Cancer Nanomedicine: From Drug Delivery to Imaging

See allHide authors and affiliations

Science Translational Medicine  18 Dec 2013:
Vol. 5, Issue 216, pp. 216rv4
DOI: 10.1126/scitranslmed.3005872


Nanotechnology-based chemotherapeutics and imaging agents represent a new era of “cancer nanomedicine” working to deliver versatile payloads with favorable pharmacokinetics and capitalize on molecular and cellular targeting for enhanced specificity, efficacy, and safety. Despite the versatility of many nanomedicine-based platforms, translating new drug or imaging agents to the clinic is costly and often hampered by regulatory hurdles. Therefore, translating cancer nanomedicine may largely be application-defined, where materials are adapted only toward specific indications where their properties confer unique advantages. This strategy may also realize therapies that can optimize clinical impact through combinatorial nanomedicine. In this review, we discuss how particular materials lend themselves to specific applications, the progress to date in clinical translation of nanomedicine, and promising approaches that may catalyze clinical acceptance of nano.


Nanomaterials come in many flavors. They include lipid-based vehicles (liposomes, solid lipid nanoparticles, and micelles) (13); polymer carriers, such as hydrogels, polymersomes, dendrimers, and nanofibers (416); metallic nanoparticles (gold, silver, and titanium) (1720); carbon structures [nanotubes, nanohorns, nanodiamonds (NDs), and graphene] (2130); and inorganic particles, such as silica (3134). These nanomaterials have been envisioned as drug and imaging agent delivery vehicles (or even as drugs and imaging agents themselves). Nanomedicine applications range from cancer to inflammation and regenerative medicine—in essence, the gamut that is clinical medicine. It is becoming clear, however, that different classes of materials are optimal for specific applications. For example, albumin is an effective solubilization platform for hard-to-deliver chemotherapeutics, such as paclitaxel ( NCT00733408). Metallic particles are promising photothermal therapeutic agents (35). Nanocarbons have mediated unprecedented increases in magnetic resonance (MR) efficiency for improving imaging-based diagnosis (36). In addition, the rational and scalable design of targeted polymeric nanoparticles because of the ability to tailor their chemical makeup and properties have enabled the stabilization of small interfering RNA (siRNA) therapies without the need for toxic polycationic agents, as well as small-molecule therapies. These attributes have catalyzed their translation into the clinic with improved efficacy over current standards. Recent studies have also examined how combining the innate attributes of varying classes of nanomaterials to both induce biological responses and respond to biological signals can be used to shrink tumors (32, 37, 38).

More than 7 million people died of cancer in 2008 worldwide, and it continues to be the second leading cause of death in the United States (39, 40). Successful treatment and imaging of cancer relies on several capabilities that can be uniquely addressed via nanomedicine. Nanomaterial sizes match the length scales of tumor interendothelial junctions, allowing for efficient penetration and retention (EPR) for improved localization (Fig. 1). Versatile surface chemistry and surface area–to–volume ratios of nanoparticles mediate potent drug binding or high drug-loading capacities to improve retention, efficacy, and safety. The ability to cofunctionalize nanoparticles with targeting moieties, imaging agents, and drug payloads enables the synthesis of multimodal complexes that may simultaneously provide patients with improved treatment specificity and highly sensitive imaging capabilities to monitor treatment progress and outcomes. Highlighting these attributes, and discussing remaining challenges, which, when addressed, may lead to the fruition of these approaches in the clinic, serves as the foundation for a focused look at how nanotechnology can uniquely affect cancer.

Fig. 1. Passive versus targeted nanomedicine delivery and retention.

Nanoparticles larger than 8 nm in size can passively target tumors through preferential passage through larger interendothelial junctions (40 nm to 1 μm) compared to those of healthy tissue (≤8 nm). Large junctions are a key characteristic of the irregular tumor vasculature. Nanoparticles can also be conjugated with targeting agents, such as antibodies specific to proteins more highly expressed in tumors than healthy tissue, to actively target tumors. These proteins include epidermal growth factor receptor (EGFR/Her1), human epidermal growth factor receptor 2 (Her2), folate receptor, endoglin (CD105), prostate-specific membrane antigen (PSMA), epithelial cell adhesion molecule (EpCAM), CD20, CD44, CD90, and CD133. Once nanoparticles enter tumors, defective lymphatic drainage of nanoparticles results in enhanced retention.


