Physics in Medicine & Biology

PAPER You may also like

- Cationic radionuclides and ligands for
SPECT imaging of 226Ac as a theranostic isotope targeted therapeutic radiopharmaceuticals
Bayirta V. Egorova, Olga A. Fedorova and

for 225Ac radiopharmaceutical development

Stepan N. Kalmykov

- Production of medical isotopes from a

thorium target irradiated by light charged
To cite this article: Helena Koniar et al 2022 Phys. Med. Biol. 67 185009 particles up to 70 MeV
C Duchemin, A Guertin, F Haddad et al.

- A nephron-based model of the kidneys for

macro-to-micro -particle dosimetry
Robert F Hobbs, Hong Song, David L
View the article online for updates and enhancements. Huso et al.

This content was downloaded from IP address 193.54.110.57 on 12/01/2023 at 14:41

 Phys. Med. Biol. 67 (2022) 185009 https://doi.org/10.1088/1361-6560/ac8b5f

PAPER

RECEIVED
SPECT imaging of 226Ac as a theranostic isotope for 225Ac
29 April 2022
REVISED
radiopharmaceutical development
2 August 2022
ACCEPTED FOR PUBLICATION
Helena Koniar1,2 , Cristina Rodríguez-Rodríguez2,3 , Valery Radchenko1,4, Hua Yang1,5 , Peter Kunz5,6,
19 August 2022 Arman Rahmim2,7 , Carlos Uribe7,8 and Paul Schaffer1,5,7
PUBLISHED 1
Life Sciences Division, TRIUMF, Vancouver, Canada
12 September 2022 2
Department of Physics and Astronomy, University of British Columbia (UBC), Vancouver, Canada
3
Faculty of Pharmaceutical Sciences, University of British Columbia (UBC), Vancouver, Canada
4
Department of Chemistry, University of British Columbia (UBC), Vancouver, Canada
5
Department of Chemistry, Simon Fraser University, Burnaby, Canada
6
Accelerator Division, TRIUMF, Vancouver, Canada
7
Department of Radiology, University of British Columbia (UBC), Vancouver, Canada
8
Functional Imaging, BC Cancer, Vancouver, Canada
E-mail: hkoniar@triumf.ca

Keywords: actinium-226, actinium-225, targeted alpha therapy, radiopharmaceutical therapy, theranostic pair, preclinical imaging, SPECT

Abstract
Objective. The development of alpha-emitting radiopharmaceuticals using 225Ac (t½=9.92 d)
benefits from the quantitative determination of its biodistribution and is not always easy to directly
measure. An element-equivalent matched-pair would allow for more accurate biodistribution and
dosimetry estimates. 226Ac (t½=29.4 h) is a candidate isotope for in vivo imaging of preclinical 225Ac
radiopharmaceuticals, given its 158 keV and 230 keV gamma emissions making it suitable for
quantitative SPECT imaging. This work aimed to conduct a performance assessment for 226Ac
imaging and presents the first-ever 226Ac SPECT images. Approach. To establish imaging performance
with regards to contrast and noise, image quality phantoms were scanned using a microSPECT/CT
system. To assess the resolution, a hot rod phantom with cylindrical rods with diameters between 0.85
and 1.70 mm was additionally imaged. Two collimators were evaluated: a high-energy ultra-high
resolution (HEUHR) collimator and an extra ultra-high sensitivity (UHS) collimator. Images were
reconstructed from two distinct photopeaks at 158 keV and 230 keV. Main results. The HEUHR
SPECT image measurements of high activity concentration regions were consistent with values
determined independently via gamma spectroscopy, within 9% error. The lower energy 158 keV
photopeak images demonstrated slightly better contrast recovery. In the resolution phantom, the
UHS collimator only resolved rods …1.30 mm and …1.50 mm for the 158 keV and 230 keV
photopeaks, respectively, while the HEUHR collimator clearly resolved all rods, with resolution <0.85
mm. Significance. Overall, the feasibility of preclinical imaging with 226Ac was demonstrated with
quantitative SPECT imaging achieved for both its 158 keV and 230 keV photopeaks. The HEUHR
collimator is recommended for imaging 226Ac activity distributions in small animals due to its
resolution <0.85 mm. Future work will explore the feasibility of using 226Ac both as an element-
equivalent isotope for 225Ac radiopharmaceuticals, or as a standalone therapeutic isotope.

1. Introduction

Radiopharmaceutical therapies (RPT) with alpha-emitting radionuclides can provide effective treatment for
aggressive cancers by directly targeting cancerous cells while minimizing damage to neighbouring healthy tissue
(Sgouros 2020). This is due to the highly localized energy deposition and short range of alpha radiation (Kim and
Brechbiel 2012). 225Ac (t½=9.92 d) has considerable potential for therapeutic use due to its successive four
alpha emissions in its decay chain (Miederer et al 2008). Figure 1 sumarizes the 225Ac decay chain and its physical

© 2022 Institute of Physics and Engineering in Medicine

Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 1. The 225Ac (top) and 226Ac (bottom) decay chains. Photons with branching rations >6% are shown.

