Exploring [11C]CPPC as a CSF1R-targeted PET imaging marker for early Parkinson's disease severity

Abstract

BACKGROUNDMicroglia-mediated brain immune changes play a role in the pathogenesis of Parkinson's disease (PD), but imaging microglia in living people with PD has relied on positron emission tomography (PET) ligands that lack specificity in labeling immune cells in the nervous system. We aimed to develop imaging of colony stimulating factor 1 receptor (CSF1R) as a microglial-sensitive marker of innate immunity.METHODSIHC using a CSF1R antibody evaluated colocalization with Iba-1 in PD (n = 4) and control (n = 4) human brain samples. Autoradiography using a CSF1R tritiated ligand in human brain samples from individuals with PD (n = 5) and in a control group (n = 4) was performed to obtain Bmax. PET imaging using a CSF1R radioligand was performed in 10 controls and 12 people with PD, and VT was compared between groups and correlated with disease severity.RESULTSIHC of CSF1R in human brain samples shows colocalization with Iba-1 and is significantly increased in brain samples from individuals with PD compared with individuals in a control group. Autoradiography revealed significantly increased CSF1R ligand binding in the inferior parietal cortex of patients with PD. [11C]CPPC PET showed higher binding in people with moderate PD compared with people in a control group and ligand binding correlated with more severe motor disability and poorer verbal fluency.CONCLUSIONThis study underscores the significance of CSF1R imaging as a promising biomarker for brain immune function in Parkinson's disease, which may be associated with cognitive and motor disease severity.FUNDINGPET imaging: the Michael J. Fox Foundation and the RMS Family Foundation. Radiotracer development: NIH (R01AG066464 and P41 EB024495). Postmortem brain tissues: NIH P30 AG066507 and BIOCARD study NIH U19 AG033655.

Keywords: Inflammation; Innate immunity; Neurodegeneration; Neuroscience; Parkinson disease.

Figure 1. Increased levels of CSF1R in brains of human patients with PD (N = 4).

(A) Representative confocal images with DAPI (blue), IBA1 (green), and CSF1R (red) in the midbrain and cingulate cortex of age-matched individuals who were in the healthy control and PD groups. White dashed lines indicate the region for high-magnification images. Scale bars: 200 μm (low-magnification images); 50 μm (high-magnification images). (B) CSF1R⁺ cells with IBA1⁺ cells are quantified (mean ± SEM; dots=individual data points); P values from independent, 2-tailed Student’s t tests. *P < 0.05, PD versus healthy control. (C) Representative immunoblots with CSF1R and β-actin antibodies in the midbrain (MB), cingulate cortex (Cing), posterior cingulate cortex (Post), temporal cortex (Temp), cerebellar cortex (CB) and caudate (Caud) of age-matched individuals who were in the healthy control and PD groups. (D) Relative CSF1R levels normalized to β-actin was quantified (n = 4). Individual data are shown as dots, bars are the mean ± SEM. P values were determined by unpaired 2-tailed Student’s t tests. *P < 0.05, **P < 0.005, ***P < 0.0005, PD versus healthy control.

Figure 2. In vitro CSF1R autoradiography with [³H]JHU11761 in human inferior parietal cortex with and without Parkinson’s disease.

Labels show diagnosis of Parkinson’s disease (PD) or control (C) with donor age and sex. Tritium scales standards on the left depict densities beginning at 5.89 nmol/g (1; yellow) and serially decrease until (7; blue), 0.09 nmol/g. Gray matter B_max is indicated to the left of each case.

Figure 3. B_max of ³H-JHU11761 from human frozen sections in Healthy Controls (N = 6) or people with Parkinson’s disease (N = 6).

Samples from (A) inferior parietal cortex gray matter (IPCGM) and white matter (IPCWM), and (B) white matter and (C) gray matter B_max in the caudate nucleus (CN), midbrain (MB), and basal ganglia (BG). Boxes indicate the interquartile (25th–75th) range, whiskers show 1.5 times interquartile range, the line in the box indicates the median and dots indicate outliers greater than or less than 1.5 times the interquartile range. IPCGM and IPCWM are shown separately due to difference in y-axis scale. Student’s independent t test was used for comparing B_max in HC vs PD, *P < 0.05.

Figure 4. Regional total volume of distribution (V_T) of [¹¹C]CPPC, a CSF1R PET radioligand, in people with mild or moderate PD, defined by motor disability, and controls of a similar age.

Boxes indicate the interquartile (25th–75th) range, whiskers show 1.5 times interquartile range, the line in the box indicates the median and dots indicate outliers greater than or less than 1.5 times the interquartile range. *P < 0.05 and **P < 0.004 for ANOVA and post hoc test between moderate PD (N = 4) and mild PD (N = 10) and or moderate PD vs healthy controls (N = 10). ACC, anterior cingulate cortex; CBx, cerebellar corte;, FC, frontal cortex; Hp, hippocampus; OC, occipital cortex; PC, parietal cortex; PCC, posterior cingulate cortex; Str, striatum; TC, temporal cortex; Th, thalamus.

Figure 5. Correlation between regional [¹¹C]CPPC V_T and motor disability in Parkinson’s disease.

Scatter plots and linear relationship between [¹¹C]CPPC V_T and ADL disability from motor symptoms in people with PD (n = 12), measured by MDS-UPDRS Part II in regions of interest that showed a statistically significant relationship (P < 0.005) using Pearson correlation, including (A) Brainstem and (B) Temporal cortex.