Nanoparticles in the clinic

Abstract Nanoparticle/microparticle‐based drug delivery systems for systemic (i.e., intravenous) applications have significant advantages over their nonformulated and free drug counterparts. For example, nanoparticle systems are capable of delivering therapeutics and treating areas of the body that other delivery systems cannot reach. As such, nanoparticle drug delivery and imaging systems are one of the most investigated systems in preclinical and clinical settings. Here, we will highlight the diversity of nanoparticle types, the key advantages these systems have over their free drug counterparts, and discuss their overall potential in influencing clinical care. In particular, we will focus on current clinical trials for nanoparticle formulations that have yet to be clinically approved. Additional emphasis will be on clinically approved nanoparticle systems, both for their currently approved indications and their use in active clinical trials. Finally, we will discuss many of the often overlooked biological, technological, and study design challenges that impact the clinical success of nanoparticle delivery systems.


| I N T R O D U C T I O N
Nanoparticle/microparticle delivery systems are widely investigated preclinically with many particle-based formulations and technologies having already been introduced in the clinic. [1][2][3][4][5] Oral, local, topical, and systemic (e.g., intravenous) administration are all proven methods that have been Food and Drug Administration (FDA)-approved for the delivery of nanoparticles/microparticles, depending on the desired application or targeted site. For example: (a) oral delivery of particles has been approved clinically for imaging applications (e.g., Gastromark), 6 (b) local delivery of particles has been widely used in the clinic as depot delivery systems for the extended delivery of a variety of biologics including peptides and other small molecules (e.g., DepoCyt), 4 (c) topical application of particles has been approved clinically to increase penetration of biologics across the skin barrier (e.g., Estrasorb), 7 and (d) systemic delivery of particles has been approved clinically for treating a variety of cancers (e.g., Doxil) 8 and other diseases. Given the utility and success of these clinical examples, preclinical research efforts for each of these delivery methods continue to increase with particular attention placed on developing new applications and further improving their delivery and efficacy.
Of these delivery methods, intravenously administered nanoparticles receive the most attention, both preclinically and clinically. The increased interest for intravenous delivery is not surprising given that nanoparticles delivered systemically have direct access to nearly all parts of the body and thus have the most potential to influence clinical care. For this same reason, systemically delivered nanoparticles also face exceedingly difficult challenges with regards to both the delivery aspect (e.g., biological challenges) 9,10 and the regulatory aspect (e.g., study design and approval challenges). 11,12 This review focuses on the clinical translation of intravenously administered nanoparticles, with additional emphasis on the challenges faced by nanoparticles from a clinical and translational point of view. Specifically, the biological, technological, and study design challenges facing the clinical translation of nanoparticles will be discussed. Comprehensive lists of intravenous nanoparticle technologies that are either approved or currently in clinical trials will be provided to highlight the current clinical landscape. organic nanoparticles (e.g., polymeric, liposomes, micelles, etc.). Inorganic nanoparticles have been successful in preclinical studies, are being developed in the clinic for a variety of applications including intraoperative sentinel lymph node imaging and thermal ablation of tumors, and have already been approved for imaging applications and anemia treatment (Figure 1). [13][14][15] Alongside this, organic nanoparticles have also exhibited substantial success in the clinic where they are currently being developed for broad applications ranging from vaccination, to hemostasis, to long-lasting depot delivery systems, to topical agents for systemic delivery through the skin. [1][2][3][4][5] More relevant to this review are nanoparticle formulations that are delivered intravenously, and in this realm, organic nanoparticles predominantly fall into two categories: (a) nanoparticles for gene therapy applications 22,23 or (b) nanoparticles for delivery of small molecule drugs for cancer treatment (e.g., head and neck, melanoma, breast, metastatic, etc.). 24,25 Organic nanoparticle formulations for other applications (e.g., vaccines, fungal treatments, etc.) are also in development and will be highlighted here ( Figure 1).

| N A N O P A R T I C L E T Y P E S , A P P L I C A T I O N S , A D VA N T A G E S , A N D P O T E N T I AL
The main reasons behind the interest in nanoparticle technologies are that: (a) in the case of organic nanoparticles, they possess distinct advantages over many intravenously administered pharmaceuticals and biologics, and (b) in the case of inorganic nanoparticles, many stimuli responsive functions are possible based on specific colloidal assemblies. Organic nanoparticles can be designed and formulated to offer enhanced drug protection, controlled release, extended circulation, and improved targeting to diseased tissues as compared to their free drug counterparts. 25,26 Likewise, inorganic nanoparticles benefit from these same advantages, and additionally from stimuli-responsive functions arising from their surface plasmon resonance (e.g., thermal heating or imaging) or magnetic responsiveness (e.g., magnetic resonance imaging [MRI] imaging or magnetic targeting) that individual drugs or other molecules (e.g., noncolloidal) do not offer. 2,27 Given these advantages, it has been a long-held idea that nanoparticles have the potential to dramatically change clinical care by introducing new, or improving upon current, therapies. A large portion of the interest in nanoparticles stems from their potential as a platform delivery system, with the capability of exchanging specific design features (e.g., targeting antibodies, the encapsulated drug, and control over how/when the diseased site interacts with this drug) in a "plug-and-play" format to treat additional or other diseases.