properties. While the theraputic potency of 225Ac greatly benefits from successive alpha emissions, dosimetry
estimates are complicated by potential daughter migration away from the targeted site (de Kruijff et al 2015,
Tronchin et al 2022). Imaging-based pharmacokinetic testing is required to understand biodistribution and
estimate dosimetry in vivo (Sgouros et al 2021). However, 225Ac itself does not have any imageable gamma or
positron emissions (McDevitt et al 1998). Indirect single photon emission computed tomography (SPECT)
imaging is only possible with emissions from its progeny radionuclides 221Fr (t1/2=4.80 m) and 213Bi
(t1/2=45.6 m) (Robertson et al 2017), and relies on the assumption that both 225Ac and its progeny isotopes are
retained at the target site. Long-lived progeny in the 225Ac decay chain can migrate from the target site before
they decay; and while they are imagable, 221Fr and 213Bi have been shown to accumulate in the kidneys, urine and
other organs (Beninson et al 1979, Jaggi et al 2005, Schwartz et al 2011, de Kruijff et al 2015, 2019). Thus, there
are inherhent errors introduced when imaging progeny nuclides to quantify the biodistribution and dosimetry
of 225Ac. Moreover, injected activities of 225Ac are low resulting in very few gamma detections from 221Fr and
213
Bi, which makes it difficult to obtain a reasonable image within a clinical time frame. Due to a lack of an
element-equivalent imageable isotope of 225Ac, performance evaluation of 225Ac radiopharmaceuticals is
currently assessed by either performing invasive organ-harvesting biodistribution studies in preclinical models.
In clinical settings, 225Ac radiopharmacuticals are paired with an imageable isotope, such as 68Ga (Baum and
Kulkarni 2012), 177Lu (Kratochwil et al 2016), 111In (Kelly et al 2020) and 89Zr (Solomon et al 2020). Absorbed
dose is inferred with the underlying assumption of equivalent uptake and residency between 225Ac and
imageable forms of a radiopharmaceutical using other isotopes. Imaging with alternative isotopes is effective for
therapeutic target verification; however, obtaining direct and quantitative dosimetry information can be
challenging because of potential changes in the pharmacokinetics for two separate compounds containing non-
equivalent isotopes (Miller et al 2022). The importance of element-equivalent matched pairs has been
demonstrated in yttrium with 86Y/90Y (Rösch et al 2017), astatine with 211At/209At (Crawford et al 2018) and
lead with 203Pb/212Pb (McNeil et al 2021).
The availability of an element-equivalent matched-pair of actinium isotopes would allow for more accurate
biodistribution and dosimetry estimates. Quantitative pharmacokinetic information can be used to optimize the
therapeutic properties of a preclinical radiopharmaceutical. 226Ac (t½=29.4 h) is one candidate for in vivo
imaging studies of 225Ac-labeled compounds. Figure 1 summarizes the 226Ac decay chain and its physical
properties. First, element-equivalence maintains pharmacokinetic properties in vivo, and second, the emission
of gamma photons (158 keV and 230 keV) from the decay to 226Th are ideal energies for quantitative SPECT
imaging. 226Ac also holds promise as a therapeutic isotope in itself because of its rapidly decaying progeny after
226
Th accounting for a total of four alpha emissions. The theraputic potential of 226Th has been recognized with
prelimiary studies investigating in vivo 230U/226Th generators (Morgenstern et al 2008, Ferrier et al 2020, Friend
et al 2020). Current availability of 226Ac is limited, however, TRIUMF (Vancouver, Canada) can produce
quantities suitable for preclinical imaging studies.

2
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

This work aimed to conduct a performance assessment for 226Ac quantitative SPECT imaging and, to the
best of our knowledge, is the first attempt to image 226Ac, and as such there are currently no established methods
for imaging 226Ac. SPECT images of 225Ac progeny, 221Fr (218 keV; 11.4%) and 213Bi (440 keV; 25.9%), have
been acquired with a VECTor microSPECT/PET/CT system (Robertson et al 2017). This previous work serves
as a basis for our performance assessment of VECTor for imaging 226Ac decay via its 158 keV (17.5%) and 230
keV (26.9%) gamma emissions. The in vivo applications of this novel imaging technique could provide
invaluable pharmacokinetic information for evaluating preclinical Ac-based radiopharmaceutical properties
such as rate of accumulation and uptake at the target site, radiation dosimetry estimates at the tumour for
efficacy, and dosimetry to surrounding normal tissue for safety. Ultimately, this imaging technique could
facilitate faster development of 226Ac/225Ac radiopharmaceuticals in the preclinical stage by optimizing which
are best suited for investigating with future therapy studies.

2. Materials and methods

2.1. 226Ac physical properties, production, and quantification

226
Ac decays to 226Th via β-decay with a branching ratio of 83% (see figure 1). From 226Th, the decay chain
cascades through four alpha emissions to stable 210Pb, which could be leveraged for their therapeutic effects.
From this decay scheme, only the initial decay of 226Ac is imageable via either its 158 keV (17.5%) and 230 keV
(26.9%) gamma emissions. Other gamma emissions occur with branching ratios <6% making it challenging to
generate images representing activity distributions of specific isotopes.
TRIUMF (Vancouver, Canada) can produce small quantities of 226Ac suitable for preclinical studies with
high purity and specific activity. With TRIUMF’s ISAC (Isotope Separation and Acceleration) facility, ISOL
(isotope separation on-line) targets are irradiated with up to 100 μA of a 480 MeV proton beam from TRIUMF’s
main cyclotron (Kunz et al 2020, Fiaccabrino et al 2021). Detailed information on isotope yields is available in
the ISAC Isotope Database (Kunz 2022). A full description of this 226Ac production method will be presented
under a separate cover, and is only briefly reported here. To produce 226Ac at ISAC, a uranium carbide target was
bombarded with a continuous 22 μA proton beam with an energy of 480 MeV resulting in a heterogenous ion
beam with a wide range of spallation and fission products. 226Ac was seperated by its mass over charge ratio with
a mass seperator magnet to produce an A=226 beam. With an energy of 20 keV, the A=226 beam was
deposited in a thin ammonium chloride layer coating an alumium implantation target. The ion beam current
during the 12 h implantation period was on average 45 pA. The main beam components at A=226 include the
elements Ac, Ra and Fr. Isotopes of other elements (Th, Pa, U) that may be produced are not released from the
uranium carbide target due to the formation of non-volatile carbides. Through gamma and alpha spectroscopy
the yield of 226Ac was determined as ∼1 × 108 ions s−1 (∼16 pA). 226Fr has a half-life of 48 s and decays quickly
into 226Ra. Therefore, the remaining 29 pA of the total ion beam current, accumulated over the 12 h
implantation period, resulted in the collection ∼8 × 1012 atoms or ∼110 Bq of 226Ra. The salt layer and
implanted isotopes were desolved from the implantation target with small volumes (<0.5 ml) of water. For this
study, a total of 33.8±2.7 MBq of 226Ac was retrieved and available for imaging, with high radionuclidic purity.
The extrapolated activity of 110 Bq of co-implanted 226Ra is negligable and, since 226Ra is also a daughter isotope
of 226Ac, no purification steps were taken for this phantom imaging study. For future in vivo studies, 226Ac and
226
Ra can be seperated with a DGA column to increase radionuclide purity prior to labelling (Zielinska et al 2007,
Mastren et al 2017, Radchenko et al 2017).
Samples of the radioactive solution were measured with a high-purity geranium (HPGe) detector (Canberra
Industries, Meriden, CT) and activity concentrations were determined by gamma spectroscopy with the Genie
2000 software package (Canberra Industries, Meriden, CT). Detector dead time was <0.5% and counts were
acquired for 15 min. Activity measurements have an estimated uncertainty of 8%, introduced from calibration
source uncertainty and volume measurement uncertainty when collecting samples (Crawford et al 2017).

