Gamma camera calibration and validation for quantitative SPECT imaging with 177Lu

Author’s Accepted Manuscript GAMMA CAMERA CALIBRATION AND VALIDATION FOR QUANTITATIVE SPECT IMAGING WITH 177Lu M. D’Arienzo, M. Cazzato, S. Ungania, G. Iaccarino, M. D’Andrea, M.L. Cozzella, A. Fazio, A. Fenwick, L. Johansson, M. Cox, L. Strigari, P. De Felice

PII: DOI: Reference:

www.elsevier.com/locate/apradiso

S0969-8043(16)30090-2 http://dx.doi.org/10.1016/j.apradiso.2016.03.007 ARI7419

To appear in: Applied Radiation and Isotopes Received date: 21 September 2015 Revised date: 9 February 2016 Accepted date: 7 March 2016 Cite this article as: M. D’Arienzo, M. Cazzato, S. Ungania, G. Iaccarino, M. D’Andrea, M.L. Cozzella, A. Fazio, A. Fenwick, L. Johansson, M. Cox, L. Strigari and P. De Felice, GAMMA CAMERA CALIBRATION AND VALIDATION FOR QUANTITATIVE SPECT IMAGING WITH 177Lu, Applied Radiation and Isotopes, http://dx.doi.org/10.1016/j.apradiso.2016.03.007 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

GAMMA CAMERA CALIBRATION AND VALIDATION FOR QUANTITATIVE SPECT IMAGING WITH 177Lu M. D’Arienzo1, M. Cazzato2, S. Ungania2, G. Iaccarino2, M. D’Andrea2, M. L. Cozzella1, A. Fazio1, A. Fenwick3, L. Johansson3, M. Cox3, L. Strigari2, P. De Felice1 1

National Institute of Ionizing Radiation Metrology, ENEA CR Casaccia, Via Anguillarese 301, I-00123 Rome, Italy 2 Istituto Regina Elena, Rome, Italy, Via Elio Chianesi 53, 00144 Rome, Italy 3 National Physical Laboratory, Teddington TW11 0LW, UK

Abstract Over the last years 177Lu has received considerable attention from the clinical nuclear medicine community thanks to its wide range of applications in molecular radiotherapy, especially in peptide-receptor radionuclide therapy (PRRT). In addition to short-range beta particles, 177Lu emits low energy gamma radiation of 113 keV and 208 keV that allows gamma camera quantitative imaging. Despite quantitative cancer imaging in molecular radiotherapy having been proven to be a key instrument for the assessment of therapeutic response, at present no general clinically accepted quantitative imaging protocol exists and absolute quantification studies are usually based on individual initiatives. The aim of this work was to develop and evaluate an approach to gamma camera calibration for absolute quantification in tomographic imaging with 177Lu. We assessed the gamma camera calibration factors for a Philips IRIX and Philips AXIS gamma camera system using different reference geometries, both in air and in water. Images were corrected for the major effects that contribute to image degradation, i.e. attenuation, scatter and dead-time. We validated our method in non-reference geometry using an anthropomorphic torso phantom provided with liver cavity uniformly filled with 177 LuCl3. Our results showed that calibration factors depend on the particular reference condition. In general, acquisitions performed with the IRIX gamma camera provided good results at 208 keV, with agreement within 5 % for all geometries. The use of a Jaszczak 16 mL hollow sphere in water provided calibration factors capable of recovering the activity in anthropomorphic geometry within 1 % for the 208 keV peak, for both gamma cameras. The point source provided the poorest results, most likely because scatter and attenuation correction are not incorporated in the calibration factor. However, for both gamma cameras all geometries provided calibration factors capable of recovering the activity in anthropomorphic geometry within about 10 % (range −11.6 % to +7.3 %) for acquisitions at the 208 keV photopeak. As a general rule, scatter and attenuation play a much larger role at 113 keV compared to 208 keV and are likely to hinder an accurate absolute quantification. Acquisitions of only the 177Lu main photopeak (208 keV) are therefore recommended in clinical practice. Preliminary results suggest that the gamma camera calibration factor can be assessed with a standard uncertainty below (or on the order of) 3 % if activity is determined with equipment traceable to primary standards, accurate volume measurements are made, and an appropriate chemical carrier is used to allow a homogeneous and stable solution to be used during the measurements.

Keywords: Quantitative Imaging, Molecular Radiotherapy, Gamma Camera, Calibration, Radionuclide Therapy

1. INTRODUCTION Owing to its favorable physical properties and decay characteristics, lutetium-177 (177Lu) is an attractive radionuclide used for a number of therapeutic applications. Over the last 15 years this radionuclide has been successfully applied to peptide receptor radionuclide therapy (PRRT) in the form of 177Lu-[DOTA0, Tyr3]octerotate (177Lu-DOTATATE) or 177 Lu-[DOTA0, Tyr3]octerotite (177Lu-DOTATOC) to treat neuroendocrine tumors and other somatostatin-receptor expressing neoplasms (Kwekkeboom, 2004; Kwekkeboom et al., 2008; Bodei et al., 2011; Sandström et al., 2012; Sandström et al., 2009; Larsson et al., 2012). Other examples of therapeutic strategies with 177Lu include bone metastatic pain palliation in the form of 177Lu-EDTMP (Agarwal et al., 2014; Yuan et al., 2013; Shinto et al., 2014) and radiosynovectomy using 177Lu- phytate complex (Jalilian et al., 2014). In addition to eradicating tumors, peptides labeled with 177Lu allow the direct visualization of the radiopharmaceutical biodistribution thanks to the emission of two imageable gamma emissions. Because of the rather large cross section 177Lu can be produced directly by thermal neutron bombardment of natural 176Lu2O3 target (natural abundance of 176Lu is 2.6%) or enriched 176Lu2O3 (enriched in 176Lu up to 60–80%, Pillai et al., 2003; Dash et al., 2015). In addition, carrier-free 177Lu can be obtained indirectly by neutron capture of 176Yb target (Yb2O3) following radiochemical separation of 177Lu from Yb isotope. It is a low-energy beta-emitting radionuclide (maximum and average beta energies 498.3 keV and 149.4 keV, respectively) with two low-abundance gamma photons of 112.95 keV (6.20 %) and 208.36 keV (10.38 %) (BIPM, 2004), both suitable for imaging with commercial gamma camera systems. In order to optimize both the effectiveness of PRRT therapy and the absorbed dose to organs at risk, accurate determination of the activity of 177Lu within a defined anatomyindexed volume in a patient is mandatory. The general consensus is that the collection of accurate quantitative data is a critical step in dosimetry and treatment planning in molecular radiotherapy. In fact, inaccuracies in absolute quantification are likely to result in reduced efficacy or increased incidence of adverse side effects. Such an outcome especially applies to PRRT where a number of studies on patients treated with 177Lulabelled peptides have shown possible side effects, typically observed in kidneys and bone marrow (Sandström et al., 2012; Paganelli et al., 2014), albeit with an improved quality of life. The current state of the art for quantitative activity assessment is the use of single photon emission computed tomography (SPECT). It is generally believed that activity quantification using tomographic imaging is theoretically superior to that obtained with planar imaging as problems of organ overlap may be overcome, contrast for small regions may be reduced, and more accurate information regarding heterogeneous activity uptake (and ultimately absorbed dose) may be obtained (Siegel et al., 1999). However, even absolute activity quantification via tomographic imaging is affected by inherent limitations. Attenuation and scatter of photons degrade image quality, and the accuracy of

