Quantification of [¹¹C]-R-PK11195 PET

Brain

Analysis methods used in literature

Schuitemaker et al (2007a) published an extensive comparison of methods for producing parametric images of [¹¹C]-R-PK11195 binding. They suggest that when plasma input is available the Logan graphical analysis should be used (30-60 min linear fit), and with reference region input RPM1 (original version by Gunn et al., 1997) should be used, provided that the range of basis functions is carefully optimized.

Image-derived input function can be obtained from carotid arteries in sufficient quality to detect glial activation in Parkinson's disease (Kang et al., 2018).

In a study of MS patients and healthy controls, the fractions of parent radioligand in plasma did not differ between groups, sexes, or treatment categories, suggesting that plasma metabolite correction could be based on population average (de Souza et al., 2021).

2-tissue compartment model with plasma input

Kropholler et al (2005) and Jučaite et al (2012) have validated the use of 2-tissue compartment model in estimating the total distribution volume (V_T) and binding potential (k₃/k₄). V_B was fitted, but K₁/k₂ was fixed to whole cortex value (Kropholler et al., 2005). With another tracer for peripheral benzodiazepine receptor ([¹¹C]DAA1106) K₁/k₂ was found to differ among individuals (Ikoma et al., 2007), suggesting that k₃/k₄ should be preferred over V_T.

Reference tissue input

Because glial cells are located everywhere in the brain, there is no true reference region for [¹¹C]-R-PK11195 binding. Therefore, cluster analysis has been applied in extracting a reference tissue curve from the dynamic image, and it is used as input function for the simplified reference tissue model (Banati et al., 2000; Kropholler et al., 2006 and 2007). [¹¹C]-R-PK11195 PET with cluster analysis derived reference input curve can be used to assess longitudinal changes and treatment effects in microglial activation in MS (Sucksdorff et al., 2019).

However, the unsupervised tissue classification does not succeed in finding a reference tissue curve in all cases, and therefore a supervised clustering algorithm has been developed and validated (Turkheimer et al., 2007), and Matlab software Super-PK is available for this purpose. The resulting pseudo-reference region curve was then used to estimate binding potential (BP_ND) with simplified reference tissue model (SRTM) and rank-shaping regularized exponential spectral analysis (RS-ESA). Wavelet-based Logan plot, basis pursuit and SRTM give better ICC than ratio or traditional Logan method (Anderson et al., 2007).

For certain diseases it has been shown that cerebellum or certain cortical regions do not have increased microglial burden, and then these regions can be used as reference region for reference tissue model (Gerhard et al., 2002 and 2005; Kropholler et al., 2007c; Holmes et al., 2016) to assess regional changes. Cerebellar grey matter can be used as reference tissue with SRTM for quantifying TSPO expression in human glioma (Su et al., 2015). DVR maps were also calculated with Logan plot using the cerebellar cortex as input function in patients with chronic fatigue syndrome (Nakatomi et al., 2014). Jeon et al (2017) and Seo et al (2021) have used cerebellum as reference region to calculate DVR maps of the brains of patients with chronic pain, and used an alternative form of reference input Logan plot (Zhou et al., 2009), which is less sensitive to noise.

TSPO level in the normal brain is very low, and thus the blood vessel uptake of TSPO ligands becomes predominant. Vascular component can be included in SRTM as a linear term for both the target and reference region, increasing the [¹¹C]-R-PK11195 BP_ND difference between AD patients and control subjects (Tomasi et al., 2008).

It should be noted that the test-retest reliability and convergent validity of measures obtained without arterial input function may be poor (Plavén-Sigray et al., 2018).

Ratio method with white matter as reference tissue

Hammoud et al (2005) validated by simulations the calculation of tissue-to-white matter ratio as a parameter related to binding potential. They calculated the ratio from 10 to 60 min p.i.. Unfortunately this method was not included in the comparison by Schuitemaker et al (2007b).

Suggested analysis method for Turku

[¹¹C]-R-PK11195 has radioactive metabolites in the plasma and at least [¹¹C]CH₂O (formaldehyde) easily penetrates the blood-brain barrier (De Vos et al., 1999). The uptake of labelled metabolites in the brain precludes perfect quantification of peripheral benzodiazepine receptors, but an index related to the receptor concentration can still be achieved.

When plasma curves corrected for radioactive metabolites (Roivainen et al., 2009) are available, the method of Kropholler et al (2005) is preferable choice for analysis method for regional data. To calculate parametric V_T images the Logan graphical analysis is recommended (Schuitemaker et al., 2007a), although a strictly linear phase can not be achieved.