2.2. SPECT acquisition and image reconstruction

Imaging was performed with the Versatile Emission Computed Tomography (VECTor) micro PET/SPECT/CT
imaging system (MILabs, Utrecht, Netherlands) (Goorden et al 2013). The imaging system is capable of both
SPECT/PET imaging with three large-field-of-view NaI gamma detectors in a triangular arrangement and x-ray
computed tomography (CT) scanning. Cylindrical collimators with rings of clustered pinholes can be
interchanged and optimized for resolution and sensitivity to best suite different energies from annihalation
photon or gamma emissions and various preclinical applications. The clustered pinhole design of the
collimators used with VECTor apply a focusing gemoetry on a central scanning volume that allows higher
sensitivity to a specific organ or tumour (Goorden et al 2013). The bed is scanned in the axial direction allowing

3
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Table 1. Phantom imaging experiments, with total activity filled, acquisition duration, and total counts in each photopeak energy window.

Total scan Counts

Phantom Total activity (MBq) Collimator Bed positions time (min) (millions)

Point source 10.14±0.81 HEUHR 18 35 158 keV 8.39

UHS 10 15 230 keV 20.38
158 keV 16.60
230 keV 9.64
Contrast and
uniformity
Hot insert 2.72±0.22 HEUHR 33 60 158 keV 3.75
230 keV 4.10
Warm insert 9.53±0.76 UHS 17 60 158 keV 18.77
230 keV 23.90
Resolution 7.14±0.57 HEUHR 19 60 158 keV 3.49
230 keV 3.86
UHS 8 60 158 keV 18.80
230 keV 23.58

the central field of view (CFOV) to collect photons throughout the entire volume of interest with a scanning
focus method of SPECT acquisition (Vastenhouw and Beekman 2007).
While MILabs produces many cylindrical collimators compatible with VECTor, two collimators were
investigated for 226Ac imaging and performance evaluation. First, the extra ultra-high sensitivity (UHS)
collimator was investigated since it is ideal for low energy (<350 keV) and provides higher sensitivity favourable
for low-activity imaging (Ivashchenko et al 2015a). The UHS collimator is a lead cylinder with an inner diameter
of 46 mm and consists of 54 conical pinholes with diameter 2.0 mm positioned in four rings. The CFOV size is
12 × 12 × 7 mm. For the UHS collimater, the peak sensitivity was 13 080 cps MBq−1 and reconstructed spatial
resolution was 0.85 mm when assessed with 99mTc. Second, the high energy ultra-high resolution (HEUHR)
collimator was investigated since it is ideal for a broad energy range and provides higher spatial resolution (Miwa
et al 2015). The HEUHR collimater is a tungsten cylinder with an inner diameter of 44 mm and consists of 48
clusters of four pinholes with diameter 0.7 mm positioned in four rings. The CFOV size is 12 × 12 × 9 mm. For
the HEUHR collimator, the peak sensitivity was 2800 cps MBq−1 and 2899 cps MBq−1 and the reconstructed
spatial resolution was 0.5 mm and 0.8 mm when assessed with 99mTc and 18F, respectively. We characterize both
the HEUHR and UHS collimators ability to image 226Ac to determine which is best suitable for future in vivo
applications in mice models.
The high activity of 226Ac available for our imaging study permits shorter data acquisition times to obtain
sufficient counts for image reconstruction. These shorter data acquisition times could help to facilitate in vivo
imaging studies. Table 1 reports the collimator, number of bed positions, scan duration, and total counts in each
photopeak for each image acquisition. The total phantom scan times in this study are all 1 h or less, which is
promising for future in vivo experiments when scan times are restricted by animal welfare considerations around
anesthesia. Further, for comparable count statistics with the 225Ac progeny 221Fr and 213Bi, the total scan times
required are much longer, ∼6 h total, and is not as well suited for animal models (Robertson et al 2017).
Data was acquired in list mode and sorted into 512 energy bins with 2.34 keV width. Figure 2 illustrates the
resulting energy spectrum for 226Ac from scanning a calibration point source. Two distinct photopeaks in the
energy spectrum corresponding to 226Ac’s 158 keV and 230 keV emissions were defined with energy windows of
18% width for individual image reconstruction. VECTor’s pixel-based ordered subset expectation maximation
(POSEM) iterative reconstruction algorithm was used for all image reconstructions (Branderhorst et al 2010).
SPECT images were reconstructed using a system matrix optimized for 140 keV photons. Decay corrected
images with 0.4 mm wide voxels were reconstructed with 16 subsets and 3 iterations. This combination of
subsets and iterations optimized the contrast-to-noise ratio in resolution phantom images. With 48 MLEM-
equivalent iterations (1 subset, 48 iterations), the contrast-to-noise ratio for most images was minimized across
different photopeaks and collimator combinations. Therefore, 16 subsets and 3 iterations were chosen for all
image reconstructions to optimize image quality and reconstruction time. Using the triple energy window
(TEW) method, SPECT images were corrected for background and scatter with a high and low energy
background window of 3% width. Attenuation correction was implemented by co-registration of subsequently
acquired CT images with the resulting SPECT images. After co-registration with CT images, the SPECT images
were resampled to 0.17 mm wide voxels.
Reconstructed images were analyzed with scripts written in Python (v3.8.8, https://python.org/) using the
NiBabel (v3.2.1, https://nipy.org/nibabel/) package to read the SPECT and CT NIfTI files exported by

4
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 2. 226Ac energy spectra acquired by the VECTor scanner for both HEUHR (blue) and UHS (green) collimators, normalized by
acquisition time and decay corrected activity. Energy spectra were acquired with a uniform point source containing 10.14 ± 0.81 MBq
of 226Ac. The energy windows used to reconstruct the 158 keV and 230 keV gamma emissions are shaded in blue and red, respectively.
The grey shaded regions indicate the energy windows defined to perform background and scatter correction. The low energy peak ∼90
keV in the spectra is due to a combination of characteristic x-rays emissions from the collimator materials (tungsten and lead) and
backscatter and Compton edges from the 158 keV and 230 keV gammas.

