Organically modified silica

SCFR: Stem Cell Growth Factor Receptor

2.1.1. Porphyrin-based nanoformulations

Porphyrin derivatives are a major class of photosensitizer. Porfimer sodium or Photofrin is a member of this class and has been clinically approved for PDT of several pre-cancerous lesions and malignancies in the USA. The current clinical formulation does not involve a nanoparticle vehicle; however several promising pre-clinical studies have been carried out with nanoformulations of porphyrin photosensitizers. For example, Chen et al have used human serum albumin nanoparticles for delivery of 5,10,15,20-tetrakis(m-hydroxyphenyl)porphyrine (mTHPP; a porphyrin derivative) and pheophorbides (chlorin derivatives) to leukemia cells [29–31]. It was found that the nanoparticles were taken up through lysosomal mechanisms and caused ∼50% cell death due to apoptosis [31]. Porphyrin derivatives such as Photofrin and protoporphyrin IX have been formulated in several different nanoparticle systems such as metal oxide, chitosan, polymeric, silica and gold nanoparticles (examples shown in Figures 3A and 3B) [32–53]. While majority of these formulations have envisioned the EPR mechanism to be utilized to passively target solid tumors, some have utilized receptor-mediated active targeting to enhance cell-selective delivery. For example, Yin et al conjugated the AS1411 aptamer to multifunctional iron oxide nanoparticles packaging 5,10,15,20-tetrakis(1-methylpyridinium-4-yl) porphyrin (TMPyP4) for multimodal imaging and subsequent PDT of breast cancer [34]. Other examples where porphyrin PS have been packaged in nanoparticles capable of active targeting of cancer cells include mannose receptor-targeted silica nanoparticles, CD22-targeted viral nanoparticles and folate receptor-targeted polymeric nanoparticles [39, 40, 44, 54].

Figure 3.

Representative images showing the feasibility of packaging photosensitizers in nanoparticle vehicles for delivery to cancer cells, in vitro; (A) percentage of breast cancer cell death caused by PDT with protoporphyrin IX loaded gold nanoparticles of different sizes compared to free drug; (B) intracellular uptake of porphyrin-derivative loaded silica nanoparticles after 4 hours incubation (ii) compared to cells alone (i); (C) F3 peptide-targeted polyacrylamide nanoparticles (i) loaded with 2-devinyl-2-(1- hexyloxyethyl) pyropheophorbide (HPPH), a chlorin-derivative, are taken up into 9L gliosarcoma cells faster than nontargeted analogs (ii), as corroborated by the quantitative fluorescence data (iii); (D) confocal microscopy of bladder carcinoma cells incubated with zinc (II) phthalocyanine loaded silica-coated upconversion nanoparticles, apoptotic cells are indicated by arrows; (E) increased percentage of cell death due to apoptosis and necrosis in breast cancer cells when treated with polymeric micelle-encapsulated silicon phthalocyanine photosensitizer Pc 4; (F) dose-response curve in breast cancer cells shows enhanced PDT efficacy when treated with nanoparticle encapsulated methylene blue compared to free drug; (G) cell death (red) following 2 hrs of incubation in glioma cells with nanoparticle-encapsulated methylene blue and irradiation compared to nonirradiated cells (green) and (H) decrease in ovarian cancer cell viability following treatment with free hypericin (circles) compared to nanoparticle-encapsulated hypericin (squares) at various drug dosages.

Adapted with permission from [36, 49, 53, 55–59]

