From design to the clinic: practical guidelines for translating cardiovascular nanomedicine

Abstract

Cardiovascular diseases (CVD) account for nearly half of all deaths in Europe and almost 30% of global deaths. Despite the improved clinical management, cardiovascular mortality is predicted to rise in the next decades due to the increasing impact of aging, obesity, and diabetes. The goal of emerging cardiovascular nanomedicine is to reduce the burden of CVD using nanoscale medical products and devices. However, the development of novel multicomponent nano-sized products poses multiple technical, ethical, and regulatory challenges, which often obstruct their road to successful approval and use in clinical practice. This review discusses the rational design of nanoparticles, including safety considerations and regulatory issues, and highlights the steps needed to achieve efficient clinical translation of promising nanomedicinal products for cardiovascular applications.

1. Introduction

Cardiovascular nanomedicine aims to improve diagnosis and treatment of cardiovascular diseases (CVD), which are responsible for the majority of deaths worldwide.¹ Regarding diagnostics, the goal is to move the current imaging agents to a new level allowing the detection and characterization of CVD at an early stage. The therapeutic aim is to move forward from conventional drug therapies that lead to full systemic exposure to targeted drug delivery using nanosystems that minimize the systemic side-effects and enhance drug localization and efficacy in atherosclerotic and thrombotic lesions. With hybrid nanoparticles (so called ‘theranostics’) one could combine imaging and treatment, to enable monitoring patients’ responses to therapy.

The possible applications of nanoparticles in the management of CVD range from ultra-sensitive monitoring of cardiovascular markers, through detection and characterization of plaques and aneurysms, in situ detection of thrombosis, imaging of inflammation in myocardial infarction (MI), to the targeted delivery of atheroprotective, or thrombolytic drugs (Figure 1).^2–10 Cell labelling with nanoparticles for cell-based therapies can also be envisioned to enhance stent endothelialisation and improve myocardial regeneration.^14–16

Figure 1.

Possible applications of nanoparticles (NP) for diagnosis and therapy in CVD patients. The used photoacoustic and MR images are reproduced with permission from Blázquez R et al.¹¹; from Wu C et al.¹²; and from Yilmaz et al.¹³

However, bringing a medicinal product into the clinical arena is a challenging and time/cost-consuming process. Extensive in vitro and in vivo preclinical studies are required before first-in-man clinical safety trials can be initiated. In the USA, the Food and Drug Administration (FDA) is the responsible supervising agency to decide if a drug or a medical device is allowed to enter clinical trials and whether, upon completion of the clinical development programme, it will be approved for marketing. In Europe, the European Medicines Agency (EMA) is the regulatory body responsible for drug market approval. The classical process of drug development, testing, and approval is estimated to take around 10–15 years, with costs of roughly around 1 billion USD per product according to some estimates.¹⁷ Both the EMA and the FDA have stated that no new regulations are needed for approval or commercialization of nanomedicines, considering that the existing regulatory framework is valid and accurate. There is, however, a need for elaboration of the regulatory framework to accommodate for special safety and quality aspects that complex nanotechnology products can entail, and a need to improve technical guidance documents used for the application and implementation of existing regulatory frameworks.¹⁸

A standardized definition of ‘nanoparticle’ varies among organizations and countries. Although a specified size limit is not always relevant for scientific or medical applications, it is needed for regulatory purposes and is defined (albeit differently) both in the EU and in the USA. The definition implemented in the EU states that nanomaterial is a ‘natural, incidental, or manufactured material containing particles, in an unbound state, or as an aggregate, or as an agglomerate and where, for 50% or more of the particles in the number size distribution, one or more external dimensions is in the size range 1–100 nm’, but exceptions are possible to the percentage and the upper limit of 100 nm, especially in the pharmaceutical sector.¹⁹ The FDA defines nanomaterial as any material with at least one dimension smaller than 1000 nm and a nanoparticle as an object with all three external dimensions in the 1–100 nm size range.

Despite the costs and regulatory obstacles, about 250 nanomedicinal products, mostly for cancer treatment, were listed by FDA as approved, or were in different phases of clinical trials in 2013,²⁰ whereby the market is dominated by liposomal and polymeric nanomedicines.^17,20,21 Medicinal nanosystems in the form of e.g. liposomes (i.e. Doxil^®, AmBisome^®), or PEG-conjugated proteins (Adagen^®, Neulasta^®) have already been granted marketing authorization within the EU and the USA under the existing pharmaceutical legislation. As for any medicinal product, the authorities evaluate any marketing application by the established principles of benefit/risk analysis, rather than solely on the basis of the technology per se. The majority of nanosystems approved thus far were relatively simple and aimed at improving stability, half-life, bioavailability, and safety of existing drugs. It is likely that some new nanotechnology products that reach clinical trials will gain in complexity, as the technology gradually advances and treatment goals become ever more ambitious. According to the current EU directives, the decision whether a nanomedicinal product is a medicine or a medical device, which determines the applicable regulatory regime, is based on the principal mode of action. With medical devices, the mode of action is physical (mechanical or chemical) while a medicinal product acts by pharmacological, immunological, or metabolic means.²² However, future nanomedical products may span the regulatory boundaries between medicinal products and medical devices, and for those nanomedicines which have a complex mode of action, this decision may prove difficult as their activity might depend on both physicochemical/mechanical and pharmacological properties.^23,24

As compared with the vast number of experimental research reports focusing on cardiovascular applications of nanoparticles that have been published in the recent years (reviewed in Refs25–28), and the reported clinical trials remain very scarce in this field. This is likely due to the rational design of non-cytotoxic nanosystems and hurdles related to scale up, good manufacturing practice (GMP)-grade production, quality control, and full pre-clinical assessments which are required before clinical studies can be started (Figure 2). Based in particular on the experience from the European Commission funded NanoAthero project (‘Nanomedicine for target-specific imaging and treatment of atherothrombosis—Development and initial clinical feasibility’ http://www.nanoathero.eu/), this review addresses the main translational steps and challenges that cardiovascular nanomedicines encounter on their development from bench to bedside.

Figure 2.