When plasma curves are not available, the very simple method by Hammoud et al (2005) seems like worth testing. Ratio image can be calculated e.g. using program imgratio. However, if precise quantitation is needed and extraction of valid reference tissue curves is possible, then RPM1 method is recommended (Schuitemaker et al., 2013), using PMOD or imgbfbp. The supervised cluster analysis can be applied to extract the reference tissue curve (Rissanen et al., 2014), and it can be used as input function in Logan plot for calculating regional DVR and in SRTM analysis for calculating BP_ND images for SPM (Schuitemaker et al., 2013; Sucksdorff et al., 2017 and 2019; Rissanen et al., 2018). Set the range of basis functions to the values determined for use with [¹¹C]-R-PK11195 in TPC by Jouni Tuisku (Sucksdorff et al., 2019).

Rheumatoid arthritis

¹¹C-R-PK11195 PET imaging allows noninvasive in vivo imaging and quantification of macrophages in rheumatoid synovitis, and possibly even in subclinical synovitis (van der Laken et al., 2008; Kropholler et al., 2009). Noninvasive visualization of macrophages may be useful both for detecting early synovitis and for monitoring synovitis activity during treatment (van der Laken et al., 2008; Gent et al., 2012; Roivainen et al., 2013).

Analysis methods used in literature

van der Laken et al (2008) reported that 1-tissue compartment model with arterial plasma input can be used to estimate regional volume of distribution (V_T) of [¹¹C]-R-PK11195 in synovial tissue, and that [¹¹C]-R-PK11195 uptake in synovial tissue was due to binding to TSPO on macrophages.

van der Laken et al (2008) also found a good correlation between V_T and SUV_40-60, as well as good correlation between SUV_40-60 and macrophage infiltration in synovial tissue. Kropholler et al (2009) recommended calculating SUV_20-40 in clinical use. Therefore, the scanning procedure could be simplified and shortened to a 20-minute static scan of joints of interest. This would make it a method that could be applied in routine clinical practice (van der Laken et al., 2008; Kropholler et al., 2009), and in whole-body imaging.

A simple scoring based on visual analysis may be useful in following arthritis activity (Gent et al., 2012 and 2014).

Suggested analysis method for Turku PET Centre

When plasma curves corrected for radioactive metabolites are available the 1-tissue compartment model fitting of van der Laken et al (2008) is preferable choice for analysis method for regional data. To calculate parametric V_T images the Logan graphical analysis may be recommended (imgdv).

For clinical routine analysis, and when plasma curves are not available, the SUV images can be computed, preferably from 20 to 40 min after injection.

Lungs

[¹¹C]-R-PK11195 could be used to assess lung inflammation. 3TCM provided k₃/k₄ ratios (BP_ND) which associate with regional macrophage score (Bitker et al., 2022).

Processing of plasma data in TPC

A script called PK11195_input.bat is available for processing plasma input data. As input it needs four data files, and haematocrit for plasma-blood TAC conversion:

1. Blood data file from online blood sampler (*.bld, *.alg, *.lis, *.txt),
2. Plasma TAC file from manual sampling (*ap.kbq),
3. count-rate file (*.cr, *.r, *.hc, *.head or *.dft),
4. fraction file of parent tracer in plasma (*.rat),
5. Haematocrit (e.g. 0.40)

In addition, user has to give the names of output files:

1. Parent tracer TAC in plasma (e.g. *apc.kbq),
2. Metabolite TAC in plasma (e.g. *apm.kbq),
3. Blood TAC (*ab.kbq).

Output TACs are calibrated and corrected for physical decay, and delay-time). In addition, the script will create SVG and/or PNG images where the user can verify how fit of an exponential function into the fraction data succeeded, how delay correction succeeded, and how the resulting curves look like.

Dispersion correction is not applied in this script because the effect of dispersion is minimal in case of an ¹¹C labelled tracer with relatively slow kinetics.

Script can be used in Windows command-prompt window (CLI) directly, or a script can be written that processes all studies.

See also:

• PET imaging of the translocator protein (TSPO)
• White blood cells
• Processing input data

---

Literature

Anderson AN, Pavese N, Edison P, Tai YF, Hammers A, Gerhard A, Brooks DJ, Turkheimer FE. A systematic comparison of kinetic modelling methods generating parametric maps for [¹¹C]-(R)-PK11195. Neuroimage 2007; 36: 28-37. doi: 10.1016/j.neuroimage.2007.02.017.

Banati RB, Newcombe J, Gunn RN, Cagnin A, Turkheimer F, Heppner F, Price G, Wegner F, Giovannoni G, Miller DH, Perkin GD, Smith T, Hewson AK, Bydder G, Kreutzberg GW, Jones T, Cuzner ML, Myers R. The peripheral benzodiazepine binding site in the brain in multiple sclerosis. Quantitative in vivo imaging of microglia as a measure of disease activity. Brain 2000; 123:2321-2337. doi: 10.1093/brain/123.11.2321.

Banati RB. Visualising microglial activation in vivo. Glia 2002; 40(2): 206-217. doi: 10.1002/glia.10144.