activity estimates varies with the object size due to the limited spatial resolution, dead time and partial volume effects. Proper compensation techniques are thus required. Gamma camera calibration is a prerequisite for absolute quantification studies via tomographic imaging, i.e. the system calibration factor must be preliminarily assessed to convert the reconstructed SPECT voxel values to absolute activity or activity concentration. Calibration procedures are currently performed either in-air or in-water using a radionuclide source with a known activity concentration (Dewaraja et al., 2012). Calibration in air consists of an acquisition of a small volume of activity (point-like source, Petri dish, line or spherical source) aimed at determining the in-air camera sensitivity (Willowson et al., 2008; Wit et al., 2006; Shcherbinin et al., 2008; He et al., 2005). Because repeated calibration experiments are generally needed to cope with changes in camera sensitivity with time, this calibration procedure might be practical to implement in a clinical environment. However, the main drawback of this approach is that it might not be sufficiently accurate if SPECT data are reconstructed with inadequate corrections for scatter, attenuation and other effects that contribute to image degradation. Alternatively, calibration can be performed in-water using an extended volume source (cylindrical phantoms filled with uniform activity or hot spheres in uniform background activity) in order to approximate the clinical conditions encountered in patient studies (Zeintl et al., 2010; Koral et al., 2005). Although more demanding, this procedure is likely to provide more reliable results as attenuating and scattering conditions are similar to those encountered in patient studies. No matter what approach is used, calibration measurements have to be performed using the same algorithms and gamma camera settings as in the patient study. At present there are no validated standard protocols, or any established methods for calibration and verification of system performance. Gamma camera calibrations are usually performed upon individual initiatives through the implementation of personalised approaches and system performances are therefore rarely comparable. As a consequence, it is generally accepted that quantitative imaging in molecular radiotherapy (MRT) suffers from a significant degree of inaccuracy. A recent study by Anizan et al. (Anizan et al., 2014) claimed the importance of accurate assessment of the gamma camera calibration factor and analysed the impact of variability factors on its repeatability and reproducibility. Furthermore, the International Atomic Energy Agency (IAEA) in 2009 initiated a Cooperative Research Project with the purpose of addressing the lack of harmonised protocols or guidelines in the clinical MRT practice (E21007 Development of Quantitative Nuclear Medicine Imaging for Patient Specific Dosimetry (http://www-naweb.iaea.org/nahu/DMRP/crp.htm). Along these lines, an EURAMET Joint Research Project was undertaken from 2012 to 2015 (MetroMRT, Metrology for Molecular Radiotherapy (D’Arienzo et al., 2014) with the purpose of providing a robust metrological basis for quantitative imaging and dosimetry in MRT, directed towards facilitating its universal adoption in clinical practice. A major part of this work was to develop and evaluate an approach to gamma camera calibration for absolute quantification in tomographic imaging with 177Lu. Calibration studies were performed on an IRIX and an AXIS gamma camera system (Philips Medical Systems, Cleveland, USA). We assessed the gamma camera calibration factors using four different reference geometries and correcting for the major effects that contribute to image degradation. Image reconstruction and processing were performed using commercially available software. For all reference geometries the impact of scatter and

attenuation on activity quantification was assessed for both photopeaks. We used available technology and correction algorithms that were not resource intensive to ensure clinical practicability of the calibration procedure. We validated our method in nonreference geometry using an anthropomorphic torso phantom provided with liver cavity uniformly filled with 177LuCl3.

MATERIALS AND METHODS I)

Reference and non-reference conditions: phantom preparation

Both for the IRIX and AXIS gamma camera the calibration coefficient was assessed using four different reference geometries: a point source in air, a 16 mL Jaszczak sphere surrounded by attenuating medium (non-radioactive water), a 16 mL Jaszczak sphere in air and a 20 cm diameter cylinder filled with water uniformly mixed with radioactive 177 LuCl3. For each phantom, a stock solution with a uniform activity concentration was prepared. Filling of the phantoms was realized by using a calibrated four decimal places analytic balance (Ohaus® Analytical Plus) and performing each time triple weighing of the sources. Table 1 shows the main information concerning the reference phantoms. Activity measurements given in the table were performed at the Italian National Institute of Ionizing Radiation Metrology. Adsorption of radionuclides on the inner walls of plastic phantoms may lead to inhomogeneous radionuclide distribution that can negatively affect quantitative imaging studies (Park et al., 2008). Furthermore the use of tap water should be avoided as minerals and other chemical impurities might stick to the phantom walls or combine with the radiopharmaceutical, changing the radionuclide distribution. Therefore a favourable chemistry must be used during the calibration procedure in order to have a uniform and stable solution. In the present study phantom preparation was performed using ultrapure water (18 MΩ). Furthermore, Diethylenetriaminepentaacetic Acid (DTPA) at a concentration of 25 µg g–1 was used as a carrier solution in order to prevent radioactive lutetium from sticking to the phantom walls and to guarantee a homogeneous radionuclide solution. Table 1. Reference and non-reference phantoms used for activity quantification measurements. The total activity used in the phantom is reported along with the nominal volumes of the sources Reference geometry Nominal volume Activity (MBq) (mL) IRIX AXIS 1) Point source 0.2 1 367.3 1 369.3 2) Jaszczack sphere in air 16 178.8 175.7 3) Jaszczack sphere in water 16 178.0 177.1 4) 20 cm cylindrical phantom 4 265 558.6 562.3 Non reference geometry 5) Anthropomorphic torso phantom – liver insert

Nominal volume (mL) 1200

Activity (MBq) IRIX AXIS 969.5 971.8

II)

Measurements of 177Lu sources and source preparation

In 2009 the Italian National Institute of Ionizing Radiation Metrology (ENEA-INMRI) took part in the CCRI(II)-K2.Lu-177 international key comparison of 177Lu (Zimmerman, 2013). Therefore 177Lu activity of the sources used in this work is traceable to the ENEA-INMRI 177Lu primary standard through the ENEA-INMRI activity secondary standard systems. In particular, all 177LuCl3 sources were measured through gamma-ray spectrometry using an HPGe-COAX detector to assess the possible presence of 177mLu. Measurements of sources 2 and 3 were also performed using two different ionization chambers (Centronic IG11-A20, NPL-CRC) that lead to a reduction of the stated uncertainty (Table 2). Hereinafter the activity concentration measured at ENEA-INMRI will be referred to as “true activity”.

