Submillimeter resolution positron emission tomography for high-sensitivity mouse brain imaging

Positron emission tomography (PET) is a powerful molecular imaging technique that can provide functional information of living objects. However, the spatial resolution of PET imaging has been limited to around 1 mm which makes it difficult to visualize mouse brain functions in detail. Here we report an ultrahigh resolution small animal PET scanner we developed that can provide a resolution approaching 0.6 mm to visualize the mouse brain functions with unprecedented detail. Methods: The ultrahigh resolution small animal PET scanner had 52.5 mm inner diameter and 51.5 mm axial coverage. The PET scanner consisted of 4 rings each of which had 16 DOI detectors. Each DOI detector consisted of a 3-layer staggered LYSO crystal array with a pitch of 1 mm and 4×4 SiPM array. The physical performance was evaluated in accordance with the NEMA NU4 protocol. The spatial resolution was evaluated with various resolution phantoms. In vivo glucose metabolism imaging of mouse brain was performed. Results: The peak absolute sensitivity was 2.84% with an energy window of 400-600 keV. The 0.55 mm rod structure of a resolution phantom was resolved using the iterative algorithm. The in vivo mouse brain imaging with ¹⁸F-FDG showed clear identification of cortex, thalamus, and hypothalamus which were barely distinguishable in a commercial preclinical PET scanner that we used for imaging comparison. Conclusion: The developed ultrahigh resolution small animal PET scanner is a promising molecular imaging tool for neuroscience research using rodent models.

TRACKING INNATE IMMUNE ACTIVATION IN A MOUSE MODEL OF PARKINSON'S DISEASE USING TREM1 AND TSPO PET TRACERS

Parkinson's disease (PD) is associated with aberrant innate immune responses, including microglial activation and infiltration of peripheral myeloid cells into the central nervous system (CNS). Methods to investigate innate immune activation in PD are limited and have not yet elucidated key interactions between neuroinflammation and peripheral inflammation. Translocator protein 18 kDa (TSPO)-positron emission tomography (PET) is a widely evaluated imaging approach for studying activated microglia and peripheral myeloid lineage cells in vivo, however it is yet to be fully explored in PD. Herein we investigate the utility of TSPO-PET, in addition to PET imaging of triggering receptor expressed on myeloid cells 1 (TREM1) - a novel biomarker of proinflammatory myeloid cells - for detecting innate immune responses in the 6-hydroxydopamine (6-OHDA) mouse model of dopaminergic neuron degeneration. Methods: C57/BL6J and TREM1-knockout mice were stereotaxically injected with 6-OHDA in the left striatum; control mice were saline-injected. At day 7 or 14 post-surgery, mice were administered ¹⁸F-GE-180, ⁶⁴Cu-TREM1-mAb or ⁶⁴Cu-Isotype control-mAb and imaged by PET/CT. Ex vivo autoradiography (ARG) was performed to obtain high resolution images of tracer binding within the brain. Immunohistochemistry was conducted to verify myeloid cell activation and dopaminergic cell death and quantitative PCR and flow cytometry were completed to assess levels of target in the brain. Results: PET/CT images of both tracers showed elevated signal within the striatum of 6-OHDA-injected mice compared to those injected with saline. ARG afforded higher resolution brain images and revealed significant TSPO and TREM1 tracer binding within the ipsilateral striatum of 6-OHDA- compared to saline-injected mice at both 7- and 14-days post-toxin. Interestingly, ¹⁸F-GE-180 enabled detection of inflammation in the brain and peripheral tissues (blood and spleen) of 6-OHDA mice, whereas ⁶⁴Cu-TREM1-mAb appeared to be more sensitive and specific for detecting neuroinflammation, in particular infiltrating myeloid cells, in these mice, as demonstrated by flow cytometry findings and higher tracer binding signal-to-background ratios in brain. Conclusion: TSPO- and TREM1-PET tracers are promising tools for investigating different cell types involved in innate immune activation in the context of dopaminergic neurodegeneration, thus warranting further investigation in other PD rodent models and human postmortem tissue to assess their clinical potential.

