Manganese-Enhanced MRI of the Brain in Healthy Volunteers

D.M. Sudarshana; G. Nair; J.T. Dwyer; B. Dewey; S.U. Steele; D.J. Suto; T. Wu; B.A. Berkowitz; A.P. Koretsky; I.C.M. Cortese; D.S. Reich

doi:10.3174/ajnr.A6152

Abstract

BACKGROUND AND PURPOSE: The manganese ion is used as an intracellular MR imaging contrast agent to study neuronal function in animal models, but it remains unclear whether manganese-enhanced MR imaging can be similarly useful in humans. Using mangafodipir (Teslascan, a chelated manganese-based contrast agent that is FDA-approved), we evaluated the dynamics of manganese enhancement of the brain and glandular structures in the rostral head and neck in healthy volunteers.

MATERIALS AND METHODS: We administered mangafodipir intravenously at a rate of 1 mL/minute for a total dose of 5 μmol/kg body weight. Nine healthy adult volunteers (6 men/3 women; median age, 43 years) completed baseline history and physical examination, 3T MR imaging, and blood work. MR imaging also followed mangafodipir administration at various time points from immediate to 7 days, with delayed scans at 1–3 months.

RESULTS: The choroid plexus and anterior pituitary gland enhanced within 10 minutes of infusion, with enhancement persisting up to 7 and 30 days, respectively. Exocrine (parotid, submandibular, sublingual, and lacrimal) glands also enhanced avidly as early as 1 hour postadministration, generally resolving by 1 month; 3 volunteers had residual exocrine gland enhancement, which resolved by 2 months in 1 and by 3 months in the other 2. Mangafodipir did not affect clinical parameters, laboratory values, or T1-weighted signal in the basal ganglia.

CONCLUSIONS: Manganese ions released from mangafodipir successfully enable noninvasive visualization of intra- and extracranial structures that lie outside the blood-brain barrier without adverse clinical effects, setting the stage for future neuroradiologic investigation in disease.

ABBREVIATIONS:

FA: flip angle
GRE: gradient recalled-echo
MEMRI: manganese-enhanced MRI

Paramagnetic contrast agents play a key role in MR imaging by influencing local magnetic fields and widening tissue-specific differences to enhance intrinsic structural information. Functional contrast agents, an emerging field, provide an additional layer of molecular and cellular information through their ability to probe intracellular processes. In preclinical studies, the manganese ion (Mn²⁺), a functional contrast agent, has garnered increasing attention due to its ability to accumulate within cells via L-type calcium channels and induce a strong reduction in intracellular T1 relaxation time, as well as travel along neural circuits.¹ The application of manganese-enhanced MR imaging (MEMRI) has thus facilitated a greater understanding of brain laminar structure, neural activity, and neural connectivity.^{2⇓⇓⇓⇓–7}

The application of MEMRI in humans has been impeded by the cardiotoxicity and potential neurotoxicity of high doses of free Mn²⁺ ions.^8,9 Mangafodipir trisodium is a chelated manganese-based contrast agent that is FDA-approved for liver and pancreas imaging.¹⁰ After intravenous administration, mangafodipir undergoes simultaneous dephosphorylation and transmetallation, resulting in a slow release of manganese ions; this is thought to curtail the hazardous effects seen with bolus doses of manganese chloride (MnCl₂).^11,12 Preliminary studies in the human brain have demonstrated enhancement of structures outside the blood-brain barrier (choroid plexus and pituitary gland), with half-lives substantially longer than those of gadolinium chelates, which remain extracellular.¹³ However, detailed characterization of this enhancement has not been reported, to our knowledge.

The purpose of this study was to use MEMRI to evaluate the dynamics and pattern of manganese enhancement in the brain and rostral head-and-neck structures. Because of the limited availability of mangafodipir and of participants for repeat scans, temporal sampling was sparse for individuals but attenuated across the study population. We also performed quantitative T1 mapping to estimate contrast uptake and washout. Furthermore, we evaluated T1-weighted signal in the basal ganglia and cerebral cortex to assess retention of manganese ions.