Currently, most nanomedicine studies focus on a single drug to treat cancer, lower toxicity, etc., which can benefit the patient to an extent. Nanomedicine, however, will have the greatest impact when administered in combination with traditional clinical therapies, like radiation, cell therapy, and small-molecule and biological drugs. Current combination therapy using unmodified drug compounds is often based on existing additive guidelines for single therapy treatment, which precludes optimization of combinatorial therapy treatment for maximal efficacy and safety. For example, docetaxel in combination with doxorubicin, followed by cyclophosphamide-methotrexate-fluorouracil (CMF) therapy, has been compared to other variations to this combination for operable breast cancer treatment (NCT00174655). With regard to the potential of nanomedicine and combination therapy, a recent study in animals suggests that timed combination therapy—where an siRNA and a drug are released sequentially from liposomal nanoparticles—might even be the key to overcoming chemoresistance (38).

This review will examine cancer nanomedicine progress toward the clinic with the theme of application-dependent selection of nanoparticle agents. One important way to rationally select nano platforms is structure. For example, cyclodextrins can greatly improve nucleic acid stability to improve transfection efficacy. Gold nanoparticles are particularly effective at crossing the blood-brain barrier (BBB), which can open new avenues for glioblastoma treatment. Photoactive nanoparticles can modulate changes in their own temperature for spatially controlled therapy. Furthermore, faceted properties of materials such as NDs can mediate marked enhancements in MR imaging (MRI) contrast. This review will also examine how algorithm-defined strategies can simultaneously optimize for efficacy, safety, and pharmacokinetics (PK), among other parameters. Ultimately, the rationally designed therapy or imaging method could save time and energy in making sure nanomedicine will translate to those who need it most: the patients.


The use of nanomaterials to treat and image cancer is arguably the most active area of nanomedicine research. Intrinsic properties like surface charge, structure, and biocompatibility can be harnessed for cancer therapy. Size, which can be controlled during synthesis, has been shown to play a major role in nanomedicine efficacy. In particular, extensive work with liposomal and polymer drug delivery vehicles has revealed that complexes greater than 8 nm in diameter preferentially home to tumor tissue versus healthy tissue (Fig. 1) (1). This is due to tumor interendothelial junctions ranging from 40 to 80 nm, on average, and as large as 1 mm, whereas interendothelial junctions of healthy tissues are ≤8 nm wide (41). This preferential uptake combines with defective lymphatic drainage of nanoparticle complexes to promote EPR in tumor tissue (Fig. 1) (1). With this particular approach, efficient intratumoral localization of nanoparticles, and resulting improvements in treatment specificity, can be mediated without functionalizing the vehicle with a targeting moiety. However, harnessing the EPR effect in drug development may be dependent on characteristics that influence tumor vasculature such as tumor type, size, and organ location (42). Therefore, EPR-mediated drug delivery may likely be indication-dependent or reliant on combination therapy.

Although EPR mediates enhanced delivery of larger nano-drug complexes to tumors compared with free small-molecule drugs, increased size can adversely affect nano-drug circulation half-life. Larger particles are more easily recognized and cleared by the liver and phagocytic cells of the reticuloendothelial system (43). Rapid clearance can be partially alleviated by adding polymer coatings, like poly(ethylene glycol) (PEG). PEGylated liposomes 200 nm in diameter, however, are still cleared faster than 100-nm PEGylated liposomes (44). When these studies are taken together, optimal drug delivery complex size for cancer therapy likely ranges from 8 to 100 nm, when relying on the EPR effect. These size requirements are likely to extend from conventional liposomal and polymeric complexes to other nanomedicine platforms, such as metal- or carbon-based ones (45, 46). Targeted drug delivery complexes likely have a wider range of optimal size. Further studies, however, are needed to determine how large these complexes can be and still maintain acceptable body retention rates.

Passive tumor targeting by nanoparticles has proven successful in humans, with liposomal doxorubicin (Doxil) already being clinically used for breast cancer (47). Yet, there are other exciting developments in the field of cancer nanomedicine that capitalize on the unique and intrinsic properties of certain materials for targeted delivery of therapeutics (Fig. 2), including those that are difficult to solubilize, challenging to deliver, or unstable (that is, easily degraded) in vivo (4852). Although the field is vast, some promising examples for translation are described herein.

Fig. 2. Designer nanomedicine.

Nanomedicine platforms can take on any number of properties to tailor therapy or imaging, and improve chances of translation.


Aiming at a target

Drug-loaded nanoparticles can be targeted to surface molecules overexpressed in cancer cells or tumor vasculature. By adding a targeting component, nanoparticles can further improve their efficacy of tumor-specific drug delivery beyond EPR-mediated tumor homing (Fig. 1). Recently, a poly(lactic-co-glycolic acid) (PLGA)/PEG–based nanoparticle, named BIND-014, was designed to target prostate cancer. The polymeric particle, with a PSMA-targeting moiety on the surface and a docetaxel payload, has been tested in preclinical studies and in humans in early-phase clinical trials for metastatic lung cancer and tonsillar cancer (53). The first studies showed the importance of targeted drug delivery and prolonged circulation toward improved therapeutic efficacy and safety via an optimized multimodal polymer nanoparticle where BIND-014 effective concentrations were only 20% of those needed with free docetaxel, which is an important consideration for reducing toxicity of chemotherapeutics. BIND-014 administration to prostate, breast, and lung xenografted tumors that normally (in humans) show little to no response to unmodified docetaxel treatment resulted in a significant therapeutic effect in animals (53). To date, clinical trials pertaining to examining the efficacy of BIND-014-001 in advanced and metastatic cancers are active (NCT01300533), and another clinical trial pertaining to the application of BIND-014-005 has been approved for non–small cell lung cancer (NCT01792479).