Clinical translation scheme. In vitro studies on imaging and drug-delivery nanosystems produced in the laboratory scale represent the largest shelves in the pyramid. The number of nanomedicinal products reaching and passing the regulatory and toxicological hurdle to enter clinical trials remains very low.

2. Translation hurdle: clinical trials in cancer vs. CVD

The search for clinical trials on the homepage clinicaltrials.gov of the US government dealing with ‘nanoparticles’ and ‘cardiovascular diseases’ delivered 13 results, whereas the search for ‘nanoparticles’ and ‘cancer’ lists 176 performed or ongoing clinical trials. Although not all clinical trials can be found on clinicaltrials.gov, this indicates the considerable challenge in putting cardiovascular nanomedicines on the road to clinical trials when compared with anti-cancer nanodrugs. The 13 listed CVD trials include contrast agents for improved imaging of cardiovascular inflammation and enhanced diagnosis of acute coronary syndrome, nanodrugs for prevention of restenosis after revascularization, and plasmonic photo-thermal therapy (PPTT) of atherosclerotic plaques. Among the completed and published trials, several were related to the clinical use of iron oxide nanoparticles for improved detection and characterization of atherosclerotic plaques [ferumoxtran (Sinerem^®)^29–31] or aortic aneurysms [ferumoxtran (Sinerem^®)⁵] and detection of inflammation in MI [ferumoxytol (Feraheme^®)^13,32,33] Recently, in association with the NanoAthero project, a clinical study was done by the group of Stroes et al.,³⁴ investigating the utility of ferumoxytol for carotid plaque imaging. Concerning therapeutic application of nanoparticles in CVD, only a few studies have been reported so far. Some examples are thus briefly highlighted. In the BLAST trial, the safety and anti-restenotic efficacy of a single intravenous bolus of liposomal alendronate, which transiently modulates monocyte function, was examined in patients undergoing bare metal stent placement.³⁵ An angiographic assessment of late lumen loss at 6 months post-implantation demonstrated that the treatment effectively reduced the late loss in the inflammatory patient subgroup, but not in the entire liposomal treatment cohort. The NANOM first-in-man trial, published in 2015, investigated the feasibility of atherosclerosis treatment by reducing the total atheroma volume with PPTT.³⁶ Silica-gold nanoparticles with photothermal properties were delivered on bioengineered artery patch containing stem cells, or via an intravenous catheter under magnetic guidance, followed by irradiation using near-infrared laser.³⁶ This clinical trial showed that both forms of administration were superior to stenting and that photothermal destruction of atheroma tissue resulted in a reduction of the plaque volume down to 37.8% of initial plaque burden, whereas stenting resulted in 52.9% reduction of plaque burden.³⁶ It is however questionable, to what extent this technique may prove applicable in the clinical routine everywhere. Within the NanoAthero project, a clinical trial was also performed to test the treatment of atherosclerotic plaques with a targeted nanomedicine containing prednisolone (Nanocort^®). In that study, prednisolone was encapsulated into liposomal nanoparticles, which increased the plasma half-life of the drug.³⁷ After systemic intravenous infusion in an antecubital vein, it was demonstrated that the nanoparticles were localized in the macrophages isolated from atherosclerotic plaques harvested from the iliac arteries, thereby lending proof-of-concept that intravenous liposomes can successfully target inflammatory cells within the atheroma. The clinical studies did not demonstrate that the delivered prednisolone have an anti-inflammatory effect measured by positron emission tomography (PET) imaging in the atherosclerotic lesions,³⁸ which might be related to in vitro observations that macrophages become lipotoxic, exemplified by enhanced lipid loading, ER stress, and apoptosis.³⁹ Summarizing the previous clinical trials, it is clear that the application of nanomedicine in CVD patients is still in its infancy and great effort will likely be needed to enforce a clinical breakthrough.

3. Rational design of nanosystems: safety by design

Imaging is a crucial aspect for risk stratification as well as for the selection of subsequent therapy and follow-up monitoring. Nanoparticulate contrast agents have been shown to improve the detection and characterization of CVD, but their application in potentially healthy subjects raises a particularly high-safety hurdle. Nanoparticle-based drug delivery systems are an attractive platform to improve the efficacy and reduce the systemic toxicity of cardiovascular drugs. For therapeutic applications, novel nanoparticle formulations including drug-carrying liposomes, lipid nanoparticles, superparamagnetic iron oxide nanoparticles (SPIONs), and polyacrylates are currently being developed by our groups and others.^37,40–45 Additional modifications, including functionalization with targeting ligands and/or imaging agents are often necessary to locally deliver the therapeutic nanoparticles and to monitor the effect of the treatment. This can result in a complex multicomponent nanosystem (Figure 3), which will require many synthesis and manufacturing steps as well as a range of quality controls, which leads to a tremendous cost increase.⁴⁶

Figure 3.

Example design of nanoparticles to achieve a targeted drug delivery and/or imaging agent. Due to the small size of nanoparticles, their surface area is large. This provides various possibilities of surface modifications e.g. by coating in order to stabilize and prevent aggregation of nanoparticles, and to allow conjugation of ligands or drugs.

A more rational design of the particles may partly help to reduce costs and increase chances of achieving successful clinical translation. Setting aims, i.e. selecting the disease process to be addressed and the intended application of nanosystem (diagnostics vs. therapeutic) is the first step in the development process. The design and production at the lab scale must be followed by a complete physicochemical characterization of nanosystems, using the available specifically adapted evaluation methods.^47,48 As different techniques for nanoparticle characterization are available (e.g. transmission electron microscopy, Fourier transform infrared spectroscopy, and dynamic light scattering), each of them featuring its own advantages and limitations, the characterization data obtained with several different measurement methods should be compared with ensure reliable results.

In-house physicochemical characterization and storage stability evaluation of produced batches limits the costs in the early development stage. Within the NanoAthero project, nanoparticle characterization and stability evaluation were performed in parallel in-house by nanoparticle providers and also by a selected independent partner, in order to demonstrate comparability and allow for reproducibility validation. In this respect, the Nanotechnology Characterization Laboratory (NCL) in the US (https://ncl.cancer.gov/), and more recently its European counterpart, European Nanomedicine Characterization Laboratory (EU-NCL, http://www.euncl.eu/), provide independent trans-disciplinary testing infrastructures covering a large set of preclinical validated characterization assays (physical, chemical, in vitro, and in vivo biological testing).

