Melanoma is a progressive disease that claims many lives each year due to lack of therapeutics effective for the long-term treatment of patients. Currently, the best treatment option is early detection followed by surgical removal. Better melanoma therapies that are effectively delivered to tumors with minimal toxicity for patients are urgently needed. Nanotechnologies provide one approach to encapsulate therapeutic agents leading to improvements in circulation time, enhanced tumor uptake, avoidance of the reticulo-endothelial system, and minimization of toxicity. Liposomes in particular are a promising nanotechnology that can be used for more effective delivery of therapeutic agents to treat melanoma. Liposomes delivering chemotherapies, siRNA, asODNs, DNA, and radioactive particles are just some of the promising new nanotechnology based therapies under development for the treatment of melanoma that are discussed in this review.
The skin is the largest organ of the body, and its functions include protecting the body from injury and pathogens, maintaining fluid retention, and controlling body temperature (Balch, 1998). The skin can be affected by many ailments of which one is cancer. Three main types of skin cancer are basal cell carcinoma, squamous cell carcinoma, and melanoma (Balch, 1998). Melanoma is the most deadly skin cancer developing from melanocytes. Melanocytes are found at the base of the epidermis and produce melanin, which is a pigment found in skin, hair, and eyes (Balch, 1998). Melanoma is a cancer of these melanocytes and results in a progressive disease that arises from either a single melanocyte or dysplastic nevi (Elder, 1999; Elder et al., 1993). Melanoma only accounts for ∼5% of the three major forms of skin cancer, but has the highest mortality of all skin cancers (Balch, 1998). It was estimated by the American Cancer Society that in 2008 there would be 62 480 new cases of melanoma, with 8 420 cases resulting in death. Early stage melanomas, also known as radial growth phase melanomas (RGP) develop radially outwards without breaking into the dermis, while vertical growth phase melanomas (VGP) have broken through the epidermal-dermal junction and are present in the dermis (Elder, 1999; Elder et al., 1993; Guerry et al., 1993). Melanomas in the dermis have the potential to spread throughout the skin causing cutaneous metastases to enter the bloodstream or lymphatic system leading to development of metastases in distant organs such as lung, liver, and brain (Elder, 1999). From a clinical perspective, Breslow’s tumor thickness is an important prognostic indicator of survival for melanoma patients and is an important component used in the staging of primary melanoma tumors (Garbe and Eigentler, 2007). A tumor with ≤1.0 mm depth is classified as stage I, 1.01 mm–2.0 mm as IIA, 2.01–4.0 mm as IIB, and >4.0 mm in depth is classified as stage III and above (Garbe and Eigentler, 2007).
Several reasons account for the lack of effective melanoma therapeutics. These include identification of protein targets causing the disease, development of novel agents, determination of optimal therapeutic combinations, and effective delivery of agents into tumor cells. Optimal agent delivery is required to enhance drug concentration in tumors, reduce side effects, and lower therapeutically effective doses (Wang et al., 2007a). Therefore, scientists and clinicians are focusing on identifying new methods for delivering both existing and novel agents to melanoma tumors.
Several factors are important for liposome efficacy. The size of nanoparticles is crucial for successful delivery into melanoma tumors. Vesicles <100 nm have reduced uptake into liver tissue, while vesicles >100 nm are prone to rapid clearance rates by the mononuclear phagocytic system (DeJong and Borm, 2008). Furthermore, surface modification of liposomes with polyethylene glycol (PEG) can result in prolonged circulation by masking the liposome from this system (Gabizon et al., 1994). Researchers have used a wide variety of lipids in their liposomal formulations, with some of the common lipids listed in Table 2. Heating and light can also be used to enhance nanovesicle content unloading in the body rather than solely through degradation. For example, thermosensitive nanoparticles have been developed for doxorubicin that enhance cytotoxicity about 20 times more in vitro in Lewis Lung Carcinoma cells at 42°C compared to 37°C (Na et al., 2006). In addition, route of administration is important in determining liposome efficacy. Localized administration allows for increased agent accumulation within the tumor without toxicity to the rest of the body, but does not allow for targeting of distant metastases. Alternatively, systemic administration allows for treatment of distant metastases but biodistribution of the liposome agent and toxicity must be dealt with for effective treatment. This review will focus on the use of liposomes for the more effective delivery of therapeutic agents to treat melanoma.
