Skip to main content

Imaging of Neuroinflammation in Neurodegenerative Diseases

Final Report Summary - INMIND (Imaging of Neuroinflammation in Neurodegenerative Diseases)

Executive Summary:
The INMiND project focussed on the study of molecular mechanisms of neuroinflammation (NI) in neurodegenerative diseases (NDs; NIND). In a concerted action among all partners, a large amount of cellular and in vivo studies have been performed to assess the dynamic pattern of microglial activation and its relation to neuroinflammation and neuroregeneration (toxicity and repair) at various disease stages (early, late, under therapy) in various NDs with the dual aim to develop and validate novel animal models and imaging biomarkers (PET, SPECT, MR and OI) capable to assess qualitatively and quantitatively regulation of microglial activity and function in vivo and to validate the outcome of known and newly developed neuroprotective strategies in preclinical models and patients. The work was embedded within 10 interrelated work packages (WPs, Annex Figure 1) related to basic mechanisms of NIND regulation (WP1/2), development of imaging probes for NIND detection (WP3/4), validation of targets and imaging probes in animal models of NIND (WP5/6), applications in humans with NIND (WP8/9), quantification of imaging biomarkers (WP7) and training of a new generation of cross-disciplinary scientists in the field of NIND (WP12). Beside the scientific WPs, 2 WPs dealt with management, dissemination and integration (WP10/11).
Among the main achievements of the project are the identification of novel mechanisms of microglia regulation and function under inflammatory stimuli, based on the definition of new molecular targets and the use of standardised protocols for cell isolation, culture and model induction (e.g. M1-LPS, M2-IL4) and the successful implementation of this knowledge in various NIND models to investigate the temporal profile of NIND in vivo, generate novel NIND animal models (hP2X7, α-synuclein, LPS, IL4, IL13) or evaluate pharmacological manipulations (WP1/2/5/6). The interrelation between microglial activation and neuroregeneration (NR) was characterised in vivo (WP2) and studies investigated the role of sex hormones and age- and gender-specific differences in microglial activation, NI and NR (WP1/2). A wide range of novel radiotracers for NIND targets identified at the beginning of the project (TSPO and non-TSPO such as MMP, P2X7R, CB2R, H4R, MAO-B) have been produced, optimised (WP3) and made available for human or animal studies (WP5/6/8). Especially notable [11C]SMW139 and [11C]JNJ717, two new PET tracers targeting the P2X7 receptor that were unknown before the start of INMIND. Furthermore, new knowledge on the molecular targets allowed the identification / validation of new biological targets for PET imaging of M2 microglia phenotypes (QPCT, CGRP, P2RY12 and P2RY14) (WP1/2/3/5). Several lead and first radiolabelled compounds (WP3) as well as highly sensitive micro- and nano-sized probes suitably designed for the visualisation of NI by MR and optical modalities (WP4) have been synthesised, optimised and tested. The developed MR agents enabled visualisation of circulating immune cells, intravascular inflammation markers and intracerebral targets with potential for dual PET-MR detection. Multi-modal imaging studies (PET, SPECT, MRI, OI) have been carried out in a range of pre/clinical NIND disease models and conditions to assess the dynamics of microglial activation and related NR and ND (WP5/6/8). NI imaging findings were correlated with histopathological findings and with imaging findings related to other disease-specific hallmarks (amyloidosis, astrogliosis, BBB permeability, cell density, connectivity). In these imaging studies a variety of PET tracers were extensively evaluated. Several neuroprotective and -regenerative strategies were evaluated with imaging readouts to monitor treatment response in animal models (WP1/2/5/6) and human studies (WP8). The first image-guided RPCT to validate a novel protective strategy based on the use of TNF-α inhibition in MCI subjects has been conducted, analysis of the predicted protective effect is still ongoing (WP9). To allow uniform data analysis in pre/clinical imaging studies, standardised procedures for quantitative PET analytical methods and tracer kinetic models were developed (WP5/6/7/8) and a European database of (R)-[11C]PK11195 data in healthy volunteers completed (WP7). Furthermore, to sustain the project outcome timely dissemination of generated knowledge foreground has been achieved by publication of 328 peer-reviewed papers and presentation of 442 lectures and 369 posters at 399 national and international conferences. Training of young researchers was ensured through the organisation of 31 dedicated, high-level training courses attracting in total 473 participants from Europe, North America, Asia and the Arabic World (WP12) and the graduation of 41 PhD students.
Project Context and Objectives:
Neurodegenerative diseases (NDs) such as Alzheimer’s disease (AD), frontotemporal dementia (FTD), Parkinson’s disease (PD), Huntington’s disease (HD) and amyotrophic lateral sclerosis (ALS) are the most common chronic neurological disorders. Among these, AD and PD are the most frequent ones with approximately 10% of the population above the age of 65 being affected. With aging of the population, particularly in industrialized countries the absolute number of patients with ND is expected to increase significantly within the next decades. Symptoms of AD are loss of memory, progressive impairment of cognition and neuropsychiatric disturbances. PD patients develop motor disturbances (bradykinesia, rigidity, tremor, postural instability), autonomic dysfunction, cognitive impairment and depression. HD, a monogenetic neurodegenerative disorder, affects muscle coordination and cognitive functions. In this disease, physical symptoms can begin at any age from infancy (6% of cases) to old age, but usually appear between 35 and 44 years of age. ALS has a projected life-time risk of 1/2000 and is the most common motor neuron disease. Also ALS manifests at a relative early age (~55 years). Degeneration of motor neurons leads to progressive muscle wasting/weakness, spasticity and respiratory weakness leading to death usually within 2-5 years. Common to each of these NDs is that
• normal life expectancy is reduced
• associated impairment of cognitive and motor functions is devastating at the personal and family level and has major impact on the health care system, not only with regards to direct medical but also to indirect care costs
• only symptomatic and no curative treatment options exist so far, many of which have significant side effects; e.g. for AD and PD, drugs are employed that elevate levels of affected central neurotransmitters (acetylcholine, dopamine) or bind to their corresponding receptors (e.g. dopamine receptor)
• the “protein turn-over machinery” is impaired leading to deposition of extra- and/or intra-cellular protein aggregates, which in turn induces neuroinflammation (NI) with the activation and recruitment of cells of the immune system (primarily microglia) contributing to neurotoxicity and degeneration of neurons
• diagnosis is only possible at a relatively late stage in the disease course, i.e. when clinical symptoms appear, which in PD, for example, happens when over 50% of disease-specific neurons (nigro-striatal projections) are already damaged
In AD and PD it is still unclear whether disease progression is driven by altered protein metabolism and/or activation of a microglial response. Activated microglia exist in various phenotypes depending on the nature and stage of disease. Microglial responsiveness to injury suggests that it may serve as marker for disease activity and progression. It may also represent a target for diagnostic probes.