Transferrin receptor targeting using nanoparticles has also been explored clinically in pancreatic cancer, which overexpresses the transferrin receptor in as much as 93% of the tumor cells, whereas normal tissue did not stain positively for the transferrin receptor (54). This includes recent developments in the clinical evaluation of transferrin receptor–targeted liposomal gene therapy that may improve standard drug efficacy through restoration of tumor suppressor genes such as p53 (NCT00470613) and Rb94 (NCT01517464) (Table 1). This was particularly well demonstrated with SGT-53 (SynerGene Therapeutics Inc.), a transferrin receptor–targeting liposomal nanoparticle that delivers p53-containing plasmids to tumors that overexpress the transferrin receptor. Preclinical data demonstrated enhanced gene therapy to murine metastatic pancreatic tumors, improving gemcitabine response (55). Clinical trials demonstrated that single-agent SGT-53 delivery is well tolerated and the p53 transgene can be seen specifically in metastatic tumor specimens and not in normal tissue. The trials have since been modified to include combinatorial therapy with docetaxel, as originally intended (56).

Table 1. Example nanoparticle applications in cancer drug delivery.

Table is organized by stage of translation.

View this table:

In addition to targeting cancer cells, targeting the tumor vasculature has also demonstrated to be a viable method for targeted drug delivery (Fig. 2 and Table 1). This is best exemplified by RGD-containing peptides that target αv integrins overexpressed in tumor vasculature (57). Conjugation and coadministration of nanoparticles with internalizing RGD (iRGD), a peptide that combines RGD with a tumor-penetrating CendR motif, exhibited enhanced tumor targeting, penetration, and efficacy compared to nontargeted nanoparticles (58, 59). Clinical trials are currently under way to evaluate the effect of iRGD on tumor vasculature and should pave the way to translation of iRGD-targeted delivery of drugs and nanoparticles (NCT01741597).

Capitalizing on carbon

Carbon nanomaterials, including nanotubes, nanohorns, fullerenols, graphene, and NDs, have been the subject of several promising preclinical studies owing to their versatile surface properties. Carbon materials can deliver a wide spectrum of therapeutic compounds, often with little to no modification to the drug or nanomaterial, owing to the ability to mediate π-π stacking with nanocarbon surfaces for noncovalent delivery (21, 24, 26, 27, 6063). A major concern surrounding these materials is their safety because of potential cardiopulmonary toxicity. However, multiple reports show no apparent toxicity after nanocarbon administration in mouse models (24, 26, 45, 60, 62). Carbon nanomaterials have yet to be tested in humans, as continued safety assessments are ongoing prior to U.S. Food and Drug Administration (FDA) approval for clinical studies.

A major hurdle of cancer therapy is intrinsic or acquired chemoresistance, which contributes to treatment failure in more than 90% of metastatic cancers (64). A common mechanism of chemoresistance is the overexpression of adenosine triphosphate–binding cassette (ABC) transporter proteins that can efflux a wide range of small molecules, including chemotherapeutics (65). Conventional methods of overcoming ABC transporter protein chemoresistance have involved the development of inhibitors of these proteins. Clinical trials for such drugs have met little success as potent inhibitors of multiple transporter proteins and have been associated with trial-ending toxic side effects (66). More specific inhibitors struggle to find a balance between efficacy and tolerability (67). Moreover, administering a single inhibitor is unlikely to succeed because chemoresistant tumors often express different ABC transporter proteins. Delivery of anthracyclines—a class of DNA-damaging chemotherapeutics—by NDs has been shown to overcome ABC transporter drug efflux without targeting the transporter itself or another overexpressed cancer protein (26, 68). In mouse models of anthracycline-resistant cancer, delivery of the anthracycline doxorubicin by NDs (Fig. 2) resulted in increased tumor-killing efficacy as well as lower systemic toxicity, including less myelosuppression, compared with free doxorubicin (26). The combination of safety and efficacy observed with the ND-doxorubicin agents was mediated by potent drug binding, without the need for chemical modification of the drug or the ND surface. A targeting moiety was also not present, suggesting that the efficacy of NDs was likely related to EPR.