Among the parameters to consider when designing a nanosystem is the chemical composition, which is often the most critical feature that affects nanoparticle toxicity.^49,50 Further, particle surface charge, indicated by zeta potential, has a strong influence on nanoparticle stability in suspension and in vivo toxicity.^51,52 Size is another critical factor that affects the behaviour and biological safety of nanoparticles.⁵³ For example, nanoparticles with hydrodynamic diameter smaller than 10 nm have been reported to cause undesirable effects by passing through the blood–brain barrier, and nanoparticles with diameter less than 5 nm are rapidly cleared by the kidneys, which dramatically reduces their circulation time.⁵² Particle size and shape are also likely to affect their margination, extravasation, and penetration through vascular walls, particularly in larger vessels relevant to CVD (reviewed in Refs^54,55). Previous ex vivo studies in whole blood model showed enhanced margination of micro- compared with nano-sized particles and the dependence of this effect on a high-aspect ratio of particles.^56,57 In these investigations, nanorods did not display enhanced margination compared with nanospheres,⁵⁷ indicating that binding of nanoparticles to arterial endothelium may require a margination-enhancing design and/or active targeting. Nanoparticle agglomeration is another factor with strong adverse consequences in vivo.⁵⁸ Agglomeration is influenced by the particle composition, size, and zeta potential, but also extrinsic factors, e.g. temperature, as well as pH, osmotic strength, and the presence of serum. As aggregated nanoparticles are no longer nano-sized, they undergo a rapid recognition by the reticuloendothelial system (RES) and are cleared by the liver or spleen. Moreover, their presence in the circulation may cause serious undesirable side-effects, such as clogging blood or lymphatic vessels.⁵⁹ Prevention of agglomeration is therefore required for designing a stable, clinically safe nanosystem. In this respect, PEGylation of nanoparticles appears effective in reducing their agglomeration. By creating a hydrophilic layer around the nanoparticles, PEGylation also provides a strong steric barrier to opsonin adsorption,⁶⁰ opposing nanoparticle recognition by the RES, and increasing their circulation half-life. Other methods to reduce particle agglomeration explored within the NanoAthero project included coating of SPIONs with cross-linked dextran or fatty acids^44,61 and brush-like coating of polymer nanoparticles with polysaccharides (dextran and fucoidan⁸). Careful attention should also be given to the protein corona which forms on the surface of the nanoparticles when they interact with plasma, since this can affect their toxicity and efficacy.^62,63 Taken together, detailed and standardized characterization can facilitate the prediction of nanoparticle performance in physiological conditions and is mandatory to consider before any given nanosystem can enter the preclinical in vitro and in vivo testing stages.

4. Candidate selection: a multi-criteria decision process

In the selection of the best candidate nanosystems for imaging and therapy of atherothrombosis many factors should be carefully considered. Within the NanoAthero project, a decision tree was established based on the physical and biochemical characteristics of the nanosystems developed in this project (Figure 4). The most important selection criteria are briefly outlined below with short commentaries.

Figure 4.

Candidate selection criteria. ADME, absorption, distribution, metabolism, and elimination; GRAS, generally recognized as safe; PDI, polydispersity index.

4.1 Product physicochemistry

Physicochemical properties of the nanoparticles are the critical determinants of their safety and in vivo performance. Intensive efforts are being developed by the NCL and EU-NCL initiatives, as well as by different research groups, including the NanoAthero consortium, to propose standardized methods for measurements of key parameters such as particle diameter and zeta potential.^64–66 Apart from the parameters listed in the Section 3 (diameter, charge, and polydispersibility), pH and osmolarity of the final dispersion for long-term storage and for injection should be considered when selecting suitable candidates. In case of nanosystems containing drugs or contrast agents, encapsulation/binding efficiency and the amount of drug or contrast agent per particle are the important selection parameters.

4.2 Ingredient quality (safety, sterilisability, and pyrogen content)

The quality of the starting materials is an important point to be considered. Preferably, raw materials with existing pharmacopoeia reference (i.e. Ph.Eur., USP) or medical-grade substances should be used for nanoparticle synthesis. The safety of ingredients can be confirmed using the GRAS (Generally Recognized as Safe) Substances Database (https://www.accessdata.fda.gov/scripts/fdcc/? set=SCOGS).

The final drug product should obviously be sterile. Sterilisability of the produced nanoparticles must therefore be ensured, which is in most cases achieved with (redundant) sterile filtration through <0.2 μm filters right before filling into the sterile dosage units. Depending on the chemical/biological components and the production process, the final nanosystem may contain bacterial endotoxins,⁶⁷ which can cause adverse effects upon in vivo administration, potentially leading to the organ damage. The FDA-recommended high-sensitivity bacterial endotoxin LAL test (limulus amoebocyte lysate assay) is commonly used in preclinical pharmaceutical development, but many nanoparticles interfere with the assay.^67,68 In our project, there was a case of endotoxin contamination of an additive, the commercially available bovine serum albumin used as a coating to improve the biocompatibility of one type of SPIONs. This resulted in unexpected inflammatory effects of these particles and additionally necessitated a complete and very costly purification of the synthesis unit to avoid cross-contamination. To overcome this problem, clinical grade serum albumin of human origin was used for further development of these particles.

4.3 Manufacturability (process, cost of goods)

The manufacturing process of a nanomedicinal product may involve a multi-step procedure requiring a number of excipients, which can drastically increase the cost and represents an additional production hurdle. The research and development methods often involve a low-volume production and scaling up the process may entail serious difficulties for some nanoparticles, and be easier for others.⁶⁹ For instance, scaled-up production of lipid nanoparticles is relatively easy to implement, and has been documented for more than 25 years in the medical field.⁷⁰ Apart from this, the costs and availability of raw materials must be considered, as well as the batch-to-batch reproducibility of physicochemical characteristics.

