Review pubs.acs.org/CR
Nanoparticles in Photodynamic Therapy Sasidharan Swarnalatha Lucky,†,§ Khee Chee Soo,‡ and Yong Zhang*,†,§,∥ †
NUS Graduate School for Integrative Sciences & Engineering (NGS), National University of Singapore, Singapore, Singapore 117456 ‡ Division of Medical Sciences, National Cancer Centre Singapore, Singapore, Singapore 169610 § Department of Biomedical Engineering, Faculty of Engineering, National University of Singapore, Singapore, Singapore 117576 ∥ College of Chemistry and Life Sciences, Zhejiang Normal University, Zhejiang, P. R. China 321004 Notes Biographies Acknowledgments References
1. INTRODUCTION: PHOTODYNAMIC THERAPY FOR CANCER Cancer accounted for 7.6 million deaths (around 13% of all deaths) worldwide in 2008.1 Almost 13 million cancer cases are newly diagnosed every year, and deaths are projected to rise, with an estimated 13.1 million in 2030. Although regular screening and surveillance programs as well as early intervention are the best ways to improve the outcome and survival, efforts need to be divided at finding better cancer therapeutic options that are effective, efficient, affordable, and acceptable to patients. The conventional cancer treatment options are chemotherapy, radiotherapy, and surgery, and more recently small molecule-based therapies and immunotherapy along with a combination of these strategies are being practised. However, chemotherapy is often associated with systemic sideeffects, high recurrence rate is associated with surgical resection of tumors, while radiation therapy is limited by the cumulative radiation dose. While refinement of the conventional cancer treatment modalities is important, research has also focused on developing alternate treatment modalities that are safe, potent, and cost-effective. Photodynamic therapy (PDT) is an alternative tumor-ablative and function-sparing oncologic intervention. Since its inception in early 1900s and its first modern demonstration by Dougherty et al. in 1975,2 PDT has undergone extensive investigations and has emerged as a disease site specific treatment modality. Essentially, it involves the administration of a tumor-localizing photosensitizer (PS) followed by local illumination of the tumor with light of a specific wavelength to activate the PS. The excited PS then transfers its energy to molecular oxygen, thus generating cytotoxic reactive oxygen species (ROS), such as singlet oxygen (1O2) that can oxidize key cellular macromolecules leading to tumor cell ablation.3 Hence, PDT employs 3 nontoxic components that by its own do not have any toxic effects on the biological systems, unlike chemotherapy drugs that induce systemic toxicity and ionizing light of radiation therapy that damages neighboring normal tissues. Clearly, PDT has its own merits compared to the conventional treatment methods due to
CONTENTS 1. Introduction: Photodynamic Therapy for Cancer 2. Principle of Photodynamic Therapy 2.1. Photodynamic Reaction 2.2. Mechanism of Tumor Destruction 2.2.1. Direct Tumor Cell Kill 2.2.2. Vascular Damage 2.2.3. Inflammatory and Immune Response 3. Photosensitizers 3.1. 1st Generation Photosensitizers 3.2. 2nd Generation Photosensitizers 3.3. 3rd Generation Photosensitizers 4. Current Limitations of Photodynamic Therapy 4.1. Tumor Selectivity 4.2. Formulation of Photosensitizers 4.3. Tissue Penetration Depth 5. Nanoparticles in Photodynamic Therapy 5.1. Nanoparticles As Delivery Vehicles of Photosensitizers 5.1.1. Biodegradable Nanoparticles 5.1.2. Nonbiodegradable Nanoparticles 5.2. Nanoparticles as Downconverting Photosensitizers 5.2.1. Fullerenes 5.2.2. Titanium Dioxide Nanoparticle 5.2.3. Zinc Oxide Nanoparticle 5.3. Nanoparticles as Energy Transducers 5.3.1. X-ray Activatable Nanoparticle 5.3.2. Quantum Dots 5.3.3. Two-Photon Absorbing Nanoparticles 5.3.4. Upconversion Nanoparticles 6. Conclusion and Future Perspectives Author Information Corresponding Author © 2015 American Chemical Society
2036 2036 2037 2037
1990 1991 1991 1991 1991 1993 1994 1995 1995 1995 1996 1996 1997 1997 1997 1998 1998 1998 2005 2011 2011 2012 2012 2013 2013 2014 2014 2016 2034 2036 2036
Received: August 3, 2014 Published: January 20, 2015 1990
DOI: 10.1021/cr5004198 Chem. Rev. 2015, 115, 1990−2042
Chemical Reviews
Review
Figure 1. Schematic illustration of a typical photodynamic reaction. Adapted from ref 10. 2002, Elsevier.