Although anthracycline delivery by NDs improved therapeutic efficacy compared to traditional chemotherapy—at least in mouse models—even more impressive therapeutic gains can be achieved using a targeting component. Hybrid drug delivery complexes containing epirubicin (a chemotherapeutic) and an EGFR antibody (Fig. 2) were delivered to mice bearing triple-negative breast cancer (TNBC) human breast cancer (MDA-MB-231) xenografts (Table 1) (45). Drug-loaded EGFR-targeted complexes maintained the enhanced safety profile of their untargeted counterparts while increasing homing to EGFR-positive mammary tumors. This resulted in increased tumor response and, in some cases, complete regression of these highly aggressive tumors.

These studies demonstrate a clear clinical potential for untargeted and targeted carbon nanomedicine. However, judging by the lack of nanocarbon clinical trials being conducted, continued toxicity and clearance assessment in rodent and large animal models remains a critical precursor to translation.

Silencing genes

One area where the intrinsic properties of nanoparticles can improve drug delivery is RNA interference (RNAi). RNAi is a promising method for silencing cancer-causing genes, but is hampered by enzymatic degradation of small RNAs in vivo. As the most clinically developed example, cyclodextrin-based polymers (CDPs) are well tolerated and can self-assemble to protect and deliver siRNA to cancer cells (Fig. 2) (6971). Following successful preclinical studies, these CDPs were used in the first siRNA-based cancer therapy clinical trial targeting patients with recurrent or metastatic solid malignancies (NCT00689065) (69, 71). Calando Pharmaceuticals developed CALAA-01, a CDP-siRNA complex functionalized with a transferrin-targeting ligand for the treatment of metastatic melanoma, PEG to promote stability and biocompatibility in vivo, and an siRNA to silence the gene RRM2 (ribonucleotide reductase M2 subunit) (Table 1). A striking element of this approach is the ability for self-assembly of the nanoparticles through the interaction of the positively charged cyclodextrin chains around the negatively charged siRNA to improve nucleic acid stability. The process of translating the CDP-siRNA complex included stabilizing RNA therapeutics, manufacturing scale-up [for example, chemistry, manufacturing, and controls (CMC) compliance], and ensuring delivery specificity to melanoma solid tumors. Although the clinical trials are still ongoing, there is evidence of siRNA activity in tumor biopsies from treated patients (69).

In addition to CDP-based siRNA delivery, a previous clinical study used liposomal bcr-abl siRNA for imatinib-resistant chronic myeloid leukemia (CML) treatment that resulted in reduced bcr-abl gene expression in peripheral blood samples and reduced extramedullar CML node size (72). However, transfection efficiency decreased by the third round of administration, in turn reducing the ability to affect bcr-abl mRNA levels. A stable and rationally designed delivery system could perhaps resist serum ribonuclease activity that limits transfection efficiency. To improve delivery and increase the circulatory half-life of siRNA, lipid-based carriers with tumor-targeting and tumor-penetrating moieties have been developed (51) (Table 1). These “siRNA nanocomplexes” were designed to display a cyclic nonapeptide on its surface, LyP-1. LyP-1 targets p32—a mitochondrial protein overexpressed on stressed tumor and tumor-associated cells—and is also able to penetrate cells after being proteolytically processed by endogenous proteases. This ensured that the particle was more toxic to tumors than to healthy tissue, because the payload was only released once inside the cancer cell. The nanocomplexes were loaded with siRNA against the oncogene inhibitor of DNA binding 4 (ID4) and tested in mouse models of human ovarian cancer. Tumor-penetrating nanoparticles functionalized with ID4-silencing siRNA suppressed tumor growth by 82%, whereas control injections (saline and untargeted control nanoparticles) did not suppress tumor growth.

As another example of how nanoparticles deliver siRNA—except with the nucleic acids displayed on the outside rather than encapsulated—Jensen et al. created gold nanoparticle–based SNAs (73). These SNAs were recently used for RNAi against the Bcl2L12 oncogene in a mouse glioblastoma model. SNAs exhibited no apparent toxicity, and owing to their resistance to nuclease-triggered degradation, they mediated stable and persistent gene transfection without the need for polycationic polymers. SNAs exhibited substantial accumulation in the brain tissue of healthy mice with an intact BBB after systemic injection, providing a strong basis for applications in brain cancer therapy, given the fact that the presence of the BBB has blocked many candidate therapies (Fig. 2).