| C L I N I C A L L Y A P P R O V E D N A NOP A R T I C L E S / MI C RO P A RT I C LE S
Currently, there are a number of nanoparticle therapeutics, imaging agents, and technologies that have been approved for clinical use, either by the FDA in the United States, or the European Medicines Agency (EMA) in the European Union (Table 1). In this section, we will highlight the currently approved nanoparticles and their clinical indications.

| Cancer nanoparticle medicines
Many clinically approved nanoparticle formulations are used in treating various cancers at a variety of stages. Interestingly, all but one of these systems (Abraxane) is liposomal systems encapsulating an anticancer drug. Doxil, polyethylene glycol (PEG) functionalized liposomal doxorubicin, was the first approved (FDA 1995) cancer nanomedicine. 8 Soon after, other liposomal formulations such as liposomal daunorubicin (DaunoXome), 28 liposomal vincristine (Marqibo), 29 and most recently liposomal irinotecan (Onivyde) 30 were approved by the FDA, whereas non-PEGylated liposomal doxorubicin (Myocet) 31 and liposomal mifamurtide (MEPACT) 32 were approved by the EMA. The lone nonliposomal nanoparticle system currently approved for cancer treatments is Abraxane, which is an albumin-bound paclitaxel nanoparticle. 33 The majority of these formulations are not PEGylated, with the exception of Doxil and Onivyde, 34 which is perhaps surprising given the widely known advantages even small amounts of PEG have shown to confer to nanoparticle delivery systems. [35][36][37] Additionally, all of these formulations are passively targeted, with no active or chemicalbased targeting moieties; again, this is despite the proven advantages of active-targeting in preclinical settings. 25,26,38 It is likely that the other advantages, notably their reduced toxicity stemming from their ability to preferentially accumulate at tumor sites and limit off-target side effects via the enhanced permeation and retention (EPR) effect, 39 are responsible for the success and increased efficacy that these approved particles have over their free drug counterparts.

| Iron-replacement nanoparticle therapies
Another clinical area where nanoparticles have made a significant impact is in iron-replacement therapies for treatment of anemia (Table   1). [40][41][42] In these applications, the nanoparticle (iron-oxide colloids) is the therapeutic with the goal being to increase iron concentration in the body. 43 These nanoparticle approaches originated from the need to address toxicity issues associated with the injection of free iron. 40,42 Using colloidal iron coated with sugars, many of these toxicity issues were resolved. 40,42 It should be noted that nanoparticles indicated for iron-replacement undergo vastly different approval procedures, by both the FDA and EMA, as they are nonbiological complex drugs; it is a widely held belief that additional factors, stemming from their colloidal and nanoparticle nature, need to be considered during their approval (e.g., manufacturing conditions). 41,44