III)

Acquisition parameters and quantitative reconstruction

Phantom acquisitions were performed using a three-head 3/8" (9.5 mm) NaI crystal IRIX and a double-head 3/4" (19 mm) NaI crystal AXIS gamma camera (both having 39.3 cm × 53.3 cm useful field of view. As a general rule, low-energy collimators are not suitable for energies above ~160 keV as septal penetration would generate star artefacts. Therefore medium energy high resolution (MEHR) parallel hole collimators were used for both systems. Acquisitions were performed using a 128 × 128 matrix (4.67 mm cubic voxel size), angular step of 4° over 360° (90 frames). Acquisitions were made with the phantoms in the centre of the field of view. For each phantom acquisitions of both photopeaks (113 keV, 208 keV) were performed. All data reconstruction was performed using the built-in software from the vendor (Odyssey LX 9.4C), while image analysis was performed with MIM workstation (MIM 6.1.7, MIM Software Inc., Cleveland, Ohio). Images were reconstructed using maximum likelihood expectation maximization (ML-EM) algorithm with 20 iterations. Since phantoms with simple geometries uniformly filled with radionuclide were used in this experiment, attenuation correction was performed using Chang’s algorithm (Chang, 1978) considering attenuation coefficients 0.165 cm–1 and –1 0.135 cm for the 113 keV and 208 keV photopeaks, respectively. Scatter correction on SPECT images was performed applying the triple energy window (TEW) technique on reconstructed images, based on the evaluation of the primary counts and the scattered component. The TEW window method for scatter correction is straightforward to implement and proved to give accurate scatter estimates in phantoms (Ogawa et al., 1991; Narita et al., 1996; Ichihara et al., 1993). For this purpose, two main windows were centered over the two photopeak energies, with 15 % energy window over the 113 keV photopeak (104.5 keV to 121.5 keV), and a 20 % window over the 208 keV photopeak (187.2 keV to 228.8 keV). In addition, two scatter windows on both sides of the main photopeaks were acquired. For the 113 keV photopeak, two windows of 10 % below (93.2 keV to 104.5 keV, centered at 98.8 keV) and above (121.5 keV to 132.8 keV, centered at 127.1 keV) were used, while for the 208 keV photopeak two windows of 10 % below (166.4 keV to 187.2 keV, centered at 176.8 keV) and above (228.8 keV to

249.6 keV, centered at 239.2 keV) were set up (figure 1). For each photopeak the scatter E / keV contribution at energy E (keV), C scatter , was assessed using the following equations:

 C 98.8 keV C 127.1keV keV C 113  scatter   W  98.8 keV W 127.1keV

 W113keV   2 

 C 176.8 keV C 239.2keV keV C 208  scatter   W  176.8 keV W 239.2keV

(1)

 W 208 keV   2 

where CE /keV are the counts in the windows centered at energy E (keV) and WE /keV are the corresponding window widths. To obtain the corrected image, Ccorr , the scattered E / keV photons C scatter are subtracted pixel-by-pixel from the count acquired with the main

window, Ctotal : E /keV Ccorr  Ctotal  C scatter

Figure 1. 177Lu spectrum as acquired from the AXIS gamma camera. Acquisition windows according to the TEW technique are reported in black (main photopeaks, sparse line pattern) and grey (scatter windows, dense line pattern). Both photopeaks show approximately the same relative intensity. The higher relative intensity of the lower photopeak is attributable to downscatter from the 208 keV,to bremsstrahlung photons generated from narrow-angle Compton scatter of 113 keV primary photons.

IV)

Dead Time

Dead-time count loss may result in a significant quantitation inaccuracy in SPECT imaging. In fact, at high count rates, the events detected by a gamma camera may occur in such rapid succession that the voltage pulses generated by the photomultiplier tubes overlap temporally. This means that the gamma camera calibration factor is a decreasing function of the count rate. It should be noted that all systems behave as a combination of paralyzable and non-paralyzable systems. However, as one component may prevail over the other, SPECT systems are often identified as paralyzable or non-paralyzable. In the present study both gamma cameras were considered to have a purely non-paralyzable deadtime, as suggested by the manufacturer. In the present study, dead-time corrections were assessed using the dual source method consisting of three count-rate measurements using two sources of similar activity, with the detector exposed to each source individually as well as to both concurrently. Measurements were performed without collimator and using two 99mTc syringe sources. Acquisition duration was 120 s. According to the dual source algorithm the gamma camera dead time, is given by the following formula (Cherry et al., 2012; Early and Sodee, 1995):



2(m12  mb ) m  m2  2mb ln( 1 ) 2 (m1  m2  2mb ) m12  mb

(2)

where m1 and m2 are the counting rates from sources 1 and 2, m12 is the combined counting rate and mb is the background counting rate. For non-paralyzable systems, the true counting rate Rt can be calculated from the observed counting rate R0 as follows:

Rt 

R0 (1   R0 )

(3)

Dead-time corrections were then applied on reconstructed images according to equation (3) V)

Determination of calibration factor

For each reference condition, the system calibration factor, Src, from decay-corrected counts per second to Becquerel was derived using the following equation (Ritt et al., 2011):

 T T  T    T  R Src   exp  0 cal  ln2    acq  ln2   1  exp   acq  ln2    T  T    T  V  ca  1/2   1/2    1/2 