Therefore, the main aims of the INMiND project were (i) to increase the understanding of the basic mechanisms of regulation of microglial activity and related functions (toxicity and repair) by imaging their dynamics in vivo; (ii) to employ activated microglia as a biomarker of NIND disease activity in animal model systems and patients with NIND (diagnosis); and (iii) to non-invasively follow the neuroprotective effects of immunomodulatory therapies in patients with NIND in vivo in order to preserve neuronal function as early as possible in the disease course. With that, INMiND aimed to identify molecular target / imaging biomarker combinations, which can be used for early diagnosis, assessment of disease activity and therapy (theranostics). Knowledge and paradigms generated by the INMiND project will also be translatable to the development of therapeutic paradigms for a broad range of NINDs.

To reach these goals the INMiND project aimed to (i) investigate the concerted action of protein-deposition and inflammatory response on the neurodegenerative and neuroregenerative processes, not only at the cellular level but also in the living system (animal model, patient); (ii) investigate the role of microglial activation on the disease process (early diagnosis, activity, response to therapy) employing improved and innovative imaging technologies, which will allow for assessment of the dynamics of microglial activity in vivo over time; (iii) investigate genetic and pharmacological strategies to modulate inflammatory responses favouring neuroprotection and neuroregeneration; (iv) develop disease markers for early diagnosis to allow for application of neuroprotective/ immunomodulatory strategies that are aimed at preventing inflammation-induced neuronal death as early as possible in the disease course.