4.4 Stability

Although it is not strictly required for human application, long-term stability on storage of nanosytems is a prerequisite for a nanoproduct to be marketable. Ideally, the shelf life should be equal to or longer than 6 months. The parameters to consider include colloidal stability and chemical stability of drugs and excipients on storage, but also potential leakage of drug or contrast agent from the nanoparticles. Within NanoAthero, the standardized physicochemical characterization of nanosystems was performed at 1 month post-preparation date and—to determine the long-term particle stability—the subsequent measurements were performed after 3, 6, and 12 months of storage at 4°C in the respective nanoparticle dilution media. The acceptable variation was set to 10% diameter variation, and 20% polydispersity index variation at PDI of maximally 0.25.

4.5. Toxicity/biocompatibility

One of the factors that may influence the particle behaviour and toxicity is stability in biological fluids (serum-containing cell culture media, plasma, whole blood).⁷¹ Analysis of nanoparticle agglomeration in plasma and blood is therefore mandatory. The screening of nanosystems should first be done in vitro, to assess the potential toxicity of nanoparticles towards blood cells, other first-contact cells (e.g. endothelial cells in the case of intravenous application) and the actual target cells. Other undesired effects, including haemolytic reactions, platelet, and complement activation, reactive oxygen species production can relatively quickly be evaluated by in vitro tests.^44,72 After in vitro screening to select the constructs with adequate haemo- and biocompatibility, proof of principle studies in in vivo models and GMP-compliant manufacturing process are required, that are followed by regulatory toxicity studies in animals, usually rats and mice (see Section 7.2).⁷³

4.6. Efficacy

To some extent, the in vivo performance and potential efficacy of nanosystems can be predicted with in vitro or ex vivo models or phantoms,^44,74,75 but the ultimate preclinical proof of efficacy requires an animal model of disease. In the NanoAthero project, the characterized nanosystems containing imaging agents (radionuclides, iron oxides, micellar formulations containing gadolinium) were tested in appropriate animal models: mouse or rabbit models of atherosclerosis and a rat model of thrombosis.⁸ Dedicated small animal magnetic resonance imaging (MRI) coils and a 3T MRI system were used for imaging. To verify the accumulation of the nanosystems in the diseased region, histological analysis of the imaged sections was performed post-mortem. Single photon emission computed tomography/computed tomography (SPECT/CT) and PET analyses, including the grafting of tracers was also performed, as well as in vivo and ex vivo fluorescence imaging after nanoparticle labelling.

For therapeutic purposes, the nanosystems containing compounds with anti-inflammatory activity were pre-screened in vitro⁷⁴ and then tested in the apolipoprotein-E (apoE)-knockout mouse model,⁴¹ followed by selection of promising candidates.

4.7. Pharmacokinetics and biodistribution

The determination of pharmacokinetics (PK) and biodistribution is usually done by the detection of particle- and/or drug-bound radiolabels in animal tissues harvested at different time points. Radiolabelling (³H or other radionuclides) allows imaging of the biodistribution of nanoparticles on tissue sections in rodents and the quantification of the percent of injected dose in different organs and body fluids. Full-body autoradiography or selected tissues sampling for well counting should identify the main-targeted organs (usually liver, kidney, and spleen). Biodistribution estimation derived from in vivo nuclear imaging using nanoparticles labelled with gamma- or positron-emitters radionuclides is an alternative. While the accuracy of the measurement is lower when compared with direct tissue sampling, this approach allows for iterative assessments in a single animal, enabling either a marked reduction of the number of animals sacrificed for a given experimental protocol, and/or an increase in the number of measurements.⁷⁶ Fluorescence imaging, having the advantages of lower cost, detection below cellular level by microscopy techniques, is another alternative, but its main limitation is that it is not truly quantitative (semi-quantitative analysis).

In case of nanoparticles containing a drug payload, the drug can be labelled with ¹⁴C. By using dual ¹⁴C and ³H-detection, parallel quantification and comparison of the biodistribution between the free drug and nanoparticle-conjugated drug is possible.^45,77Ex vivo validation by autoradiography allows localizing a radioactive material within particular tissues or cells with high sensitivity and quantitative estimation of the delivered amount of (nano)drug.^78,79 These data are of critical importance to determine the ability of nanoparticles to target and deliver their drug payload to particular tissues, but the expense and the efforts required for these investigations are considerable.

4.8. Clinical « acceptability »

Novel nanodrugs are commonly greeted with a degree of concern and reserve in fear of their potential nanotoxicity, unless the nanosystem carrier is well established and/or carries an approved drug. Acceptance is usually less of an issue in high-medical-need indications.⁸⁰ Quite obviously also the administration route is a factor of importance, whereby oral administration is preferred by patients.⁸¹

However, parenteral administration and in particular intravenous injection is often the only feasible way cardiovascular contrast agents and nanodrugs should be given, which requires admission of a patient to a hospital or outpatient clinic, and significantly increases the costs.

5. In vitro proof of safety and efficacy

Nanomedicine offers unique possibilities in terms of CVD management, but despite these exciting possibilities, it is clear that nanomedicines can also entail new and sometimes unforeseen risks. The impact of nanoparticles on biological pathways and their toxic effects on the human body can be difficult to predict. Due to the interference of the particles with the traditional photometric cytotoxicity assays, routinely used tests such as lactate dehydrogenase assay or 3-(4, 5dimethylthiazol-2-yl)2, 5diphenyltetrazolium bromide assay can produce false-positive or false-negative results.^82,83 Toxicity of engineered nanoparticles can be over- or underestimated due to their influence on absorbance of light in the visible spectrum, quenching of fluorescence, or even adsorption of the dye to their surface.⁸³ Suitable in vitro assays must thus be chosen and validated to enable a meaningful in vitro toxicity evaluation. Several organizations, including the NCL or the International Organization for Standardization underscore the importance of a general standardization of in vitro toxicity assessment within nanotechnology and nanomedicine.⁴⁷ Here, an essential point is the batch-to-batch reproducibility, because safety or efficacy evaluations are not reliable if batch-to-batch reproducibility is insufficient.