2.1.2. Chlorin-based nanoformulations

Chlorins form another major class of photosensitizers with a chemical structure similar to chlorophyll. The most popular drug in this class is chlorin e6, which has been studied in a variety of nanovehicles including chitosan, human serum albumin, silica, iron oxide, hyaluronic acid and various polymeric nanoparticles. Similar to the porphyrin-based nanoformulations, these chlorin-loaded nanoparticles were largely designed to utilize the EPR-based passive targeting mechanisms in breast, cervical, colon and brain cancers. However, a few approaches have also utilized active receptor-mediated targeting. For example, Benachour et al used a neuropilin-I targeting peptide on silica nanoparticles loaded with 5-(4-carboxyphenyl)-10,15,20-triphenylchlorin (TPC) to target tumorangiogenic vessels to achieve close to 100% cell death in vitro [60]. Yoon et al developed CD44 tumor-targeting hyaluronic acid nanoparticles (HANPs) as the carrier for chlorin e6 to achieve nearly 100% cell death in two different human colon cancer cell lines [61]. Similarly, folate receptor targeting was utilized by Li et al on pheophorbide A loaded heparin nanoparticles for delivery to cervical cancer [62]. More recent research with chlorin-based nanoformulations involves nanoparticles designed for combining imaging and PDT. For example, Huang et al have reported on chlorin e6 conjugated magnetic nanoparticles for concomitant MR imaging and PDT of gastric cancer [63]. Similarly, Wang et al developed tumor homing F3 peptide functionalized polyacrylamide nanoparticles loaded with 2-devinyl-2-(1- hexyloxyethyl) pyropheophorbide (HPPH) for the “see and treat” strategy through concurrent fluorescence imaging and PDT of gliosarcoma and breast carcinomas (example shown in Figure 3C) [49]. Another recent nanovehicle formulation of chlorins involve a unique platform called ‘upconversion nanoparticles’, that has been studied by a few groups for PDT of glioma, cervical and breast cancers [64, 65]. Upconversion is a process where sequential absorption of multiple photons can lead to emission of light at wavelengths shorter than the excitation wavelength (usually excitation in the infra-red range to render emission in the visible range) [66]. This emitted light can in turn activate the photosensitizer if upconversion elements (e.g. rare earth metals like Er³⁺) are formulated along with photosensitizers within the same vehicle. Hence this approach is potentially promising towards resolving challenges of ensuring light delivery in vivo.

2.1.3. Phthalocyanine-based nanoformulations

Phthalocyanine based photosensitizers, such as silicon phthalocyanine 4 (Pc 4) and zinc (II) phthalocyanine have been studied in several nanovehicles such as micelles, gold nanoparticles and silica nanoparticles. Phthalocyanine derivatives have also been packaged in mesoporous silica nanoparticles, PLGA nanoparticles, dendrimers, liposomes and LDL nanoparticles, without any receptor-targeted ligand decorations and tested on a variety of cancer cells in vitro. For example, Ricci-Junior et al have studied the delivery of zinc (II) phthalocyanine to murine lymophama cells via nontargeted PLGA nanoparticles to yield 60% cell death in vitro [67, 68]. Due to the lack of active targeting abilities and limited tumor selectivity, many of these non-targeted nanoparticle formulations have not progressed into efficient in vivo studies. In our research we have packaged Pc 4 in passively targeted polymeric nanoparticles for treatment of human breast and epidermoid carcinomas [56]. We later refined the nanoformulation with EGFR-specific peptides for active-targeting of head and neck carcinomas (Figure 3E) [69, 70]. Due to the high PDT-potency of Pc 4 and the superior cell-targeting capability of the nanoparticle platform, we were able to achieve high intracellular uptake and close to 100% cell death in vitro with drug dosages on the nanomolar scale [56, 69, 70]. Pc 4 has also been used in EGF-targeted gold nanoparticles by Cheng et al to target human cervical and rat glioma. However, it was found that to achieve effective levels of PDT in vitro, much higher doses of Pc 4 was required. This is possibly due to the fact that the drug was adsorbed on the nanoparticle surface where it is exposed to the plasma environment and prone to desorptive loss, rather than packaged within the nanoparticle [71–73]. Gold nanoparticles have also been used to package other phthalocyanine photosensitizers, and have been decorated with targeting moieties such as T-antigen specific lectins and anti-HER2 antibodies for selective delivery to colon and breast cancer [74–76]. Upconverting nanoparticles have been used for passive targeting of murine hepatoma and bladder cancer as well as human cervical cancer using PS such as zinc (II) phthalocyanine (e.g. Figure 3D) [55, 77, 78]. Zhao et al were able to achieve only 40% in vitro cell death on murine liver cancer cells but were able to successfully use the lanthanide-doped upconversion nanoparticles as a successful T₁ and T₂ contrast agent [77]. Qiao used similar nanoparticles but varied drug concentration, irradiation power and duration to yield up to 80% cell death along with successful imaging. However, it is unclear how much toxicity is due to the nanoparticle vehicle rather than the PDT induced effect. For example, Guo et al showed 50% cell death was caused by the upconversion nanoparticles alone [55]. These toxicity issues could prevent the nanoparticles from being used in vivo, and therefore more studies with these particles are needed to establish safety.