PS can then directly interact with a substrate, such as the cell membrane or a molecule, and transfer a proton or an electron to form a radical anion or cation, respectively, which then reacts with oxygen to produce oxygenated products such as superoxide anion radicals, hydroxyl radicals, and hydrogen peroxides (type I reaction). Alternatively, the energy of the excited PS can be directly transferred to molecular oxygen (itself a triplet in the ground state), to form 1O2 (type II reaction). The energy required for the transition of oxygen from triplet ground state to excited singlet state is 22 kcal mol−1, which corresponds to a wavelength of 1274 nm.12 Thus, a relatively small energy is only needed to produce 1O2.13 The byproducts formed as a result of the type I and type II reactions are responsible for the cell-killing and therapeutic effect in PDT. It is to be noted that both type I and type II reactions can occur simultaneously, and the ratio between these processes depends on the type of PS, as well as the concentrations of molecular oxygen and substrate present.11 However, most of the studies indicate type II reactions, hence 1O2 play a dominant role in PDT.14
its minimal invasiveness, repeatability without cumulative toxicity, excellent functional and cosmetic results, reduced long-term morbidity, and improved quality of life of the patients. Over the last four decades PDT has proven to be effective in superficial bladder cancer,4 early and obstructive lung cancer,5 Barrett’s esophagus,6 head and neck cancers,7 and skin cancer.8 It is also being used as an adjunctive therapy following surgical resection of tumor, to reduce residual tumor burden.9 Despite the widespread and rapidly growing applications, PDT has yet to gain clinical acceptance as a first-line oncological intervention due to certain limitations including lack of an ideal PS, challenges in formulating PS, choosing the right light dosimetry for a complete and effective treatment, difficulties in planning the treatment and monitoring the treatment response; which will be discussed in detail in this review. The application of nanoparticles in PDT has been a major stride forward in resolving some of the challenges associated with classic PS. Since a comprehensive review, encompassing the theory behind the design of nanoconstruct to its application in PDT, is beyond the scope of this review, we have attempted to provide an overview of the present status and prospects of such nanoparticles by highlighting its development in each phase and giving special emphasis on multifunctional theranostic agents, by taking specific illustrations from recently published articles. The examples given in this review do not mean that the pioneering contributions made by a large number of researchers are neglected. Special interest is devoted to the last section on “Upconversion Nanoparticles” (UCNs), a multifaceted tool that due to its recent accelerated progress shows great potential in augmenting the scope of PDT in the treatment of solid tumors. For detailed information on other nanoparticles, readers are requested to refer to key publications and review articles provided in each section.