| Nanoparticle/microparticle imaging agents
Alongside colloid-based iron-replacement therapies, similar iron-oxide nanoparticles are clinically approved as contrast agents for MRI (Table   1). 45,46 For imaging applications, the innate magnetic responsiveness of iron-oxide nanoparticles is used with MRI to generate contrast for imaging a variety of cancers and pathologies. 47,48 The combination of an iron-oxide nanoparticle's MRI responsiveness and small size, which facilitates preferential uptake in tumors, provides accurate and precise imaging of cancerous tissues. Interestingly, the majority of colloidal iron-oxide imaging agents have been discontinued in the United States and most of Europe. 13 In addition to MRI contrast enhancers, particles can be used as intravenous ultrasound enhancing agents. In these cases, particles typically take the form of micron-sized microbubbles. 49,50 These microbubbles provide a means to enhance contrast by stabilizing and encapsulating air bubbles, which are near-perfect reflectors of ultrasound and would otherwise rapidly dissolve in blood if not encapsulated/formulated. 49 Few of these products are approved and currently used in the clinic, for example, Definity (FDA approved) and SonoVue (EMA approved) are fluorocarbons or sulfur hexafluoride encased in lipid shells, respectively. Optison (FDA and EMA approved) is another ultrasound contrast agent formulated as human serum albumin encased perflutren.
3.4 | Nanoparticles for vaccines, anesthetics, fungal treatments, and macular degeneration Nanoparticles, or in these cases liposomes, are also used in a number of other clinical applications ( Table 1). The first of these is Diprivan, 51 which was FDA approved in 1989 as a general anesthetic. 52 Two vaccines, Epaxal for vaccination against hepatitis A 53 and Inflexal V for vaccination against influenza, 54 are liposomal systems that have been approved in many European countries. Interestingly, these two vaccines use their viral glycoprotein-liposomal template as the primary adjuvant, 55 with Epaxal doing so in lieu of traditional adjuvants such as  56 In doing so, toxicity is dramatically reduced as the pharmacokinetics and tissue distribution is improved via liposomal encapsulation. Furthermore, the liposomal formulation addresses a significant issue of the free drug form of amphotericin B, which is its insolubility in pH 7 saline. While not true liposomes, other FDA approved lipid-complexed formulations of amphotericin B exist, such as Abelcet and Amphotec. 57 Visudyne® is a light-activated liposomal formulation of verteporfin. Liposomal encapsulation offers enhanced uptake in proliferating cells which particularly enhances targeting and subsequent uptake by targets neovascular areas, which, following light stimulation damages the endothelium and blocks local blood vessels to prevent and treat neovascularization. 58

| C U R R E N T N A N O PA R T I C L E / M I C R O P A R T IC L E CL I N I C A L TR I AL S
Given the successes of many of these formulations in the clinic and commercial realm, significant efforts continue to explore currently approved nanomedicines as well as developing new ones. Here, we will: (a) briefly review the current clinical trial landscape for currently approved nanoparticles (Table 1), (b) review the current clinical trial landscape regarding cutting-edge nanoparticle formulations which are seeking approval (Table 2), and (c) highlight key technologies attempting to integrate targeting and stimuli-responsive functions into nanoparticle delivery systems.

| Previously approved nanoparticles
By seeking approval for additional indications, currently approved nanoparticle systems experience a more direct path to clinical approval as compared to a newer, developing, technology. This is because already approved nanoparticles have proven their safety and efficacy in humans and, if commercialized, likely meet good manufacturing practice (GMP) standards.

| Cancer nanoparticle medicine
As cancer nanomedicines were approved by the FDA over 20 years ago, it is not surprising that these currently approved nanoparticles are  It is unlikely that these approved products will resurface in the clinic given that the manufacturer no longer produces them, either for clinical or research purposes. However, ferumoxytol (Feraheme or Rienso), which is approved for iron-replacement therapies is broadly investigated for imaging applications in the clinic. Indeed, ferumoxytol is the most widely investigated iron-oxide particle with the majority of clinical trials focused on imaging of various cancers or other pathologies    JVRS-100's efficacy has been proven in preclinical studies as a means to stimulate the immune system to fight against leukemias. 81 Considering that nanoparticles are more often designed to avoid immune system and/or complement activation, the methodology behind delivery system is less intuitive; however, it is clear that JVRS-100 is designed to utilize key features of nanoparticles, notably their uptake by the reticulo-endothelial system and abilities to encapsulate and protect DNA, to act as an ideal system for immune system activation. Other systems ( 188 Re-BMEDA-liposome) are incorporating radioactive isotopes into liposomes as a means to target the delivery of the radionuclides to tumors. A number of other novel delivery nanoparticle systems are in clinical trials for cancer therapeutics such as targeted and stimuli responsive nanoparticles systems. These targeting and stimuli-responsive systems will be highlighted in the next sections.

| Applications other than cancer, iron-replacement, or imaging
Additional clinical trials are focused on testing nanoparticles for applications other than cancer, iron-replacement, or imaging (Table 3)

| Advanced nanoparticle systems
While the majority of nanoparticle delivery systems in clinical trials build on technologies that are long-established in their clinical utility (e.g., liposomes) or are already approved (e.g., Abraxane), some introduce aspects which are standard in academic and preclinical settings.
Here, we will highlight systems that are moving nontraditional clinical nanoparticle formulations forward; specifically, nanoparticle systems that leverage chemical-targeting or stimuli-responsive therapeutic release (Table 3).