Full biocompatibility (including haemo-, cyto-, and immune compatibility) of the nanosystems is absolutely essential, as the target population of CVD patients may be prone to critical responses to any incompatibility. Therefore, developing a systematic workflow to analyse the biological effects of nanoparticles under standardized conditions is particularly relevant. All nanosystems intended for intravascular administration should be tested for their potential toxicity towards primary human endothelial cells. Using two complementary methods for long-term in vitro monitoring in parallel is recommended, as one single method may increase the risk of bias. In our opinion, real-time cell analysis and live-cell microscopy represent the suitable methods for parallel testing of the toxicity of nanoparticles in static in vitro conditions.⁶⁶ Importantly, no interference resulting from the presence of nanoparticles should be detectable by real-time cell analysis in the absence of cells, which was indeed the case in our studies. This clearly underscores the suitability of the techniques, we used for the future standardized nanotoxicology studies. Beyond analysis of nanoparticle effects on cell viability in static culture conditions, investigating the effects of circulating nanoparticles on endothelial monolayer under physiological-like shear stress conditions allows performing the in vitro assays in dynamic conditions corresponding to the physiological environment of endothelial cells.⁶⁶ Among the nanosystems evaluated positively in cell-compatibility studies, selected candidates should undergo detailed analyses to exclude haemolysis, coagulation, platelet activation and aggregation, leucocyte activation, and complement activation.⁴⁴

Concerning possible compounds in evaluation for therapeutic applications, an in vitro screening setup for selected promising compounds/formulations and their potential athero-protective effects should be established. For instance, within NanoAthero we selected a range of in vitro assessments that address several pivotal pathological pathways in atherosclerotic plaques. These assays revealed that pterostilbene, simvastatin, and the liver X receptor agonist T0901317 were the most promising atheroprotective compounds to be integrated into nanosystems for plaque therapy⁷⁴ and three simvastatin-loaded nanocarriers, including high-density lipoprotein nanoparticles, PEGylated liposomes, and polymeric micelles, were subsequently evaluated in vivo, in apoE-deficient mice.⁴¹

6. Preclinical animal models

Multiple animal models are available that address CVD in different species, including rodents, and larger animals (rabbits, pigs, and non-human primates).^84,85 To date, genetically engineered hyperlipidaemic mice are among the most widely used models of atherosclerosis, but several transgenic,⁸⁷ models in alternative species (rat and pig) have also been created.^86,87 While mouse models of atherosclerosis are inexpensive and highly valued as a tool to identify the molecular mechanisms of the disease that can be targeted by novel (nano)medicines, they lack multifactorial background of atherosclerosis and have limited predictive value as the lipid profile and metabolism of mice, as well as the plaque composition are different from humans.⁸⁸ Additional drawback is the small size of these animals, which limits the availability of biological samples, as well as the possibility of morphological and functional imaging of atherosclerosis. Despite of these drawbacks, mice still represent a model of choice for initial drug testing or biodistribution studies, and the continuing efforts to develop transgenic models, e.g. apoE3Leiden/cholesteryl ester transfer protein (CETP) mice,⁸⁹ aim at better reproduction of human disease characteristics.

To promote clinical translation of emerging nanomedicinal products, larger animal models suitable for interventional procedures and imaging are advantageous. Rabbits represent a cost-efficient model of atherosclerosis with similarities to human lipoprotein profile, CETP expression, and size large enough to allow tissue sampling and imaging in clinically used scanners. The limitation of the rabbit model is that lesion complications observed in humans (haemorrhage, ulcerations, and thrombosis) are usually absent and their foam cell and macrophage load is increased compared with human plaques. Pigs and non-human primates represent two atherosclerosis models considered optimal to reflect the disease in patients, because of their similarities to humans in terms of metabolism, cardiovascular anatomy, and physiology. Human-like complex plaque morphology⁹⁰ and instability traits have been reported in these animals.^91,92 Despite the disadvantages of the large animal models, including the great expense and ethical considerations, these models allow the best extrapolation of findings to humans, thus contributing to the development of emerging therapies.

Detailed recommendations on design and performing animal studies in common models of atherosclerosis have been recently published in a statement of American Heart Association.⁸⁵

7. In vivo safety: a prerequisite for approval

Toxicology assessment of nanomedicines in vivo is in principle not very different from conventional drug products, albeit that specific potential nanomedicine-related safety issues in humans need to be looked for in special animal models. These are listed below.

7.1. Complement activation-related pseudoallergy assessment

To characterize, predict, and prevent pseudoallergic reactions to nanomedicines, which often arise following their first intravenous administration, EMA recommends the detection of the Complement activation-related pseudoallergy (CARPA). The unique in vivo porcine model of CARPA allows evaluation of the risk of—otherwise unpredictable—acute cardiopulmonary distress, which can be severe or occasionally lethal, and therefore, unacceptable for CVD patients.⁹³ The CARPA tests in pigs should include both single dose and repeat-dose administration, corresponding to the predicted use of the final nanosystem.⁹⁴ The candidates that passed the CARPA evaluation successfully without inducing hypersensitivity reaction (i.e. were CARPA-negative) can subsequently enter the regulatory toxicity studies.^44,72

Additionally, many nanomedicines undergoing development or approved as products include a coating to improve stability, minimize aggregation, and prolong circulation time. The presence of coating has the potential to impact on bio-molecular and cellular interactions of nanoparticles upon in vivo administration. For example, naturally occurring anti-PEG antibodies (IgM) have been detected in nearly 25% of healthy donors with no known exposure to PEG, indicating a growing prevalence of PEG exposure (e.g. in cosmetics or processed foods) and an increased risk of immunogenicity/antigenicity.⁹⁵ Anti-PEG antibodies may also lead to increased clearance of PEGylated nanomedicines upon administration, thus reducing their biological activity. Although no specific animal models have been recommended for testing PEG antigenicity, it is important to monitor the patients for the presence of anti-PEG antibodies prior to and during the administration of PEGylated nanomedicines.⁹⁶

7.2. Regulatory toxicity studies

Prior to clinical trials, the authorities require preclinical safety and PK evaluation in animals under good laboratory practice (GLP) regulations. As such, the regulatory toxicity studies are commonly outsourced to an approved and fully equipped Contract Research Organization (CRO). The non-clinical safety assessment for marketing approval of a pharmaceutical usually includes pharmacology studies, general toxicity studies, toxicokinetic (TK), and non-clinical pharmacokinetic studies, reproduction toxicity studies and genotoxicity studies. The non-clinical safety studies should be adequate to characterize potential adverse effects that might occur under the conditions of the clinical trial to be supported. The choice of the more adequate panel of studies to address safety of novel nanosystems should be guided mainly by the dosage expected to be used in humans and also by the specific characteristics of the nanoparticles.