Materials and Methods

Participant Recruitment

Participants were enrolled in the National Institutes of Health institutional review board–approved study, “Manganese-Enhanced Magnetic Resonance Imaging in Healthy Volunteers and People with Multiple Sclerosis” (NCT01326715). Each participant provided written informed consent. Healthy volunteers were recruited between October 2013 and June 2017. Key inclusion criteria were the following: 18–70 years of age, normal complete blood count, normal findings of kidney function and liver function tests, and normal brain MR imaging findings within 1 year before enrollment. Key exclusion criteria were the following: hypertension (>160/100 mm Hg), previous or current alcohol or substance abuse, occupational exposure to manganese, family history of Parkinson disease, pregnancy/lactation, or treatment with calcium channel blockers.

Screening Visit and Mangafodipir Infusion

The screening visit consisted of a physical examination, medical history, laboratory testing, and routine clinical 3T MR imaging to ensure absence of structural abnormalities. Mangafodipir trisodium was infused intravenously (50 μmol/mL at 1 mL/minute, for a total dose of 5 μmol/kg) after a baseline scan. Vital signs were monitored before, during, and after infusion.

MEMRI Protocol

Images were acquired on a Magnetom Skyra (Siemens, Erlangen, Germany) 3T scanner with a body-transmit and a 32-channel receiver head coil. Sequences acquired at each MEMRI time point (as detailed in the Table) included proton-density and T2-weighted (turbo spin-echo: TR = 5000 ms, TE = 18/82 ms, FOV = 256 mm, 1-mm in-plane resolution, 54 slices of 3-mm thickness), multiple T1-weighted (gradient-recalled-echo [GRE]: TR = 7.8 ms, TE = 3 ms, FOV = 256 mm, 1-mm isotropic resolution, 192 slices, flip angles [FAs] of 3°, 6°, 9°, 12°, 15°, and 18°), T1-weighted MPRAGE (turbo flash: TR = 1900 ms, TE = 2.5 ms, TI = 900 ms, FOV = 256 mm, FA = 9°, 1-mm isotropic resolution, 176 slices), or MP2RAGE (turbo flash: TR = 5000 ms, TE = 3 ms, TI = 700 and 1400 ms, FA = 4°, FOV = 256 mm, 1-mm isotropic resolution, 176 slices), T2-weighted FLAIR (inversion recovery sampling perfection with application-optimized contrasts by using different flip angle evolutions [IR SPACE sequence; Siemens]: TR = 4800 ms, TE = 353 ms, TI = 1800 ms, FOV = 256 mm, 1-mm isotropic resolution, 176 slices), and T2*-weighted (multishot echo-planar imaging [EPI]: TR = 64 ms, TE = 35 ms, FA = 10°, FOV = 250 mm, 0.65-mm isotropic resolution, 256 slices, EPI factor = 15).

View this table:

Healthy volunteer study protocol

A B₁ map using a modified Bloch-Siegert method¹⁴ (turbo flash sequence: TR = 19 ms, TE = 9.8 ms, FOV = 256 mm, FA = 8°, 4-mm isotropic resolution, 48 slices) was acquired, and T1 maps were calculated using this and the multiple GRE scans in Medical Image Processing, Analysis and Visualization 8.0.2 (MIPAV, Version 8.0.2; https://mipav.cit.nih.gov).

Seven of the 9 participants received gadolinium as part of other National Institutes of Health MR imaging studies, and those scans were available for retrospective analysis. The gadolinium scan included here was from a T1-weighted GRE scan (TR = 8.1 ms, TE = 2 ms, FOV = 256 mm, FA = 18°, 1-mm isotropic resolution, 176 slices).

One participant (healthy volunteer 4) was excluded from analysis due to an incorrect sequence acquisition. Images were interpreted clinically by an experienced neuroradiologist (D.S.R.).