The following Objectives, which were embedded within a translational and reverse-translational strategy were the focus of the INMiND project:
1. Identify novel mechanisms of regulation and function of microglia under inflammatory stimuli and in ND-models (WP1/2), which will support the development of (i) novel imaging biomarkers of NI and (ii) therapeutic strategies of NIND. To achieve the aims of this objective, within WP1 the basic pathophysiological mechanisms of microglia activation were investigated by studying the expression of genes and proteins identified as markers of specific phenotypes and by evaluating selected features of microglia activity. Experimental systems, such as (i) microglia cells in culture derived from animal models of NIND (AD, PD, HD, ALS) and (ii) newly generated transgenic and multimodality reporter animals were implemented and enabled to evaluate the dynamics of microglia phenotype occurrence in vitro as well as in vivo during disease progression in a non-invasive manner. To elucidate the gender involvement in NI and ND, most of the in vivo studies were carried out in animals of both sexes and with pharmacological manipulation to assess the genetic and hormonal contribution to microglia activation during NI and ND. Novel molecular targets were identified to improve the assessment of microglia activation and served as specific targets for new imaging biomarkers developed in WP3/4. By the combined activities of WP1/2/3/4, first novel probes for known markers of microglia activation were generated and validated promptly in preclinical (WP3) and translational studies (WP5). As a long-term goal, dedicated probes specific for the M2 phenotype were developed, which will allow to directly assess this protective activation pathway of microglia in vivo.
2. Identify the interplay between microglial activation and neuroregeneration induced by neural stem cells (WP2). The reciprocal signalling activities between microglia phenotypes/activation states and NPCs were studied in specifically established culture systems and animal models and the impact on neurogenesis and regeneration in vivo was assessed.
3. Identify molecular targets for ND therapeutics and advanced imaging probes, which bind to activated microglia and BMD cells (WP1/2/3/4/5/6). This objective concentrated on the ‘phenotypic profile’ of microglia and its suitability as target for ND therapeutics and advanced imaging probes through the design, development and evaluation of imaging biomarkers for non-invasive detection (diagnosis, therapy follow-up) of NIND by PET, SPECT (WP3), MRI and OI (WP4). Radiotracer development focussed on several pre-identified targets (TSPO, P2X7R, H4R, CB2R, MMP-9, COX-2, MAO-B), as well as on several new ones assessed within WP1/2 (P2RY12, P2RY14, QPCT, CGRP). In a translational and reverse-translational manner, continuous optimisation of the imaging biomarkers enabled application for the assessment of therapy effects in NIND. The design of the MRI/OI probes (WP4) primarily aimed at obtaining complementary information with respect to the developed PET/SPECT probes (WP3) and focussed mainly on MRI/OI probes to assess the role of blood infiltrating macrophages/BMD cells in NI and for targeting NI biomarkers such as selectins, integrins, MMP or TSPO.
4. Quantitatively assess dynamics of microglial activation in various ND models in conjunction with other disease-specific imaging markers (e.g. Aβ, neurotransmitter function) using standard and newly developed probes (WP5/6/7). In a panel of animal models for NI the course of microglial activation in vivo together with that of ND was assessed (WP5) and extensively validated with existing and newly developed probes (WP3/4) to non-invasively quantify NI. Our experimental models ranged from acute challenges triggering very intense peaks of microglial activation to more clinically relevant models of ND in which the microglial reaction progressively builds up and is comparatively modest. In vivo results were correlated with ex vivo data for assessment of target upregulation and microglial activation by means of autoradiography, immunohistochemistry and (for some experiments) microdialysis (radioligand pharmacokinetics in blood and extracellular brain fluid) and in primary microglial cultures. The primarily analysed imaging target for NI was the one best known up to now (TSPO) and (R)-[11C]PK11195 was used as standard imaging biomarker with which newly developed tracers were compared. Candidate drugs for targets that are known to be upregulated in microglia were explored for labelling (WP3/4) and subsequently tested in vivo (WP5/6). Finally, new biological targets identified to be up-regulated in specific subtypes of reactive microglia in WP1/2 served for the development of a new generation of phenotype-specific compounds (WP3/4) and were systematically investigated in various animal models of NIND (WP5/6). Most importantly, this approach supported not only the generation of molecules that allow non-invasive quantification (WP7) of reactive microglia in living subjects, but also identification of putative targets for therapeutic intervention. The goal of assessing the dynamic course of microglial activation in vivo was combined with Objective 6 to investigate the relationship between therapeutic intervention of NI and protection of neuronal function and outcome employing (i) in vivo imaging biomarkers for amyloid-β ([11C]PIB), cholinergic neurons ([11C]MP4A), dopaminergic neurons ([18F]LBT-999; [123I]ioflupane) and glucose metabolism ([18F]FDG); (ii) behavioural analysis; and (iii) post-mortem autoradiography and/or histopathological analysis. Furthermore, in a subset of ND models all generated expertise was employed to develop improved animal model systems, including multimodal imaging and reporter-based systems, enabling in vivo studies in the same animal to investigate the interrelationship between NI and ND over time (longitudinal studies) and space (locations in the brain) and the role of endogenous neural stem cells (endoNSC) in ND progression and outcome (WP1/2, WP6). Different imaging