2.2. Mechanism of Tumor Destruction
Following PDT, the extent of photoinduced tumor destruction depends on various factors such as the type of PS, its concentration and localization in the tumor microenvironment or its subcellular location at the time of irradiation, the time between PS administration and light irradiation (drug-light interval, DLI), light fluence rate, the total fluence, the type of tumor, and its level of oxygenation.3a,15 PDT’s downstream targets include tumor cells, tumor as well as normal microvasculature, and the inflammatory and immune host system. The ROS generated by the photodynamic reaction is the key effector causing irreversible damage to the tumor cells and microvasculature, ensuing a plethora of inflammatory and immune response; a combination of which often helps to achieve a long-term tumor control.15b 2.2.1. Direct Tumor Cell Kill. Generally, the site of photodamage is considered to coincide with the location of accumulation of the PS in the tumor cells, as 1O2 has a very short life (half-life: 0.03 to 0.18 ms),16 restricting its diffusion in biological systems to a radius of 1000 Ce6 molecules/UCN). Due to the presence of the biocompatible polymer PEG, these UCNs carrying Ce6 had better circulation half-life than free-Ce6. Particles mainly accumulated in the liver and spleen; however the amount dropped considerably 7-days post injection, reducing long-term toxic effects. Following in vivo PDT at 980 nm, there was a 6-fold decrease in tumor size among the treated group, thus clearly demonstrating its potential as PDT agent. Thereafter, our team reported therapeutic efficiency of mesoporous silica coated UCNs coloaded with two different PS, ZnPC and MC-540 (Figure 24), using mouse melanoma model in a stepwise fashion.235b First, the melanoma cells were prelabeled with PS loaded UCNs and then injected subcutaneously to develop UCN loaded tumors in mice. This was followed by assessment of PDT efficacy by injecting PS loaded UCNs intratumorally into tumor bearing mice and subsequently intravenous administration of FA conjugated- PEGylated- PS loaded UCNs was performed. This was the first time that two different PSs absorbing at two different wavelengths (660 and 540 nm) of
the emitted upconverted light was being utilized for maximum PDT efficacy. The results were quite promising in that all the three modes of UCN administration produced significant reduction in tumor volume compared to the untreated group. However, a very long irradiation time of 1−2 h (at 980 nm, power density of 415 mW cm−2), was employed for in vivo PDT. This could have been due to the inefficient energy transfer from UCNs to the PS probably due to the limited PS loading capacity ( 0.05 for not significant (**) pairs. Reproduced with permission from ref 266. 2014 American Chemical Society.
studies could have led to the observed difference in the excretion routes and clearance of the nanoparticles. But it is to be noted that, although the nanoparticles remained in the animal for over 90 days, no appreciable variations were observed in the blood biochemical, hematological or histopathological analysis. Presence of targeting ligands were found to channel the UCNs to the tumor and reduce the percentage of injected dose that accumulated in other organs such as the liver and spleen. Cui et al. administered FA conjugated UCNs intravenously, that exhibited high tumor targeting efficiencies and accumulated in the tumor as early as 4 h post administration, whereas the unconjugated UCNs became visible only after 12 h.234c The particles first accumulated in the liver, following which they were gradually excreted into the intestines by 24 h and the absence of fluorescence signals from the loaded NIR dye (ICGDer-01) at 48 h indicated complete clearance of the UCNs from the system via enterohepatic route. A significant variation in the biodistribution was observed in the presence of a targeting moiety FA at 24 h in a tumor bearing mice, where FAconjugated UCN levels were much higher in tumor but lower in organs like lung, liver, and kidney, compared to the level of unmodified UCNs. Due to the huge diversity in the nature, size,