Table 2. Common examples of lipids used in making liposomes
Provides a neutral component to a cationic liposome
Most cancer chemotherapeutic drugs also harm normal surrounding cells and tissue. These side effects can be curtailed by encapsulating these toxic drugs, such as doxorubicin in liposomes to form a ‘protective cage’ around the drug molecules (Figure 2) (Kawakami and Hashida, 2007). These envelopes increase circulation of drugs in the body and enable controlled release. Lipids can be selected for liposomal formulations that enhance these properties. Helper lipids such as 1,2-Dioleoyl-sn-Glycero-3-Phosphoethanolamine (DOPE) and cholesterol are often added to increase stability, assist liposome formulation, and enhance cellular uptake (Table 2) (Farhood et al., 1995). Additionally, pegylated lipids can be added to liposomal formulations to enhance duration of circulation and prevent uptake by the reticulo-endothelial system (Table 2) (Gabizon et al., 1994). These parameters were taken into account when designing liposomal formulations that would encapsulate doxorubicin or incorporate ceramide into the lipid bilayer.
The best-studied liposomal formulation is one containing the alkylating agent, doxorubicin. Doxorubicin functions by intercalating into DNA and has been used in the liposomal formulation known as Doxil. Liposomal doxorubicin showed increased time in circulation and decreased cardiovascular related toxicity when compared with free doxorubicin (Batist et al., 2001). While approved for use in other cancers, phase II trials of Doxil in melanoma patients were stopped due to inactivity (Ellerhorst et al., 1999; Smylie et al., 2007). In contrast, encapsulated doxorubicin liposomes combined with cyclophosphamide in an experimental pulmonary metastatic melanoma mouse model showed high antitumor effect (Shiraga et al., 2008).
A major drawback for successful gene therapy to treat melanoma has been a delivery method to introduce the gene into tumor cells. Viral methods can be efficient but use has been limited due to potential for disrupting functioning of important genes, as viral vectors may insert randomly into areas of the host genome that would disrupt normal function (Lv et al., 2006). Liposomes provide an alternative in which the nucleic acids are protected and delivered into cells without concerns of integration (Akhtar and Benter, 2007).
Cationic liposomes, with or without modifications are frequently used because they spontaneously interact with and load negatively charged nucleic acids (Vorhies and Nemunaitis, 2007). Additionally, cationic lipids are thought to interact with the plasma membranes of cells, thereby, enhancing uptake (Kane et al., 2006; Santel et al., 2006). Another potential advantage of using cationic liposomes is that they are preferentially taken up by the leaky tumor vasculature therefore, increasing the amount of siRNA delivered to the tumor (Hashida et al., 2005). Unfortunately, cationic liposomes have more associated toxic side effects than neutral liposomes. There are reports concerning the rapid elimination from the blood, entrapment by the reticulo-endothelial system, and binding of negatively charged serum proteins for cationic liposomes. In addition, macrophages show preferential uptake of charged liposomes, which can trigger adverse immune reactions (Daniel and Smith, 1999; Hashida et al., 2005). However, inclusion of pegylation can reduce these side effects. Despite these associated toxicities, liposomal delivery is still a safer alternative than viral based vectors, which can induce toxic immune responses and alter gene expression due to viral integration (Akhtar and Benter, 2007).