modalities were used to evaluate and validate the most suitable, most sensitive, and least mutually interfering quantitative readouts for NIND established in WP5. This also included validation of intrinsic MRI contrast parameters as imaging markers for early, progressive and late ND. To study the influence of NI on proliferation and recruitment of endoNSCs, different imaging modalities were available in different teams (OI, µPET, µMRI) and used because of their complementary information on endoNSC behaviour (proliferation, migration, differentiation) in vivo. The dynamics of NI in relation to neurodegeneration and neuroregeneration in vivo was studied in a wide range of animal models of NIND to get the best insights into NI-related disease dynamics. As correct interpretation of PET images is not possible without quantitative studies, the project focussed on the development of methods for quantification of microglia-related probe accumulation in vivo (WP7) such as full kinetic analysis, plasma input models, reference tissue models, supervised cluster and other novel clustering analysis methods, optimised MRI and non-MRI based partial volume correction methods or simplified (semi)-quantitative analysis methods. This is crucial, as (i) in vivo uptake of a ligand depends on physiological processes (delivery, perfusion, extraction, binding to a receptor); (ii) only binding to a receptor associated with activated microglia is relevant requiring a kinetic model to extract binding information from imaging data; and (iii) quantification is essential for assessment of disease progression and modification by therapy. Animal models of NIND are limited and correlate only in part with the entire spectrum of human NIND pathology. Therefore, the true value of studies in the preclinical situation arises when they are being translated into human application (Objective 5). In the INMiND project there has been a very strong interaction between the studies in animals (WP5/6) and in humans (WP8/9) in a translational and reverse-translational strategy. In summary, we investigated the dynamics of NI in relation to neurodegeneration and neuroregeneration in vivo in a wide range of animal models of NIND to get the best insights into NI-related disease dynamics.
5. Translate new imaging paradigms for microglial activation into patients with various ND (AD, MCI, PD, HD, ALS) and multiple sclerosis (MS) with and without immunomodulatory therapies (WP7/8/9). The aims within this objective were achieved in WP8/9 by (i) evaluation of new tracers for microglial activation markers (TSPO, P2X7R, CB2R, H4R, MMP9, COX2, newly identified targets) developed and implemented through WP1/2/3/4/5/6 in various forms and stages of NINDs, including genetic (“at risk”) patient groups in longitudinal studies and correlate the PET measures with clinical parameters; (ii) study of non-ND disorders yielding complementary information on the in vivo assessment of NI that are of relevance for ND; and (iii) use of (R)-[11C]PK11195 (or an improved marker) as an outcome measure of immune-modulation therapies. From the assumption that disease pathogenesis is ongoing for years, if not decades, before clinical onset, earlier intervention studies in subjects with a high risk of developing AD are needed. In line with this, an interventional trial was started in amnestic MCI patients aiming to influence amyloid deposition to investigate further the dynamic interplay between amyloid and microglial activation over time. This randomised placebo-controlled trial in MCI subjects with increased retention of in vivo amyloid examined the effects of an early immunomodulatory therapy aimed at reducing detrimental microglial activation, while maintaining the physiological and regenerative microglia functions, and its relationship to clinical and in vivo neuropathological outcomes. In this study multi-modal PET/MRI imaging with [18F]florbetapir and (R)-[11C]PK11195 were performed before intervention and at (long-term) follow-up and correlated to clinical phenotype. The final aim is to (i) establish new diagnostic avenues for NI for disease diagnosis and therapy follow-up in vivo and (ii) study the engagement of the adaptive immune response on disease progression using high-end imaging as well as clinical end-points to provide the ultimate goal of image-guided disease staging and therapy assessment in vivo.
6. Validate protective strategies for neurons and axons for improved disease outcome (WP1/2/5/6/8/9). INMiND investigated outcome measures of immuno-modulatory and potentially neuroprotective strategies in cell culture (WP1/2), animals (WP5/6) and humans (WP8/9) using high-end and complementary imaging modalities (longitudinal BLI, PET/SPECT and combined MRI measures, e.g. for microglial activity, neuronal integrity, astrocytosis, perfusion, brain metabolites, DKI/DTI) in conjunction with invasive measures (e.g. neuropathological and immunohistochemical techniques) to demonstrate and validate the efficacy of interventional paradigms in the in vivo application by the use of drugs with known anti-inflammatory action (NSAID; estrogen; PPARγ agonists; FK506; minocycline), as well as by novel strategies that included the use of drugs targeting molecules strongly upregulated in reactive microglia, e.g. pharmacologically relevant doses of the non-radiolabelled versions of the used PET tracers.
7. Train a new generation of cross-disciplinary scientists in the field of NIND for biomedical research (WP12). This has been successfully achieved within WP12 by (i) organisation and control of efficient and complementary high-level training activities within the field of INMiND-related research subjects at dedicated INMiND training sites and which were open to the Pan-European and broader research community, (ii) promotion and support of these training activities, and (iii) facilitation of the exchange and mobility of students and researchers within the consortium.
Project Results:
Per project objective the following achievements and results have been accomplished and are specified more in detail per WP in the paragraphs below.