Pulling the trigger

Photothermal therapy involves the use of heat to kill cancer cells, which has gained interest because the inducible and noninvasive nature of this triggered therapy can be applied to a wide range of solid tumors (Table 2). A novel class of materials based on a gold nanoshell encapsulating a silica core (Fig. 2) has been shown to be effective in near-infrared laser-induced cancer therapy in multiple cancer models, including lung, head and neck, prostate, and gliomas (46, 7476). Unique properties that make these nanoshells clinically promising include biocompatibility of the particles as well as the tunability of the surface plasmon resonance of the gold metal shell based on the modulation of the ratio of the dielectric core radius to shell thickness. This architecture allows for highly efficient and localized conversion of light to heat. These two properties are critical to developing a safe product that can be used for heat therapy in a targeted area (as wide as 2 mm) with minimal laser power requirements (10 to 100 mJ/cm2) (77, 78). In a recent mouse study, these gold-coated silica particles were labeled with VEGF to target the blood vessels of intracerebral tumors (35). Following administration and photothermal ablation, the vasculature of diseased tissue was ruptured, whereas healthy tissue was not damaged. This nanoshell technology has been commercialized by Nanospectra Biosciences as AuroLase, with clinical trials in head and neck cancer (NCT00848042) and primary and metastatic lung tumors (NCT01679470) already under way.

Table 2. Example nanoparticle applications in cancer therapy.

Table is organized by stage of translation.

View this table:

In addition to photothermal therapy, several other triggered modalities based on radio- or magneto-stimulus are being explored, including magnetic and radio-based methods for enhancing drug delivery and efficacy (Table 2) (79, 80). These methods may be more easily translated because MRI and radiotherapy are already commonly used in cancer treatment. Radiotherapy is a standard method of tumor treatment that relies on greater absorption of x-rays in tumor tissue compared to nearby normal tissue. Radio-induced nanomedicine, specifically using hafnium oxide nanoparticles, has demonstrated enhanced energy deposit in nanoparticle-containing subcellular structures of cells, which results in enhanced energy release and subsequent localized cellular destruction from nanoparticle clusters. Multiple xenograft tumor models demonstrated greater sensitivity to ionizing radiation treatment when pretreated with hafnium oxide nanoparticles compared to radiation treatment alone. This approach is currently being evaluated clinically in patients with soft tissue sarcoma of the extremity or advanced squamous cell carcinoma of the oral cavity (NCT01946867 and NCT01433068) (80).

Immune system plays defense

Immunotherapy involves the activation or suppression of immune responses for disease treatment. In essence, immunotherapy capitalizes on our natural defense system to prevent the growth (or occurrence) of cancer. Emerging immunotherapeutic strategies for cancer involve the delivery of agents, such as tumor-specific antigens (Fig. 2), that can prime or stimulate one’s own immune system to actively seek out and destroy cancer cells that have these antigens (Table 2). As a new area of research, nanoparticles may be used to modulate the immune response to tumors (81). Nanotechnology has made the passive and targeted delivery of antigens and adjuvants possible. In addition, nanomaterials have helped overcome biological barriers that have stunted the development of more conventional approaches to adjuvant delivery such as oral delivery. This is particularly evident when delivering adjuvants and antigens orally through enhanced uptake by mucosal immune cells, as well as delivering easily degradable adjuvants, such as RNA (82, 83). Because of the preclinical successes of nano-based immunotherapy, several clinical trials (NCT00291473 and NCT00651703) are currently underway to translate these approaches to a variety of human diseases ranging from cancer to allergies (84, 85) (Table 2).

Nanoparticles have been recently used in cancer immunotherapy vaccines because they can co-deliver multiple agents, such as an antigen and adjuvant, in a single, biocompatible platform with enhanced uptake and efficiency compared to conventional treatment [for example, standard vaccines and dendritic cell (DC) therapy]. Cancer vaccines impair cancer progression or prevent occurrence by delivering specific tumor antigens that induce the generation of antigen-specific cytotoxic T lymphocytes (CTLs). Strong CTL response often requires the coadministration of adjuvants. Nanoparticles are well suited for this application because they can ensure delivery of both an adjuvant and an antigen to the same immune cell, whereas standard vaccine formulations are more inefficient and require concentration-dependent levels of both adjuvant and antigen to achieve similar effects. One of the first nanoparticle cancer vaccine delivery vehicles was based on a truncated Her2 protein complexed with hydrophobic polysaccharides into nanoparticles that enhanced the uptake of Her2. Delivery of both an antigen and an adjuvant in a single nanoparticle in mice resulted in the stronger production of anti-Her2 antibodies compared to vaccination with Her2 protein alone as well as CTL-mediated repression of Her2-expressing murine tumors in vivo (86). This work is currently in clinical trials (NCT00291473) where it has been relatively well tolerated and elicits similar immune responses in humans (84, 87). Immune response was seen in 93% of patients, which is compelling when compared to clinical trials with E75, an immunogenic peptide from Her2/neu protein, that saw postvaccine delayed-type hypersensitivity response in 74% of treated patients (88).

