Non-invasive PET Imaging of PARP1 Expression in Glioblastoma Models

Abstract

Purpose: The current study presents [(18)F]PARPi as imaging agent for PARP1 expression.

Procedures: [(18)F]PARPi was generated by conjugating a 2H-phthalazin-1-one scaffold to 4-[(18)F]fluorobenzoic acid. Biochemical assays, optical in vivo competition, biodistribution analysis, positron emission tomography (PET)/X-ray computed tomography, and PET/magnetic resonance imaging studies were performed in subcutaneous and orthotopic mouse models of glioblastoma.

Results: [(18)F]PARPi shows suitable pharmacokinetic properties for brain tumor imaging (IC50 = 2.8 ± 1.1 nM; logPCHI = 2.15 ± 0.41; plasma-free fraction = 63.9 ± 12.6 %) and accumulates selectively in orthotopic brain tumor tissue. Tracer accumulation in subcutaneous brain tumors was 1.82 ± 0.21 %ID/g, whereas in healthy brain, the uptake was only 0.04 ± 0.01 %ID/g.

Conclusions: [(18)F]PARPi is a selective PARP1 imaging agent that can be used to visualize glioblastoma in xenograft and orthotopic mouse models with high precision and good signal/noise ratios. It offers new opportunities to non-invasively image tumor growth and monitor interventions.

Keywords: Glioblastoma; Imaging; Orthotopic; PARP1; PET.

Fig. 1

Structure of [^18/19F]PARPi and biophysical properties. a 4-(4-Fluoro-3-(piperazine-1-carbonyl)benzyl)phthalazin-1(2H)-one (54.5 μmol) was conjugated with 4-fluorobenzoic (65.5 μmol) to yield the final [¹⁹F]PARPi. b Key pharmacokinetic properties of [¹⁹F]PARPi. logP_CHI=logP based on the chemical hydrophobicity index; logP_O/W=logP based on the octanol/water partition coefficient; CHI=chemical hydrophobicity index; IC₅₀=half maximal inhibitory concentration.

Fig. 2

Specificity of [¹⁹F]PARPi uptake. U251 MG or U373 MG were incubated either alone with PARPi-FL (500 nM) or with PARPi-FL (500 nM) plus [¹⁹F]PARPi (tenfold excess), or PARPi-FL (500 nM) plus olaparib (tenfold excess). [¹⁹F]PARPi and olaparib compete for the same binding sites as PARPi-FL, resulting in a reduction of fluorescence intensity in case of specific binding. a Confocal microscopy of U251 MG cells after PARPi-FL or [¹⁹F]PARPi /PARPi-FL injection. Green PARPi-FL signal, blue Hoechst nuclear stain. b Quantification of fluorescence intensity in cell nuclei in U251 MG and U373 MG cells after incubation with only PARPi-FL, [¹⁹F]PARPi/ PARPi-FL, or olaparib/PARPi-FL. Error bars are mean ±standard deviation (SD). P values were calculated with Student’s t-tests, unpaired; *P<0.05.

Fig. 3

[¹⁸F]PARPi in vivo pharmacokinetics. a In vivo blood half-life of [¹⁸F]PARPi (n=3). Blood was collected at different time points (5, 15, 30, 45, 60, 90, 120 min), weighted, and counted. Results expressed as percent injected dose per gram (%ID/g). b Selected tumor-to-non-target-tissues ratio of [¹⁸F]PARPi (n=12) calculated from biodistribution data. c Biodistribution study in selected tissues. After injection of [¹⁸F]PARPi into mice with tumor xenografts, mice were euthanized at 2 h post-injection and radioactivity in organs was measured (n=6/group). Error bars are mean±SD. P values were calculated with Student’s t-tests, unpaired; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Fig. 4

[¹⁸F]PARPi localization in orthotopic U251 MG tumor-bearing mouse brain. a Autoradiography and H&E staining of U251 MG tumor-bearing brain sections of animals injected with [¹⁸F]PARPi or olaparib/[¹⁸F]PARPi. Yellow arrows indicate location of tumor tissue. Quantification of activity on autoradiographic sections in b orthotopic U251 MG tumors and mouse brain and c muscle of tumor-bearing mice that were unblocked and blocked, as well as unblocked healthy mice (n=6). Error bars are mean±SD. P values were calculated with Student’s t-tests, unpaired; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001.

Fig. 5

In vivo whole body PET/CT imaging of [¹⁸F]PARPi in orthotopic brain tumor-bearing mice. a Fused PET/CT coronal images of a brain orthotopic U251 MG tumor-bearing mouse acquired at 2 h post-injection of [¹⁸F]PARPi (left) or the blocking agent olaparib (500-fold excess) followed by [¹⁸F]PARPi (right). b PET quantification of U251 MG tumors from images acquired at 30 min and 2 h post-injection (n=10). c Representative 3D PET/CT images of orthotopic U251 MG tumor-bearing mice after injection of [¹⁸F]PARPi (left) and after pre-injection (30 min before) of blocking agent and [¹⁸F]PARPi (500-fold excess olaparib) (right). Images were acquired at 2 h post-injection. Error bars are mean±SD. P values were calculated with Student’s t-tests, unpaired; *P<0.05.

Fig. 6

In vivo whole body PET/MRI imaging of [¹⁸F]PARPi in orthotopic brain tumor-bearing mice. Coronal view of [¹⁸F]PARPi PET images, contrast-enhanced MRI, and fused PET/MRI of orthotopic U251 MG tumor-bearing mice. PET scans were acquired 2 h after injection of [¹⁸F]PARPi or olaparib/[¹⁸F]PARPi, and MRI scans were acquired 1 min after Magnevist injection. Top row: MRI, PET, and co-registered PET/MRI for a mouse receiving only [¹⁸F]PARPi. Bottom row: MRI, PET, and co-registered PET/MRI for a mouse receiving [¹⁸F]PARPi after a 500-fold excess of olaparib.