In a phase II trial, melanoma patients were treated intralesionally with allovectin-7. Allovectin-7 is composed of plasmid DNA containing HLA-B7 heavy chain and beta2-microglobulin encapsulated in DMRIE/DOPE (1,2-dimyristyloxypropyl-3-dimethyl-hydroxyethyl ammonium bromide/dioleoylphosphatidyl-ethanolamine). Allovectin-7 has an excellent safety profile with no observable dose related toxicities (DeJong and Borm, 2008). While the overall mechanism of action for allovectin-7 remains uncertain, an overall response rate of 9% was observed for melanoma (DeJong and Borm, 2008).
Several studies have also utilized mRNAs directly loaded into liposomes instead of using DNA. This removes the requirement for transcription of the gene of interest by the cellular machinery. An example is the mRNA-based vaccine using L27Melan-A/MART-126-35 (MART-1) which produces an epitope that is a melanoma associated antigen (Schumacher et al., 2005). The liposomal formulation consisted of phosphatidylcholine and 1,2-Dipalmitoyl-sn-glycero-3-phospho-rac-(1-glycerol) (DPPG) in a 70:30 molar ratio (Schumacher et al., 2005). MART-1 mRNA induced proliferation of CD4+ T cells, which significantly increased the generation of L27Melan-A/MART-126-35 specific cytotoxic T cells (Schumacher et al., 2005). These effector cells were then able to effectively kill any melanoma cells that were expressing MART-1. Also, MART-1 has been used as a vaccine in a liposome formulation of L-histidine-(N,N-di-n-hexadecylamine)ethylamide (HDHE) and cholesterol to prevent progression and metastasis of the murine B16 melanoma cell line (Mockey et al., 2007). Human MART-1 antigen has a 68.6% amino-acid sequence identity with that found in mice, and as a result, was able to induce an immune response against murine B16 melanoma cells (Mockey et al., 2007). Systemic administration prevented tumor progression of B16 melanoma cells in mice (Mockey et al., 2007).
BAX is a proapoptotic gene that also functions as a tumor suppressor in most cancers (Okumura et al., 2008). One study has used BAX mRNA incorporated in cationic liposomes composed of 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) and DOPE in a 1:1 molar ratio to treat gingival human malignant melanoma (Okumura et al., 2008). Caspase-3 activity and TUNEL-positive cells were increased after treatment with BAX mRNA encapsulated in liposomes. Tumors were 36.7% smaller than saline control groups, 10 days after treatment had ended (Okumura et al., 2008).
shRNA or siRNA targeting genes deregulated during melanoma development
The specificity of siRNA makes it a potentially effective therapy for many conditions including, infection by viruses such as hepatitis B, ischemia, depression, age-related macular degeneration, and cancer (Aagaard and Rossi, 2007; Aigner, 2007; Foster et al., 2001; Grandage et al., 2005). However, in vivo delivery of siRNA to target cells is problematic (Aagaard and Rossi, 2007; Aigner, 2007). Naked siRNAs do not readily cross the plasma membrane and are rapidly degraded by RNases found in serum (Pal et al., 2005). Development of an effective delivery method that protects siRNAs from degradation and delivers them into tumors is essential to utilizing RNAi technology for melanoma patient treatment.
Recently, the PI3K/Akt and MAPK pathways have been targeted in melanoma cells using siRNA to Akt3 and V600EB-Raf loaded into cationic nanoliposomes (Tran et al., 2008a). These nanoliposomes were applied topically to tumor bearing mice that had been pretreated with ultrasound for skin permeabilization (Tran et al., 2008a). The ultrasound effectively permeabilized the skin allowing for successful uptake of the liposomal siRNA complexes into the melanoma tumors. The combination of the two siRNAs resulted in enhanced synergistic reduction in tumor size compared with liposomes containing single siRNAs (Figure 4) (Tran et al., 2008a).