Antigen-presenting cells (APCs) are required for CTL response to an antigen. DCs have been identified as the most potent APCs and have subsequently been used for vaccines. Conventional DC-based vaccines involve the isolation and expansion of antigen-exposed DCs in vitro, followed by reintroduction of activated DCs into cancer patients. Although DC vaccines are a promising strategy that has seen some success in clinical trials, this procedure remains costly and inefficient. Several studies suggest that, owing to their versatile surface chemistry and functionalization, nanoparticles can be used as modular platforms to efficiently deliver antigens and prime DCs for an anticancer immune response (89, 90). These studies often incorporate antigen exposure with DC chemoattractants and DC activators, such as Toll-like receptor (TLR) ligands. For instance, polylactide-co-glycolide (PLG) was used to form matrices that expressed the TLR9 ligand CpG-ODN and also released the cytokine GM-CSF as well as melanoma tumor antigens (90). This work demonstrated that GM-CSF was capable of recruiting DCs to the PLG matrices in vivo in mice, whereas CpG-ODN activated the recruited DCs. Exposure of these activated DCs to melanoma tumor antigens by PLG matrices resulted in a potent CD8+ CTL response and an enhanced impairment of melanoma tumor progression. Translation of this work into the clinic should lead to a cheaper and faster DC-based vaccine approach compared to current methods that require the removal, ex vivo activation, and reimplantation of a patient’s DCs. Indeed, phase 1 trials are underway (NCT01753089).


Many of the advantages that are offered by nanotechnology for drug delivery, such as improved circulation half-life, EPR, and reducing toxicity, are also beneficial to imaging. However, for nanomedicine to benefit the imaging community in a translational setting, it is important that properties unique to the nano-imaging agents be harnessed. Conventional approaches to making nano-imaging agents include high-density encapsulation of contrast agents such as gadolinium (Gd) or manganese oxide into liposomes and PLGA, or loading the contrast agents onto the surface of nanoparticles, such as carbon nanotubes, gold, or NDs. These strategies can improve circulation time and mediate tumor localization through EPR, thus improving imaging capabilities (Table 3). Perhaps one of the most promising nano-imaging approaches is the use of superparamagnetic iron oxide nanoparticles, which are currently in clinical trials for lymph node imaging and pancreatic cancer staging (NCT00920023) (91). Iron oxide nanoparticles are paramagnetic, which can be imaged after the application of an external magnetic field. Coating iron oxide nanoparticles with lipid-like materials has resulted in improved control of particle size, dispersal, and, importantly, biocompatibility (92). Iron oxide degradation in vivo is also possible, which provides clear pathways for clearance after administration (93).

Table 3. Example nanoparticle applications in cancer imaging.

Table is organized by translational status.

View this table:

Although established imaging platforms, such as iron oxide, have transitioned into clinical studies, emerging approaches using novel materials that can uniquely enhance imaging safety and efficacy are being developed. With regard to cancer nanomedicine (94, 95), nano-enhanced imaging can be used to track therapy or diagnose disease. In particular, nanoparticles have been designed to enhance contrast agent efficiency (96). Some nanoparticles, such as self-assembling protein cages, can be used to increase packaging of paramagnetic ions (97). Such systems have demonstrated enhanced localized relaxivity over free paramagnetic ions. ND-based MRI is a revolutionary approach to MRI because the ND surface mediates a per-Gd relaxivity (contrast efficiency) increase of 12-fold (36). This is among the highest reported per-Gd relaxivity values owing to the faceted ND surface, which potently attracts water. The ability of the NDs to facilitate water molecule interaction with Gd enables higher MR signal, and in turn, a lower Gd concentration is needed to achieve the clinically required brightness and contrast. Clinically speaking, one order of magnitude reduction in the amount of Gd needed would represent a game-changing advance in medicine because this would allow for simultaneous high-contrast imaging while also markedly reducing the amount of contrast agent needed to achieve this. Furthermore, this may increase the population of Gd-amenable patients who were previously precluded from Gd-based MRI because of renal toxicity. In addition to ND-Gd, gadofullerenes, gadonanotubes, and silica-based imaging agents have demonstrated per-Gd relaxivity increases over the delivery of Gd alone (98, 99).

MRI is a standard of care for preoperative diagnostic imaging of tumors, yet intraoperative MRI has proven to be difficult and not very accurate. Kircher et al. developed a trimodal MRI-photoacoustic-Raman (MPR) nanoparticle imaging complex. This multimodal particle used both photoacoustic and Raman imaging for imaging during tumor resection combined with accurate MRI preoperative evaluation of tumors (100). MPRs accumulated preferentially in tumors via EPR and successfully delineated brain tumor margins preoperatively and intraoperatively with a single injection in a human glioblastoma mouse model.