Objective 1: Identification of novel mechanisms of regulation and function of microglia under inflammatory stimuli and in ND models.
(i) improved knowledge on novel mechanisms of regulation of microglial activation and functions that are induced by selected endogenous regulatory systems;
(ii) delivery of candidate molecules serving the aim of objectives 3, 5 and 6;
(iii) additional basic elements assessing the therapeutic effects of drugs widely prescribed to the elderly population (e.g. PPARγ agonists; estrogenic drugs) which act through these endogenous systems;
(iv) identification of selected drugs endowed with beneficial potential in NIND;
(v) optimised imaging biomarkers for the assessment of disease progression and response to therapy;
(vi) newly generated animal models, which allow investigations on the occurrence of specific microglia phenotypes at defined disease stages and of the expression of NI imaging targets under longitudinal studies of ND conditions.

Objective 2: Identification of the interplay between microglia activation and neuroregeneration induced by neural stem cells.
(i) further developed and newly established neural microglia culture systems as well as NPCs-microglia co-cultures;
(ii) improved understanding of the various functions and molecular signatures of different microglial phenotypes (M1/M2/Mreg) in the in vivo system;
(iii) identified activation state-specific and ND-specific microglia-derived activities on NPCs and regeneration;
(iv) identified NPCs-derived activities on microglia activation and microglia phenotypes in ND;
(v) improved knowledge of gender-dependent effects on NI and neurogenesis/regeneration in ND.

Objective 3: Identification of molecular targets for ND therapeutics and development of advanced imaging probes which target activated microglia and BMD cells.
(i) more favourable pharmacokinetics;
(ii) easier process of quantification;
(iii) biological target;
(iv) microglia “phenotypic profile”.

Objective 4: Quantitative assessment of the dynamics of microglial activation in various ND models in conjunction with other disease-specific imaging biomarkers (e.g. Aβ, neurotransmitter function) using standard and newly developed probes.
(i) implementation of improved markers for quantitative assessment of TSPO;
(ii) generation of new complementary molecular target/imaging biomarker combinations (P2X7R, H4R, CB2R, COX2, MMP9, MAO-B);
(iii) discovery of new molecular target/imaging biomarker combinations (P2RY12, P2RY14, QPCT, CGRP discovered and developed in WP1/2/3/4);
(iv) comprehensive investigation of the discovered new molecular target/imaging biomarker combinations in a panel of NIND model systems by non-invasive as well as invasive (validation) measures;
(v) new insights into the in vivo dynamics, not only of single NI markers, but also in the concerted interplay of multiple NI markers in disease pathogenesis in the in vivo system;
(vi) direct correlation of the extent and degree of NI as assessed in vivo with resulting neuronal damage as measured by a series of imaging biomarkers for the integrity of cholinergic (AD), dopaminergic (PD) and serotonergic neurons as well as global glucose metabolism and advanced neuronal and axonal MRI markers of neuronal damage in vivo.

Objective 5: Translation of new imaging paradigms for microglial activation into patients with various ND (AD, PD, HD, ALS) and MS with and without immunomodulatory therapies.
(i) improved technical development on
a. how to assess microglia activation (based on improved TSPO ligands and new biomarkers for P2X7R, CB2R, H4R, COX2, MMP9, MAO-B and further targets defined) and the related astroglia response in NIND in vivo
b. how to assess various disease parameters (e.g. for AD: amyloid load, microglial activation, cholinergic function) together with morphological and clinical parameters in the same subject over time;
(ii) improved knowledge on how NI processes influence the course of ND in humans, and how we may be able to therapeutically intervene with these processes on the basis of the full range of
a. proof-of-principle studies in various NIND and other brain diseases (including immunohistochemical correlations)
b. studies at very early onset of NIND (genetic “at risk”; MCI)
c. longitudinal studies
d. interventional studies (immune-modulation);
(iii) improved knowledge on disease pathophysiology;
(iv) development of new, innovative imaging-guided therapy paradigms (theranostics).

Objective 6: Validation of protective strategies for neurons and axons to improve disease outcome.
(i) combined assessment by various and complementary imaging modalities of
a. microglial alteration and of
b. neuronal function, neurotransmission and -integrity in physiologically intact organisms in vivo (animal models, humans);
(ii) in vivo quantification of the dynamics of NI-driving molecular mechanisms and their impact on neuronal integrity at various disease stages (onset, aging, immune-modulation);
(iii) validation of putative effective anti-inflammatory therapies that represent possible protective strategies for neurons and axons for improved disease outcome by a broad panel of in vivo measurements of molecular parameters;
(iv) improved knowledge on the impact of NI on neurodegenerative processes to pave the way for an improved understanding of how neuro-regenerative strategies for NINDs could be developed further.