2.1.4. Other miscellaneous photosensitizers

Besides the above-described major classes of photosensitizers, several other photosensitizer molecules have also been investigated in nanoparticle-based formulations. Among these, cyanine IR-768 and indocyanine green (ICG) are cyanine compounds that can be activated by a higher wavelength (near infra-red, NIR) than most other photosensitizers, thus allowing for higher tissue penetration of the activating light. Pietkiewicz et al used oilcored poly(n-butyl cyanoacrylate) nanocapsules to deliver cyanine IR-768 to doxorubicin-sensitive and –resistant human breast cancer cell lines. They were able to show almost 100% cell death in vitro in both cell lines which shows promise towards using PDT in chemotherapy resistant cancers [79]. One limitation of the NIR or IRactivable photosensitizers is that due to their higher wavelength of activation, the energy transferred to the photosensitizer is lower resulting in a lower singlet oxygen yield. Despite this, several groups have studied ICG loaded nanoparticles for treatment of deep tissue cancers [80–84]. For example, Barth et al developed ICGloaded calcium phophosilicate nanoparticles conjugated with anti-CD117 antibody for active-receptor mediated targeting of leukemia stem cells. They were able to achieve 50% cell death in vitro before moving on to in vivo work [84]. Hypericin is a natural extract of St. John's wort previously used in cancer detection. It has recently gained popularity in PDT research due to its ability to cause photosensitivity when ingested past a threshold concentration. Like many photosensitizers, hypericin is highly hydrophobic. Zeisser-Labouebe et al have demonstrated formulation of hypericin in poly(lactic acid) (PLA) and poly(lactic-co-glycolic acid) (PLGA) nanoparticles for passive delivery to ovarian cancer (e.g. Figure 3H) [59]. In their studies it was found that they were able to use lower dosages to achieve 100% PDT-induced cell death in vitro when drug was delivered via nanoformulation compared to free drug [59]. Zhou et al experimented with the concept of heavy atom addition to photosensitizer-loaded silica nanoparticles to increase singlet oxygen production for enhanced PDT of cervical cancer. They used another less studied photosensitizer, hypocrellin A, and found that they could get up 90% cell death in vitro [85]. Fullerene cages such as C₆₀ are a relatively new class of photosensitizers known for their carbon-only closed-cage polyhedron shape. Due to their plethora of pi-bond electrons, they can successfully emit fluorescence and participate in the cascade of energy transfer events that leads to the creation of singlet oxygen. They have been conjugated to the surface of viral nanoparticles, cyclodextrin nanoparticles and other fullerene nanocages and have been studied for the treatment of cervical and prostate cancer [86–88]. Methylene blue (MB) is another biological dye that has gained recent popularity as a photosensitizer due to its strong absorption band at 630 nm. It is often used as a model photosensitizer because its absorption wavelength is similar to Photofrin. A variety of nanoparticles have been studied with MB including apoferritin nanocages, polyacrylamide nanoparticles and silica nanoparticles (Figure 3G) [58, 89–91]. Khdair et al have developed aerosol alginate nanoparticles for simultaneous delivery of doxorubicin and MB for a chemotherapy-PDT combination treatment strategy (e.g. Figure 3F). Though lacking any active targeting motif, they were able to successfully enhance cytotoxicity in drug-resistant ovarian cancer cells [57, 92].

2.2. In vivo studies in photodynamic nanomedicine

As stated previously, majority of research reported regarding in nanoparticle-formulation based PDT involves in vitro studies, probably because many of the in vivo challenges regarding consistent levels of drug delivery, siteselective light delivery and sufficient oxygen levels in tumors have not been completely resolved yet. Nevertheless, several preclinical studies have been carried out using appropriate in vivo animal models, in subcutaneous and orthotopic xenografts, showing promising results.

Porphyrin derivatives, particularly first generation photosensitizers like porfimer sodium (Photofrin), have been studied in various murine models with different nanoparticles. Hu et al developed meso-tetra (carboxyphenyl) porphyrin (TCPP) -loaded PLGA nanoparticles and TCPP-nanoparticles to treat human colon cancer xenografts in mice [41]. It was found that mice treated with TCPP-nanoparticles and the TCPP-loaded PLGA nanoparticles demonstrated significant tumor regression over time though the tumors did eventually grow back to their original volume or higher after 25 days [41]. In a similar study, Reddy et al used F3 peptide-targeted iron oxide nanoparticles loaded with Photofrin to increase the median survival of mice with orthotopic rat gliomas from 7 days (with saline treatment) to 33 days [46]. Some groups have attempted to develop multifunctional porphyrin derivative loaded nanoparticles for dual imaging and PDT. For example, Sun et al used magnetic chitosan nanoparticles conjugated to 2,7,12,18-tetramethyl-3,8-di-(1-propoxyethyl)- 13,17-bis-(3-hydroxypropyl) porphyrin (PHPP) to treat mice infected with human colon cancer xenografts [35]. Following magnetically induced targeting, they were able to induce significant reduction in tumor size compared with the nontargeted tumors (Figure 4A) [35].

