Gadolinium Deposition in Neurology Clinical Practice

Artificial Intelligence to Reduce or Eliminate the Need for Gadolinium-Based Contrast Agents in Brain and Cardiac MRI: A Literature Review

ABSTRACT

Brain and cardiac MRIs are fundamental noninvasive imaging tools, which can provide important clinical information and can be performed without or with gadolinium-based contrast agents (GBCAs), depending on the clinical indication. It is currently a topic of debate whether it would be feasible to extract information such as standard gadolinium-enhanced MRI while injecting either less or no GBCAs. Artificial intelligence (AI) is a great source of innovation in medical imaging and has been explored as a method to synthesize virtual contrast MR images, potentially yielding similar diagnostic performance without the need to administer GBCAs. If possible, there would be significant benefits, including reduction of costs, acquisition time, and environmental impact with respect to conventional contrast-enhanced MRI examinations. Given its promise, we believe additional research is needed to increase the evidence to make these AI solutions feasible, reliable, and robust enough to be integrated into the clinical framework. Here, we review recent AI studies aimed at reducing or replacing gadolinium in brain and cardiac imaging while maintaining diagnostic image quality.

Gadolinium during human pregnancy following administration of gadolinium chelate before pregnancy

Gadolinium (Gd), a toxic rare earth element, has been shown to dissociate from chelating agents and bioaccumulate within tissues, raising concerns about the possibility of their remobilization during pregnancy with subsequent free Gd exposures to developing fetuses. Gd chelates are among the most commonly used magnetic resonance imaging (MRI) contrast agents. This investigation was undertaken after the detection of elevated Gd (800-1000× higher than the usual rare earth element levels) in preliminary unpublished studies from the placentae of subjects in the NIH ECHO/UPSIDE Rochester Cohort Study and unpublished studies from placentae analyzed in formalin-fixed placental specimens from Surgical Pathology at the University of Rochester. Fifteen pregnancies with elevated Gd were studied (12 first pregnancies and 3 second pregnancies). Maternal bloods were collected from all three trimesters, maternal, and cord (fetal) bloods at delivery as well as placental tissue. Breastmilk was also collected from selected mothers. It was determined that Gd was present in maternal bloods from all three trimesters, and in cord bloods and breastmilk in both first and second pregnancies. These results emphasize the need to fully appreciate the implications of pre-pregnancy exposure to Gd chelates and its potential effects on maternal and fetal health.

Imaging surveillance in multiple endocrine neoplasia type 1: Ten years of experience with somatostatin receptor positron emission tomography

Guidelines for multiple endocrine neoplasia type 1 (MEN1) recommend intensive imaging surveillance without specifying a superior regimen, including the role of somatostatin receptor imaging (SRI) with positron emission tomography (PET). The primary outcomes were to: (1) Assess change in treatment of duodenal-pancreatic neuroendocrine neoplasms (DP-NENs), bronchopulmonary NENs, and thymic tumors attributed to use of SRI PET/computed tomography (CT) and (2) estimate radiation from imaging and risk of cancer death attributed to imaging radiation. This was a retrospective single center study, including all MEN1 patients, who had had at least one SRI PET/CT. A total of 60 patients, median age 42 (range 21-54) years, median follow-up 6 (range 1-10) years were included. Of 470 cross sectional scans (MRI, CT, SRI PET/CT), 209 were SRI PET/CT. The additional information from SRI PET had implications in 1/14 surgical interventions and 2/12 medical interventions. The estimated median radiation dose per patient was 104 (range 51-468) mSv of which PET contributed with 13 (range 5-55) mSv and CT with 91 mSv (range 46-413 mSv), corresponding to an estimated increased median risk of cancer death of 0.5% during 6 years follow-up. SRI PET had a significant impact on 3/26 decisions to intervene in 60 MEN1 patients followed for a median of 6 years with SRI PET/CT as the most frequently used modality. The surveillance program showed a high radiation dose. Multi-modality imaging strategies designed to minimize radiation exposure should be considered. Based on our findings, SRI-PET combined with CT cannot be recommended for routine surveillance in MEN1 patients.