Later, Cheng et al. conducted a thorough assessment of the pharmacokinetics, long-term biodistribution and toxicity of 30 nm sized PEG or PA modified UCNs.271 Although, the blood circulation curves of UCN-PEG and UCN-PA followed a twocompartmental model, the blood circulation half-life of UCNPEG was significantly greater than UCN-PA, which was attributed to the biocompatible PEG coating on the UCN surface and its improved stability in physiological solutions. In vivo UCL imaging showed that UCN-PA accumulated in liver within 5 min, but UCN-PEG uptake was slower and observed in liver at 30 min. Ex vivo UCL imaging revealed accumulation of UCNs in the liver, spleen, lung and bone at 1 day, but no signal was detected by day 7. However, the yttrium ion levels in these organs did not drop much over a course of 90 days, except in the lung, which showed a rapid decrease of yttrium ion concentration after 3 days. The presence of nanoparticle aggregates in the TEM of liver slices from mice 7 days post injection, indicates partial decomposition of UCNs inside the macrophage cells after RES uptake and loss of their UCL emission signals over time. Thus, the study concluded that nanoparticle residues would stay inside the mouse body for more than 90 days, which is in stark contrast to the previous reports. Perhaps the size difference of UCNs in these two 2031
DOI: 10.1021/cr5004198 Chem. Rev. 2015, 115, 1990−2042
Chemical Reviews
Review
Figure 28. (A) Schematic illustration of the synthetic procedure of UCMSNs. Gd-UCNPs were prepared by epitaxial growth NaGdF4 layer on NaYF4:Yb/Er/Tm through a typical thermal decomposition process. A dense silica layer was then coated on Gd-UCNPs by a reverse microemulsion method, designed as Gd-UCNPs@SiO2. Subsequently, a mesoporous silica shell was deposited on Gd-UCNPs@SiO2 via the template of CTAC, designed as Gd-UCNPs@SiO2@mSiO2. Finally, UCMSNs were successfully fabricated based on a “surface-protected hot water etching” strategy. (B) Transmission electron microscopic (TEM) images of (a) Gd-UCNPs (NaYF4:Yb/Er/Tm@NaGdF4), (b) Gd-UCNPs@SiO2, (c) Gd-UCNPs@ SiO2@mSiO2, and (d) UCMSNs (scale bar = 50 nm) (C) In vivo T1−MRI images of a 4T1-tumor bearing mouse after intravenous injection of UCMSNs at designated time points: (a) 0 min; (b) 5 min; (c) 10 min; (d) 15 min. (e) Comparison of MRI signal intensity of the corresponding 4T1-tumor at different time points from 0 to 15 min. (f−h) Ex vivo NIR−NIR upconversion luminescent imaging (UCL) of dissected organs after the intravenous injection of UCMSNs: 1, heart; 2, liver; 3, spleen; 4, lung; 5, kidney; 6, tumor (i−k). (D) Tumor growth and relative tumor volumes of different groups post intratumoral injection of UCMSNs-HP-Dtxl and corresponding treatment. Control groups received PBS. (E) Tumor growth and relative tumor volumes of different groups post intravenous injection of UCMSNs-HP-Dtxl and corresponding treatment. Reproduced with permission from ref 264. 2014 Elsevier.
composition, charge, surface properties, and presence or absence of additional functional moieties on the surface of
UCNs, it becomes very hard to even speculate the biodistribution and toxicity of these nanomaterials, with the 2032
DOI: 10.1021/cr5004198 Chem. Rev. 2015, 115, 1990−2042
Chemical Reviews
Review
Table 5. Summary of the Published Efforts on Biodistribution and Clearance of UCNs in Animal Models
particular organ still remain unanswered. Long-term comparative studies on the biodistribution, metabolism, degradation, and toxicity profiles of various UCNs of different size, charge, and surface functionalities are necessary to evaluate the
limited knowledge available from the literature to date. Several questions regarding the ultimate fate of these nanomaterials in the system, the possibility of their degradation into toxic byproducts, and their effects following prolonged residence in a 2033
DOI: 10.1021/cr5004198 Chem. Rev. 2015, 115, 1990−2042
Chemical Reviews
Review
Figure 29. (A) Real-time in vivo UCL imaging of athymic nude mice after intravenous injection of PA-UCNs (15 mg kg−1) at different time points. Column 3: overlays of UCL and brightfield images of mice. Column 6: overlays of UCL and brightfield images of dissected mice. (B) Biodistribution of particles in organs of mice. Error bars were based on triplet measurements. (C) H&E-stained tissue sections from mice 115 days postinjection with PA-UCNs (a, c, e, g, i, and k) and mice receiving no injection (b, d, f, h, j, and l). Tissues were harvested from heart (a, b), spleen (c, d), liver (e, f), lung (g, h), kidney (i, j) and blood smear (k, l). Reproduced with permission from ref 270. 2010 Elsevier.