Synthesis, preclinical evaluation, and a pilot clinical PET imaging study of 68Ga-labeled FAPI dimer

Cancer-associated fibroblasts (CAFs) are crucial components of the tumor microenvironment. Fibroblast activation protein (FAP) is overexpressed in CAFs. FAP-targeted molecular imaging agents, including FAP inhibitor (FAPI)-04 and FAPI-46, have shown promising results in tumor diagnosis. However, these molecules have relatively short tumor-retention time for peptide-targeted radionuclide therapy applications. We aimed to design a ⁶⁸Ga-labeled FAPI dimer (denoted as ⁶⁸Ga-DOTA-2P(FAPI)2) to optimize the pharmacokinetics and evaluate whether this form is more effective than its monomeric analogs. Methods: ⁶⁸Ga-DOTA-2P(FAPI)2 was synthesized based on the quinoline-based FAPI variants (FAPI-46), and its binding properties were assayed in CAFs. Preclinical pharmacokinetics was determined in FAP-positive patient-derived xenografts (PDXs) using small-animal PET and biodistribution experiments. The effective dosimetry of ⁶⁸Ga-DOTA-2P(FAPI)2 was evaluated in three healthy volunteers, and PET/ CT imaging of ⁶⁸Ga-FAPI-46 and ⁶⁸Ga-DOTA-2P(FAPI)2 was performed in three cancer patients. Results: ⁶⁸Ga-DOTA-2P(FAPI)2 was stable in phosphate-buffered saline and fetal bovine serum for 4 h. The FAPI dimer showed high affinity and specificity for FAP in-vitro and in-vivo. The tumor uptake of ⁶⁸Ga-DOTA-2P(FAPI)2 was approximately two-fold stronger than that of ⁶⁸Ga-FAPI-46 in PDXs, while the healthy organs showed low tracer uptake and fast body clearance. The effective dose of ⁶⁸Ga-DOTA-2P(FAPI)2 was 1.19E-02 mSv/MBq, calculated using OLINDA. Finally, PET/CT scans in three cancer patients revealed higher intratumoral uptake of ⁶⁸Ga-DOTA-2P(FAPI)2 than that of ⁶⁸Ga-FAPI-46 in all tumor lesions (maximum standardized uptake value: 8.1-39.0 vs. 1.7-24.0, respectively; P < 0.001). Conclusion: ⁶⁸Ga-DOTA-2P(FAPI)2 has increased tumor uptake and retention properties compared to ⁶⁸Ga-FAPI-46, and it could be a promising tracer for both diagnostic imaging and targeted therapy of malignant tumors with positive expression of FAP.

Synaptic vesicle glycoprotein 2A is affected in the CNS of Huntington's Disease mice and post-mortem human HD brain

Synaptic dysfunction is a primary mechanism underlying Huntington's Disease (HD) progression. This study investigated changes in synaptic vesicle glycoprotein 2A (SV2A) density by means of ¹¹C-UCB-J microPET imaging in the central nervous system (CNS) of HD mice. METHODS: Dynamic ¹¹C-UCB-J microPET imaging was performed at clinically relevant disease stages (at 3, 7, 10, and 16 months, M) in the heterozygous knock-in Q175DN mouse model of HD and WT littermates (n = 16-18/genotype and time point). Cerebral ¹¹C-UCB-J analyses were performed to assess genotypic differences during pre-symptomatic (3M) and symptomatic (7-16M) disease stages. ¹¹C-UCB-J binding in the spinal cord was quantified at 16M. 3H-UCB-J autoradiography and SV2A immunofluorescence were performed post-mortem in mouse and human brain tissue. RESULTS: ¹¹C-UCB-J binding was declined in symptomatic heterozygous mice compared to WT littermates in parallel with disease progression (7M: p<0.01, 16M: p<0.0001). Specific ¹¹C-UCB-J binding was detectable in the spinal cord, with symptomatic heterozygous mice displaying a significant reduction (p<0.0001). 3H-UCB-J autoradiography and SV2A immunofluorescence corroborated the in vivo measurements demonstrating lowered SV2A in heterozygous mice (p<0.05). Finally, preliminary analysis of SV2A in post-mortem human brain suggested lower SV2A in HD gene carrier compared to nondemented control. CONCLUSION: ¹¹C-UCB-J PET detects SV2A deficits during symptomatic disease in heterozygous mice in both brain and spinal cord, offering a novel marker of synaptic integrity widely distributed in CNS. Upon clinical application, ¹¹C-UCB-J PET imaging yields promise for SV2A measurement in patients with HD during disease progression and following disease-modifying therapeutic strategies.