VECTor, the Matplotlib (v3.5.1, https://matplotlib.org/) Python library for creating visualizations, the scikit-
image (v0.18.1, https://scikit-image.org/) package for defining regions of interest (ROIs) and line profiles, and
the NumPy (v1.22.3, http://numpy.scipy.org/) Python library for data analysis.
To enable quantitative SPECT images, calibration factors for each collimator and photopeak combination
were determined. A uniform 226Ac source in an Eppendorf tube containing 50.7±4.1 MBq ml−1 of 226Ac was
scanned with 35 min and 15 min acquisitions for the HEUHR and UHS collimators, respectively. ROIs were
drawn in the SPECT images of the point source to correlate the average voxel intensity with the activity
concentration determined by gamma spectroscopy for each collimator and photopeak combination. These
calibration factors were applied to the SPECT images of the contrast and uniformity phantom and the resolution
phantom to directly measure activity concentrations in their voxels. These calibrations can also make future
SPECT images quantitative when they are reconstructed with the same methods.

2.3. Image contrast and uniformity phantom

Image contrast and response uniformity were assessed with a SPECTIQ cylindrical acrylic phantom, which is a
smaller version of the NEMA NU-4 micro-PET image quality phantom (Harteveld et al 2011, Visser et al 2011).
This phantom contains various activity concentrations to assess the imaging systems contrast abilities and
uniform response. The contrast region of the SPECTIQ phantom consists of two small cyclindrical insterts: a
‘hot’ region with high activity concentration and an activity-free ‘cold’ region. The hot and cold inserts are both
encased by a ‘warm’ background region with low activity concentration. These conditions recreate
concentrations that are seen in vivo, where a small region with high activity, like a tumour or bladder, is
surrounded by a background with low activity concentration. The hot and warm regions were filled with 226Ac
concentrations of 6.76±0.54 MBq ml−1 and 1.46±0.12 MBq ml−1, respectively. The uniformity region of the
SPECTIQ phantom consists of only the warm cylinder, and the hot and cold cylinders do not extend into this
region.
Contrast measurements were determined as described by Robertson et al (2017) and are briefly summarized
here. Cylindrical ROIs were centred over the hot, warm, and cold regions with diameters 85% the physical
diameter and positioned using CT images (see figure 3(a)). For the quantitative accuracy and contrast recovery
calculations and image line profile measurements only, a 1 mm FWHM Gaussian filter was applied to the
reconstructed SPECTIQ phantom images. Unflitered images were used to assess noise and background
variablity. Quantitative accuracy was calculated by the ratio of mean activity concentration measured in the hot
and warm ROIs to the activity concentration determined intependently with gamma spectroscopy. Contrast
recovery between different activity concentrations was calculated as defined by the NEMA NU 2-2007 standards
(National Electrical Manufacturers Association 2007). Contrast recovery between the hot and warm region was
defined as:
Ih / Iw - 1
CRh - w = *100%
ah / a w - 1

5
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 3. Contrast and uniformity phantom. (left) CT image cross section of the contrast region’s hot and cold inserts. (centre) SPECT
image cross section of the contrast region’s hot and cold regions; the blue ROIs were used to calculate contrast metrics and the red line
was used to measure intensity profiles. (right) SPECT image cross section of the uniformity region with low warm activity, with blue
ROIs used for noise calculations, and green ROIs with diameters equal to the hot and cold inserts used for background variability
calculations.

and contrast recovery between the warm and cold regions was defined as:
I
CR w- c = ⎛1 - c ⎞*100%,
⎜ ⎟

⎝ I w⎠

where Ih, Iw , and Ic represent the mean activity concentrations measured within the hot, warm, and cold ROIs,
respectively, and ah and a w represent the activity concentrations determined by gamma spectroscopy
independently. Contrast insert profiles across the hot and cold regions were also measured on a single image
plane to analyze their intensity profiles (see figure 3(b)). Image uniformity was assessed with radial line profiles
through the uniform warm region. Due to the low activity concentration in this region, images were averaged
axially over 25 adjacent planes (5 mm) before the uniformity region line profile was measured. A radial line
profile through the contrast region of the phantom was also measured from a single image plane.
Image response variability to uniform activity concentration was also assessed within the phantom’s
uniformity region with two metrics. Firstly, noise was assessed as the coefficient of variation (COV) of the
average voxel intensity in the warm ROI (blue circle in figure 3(c)) across 45 adjacent planes (7.6 mm). Secondly,
background variability was assessed with 9 ROIs (green circles in figure 3(c)) located within the warm region, in a
method similar to Tong et al (2010). Background variability was calculated as the COV of the mean value from
ROIs across 6 non-adjacent image slices 1.86 mm apart, for a total of 54 ROIs.

2.4. Image resolution phantom

A hot rod phantom withclusters of cylindrical holes with diameters 1.70 mm, 1.50 mm, 1.30 mm, 1.10 mm, 0.95
mm, and 0.85 mm was used to assess the imagin system’s ability to resolve small objects. All the rods were filled
with 16.9±1.3 MBq ml−1 of 226Ac. Walker et al (2014) defined inter-rod contrast metrics that were used to
analyze the resulting SPECT images and quantify the resolvability of small hot regions. Circular ROIs were
placed on and between the rods as defined within CT images, with diameters 80% the physical diameter (see
figure 4). Measurements were repeated on 35 planes, for a total axial thickness of 5.9 mm. The inter-rod contrast
is defined as:
h¯d - b¯d
Cd = ,
h¯d
where d is an index of the rod diameter, h¯d is the ROI mean for all planes and rods with diameter index d placed
on top of rods, and likewise b¯d is the ROI mean for ROIs placed in between the rods. The ideal value for this
metric is 100%, however, rods with contrast greater than 20% are visually well resolved.
To assess the imaging system’s accuracy in quantifying activity concentrations in small regions, the recovery
coefficient (RC) was measured for each cluster of rods. The RC is defined as:
h¯d
RC = ,
h0
where h 0 is the activity concentration determined with gamma spectroscopy and h¯d is the mean activity
concentration measured within the ROIs for rods with diameter d. The ideal value for this metric is 1.
To assess the effect of lower total counts, either from lower activity concentrations or shorter scan times,
images of the resolution phantom were reconstructed using 20%, 5%, and 1% of the total counts, simulating

6
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 4. (left) CT image cross section of the resolution phantom’s hot rod clusters. (right) SPECT image cross section of the
resolution phantom, with red ROIs representing hot regions and blue ROIs representing cold regions used for inter-rod contrast and
recovery coefficient calculations.

Figure 5. 158 keV and 230 keV SPECT images acquired with both the HEUHR and UHS collimators of the SPECTIQ phantom’s
contrast region (top) and uniformity region (bottom). Images were filtered with a 1 mm FWHM Gaussian filter.

data acquired with activity concentrations of 3.37, 0.84, and 0.17 MBq ml−1. Inter-rod contrast and RC were also
assessed in these images.