| Targeted delivery systems
Interestingly, while nanoparticles and targeting antibodies are both approved for clinical use, systems combining these two technologies are lacking in both approved products and in clinical trials (Table 3). 89 Considering the substantial preclinical research efforts into, and the proven benefits of, actively targeted nanoparticle drug delivery systems, 24,25,38,90 this is somewhat surprising. Still, few technologies are being investigated in the clinic using chemically targeted means to enhance delivery of a number of therapeutics. Particular attention should be placed on Calando Pharmaceutical's CALAA01 siRNA formulation, which was the first example of a targeted nanoparticle formulation for siRNA delivery. 60,61 This system used a cyclodextrinbased particle which contained PEG, an siRNA designed to knockdown RRM2, and a transferrin receptor targeting protein, which when formulated together in a single nanoparticle was successful in achieving knockdown of the protein RRM2 (Figure 2ai). 60

| Stimuli-responsive nanoparticles (nonimaging applications)
The majority of stimuli-responsive nanoparticles are inorganic systems used for imaging applications (Table 3). However, inorganic nanoparticles made from materials such as gold or iron oxide can be  used for other stimuli responsive functions, such as thermal heating or magnetic control. AuroLase, a gold nanoparticle designed to absorb light to thermally ablate solid tumors, is in a current trial as a site-selective therapy for treatment of primary or metastatic lung tumors. Given that no active drug is used, and AuroLase is externally activated at the target site, this specific therapy has the potential for significantly reduced side effects. These efforts build on extensive preclinical testing, 98-100 with early clinical results pointing to excellent tolerability in humans. 101 Magnablate, an iron-oxide nanoparticle, is being developed for similar application, except that magnetic fields are used to stimulate the nanoparticles for thermal ablation. Hafnium oxide nanoparticles (NBTXR3), developed by Nanobiotix, utilize an external radiation source to enhance cell death at the radiation site via release of electrons. In preclinical animal models, NBTXR3 showed antitumor effects with similar to standard radiation therapies 102 and early clinical results suggest a good safety profile in humans as well as encouraging antitumor results. 103 As these therapies attack cancer cells via a physical mechanism, it is likely that they will benefit from synergistic pairings with other, more chemicalbased, treatments.
Nanoparticle systems based on organic component can also be designed to be stimuli-responsive. Building on the success of Doxil, Celsion Corporation is currently developing ThermoDox® as a temperature-sensitive version of liposomal doxorubicin. ThermoDox is a heat-sensitive liposome that release doxorubicin upon exposure to high temperatures (428C). Clinical trials for ThermoDox® highlight tolerability in breast cancer patients. 104 Additional clinical studies are underway for the treatment of hepatocellular carcinoma, breast cancer at the chest wall, and liver metastases (Table 3). In these studies, increased local temperature is achieved via microwave hypothermia, ultrasound, or radiofrequency thermal ablation. While these highlighted inorganic systems respond to external stimuli such as near infrared (NIR) lasers or magnets or radiation sources, other stimuliresponsive systems take advantage of unique or key biological conditions or situations to control drug release. Two other clinically investigated particles, LiPlaCis and RadProtect, respond to local biological cues to release their therapeutic. In the case of LiPlaCis, the liposomes degrade more rapidly in presence of phospholipase A2, 105 which is more abundant in tumor tissues 106  is not linked to uptake in a target cell, this method can be used to control the rate of amifostine release into the blood.

| T HE M A I N C HA L LE NG ES
Each individual nanoparticle formulation will face unique challenges in their clinical translation yet the majority of nanomedicines will encounter many of the same challenges. These challenges are biological, technological, and study design related. Here, we will focus on key challenges the majority of intravenous nanoparticle formulations face and how these challenges present unique issues from a clinical and translational point of view.

| Biological challenges
Biological challenges including modulating biodistribution or controlling passage of nanoparticles across biological barriers and into target cells limit the effectiveness of all nanoparticle formulations. As many preclinical studies and academic groups predominately focus on these challenges, we will not review them in detail, as it has been done previously. 11,[107][108][109] Here, we will focus on how clinically approved and clinically investigated nanomedicines are addressing these challenges and additionally highlight alternative clinical technologies (e.g., FDA-approved targeting antibodies) that complement nanoparticle delivery systems. Additional focus will be placed on how both differences in animal and human diseases, and human disease heterogeneity, influence preclinical and clinical nanomedicine efforts.