1

(4)

where R is the counting rate measured in the VOI, V is the volume of the solution used for phantom preparation, ca is the activity concentration as determined at ENEA-INMRI, T0 is the acquisition start time, T1/2 is the radionuclide physical half-life, Tcal is the time of activity calibration, Tacq is the acquisition duration. The product V  ca represents the total

activity injected into the phantom. The first term in brackets corrects for the radioactive decay from the time of calibration until the start time of the acquisition. The second and third terms correct for the time duration of the acquisition and calculate the mean counts considering an exponential decay during acquisition, respectively. For reference geometries 1, 2 and 3 (Table 1) the system calibration factor was calculated in the reconstructed image by drawing a VOI as large as the physical dimension of the source plus a 3 cm margin in order to avoid partial volume effects. For reference geometry 4 (cylindrical phantom) counting rate was measured in a uniform area in the middle of the phantom to avoid border effects. VI)

Validation in anthropomorphic geometry

We validated the calibration method using an anthropomorphic torso phantom (Data Spectrum) provided with a liver insert of 1200 mL nominal volume. The liver cavity was filled with a known amount of 177LuCl3 (table 1). The SPECT data in anthropomorphic geometry were were acquired and reconstructed using the same settings as those used under reference conditions. The same corrections algorithms were used with the same energy windows. We manually drew 5 small spherical VOIs (16 mL each), evenly distributed in the liver volume, and calculated the average counting rate R A and associated standard deviation,  R . The calibration factors Src obtained in reference Anthro conditions were used for the absolute quantification of the activity concentration cVol within the liver insert. The ability of each calibration factor to determine the activity concentration in anthropomorphic geometry was assessed using the following formula:

Anthro cVol 

RA / VVOI S rc

(5)

where VVOI is the volume of the VOI.

VII)

Uncertainty Analysis

Uncertainty in the calibration factor Uncertainties in the parameters T0, T1/2, Tcal, Tacq in equation 4 were considered negligible. Therefore the relative combined standard uncertainty (%) of the system calibration factor is given by the sum in quadrature of the relative standard uncertainties in the activity concentration, uca , the phantom volume, uV , and the counting rate, uR :

uSrc 

u   u   u  2

ca

2

V

2

R

(6)

An estimation of the standard uncertainties associated with radionuclide activity concentration measurements, uca , is reported in table 2. The standard uncertainty in the volume of the point source, uV , was assumed to be negligible as accurate volume measurements were possible using a calibrated four decimal places balance. Relative standard uncertainties on volumes, uV , are reported in table 2. Finally, the standard deviation of the gamma camera long-term stability was assumed to represent the standard uncertainty in the counting rate, uR . Repeated measurements of the same source over a 6month period provided a 0.7 % standard deviation in the counting rate for the AXIS and 1.3 % for the IRIX gamma camera (table 2). Uncertainty in the absolute quantification Anthro Absolute quantification of the activity concentration cVol within the liver insert was calculated using equation 5. The uncertainty in the volume, uVVOI , was considered negligible as VVOI is the volume of the VOI digitally computed by the image analysis software. Under this assumption, the final relative combined standard uncertainty of the activity concentration measured in anthropomorphic geometry (equation 5) is:

Equation 1

uc Anthro  Vol

u   u  2

Src

2

RA

with uRA = 3 % for both gamma cameras, calculated as the standard deviation over the five spherical VOIs outlined on the anthropomorphic phantom. Table 2. Uncertainty table for the calibration factor (equation 6). uca (%) uV (%) u R (%) 1) Point source 2) Jaszczack sphere in air 3) Jaszczack sphere in water 4) 20 cm cylindrical phantom

IRIX/AXIS 3.0 – 1.1 0.1 1.1 0.1 3.0 1.0

IRIX 1.3 1.3 1.3 1.3

AXIS 0.7 0.7 0.7 0.7

RESULTS Quantitative results in reference geometry Measured dead times were 2.30 µs and 1.46 µs for the IRIX and AXIS gamma camera, respectively. Dead time correction varied between 0.1 % and 2.3 % (9.9 kcps) for IRIX and between 0.1 % and 2.1 % (14.1 kcps) for AXIS. In both cases the higher dead time correction was obtained for acquisition with the point source (1.36 GBq, table 1). Dead time correction in anthropomorphic geometry was 1.1 % for both imaging systems.

Figure 2. Axial slice of the reference geometries used for gamma camera calibration.

The calibration factors Src obtained under reference conditions are reported in tables 3 and 4 for both gamma cameras and for both photopeaks. A good consistency was found for the IRIX gamma camera for the 208 keV photopeak with a 3.0 % relative standard deviation in the calibration factors obtained for different reference geometries. Calibration factors calculated for the AXIS gamma camera using the 208 keV photopeak provided poorer results, with 9.0 % relative standard deviation. Calibration factors assessed using the lower photopeak returned higher dispersion, with 17.2 % standard deviation for the IRIX gamma camera and 15.0 % for the AXIS. Of note, in both cases the calibration factor obtained using the point-source as reference geometry is significantly higher than those obtained using other geometries. Figure 2 shows the reference sources as acquired with the AXIS gamma camera. Table 3. Calibration factors for the IRIX gamma camera, for both photopeaks, with relative standard uncertainties. IRIX (208 keV)

IRIX (113 keV)

Reference geometry

S rc (s–1 MBq–1)

uSrc (%)

S rc (s–1 MBq–1)

uSrc (%)

Point source Jaszczack sphere in air Jaszczack sphere in water 20 cm cylindrical phantom Relative standard deviation

7.56 7.36 7.09 7.13 3.0 %

3.3 1.7 1.7 3.4

5.81 4.36 4.51 3.97 17.2 %

3.3 1.7 1.7 3.4

Table 4. Calibration factors for the AXIS gamma camera, for both photopeaks, with relative standard uncertainties. AXIS (208 keV)

AXIS (113 keV)

Reference geometry

S rc (s–1 MBq–1)

uSrc (%)

S rc (s–1 MBq–1)

uSrc (%)

Point source Jaszczack sphere in air Jaszczack sphere in water 20 cm cylindrical phantom Relative standard deviation

8.67 7.82 7.28 8.84 9.0 %

3.1 1.3 1.3 3.2

4.91 3.91 3.57 3.73 15.0 %

3.1 1.3 1.3 3.2

Quantitative results in anthropomorphic geometry

Tables 5 and 6 show the activity concentrations determined within the anthropomorphic phantom using the sensitivity factors measured for the different reference geometries (equation 5). As a general rule, acquisitions performed using the higher photopeak provide better estimates of the recovered activity concentration in the anthropomorphic phantom. This is especially true for the IRIX gamma camera, where the agreement with the true value is approximately within ± 5 % for all reference geometries (Table 5). For both gamma cameras, the activity concentration measured using the point source sensitivity factor provided poorer results. A severe underestimation of the activity concentration in the anthropomorphic phantom was found in particular for the lower photopeak. Calibration factors assessed using the Jaszczak sphere in water provided the best estimates for the 208 keV photopeak for both systems, with activity concentration values within 1 % of the true value.