3. Results

3.1. Image contrast and uniformity

Figure 5 shows a single plane of SPECT images in the contrast region and uniformity region of the SPECTIQ
phantom and figure 6 shows intensity profiles through the these two regions. Quantitative accuracy, contrast
recovery, and noise measurements from SPECT images of the contrast and uniformity phantom acquired with
the HEUHR and UHS collimators are summarized in table 2.
For the HEUHR collimator SPECT images, the activity concentration measured in the hot region insert was
within 8.3% of the expected values determined independently with gamma spectroscopy. The error in
quantitative accuracy was greater for the UHS collimator SPECT images, up to 16.7%. The activity
concentrations measured in the uniform warm region differed from the gamma spectroscopy values <11.8% for
all photopeak reconstruction and collimator combinations.
Contrast recovery between the hot–warm region was highest for the HEUHR collimator, with 99.3%–
100.1% contrast recovery compared to 86.6%–92.9% contrast recovery for the UHS collimator. In both
collimators, the lower energy 158 keV images had slightly better contrast recovery compared to the 230 keV
images. Contrast recovery between the warm–cold regions were also higher in the HEUHR collimator (86.5%–

7
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 6. Activity concentration profiles through the uniformity region (left) and contrast region (right). The ‘true’ profile in solid grey
represents the ideal line profile with the activity concentration determined via gamma spectroscopy and its associated error in dashed
grey. Profiles from 158 keV and 230 keV photopeak images are shown in blue and red, respectively.

Table 2. Image contrast and uniformity results, with expected ideal values.

Photopeak Quantitative accuracy Contrast recovery Uniformity

Collimator energy
Hot region Warm region Hot– Warm– Background varia-
mean (%) mean (%) warm (%) cold (%) Noise (%) bility (%)

HEUHR 158 keV 91.7 91.6 100.1 86.5 1.7 7.1

230 keV 91.9 92.4 99.3 87.9 2.8 7.9
UHS 158 keV 83.3 88.2 92.9 86.1 4.4 6.4
230 keV 84.5 94.4 86.6 83.7 2.5 6.3
Ideal 100 100 100 100 0 0

87.9%) than the UHS collimator (83.7%–86.1%). Images of the contrast phantom (figure 5) visually support
these observations.
Image noise across several planes in the uniform warm region was lowest for the HEUHR collimator 158 keV
image, with 1.7% response variability. However, background variability was slightly higher in the HEUHR
collimator (7.1%–7.9%) compared to the UHS collimator (6.3%–6.4%).

3.2. Image resolution

Figure 7 shows SPECT images of the resolution phantom. Visually it is clear that the HEUHR collimator clearly
resolves all rods for both the 158 keV and 230 keV reconstructions, indicating a resolution <0.85 mm. As seen in
figure 8, inter-rod contrast values are >50% for all rod diameters with the HEUHR collimator, and peaked at
91% with the 1.70 mm diameter rods, in both 158 keV and 230 keV images. The UHS collimator resolves rods
…1.30 mm and …1.50 mm for the 158 keV and 230 keV images, respectively. Inter-rod contrast values for the
UHS collimator peak for the 1.70 mm rod at 60% and 75% for the 158 keV and 230 keV, respectively. By
reconstructing with fewer counts, the inter-rod contrast and resolvability of the rods remains unchanged as the
activity concentration is decreased to 0.17 MBq ml−1 (see figure 10).
Recovery coefficients (RC) for all resolved rods are shown in table 3. RC values range between 0.44 and 0.82
for the largest 1.70 mm rod and decreases with smaller rod diameters. The 158 keV images have slightly higher
RC values than the higher energy 230 keV images for both the HEUHR and UHS collimators. Profiles through
rods of diameter 0.95 mm, 1.30 mm, and 1.70 mm are shown in figure 9 and can be compared to the known
activity concentration of 16.9±1.3 MBq ml−1. As the activity concentration is decreased, RC values changed
very little (see figure 11).

4. Discussion

Herein we demonstrate the ability to quantitatively image 226Ac. Overall, the HEUHR collimator showed the
most optimal image characteristics with spatial resolution <0.85 mm (figure 7) and quantitative accuracy within

8
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 7. SPECT images of the resolution phantom reconstructed from the 158 keV and 230 keV photopeaks acquired with both the
HEUHR and UHS collimators.

Figure 8. Inter-rod contrast measurements from the SPECT images of the resolution phantom.

Figure 9. Activity concentration profiles through 0.95 mm diameter rods (left), 1.30 mm diameter rods (centre), and 1.70 mm
diameter rods (right). The activity concentration measured within each rod from SPECT images can be compared to the known value
of 16.9 ± 1.3 MBq ml−1.

Table 3. Recovery coefficients from the resolution phantom. Values omitted for rods that are not
resolved (inter-rod contrast <0.2).

Rod diameter (mm)

Collimator Photopeak energy 0.85 0.95 1.10 1.30 1.50 1.70

HEUHR 158 keV 0.34 0.43 0.56 0.64 0.80 0.82

230 keV 0.34 0.41 0.55 0.65 0.76 0.81
UHS 158 keV — — — 0.31 0.44 0.55
230 keV — — — — 0.36 0.44

9
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 10. Inter-rod contrast at reduced activity concentrations, simulated by reduced count rates.