The following toxicology tests are the main ones required before nanomedicinal product trials in humans: (i) safety pharmacology, a core battery according to ICHS7A, ICHS7B including the assessment of effects on cardiovascular, respiratory and central nervous systems (QT prolongation, respiratory function, and Irwin Test); (ii) TK and PK studies to determine plasma PK and elimination, as well as the validation of analysis methods in relevant species for repeated dose studies; (iii) acute toxicity studies, mainly based on single dose or expanded acute toxicity studies in two mammalian non-primate species. Animals are monitored over 14 days for body weight, organ weight indices, as well as behavioural, biochemical, and histopathological changes. The maximum tolerated dose and the no observed adverse effect level should be obtained; (iv) repeated dose toxicity studies over 2 weeks (minimum duration), in two mammalian species (rodent and non-rodent); (v) local tolerance studies in rabbit, using routes relevant to the proposed clinical administration route; (vi) genotoxicity studies including gene mutation (Ames Test) and chromosomal damage test (human lymphocytes). The requirement to execute this complete panel of tests is related to the effective dose that it is supposed to be used. For instance, not all these tests are required for a PET/SPECT nanosystem for imaging using microdoses and a single injection.

More specific studies should be taken into account with respect to different peculiarity of the investigated nanosystems (i.e. iron determination for iron-based nanosystems, rate and location of a drug released by a liposome system, etc.). All the above mentioned studies are necessary for the evaluation of benefits and risks for patients, but due to the large costs of procedures, often constitute a first major financial hurdle for a given nanoproduct.

8. Production scale-up and GMP-compliant synthesis

Another major hurdle to overcome in the process of approval of a nanomedicine for clinical use relates to its scale-up and production under GMP regulation. At many academic institutes, adequate facilities and expertise for scaled-up production and manufacturing under GMP are lacking, and therefore, these activities need to be outsourced to a fully licensed manufacturer capable of handling nanomedicinal products. Very often significant pharmaceutical development has to be done before a process can be scaled up and brought under GMP. A major issue with nanomedicinal products is their sterilization, where one is mostly condemned to sterile filtration through 0.2 μm filters or—if particles are around or larger than 200 nm—needs to implement an aseptic manufacturing method, which comes with its own challenges.

Besides a robust manufacturing process, also the quality control, which includes the release specifications of the product and the implementation of the full set of characterization assays has to be prepared. All assays have to be verified or qualified before GMP manufacturing. Finally one must ensure that containers, closures and packaging material are of the right quality and fully compatible with the product.

9. Preparation of regulatory dossiers for local/national authorizations of clinical trials involving nanomedicines

For all pharmaceutical/medicinal products, non-clinical and clinical information has to be compiled in the format of an Investigational Medicinal Product Dossier (IMPD) and an Investigator’s Brochure (see below). This documentation is required for clinical trial approval, as well as for the final product dossier and usually requires the specific expertise of a dedicated academic or industrial CRO.

In order to inject any nanomedicinal product (ranging from a macromolecular assembly to a complex nanoparticulate structure loaded with drug and/or contrast agent) in humans, several steps have to be followed, in accordance with the specific guidelines (manufacture of sterile medicines, manufacture of experimental drugs and manufacture of radiopharmaceuticals). The first step involves the IMPD preparation. This document compiles all information related to the drug substance (Part S) and the investigational medical product under test (Part P). The drug substance can be a natural or a synthetic compound, and the product is the nanoformulation of the drug.

Part S describes (i) the origin and the structure of the drug substance, (ii) its manufacturing process and process controls, (iii) the control of materials and critical steps, (iv) its composition and the impurities, (v) the full control of the drug substance (specifications, analytical procedures, validation of analytical procedures, batch analyses, and justification of specifications), (vi) the container closure system, and (vii) the stability under long term and accelerated storage conditions. Part P describes the nanoformulation of the drug and the pharmaceutical development, as well as the same information as in Part S except an additional specific control of excipients. For the development of a radiopharmaceutical for PET or SPECT imaging-based diagnostics, an additional IMPD Part P has to be completed. The reason for this is that the cold nanoformulation of the drug described in the IMPD Parts S and P is not the final product to be injected into humans, so that the final formulation with the added radionuclide is considered as a new medicinal product under test.

The second document to be completed is the Investigator’s Brochure composed of five chapters that assemble all the non-clinical and clinical information available about the investigational product that is relevant in the outlook of administration to human subjects. The first chapter is a short review that deals with the biological properties of the medical product and their effects in humans in accordance with the medical indications. The second chapter is a summary of the main results of the IMPD Part S and P. The third chapter contains a scientific description of all the preclinical results (in vitro and in vivo pharmacology; biodistribution, PK, and dosimetry, if necessary). In addition, a special focus is placed on a battery of toxicology studies: acute oral toxicology, extended single dose toxicity, CARPA, and genotoxicity. The fourth chapter compiles all the data obtained previously in humans with the use of the investigated medical product. The last chapter provides the investigators with a guidance summarizing the information essential for a clinical study (therapeutic indications, posology and administration route, contra-indications, special warnings, and reference safety information).

Finally, the interventional research protocol (study protocol) and a dedicated document dealing with information to be specified for clinical trials on first administration in humans (if necessary) have to be prepared. The study protocol should indicate the scientific justification for the trial, the objectives, a description of the trial, the procedure of the trial and eligibility criteria, the treatment administered to study participants, and the efficacy assessment. Furthermore, this document must contain information regarding regulatory issues such as specific committees for the trial, safety assessment (risks and restrictions added by the study), data management, statistical aspects, quality control and insurance, ethical and legal considerations, as well as funding and insurance issues.