Figure 4.

Representative images of in vivo studies of nanoformulation-mediated PDT; (A) T2 weighted image of porphyrin derivative loaded magnetic chitosan NPs showing reduction in magnetically targeted tumor (left) compared to nontargeted tumor (right); (B) Tumor volume after treatment with chlorin-conjugated iron oxide NPs showing tumor stagnation in the group receiving magnetically localized NPs followed by subsequent PDT; (C) Tumor targeting ability of chlorin-loaded NPs compared to free drug as seen by fluorescence imaging of the tumor hotspot and (D) tumor growth suppression caused by phthalocyanine-loaded NPs in hepatoma xenografts.

Adapted with permission from [35, 63, 99, 108].

Chlorin-based PDT nanoformulations have also progressed into in vivo studies. Chlorin e6 (Ce6) has been loaded inside human serum albumin (HSA), glycol chitosan (HGC), iron oxide, hyaluronic acid (HA) and upconversion nanoparticles for in vivo PDT investigations [61, 63, 64, 98, 99]. Several research groups have developed Ce6-HSA nanoparticles, Ce6-chitosan nanoparticles and Ce6-hyaluronic acid nanoparticles for the study of colon cancer. The nanoparticle formulations exhibit significant growth inhibition of colon cancer xenografts in mouse models compared to treatment with free Ce6 (Figure 4C) [61, 98, 99]. Another commonly used chlorin derivative is 2-[1-hexyloxyethyl]-2-devinyl pyropheophorbide-a (HPPH). HPPH-loaded polyacrylamide nanoparticles (PAA) developed by Wang et al and Gupta et al have been studied in multiple murine models [102, 111]. Wang et al demonstrated that the rate of growth inhibition in a murine colon cancer xenograft model was dependent on both the HPPH dosage and light fluence [102]. Using the same tumor model, Gupta et al showed that formulations of HPPH-loaded PAA nanoparticles significantly inhibited the growth of the tumors compared to saline treatment [111]. Similar to the work discussed previously with porphyrin derivatives, iron oxide nanoparticles have been studied for concomitant imaging and PDT. Huang et al developed a single platform consisting of Ce6 conjugated iron oxide nanoparticles for magnetically guided drug delivery and subsequent PDT. It was found that tumor size stayed relatively stagnant over a twenty eight day period following nanoparticle administration and PDT (Figure 4B) [63]. A newer area of research involves upconversion nanoparticles which convert lower level excitation photons to higher energy NIR photons for enhanced PDT. Park et al used Ce6-conjugated upconversion nanoparticles (UCNPs) to treat human glioma xenografts in mice [64]. IV injection of UCNPs resulted in significant tumor regression compared to controls.

Phthalocyanine photosensitizers have also been studied in several animal models. For example, Tu et al developed zinc-phthalocyanine (ZnPc) loaded mesoporous silica nanoparticles (MSNs) to treat human hepatoma xenografts in mice (Figure 4D) [108]. They were able to increase the lifetime of the treated animals from 16 days to 40 days. Cheng et al used silicon phthalocyanine 4 (Pc 4)-conjugated gold nanoparticles to treat orthotopic human gliomas in a murine model and found that they were able to induce tumor necrosis [73]. Using a different approach, Nishiyama et al loaded phthalocyanine dendrimers inside of polymeric micelles and used them to treat human lung cancer xenografts in mice [109]. Thirty days after treatment, tumors treated with nanoparticles had only grown to 10 times their starting size, while tumors treated with a control grew to 25 times their starting size. As stated previously, we have developed EGFR-targeted Pc 4-loaded polymeric micelles and have demonstrated their capability for tumor-targeted delivery of Pc 4 for enhanced PDT. Our ongoing studies are focused on evaluating this Pc 4 formulation for PDT of head and neck carcinomas in mouse models.