A Novel Radiotracer for Molecular Imaging and Therapy of Gastrin-Releasing Peptide Receptor Positive Prostate Cancer

The gastrin-releasing peptide receptor (GRPR) is overexpressed in many solid malignancies, particularly in prostate and breast cancers, among others. We synthesized ProBOMB2, a novel bombesin derivative radiolabeled with ⁶⁸Ga and ¹⁷⁷Lu, and evaluated its ability to target GRPR in a preclinical model of human prostate cancer. Methods: ProBOMB2 was synthesized on solid phase using Fmoc chemistry. The chelator 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid was coupled to the N-terminus and separated from the GRPR-targeting sequence by a cationic 4-amino-(1-carboxymethyl)-piperidine spacer. Binding affinity for both human and murine GRPR was determined using a cell-based competition assay, while a calcium efflux assay was used to measure the agonist/antagonist properties of the derivatives. ProBOMB2 was radiolabeled with ¹⁷⁷Lu and ⁶⁸Ga. SPECT and PET imaging, and biodistribution studies were conducted using a preclinical prostate cancer model of male immunocompromised mice bearing GRPR-positive PC-3 human prostate cancer xenografts. Results: Ga-ProBOMB2 and Lu-ProBOMB2 bound to PC-3 cells with a Ki of 4.58±0.67 and 7.29±1.73 nM, respectively. ⁶⁸Ga-ProBOMB2 and ¹⁷⁷Lu-ProBOMB2 were radiolabeled with a radiochemical purity greater than 95%. Both radiotracers were primarily excreted via the renal pathway. PET images of PC-3 tumor xenografts were visualized with excellent contrast at 1 h and 2 h post-injection (p.i.) with ⁶⁸Ga-ProBOMB2, and very low off-target organ accumulation. ¹⁷⁷Lu-ProBOMB2 enabled clear visualization of PC-3 tumor xenografts by SPECT imaging at 1 h, 4 h, and 24 h p.i. ¹⁷⁷Lu-ProBOMB2 displayed higher tumor uptake than ⁶⁸Ga-ProBOMB2 at 1 h p.i. ¹⁷⁷Lu-ProBOMB2 tumor uptake at 1 h, 4 h, and 24 h p.i. was 14.9±3.1, 4.8±2.1, and 1.7±0.3 %ID/g, respectively. Conclusion: ⁶⁸Ga-ProBOMB2 and ¹⁷⁷Lu-ProBOMB2 are promising radiotracers with limited pancreas uptake, good tumor uptake, and favorable pharmacokinetics for imaging and therapy of GRPR-expressing tumors.

PET tracing of biodistribution for orally administered 64Cu-labeled polystyrene in mice