8.3% in regions with high activity concentrations (table 2). With the HEUHR collimator, the 158 keV and 230
keV reconstructions both demonstrated quantitative accuracy, high contrast recovery, and low noise and
background variation in the contrast phantom (see table 2). With respect to contrast metrics, 226Ac imaging also
outperforms SPECT images obtained from 225Ac’s progeny isotopes 221Fr (218 keV) and 213Bi (440 keV). As
characterized by Robertson et al (2017), with the HEUHR collimator, contrast recovery values peaked at 85.5%
for 221Fr while the values for 158 keV and 203 keV are 100.1% and 99.3%, respectively. Further noise is reduced
in the 226Ac images, with 1.7% in the HEUHR 158 keV images while the noise in 221Fr HEUHR was reported as
15%. This validates potential future applications of 226Ac for in vivo quantitative imaging since activity
concentrations can be accurately determined in small tumours with high uptake or critical organs with low
uptake. Quantitative accuracy is especially important for dosimetry estimates to optimize therapeutic effects.
The HEUHR collimator resolved all rods (0.85–1.7 mm) for both the 158 keV and 230 keV images, with
nearly identical inter-rod contrast values. Our HEUHR 158 keV and 230 keV images have comparable inter-rod
contrast values to those of 225Ac’s progeny nuclide 221Fr (218 keV) (Robertson et al 2017). The UHS collimator
only resolved rods 31.30 mm and 31.50 mm for the 158 keV and 230 keV images, respectively. Superior spatial
resolution with the lower energy 158 keV photon is expected, since spatial resolution has been shown to decrease
at higher energies for VECTor (Goorden et al 2013, Robertson et al 2017, Crawford et al 2018). These results are
consistent with other VECTor phantom imaging studies. With a hotrod phantom and the same HEUHR
collimator, Goorden et al (2013) found the resolution was <0.8 mm for the PET isotope 18F (511 keV) and <0.5
mm for the SPECT isotope 99mTc (140 keV). Since 226Ac’s two gamma emissions, 158 keV and 230 keV, are
close, but slightly larger in energy to that of 99mTc (140 keV), the resolvability of 226Ac is likely between 0.5 and
0.8 mm. Further imaging studies are needed to discern the resolvability limits of 226Ac. By reconstructing images
with reduced total count rates, the resolution capabilities are maintained above activity concentrations of 0.17
MBq ml–1. For future animal studies, the implications of this are important to consider as activity
concentrations are lower for in vivo studies than in phantom studies. Although these reduced activity
concentrations are achieved by artificially reducing the total count, these results could be repeated
experimentally with in vivo studies and lower injected activities.
Relative to the higher energy 230 keV image, the 158 keV image had slightly better recovery coefficients,
which is a metric for the accuracy of activity concentration measured within the rods. Given the scale of

10
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Figure 11. Recovery coefficients for rods at reduced activity concentrations.

anatomical structures in mouse models, reconstructing images with the 158 keV photopeak is advantageous for
achieving quantitative accuracy in dosimetry estimates. The RC values for 225Ac’s progeny nuclides 221Fr (218
keV) and 213Bi (440 keV) were reported as 0.73 and 0.97 for the 1.70 mm rods, respectively (Robertson et al
2017). The RC values measured from the 230 keV and 158 keV images (see table 3) have comparable values for
those of to those of 221Fr (218 keV), however, 213Bi (440 keV) has much higher RC values. The RC values for 226Ac
images are all <0.82 for all rod diameters, which can result in less quantitative accuracy and underestimating
dosimetry in mm-scale organs in animal studies. This is one limitation of the HEUHR collimator since it is
designed for higher energy photons (>350 keV). A collimator that is better suited for lower photon energies
could help to rectify this disadvantage. Collimators optimized for lower energy 99mTc (140 keV) SPECT imaging
have shown quantitative accuracy and can be evaluated in future work for their suitability in 226Ac imaging
(Ivashchenko et al 2015b).
These results support feasible in vivo imaging of 226Ac given the high number of counts and short acquisition
times needed for image reconstruction. An in vivo study with VECTor usually requires ∼106–107 photopeak
counts for image reconstruction. For mice, the number of bed positions scanned is larger than for the phantoms;
86 and 45 total for the HEUHR and UHS collimators, respectively. Additionally, the scanned volume is larger in
a mouse than in a phantom, resulting in lower activity concentrations. This increases acquisition time required
to achieve the targeted number of counts for image reconstruction. However, as indicated by the results from the
contrast phantom images, quantitative accuracy was >90% of expected values in both the hot and warm regions
despite 5-fold activity concentration difference between the regions. This indicates that with fewer total counts,
either by lower activity concentrations or shorter acquisition times, retains the quantitative accuracy in an in vivo
scan. The quanitative accuracy seen in SPECT image measurements is on the same order of precision as activity
measured via gamma spectroscopy with 8% relative uncertainty. Further, the results from reconstructing
resolution phantom images with reduced total counts indicate that acceptable image quality can be achieved
with activity concentrations …0.17 MBq ml−1. These results are expected given that with higher spatial
resolution in an imaging system, image contrast recovery improves rapidly, and fewer counts are required to
achieve the same image quality seen in a system with poorer spatial resolution (Muehllehner 1985).

11
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

The novel chelator crown has shown high tumour uptake and low liver accumulation when attached to the
peptides TATE and αMSH, targeting neuroendocrine and melanoma cancers, respectively (Yang et al
2020a, 2020b). With tumour accumulation >10% ID g−1, injected activities ∼2 MBq will achieve activity
concentrations >0.17 MBq ml−1 and produce high quality SPECT images within a 1 h scan time appropriate for
murine studies. Ex vivo biodistribution studies with 225Ac-crown compounds demonstrated %ID g−1 in
tumours appropriate for 226Ac SPECT imaging, and will be further investigated with in vivo 226Ac SPECT
imaging in future work. This activity for imaging is greater than therapeutic levels, limiting this imaging
modality to applications in pre-clinical radiopharmaceutical development. However, with future studies,
injected activities for imaging can be further optimized while maintaining high image quality.
Currently, 225Ac radiopharmaceuticals are typically paired with alternative imageable isotopes including
68
Ga, 177Lu, 111In, and 89Zr (Baum and Kulkarni 2012, Kratochwil et al 2016, Kelly et al 2020, Solomon et al
2020). 177Lu has been used for intra-therapy dosimetry of 225Ac as it is often administered in tandem therapies
(Khreish et al 2020, Rosar et al 2021). However, biodistribution studies have shown large differences between
225
Ac and 177Lu radiopharmaceuticals, which could severely underestimate the healthy tissue absorbed dose
(Graf et al 2014). In order to achieve quantitative accuracy of 225Ac dosimetry, 226Ac could be used for intra-
therapy dosimetry, since this element-equivalent pair has identical intrinsic chemical behaviour (Miller et al
2022). Furthermore, 226Ac would provide a therapeutic effect due to its four high-energy alpha emissions.
Future work will evaluate 226Ac dosimetry estimations, as this isotope is currently limited to preclinical
applications in murine models where dosimetry concerns are less significant than in clinical applications.
Further understanding of dosimetry and theraputic potency of 226Ac lends the option of operating at different
injected activities depending on whether the outcome is imaging- or therapy-based.

5. Conclusion

The fesability of SPECT imaging with 226Ac has been demonstrated with high quantitative accuracy and spatial
resolution. For small animal imaging, our results are conducive for imaging activity distributions in the
development of preclinical 225Ac radiopharmaceuticals. Future work will explore the feasibility of using 226Ac
both as an element-equivalent isotope for 225Ac radiopharmaceuticals and as a standalone therapeutic isotope.