Objective 7: Training of a new generation of cross-disciplinary scientists in the field of NIND for biomedical research
(i) 31 training courses
(ii) 125 days of training activity
(iii) 473 trainees, including 189 students
(iv) 134 trainees from other INMiND partners
(v) 8 trainees from SMEs and industry, including 5 trainees from INMiND SMEs
(vi) trainees from 23 different countries
(vii) successful establishment of the European Molecular Imaging Doctoral School network (emids) involving 3 INMiND partners
(viii) promotion and support ensured by the French Infrastructure France Life Imaging (FLI) leading to the financial support of 24 PhD students
(ix) sustainability of the training actions with the submission of 2 European Training Networks project proposals (H2020/MSCA) coordinated by 2 INMiND partners in 2017 and 2018 WP1 - Basic mechanisms of microglia activation and new molecular targets

Standard protocols for microglia imaging and biochemical characterisation in vitro have been defined. In particular, specific murine markers such as:
• General microglia mRNA markers: Iba1, C1qa, Slc13a3, Speg, Fcrls
• M1 mRNA markers: Irg1, Cxcl2, Csf3, Cxcl10
• M2 mRNA markers: Ym1, Fizz1, Mrc-1
have been tested, validated and used as endpoints to substantiate a novel M2 murine model, the IL4 ICV model. To this purpose, the responses to the M2 stimulus of microglia in different cerebral areas have been assessed. Most of the selected murine biomarkers can be considered as reliable: for this reason, they have been used as standard markers to assess the activation state of microglia in murine brain areas. Basic research mainly focused on the anti-inflammatory (M2) phenotype of microglia; in particular, the molecular mechanisms involved in the expression of the mRNA of typical M2 markers in mouse mixed glia and pure microglia cell cultures stimulated with IL4 have been studied. Investigation on specific anti-inflammatory pathways, as Stat6, Jak1 and Jak3 revealed that the mRNA induction of M2 genes requires prior expression of Stat6/Jak1/Jak3-dependent proteins. Genome-wide analyses carried out after IL4 treatment in the absence or presence of a Jak3 inhibitor led to the identification of a Jak3-dependent gene that was not previously reported. The described results improved our knowledge on novel mechanisms of regulation of microglial activation and functions that are induced by selected endogenous regulatory systems (Objectives 1.i and 2.ii).
A protocol for the isolation of adult mouse brain microglia has been developed and validated (Pepe G, Calderazzi G, De Maglie M, Villa AM, Vegeto E. Heterogeneous induction of microglia M2a phenotype by central administration of interleukin-4. J Neuroinflammation. 2014 Dec 31;11:211. doi: 10.1186/s12974-014-0211-6); also rat and human microglia primary cultures with high purity have been obtained. Moreover, co-culture of primary microglia and neurons have been established (Objective 2.i). Primary neurons and N2a cells were exposed to medium conditioned by activated microglia (also obtained from APP/PS1 mice) to assess autophagy markers. Autophagy was compromised in a time-dependent manner, however, this effect was prevented by NLRP3 KO (Objective 2.i and 2.ii). To assess the effect of acute peripheral inflammation on learning, memory and synaptic plasticity, young and old wild-type and APP/PS1 transgenic mice were challenged with LPS. Peripheral administration of LPS led to a reduction of pre- and post-synaptic proteins, activation of microglia, and deficient performances in behavioural paradigms in old animals, which were accentuated by the APP/PS1 genotype and sustained long after immune challenge (Objective 6.iv). A dysregulation of the processes controlling the synthesis and degradation of the ECM has been also observed in early stages of AD. ECM changes affecting brain microvascular functions could therefore drive disease progression and provide potential new early investigational biomarkers in AD (Objective 3.iii) (Lepelletier FX, Mann DM, Robinson AC, Pinteaux E, Boutin H. Early changes in extracellular matrix in Alzheimer's disease. Neuropathol Appl Neurobiol. 2017 Feb;43(2):167-182. doi: 10.1111/nan.12295). By targeted injection of a lentiviral vector (LV) or implantation of genetically modified mesenchymal stem cells (MSC), in order to express IL-13 in the corpus callosum (CC) of cuprizone-treated mice, it has been demonstrated that modulation of inflammatory processes reduces oligodendrocyte loss and demyelination (Objective 6.iv) (Guglielmetti C, Le Blon D, Santermans E, Salas-Perdomo A, Daans J, De Vocht N, Shah D, Hoornaert C, Praet J, Peerlings J, Kara F, Bigot C, Mai Z, Goossens H, Hens N, Hendrix S, Verhoye M, Planas AM, Berneman Z, van der Linden A, Ponsaerts P. Interleukin-13 immune gene therapy prevents CNS inflammation and demyelination via alternative activation of microglia and macrophages. Glia. 2016 Dec;64(12):2181-2200. doi: 10.1002/glia.23053). The NF-κB reporter mouse has been exploited to follow the progression of EAE by means of non-invasive in vivo molecular imaging. These experiments provide new elements useful to characterise microglial activation and function in ND (Objective 4.v).