The national competent authority issues a clinical trial authorization upon reviewing study protocol and the Investigator’s Brochure. The content and format of the protocol must comply with Community guideline on Good Clinical Practice (CPMP/ICH/135/95). In parallel, the subject information leaflet and informed consent form are prepared, informing the patients on the nature, scope and possible consequences of the study. These must be in language and terms understandable to the participants. Local authorization of clinical trials is obtained through institutional ethical committees based on the submitted study protocols, the case report forms, subject information leaflet, and informed consent forms, according to the National and European legislation (Helsinki, National codes of Public Health, the principles GCP, Bioethics law, European Dir. 95/46/CE) (Figure 5).

Figure 5.

List of regulatory documents to prepare and submit for national authorization of Phase 1 clinical trials.

10. Clinical adoption challenges

The implementation of new technologies in healthcare faces multiple challenges, including institutional interests, availability of appropriate infrastructure and clinical skills for preparing patients and their treatment, the administration and the monitoring of treatment outcomes, the enrolment of patients in the clinical trials and the evaluation and acceptance endpoints,⁹⁷ and last but certainly not least, country-specific reimbursement structures and affordability. Thus, recent experience with the biologicals (PCSK9-antibodies) has taught that even highly effective and safe interventions⁹⁸ will be implemented at a low pace, if the price is considered to be out of balance with the offered advantages. Studies in several EU member states concerning the attitudes of the public to nanomedicine revealed a global support, because nanotechnology in medicine is expected to bring medical progress, but the potential for safety risks is also often cited.^99,100 However, the perception of risks differs very significantly between the use of industrial nanomaterials (generally of inorganic nature) and the use of medical nanomaterials. Despite the fact that the regulation, control and approval between the two categories of nanomaterials are very different, with an incomparably more stringent regulation in nanomedicine, the implementation of technologies that involve significant use of manufactured nanoparticles may face resistance from those patients who perceive nanotechnologies to be associated with unseen future risks.¹⁰¹ This implies a need to engage with the public and especially with patients’ organizations as the introduction of nanomedicinal products proceeds.¹⁰² Interaction should be sought already at an early stage with relevant patients’ organizations and also with medical staff who would eventually become the end users of the nanomedicines. The contact to patients' organizations can be established before or during designing clinical trials, but care is needed to ensure that the safety aspects are sensitively handled, and that sufficient support and information is provided to patients and carers.

To create a platform for dissemination among the patients’ communities and the general public, within the scope of the NanoAthero project Edinetics Ltd. developed the Democs card game entitled ‘Nanomedicine for Atherosclerosis’, which provides an inexpensive and entertaining way to engage, inform and discuss the benefits and risks of novel nanomedicines with a broader public.

11. Summary and conclusions

The potential clinical impact of nanotechnology in terms of CVD diagnosis, management and risk assessment to ultimately reduce the global disease burden cannot be overestimated. The translation of basic studies into clinical trials clearly represents the biggest challenge in this field, because developing and bringing a novel nanomedical product to the clinic is a process that involves multidisciplinary efforts of biologists, chemists, pharmacists, bio-engineers and clinicians (Figure 6), and requires strong expertise in safety issues, healthcare structures, GMP-compliant production and marketing.

Figure 6.

Multidisciplinarity in nanomedicine. Bringing the nanomedicinal product into the clinic requires co-operative efforts of experts from different areas of science, technology, healthcare, and industry.

Whereas about 20% of the approved nanodrugs are indicated for the treatment of cancer, according to current estimates,^17,20 cardiovascular nanomedicines represent about 1% of the market. While approved anti-cancer nanomedicines generally alter the toxicological profile of the encapsulated drugs in patients, they do not really enhance local antitumor efficacy, which allegedly results from poor, erratic and heterogenous drug delivery in tumour tissues.¹⁰³ Indeed, given the high medical need, the translational hurdle is generally lower for anti-cancer nanomedicines, but the absence of a real breakthrough in terms of improved drug delivery and anti-tumour effect may slow down the momentum in the development of nanomedicines for other indications. It is clear that the clinical relevance of nanomedicine, both in oncology and cardiology, will depend on rational design of particles for which sufficient delivery to target tissues can be ensured.¹⁰³ For this purpose, extensive fundamental studies on nanoparticle interactions with vascular endothelium in the presence of blood cells will be essential to determine the relationship between physicochemical properties of nanoparticles and their delivery efficiency. In terms of safety, cardiovascular nanomedicine is further expected to benefit from standardized definitions and clear guidelines, but also from reliable, interference-free assays serving as nanotoxicity screening tools.

Addressing the key steps in the process of nanomedicinal product translation (Tables 1and2), this article intends to help researchers and clinicians better understand the development hurdles and regulatory requirements concerning new (nano)medicines, highlighting the tension that exists between these complexities on the one hand and the feasibility and affordability desired by the cardiovascular clinical arena on the other hand. As many early-stage innovative nanomedicine development efforts take place within academia, beyond the large R&D budgets of the pharmaceutical industry, there can only be hope that more funding options become available in the field of CVD to perform systematic basic studies concerning the mechanisms of nanoparticle transport, their interactions with cells and disease targeting efficacy that should in the future guide their improved design. Large scale funding and/or pharmaceutical industry investments will be necessary to help promising new nanomedicinal drug products reach the clinical stage in which proof of efficacy and added therapeutic benefit can really be shown in patients.

Table 1.

Translation checklist

Development step	Performed (Y/N)	Qualified (Y/N)
Characterization and physicochemical evaluation
Size and charge analysis
Dispersibility analysis in complex media
Degradation products analysis
Stability and shelf life
Costs analysis
Manufacturability and scale-up
Sterisability, pyrogen content check
In vitro safety
Target cell response analysis
First-contact cell response
Haemolytic response
Immune response
Thrombogenicity analysis
In vivo evaluation
In vivo efficacy in appropriate animal model
CARPA
Biodistribution
Regulatory toxicology and PK
GMP-compliant production
Scale-up
GMP synthesis
Approval for clinical use
IMPD, preparation of regulatory dossiers
Ethical approval
Evaluation of clinical adoption readiness

CARPA, complement activation-related pseudoallergy; GMP, good manufacturing practice; IMPD, investigational medicinal product dossier; PK, pharmacokinetics..

Table 2.