Consensus statement for metastatic surveillance of uveal melanoma in Scotland

Ophthalmic treatments are successful in managing uveal melanomas achieving good local control. However, a large number still metastasise, primarily to the liver, resulting in mortality. There is no consensus across the world on the mode, frequency, duration or utility of regular liver surveillance for metastasis and there are no published protocols. The Scottish Ocular Oncology Service (SOOS) constituted a Scottish Consensus Statement Group (SCSG) which included ocular oncologists, medical oncologists, radiologists and a uveal melanoma patient as a lay member. This group carried out an extensive review of literature followed by discussions to arrive at a consensus regarding surveillance planning for posterior uveal melanoma patients in Scotland. The Consensus Statement would provide a framework to guide each patient's surveillance plan and provide all patients with clarity and transparency on the issue. The SCSG was unable to find adequate evidence on which to base the strategy. The consensus statement recommends a risk-stratified approach to surveillance for these patients dividing them into low to medium-risk and high-risk groups defining the mode and duration of surveillance for each. It supplements the UK-wide Uveal Melanoma National Guidelines and allows a more uniform consensus-based approach to surveillance in Scotland. It has been adopted nationally by all health care providers in Scotland as a guideline and is available to patients on a publicly accessible website.

Shape programmable T1-T2 dual-mode MRI nanoprobes for cancer theranostics

The shape effect is an important parameter in the design of novel nanomaterials. Engineering the shape of nanomaterials is an effective strategy for optimizing their bioactive performance. Nanomaterials with a unique shape are beneficial to blood circulation, tumor targeting, cell uptake, and even improved magnetism properties. Therefore, magnetic resonance imaging (MRI) nanoprobes with different shapes have been extensively focused on in recent years. Different from other multimodal imaging techniques, dual-mode MRI can provide imaging simultaneously by a single instrument, which can avoid differences in penetration depth, and the spatial and temporal resolution of multiple imaging devices, and ensure the accurate matching of spatial and temporal imaging parameters for the precise diagnosis of early tumors. This review summarizes the latest developments of nanomaterials with various shapes for T₁-T₂ dual-mode MRI, and highlights the mechanism of how shape intelligently affects nanomaterials' longitudinal or transverse relaxation, namely sphere, hollow, core-shell, cube, cluster, flower, dumbbell, rod, sheet, and bipyramid shapes. In addition, the combination of T₁-T₂ dual-mode MRI nanoprobes and advanced therapeutic strategies, as well as possible challenges from basic research to clinical transformation, are also systematically discussed. Therefore, this review will help others quickly understand the basic information on dual-mode MRI nanoprobes and gather thought-provoking ideas to advance the subfield of cancer nanomedicine.

The Right Imaging Protocol for the Right Patient

OBJECTIVE

This article provides a high-level overview of the challenge of choosing the right imaging approach for an individual patient. It also presents a generalizable approach that can be applied to practice regardless of specific imaging technologies.

ESSENTIAL POINTS

This article constitutes an introduction to the in-depth, topic-focused analyses in the rest of this issue. It examines the broad principles that guide placing a patient on the right diagnostic trajectory, illustrated with real-life examples of current protocol recommendations and cases of advanced imaging techniques, as well as some thought experiments. Thinking about diagnostic imaging strictly in terms of imaging protocols is often inefficient because these protocols can be vague and have numerous variations. Broadly defined protocols may be sufficient, but their successful use often depends largely on the particular circumstances, with special emphasis on the relationship between neurologists and radiologists.

Can Deep Learning Replace Gadolinium in Neuro-Oncology?: A Reader Study

MATERIALS AND METHODS

This monocentric retrospective study leveraged 200 multiparametric brain MRIs acquired between November 2019 and February 2020 at Gustave Roussy Cancer Campus (Villejuif, France). A total of 145 patients were included: 107 formed the training sample (55 ± 14 years, 58 women) and 38 the separate test sample (62 ± 12 years, 22 women). Patients had glioma, brain metastases, meningioma, or no enhancing lesion. T1, T2-FLAIR, diffusion-weighted imaging, low-dose, and standard-dose postcontrast T1 sequences were acquired. A deep network was trained to process the precontrast and low-dose sequences to predict "virtual" surrogate images for contrast-enhanced T1. Once trained, the deep learning method was evaluated on the test sample. The discrepancies between the predicted virtual images and the standard-dose MRIs were qualitatively and quantitatively evaluated using both automated voxel-wise metrics and a reader study, where 2 radiologists graded image qualities and marked all visible enhancing lesions.

