Elsevier

World Neurosurgery

Volume 78, Issue 6, December 2012, Pages 709-711
World Neurosurgery

Peer-Review Report
Frontiers
The Use of Nanoparticles as Contrast Media in Neuroimaging: A Statement on Toxicity

https://doi.org/10.1016/j.wneu.2011.08.013Get rights and content

The use of nanoparticles in diagnostic imaging is rapidly gaining utility and acceptance. A handful of iron oxide nanoparticle compounds have already been approved by the U.S. Food and Drug Administration for clinical use with a favorable acute safety profile. However, because use of these agents is still in its early stages, long-term clinical data has yet to become readily available. The ability of these particles to interact with cellular biology at a molecular level does have theoretical injurious potential. As with any medical intervention, its relative risks must be clearly understood. This article discusses the safety profile and potential toxicities of nanoparticles as used in diagnostic imaging, and serves to inform the prescribing physician of relative and potential risk to the patient.

Introduction

The use of superparamagnetic iron oxide nanoparticles (SPIO) and ultrasmall superparamagnetic iron oxide nanoparticles (USPIO) continues to gain utility and acceptance in the advancement of neuroimaging. As with any new diagnostic or treatment modality, the excitement must be met with equal if not greater caution to possible side effects. In this supplement we discuss the current understood and potential toxicities of the use of nanoparticles in diagnostic imaging.

The metabolism and pharmacokinetics of (U)SPIO continue to be under investigation. Iron oxide nanoparticles are readily endocytosed by circulating macrophages and further incorporated into cytoplasmic lysosomes. Although it has been generally agreed that inherent cellular iron metabolism ultimately regulates excretion and storage, the process by which (U)SPIO converts the contrast agent into biocompatible iron has not been fully described. Furthermore, the mass of injected iron nanoparticles may exceed 200 mg, whereas daily physiologic requirements for erythropoiesis are about 30 mg, emphasizing the need to understand mechanism for dealing with this iron surplus (1). To further explore the decomposition of various (U)SPIO contrast agents in vitro, Lévy et al. recreated the rich enzymatic and acidic environment of the lysosome (4). Although the different nanoparticles demonstrated variable kinetics based on their coatings, the concentration of intact nanoparticles decreased with time, correlating with an increase in the nontoxic form of free iron ions (Fe[III]) (Figure 1). However, the magnetic properties and ultimate structure of the remaining nanocrystals did not change in the lysosomal environment. Fe(III) is readily chelated by citrate, also present in the lysosomal environment, and this bound form may be subject to physiologic iron storage and metabolism. Although this study demonstrates a potential mechanism and molecular end product of (U)SPIO metabolism, it remains to be seen how our biologic processes manage the increase in iron concentration. Despite this poorly understood metabolic process, the use of (U)SPIO as an intravenous contrast agent is considered to be relatively safe, with little to no major toxicities having been reported.

Ralph Weissleder is one of the original investigators and continues to explore the utility of (U)SPIO-based agents for specific enhanced imaging of various organ tissues. His initial work described these agents' affinity for the hepatic, splenic, and lymphatic tissues. With modifications of the outer shell of these nanostructures, he has also demonstrated important utility in neuroimaging. Two decades ago, he demonstrated the safety profile of SPIO in various animal models. Both rats and dogs were given AMI-25, an SPIO already being used in clinical trials, and studied for effects of acute toxicity, subacute toxicity, and mutagenicity (10). Animals with the highest dose injection did not show any evidence of acute toxicity other than discolored mucosa. There was no evidence of increased mortality or morbidity, changes in feeding habits, or body weight in more than 3 weeks of observation. The only significant changes noted were increases in serum iron concentration and hematocrit, both of which remained at the upper limit of normal. The Ames test for mutagenicity did not demonstrate any evidence of cellular effect of AMI-25 up to its highest dosage concentration. Histopathology did reveal iron deposits in hepatic and reticuloendothelial tissue without any evidence of cell damage. Although iron toxicity remains a concern, physiologic iron metabolism was hypothesized to be capable of handling the increase in serum iron caused by this agent. Ros et al. later demonstrated the safety of intravenous ferumoxides in imaging of hepatic lesions in humans. Their study did not demonstrate any adverse effects in over 200 patients with focal hepatic lesions; however, follow-up was limited (7).

These studies among others have led the U.S. Food and Drug Administration to approve various superparamagnetic iron oxide nanoparticles for clinical use for imaging of abdominal viscera; however, these agents remain off-label for the use of neuroimaging. There is potential concern for long-term neurotoxicity given these agents' prolonged half-life and opsonization by neuroglial cells. Furthermore, free iron has been associated with the formation of free radicals, which would be particularly harmful to neural tissues already weakened by pathologic processes. Muldoon et al. examined the neurotoxicity of various SPIO nanoparticles in an animal model. Rats were given either an intracerebral inoculation or an intravenous administration with physiologic blood–brain barrier disruption of ferumoxytol, ferumoxides, ferumoxtran-10, or its laboratory preparation MION-46 (5). Tumor model rats were also given intravenous administration of these agents. Animals followed up clinically and with serial magnetic resonance imaging for 3 months before being killed and pathologically examined. In all specimens, iron deposits were noted in the various pathologic tissues and direct inoculation sites. Even in the direct inoculation group, there was no evidence of histopathologic change in myelin tracts, the only pathology being attributed to the trauma of the needle track. However, despite the promising safety profile of these nanoparticles demonstrated by this group, this study is still limited by its follow-up time and limited exploration systemic toxicity.