lagging behind is a thorough understanding of how they behave as a collective entity and affect complex biological systems. In this regard, there is an urgent need for in-depth characterization of such multifunctional therapeutic agents to check if they cross-talk, quench, or interfere with each other’s capabilities impeding the overall outcome. In addition to that, standardized physicochemical characterization, validation, and safety protocols must be established to achieve pure and homogeneous nanoparticles that can yield precise and reproducible results. Furthermore, the fact that none of such multifunctional theranostic agents have been approved by FDA for clinical application is undeniably worrying. Again, the major bottleneck for the translation of these theranostic agents into clinical practice is the absence of standardized procedures to assess its in vivo biodistribution, detailed pharmacokinetic pharmacodynamic (PK−PD) analysis, short-term and long-term effects of the nanomaterials in vivo, as well as its clearance from the system. This was exactly the reason behind the establishment of National Characterization Laboratory by the U.S. National Cancer Institute (NCI) in 2004, which has developed standardized evaluation protocols for rigorous characterization of nanoparticle’s physicochemical properties, in vitro immunological and cytotoxic characteristics, as well as administration, distribution, metabolism, and elimination (ADME)/toxicity profiles in animal models; to facilitate the regulatory review of nanotechnology-based cancer therapeutics and fast track the clinical trial process. Today, a decade later, more than 300
usefulness of this class of nanomaterials before its further development and application in the clinics.
6. CONCLUSION AND FUTURE PERSPECTIVES PDT is undoubtedly a promising therapeutic option in the management of cancer. However, it has still not gained acceptance as a first line treatment option mainly due to the shortcomings of classical PSs. The application of nanoparticles in the field of PDT is an extremely promising avenue for future technological breakthroughs. Nanoparticles can act as carriers of hydrophobic PSs and transport high PS “payload” to the tumor site via the EPR effect. Additional surface modification and functionalization with targeting moieties could further enhance the selective accumulation of the PS loaded nanoparticles at the target site. In recent years, much effort has been channelled in the development of multifunctional theranostic agents where a number of diverse diagnostic and therapeutic functionalities have been incorporated in a single nanoplatform for a “see and treat” approach. The versatility of such agents allows delivery of drugs to the diseased/target site with enhanced spatial specificity while achieving simultaneous realtime imaging, monitoring, as well as therapeutic capabilities. While, an in-depth knowledge and understanding of nanomaterial chemistry, nanoscale manufacturing and synthesis has evolved tremendously in order to develop novel nanoplatform incorporated with sophisticated functionalities, what is now 2034
DOI: 10.1021/cr5004198 Chem. Rev. 2015, 115, 1990−2042
Chemical Reviews
Review
being carried out at the solution- and cell-based level. Solutionbased studies help to predict nanoparticle toxicity by assessing the particle size, size distribution, morphology, composition, dispersion, surface area, surface chemistry, and surface reactivity. It is to be noted that nanoparticles behave differently in different solutions and its dispersion in solutions rarely reflect its distribution at the primary particle size. Nanoparticle agglomeration could lead to size-, surface area-, and dosedependent toxicity.277 Furthermore, it raises concerns about efficiency of ROS production due to the altered photophysical characteristics of the agglomerated PS loaded UCNs. Hence, it is highly essential to study its dispersion characteristics and stability in physiologically relevant buffers over a period of time, before its characterization in vitro. However, this is hardly performed in any study until date, perhaps due to which its efficacy seem to be undermined under in vitro and in vivo conditions. A better knowledge of the characteristics of UCNs at conditions close to the physiological setting will further streamline its optimization very early during the evolution of the construct and steer its development as a better PDT agent. For effective PDT, optimization of the PDT parameters such as the DLI, laser dosimetry including the fluence, fluence rate, and time of irradiation is necessary. For PDT using UCNs, a laser power density ranging between 0.4 W cm−2 to 134 W cm−2 and fluence >1000J cm−2 have been typically used. However, in clinical application and PDT experiments using conventional PS, power density