Plastics are used commonly in the world because of their convenience and cost effectiveness. Microplastics, an environmental threat and human health risk, are widely detected in food and consequently ingested. However, degraded plastics are found everywhere, creating an environmental threat and human health risk. Therefore, real-time monitoring of orally administered microplastics to trace them in the body is tremendously important. Methods: In this study, to visualize their absorption path, we labeled polystyrene with [⁶⁴Cu]Cu-DOTA. We prepared radiolabeled polystyrene with ⁶⁴Cu. Afterward, [⁶⁴Cu]Cu-DOTA-polystyrene was orally administered to mice, and we evaluated its transit and absorption using PET imaging. The absorption path and distribution of [⁶⁴Cu]Cu-DOTA-polystyrene were determined using PET over 48 h. Ex vivo tissue radio–thin-layer chromatography (TLC) was used to demonstrate the existence of [⁶⁴Cu]Cu-DOTA-polystyrene in tissue. Results: PET images demonstrated that [⁶⁴Cu]Cu-DOTA-polystyrene began to transit to the intestine within 1 h. Accumulation of [⁶⁴Cu]Cu-DOTA-polystyrene in the liver was also observed. The biodistribution of [⁶⁴Cu]Cu-DOTA-polystyrene confirmed the distribution of [⁶⁴Cu]Cu-DOTA-polystyrene observed on the PET images. Ex vivo radio-TLC demonstrated that the detected γ-rays originated from [⁶⁴Cu]Cu-DOTA-polystyrene. Conclusion: This study provided PET evidence of the existence and accumulation of microplastics in tissue and cross-confirmed the PET findings by ex vivo radio-TLC. This information may be used as the basis for future studies on the toxicity of microplastics.

A longitudinal PET/MR imaging study of colony stimulating factor-1 receptor-mediated microglia depletion in experimental stroke

Microglia-induced neuroinflammation after stroke contributes to the exacerbation of postischemic damage but also supports neurorestorative events. Longitudinal molecular imaging of microglia-targeted therapies will support the assessment of target engagement, therapy efficacy, and deciphering of the mode of action. We investigated the effects of chronic colony-stimulating factor 1 receptor (CSF-1R) inhibitor–mediated microglia depletion on translocator protein (TSPO)–dependent neuroinflammation and cerebrovascular parameters using PET/MRI. Methods: Forty C57BL/6 mice underwent a 30-min transient occlusion of the middle cerebral artery and were randomly assigned to either a control group or a group treated with CSF-1R inhibitor (PLX5622). Eight mice per group were used for N,N-diethyl-2-(2-(4-(2-¹⁸F-fluoroethoxy) phenyl)5,7dimethylpyrazolo[1, 5a]pyrimidin-3-yl)acetamide (¹⁸F-DPA-714) (TSPO) PET imaging on days 7, 14, 21, and 30 after ischemia and behavioral tests before and after surgery. An extra group of 8 mice underwent MRI, including T2-weighted (infarct), perfusion-weighted (cerebral blood flow), and diffusion-weighted (water diffusion, cellular density) sequences, on days 1, 3, 7, 14, 21, and 30. Ex vivo analysis (immunoreactivity, gene expression) was performed to characterize the inflammatory environment. Results: We demonstrated that long-term CSF-1R inhibition transiently decreased the TSPO PET signal within the infarct. Residual TSPO activity was partly due to a potentially resistant Iba-1–positive cell populations with low CSF-1R and transmembrane 119 expression. The decrease in selected pro- and antiinflammatory marker expression suggested an apparent global dampening of the neuroinflammatory response. Furthermore, the temporal changes in the MRI parameters highlighted treatment-induced effects on reperfusion and tissue homeostasis, associated with impaired motor function at late stages. Conclusion: Longitudinal TSPO PET/MRI allows the assessment of target engagement and optimization of drug efficiency. PLX5622 has promising immunomodulatory effects, and the optimal therapeutic time window for its application needs to be defined.

11C-PIB and 124I-antibody PET provide differing estimates of brain amyloid-beta after therapeutic intervention