Among other photosensitizer nanoformulations, indocyanine green (ICG)-loaded calcium phosphosilicate nanoparticles developed by Barth et al were tested in a mouse model to determine treatment efficacy against orthotopic murine chronic myeloid leukemia [84]. This formulation was found to both decrease the size of the tumors and increase the life span of the mice tested. He et al found methylene blue-loaded phosponateterminated silica nanoparticles led to tumor necrosis and shrinkage in mice upon light activation [91].

3. Current state-of-the-art and challenges for nanoparticle-based PDT

3.1. Nanoparticle design parameters

As evident from Table 2 and Table 3, nanoparticles provide an effective way to package a variety of photosensitizer molecules and deliver them via passive and active targeting pathways to tumor cells. Table 5 lists the various nanoparticle systems (along with representative construct components) that have been studied for photosensitizer delivery. In this context, it is also important to recognize the critical design parameters for nanoparticulate vehicles, as listed in Table 4, that influence efficient packaging and enhanced delivery of photosensitizers to deep tissue tumors in vivo. The plasma half-life of the photosensitizer-loaded nanovehicles is dependent upon renal clearance extent and also on the rate of reticuloendothelial system (RES) uptake-based clearance (e.g. in the liver). The renal elimination can be avoided by fabricating nanovehicles that are bigger than renal filtration threshold of 15 nm in diameter [113, 114]. The RES uptake can be avoided by minimizing plasma protein adsorption (opsonization) on the nanovehicle surface. It has been shown that making nanovehicle surface hydrophilic (e.g. by grafting polyethylene glycol (PEG) brushes onto the surface) or modulating the geometric parameters (size and shape) of the vehicle, can modulate the protein adsorption and macrophagic interaction, as well as, biotransport (e.g. margination to the wall and extravasation) properties, and thereby can enhance plasma half-life [115–122]. The increased half-life can allow for increased number of passages of the PS-loaded nanovehicles past the tumor site, thereby enhancing the possibility of EPR-mediated increased accumulation within the tumor stroma, resultant increased bioavailability of the encapsulated photosensitizer drug within the tumor, and significantly enhanced PDT efficacy, while avoiding side-effects in healthy tissues. An additional level of tumor-selectivity can be achieved by modifying the PS-loaded nanovehicle surface with antibodies, antibody fragments and ligands that bind to specific receptors upregulated on the cancer cells surface and thereby facilitate receptor-mediated cellular internalization of the payload [46–49, 56, 69, 70, 123–126].

Table 3. In vivo PDT studies reported on nanoparticle-based photosensitizer formulations.

Photosensitizer	NP Vehicle	Targeting Motif	Cell Type	Ref.
Porphyrin derivatives (hematoporphyrin, porfimer sodium etc.)	PLGA nanoparticles	N/A (Passive)	SW480 (human colon cancer)	[41]
	Polyacrylamide nanoparticles	F3 peptide	9L (orthotopic rat glioma)	[46, 112]
	Magnetic chitosan nanoparticles	N/A (Passive)	SW480 (xenograft human colon)	[35]
	Silica nanoparticles	N/A (Passive)	HeLa (human colon cancer) HCT 116 (human colon cancer) A549 (human lung cancer)	[37, 38]
Chlorin derivatives (chlorin e6, mTHPC, pheophorbide etc)	Human serum albumin nanoparticles	N/A (Passive)	HT-29 (human colon cancer)	[99]
	Chitosan nanoparticles	N/A (Passive)	HT-29 (human colon cancer)	[98]
	Iron oxide nanoparticles	N/A (Passive)	MGC-803 (human gastric cancer)	[63]
	NaYF4:Yb,Er/NaGdF4 nanoparticles	N/A (Passive)	U87MG (human glioma)	[64]
	Hyaluronic acid nanoparticles	Hyaluronic acid targeted to CD44	HT-29 (human colon cancer)	[61]
	PLGA nanoparticles	N/A (Passive)	Rhabdomyosarcoma (orthotopic)	[101]
	Polyacrylamide nanoparticles	N/A (Passive)	Colon-26 (murine colon cancer)	[102, 111]
Phthalocyanine derivatives (zinc (II) phthalocyanine, silicon phthalocyanine 4 etc.)	Mesoporous silica nanoparticles	N/A (Passive)	H22 (human hepatoma)	[108]
	Dendrimers	N/A (Passive)	A549 (human lung cancer)	[109]
	Gold nanoparticles	EGF	Gli36 (orthotopic human glioma)	[73]
Indocyanine green	Calcium phosphosilicate	Anti-CD117 antibody	32D-p210-GFP (orthotopic murine chronic myeloid leukemia)	[84]
Methylene blue	Phosphonate-terminated silica nanoparticles	N/A (Passive)	HeLa (human cervical cancer)	[91]

Table 5. Examples of nanoparticles systems along with components currently being studied for photosensitizer formulations.