Acknowledgments

This study was financially supported by the National Science and Engineering Research Council of Canada
(NSERC) Discovery Grants (SAPIN-2021–00030 (PK), RGPIN-2021–04093 (PS), RGPIN-2018–04997 (VR)),
Canada Foundation for Innovation (CFI) (project no. 25413), and the Government of Canada’s New Frontiers
in Research Fund—Exploration (NFRFE-2019–00128 (PK)). TRIUMF receives federal funding via a
contribution agreement with the National Research Council of Canada (NRC).

ORCID iDs

Helena Koniar https://orcid.org/0000-0003-0794-9115

Cristina Rodríguez-Rodríguez https://orcid.org/0000-0002-3313-4422
Hua Yang https://orcid.org/0000-0003-1833-9515
Arman Rahmim https://orcid.org/0000-0002-9980-2403
Carlos Uribe https://orcid.org/0000-0003-3127-7478
Paul Schaffer https://orcid.org/0000-0002-6392-8792

References
Baum R P and Kulkarni H R 2012 THERANOSTICS: from molecular imaging using Ga-68 labeled tracers and PET/CT to personalized
radionuclide therapy—the bad berka experience Theranostics 2 437–47
Beninson D, Dunster H, Jacobi W, Jammet H, Liniecki J, Meinhold C, Moiseev A, Rowley K, Sinclair W and Takahashi S 1979 ICRP
Publication 30 (Part 1): Limits for intakes of radionuclides by workers Ann. ICRP 2 67 https://icrp.org/publication.asp?id=ICRP%
20Publication%2030%20(Part%201)
Branderhorst W, Vastenhouw B and Beekman F J 2010 Pixel-based subsets for rapid multi-pinhole SPECT reconstruction Phys. Med. Biol.
55 2023–34
Crawford J R, Robertson A K H, Yang H, Rodríguez-Rodríguez C, Esquinas P L, Kunz P, Blinder S, Sossi V, Schaffer P and Ruth T J 2018
Evaluation of 209At as a theranostic isotope for 209At-radiopharmaceutical development using high-energy SPECT Phys. Med. Biol.
63 045025
Crawford J R, Yang H, Kunz P, Wilbur D S, Schaffer P and Ruth T J 2017 Development of a preclinical 211Rn/211At generator system for
targeted alpha therapy research with 211At Nucl. Med. Biol. 48 31–5