PET imaging of amyloid-β (Aβ) has become an important component of Alzheimer disease diagnosis. ¹¹C-Pittsburgh compound B (¹¹C-PiB) and analogs bind to fibrillar Aβ. However, levels of nonfibrillar, soluble, aggregates of Aβ appear more dynamic during disease progression and more affected by Aβ-reducing treatments. The aim of this study was to compare an antibody-based PET ligand targeting nonfibrillar Aβ with ¹¹C-PiB after β-secretase (BACE-1) inhibition in 2 Alzheimer disease mouse models at an advanced stage of Aβ pathology. Methods: Transgenic ArcSwe mice (16 mo old) were treated with the BACE-1 inhibitor NB-360 for 2 mo, whereas another group was kept as controls. A third group was analyzed at the age of 16 mo as a baseline. Mice were PET-scanned with ¹¹C-PiB to measure Aβ plaque load followed by a scan with the bispecific radioligand ¹²⁴I-RmAb158-scFv8D3 to investigate nonfibrillar aggregates of Aβ. The same study design was then applied to another mouse model, App^NL-G-F. In this case, NB-360 treatment was initiated at the age of 8 mo and animals were scanned with ¹¹C-PiB-PET and ¹²⁵I-RmAb158-scFv8D3 SPECT. Brain tissue was isolated after scanning, and Aβ levels were assessed. Results:¹²⁴I-RmAb158-scFv8D3 concentrations measured with PET in hippocampus and thalamus of NB-360–treated ArcSwe mice were similar to those observed in baseline animals and significantly lower than concentrations observed in same-age untreated controls. Reduced ¹²⁵I-RmAb158-scFv8D3 retention was also observed with SPECT in hippocampus, cortex, and cerebellum of NB-360–treated App^NL-G-F mice. Radioligand in vivo concentrations corresponded to postmortem brain tissue analysis of soluble Aβ aggregates. For both models, mice treated with NB-360 did not display a reduced ¹¹C-PiB signal compared with untreated controls, and further, both NB-360 and control mice tended, although not reaching significance, to show higher ¹¹C-PiB signal than the baseline groups. Conclusion: This study demonstrated the ability of an antibody-based radioligand to detect changes in brain Aβ levels after anti-Aβ therapy in ArcSwe and App^NL-G-F mice with pronounced Aβ pathology. In contrast, the decreased Aβ levels could not be quantified with ¹¹C-PiB PET, suggesting that these ligands detect different pools of Aβ.

Glitter in the Darkness? Non-fibrillar β-amyloid Plaque Components Significantly Impact the β-amyloid PET Signal in Mouse Models of Alzheimer's Disease

Objective: β-amyloid PET (Aβ-PET) is an important tool for quantification of amyloidosis in the brain of suspected Alzheimer's disease (AD) patients and transgenic AD mouse models. Despite the excellent correlation of Aβ-PET with gold standard immunohistochemical assessments, the relative contributions of fibrillar and non-fibrillar Aβ components to the in vivo Aβ-PET signal remain unclear. Thus, we obtained two murine cerebral amyloidosis models that present with distinct Aβ plaque compositions and performed regression analysis between immunohistochemistry and Aβ PET to determine the biochemical contributions to Aβ-PET signal in vivo. Methods: We investigated groups of AppNL-G-F and APPPS1 mice at three, six and 12 months of age by longitudinal ¹⁸F-florbetaben Aβ-PET and with immunohistochemical analysis of the fibrillar and total Aβ burdens. We then applied group level inter-modality regression models using age and genotype matched sets of fibrillar/ non-fibrillar Aβ data (predictors) and Aβ-PET results (outcome) for both transgenic models. An independent group of double-hit APPPS1 mice with dysfunctional microglia due to knock-out of triggering receptor expression on myeloid cells 2 (Trem2-/-) served for validation and evaluation of translational impact. Results: Neither fibrillar nor non-fibrillar Aβ content alone sufficed to explain the Aβ-PET findings in either transgenic AD model. However, a regression model compiling fibrillar and non-fibrillar Aβ together with the estimate of individual heterogeneity and age at scanning could explain a 93% of variance of the Aβ-PET signal (P<0.001). Fibrillar Aβ burden had a 16-fold higher contribution to the Aβ-PET signal when compared to non-fibrillar Aβ. However, given the relatively greater abundance of non-fibrillar Aβ, we estimate that non-fibrillar Aβ produced 79±25% of the net in vivo Aβ-PET signal in AppNL-G-F mice, and 25±12% in the APPPS1 mice. Corresponding results in separate groups of APPPS1/Trem2-/- and APPPS1/Trem2+/+ mice validated the calculated regression factors and revealed that the altered fibrillarity due to Trem2 knockout impacts the Aβ-PET signal. Conclusion: Taken together, the in vivo Aβ-PET signal derives from the composite of fibrillar and non-fibrillar Aβ plaque components. While fibrillar Aβ has inherently higher PET tracer binding, the greater abundance of non-fibrillar Aβ plaque in AD model mice contributes importantly to the PET signal.