Table 5. Measured concentration in anthropomorphic geometry for the IRIX gamma camera, for both photopeaks for different reference geometries. Activity concentration in the liver cavity at the –1 –1 time of acquisition was 0.808 kBq mL ± 0.024 kBq mL . Reference geometry Measured activity concentration Deviation uc Anthro (%) Vol (%) IRIX (208 keV) (MBq mL–1) Point source 0.79 4.4 –2.2 Jaszczack sphere in air 0.84 3.5 4.2 Jaszczack sphere in water 0.81 3.5 0.5 20 cm cylindrical phantom 0.84 4.5 3.7

Reference geometry Point source Jaszczack sphere in air Jaszczack sphere in water 20 cm cylindrical phantom

Measured activity concentration IRIX (113 keV)(MBq mL–1) 0.68 0.88 0.91 1.00

uc Anthro (%) Vol

4.4 3.5 3.5 4.5

Deviation (%) –15.3  9.2  12.9  23.8 

Table 6. Measured concentration in anthropomorphic geometry for the AXIS gamma camera, for both photopeaks for different reference geometries. Activity concentration in the liver cavity at the time of acquisition was 0.810 ± 0.024 kBq/mL. Reference geometry Measured activity concentration Deviation uc Anthro (%) –1 Vol (%) AXIS (208 keV) (MBq mL ) Point source 0.73 4.3 –9.9  Jaszczack sphere in air 0.87 3.3 7.3  Jaszczack sphere in water 0.81 3.3 –0.1  20 cm cylindrical phantom 0.71 4.4 –11.6 

Reference geometry Point source Jaszczack sphere in air Jaszczack sphere in water 20 cm cylindrical phantom

Measured activity concentration AXIS (113 keV) (MBq mL–1) 0.58 0.79 0.72 0.76

uc Anthro (%) Vol

4.3 3.3 3.3 4.4

Deviation (%) –28.7  –1.7  –10.5  –5.9 

Impact of correction factors for different reference geometries Scatter and attenuation are highly dependent on the geometry and composition of the source and media. However, regardless of which reference geometry is used, scatter and attenuation play a major role for acquisitions at the lower photopeak. In particular, the higher scatter fraction at the lower photopeak is attributable to down-scatter the 208 keV , to bremsstrahlung photons generated from the beta decays of 177Lu and to narrow-angle Compton scatter of 113 keV primary photons. For acquisition at 113 keV, scatter radiation contributes to approximately 50 % of the total counts both for the Jaszczak sphere in water and the cylindrical phantom (figure 3b and 3d, round and square symbols). For the same phantoms, scatter contribution is in the range 10 % to 20 % for acquisitions at 208 keV (figure 3a and 3c, round and square symbols). These results hold true for both gamma cameras. In the absence of an attenuating medium, the correction factors when using hot Jaszczak sphere in air are less affected by scatter and attenuation effects (figure 3a-3d, triangle symbol). As expected, attenuation correction without scatter compensation leads to appreciable overestimations of the activity concentration in all reference geometries. This is particularly true for acquisitions at 113 keV (figure 3a-3d).

A

B

C

D

Figure 3. Impact of correction factors in the reference geometries for the IRIX (A and B) and AXIS (C and D) gamma cameras. For each reference geometry the fully corrected image (scatter + attenuation) was taken as the benchmark and the percentage deviation from the benchmark data is reported on the Y axis. Figures A and C show acquisitions at the main photopeak (208  keV) while acquisitions at 113 keV are reported in figures B and D.

DISCUSSIONS Despite the growing trend towards developing patient-specific dosimetry models based on quantitative tomographic information, progress in this area has been hindered by the lack of appropriate standards for calibrating the imaging systems and the lack of validated methodologies for absolute activity quantification in patients. So far, only a few literature studies have been dedicated to gamma camera calibration for quantitative imaging with 177Lu. Of note, each author proposed a personalized approach for the assessment of gamma camera calibration factor. Beauregard (Beauregard et al., 2011) performed sensitivity studies using both a point-like source (Eppendorf tube) in air and a point-like source surrounded by scattering/attenuating medium (saline bags). The authors validated this approach on a phantom using a large fillable cylinder with two hot cylindrical inserts. After corrections for scatter, attenuation and dead time, the authors obtained deviations from the calibrated activity in the range –14.9 % to +4.3 %. In another study, Shcherbinin (Shcherbinin et al., 2012) performed sensitivity measurements using a point source in air while validation was carried out using a cylindrical 70 mL container placed off-center in an empty Jaszczak phantom and in the same phantom filled with inactive water. After correcting for attenuation, scatter, resolution loss, contamination and partial volume effects, they obtained a maximum deviation of –1.6 % in the selected ROI and +17.7 % in the entire field of view. Recent work by Guerriero (Guerriero et al., 2013) described a calibration method based on a hollow anthropomorphic torso phantom with a set of hollow spheres and additional home-made inserts. In another study, Sandström (Sandström et al., 2009) performed calibration of the SPECT imaging system using a 100 mL sphere filled with a known amount of radioactivity placed inside a thorax phantom.