12
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

de Kruijff R M, Raavé R, Kip A, Molkenboer-Kuenen J, Morgenstern A, Bruchertseifer F, Heskamp S and Denkova A G 2019 The in vivo fate
of 225Ac daughter nuclides using polymersomes as a model carrier Sci. Rep. 9 11671
de Kruijff R M, Wolterbeek H T and Denkova A G 2015 A critical review of alpha radionuclide therapy—how to deal with recoiling
daughters? Pharmaceuticals 8 321–36
Ferrier M G et al 2020 Thorium chelators for targeted alpha therapy: rapid chelation of thorium-226 J. Label. Compd. Radiopharm. 63
502–16
Fiaccabrino D E, Kunz P and Radchenko V 2021 Potential for production of medical radionuclides with on-line isotope separation at the
ISAC facility at TRIUMF and particular discussion of the examples of 165Er and 155Tb Nucl. Med. Biol. 94–95 81–91
Friend M T, Mastren T, Parker T G, Vermeulen C E, Brugh M, Birnbaum E R, Nortier F M and Fassbender M E 2020 Production of 230Pa by
proton irradiation of 232Th at the LANL isotope production facility: precursor of 230U for targeted alpha therapy Appl. Radiat. Isot.
156 108973
Goorden M C, van der Have F, Kreuger R, Ramakers R M, Vastenhouw B, Burbach J P H, Booij J, Molthoff C F M and Beekman F J 2013
VECTor: a preclinical imaging system for simultaneous submillimeter SPECT and PET J. Nucl. Med. 54 306
Graf F et al 2014 DNA double strand breaks as predictor of efficacy of the alpha-particle emitter Ac-225 and the electron emitter Lu-177 for
somatostatin receptor targeted radiotherapy PloS One 9 e88239–88239
Harteveld A A, Meeuwis A P W, Disselhorst J A, Slump C H, Oyen W J G, Boerman O C and Visser E P 2011 Using the NEMA NU 4 PET
image quality phantom in multipinhole small-animal SPECT J. Nucl. Med. 52 1646
Ivashchenko O, van der Have F, Goorden M C, Ramakers R M and Beekman F J 2015a Ultra-high-sensitivity submillimeter mouse SPECT
J. Nucl. Med. 56 470–5
Ivashchenko O, van der Have F, Villena J L, Groen H C, Ramakers R M, Weinans H H and Beekman F J 2015b Quarter-millimeter-
resolution molecular mouse imaging with U-SPECT Mol. Imaging 14 7290.2014.00053
Jaggi J S, Seshan S V, McDevitt M R, LaPerle K, Sgouros G and Scheinberg D A 2005 Renal tubulointerstitial changes after internal irradiation
with α-particle–emitting actinium daughters J. Am. Soc. Nephrol. 16 2677–89
Kelly V J, Wu S-T, Gottumukkala V, Coelho R, Palmer K, Nair S, Erick T, Puri R, Ilovich O and Mukherjee P 2020 Preclinical evaluation of
an (111)In/(225)Ac theranostic targeting transformed MUC1 for triple negative breast cancer Theranostics 10 6946–58
Khreish F, Ebert N, Ries M, Maus S, Rosar F, Bohnenberger H, Stemler T, Saar M, Bartholomä M and Ezziddin S 2020 225 Ac-PSMA-617/
177Lu-PSMA-617 tandem therapy of metastatic castration-resistant prostate cancer: pilot experience Eur. J. Nucl. Med. Mol. Imaging
47 721–8
Kim Y-S and Brechbiel M W 2012 An overview of targeted alpha therapy Tumor Biol. 33 573–90
Kratochwil C, Bruchertseifer F, Giesel F L, Weis M, Verburg F A, Mottaghy F, Kopka K, Apostolidis C, Haberkorn U and Morgenstern A
2016 225 Ac-PSMA-617 for PSMA-targeted α-radiation therapy of metastatic castration-resistant prostate cancer J. Nucl. Med. 57
1941–4
Kunz P et al 2020 Medical isotope collection from ISAC targets Eur. Phys. J. Web Conf. 229 06003
Kunz P 2022 ISAC Yield Database Onlinehttps://yield.targets.triumf.ca/search/yield/data
Mastren T et al 2017 Simultaneous separation of actinium and radium isotopes from a proton irradiated thorium matrix Sci. Rep. 7 8216
McDevitt M R, Sgouros G, Finn R D, Humm J L, Jurcic J G, Larson S M and Scheinberg D A 1998 Radioimmunotherapy with alpha-emitting
nuclides Eur. J. Nucl. Med. 25 1341–51
McNeil B L, Robertson A K H, Fu W, Yang H, Hoehr C, Ramogida C F and Schaffer P 2021 Production, purification, and radiolabeling of the
203Pb/212Pb theranostic pair EJNMMI Radiopharm. Chem. 6 6
Miederer M, Scheinberg D A and McDevitt M R 2008 Realizing the potential of the actinium-225 radionuclide generator in targeted alpha
particle therapy applications Adv. Drug Deliv. Rev. 60 1371–82
Miller C, Rousseau J, Ramogida C F, Celler A, Rahmim A and Uribe C F 2022 Implications of physics, chemistry and biology for dosimetry
calculations using theranostic pairs Theranostics 12 232–59
Miwa K, Inubushi M, Takeuchi Y, Katafuchi T, Koizumi M, Saga T and Sasaki M 2015 Performance characteristics of a novel clustered
multi-pinhole technology for simultaneous high-resolution SPECT/PET Ann. Nucl. Med. 29 460–6
Morgenstern A, Apostolidis C, Bruchertseifer F, Capote R, Gouder T, Simonelli F, Sin M and Abbas K 2008 Cross-sections of the reaction
232Th(p,3n)230Pa for production of 230U for targeted alpha therapy Appl. Radiat. Isot. 66 1275–80
Muehllehner G 1985 Effect of resolution improvement on required count density in ECT imaging: a computer simulation Phys. Med. Biol.
30 163–73
National Electrical Manufacturers Association 2007 NEMA Standards Publication NU 2-2007: Performance Measurements of Positron
Emission Tomographs (Rosslyn, VA: National Electrical Manufacturers Association)
Radchenko V et al 2017 Radiometric evaluation of diglycolamide resins for the chromatographic separation of actinium from fission
product lanthanides Talanta 175 318–24
Robertson A K H, Ramogida C F, Rodríguez-Rodríguez C, Blinder S, Kunz P, Sossi V and Schaffer P 2017 Multi-isotope SPECT imaging of
the 225Ac decay chain: feasibility studies Phys. Med. Biol. 62 4406–20
Rosar F, Hau F, Bartholomä M, Maus S, Stemler T, Linxweiler J, Ezziddin S and Khreish F 2021 Molecular imaging and biochemical response
assessment after a single cycle of [(225)Ac]Ac-PSMA-617/[(177)Lu]Lu-PSMA-617 tandem therapy in mCRPC patients who have
progressed on [(177)Lu]Lu-PSMA-617 monotherapy Theranostics 11 4050–60
Rösch F, Herzog H and Qaim S M 2017 The beginning and development of the theranostic approach in nuclear medicine, as exemplified by
the radionuclide pair (86)Y and (90)Y Pharmaceuticals 10 56
Schwartz J, Jaggi J S, O’Donoghue J A, Ruan S, McDevitt M, Larson S M, Scheinberg D A and Humm J L 2011 Renal uptake of bismuth-213
and its contribution to kidney radiation dose following administration of actinium-225-labeled antibody Phys. Med. Biol. 56 721–33
Sgouros G 2020 Dosimetry, radiobiology and synthetic lethality: radiopharmaceutical therapy (RPT) with alpha-particle-emitters Target.
Alpha Ther. 50 124–32
Sgouros G, Frey E, Du Y, Hobbs R and Bolch W 2021 Imaging and dosimetry for alpha-particle emitter radiopharmaceutical therapy:
improving radiopharmaceutical therapy by looking into the black box Eur. J. Nucl. Med. Mol. Imaging 49 18–29
Solomon V R et al 2020 Nimotuzumab site-specifically labeled with (89)Zr and (225)Ac using spytag/spycatcher for PET imaging and alpha
particle radioimmunotherapy of epidermal growth factor receptor positive cancers Cancers 12 3449
Tong S, Alessio A M and Kinahan P E 2010 Noise and signal properties in PSF-based fully 3D PET image reconstruction: an experimental
evaluation Phys. Med. Biol. 55 1453–73
Tronchin S, Forster J C, Hickson K and Bezak E 2022 Dosimetry in targeted alpha therapy. a systematic review: current findings and what is
needed Phys. Med. Biol. 67 09TR01
Vastenhouw B and Beekman F 2007 Submillimeter total-body murine imaging with U-SPECT-I J. Nucl. Med. 48 487

13
Phys. Med. Biol. 67 (2022) 185009 H Koniar et al

Visser E P, Harteveld A A, Meeuwis A P W, Disselhorst J A, Beekman F J, Oyen W J G and Boerman O C 2011 Image quality phantom and
parameters for high spatial resolution small-animal SPECT Nucl. Instrum. Methods Phys. Res. Sect. Accel. Spectrometers Detect. Assoc.
Equip. 654 539–45
Walker M D, Goorden M C, Dinelle K, Ramakers R M, Blinder S, Shirmohammad M, van der Have F, Beekman F J and Sossi V 2014
Performance assessment of a preclinical PET scanner with pinhole collimation by comparison to a coincidence-based small-animal
PET scanner J. Nucl. Med. 55 1368
Yang H, Gao F, Yuan Z, Rodriguez-Rodriguez C, Merkens H, Robertson A, Radchenko V, Causey P, Benard F and Schaffer P 2020a A novel
actinium bifunctional chelator crown and biodistribution of Ac-225-Crown-TATE J. Nucl. Med. 61 1235
Yang H et al 2020b Synthesis and evaluation of a macrocyclic actinium-225 chelator, quality control and in vivo evaluation of 225Ac-crown-
αMSH peptide Chem.—Eur. J. 26 11435–40
Zielinska B, Apostolidis C, Bruchertseifer F and Morgenstern A 2007 An improved method for the production of Ac‐225/Bi‐213 from Th‐
229 for targeted alpha therapy Solvent Extr. Ion Exch. 25 339–49

14