Mesothelin/CD3 half-life extended bispecific T-cell engager molecule shows specific tumor uptake and distributes to mesothelin and CD3 expressing tissues

BiTE ® (bispecific T-cell engager) molecules exert antitumor activity by binding one arm to CD3 on cytotoxic T-cells and the other arm to a tumor-associated antigen. We generated a fully mouse cross-reactive mesothelin (MSLN)-targeted BiTE molecule that is genetically fused to a Fc-domain for half-life extension, and evaluated biodistribution and tumor targeting of a zirconium-89 (89Zr)-labeled MSLN HLE BiTE molecule in 4T1 breast cancer bearing syngeneic mice with positron emission tomography (PET). Biodistribution of 50 µg 89Zr-MLSN HLE BiTE was studied over time by PET imaging in BALB/c mice and revealed uptake in tumor and lymphoid tissues with an elimination half-life of 63.4 hours. Compared to a non-targeting 89Zr-control HLE BiTE, the 89Zr-MLSN HLE BiTE showed a 2-fold higher tumor uptake and higher uptake in lymphoid tissues. Uptake in the tumor colocalized with mesothelin expression, while uptake in the spleen colocalized with CD3 expression. Evaluation of the effect of protein doses on the biodistribution and tumor targeting of 89Zr-MSLN HLE BiTE revealed for all dose groups that uptake in the spleen was faster than in the tumor (day 1 vs day 5). The lowest dose of 10 µg 89Zr-MSLN HLE BiTE had higher spleen uptake and faster blood clearance compared to higher doses of 50 µg and 200 µg. 89Zr-MSLN HLE BiTE tumor uptake was similar at all doses. Conclusion: The MSLN HLE BiTE showed specific tumor uptake and both arms contributed to the biodistribution profile. These findings support the potential for clinical translation of HLE BiTE molecules.

The Latest Advances in Imaging Crosstalk Between the Immune System and Fibrosis in Cardiovascular Disease

Inflammation and fibrosis are hallmarks of tissue repair process and organ failure progression in cardiovascular diseases. Paradigm-shifting research on diverse immune cell populations within the cardiovascular system have enabled discovery of new biomarkers fostering development of diagnostic and therapeutic agents at the molecular level to better manage cardiovascular diseases. To date, a variety of molecular imaging agents have been developed to visualize the biomarkers expressed on immune cells and fibroblasts within their crosstalk network, which drives the pathogenesis of fibrosis triggered by both innate and adaptive immunity. Herein, key biomarkers up-regulated in the immune-fibrosis axis are discussed. The promising molecular imaging agents to reveal this critical pathological process are summarized.

Longitudinal imaging of T-cells and inflammatory demyelination in a preclinical model of multiple sclerosis using 18F-FAraG PET and MRI