The effects of pharmacologic manipulation of nuclear receptors (Estrogen Receptor α, ERα) and ion channels on the pro- and anti-inflammatory activation of microglia have been extensively investigated (Objective 1.iii). In particular, the effects of estrogen treatment on polarisation states have been investigated and the molecular signalling pathways involved have been identified and validated in vitro. A previously unreported ERα-mediated effect of E2 on the inflammatory machinery has been identified: experiments on RAW 264.7 cells show that the E2-dependent activation of ERα shortens the LPS-induced pro-inflammatory phase thus triggering the resolution of inflammation toward the IL10-dependent “acquired deactivation” phenotype, which is responsible for tissue remodelling and the restoration of homeostatic conditions (Objective 1.iii) (Villa A, Rizzi N, Vegeto E, Ciana P, Maggi A. Estrogen accelerates the resolution of inflammation in macrophagic cells. Sci Rep. 2015 Oct 19;5:15224. doi: 10.1038/srep15224). In addition, experiments assessing the involvement of the IK channels (KCa3.1 and KCa1.1) in microglia activation revealed an important role in the downstream activation pathways of IL4. New small molecules acting on the KCa3.1 channel have been synthesised (Objectives 1.ii and 1.iv) (Blomster LV, Strøbaek D, Hougaard C, Klein J, Pinborg LH, Mikkelsen JD, Christophersen P. Quantification of the functional expression of the Ca2+-activated K+ channel KCa3.1 on microglia from adult human neocortical tissue. Glia. 2016 Dec;64(12):2065-2078. doi: 10.1002/glia.23040). Also the action exerted by various TSPO ligands on the inflammatory response elicited in pure rat microglial cells by LPS and amyloid treatment has been assessed, showing that the TSPO compounds are able to inhibit the LPS response (Objectives 1.ii and 1.iv). The effect of leukotrienes and leukotriene signalling inhibition on the mRNA expression levels of the most promising M2 marker P2Y12 have been evaluated. Results show a significant reduction of the expression of the marker when treating leukotriene-stimulated BV-2 cells with a leukotriene receptor inhibitor in comparison to leukotriene treatment alone (Objectives 1.ii and 1.iv).
In order to assess the microglia phenotypic profile (Objective 3.iv) in physiological conditions, and to evaluate the influence of genetic sex, RNA-sequencing (RNAseq) analysis has been performed on microglia isolated from the brain of adult mice of both sexes (Objective 2.v). Moreover, to determine the microglia phenotypic profile following estrogen replacement, analyses of the transcriptome in microglia and macrophages have been performed by means of RNAseq also on estrogen-treated ovariectomised mice. This experiment allowed to investigate in these models the effect of intracellular receptor manipulations on NI (Objective 1.iii) and to characterise in vivo the major molecular pathways activated by estrogens in microglia (Villa A, et al. The sex of microglia. Under review). In addition, PPAR animal models have been generated and used to assess the involvement of these intracellular receptors in ALS (Objective 1.iii) (Benedusi V, Martorana F, Brambilla L, Maggi A, Rossi D. The peroxisome proliferator-activated receptor gamma (PPARgamma) controls natural protective mechanisms against lipid peroxidation in amyotrophic lateral sclerosis. J Biol Chem. 2012 Oct 19;287(43):35899-911. doi: 10.1074/jbc.M112.366419). Taken together, these results provide novel information on the basic mechanisms of microglial activation and on molecular signalling pathways activated by endogenous regulators of NI (Objectives 1.i and 2.ii).
Most of these experiments have been carried out in animals of both sexes and after manipulation of sex hormone signalling. The large body of evidence so far generated suggests that sex differences exist in microglia functions in health and disease. The sex hormone estradiol plays an important role in this sexual dimorphism. The role of estrogens in neuroinflammation and neurodegeneration has been extensively reported in Villa A, Vegeto E, Poletti A, Maggi A. Estrogens, neuroinflammation, and neurodegeneration. Endocr Rev. 2016 Aug;37(4):372-402. doi: 10.1210/er.2016-1007 (Objective 2.v).
New candidate molecular markers (P2RY12, RAMP1, CALCRL, QPCT, P2RY14) for M2 polarisation state have been identified as potential targets for PET imaging of M1/M2 phenotypes in rodents and humans. The proposed markers have been screened by biologists (P2, P3, P6 as part of WP1) and radiochemists (P4, P8, P32, P13a, P16 and P23 as part of WP3) to assess the biological and chemical suitability of each target. The most promising biomarkers (P2RY12, RAMP1, CALCRL) have been investigated and tested in microglia isolated from adult mice and highly pure human microglia primary cultures, thus identifying their role as possible translational M2 markers (Objectives 1.v and 4.iii). Moreover, the expression of the most performing biomarker (P2RY12) has been tested in rodent models of acute neuroinflammation (pMCAO) and in human samples of stroke patients, revealing a differential expression of P2RY12 following the pathology. These data have been provided to WP3 together with murine (brains of mice treated with IL4 ICV) and human (tissue sections of stroke patients) samples to develop and test new radiotracers for the assessment of activated microglia (Objective 4.iii) (Villa A, et al. Identification of new molecular targets for PET imaging of microglial anti-inflammatory phenotype. Submitted).
Finally, the M1 reporter vector has been developed and validated in vitro (Objective then M1 reporter animals have been generated and the reporter has been validated in vivo (Objective (Rizzi N, Rebecchi M, Levandis G, Ciana P, Maggi A. Identification of novel loci for the generation of reporter mice. Nucleic Acids Res. 2017 Apr 7;45(6):e37. doi: 10.1093/nar/gkw1142). This reporter animal has been used in experiments aimed at following the progression of inflammation in EAE by means of non-invasive in vivo molecular imaging (Objective 6.ii). Moreover, a full set of new constructs for M2 phenotype imaging (based on the fullArgI-tk-TdTomato-ires-Luc2 construct) has been generated, validated (Objective and used to investigate in vitro the temporal profile of IL4-mediated Arg-1 expression in RAW 264.7 cells primed with LPS, and the effects of endogenous modulators of the inflammatory response (e.g. estrogens) (Objectives 1.iii 3.iv and 6.i). Extensive experiments revealed that these sets of constructs were unsuitable for the generation of the reporter mouse. For this reason, an alternative set of constructs, based on STAT6 response element, were generated, and in vitro validated. The final tests on the specific construct necessary for the generation of the M2 reporter animal model for imaging NIND are currently ongoing (Objective WP2 - Neurogenesis / regeneration in ND-associated NI