Fenwick et al. (Fenwick et al., 2014) have recently proposed a practical phantombased method to determine and validate a traceable calibration factor for a SPECT/CT system using 177Lu. It consists of three measurements using an elliptical Jaszczak phantom containing an active sphere placed in different locations inside the phantom. The calibration factor for the camera was determined using the arithmetic mean of the three measurements. The proposed calibration method demonstrated the ability to assess the activity concentration in an anthropomorphic geometry to within 5 % of the true activity. In the present work four different reference conditions were analyzed for gamma camera calibration and validated in anthropomorphic geometry. The use of a hot Jaszczak sphere in non-radioactive water provided calibration factors values capable of measuring the activity in anthropomorphic geometry within 1 % for the 208 keV peak, for both investigated gamma cameras. Our results are consistent with those reported by Fenwick et al. (Fenwick et al., 2014). The point source provided the poorest results, most likely because scatter and attenuation correction are not incorporated during image reconstruction. In fact it is generally believed (Dewaraja et al., 2012) that if gamma camera calibration is performed with a source geometry mimicking scatter and attenuation properties in patient imaging (such as a tank of uniform activity or hot spheres in uniform background activity) the effects of imperfect compensation will be partly reduced. As a general rule, acquisitions at the lower photopeak provided results with larger deviations from the true activity concentration. In fact, scatter and attenuation play a major role at 113 keV and are likely to hinder accurate absolute quantification. Our results corroborate previous investigations demonstrating inferior quantitative results when using the lower photopeak (de Nijs et al., 2014; Sanders et al., 2014). For both gamma cameras, if only the higher photopeak is acquired, all geometries provided accuracies ranging from –11.6 % to +7.3 %. Furthermore our findings are in keeping with those obtained by He and colleagues (He B., et al 2012) where inner compartments of a torso phantom were filled with known activities of 177Lu. In their study they achieved an agreement in activity estimates below 3.2 % for all organs if only the 208 keV photopeak was used, but were up to 40 % if only the 113 keV photopeak was used for quantitative SPECT. These results provide a strong argument for encouraging the clinical use of the higher 177Lu photopeak if acquisitions are performed with a medium energy collimator and if photon abundance is sufficiently large to provide good quality images. The same approach has been described by several authors (Garkavij et al., 2010; Beauregard et al., 2011; Shcherbinin et al., 2012) with varying reported success. Noteworthy, simultaneous acquisition of both photopeaks is also possible, with the aim to increase counting statistics and to improve contrast resolution (van Dalen et al., 2007, Liu et al., 2012). However the general consensus is that proper scatter correction techniques have to be applied to obtain accurate quantification when acquisitions are performed using both photopeaks (de Nijs et al., 2014). In the present study a maximum dead time correction of about 2 % was found for acquisition with the point source in air (1.36 GBq total activity). However dead time can potentially lead to significant absolute quantification inaccuracies when high activities are administered to patients, as in molecular radiotherapy with [177Lu]octreotate. Beauregard (Beauregard et al., 2011) reported dead-time corrections ranging between

6.6 % and 11.7 % in patients treated with [177Lu]octreotate and administered with activities between 8.3 GBq and 10.2 GBq. Therefore dead-time correction should be applied during the calibration procedure and is required in patient study to improve the accuracy of quantitative data. As a further point, activity measurements play a major role on the final quantification uncertainty and if activity is determined by a National Metrology Institute uncertainties can be reduced dramatically. Previous research aiming to assess factors affecting the stability and repeatability of gamma camera calibration has demonstrated that the largest source of variability is attributable to the measurement of the calibration source activity in the radionuclide activity meter (Anizan et al., 2014). In the present phantom study the gamma camera calibration factors were determined with uncertainties in the range 1.3 % to 3.4 %, while the uncertainties in the ctivity measured in complex geometry ranged from 3.3 % to 4.5 %. It is worth noting that the overall uncertainty is likely to be minimized if I) activity measurements are performed using equipment that has direct traceability to primary standards (through secondary standard systems) II) activity is dispensed using accurate volume measurements (triple weighing of the source performed with calibrated balances) and III) an appropriate chemical carrier is used throughout the calibration procedure. The current research acknowledges a few limitations that should be noted to aid interpretation of the results. Correction for attenuation was performed using the easily implementable Chang algorithm whereas modern SPECT scanners usually perform CTbased corrections. In our experience this choice had a negligible impact on the final results since both reference and non-reference phantoms had cylindrical geometry and were uniformly filled with water, thereby avoiding the presence of heterogeneous media or air gaps. Another possible limitation is that the calibration procedure and the uncertainty analysis described in the present study were performed only on 177Lu.. Furthermore, we deliberately focussed on absolute quantification in large volume objects containing a uniform activity distribution with the intent to avoid additional confounding factors, as partial volume effects. The extent to which both our methods and results can be generalized to other radionuclides and small objects certainly requires further investigation. In particular, absolute quantification in small structures close to a highuptake anatomical regions requires a detailed knowledge of spill-in and spill-out effects (Frey et al 2012).In this study we concerned ourselves with the impact that activity measurements and calibration procedures have on absolute quantification in anthropomorphic geometry. In clinical practice activity may typically be determined using a radionuclide calibrator traceable to a national standards laboratory (Zimmerman and Judge, 2007). The standard uncertainty for a typical well-calibrated field instrument can be expected to be in the region of 2 % for medium energy gamma emitters (100 keV to 2 MeV), or 5 % for low energy gamma emitters (< 100 keV) and pure beta emitters (AAPM, 2012). Therefore it is likely that at a clinical level the standard uncertainty for the gamma camera calibration factor be in the order of 5 % if well-calibrated field instrument are used for activity measurements. It is worth noting that the current general consensus is that dosimetry in radiation therapy should aim for an accuracy of ± 5 % as a change in the delivered dose to the patient by more than 5 % is likely to produce a clinically relevant change of the tumour control probability. Therefore it is desirable that in the near future the final standard uncertainty in absolute quantification in complex

phantoms be well below the expected 5 % to ensure the uncertainty in clinical absorbed dose estimations comply with the requirement of 5 % to a reference point.

CONCLUSIONS Quantitative SPECT with appropriate compensations is likely to provide accurate 177Lu activity estimates in anthropomorphic geometry using the 208 keV photopeak. Because of bremsstrahlung photons generated from the beta decays of 177Lu, downscatter from the 208 keV emission, and narrow angle Compton scatter of 113 keV primary photons, absolute quantification may be inaccurate when using the 113 keV photopeak. Acquisitions only on the upper energy window with medium-energy collimators are therefore recommended in clinical practice, provided that the administered activity is sufficiently high to produce good quality images. The use of a hot Jaszczak sphere in non-radioactive water phantom mimicking the patient geometry proved to be a suitable reference condition for calibrating a SPECT system, with the potential of measuring activity concentration in anthropomorphic geometry within a few percent. Preliminary results with 177Lu suggest that the gamma camera calibration factor can be determined with a standard uncertainty below (or of the order of) 3 % if activity measurements are performed with equipment traceable to primary standards, accurate volume measurements are made, and an appropriate chemical carrier is used during the experimental activity. Further research is needed to extend the proposed approach to other radionuclides. Absolute quantification in molecular radiotherapy is strongly in need of a robust metrological support. Significantly more research is needed, both to improve metrological traceability in MRT and to develop a standardized protocol for validation and calibration of quantitative imaging.

ACKNOWLEDGMENTS This work was supported by the EURAMET MetroMRT project (European Commission, Seventh Framework Programme, Grant Agreement N° 217257).