Lymphocytes and innate immune cells are key drivers of multiple sclerosis (MS) and are the main target of MS disease-modifying therapies (DMT). Ex vivo analyses of MS lesions have revealed cellular heterogeneity and variable T cell levels, which may have important implications for patient stratification and choice of DMT. Although MRI has proven valuable to monitor DMT efficacy, its lack of specificity for cellular subtypes highlights the need for complementary methods to improve lesion characterization. Here, we evaluated the potential of 2′-deoxy-2′-¹⁸F-fluoro-9-β-d-arabinofuranosylguanine (¹⁸F-FAraG) PET imaging to noninvasively assess infiltrating T cells and to provide, in combination with MRI, a novel tool to determine lesion types. Methods: We used a novel MS mouse model that combines cuprizone and experimental autoimmune encephalomyelitis to reproducibly induce 2 brain inflammatory lesion types, differentiated by their T cell content. ¹⁸F-FAraG PET imaging, T2-weighted MRI, and T1-weighted contrast-enhanced MRI were performed before disease induction, during demyelination with high levels of innate immune cells, and after T cell infiltration. Fingolimod immunotherapy was used to evaluate the ability of PET and MRI to detect therapy response. Ex vivo immunofluorescence analyses for T cells, microglia/macrophages, myelin, and blood–brain barrier (BBB) integrity were performed to validate the in vivo findings. Results:¹⁸F-FAraG signal was significantly increased in the brain and spinal cord at the time point of T cell infiltration. ¹⁸F-FAraG signal from white matter (corpus callosum) and gray matter (cortex, hippocampus) further correlated with T cell density. T2-weighted MRI detected white matter lesions independently of T cells. T1-weighted contrast-enhanced MRI indicated BBB disruption at the time point of T cell infiltration. Fingolimod treatment prevented motor deficits and decreased T cell and microglia/macrophage levels. In agreement, ¹⁸F-FAraG signal was decreased in the brain and spinal cord of fingolimod-treated mice; T1-weighted contrast-enhanced MRI revealed intact BBB, whereas T2-weighted MRI findings remained unchanged. Conclusion: The combination of MRI and ¹⁸F-FAraG PET enables detection of inflammatory demyelination and T cell infiltration in an MS mouse model, providing a new way to evaluate lesion heterogeneity during disease progression and after DMT. On clinical translation, these methods hold great potential for stratifying patients, monitoring MS progression, and determining therapy responses.

A Noninvasive Method for Quantifying Cerebral Metabolic Rate of Oxygen by Hybrid PET/MRI: Validation in a Porcine Model

The gold standard for imaging the cerebral metabolic rate of oxygen (CMRO2) is positron emission tomography (PET); however, it is an invasive and complex procedure that also requires correction for recirculating 15O-H2O and the blood-borne activity. We propose a noninvasive reference-based hybrid PET/magnetic resonance imaging (MRI) method that uses functional MRI techniques to calibrate 15O-O2-PET data. Here, PET/MR imaging of oxidative metabolism (PMROx) was validated in an animal model by comparison to PET-alone measurements. Additionally, we investigated if the MRI-perfusion technique arterial spin labelling (ASL) could be used to further simplify PMROx by replacing 15O-H2O-PET, and if the PMROx was sensitive to anesthetics-induced changes in metabolism. Methods: 15O-H2O and 15O-O2 PET data were acquired in a hybrid PET/MR scanner (3 T Siemens Biograph mMR), together with simultaneous functional MRI (OxFlow and ASL), from juvenile pigs (n = 9). Animals were anesthetized with 3% isoflurane and 6 mL/kg/h propofol for the validation experiments and arterial sampling was performed for PET-alone measurements. PMROx estimates were obtained using whole-brain (WB) CMRO2 from OxFlow and local cerebral blood flow (CBF) from either noninvasive 15O-H2O-PET or ASL (PMROxASL). Changes in metabolism were investigated by increasing the propofol infusion to 20 mL/kg/h. Results: Good agreement and correlation were observed between regional CMRO2 measurements from PMROx and PET-alone. No significant differences were found between OxFlow and PET-only measurements of WB oxygen extraction fraction (0.30 ± 0.09 and 0.31 ± 0.09) and CBF (54.1 ± 16.7 and 56.6 ± 21.0 mL/100 g/min), or between PMROx and PET-only CMRO2 estimates (1.89 ± 0.16 and 1.81 ± 0.10 mLO2/100 g/min). Moreover, PMROx and PMROxASL were sensitive to propofol-induced reduction in CMRO2 Conclusion: This study provides initial validation of a noninvasive PET/MRI technique that circumvents many of the complexities of PET CMRO2 imaging. PMROx does not require arterial sampling and has the potential to reduce PET imaging to 15O-O2 only; however, future validation involving human participants are required.