Type of Nanoparticle	Example of Nanoparticle Components	Reference
		Master et al [56, 70]
		Zhao et al [44]
		Sibata et al [136]
		Ichikawa et al [137]
		Nawalany et al [93]
		Ricci-Junior et al [68] Hu et al [41] Tang et al [90] Tang et al [58]
		Wang et al [102] Gao et al [50] Reddy et al [46] Hah et al [47] Wang et al [49]
		Chen et al [36] Simon et al [37] Brevet et al [39]
	$3Ag(s)+\mathit{\text{AuC}}{l}_4^{-}(aq)\to Au(s)+3\mathit{\text{AgCl}}(s)+C{l}^{-}(aq)$	Cheng et al [71] Khaing et al [53] Gamaleia et al [51] Obaid et al [74]
		Zhao et al [77] Park et al [64] Qiao et al [78] Guo et al [55]
		Chen et al [96]

Table 4. Design Parameters for Nanoparticle vehicles for Photosensitizer Delivery to Solid Tumors.

Criteria	Rationale
Size and Shape	Facilitate vascular margination and extravasation into tumor
Plasma Half-Life	Increase circulation time for tumor-selective passive accumulation
Drug Encapsulation Efficiency	Limit partitioning of drug in plasma and reduce nonspecific uptake Ensure that drug is not in a highly aggregated form
Surface Charge (zeta potential)	Minimize aggregation of nanoparticle vehicles in suspension
Controlled Release Kinetics	Maintain desirable therapeutic action specifically at the target site
Active Cell-Targeting Ability	Facilitate receptor- or membrane-mediated intracellular uptake
Ease of Formulation and Scale Up	Facilitate translation into large-scale clinical use
Possible Imaging Component	Enable image-guided therapy and image-assisted evaluation

3.2. Photosensitizer encapsulation considerations

Regarding encapsulation of photosensitizers in nanovehicles, it is important to note that unlike loading of chemotherapeutic drugs where higher encapsulation correlates to higher cytotoxic dose, for effective PDT the ideal dose may not be necessarily the maximum amount of photosensitizer that can be loaded. If the photosensitizer molecules aggregate extensively within the nanovehicle, upon photoirradiation they can self-quench the reactive oxygen species and thus reduce the photodynamic efficacy. Hence formulation of photosensitizers within nanovehicles should ensure sufficient loading that is photodynamicaly active but not aggregating and self-quenching [69]. Also, photosensitizer loading within nanovehicles should also ensure that while in circulation the drug molecules do not leak out or partition into the plasma. For this purpose, nanovehicle designs where the drug is physically trapped within the vehicle core (e.g. in micelles) or chemically conjugated to the vehicle (e.g. in certain polymer or silica based nanoparticles) may have advantage over systems where the drug molecules are adsorbed or conjugated onto the vehicle surface (e.g. in certain gold based nanoparticles).

Vehicle surface-adsorbed drug molecules can potentially desorb either by physical partitioning or chemical/enzymatic cleavage in the plasma environment in vivo, leading to premature loss of the drug even before the vehicle can reach its target site. It is interesting to note here that for chemotherapeutic drugs encapsulated or conjugated within a vehicle core, therapeutic effects may get affected due to challenges in release kinetics of the drug from the vehicle, but for PDT actual release of the photosensitizers from the vehicle may not be necessary, since the actual cytotoxic component is the reactive oxygen species. Oxygen can easily diffuse within the vehicle core, interact with the photoactivated PS, and get converted into reactive oxygen species, which can then diffuse out of the vehicle to cause cytotoxic damage. Therefore, nanoparticle vehicles that enable encapsulation of photosensitizers within the vehicle core, can potentially avoid premature drug loss and inactivation of the drug by plasma components, while allowing drug accumulation preferentially within the tumor for subsequent PDT. The versatile nature of nanoparticulate delivery vehicles also allows for multifunctional modifications that can incorporate imaging agents for diagnostic functionalities, along with photosensitizers for PDT [127–130]. For example, magnetic resonance imaging (MRI) agents like Gadolinium (T1 agent) or iron oxide (T2 agent) can be incorporated in photosensitizer-loaded nanovehicles to allow MRI guidance of tumor volume to be photoirradiated or MRI-assisted evaluation of PDT effect [131, 132]. Following similar rationale, optical probes (e.g. fluorescent molecules) can be incorporated in the vehicle for optical imaging [133–135].