RESULTS

The automated analysis of the test brain MRIs computed a structural similarity index of 87.1% ± 4.8% between the predicted virtual sequences and the reference contrast-enhanced T1 MRIs, a peak signal-to-noise ratio of 31.6 ± 2.0 dB, and an area under the curve of 96.4% ± 3.1%. At Youden's operating point, the voxel-wise sensitivity (SE) and specificity were 96.4% and 94.8%, respectively. The reader study found that virtual images were preferred to standard-dose MRI in terms of image quality (P = 0.008). A total of 91 reference lesions were identified in the 38 test T1 sequences enhanced with full dose of contrast agent. On average across readers, the brain lesion SE of the virtual images was 83% for lesions larger than 10 mm (n = 42), and the associated false detection rate was 0.08 lesion/patient. The corresponding positive predictive value of detected lesions was 92%, and the F1 score was 88%. Lesion detection performance, however, dropped when smaller lesions were included: average SE was 67% for lesions larger than 5 mm (n = 74), and 56% with all lesions included regardless of their size. The false detection rate remained below 0.50 lesion/patient in all cases, and the positive predictive value remained above 73%. The composite F1 score was 63% at worst.

CONCLUSIONS

The proposed deep learning method for virtual contrast-enhanced T1 brain MRI prediction showed very high quantitative performance when evaluated with standard voxel-wise metrics. The reader study demonstrated that, for lesions larger than 10 mm, good detection performance could be maintained despite a 4-fold division in contrast agent usage, unveiling a promising avenue for reducing the gadolinium exposure of returning patients. Small lesions proved, however, difficult to handle for the deep network, showing that full-dose injections remain essential for accurate first-line diagnosis in neuro-oncology.

Cross-modal distillation to improve MRI-based brain tumor segmentation with missing MRI sequences

Convolutional neural networks (CNNs) for brain tumor segmentation are generally developed using complete sets of magnetic resonance imaging (MRI) sequences for both training and inference. As such, these algorithms are not trained for realistic, clinical scenarios where parts of the MRI sequences which were used for training, are missing during inference. To increase clinical applicability, we proposed a cross-modal distillation approach to leverage the availability of multi-sequence MRI data for training and generate an enriched CNN model which uses only single-sequence MRI data for inference but outperforms a single-sequence CNN model. We assessed the performance of the proposed method for whole tumor and tumor core segmentation with multi-sequence MRI data available for training but only T1- weighted (T1w) sequence data available for inference, using both BraTS 2018, and in-house datasets. Results showed that cross-modal distillation significantly improved the Dice score for both whole tumor and tumor core segmentation when only T1w sequence data were available for inference. For the evaluation using the in-house dataset, cross-modal distillation achieved an average Dice score of 79.04% and 69.39% for whole tumor and tumor core segmentation, respectively, while a single-sequence U-Net model using T1w sequence data for both training and inference achieved an average Dice score of 73.60% and 62.62%, respectively. These findings confirmed cross-modal distillation as an effective method to increase the potential of single-sequence CNN models such that segmentation performance is less compromised by missing MRI sequences or having only one MRI sequence available for segmentation.

Safety and Gadolinium Distribution of the New High-Relaxivity Gadolinium Chelate Gadopiclenol in a Rat Model of Severe Renal Failure

OBJECTIVE

The aim of this study was to investigate the toxicological profile of gadopiclenol, a new high-relaxivity macrocyclic gadolinium-based contrast agent (GBCA), in renally impaired rats, in comparison with 2 other macrocyclic GBCAs, gadoterate meglumine and gadobutrol, and 1 linear and nonionic GBCA, gadodiamide.

METHODS

Renal failure was induced by adding 0.75% wt/wt adenine to the diet for 3 weeks. During the second week of adenine-enriched diet, the animals (n = 8/group × 5 groups) received 5 consecutive intravenous injections of GBCA at 2.5 mmol/kg per injection, resulting in a cumulative dose of 12.5 mmol/kg or saline followed by a 3-week treatment-free period after the last injection. The total (elemental) gadolinium (Gd) concentration in different tissues (brain, cerebellum, femoral epiphysis, liver, skin, heart, kidney, spleen, plasma, urine, and feces) was measured by inductively coupled plasma mass spectrometry. Transmission electron microscopy (and electron energy loss spectroscopy analysis of metallic deposits) was used to investigate the presence and localization of Gd deposits in the skin. Relaxometry was used to evaluate the presence of dissociated Gd in the skin, liver, and bone. Skin histopathology was performed to investigate the presence of nephrogenic systemic fibrosis-like lesions.