A particularly promising approach being developed for MRI is CEST, which uses polypeptides as biocompatible and biodegradable MRI agents. Amide, amine, and hydroxyl protons and their unique exchange rates can be assigned specific colors to produce multicolor MR probes (101, 102). Recent studies have also demonstrated the applicability of CEST compounds as pH sensors to monitor cell transplantation and viability (103). Here, mouse hepatocytes and liposomes loaded with l-arginine, a pH-sensitive CEST agent, were injected into mice. Reductions in pH after cell death resulted in decreased MR signal from the l-arginine–loaded liposomes, which served as a modality to monitor murine hepatocyte viability after transplantation. Additional studies demonstrated the use of CEST functional liposomes that carry color signatures unique to the polypeptides being carried, which result in spatially defined, multicolor imaging of popliteal lymph nodes (102). These studies exemplify how well-designed nanoparticle systems can address clinical problems that more conventional approaches are unable to achieve.


Safety is of paramount importance in any drug’s translational roadmap. Although the highlighted approaches in this Review have demonstrated remarkable potential toward clinical translation, continued clinical trials of promising platforms and development of pathways toward approval of nanomaterials will continue to shape the landscape of how nanomedicine can affect clinical practice. The cancer applications described here, where nanomedicine is poised to improve the safety and efficacy of treatment in a clinical setting—and is already doing so in clinical trials—are by no means exhaustive, because enhanced drug delivery and imaging can also benefit the fields of ophthalmology, regenerative medicine, neurodegenerative diseases, sexually transmitted diseases, infectious diseases, and global health.

The issue that follows now is the evaluation of the role nanomedicine will play, if any, in truly transforming the way that medicine is practiced. Nanoparticles can affect a wide range of diseases by improving efficacy and safety, prolonging circulation time, and enhancing delivery specificity. For cancer, whether or not a single drug can truly lead to sustained remission is a major question moving forward in nanomedicine design. The concept of combinatorial treatment is gaining traction as a required method of treatment, particularly in the field of cancer therapy (Fig. 2). Furthermore, as cancer therapy begins to marry diagnostics and therapeutics into a more personalized approach, optimal combinatorial therapy will likely differ for subsets of patients within a single type of cancer.

With regard to clinical evaluation of the best methods of combinatorial treatment, these studies are often based on the therapeutic guidance and additive combination of singular treatment with these drugs, resulting in only marginal improvements to efficacy and safety. As medical research begins to become more multidisciplinary, emerging methods, such as feedback system control (FSC) and combinatorial drug design (CDD), can be used to bring engineering approaches to rational CDD and optimization using nanomedicine. Real-time drug monitoring and personalized PK will enable more tailored nano-based therapies (104). In addition, emerging strategies including live cell interferometry can quantify changes in cellular mass and resulting fitness from populations of cells in response to drug dosing. This information can serve as a vital readout to monitor tissue response to combinatorial nanomedicine (105).

One of the emerging methods that may transform the way that nanotherapy is administered and remove its barriers to acceptance is the concept of FSC, a top-down approach that directs a cell or population of cells toward a desired phenotype through control strategies that were conventionally reserved for nonbiological scenarios (106, 107). When drug compounds with arbitrarily decided concentrations are applied to cells, their signaling pathways respond unpredictably. If a specific outcome is not met (for example, optimal apoptosis and optimal safety), FSC is a broad approach that draws from a set of mathematical search algorithms. These include Differential Evolution and the Gur Game, which select the next group of iterative drug administration conditions to rapidly achieve desired phenotypes, even among a prohibitively large testing parameter space. Although six drugs with 10 selected concentrations would each require 1,000,000 potential trials, FSC “self-guides” a cellular population toward desired phenotypic outcomes that rapidly converge with only tens of iterations. In addition, because FSC can take PK/PD (pharmacodynamics) into account during the search process, it is easily adaptable toward the implementation of nanotherapeutic combinations that are comprehensively optimized for both efficacy and safety.

In addition to FSC, additional large-scale, high-throughput methods and databases can be unified to analyze interactions at the gene, signaling pathway, and whole systems level after the coadministration of candidate drugs. CDD assesses these interactions to converge upon a directed dosage of multiple compounds that result in a targeted phenotypic outcome and simultaneously reduces off-target effects. Effective CDD uses publicly available databases in parallel with computational/mathematic resources to evaluate varying types of drug interactions and test entire parameter spaces to arrive at a combinatorial dose that satisfies designated criteria including optimized efficacy and safety (108). These include DrugBank ( and Therapeutic Targets Database (, which provide information on thousands of drug compounds and targets. The National Institutes of Health Clinical Collection ( provides information on more than 400 small molecules that are widely used in clinical trials. STRING ( is a database of more than 5 million proteins that provides insight into protein-protein interactions. These libraries can be paired with computational and modeling tools such as flux balance analysis, which has been used to design drug combinations for cancer and other disorders by accounting for both drug activity and toxicity. Petri nets can assess drug-mediated effects upon gene expression and subsequently use this information to formulate synergistic approaches to enhance treatment potency.