3.3. Tumor characteristics

While design of passively and actively targeted nanoparticle delivery systems with appropriate size, shape, charge, surface-modification and drug-loading can significantly improve the tumor-selective delivery and release of the photosensitizers, certain challenges regarding the other two components of PDT, namely, oxygen and photoactivating light, need to be resolved in parallel to improve the clinical repertoire of PDT. The presence of sufficient molecular oxygen throughout the tumor volume is crucial for effective PDT. The mechanism of PDT requires the creation of reactive oxygen species (e.g. singlet oxygen) from tissue oxygen via interaction with the photoactivated drug. However, the oxygen levels within a tumor tissue can be highly heterogenous due to dysregulated vascular distribution, leading to severely hypoxic regions [138–140]. Tumor cells in these hypoxic regions are resistant to PDT due to lack of sufficient molecular oxygen. Also, the heterogenous vascular distribution can affect the vascularly-mediated delivery of PS-loaded nanovehicles uniformly throughout the tumor volume. Furthermore, tumor cells in hypoxic regions can also become more aggressive due to various signaling mechanisms, for example induced by hypoxia inducible transcription factor 1 (HIF 1) [140]. Hence strategies are required to restore normal oxygen environment (normoxia) in such tumor regions, as well as, to achieve uniform tissue penetration of the PS-loaded nanovehicles, to make the tumors amenable to PDT and other therapies. Various therapeutic and interventional strategies, for example, molecular inhibition of HIF-1 pathways, hyperbaric oxygen therapy, ozone therapy, etc. are currently under research to restore normoxia in oxygen-deprived tumor regions [141, 142]. In parallel, interesting research is going on in the area of nanoparticle design where particle morphology, charge and tumor environment-selective degradation capabilities are being utilized to enhance the tumor penetration and uniform distribution of payload [121, 143–150]. Integration of these strategies can significantly improve the efficacy of PDT in various tumors [151, 152].

3.4. Light delivery technologies

Challenges also exist regarding the delivery of appropriate light dose throughout the entire tumor volume, partly due to the limited tissue penetration parameters of the photoactivating wavelengths and also due to the difficulty in dosimetric correlations between the extent of photoirradiation and corresponding PDT effect [26]. As mentioned previously, light with wavelengths is the deep red to near infra-red (NIR) regions have higher tissue penetration capability. Therefore photosensitizers that can be activated by wavelengths ≥ 650 nm may have advantage in deep tissue tumor applications in vivo. Regarding technological components of light delivery, significant advancements have been made in the area of laser-based illuminators and light diffusers using fiberoptic arrangements and various geometries [153–165]. Table 6 shows selected examples of lasers used for clinical PDT. Lasers have been the preferred light source due to their high power and wavelength range. Additionally, it is relatively easy to couple them to optical fibers for endoscopic use. One of the more widely used systems is the argon laser or argon laser pumped dye laser system. The laser dyes (rhodamine derivatives) can be chosen to alter the wavelength range to match the absorption of the photosensitizer being used. Argon lasers do have a disadvantage of requiring high levels of adjustment and technical support [166]. Argon-pumped dye lasers are often coupled to optical fibers for treatment of lung cancer, oral cancers and Barrett's esophageous [167–170]. These lasers are indicated for use with fiber optics due to their very small beam cross-section. At the same time, this small beam cross-section complicates their use with larger legions (e.g. skin cancers). A diffuser is often put at the end of the fiber to yield uniform distribution of photoirradiation. Semiconductor diode lasers are another type of continuous wave laser though they can also be used in pulsed mode. They have compact equipment enclosures, are generally cheaper require less maintenance and are air-cooled which makes them portable and easier to use. They are usually coupled to optical fibers for endoscopic use but the fiber output can also be expanded for use on larger legions such as the skin. Metal vapor lasers and metal vapor laser pumped dye lasers are a variation on the argon laser systems. The gold and copper vapor lasers are the most popular in this group. They have an advantage of enhanced portability over the argon lasers because they do not require specialized cooling equipment or power supplies. However, at the same time, they require long warm-up and cool-down times. Additionally, they are often considerably more expensive. Metal vapor laser systems produced pulsed outputs as opposed to continuous wave outputs which causes changes in the light delivery dynamics though this has not been shown to cause a difference in PDT response [162]. Solid state lasers are pulsed laser technology that often emit in the near infrared region. For example, the frequently used Nd:YAG laser emits at 1064 nm though frequency doubling can allow for emission at other wavelengths. The frequency-doubled emission can be coupled with pumped dye lasers to yield output in the region of photosensitizer absorption. These light delivery technologies can provide effective ways to modulate illuminator geometry and in vivo placement customized to the target tumor tissue volume. Some interesting technological advancements have also occurred in the field of nanoparticle systems that allow improved strategies for photosensitizer activation with appropriate wavelength. An interesting example is the class of ‘upconversion nanoparticles’ that were described previously in Section 2.1.2. [55, 171–173], where irradiation of tissue-penetrating infra-red wavelengths on certain rare earth elements results in their emission in the visible range which can then activate photosensitizers in their proximity. Although the PDT promise of such particle formulations has been demonstrated, the biosafety of these upconverting rare earth elements need to undergo further research to guarantee clinical translation [174].