Jučaite A, Cselenyi Z, Arvidsson A, Åhlberg G, Julin P, Varnäs K, Stenkrona P, Andersson J, Halldin C, Farde L. Kinetic analysis and test-retest variability of the radioligand [¹¹C](R)-PK11195 binding to TSPO in the human brain - a PET study in control subjects. EJNMMI Res. 2012; 2: 15. doi: 10.1186/2191-219X-2-15.

Kropholler MA, Boellaard R, Schuitemaker A, van Berckel BNM, Luurtsema G, Windhorst AD, Lammertsma AA. Development of a tracer kinetic plasma input model for (R)-[¹¹C]PK11195 brain studies. J Cereb Blood Flow Metab. 2005; 25(7): 842-851. doi: 10.1038/sj.jcbfm.9600092.

Kropholler MA, Boellaard R, Schuitemaker A, Folkersma H, van Berckel BNM, Lammertsma AA. Evaluation of reference tissue models for the analysis of [¹¹C](R)-PK11195 studies. J Cereb Blood Flow Metab. 2006; 26(11): 1431-1441. doi: 10.1038/sj.jcbfm.9600289.

Kropholler MA, Boellaard R, van Berckel BN, Schuitemaker A, Kloet RW, Lubberink MJ, Jonker C, Scheltens P, Lammertsma AA. Evaluation of reference regions for (R)-[¹¹C]PK11195 studies in Alzheimer's disease and mild cognitive impairment. J Cereb Blood Flow Metab. 2007; 27(12): 1965-1974. doi: 10.1038/sj.jcbfm.9600488.

Kropholler MA, Boellaard R, Elzinga EH, van der Laken CJ, Maruyama K, Kloet RW, Voskuyl AE, Dijkmans BA, Lammertsma AA. Quantification of (R)-[¹¹C]PK11195 binding in rheumatoid arthritis. Eur J Nucl Med Mol Imaging. 2009; 36(4): 624-631. doi: 10.1007/s00259-008-0987-7.

van der Laken CJ, Elzinga EH, Kropholler MA, Molthoff CF, van der Heijden JW, Maruyama K, Boellaard R, Dijkmans BA, Lammertsma AA, Voskuyl AE. Noninvasive imaging of macrophages in rheumatoid synovitis using ¹¹C-(R)-PK11195 and positron emission tomography. Arthritis Rheum. 2008; 58(11): 3350-3355. doi: 10.1002/art.23955.

Rissanen E, Tuisku J, Rokka J, Paavilainen T, Parkkola R, Rinne JO, Airas L. In vivo detection of diffuse inflammation in secondary progressive multiple sclerosis using PET imaging and the radioligand ¹¹C-PK11195. J Nucl Med. 2014; 55(6): 939-944. 10.2967/jnumed.113.131698.

Roivainen A, Någren K, Hirvonen J, Oikonen V, Virsu P, Tolvanen T, Rinne J. Whole-body distribution and metabolism of [N-methyl-¹¹C](R)-1-(2-chlorophenyl)-N-(1-methylpropyl)-3-isoquinoline carboxamide in man; an imaging agent for in vivo assessment of peripheral benzodiazepine receptor activity with positron emission tomography. Eur J Nucl Med Mol Imaging 2009; 36(4): 671-682. doi: 10.1007/s00259-008-1000-1.

Schuitemaker A, van Berckel BNM, Kropholler MA, Kloet RW, Jonker C, Scheltens P, Lammertsma AA, Boellaard R. Evaluation of methods for generating parametric (R)-[¹¹C]PK11195 binding images. J Cereb Blood Flow Metab. 2007a; 1603-1615. doi: 10.1038/sj.jcbfm.9600459.

Schuitemaker A, van Berckel BN, Kropholler MA, Veltman DJ, Scheltens P, Jonker C, Lammertsma AA, Boellaard R. SPM analysis of parametric (R)-[¹¹C]PK11195 binding images: plasma input versus reference tissue parametric methods. Neuroimage 2007b; 35(4): 1473-1479. doi: 10.1016/j.neuroimage.2007.02.013.

Tomasi G, Edison P, Bertoldo A, Roncaroli F, Singh P, Gerhard A, Cobelli C, Brooks DJ, Turkheimer FE. Novel reference region model reveals increased microglial and reduced vascular binding of ¹¹C-(R)-PK11195 in patients with Alzheimer's Disease. J Nucl Med. 2008; 49: 1249-1256. doi: 10.2967/jnumed.108.050583.

Turkheimer FE, Edison P, Pavese N, Roncaroli F, Anderson AN, Hammers A, Gerhard A, Hinz R, Tai YF, Brooks DJ. Reference and target region modeling of [¹¹C](R)-PK11195 brain studies. J Nucl Med. 2007; 48(1): 158-167. PMID: 17204713.

Tags: Inflammation, TSPO, Brain, Rheumatoid arthritis

---

Updated at: 2022-12-13
Created at: 2007-01-29
Written by: Vesa Oikonen, Jouni Tuisku