Nanobody nuclear imaging allows noninvasive quantification of LAG-3 expression by tumor-infiltrating leukocytes and predicts response of immune checkpoint blockade

Recent advances in the field of immune-oncology led to the discovery of next-generation immune checkpoints (ICPs). Lymphocyte activation gene-3 (LAG-3), being the most widely studied among them, is being explored as a target for the treatment of cancer patients. Several antagonistic anti-LAG-3 antibodies are being developed and are prime candidates for clinical application. Furthermore, validated therapies targeting cytotoxic T-lymphocyte–associated protein-4, programmed cell-death protein-1, or programmed cell-death ligand-1 showed that only subsets of patients respond. This finding highlights the need for better tools for patient selection and monitoring. The potential of molecular imaging to detect ICPs noninvasively in cancer is supported by several preclinical and clinical studies. Here, we report on a single-domain antibody to evaluate whole-body LAG-3 expression in various syngeneic mouse cancer models using nuclear imaging. Methods: SPECT/CT scans of tumor-bearing mice were performed 1 h after injection with radiolabeled single-domain antibody. Organs and tumors of mice were isolated and evaluated for the presence of the radiolabeled tracer and LAG-3–expressing immune cells using a γ-counter and flow cytometry respectively. PD-1/LAG-3–blocking antibodies were injected in MC38-bearing mice. Results: The radiolabeled single-domain antibody detected LAG-3 expression on tumor-infiltrating lymphocytes (TILs) as soon as 1 h after injection in MC38, MO4, and TC-1 cancer models. The single-domain antibody tracer visualized a compensatory upregulation of LAG-3 on TILs in MC38 tumors of mice treated with PD-1–blocking antibodies. When PD-1 blockade was combined with LAG-3 blockade, a synergistic effect on tumor growth delay was observed. Conclusion: These findings consolidate LAG-3 as a next-generation ICP and support the use of single-domain antibodies as tools to noninvasively monitor the dynamic evolution of LAG-3 expression by TILs, which could be exploited to predict therapy outcome.

Sensitivity and Isoform Specificity of 18F-Fluorofuranylnorprogesterone for Measuring Progesterone Receptor Protein Response to Estradiol Challenge in Breast Cancer

The purpose of this study was to evaluate the ability of 21-¹⁸F-fluoro-16α,17α-[(R)-(1'-α-furylmethylidene)dioxy]-19-norpregn-4-ene-3,20-dione (¹⁸F-FFNP) to measure alterations in progesterone receptor (PR) protein level and isoform expression in response to estradiol challenge. Methods: T47D human breast cancer cells and female mice-bearing T47D tumor xenografts were treated with 17β-estradiol (E2) to increase PR expression. ¹⁸F-FFNP uptake was measured using cell uptake and tissue biodistribution assays. MDA-MB-231 breast cancer clonal cell lines were generated that express the A or B isoforms of human PR. PR protein levels, transcriptional function, and subcellular localization were determined. In vitro ¹⁸F-FFNP binding was measured via saturation and competitive binding curves. In vivo ¹⁸F-FFNP uptake was measured using tumor xenografts and positron emission tomography. Statistical significance was determined using analysis of variance and t-tests. Results: After 48 and 72 h of E2, ¹⁸F-FFNP uptake in T47D cells was maximally increased compared to both vehicle and 24 h E2 treatment (p<0.0001 vs ethanol; P = 0.02 and P = 0.0002 vs 24 h for 48 and 72 h, respectively). T47D tumor xenografts in mice treated with 72 h E2 had maximal ¹⁸F-FFNP uptake compared to ethanol-treated mice (11.3±1.4 vs 5.2±0.81 %ID/g; P = 0.002). Corresponding tumor-to-muscle uptake ratios were 4.1±0.6, 3.9±0.5, and 2.3±0.4 for 48 h E2, 72 h E2, and ethanol-treated mice, respectively. There was no significant preferential ¹⁸F-FFNP binding or uptake by PR-A versus PR-B in the PR isoform-specific cell lines and tumor xenografts. Conclusion:¹⁸F-FFNP is capable of measuring estrogen-induced shifts in total PR expression in human breast cancer cells and tumor xenografts with equivalent isoform binding.