FSC and CDD serve as optimal dosing principles that can guide the packaging of directed drug combinations into nanomedicine delivery vehicles to maximize their clinical efficacy. This is particularly applicable for drugs, such as chemotherapeutics, that require improved circulatory stability, intratumoral retention in the cases of innate and acquired chemoresistance, and half-life. To address these challenges, promising strategies are on the horizon. These include the aforementioned use of layer-by-layer co-delivery of siRNA and small molecules for TNBC treatment (38) to overcome chemoresistance. A recent study (109) combined liposomal and silica nanoparticle delivery to treat pancreatic ductal adenocarcinoma by interfering with the formation of pericytes, which are cells that not only protect vascular endothelial cells but also typically block nanoparticle access to the vasculature. These studies demonstrate that combinatorial drug delivery through single or multinanoparticle platforms serves as a feasible and promising route toward translating rationally developed drug combinations into the clinic.


Although a nanomaterial-modified therapeutic may be a stand-alone solution for cancer owing to improved safety and efficacy over an unmodified compound, the issue of whether or not a single nanoparticle agent can effectively cure a patient of cancer should be raised. The need for an effective combination of small-molecule drugs, biologics, nucleic acids, and/or other compounds may be necessary to mediate comprehensive cancer management, or potentially a cure. The diverse properties of nanomaterials should be able to facilitate combinatorial treatment, and optimization methods such as FSC and CDD can improve the likelihood of such combinatorial therapies ending up in the clinic. Furthermore, if nanoparticles are needed as carriers, a determination of which nanoparticle would be best suited for each compound or imaging agent is required. As reviewed here, nanoparticles can have varying properties, such as aspect ratios, surface properties, and capacity for and interaction with external and biological stimuli, that will determine the best nanoparticle for a particular application (Fig. 2).

Bringing nanomedicine to market is a multidisciplinary effort, and brings unique challenges and opportunities at the intersection of manufacturing, regulation, and funding. Regulatory bodies from all over the world, including the FDA, Pharmaceuticals and Medical Devices Agency of Japan, and European Medicines Agency, have been actively developing procedures for the consideration of new drugs or imaging agents that are delivered via nanotechnology, and the general requirements for translation into clinical studies have been established (Fig. 2) (110112). Before the filing of an Investigative New Drug (IND) application, which is important for both the commencement of clinical trials and any subsequent product development, these requirements include compliance with quality control guidelines [for example, good laboratory practice (GLP) and good manufacturing practice (GMP)] to ensure material uniformity and reproducibility. Absorption, distribution, metabolism, and excretion (ADME) and toxicity studies account for acute and chronic material safety. CMC guidelines, which include analytical testing and process controls, serve as the foundation for synthesis at the kilogram and larger quantity scales, as well as achieving the selected formulation required for administration (for example, intramuscular, intravenous, and oral).

The completion of these and other requirements for FDA IND filings and subsequent clinical trials represents an ambitious threshold. Key barriers in completing these requirements include the significant costs of manufacturing scale-up, completing GLP-compliant preclinical safety, and toxicity and efficacy studies that are adequately designed so that findings can be accepted by the FDA. Despite these important challenges that must be addressed, the cancer nanomedicine field is starting to experience success in navigating the complex roadmap needed to realize its impact in the clinic. As previously mentioned, funding is a necessary part of this roadmap, and support from government agencies, venture firms, and pharmaceutical companies has served as a catalyst for the translation of nanomedicines described in this Review. This is exemplified by partnerships between BIND Therapeutics [for example, BIND-014 (53)] and AstraZeneca, Amgen, and Pfizer. With these requirements in mind, success in achieving accelerated drug approval through the FDA 505(b)(2) pathway remains to be seen for cancer nanomedicines in the pipeline. The 505(b)(2) approval of Abraxane, a paclitaxel-albumin nanoparticle for the treatment of non–small cell lung cancer in combination with carboplatin, represents a positive step forward for fast-track approvals of nanotherapeutics. This, coupled with the existing approvals for imaging drugs such as iron oxide (Feridex) and Gd-diethylenetriamine pentaacetic acid (Gd-DTPA/Magnevist), may further support the translation of nanotechnology-modified imaging compounds (Table 3).

In summary, a clearer roadmap toward solidifying how nanomedicine can move from enhancing treatment efficacy to eliminating the major diseases of our generation is coming to fruition. To this end, harnessing the highlighted unique features of nanomaterial platforms will inspire the rational design of combinatorial nanotherapeutics, which, combined with powerful new imaging agents, will underscore the true catalysis of nanomedicine translation.


  1. Acknowledgments: We would like to acknowledge the studies that we were unable to include in this Review. Competing interests: D.H. is listed as an inventor on pending patents pertaining to ND drug delivery and imaging.
View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article