While many studies utilize cationic lipids for siRNA encapsulation, due to possible toxicity neutral liposomes are sometimes used as an alternative. It has been shown that neutral 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine (DOPC) nanoparticles can incorporate siRNA targeting PAR-1 to downregulate melanoma cell growth and metastasis by inhibiting angiogenesis (Villares et al., 2008). PAR-1 is known to be overexpressed in metastatic melanoma cell lines and in patient tumors (Villares et al., 2008). Activation leads to up-regulation of genes involved in adhesion, invasion, and angiogenesis. PAR-1 siRNA DOPC liposomes were found to significantly decrease tumor size and reduce metastatic melanoma lung colony development in mice (Villares et al., 2008). Additionally, those studying other cancer types such as lung and ovarian cancer have used neutral formulations as delivery vehicles for siRNA (Lee et al., 2007; Merritt et al., 2008). Another example involved the targeting of leukocytes with neutral liposomes that used protamine to compact the siRNA (Peer et al., 2008). Used in vivo, the neutral liposomes resulted in more efficient intracellular delivery and gene silencing (Peer et al., 2008).
Delivery of multiple agents
Clinicians and scientists developing therapies for melanoma patients generally believe that to effectively treat melanoma, more than one target gene will have to be targeted simultaneously or sequentially for maximal efficacy. A drug cocktail targeting tumor processes such as proliferation, angiogenesis, and metastasis would provide a novel option for treatment of melanoma. One way to potentially improve drug combinations is to design a liposome containing multiple agents. This approach could be used to take advantage of liposomes to select agents that are incorporated into different locations in liposomes such as in the inner core and the lipid bilayer itself. By combining multiple agents in one liposome formulation, cocktails of drugs could be delivered simultaneously leading to increased patient compliance. For example, melanoma cells treated with c-myc-as targeted with GD2 and doxorubicin, reduced the IC50 by 10-fold (Pastorino et al., 2003). Furthermore, heterogeneous tumors having multiple different genotypes could be more effectively killed since multiple drugs would allow cells resistant to one drug to be killed by a second one.
For instance, liposomes containing 4-S-cysteaminylphenol (4-S-CAP) and magnetite particles were used to target melanoma tumors in mice (Ito et al., 2007; Mitragotri and Kost, 2004). 4-S-cysteaminylphenol is a tyrosine analog that specifically inhibits the tyrosinase activity found in melanoma cells (Ito et al., 2007). 4-S-cysteaminylphenol was combined with magnetite particles which generate heat when an alternating magnetic field is applied, resulting in hyperthermia mediated death (Ito et al., 2007). This combination resulted in regression of 17% of tumors in mice (Ito et al., 2007; Mitragotri and Kost, 2004).
A new area of nanotechnology is developing where a multistage delivery system that is comprised of both biodegradable and biocompatible mesoporous silicon particles has been developed for imaging and therapeutic applications, which can release multiple agents at a controlled rate. These silicon particles serve as a depot from which other nanoparticles such as liposomes are released into the body at specific sites (Tasciotti et al., 2008). This technology has not yet been tested in vivo, but it offers a promising therapeutic approach for treatment of melanoma requiring release of multiple agents over a prolonged period of time.
In order for agents targeting tumor cells to be effective, the drug must accumulate in tumors and be taken up into tumor cells. An ideal drug formulation for melanoma would preferentially accumulate in tumors or if taken up by normal cells, have little or no effect on cellular function. Liposomes can be used to accomplish this objective through several approaches (Figure 5).
It is generally accepted that tumors have ‘leaky vasculature’ due to endothelial gaps around 200 nm and irregular basement membranes (Aagaard and Rossi, 2007). As a result, leaky blood vessels in melanoma would allow liposomes to passively enter the tumors at these sites, but not in normal tissues that would lack this characteristic (Strieth et al., 2008). While this general tendency to accumulate liposome-based drugs in tumors is useful, a more directed method of targeting is highly desirable.
Use of melanoma specific antigens to target liposomes to melanoma cells
Two classically used targeting molecules are folic acid and transferrin (Hilgenbrink and Low, 2005; Singh, 1999). The levels of folic acid receptors are elevated in multiple tumor types and binding of a folic acid linked drug results in uptake by receptor-mediated endocytosis (Mattes et al., 1990; Yang et al., 2007). It was also selected as a targeting moiety due to over expression of transferrin receptor on cancer cells and endocytosis of the receptor complex enabling internalization of the drug linked to transferrin (Harding et al., 1983; Keer et al., 1990).