| Biodistribution modulation
One of the main challenges facing the clinical translation of nanomedicines is controlling their biological fate (e.g., increasing target site accumulation and decreasing off-target site accumulation). Many of the current approved and clinically investigated nanoparticles are PEGylated or PEG-terminated which limits interactions with, and rapid clearance by, immune cells. 10,37 In doing so, nanoparticles can remain in circulation for longer periods of time and increase their chances of reaching and entering target sites (e.g., tumors via EPR effect). 39,110,111 Similar effects can be achieved using hydrophilic sugar (e.g., dextran 10 ) coatings as in the case of many clinically approved iron-oxide nanoparticles. Another strategy used in the clinical (e.g., Abraxane) leverages biological proteins (e.g., human serum albumin), which likely extends the time before immune recognition as they naturally have a long circulation time. 112 Other than PEG, sugars, or serum proteins, very few other circulation-extending strategies have been able to successfully enter the clinic as an approved product, despite breakthroughs in preclinical academic studies. 113 Regarding nonapproved nanoparticles in clinical studies, one particular formulation (Cornell Dots) using integrin-targeting RGD-based peptides for increasing tumor accumulation for imaging applications, may point toward a nanoparticle future based on personalized medicine. Building on this, preclinical peptides are routinely discovered via phage displayed, especially other similar RGD-based peptides, 114,115 which may be a highly relevant strategy to discover either individually unique or disease-specific peptides in humans. 116 92 Unfortunately, nanoparticles not designed for imaging will find it difficult to achieve understanding of their biological fate. In these cases, other methods such as tissue biopsies (Figure 3d) 61 can be used to determine the extent of penetration across and into tumor or other pathological barriers. Overall, the majority of nanoparticles cannot be detected or analyzed in such a way; unfortunately, this will not only limit their analysis and interpretation of results, but it can also limit the translation of many would be successful therapies as it may not be possible to describe, and subsequently improve, clinical failures.  Figure   4a). 91,127,128 The most advanced of these strategies have proven to yield "hits" that have exhibited success in humans by optimizing ANSELMO AND MITRAGOTRI | 23 specific nanoparticle parameters (Figure 4b). 91  devices designed to mimic physiological tissues and conditions (e.g., organs-on-chips) may one day improve nanoparticle predictions of efficacy and performance. 129 Unfortunately, this remains a challenge even at the preclinical level where relevant estimates are typically generated from compartmental analyses or pharmacokinetic models. 130 As discussed above, correlation between human and animal data is essential. Early efforts to correlate animal data to human clinical data have shown agreement in some aspects (e.g., pharmacokinetics) but disagreement in other aspects (e.g., kidney toxicity). 124 As such, these differences and similarities must be considered on a case-by-case basis, however, efforts to correlate results should be implemented as soon as possible so general trends, if any, can be established.

| Study-design challenges
In contrast to the previously discussed issues are challenges relating to approval and study design in humans. Specifically, study size and the timing of nanoparticle therapies in a treatment regimen impact how results from clinical studies are perceived. As such, clinical results greatly influence future nanoparticle clinical studies; special attention must be given to ensure that clinical trials are designed to extract the most information regarding nanoparticle interactions, fate, and function while still testing key hypotheses.

| N 5 1 Clinical studies: Personalized medicine
As nanomedicine and personalized medicine efforts move forward, N 5 1 clinical studies will be required to move toward a system capa-ble of considering individual variability. 131 Many factors including genetic, environmental, and past and current treatments influence medicine efficacy. Perhaps surprisingly, many approved medications do not provide benefits to all who take them and this issue stems from original clinical study design that often overlooks trends in outliers which can affect a specific subset of patients. 131 Importantly, data regarding whether a patient either responds, or does not respond, to a given a treatment need to be collected, analyzed, and made available.
Nanomedicines provide a direct method to fine tune physicochemical properties on an individual basis that can tip the balance regarding a patient responding or not responding; however, to leverage these inherent advantages of nanoparticles, correlations between patients who either respond well or poorly and their previous medical history must be made. For example, prior medical history of all patients should be well-documented so as to determine groups which respond better to treatments in trials.

| Clinical trials as first-line therapies
The introduction of therapeutic nanoparticles in a treatment regimen during clinical trials is rarely, if ever, a first-line therapy. Typically, the only instances where this is the case is for established and approved nanoparticle systems (e.g., Doxil or Abraxane). Given that most of these therapies are not established, and in many cases their tolerability is not known, it is safest and most appropriate to investigate their efficacy as a last resort. As such, it is often the case that clinical trials are only available to patients who have stopped responding to treatments, as is the case with many cancer patients. While this represents a grand clinical challenge (i.e., treating or curing what is labeled a terminal disease), it may skew the potential of a nanoparticle therapy to benefit those who are likely still treatable.

| Studies extracting fundamental information
Empirical results are often the standard for preclinical studies. Indeed, it is similarly often the case that empirical results determine the commercial and clinical fate of nanoparticles in clinical trials; meaning, a binary result of improved survival or efficacy is enough to halt further investigations for a particular formulation. Indeed, lack of increased efficacy is unacceptable and in most of these cases this direction should be taken, however, these studies should be designed such that