Blood Work

Laboratory testing consisted of complete blood count, comprehensive metabolic panel with liver function tests and studies of creatine kinase, prolactin, thyroid stimulating hormone, and urine pregnancy (when relevant). These tests were performed at screening, week 1, month 1, and subsequent scans.

Image Processing

Image analysis was conducted using the freely available image-analysis software MIPAV. For each participant, volumes of interest (VOIs) covering the anterior pituitary gland (entire gland) and choroid plexus (3 consecutive axial slices) were manually drawn on the baseline unenhanced T1-weighted GRE (FA = 18°) scan and transferred to registered postmangafodipir images (On-line Fig 1). Exocrine glands (lacrimal, ∼3 slices; submandibular, ∼14 slices; sublingual, ∼16 slices; parotid, ∼6 slices) were manually segmented in a similar manner at each postmangafodipir time point due to their variable configuration at each scan. The VOIs of basal ganglia structures (caudate, striatum, pallidum), thalamus, cerebral cortex, and whole brain were obtained through segmentation using the semiautomated segmentation pipeline FreeSurfer (Version 5.1.0; http://surfer.nmr.mhg.harvard.edu). The mean and SD of signal intensity from T1-weighted images and the median T1 relaxation time from T1 maps were calculated in each VOI. T1-weighted signal intensity was normalized to the signal intensity from the manually segmented corpus callosum (∼11 slices) at each time point, and the fold change was calculated by dividing the normalized postmangafodipir intensity values by the normalized baseline intensity at each time point. The corpus callosum was selected for normalization due to the following: 1) prior animal and human data indicating that it does not enhance with manganese; and 2) the ease of segmentation due to its discrete borders on midline and paramedian sagittal slices.^13,15

T1 mapping was successful in 7 participants for quantification of the anterior pituitary gland, choroid plexus, exocrine glands, and corpus callosum; 5 participants had successful T1 mapping for quantification of basal ganglia substructures.

Statistical Analysis

Statistical analysis was conducted using SAS (Version 9.2; SAS Institute, Cary, North Carolina). For each structure variable (signal intensity and T1 times), the linear trapezoidal rule was used to calculate the area under the enhancement curve. The area under the curve used for analysis was the difference between the area under the enhancement curve and the area under the baseline (interval between baseline and screening scan) divided by the time length.

The enhancement was evaluated using both the Wilcoxon signed rank test and bootstrapping with 500 replicates. In addition, a piecewise linear mixed-effects model was applied to examine the pattern of manganese enhancement for each structure.

Results

Nine healthy volunteers (median age, 43 years; range, 27–55 years; 3 women) participated in this study. MR imaging time points after mangafodipir administration are listed in the Table. As described in further detail below, we noted time-varying enhancement (P < .05) of several intracranial and extracranial structures, particularly the choroid plexus, anterior pituitary gland, and exocrine glands of the head and neck (parotid, sublingual, submandibular, and lacrimal).

Choroid Plexus

On administration of mangafodipir, T1-weighted signal intensity of the choroid plexus increased within the first hour (Figs 1 and 2). In the 2 healthy volunteers scanned within the first 2 hours, enhancement was noted both quantitatively and qualitatively within the first 5 minutes after infusion (Fig 2), reaching maximum intensity in the 11- to 40-minute range. During the next 24 hours, the signal intensity decreased, achieving pre-infusion levels around 5–7 days (Figs 1 and 2). No further enhancement was noted for the remainder of the study period. Analysis of T1 maps did not show clear shortening of the T1 time (On-line Fig 2).

Fig 1.

Manganese uptake in intracranial and extracranial structures. Images from healthy volunteer 1 and healthy volunteer 2 show manganese enhancement on T1-weighted images. Time points for the anterior pituitary, lacrimal, parotid, and submandibular glands were at 5 minutes, 2 days, 7 days, and 30 days after mangafodipir administration; for the choroid plexus and sublingual gland, postmangafodipir time points were at 1 hour, 2 days, 7 days, and 30 days.