GD2 and 9.2.27 antibodies
Other factors for targeting melanoma include; GD2 and 9.2.27 antibodies (Allen et al., 2001; Pastorino et al., 2003). GD2 was originally identified as an antigen from a melanoma cell line that reacted with the patient’s serum, indicating that the patient had developed an antibody towards GD2 (Shiku et al., 1976). Importantly, the patient’s serum did not react with their own normal fibroblasts, or with lymphoid cells, platelets, granulocytes, or erythrocytes from melanoma patients whose tumors reacted with the GD2 antibody. This indicated that the GD2 antibody was fairly specific to melanoma cells (Shiku et al., 1976). GD2 was found to be present on ∼65% of melanoma cell lines making it a good choice for a targeting molecule (Shiku et al., 1976). A GD2 antibody has been coupled to a liposome and used to deliver c-myc as ODNs to melanoma xenografts resulting in a 2-month extension of life (mice were euthanized when tumors reached 2000 mm3) compared with untargeted liposomes (Pastorino et al., 2003). 9.2.27 is another monoclonal antibody that recognizes a chondroitin sulfate proteoglycan antigen, similar to rat NG2, which is found on 95% of melanomas (Berd et al., 1989; Pluschke et al., 1996) (Waldmann et al., 2002; Wang et al., 1999). Systemic administration of 9.2.27 conjugated with Bismuth-213 inhibited xenografted melanoma by 75% when compared with untreated tumors (Rizvi et al., 2005).
Monoclonal antibodies are beneficial as anti-tumor agents, but their large size can sometimes contribute to low tumor penetration and immunogenicity due to liver toxicity and non-specific uptake by bone marrow. Peptides, on the other hand are smaller in size, less immunogenic, and easier to produce than monoclonal antibodies (Chang et al., 2009; Lee et al., 2007). For instance, a novel non-small cell lung cancer specific peptide was used to help target liposomes that encapsulated doxorubicin, which led to enhanced drug accumulation in tumors and improved therapeutic index in xenografts of human lung cancer (Chang et al., 2009). These data suggest that small peptides can be used to target nanocarriers such as liposomes.
Magnetic, ultrasound, and electric targeting
Other methods of targeting liposomes to melanomas include magnetic, ultrasound, and electric targeting (Table 4) (Besic, 2007; Byrne et al., 2005; Ito et al., 2007; Mir et al., 1998; Suzuki et al., 2008). These methods use external devices such as ultrasound waves to direct particles to the site of interest (Besic, 2007). For instance, liposomes containing the ultrasound imaging gas, perfluoropropane, in combination with ultrasound were used to effectively deliver a luciferase reporter construct into tumors in mice (Suzuki et al., 2008). The 4-S-CAP experiments discussed above utilized an external magnetic source for targeting of the 4-S-CAP/magnetite liposomes to tumors (Ito et al., 2007). The amount of research in this area is small but promising.
Liposomes make good vehicles for delivery of therapeutic agents into skin because of associated hydrophobic lipid construction. However, many complexities are associated with designing the perfect liposome for effective delivery of therapeutic agents to treat melanoma. Important considerations in liposome-based therapeutic agent delivery include: (1) avoiding induction of immune response or toxicity but still protecting the drug or nucleic acids that are encapsulated; (2) increasing circulation time in the body using pegylation or other modifications; (3) targeting more than one pathway so as to have a cooperative and potentially synergistic effect; and (4) using a targeted approach through conjugation to an antibody or peptide to deliver a therapeutic cargo into the melanoma cells. Liposomes could be the next breakthrough for melanoma therapy by simultaneously targeting multiple key pathways to more effectively treat this disease.