Table 6. Examples of laser technologies currently used for PDT.

Laser Type	Wavelength Range	Fluence Rate (mW/cm2)	Method of Light Delivery
Continuous Wave Lasers
Argon laser	488 and 515 nm	500-1000	Direct or fiber optics
Argon laser pumped dye laser	500-750 nm	10-200	Direct or fiber optics
Semiconductor diode lasers	600-950 nm	≤ 700	Fiber optics
Pulsed Lasers
Metal vapor laser	UV/Vis	≤ 10	Direct or fiber optics
Metal vapor laser pumped dye laser	500-750 nm	10-500	Direct or fiber optics
Solid state laser	Usually 1064 nm but can be frequency doubled to 532, 355 and 266 nm	≤ 10	Direct or fiber optics

3.5. PDT dosimetry parameters

For efficient clinical translation of PDT technologies where nanoparticle-based tumor-targeted photosensitizer delivery is integrated with tumor-selective light delivery, it is crucial to correlate PDT parameters with the corresponding biological effect (dosimetry). According to the American Association of Physicists in Medicine (AAPM) report No. 88, for successful PDT-induced treatment (effective tumor killing), it should be ensured that the product of photoirradiation fluence rate and photoirradiation exposure time should exceed the threshold required to induce cell killing [175]. Establishing this threshold and hence the corresponding photoirradiation parameters is often a challenge, especially for deep tissue tumors where drug distribution and light distribution throughout the entire tumor volume may not be uniform. Various dosimetric techniques, for example, measurement of singlet oxygen luminescence, image-assisted evaluation of PDT-induced biological effect, measurement of fluorescence photobleaching of the photosensitizers as a surrogate parameter and measurement of total light energy absorbed by the drug per unit tissue volume, are being investigated to establish precise correlation between PDT parameters and treatment effects [176–189]. A recent interesting nanoparticle-based development in this area is the fabrication of chemoluminescent probe-incorporated nanoparticle systems that allow chemoluminescence-based monitoring of reactive oxygen species levels produced upon photoirradiation [96, 190]. Efficient integration of appropriate drug delivery mechanisms, light delivery technologies and dosimetric approaches can be envisioned to significantly enhance the translational promise of nanoparticle-based photosensitizer formulations.

Conclusion

Photodynamic Therapy (PDT) continues to gain momentum as a clinically translatable stand-alone, alternative or adjunctive treatment modality for various cancers. Successful PDT of deep tissue solid tumors will require optimization of tumor-targeted drug delivery, tumor-selective photoirradiation and dosimetric correlation of PDT-induced tumor killing effect. Nanoparticles can provide a multitude of ways to enhance the tumor-targeted photosensitizer delivery component, by enabling optimum photosensitizer encapsulation, protection from plasmainduced drug inactivation or premature drug leakage, enhanced uptake within the tumor tissue and cell, and tumor environment-specific drug release and distribution. In addition, refinement of nanoparticle designs can also allow incorporation of imaging agents, dosimetric components and light delivery components. With continued research in optimizing the drug-light-oxygen triad, clinical applications of photodynamic nanomedicine strategies customized towards various deep tissue solid tumors can be envisioned in near future.