Bernd et al. have reviewed the clinical safety and tolerability of the modern USPIO agent ferumoxtran-10 in its use for hepatic and splenic imaging (2). Data were pooled from 37 clinical studies of 1777 adults. At least 1 adverse event was reported in 23.2% of patients; 18.2% of patients had adverse events thought to be related to the study drug. Serious adverse events were reported in 2.6% of patients, but in only 0.42% of patients were these events thought to be related to ferumoxtran-10. One death was attributed to anaphylactic shock after an undiluted bolus injection of the study drug; this method of administration is no longer recommended. The most common adverse events included back pain (3%), pruritus (2.5%), and other common infusion-related reactions; half of these occurred within the first 5 minutes of infusion and resolved before the infusion was complete. The serious reactions observed (chest pain, dyspnea, desaturation) were also typical of infusion of contrast agents; patients did favorably in all but 1 case. The 1 mortality was in a patient who presented with multiple risk factors, including previous history of allergy to iodinated contrast, metastatic cancer, and poor general health. In general, there did not seem to be a correlation of adverse events with dose; however, a more rapid rate of infusion did correspond to an increase in adverse events. Patients with a history of allergy did report more hypersensitivity-type symptoms. Subgroup analysis of high-risk patients, including the elderly and those with hepatic or renal disease, did not seem to have any increased incidence of adverse events compared with the general population. Overall, the observed immediate toxicity of ferumoxtran-10 is on the order of other infused intravenous contrast agents without the evidence of acute nephropathy or nephrogenic systemic fibrosis. Again, long-term follow up data must still be collected.

The potential pro-inflammatory and allergenic effects of SPOIO seem to be enhanced by inhalational administration. Park et al. intratracheally administered mice with iron oxide nanoparticles and observed levels of inflammatory cytokines and activity of genes related to inflammation over a period of 28 days (6). They demonstrated elevated levels of many inflammatory cytokines, including interleukin-1 and tumor necrosis factor-α, as well as increased expression of matrix metalloproteinases and heat shock protein. Furthermore, microgranuloma formation was observed in the alveolar space. Interestingly, these inflammatory reactions occurred in a subacute time frame, not in the typical acute setting one would expect for an inhalational drug. This study provides evidence against inhalational use of SPIO, although only limited conclusions can be made about the inflammatory response of its parenteral administration.

Theoretical toxicity of SPIO particles at the genetic level also exists. Because nanoparticles are opsonized and metabolized by targeted cells, they have the potential to affect molecular and genetic cell behavior. One potential laboratory use for SPIO particles is to label and track stem cells (9). Because of their pluripotent capabilities and metabolic turnover, these cells are also at higher risk of metabolic alterations from molecules incorporated into their cytoplasm. Chen et al. demonstrated this effect in vitro on osteogenic differentiation in mesenchymal stem cells (3). Ferucarbotran, an ionic SPIO, inhibited osteogenic differentiation at lower doses and abolished this process at higher doses. Ferucarbotran-treated cells also demonstrated increased cell migration. This was found to be the result of iron produced within the cell as a result of metabolism of the ferucarbotran. Intracellular iron ultimately activates signaling cascades via matrix metalloproteinases and synovial sarcoma x antigen. This effect was inhibited by iron chelators that decreased the intracellular iron concentration. Although the biocompatibility of (U)SPIO molecules has been cited as a potential benefit to these agents, they do have the potential to adversely effect intrinsic pathways mediated by iron, ultimately leading to undesirable biologic outcomes.

The ability of these nanoparticles to penetrate the blood–brain barrier is a significant benefit over traditional contrast agents in neuroimaging. However, this physiologic barrier serves a significant protective function, and its pharmacologic disruption must be approached with caution, especially with agents designed to be incorporated into a cells' cytoplasm. Sharma and Sharma described the effects of normal animals treated with metal nanoparticles (including copper, gold, and silver) as a drug-delivery agent for more than 1 week (8). These animals exhibited mild cognitive impairment that was further exacerbated when the animals were subjected to hyperthermia. Although these observations do not describe the traditional dosing of nanoparticle based contrast agents, this toxicity is not to be ignored because many patients will undergo serial neuroimaging subjecting the central nervous system parenchyma to compounding doses of nanoparticles. In addition, the potential for nanoparticles to be used as a radiosensitizing agent in combination with thermotherapy makes these findings more applicable. Furthermore, the metal-specific toxicities of the nanoparticles must be explored because current contrast agents are made from iron compounds, whereas those studied by Sharma et al. included heavier metals.

As SPIO nanoparticles continue to gain popularity and efficacy in diagnostic imaging and in clinical neuroscience as a whole, we must proceed with optimistic caution. Our ability to interact with neurobiology has already evolved to incorporate micromanipulations, and is now progressing to a nanoscale. Although more precise treatments have great therapeutic potential, their potential for global injury is equally as great. The immediate clinical effects of SPIO contrast agents have been well characterized and have been found to be quite safe. It is the molecular interactions of these nanoparticles that raises concern for long-term metabolic or mutagenic effects. In terms of risk–benefit analysis, long-term effects may be less concerning for the diagnosis and treatment of high-grade malignancies with limited survival rates. However, given this technology's potential use in diagnosis (and ultimately treatment) of chronic pathologies such as Alzheimer disease and atherosclerosis, these long-term effects must be equally well characterized. The potential of nanoparticles to broaden our neurodiagnostic and neurotherapeutic capabilities is dramatic. To maximize its impact, we must continue our efforts to meticulously evaluate its safety.

Cited by (0)

Conflict of interest statement: The authors declare that the article content was composed in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

View full text