Gadolinium Retention in Human Brain, Bone, and Skin.

Abstract

HomeRadiologyVol. 300, No. 3 PreviousNext Reviews and CommentaryFree AccessEditorialGadolinium Retention in Human Brain, Bone, and SkinMichael F. Tweedle Michael F. Tweedle Author AffiliationsFrom the Department of Radiology, The Ohio State University, 1216 Kinnear Rd, Office 1650, Columbus, OH 43212.Address correspondence to the author (e-mail: [email protected]).Michael F. Tweedle Published Online:Jun 15 2021https://doi.org/10.1148/radiol.2021210957MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In See also the article by Kobayashi et al in this issue.Michael F. Tweedle, PhD, is the Stefanie Spielman professor of cancer imaging in the Radiology Department and James Comprehensive Cancer Center at the Ohio State University. He has more than 30 years of experience inventing and developing imaging agents in both industrial and academic settings.Download as PowerPointOpen in Image Viewer Gadolinium-based contrast agents (GBCAs) are among the safest parenteral pharmaceuticals in clinical use, and radiologists have used them to the benefit of hundreds of millions of patients for over 30 years. Over 90% of the gadolinium chelates in GBCAs are eliminated renally via urine within 24 hours. The appearance of a new disease, nephrogenic systemic fibrosis (NSF), found exclusively in severely ill patients with impaired renal function, was described in 2000. NSF usually manifests first as skin lesions, and severe cases worsen with fibrosis of the joints and some organs, which can be fatal. In 2006, NSF was linked to the use of a structural subclass of GBCAs that was subsequently expanded to three linear GBCAs. The contra-indication of the linear GBCAs linked to NSF cases halted the incidence of NSF. Linear GBCAs were long known to release gadolinium faster in vivo in animals than the kinetically more stable subclass, macrocyclic GBCAs. Thus, post-NSF attention focused on GBCA stability. In 2015, Kanda et al (1) discovered elevated MRI signal intensity at examinations of the dentate nucleus of patients with normal renal function who had received multiple doses of linear GBCAs, but not of macrocyclic GBCAs. McDonald et al (2) demonstrated micromolar concentrations of gadolinium in autopsy brain samples of patients with a normal renal function who received four or more cumulative doses of linear GBCAs. This retained gadolinium results in an elevated signal intensity at MRI. As a result of these findings, renewed efforts are underway to understand the in vivo trafficking and potential long-term toxicologic effects of the trace gadolinium retained in human tissues over weeks to years.In this issue of Radiology, Kobayashi et al (3) report (a) mass quantities of detected gadolinium in brain, bone, and skin tissues from deceased patients with a history of GBCA use and (b) nonzero elimination of this gadolinium from white matter and the skin. This retrospective, cross-sectional study compares a linear and a macrocyclic GBCA and focuses on the kinetics of elimination of the retained trace gadolinium. The study uses inductively coupled plasma mass spectrometry (ICP-MS) to evaluate tissue samples from 37 cadavers exposed in life to one or more doses of either linear gadobenate or macrocyclic gadoteridol. The authors paid careful attention to avoid any confounding effect of the use of multiple different GBCAs in a given patient. The study joins a small group of studies that use absolute but necessarily destructive techniques to quantify gadolinium in tissues as opposed to indirect, non–gadolinium-specific techniques like the detection of elevated MRI signal intensity (4). It adds to prior animal, clinical, and deceased human research results consistent with the conclusion that the macrocyclic GBCA showed several-fold less residual gadolinium in the target tissues than the linear GBCA. It is the first direct gadolinium measurement confirming a two-phase process of elimination of trace gadolinium residual following GBCA use.Like most studies of its kind, the study by Kobayashi et al used a small sample size. In addition, the timing, volume, and dosage of contrast material and the time from the last exposure to death were not standardized. Gadobenate is also an atypical linear GBCA that has approximately 4% hepatobiliary clearance and no links to NSF, but its in vivo stability still represents the linear subclass. Despite these limitations, the current study enhances the sparse body of knowledge of the kinetics of residual gadolinium in three important target tissues: bone, skin, and brain matter.The tissues chosen are important for different reasons. Bone has the largest gadolinium content among studied tissues so far. Skin is an early warning tissue in mild NSF cases. Finally, neural tissue is arguably the most sensitive tissue in a toxicologic sense. Kobayashi and colleagues confirm the findings of previous studies that, in general, gadolinium detected after macrocyclic GBCA use clears from human tissues faster than gadolinium detected after use of linear GBCAs. They found three- to sixfold less residual gadolinium in the brain, skin, and bone for the macro-cycle. In addition, elimination was roughly biphasic: fast in early weeks and then slower. This type of elimination is expected if most early elimination is of intact GBCA and a substantial portion of the slower phase represents elimination of gadolinium ions released from the intact GBCA. The gadolinium metabolite released in vivo from