RESULTS

Gadodiamide administrations were associated with high morbidity-mortality but also with macroscopic and microscopic skin lesions in renally impaired rats. No such effects were observed with gadopiclenol, gadoterate, or gadobutrol. Overall, elemental Gd concentrations were significantly higher in gadodiamide-treated rats than in rats treated with the other GBCAs for all tissues except the liver (where no significant difference was found with gadopiclenol) and the kidney and the heart (where statistically similar Gd concentrations were observed for all GBCAs). No plasma biochemical abnormalities were observed with gadopiclenol or the control GBCAs. Histopathology revealed a normal skin structure in the rats treated with gadopiclenol, gadoterate, and gadobutrol, contrary to those treated with gadodiamide. No evidence of Gd deposits on collagen fibers and inclusions in fibroblasts was found with gadopiclenol and its macrocyclic controls, unlike with gadodiamide. Animals of all test groups had Gd-positive lysosomal inclusions in the dermal macrophages. However, the textures differed for the different products (speckled texture for gadodiamide and rough-textured appearance for the 2 tested macrocyclic GBCAs).

CONCLUSIONS

No evidence of biochemical toxicity or pathological abnormalities of the skin was observed, and similar to other macrocyclic GBCAs, gadoterate and gadobutrol, tissue retention of Gd was found to be low (except in the liver) in renally impaired rats treated with the new high-relaxivity GBCA gadopiclenol.

Neuroimaging Considerations in Patients with Chronic Kidney Disease

Chronic Kidney Disease is a common medical condition that frequently overlaps with neurologic disease. Neuroimaging can be a useful tool to aid in the diagnoses of neurologic illness, including those that result from renal impairment. Some neuroimaging studies also have the potential to lead to adverse effects on the kidneys necessitating a thoughtful approach to selection of imaging modalities. In particular, multimodal imaging is becoming increasingly common in patients presenting with symptoms of acute stroke, a population that may be at higher risk for renal complications. This article will summarize the neuroimaging manifestations of conditions with shared renal and neurologic involvement and highlight considerations regarding the use of contrast media, nephrogenic systemic fibrosis, and metformin-associated lactic acidosis.

Gadolinium-based contrast agents: What we learned from acute adverse events, nephrogenic systemic fibrosis and brain retention

BACKGROUND

 Radiologists have been administering gadolinium-based contrast agents (GBCA) in magnetic resonance imaging for several decades, so that there is abundant experience with these agents regarding allergic-like reactions, nephrogenic systemic fibrosis (NSF) and gadolinium retention in the brain.

METHODS

 This review is based on a selective literature search and reflects the current state of research on acute adverse effects of GBCA, NSF and brain retention of gadolinium.

RESULTS

 Due to the frequent use of GBCA, data on adverse effects of these compounds are available in large collectives. Allergic-like reactions occurred rarely, whereas severe acute reactions were very rarely observed. Systemic changes in NSF also occur very rarely, although measures to avoid NSF resulted in a significantly reduced incidence of NSF. Due to gadolinium retention in the body after administration of linear MR contrast agents, only macrocyclic preparations are currently used with few exceptions. Clear clinical correlates of gadolinium retention in the brain could not be identified so far. Although the clinical added value of GBCA is undisputed, individual risks associated with the injection of GBCA should be identified and the use of non-contrast enhanced MR techniques should be considered. Alternative contrast agents such as iron oxide nanoparticles are not clinically approved, but are currently undergoing clinical trials.

CONCLUSION

 GBCA have a very good risk profile with a low rate of adverse effects or systemic manifestations such as NSF. Gadolinium retention in the brain can be minimized by the use of macrocyclic GBCA, although clear clinical correlates due to gadolinium retention in the brain following administration of linear GBCA could not be identified yet.

KEY POINTS

· Acute adverse effects are predominantly mild/moderate, rarely severe reactions occur.. · International guidelines resulted in significant reduction of nephrogenic systemic fibrosis.. · Application of macrocyclic contrast agents minimizes gadolinium retention in the brain..

CITATION FORMAT

· Bäuerle T, Saake M, Uder M. Gadolinium-based contrast agents: What we learned from acute adverse events, nephrogenic systemic fibrosis and brain retention. Fortschr Röntgenstr 2021; 193: 1010 - 1018.