REFERENCES AAPM, 2012. The Selection, Use, Calibration, and Quality Assurance of Radionuclide Calibrators Used in Nuclear Medicine, Report of AAPM Task Group 181 Agarwal, K., Singla, S., Arora, G., Bal, C., 2014. 177Lu-EDTMP for palliation of pain from bone metastases in patients with prostate and breast cancer: a phase II study. European Journal of Nuclear Medicine and Molecular Imaging 42, 79-88.

Anizan, N., Wang, H., Zhou, X., Hobbs, R., Wahl, R., Frey, E., 2014. Factors affecting the stability and repeatability of gamma camera calibration for quantitative imaging applications based on a retrospective review of clinical data. EJNMMI Research 4, 67. Beauregard, J., Hofman, M., Pereira, J., Eu, P., Hicks, R., 2011. Quantitative 177 Lu SPECT (QSPECT) imaging using a commercially available SPECT/CT system. Cancer Imaging 11, 56-66. BIPM, 2004. Monographie BIPM-5-Table of Radionuclides, Table of Radionuclides, Vol. 2 - A = 151 to 242, CEA/ LNE-LNHB, 91191 Gif-sur-Yvette, France and BIPM, Pavillon de Breteuil, 92312 Sèvres, France. Bodei, L., Cremonesi, M., Grana, C., Fazio, N., Iodice, S., Baio, S., Bartolomei, M., Lombardo, D., Ferrari, M., Sansovini, M., Chinol, M., Paganelli, G., 2011. Peptide receptor radionuclide therapy with 177Lu-DOTATATE: the IEO phase I-II study. European Journal of Nuclear Medicine and Molecular Imaging 38, 2125-2135. Chang, L., 1978. A Method for Attenuation Correction in Radionuclide Computed Tomography. IEEE Trans. Nucl. Sci. 25, 638-643. Cherry, S., Sorenson, J., Phelps, M., 2012. Physics in nuclear medicine. Elsevier/Saunders, Philadelphia. D’Arienzo, M., Capogni, M., Smyth, V., Cox, M., Johansson, L., Solc, J., Bobin, C., Rabus, H., Joulaeizadeh, L., 2014. Metrological Issues in Molecular Radiotherapy. EPJ Web of Conferences 77, 00022. de Nijs, R., Lagerburg, V., Klausen, T., Holm, S., 2014 Improving quantitative dosimetry in 177Lu-DOTATATE SPECT by energy window-based scatter corrections Nuclear Medicine Communications 35 522-533 Dash, A., Pillai, M.R., Knapp, F.F. Jr, 2015 Production of 177Lu for Targeted Radionuclide Therapy: Available Options. Nucl Med Mol Imaging 49, 85-107. Dewaraja, Y., Frey, E., Sgouros, G., Brill, A., Roberson, P., Zanzonico, P., Ljungberg, M., 2012. MIRD Pamphlet No. 23: Quantitative SPECT for Patient-Specific 3Dimensional Dosimetry in Internal Radionuclide Therapy. Journal of Nuclear Medicine 53, 1310-1325. Early, P., Sodee, D., 1995. Principles and practice of nuclear medicine. Mosby, St. Louis. Fenwick, A.J., Merrett, J.L., Johansson, L.C., Green, A.J., D’Arienzo, M., 2014 A Proposed Method of Calibration for Quantitative 177Lu Measurements using SPECT/CT, European Journal of Nuclear Medicine and Molecular Imaging, 41, Vol. 2, pp 151-705 Frey E, Humm J and Ljungberg M 2012 Accuracy and Precision of Radioactivity Quantification in Nuclear Medicine Images Seminars in Nuclear Medicine 42 208-218

Garkavij, M., Nickel, M., Sjögreen-Gleisner, K., Ljungberg, M., Ohlsson, T., Wingårdh, K., Strand, S., Tennvall, J., 2010. 177Lu-[DOTA0,Tyr3] octreotate therapy in patients with disseminated neuroendocrine tumors: Analysis of dosimetry with impact on future therapeutic strategy. Cancer 116, 1084-1092. Guerriero, F., Ferrari, M., Botta, F., Fioroni, F., Grassi, E., Versari, A., Sarnelli, A., Pacilio, M., Amato, E., Strigari, L., Bodei, L., Paganelli, G., Iori, M., Pedroli, G., Cremonesi, M., 2013. Kidney Dosimetry in 177 Lu and 90 Y Peptide Receptor Radionuclide Therapy: Influence of Image Timing, Time-Activity Integration Method, and Risk Factors. BioMed Research International 2013, 1-12. He, B., Du, Y., Song, X., Segars, W., Frey, E., 2005. A Monte Carlo and physical phantom evaluation of quantitative In-111 SPECT. Physics in Medicine and Biology 50, 4169-4185. He, B., Nikolopoulou, A., Osborne, J., Vallabhajosula, S., Goldsmith, S., 2012. Quantitative SPECT imaging with Lu-177: A physical phantom evaluation, Journal of Nuclear Medicine 53 no. supplement 1 2407 http://projects.npl.co.uk/metromrt/ (accessed 9. 2. 15). http://www-naweb.iaea.org/nahu/DMRP/crp.htm (accessed 3. 3. 15). Ichihara, T., Ogawa, K., Motomura, N., Kubo, A., Hashimoto, S., 1993. Compton scatter compensation using the triple-energy window method for single-and dualisotope SPECT, Journal of Nuclear Medicine 34, 2216–2221 Jalilian, A., Yousefnia, H., Zolghadri, S., 2014. Preparation and evaluation of Lu- 177 phytate Complex for Radiosynovectomy. World J Nucl Med 13, 22. Koral, K., Yendiki, A., Qiang Lin, Dewaraja, Y., 2005. Comparison of 3-D OSEM versus 1-D SAGE for focal total-activity quantification in I-131 SPECT with HE collimation. IEEE Trans. Nucl. Sci. 52, 154-158. Kwekkeboom, D., 2004. Radiolabeled Somatostatin Analog [177LuDOTA0,Tyr3]Octreotate in Patients With Endocrine Gastroenteropancreatic Tumors. Journal of Clinical Oncology 23, 2754-2762. Kwekkeboom, D., de Herder, W., Kam, B., van Eijck, C., van Essen, M., Kooij, P., Feelders, R., van Aken, M., Krenning, E., 2008. Treatment With the Radiolabeled Somatostatin Analog [177Lu-DOTA0,Tyr3]Octreotate: Toxicity, Efficacy, and Survival. Journal of Clinical Oncology 26, 2124-2130. Larsson, M., Bernhardt, P., Svensson, J., Wängberg, B., Ahlman, H., Forssell-Aronsson, E., 2012. Estimation of absorbed dose to the kidneys in patients after treatment with 177Lu-octreotate: comparison between methods based on planar scintigraphy. EJNMMI Research 2, 49.