Fig 2.

Kinetics of manganese enhancement in selected intracranial and extracranial structures. Quantification of T1-weighted signal intensity, normalized to the corpus callosum and measured as a fold change relative to baseline in healthy volunteers (n = 8). Each dot represents a scan at the specified time point; in some cases, a single individual had several scans within the same time window. The population mean and standard error are also shown. All structures enhanced at P < .05.

Anterior Pituitary Gland

In the 2 healthy volunteers scanned within the first 2 hours, enhancement was noted both quantitatively and qualitatively (Fig 2), but enhancement peaked in the 1- to 24-hour range (Figs 1 and 2). For most volunteers, this signal persisted for at least a week and resolved by the month 1 time point. In 2 volunteers, enhancement of the anterior pituitary gland resolved by the week 1 time point. No further enhancement of the anterior pituitary gland was observed for the remainder of the study period. Analysis of T1 values did not show a statistically significant decrease, in part because fewer data points were available for analysis (On-line Fig 2), but data suggest a possible nadir in the 1- to 24-hour range.

Exocrine Glands

The parotid, sublingual, submandibular, and lacrimal glands all demonstrated signal enhancement after mangafodipir infusion in the 1- to 24-hour range (Figs 1 and 2). The parotid and lacrimal glands had peak enhancement at approximately 5–7 days, whereas the sublingual and submandibular glands had peak enhancement in the 1- to 24-hour range (Fig 2). Probably due to their position at the edge of the head coil, T1-weighted signal in the exocrine glands was much more variable compared with enhancing intracranial structures. Whereas most volunteers had resolution of exocrine gland enhancement by month 1, two volunteers had resolution by month 2, and 1 volunteer had resolution by month 3. Quantitative T1 maps contained artifacts within several exocrine gland structures, which rendered analysis of T1 values highly variable (On-line Fig 2).

Brain

Relative to the corpus callosum, there was no evidence of manganese retention in the analyzed brain structures at various postmangafodipir time points and at month 1 (Fig 3). In particular, signal intensities in the caudate, putamen, globus pallidus, thalamus, cerebral cortex, and whole brain showed no clear change during the entire study period. Similarly, T1 maps, which are not normalized to the corpus callosum, showed no change in any of these structures or in the corpus callosum itself (On-line Fig 3).

Fig 3.

T1-weighted signal does not change in the basal ganglia, cerebral cortex, or whole brain after mangafodipir administration. Quantification of T1-weighted signal intensity, normalized to the corpus callosum and measured as a fold change relative to baseline in healthy volunteers (n = 8), in selected brain structures. Each dot represents a scan at the specified time point, summarized as mean ± standard error.

Adverse Reactions

There were no serious adverse reactions with mangafodipir administration. One volunteer reported a transient sensation of flushing and warmth, which did not require intervention. No additional symptoms were noted by the participants for the study period. Laboratory values did not significantly change after mangafodipir administration during the study period, at either week 1 or month 1.

Time Course of Manganese Enhancement

The time courses of manganese-induced signal enhancement and washout in the structures with significant changes with time are described in On-line Fig 4. For all 6 structures, the slopes before and after the knot were nonzero for the bootstrapped data. Although rates of enhancement were similar across structures, rates of washout were slowest in the exocrine glands.

Discussion

Manganese has shown promise as a functional intracellular MR imaging contrast agent through its ability to evaluate cellular integrity, activity, and neural connectivity. Manganese ion has been used as an imaging biomarker for cellular viability and structural integrity, as a surrogate for the cellular calcium ion transport, and for neural tract visualization. In this study, manganese ions released from mangafodipir in healthy volunteers resulted in enhancement of several intracranial and extracranial structures: the choroid plexus, anterior pituitary gland, and exocrine glands of the head and neck. Structures covered with a blood-brain barrier did not exhibit visible manganese enhancement. Mangafodipir was well-tolerated, without serious adverse effects or changes in laboratory values.