Barriers to translation and possible mitigation steps

Barrier to translation	Mitigation steps
Approval for novel nanomaterials/nanoparticle use in humans is challenging compared with small molecules	A clear rationale is required demonstrating the advantages of novel nanoparticles over established diagnostic/therapeutic agents, e.g. site-directed drug delivery, thus ensuring maximal therapeutic outcome while minimizing potentially negative systemic side-effects or the ability to monitor drug release/efficacy through a nanoparticle carrier plus imaging modality (theranostics). Designing nanoparticles with materials that have a history of use in humans and are easily modifiable for tailored use such as lipid or polymeric nanoparticles, more viable candidates for regulatory approval. Indeed, the first diagnostic imaging nanoparticle (64Cu-25%-CANF-Comb) to be utilized for imaging atherosclerotic plaque stability was polymeric based and entered Phase 0 human trials (ClinicalTrials.gov Identifier: NCT02498379) via an eIND pathway. Designing theranostic/therapeutic nanoparticles with similar specifications to those already FDA approved for clinical trials (e.g. AuroLase®) may expedite approval, especially if classified as a medical device rather than a medicinal product. Directing nanoparticles towards improved therapeutic responses in disease patients may prove more favourable to regulatory authorities (where side-effects may be more acceptable) than use as a diagnostic tracer in the healthy population. Diagnostic use where the target population are patients with disease states that have progressed to life threatening, where improved diagnostic information may allow earlier and robust therapeutic intervention when all other options have expired may offer a more acceptable first clinical use.
Regulatory approval for diagnostic imaging agents has, to date, been limited to trace amounts	For diagnostic imaging, nanomaterials should display high avidity for target combined with a high-sensitivity imaging modality so that administered concentrations are low. The lack of toxicity demonstrated by a low level nanomaterial dose in humans coupled with successful diagnostic/therapeutic outcomes may facilitate more rapid routes (e.g. eIND pathway in the USA) to in-human use with further agents of a similar design, e.g. same core nanoparticle but with a different targeting ligand. Currently eIND pathways (Phase 0 trials) for diagnostic imaging agents are restricted to PET probes as exampled above.
Potential toxicity of nanomaterials	Use biodegradable and inert components, preferably with a prior use in humans, e.g. liposomes, polymeric particles. Potentially immunogenic components should be protected by a biocompatible shell e.g. via pegylation. Ensure physiochemical properties, e.g. pH, charge are optimized for in vivo use. Perform sterilization procedures that adhere to regulatory guidelines. Initial studies should be done in cell culture to ascertain stability/possible aggregation in serum containing fluids and potential toxicity on cells. The extensive testing of agents on human endothelial cells under static and flow conditions is a necessity as endothelial cells would be the first point of contact for any nanomaterial administered intravascularly. Successful results in cell culture would be followed by safety testing in animals including the CARPA test in pigs and toxicology/immunogenicity studies in rodent and non-rodent species. Initial human studies may involve local delivery (e.g. topical or intra-colon) to avoid systemic distribution and demonstrate lack of adverse events before systemic administration.
Insufficient standardization between pre-clinical studies	If pre-clinical data sets are generated at multiple institutions, common standard operating protocols must be in place including identical nanomaterial properties, experimental methodology and data acquisition/analysis as required for new drug applications. This sharing of expertise rather than trying to do everything ‘in house’ allows the more rapid acquiring of these pre-requisite robust data sets. Manufacturing should take place in GLP facilities with an external independent partner facility validating quality control. Multiple analytical methods should be used to validate that the finished product meets desired specifications. Each batch must conform to desired specifications to have confidence in the safety/efficacy profile of the finished product.
Cost of manufacturing/upscaling production	Careful consideration should be given to simplifying, where possible, the manufacturing process and optimizing methodology that can be outsourced to a manufacturer for upscaling production without compromising the properties of the final product. Adequate sterilization and analytical methods suitable to nanoparticle size agents must be validated and implemented.
Nanoparticles do not maintain original physiochemical/biological properties over time precluding clinical use	Perform, early in the development stage, characterization of nanoparticles at multiple time points up to 1 year following storage at 4° to ensure stability.
Prolonged bioaccumulation in organs associated with nanoparticle elimination such as liver, spleen, or kidney	Detailed pharmacokinetic analysis including temporal and dose-response to monitor nanoparticle concentrations in different tissue compartments e.g. full body autoradiography and ex vivo tissue section assessment.
Agents do not meet expectations of efficacy	Careful selection of animal models with clinical grade data acquisition equipment can minimize likelihood of failure. Where possible, the use of large animal models (e.g. porcine) combined with a clinical specification imaging platform e.g. 3T MRI should be used for more representative and translatable data.
Lack of clinical adoption/market value	Design nanomaterials that target key unmet clinical needs such as earlier and improved diagnosis, monitoring disease progression or the efficacy of therapeutics. Benefits arising from this added clinical value would include improved patient outcomes, reduced diagnostic costs, reduced physical and emotional burden on patients and further cost savings to Health Authorities via the reduced demand for surgery and ineffective treatments due to improved decision-making arising from new diagnostic information. Nanomaterials targeted to common disease biomarkers are likely to find more widespread use (and ultimately greater financial reimbursement) compared with nanomaterials targeted to less well defined disease markers. Creating nanomaterials that can outperform current small molecular tracers, e.g. longer circulation times for improved target binding allowing visualization of new disease markers for diagnostic/therapeutic targeting.

eIND, exploratory investigative new drug application.

Acknowledgements

The authors thank Dr Jasmin Matuszak for help with clinical trial search.

Conflict of interest: none declared.

Funding

This work was supported by the EU (‘NanoAthero’ project FP7-NMP-2012-LARGE-6-309820), the DFG (German Research Foundation, grant CI 162/2-2), Manfred Roth Foundation (Fürth, Germany), the British Heart Foundation grant RE/13/5/30177; the European Commission Marie Skłodowska-Curie Individual Fellowships 661369; and the Engineering and Physical Sciences Research Council (EPSRC) grant EP/L014165/1. LETI/DTBS is part of the Arcane Labex programme, funded by the French National Research Agency (ARCANE project no ANR-12-LABX-003).