Liu, C., Brašić, J.R., Liu, X., Li, H., Xiang, X., Luo, Z., Wang, Y., Kuai, D., Zhang, G., Zaknun, .J.J. , 2012 Timing and optimized acquisition parameters for the whole-body imaging of ¹⁷⁷Lu-EDTMP toward performing bone pain palliation treatment. Nucl Med Commun. 33, 90-6. Narita, Y., Eberl, S., Iida, H., Hutton, B., Braun, M., Nakamura, T., Bautovich, G., 1996. Monte Carlo and experimental evaluation of accuracy and noise properties of two scatter correction methods for SPECT. Physics in Medicine and Biology 41, 2481-2496. Ogawa, K., Harata, Y., Ichihara, T., Kubo, A., Hashimoto, S., 1991. A practical method for position-dependent Compton-scatter correction in single photon emission CT. IEEE Transactions on Medical Imaging 10, 408-412. Paganelli, G., Sansovini, M., Ambrosetti, A., Severi, S., Monti, M., Scarpi, E., Donati, C., Ianniello, A., Matteucci, F., Amadori, D., 2014. 177 Lu-Dota-octreotate radionuclide therapy of advanced gastrointestinal neuroendocrine tumors: results from a phase II study. European Journal of Nuclear Medicine and Molecular Imaging 41, 1845-1851. Park, M., Mahmood, A., Zimmerman, R., Limpa-Amara, N., Makrigiorgos, G., Moore, S., 2008. Adsorption of metallic radionuclides on plastic phantom walls. Med. Phys. 35, 1606. Pillai, M., Chakraborty, S., Das, T., Venkatesh, M. and Ramamoorthy, N. 2003, Production logistics of 177Lu for radionuclide therapy Applied Radiation and Isotopes 59 109-118 Ritt, P., Vija, H., Hornegger, J., Kuwert, T., 2011. Absolute quantification in SPECT. European Journal of Nuclear Medicine and Molecular Imaging 38, 69-77. Sanders, J., Kuwert, T., Hornegger, J., Ritt, P. 2014. Quantitative SPECT/CT Imaging of 177Lu with In Vivo Validation in Patients Undergoing Peptide Receptor Radionuclide Therapy. Mol Imaging Biol, 17 585-593. Sandström, M., Garske, U., Granberg, D., Sundin, A., Lundqvist, H., 2009. Individualized dosimetry in patients undergoing therapy with 177Lu-DOTA-D-Phe1Tyr3-octreotate. European Journal of Nuclear Medicine and Molecular Imaging 37, 212225. Sandstrom, M., Garske-Roman, U., Granberg, D., Johansson, S., Widstrom, C., Eriksson, B., Sundin, A., Lundqvist, H., Lubberink, M., 2012. Individualized Dosimetry of Kidney and Bone Marrow in Patients Undergoing 177Lu-DOTA-Octreotate Treatment. Journal of Nuclear Medicine 54, 33-41. Shcherbinin, S., Celler, A., Belhocine, T., Vanderwerf, R., Driedger, A., 2008. Accuracy of quantitative reconstructions in SPECT/CT imaging. Physics in Medicine and Biology 53, 4595-4604.

Shcherbinin, S., Piwowarska-Bilska, H., Celler, A., Birkenfeld, B., 2012. Quantitative SPECT/CT reconstruction for 177 Lu and 177 Lu/ 90 Y targeted radionuclide therapies. Physics in Medicine and Biology 57, 5733-5747. Shinto, A., Shibu, D., Kamaleshwaran, K., Das, T., Chakraborty, S., Banerjee, S., Thirumalaisamy, P., Das, P., Veersekar, G., 2014. 177Lu-EDTMP for Treatment of Bone Pain in Patients with Disseminated Skeletal Metastases. Journal of Nuclear Medicine Technology 42, 55-61. Siegel, J.A., Thomas, S.R., Stubbs, J.B., Stabin, M.G., Hays, M.T., Koral, K.F., Robertson, J.S., Howell, R.W., Wessels, B.W., Fisher, D.R., Weber, D.A., Brill, A.B., 1999. MIRD Pamphlet No. 16: Techniques for quantitative radiopharmaceutical biodistribution data acquisition and analysis for use in human radiation dose estimates. Journal of Nuclear Medicine 40, 37S–61S. van Dalen, J., Hendrickx, B., Corstens, F., Oyen, W., 2007, Optimization scan protocol of Lu-177 in planar imaging using dual energy window scatter correction J Nucl Med. 2007; 48 (Supplement 2):425P Willowson, K., Bailey, D., Baldock, C., 2008. Quantitative SPECT reconstruction using CT-derived corrections. Physics in Medicine and Biology 53, 3099-3112. Wit, T., Xiao, J., Nijsen, J., Schip, F., Staelens, S., Rijk, P., Beekman, F., 2006. Hybrid scatter correction applied to quantitative holmium-166 SPECT. Physics in Medicine and Biology 51, 4773-4787. Yuan, J., Liu, C., Liu, X., Wang, Y., Kuai, D., Zhang, G., Zaknun, J., 2013. Efficacy and Safety of 177Lu-EDTMP in Bone Metastatic Pain Palliation in Breast Cancer and Hormone Refractory Prostate Cancer. Clinical Nuclear Medicine 38, 88-92. Zeintl, J., Vija, A., Yahil, A., Hornegger, J., Kuwert, T., 2010. Quantitative Accuracy of Clinical 99mTc SPECT/CT Using Ordered-Subset Expectation Maximization with 3Dimensional Resolution Recovery, Attenuation, and Scatter Correction. Journal of Nuclear Medicine 51, 921-928. Zimmerman, B., 2013. Report of the CCRI comparison CCRI(II)-K2.Lu-177. Metrologia 50, 06020-06020. Zimmerman, B., Judge, S., 2007. Traceability in nuclear medicine. Metrologia 44, S127S132.

Highlights      

We evaluated an approach to gamma camera calibration for SPECT imaging with 177Lu. Accurately calibrated reference activity was used to maintain a low uncertainty in the calibration of the gamma camera Calibration factors were assessed using different reference geometries, both in air and in water. We validated our method using an anthropomorphic phantom provided uniformly filled with 177LuCl3. The gamma camera calibration factor can be determined with a standard uncertainty below 3% We recommend acquisitions only on the 177Lu main photopeak (208 keV) in the clinical practice.