a GBCA is never a free ion but insoluble or protein-bound. It is eliminated about 100 times more slowly than the intact GBCA, as known from rodent studies (5). Of note, Kobayashi and colleagues found nonzero elimination of both the macrocyclic and the linear GBCAs, implying the ultimate elimination also of “deposited” free gadolinium. However, speciation to prove this inference was not part of their study. Human tissue studies that identify and quantify the detected chemical species are still needed (6).The original discovery by Kanda et al of residual elevated MRI signal intensity after multiple GBCA administrations drew attention to a previously unknown deep brain compartment for micromolar (approximately ppm) residual gadolinium. Many studies that followed used MRI signal intensity as a potential gadolinium detection tool. Like the study by Kanda et al, those studies have had value as biologic surveys of living patients due to their noninvasive nature and wide field of view. However, those MRI studies were neither gadolinium-specific (affected by endogenous paramagnetic metals that also reduce T1 in MRI) nor particularly sensitive for trace analyses. Thus, those studies have produced some variable results one would expect from a tool with these limitations. ICP-MS is a gadolinium-specific tool with more than 100-fold greater sensitivity. However, ICP-MS destroys the tissue sample, so research is restricted to biopsies and other excised tissue samples, mostly from human cadavers.In deceased humans, McDonald and colleagues (2) demonstrated that MRI signal intensity changes were linearly correlated with brain gadolinium concentration, while Murata and colleagues (7) demonstrated that bone and brain gadolinium concentrations were linearly correlated. The study by Kobayashi et al also found significant correlations between gadolinium concentration in bone and in several other tissues: the skin, dentate nucleus, globus pallidus, and white matter. This confirms and extends the prior works. The findings may allow skin and bone to become surrogates for the brain retention that presumably poses a potentially greater toxicologic risk and is not generally possible to sample in living patients.Controversy in this line of research is primarily over the lack of convincing evidence that enough free gadolinium is present in any tissue long enough to manifest an important toxicologic event. The U.S. Food and Drug Administration decided not to restrict the use of any GBCA, linear or macrocyclic, based on lack of direct evidence of any adverse health effects of gadolinium retention (8). In nearly 500 million administrations of GBCA over more than 30 years, no widespread reports traced neurologic symptoms to GBCA use. A handful of recent self-reports of brain fog, joint pain, and other symptoms are puzzling due to a lack of macrocycle versus linear differential or dose dependence that would indicate a free gadolinium role, nor any traceable anatomic or biochemical abnormalities (9,10).Gadolinium ion “deposits” and intact GBCA retention are the two likely sources of trace gadolinium retention. Intact GBCAs are well-studied, highly inert, and very unlikely to be associated with long-term toxic effects while intact. Small amounts of gadolinium ions freed from GBCA chelates are theoretically toxic because the chemical charge and size of the Gd3+ ion allow it to compete with Ca2+ for biologically important calcium sites. This phenomenon dominates toxicology studies of gadolinium ions. Replacement of calcium is the most likely mechanism of long-term retention for most of the free gadolinium, with possibly some competition with iron or precipitation with hydroxide, carbonate, or phosphate. However, the pools of these endogenous ions are enormous compared with the amounts of localized gadolinium discovered in tissues. Free gadolinium particulate deposits observed after processing brain tissue are localized to endothelium rather than neural tissues (2). These facts make finding a clinically important detrimental effect for the free gadolinium ion very unlikely in the quantities observed in renally normal patients.This is not to say that the research into trace gadolinium trafficking in humans lacks value or is not worth doing—quite the opposite. The existence of NSF was a tragic and shocking discovery. The discovery of gadolinium accumulation in the brain was surprising. Radiologists and patients need reassurance on the ultimate safety of the use of contrast agents to properly manage the risks and benefits, as is done every day for the use of ionizing radiation. The research into gadolinium trafficking in vivo relies on animal studies, living human studies, and cadaver studies, each of which bears strengths and limitations. These three types of studies cross-validated for consistency and supported by gadolinium speciation results will lead to a more thorough understanding of GBCA elimination and toxicity of any existing trace gadolinium. This is a necessary undertaking that is long overdue.Disclosures of Conflicts of Interest: Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: is a board member and 4% stock owner of Molecular Theranostics, a company developing a peptide-targeted gadolinium-based contrast agent; received consulting fees from Bracco Diagnostics, GE, and InLighta BioSciences regarding MRI contrast agents; will receive grants from Molecular Theranostics and Georgia State University. Other relationships: disclosed no relevant relationships.