Presence of other rare earth metals in gadolinium-based contrast agents

Gadolinium-based contrast agents (GBCA) are widely used to enhance tissue contrast during magnetic resonance imaging (MRI) procedures. However, free Gadolinium (Gd) is undesirable as a drug substance, due to its high toxicity. Consequently, a coordinating ligand is required to keep it in solution and to increase tolerance. In order to achieve an adequate performance, GBCA must be administered in relatively large amounts. Chelate amounts are around 13-20 g and for Gd alone, this may amount to 3.3 g. Taking into account the route of administration, impurities in GBCA may be significant. Gadolinium occurs in nature along with 16 other elements known collectively as rare earth metals (RE), which are found throughout the earth's crust in minerals such as monazite. Gadolinium oxide corresponds to 0.7-4.0% of the RE present in minerals, and the sum concentration of RE in minerals is around 4%. Rare earth metals are difficult to separate, as the chemical and physical properties of one RE are significantly similar to those of others. In this study, the presence of other RE in GBCA formulations was investigated. Different lots of Magnevist®, Viewgam®, OptiMARK®, Omniscan®, Dotarem®, and Gadovist® were analyzed. Inductively-coupled plasma mass spectrometry and atomic absorption spectrometry were used for RE determination. Procedure optimization included sample decomposition and method validation for element determination. The results showed that Sc, Y, La, Ce, Pr, Nd, Eu, Tb, Tm, Dy, Ho, and Er were present in the 22 samples analyzed. Terbium, Thulium, Europium, and Lanthanum were, on average, found in the highest amounts, which were 0.42 mg/L, 0.17 mg/L, 0.17 mg/L, and 0.16 mg/L, respectively. These results could be attributed to the similarity among Europium, Gadolinium, and Terbium. They are in sequence in the periodic table and therefore present very close ionic radii, restricting their separation. Considering the sum of all RE, Viewgam® was the most contaminated formulation (mean of 2.16 mg/L) and Magnevist® the least (mean of 0.64 mg/L). Although the RE are chemically similar, the other RE do not perform as Gd as a contrast agent; therefore, their presence in formulations may be a matter of concern.

Retention of Gadolinium in Brain Parenchyma: Pathways for Speciation, Access, and Distribution. A Critical Review

The unexpected appearance of T₁ hyperintensities, mostly in the dentate nucleus and the globus pallidus, during nonenhanced MRI was reported in 2014. This effect is associated with prior repeated administrations of gadolinium (Gd)‐based contrast agents (GBCAs) in patients with a functional blood–brain barrier (BBB). It is widely assumed that GBCAs do not cross the intact BBB, but the observation of these hypersignals raises questions regarding this assumption. This review critically discusses the mechanisms of Gd accumulation in the brain with regard to access pathways, Gd species, tissue distribution, and subcellular location. We propose the hypothesis that there is early access of Gd species to cerebrospinal fluid, followed by passive diffusion into the brain parenchyma close to the cerebral ventricles. When accessing areas rich in endogenous metals or phosphorus, the less kinetically stable GBCAs would dissociate, and Gd would bind to endogenous macromolecules, and/or precipitate within the brain tissue. It is also proposed that Gd species enter the brain parenchyma along penetrating cortical arteries in periarterial pial‐glial basement membranes and leave the brain along intramural peri‐arterial drainage (IPAD) pathways. Lastly, Gd/GBCAs may access the brain parenchyma directly from the blood through the BBB in the walls of capillaries. It is crucial to distinguish between the physiological distribution and drainage pathways for GBCAs and the possible dissociation of less thermodynamically/kinetically stable GBCAs that lead to long‐term Gd deposition in the brain.

Level of Evidence

5.

Technical Efficacy Stage

3.

Fe-HBED Analogs: A Promising Class of Iron-Chelate Contrast Agents for Magnetic Resonance Imaging

Contrast-enhanced magnetic resonance imaging is an essential tool for disease diagnosis and management; all marketed clinical magnetic resonance imaging (MRI) contrast agents (CAs) are gadolinium (Gd) chelates and most are extracellular fluid (ECF) agents. After intravenous injection, these agents rapidly distribute to the extracellular space and are also characterized by low serum protein binding and predominant renal clearance. Gd is an abiotic element with no biological recycling processes; low levels of Gd have been detected in the central nervous system and bone long after administration. These observations have prompted interest in the development of new MRI contrast agents based on biotic elements such as iron (Fe); Fe-HBED (HBED = N,N′-bis(2-hydroxyphenyl)ethylenediamine-N,N′-diacetic acid), a coordinatively saturated iron chelate, is an attractive MRI CA platform suitable for modification to adjust relaxivity and biodistribution. Compared to the parent Fe-HBED, the Fe-HBED analogs reported here have lower serum protein binding and higher relaxivity as well as lower relative liver enhancement in mice, comparable to that of a representative gadolinium-based contrast agent (GBCA). Fe-HBED analogs are therefore a promising class of non-Gd ECF MRI CA.