Standardised procedures for microglia analysis
Closely collaborating with WP1, WP2 established standardised in vitro culture systems providing microglia of defined activity states. In addition, protocols for the isolation of microglia from adult rodents were developed, and it was shown that the isolated cells retained the activation profile they acquired in vivo after injection of IL-4, and, moreover, that these protocols are also suitable for AD models. A summary of tested protocols is available for INMiND partners on the collaboration platform Basecamp (Objectives 1.i 2.i and 2.ii).

Identification of novel microglia imaging targets
To find additional candidates for in vivo imaging of microglial activation in human patients, publicly available data sets about gene expression changes in M1-/M2-polarised human macrophages, about genes involved in amyloid processing, and the catecholamine and innate immune systems in the AD pathogenesis were analysed. Moreover, a comprehensive literature review of inflammatory mediators in body fluids of AD and MCI patients was done (Brosseron F, Krauthausen M, Kummer M, Heneka MT. Body fluid cytokine levels in mild cognitive impairment and Alzheimer's disease: a comparative overview. Mol Neurobiol. 2014 Oct;50(2):534-44. doi: 10.1007/s12035-014-8657-1). In collaboration between WP1/2/3/5, these in vitro and in vivo systems have been used for validation of novel targets for imaging of neuroinflammation. Several compounds developed by WP3 for in vivo imaging showed promising pharmacological profiles (Objectives 1.ii and 4.iv).
In vivo characterisation of microglial activation and/or neuroregenerative processes
Different NIND models (e.g. AD, PD, HD, Lewy body dementia, epilepsy, stroke, EAE, demyelination, ageing) were analysed in detail. That way, for example, age was identified as an important factor determining microglial activation after demyelination in the cuprizone model. The analysis of cells possibly contributing to CNS regeneration was not limited to microglia. Thus, platelets and pericytes were shown to promote regeneration after demyelination (De La Fuente AG, Lange S, Silva ME, Gonzalez GA, Tempfer H, van Wijngaarden P, Zhao C, Di Canio L, Trost A, Bieler L, Zaunmair P, Rotheneichner P, O'Sullivan A, Couillard-Despres S, Errea O, Mäe MA, Andrae J, He L, Keller A, Bátiz LF, Betsholtz C, Aigner L, Franklin RJM, Rivera FJ. Pericytes Stimulate Oligodendrocyte Progenitor Cell Differentiation during CNS Remyelination. Cell Rep. 2017 Aug 22;20(8):1755-1764. doi: 10.1016/j.celrep.2017.08.007) and infiltrated monocytes to contribute to regeneration after stroke (Miró-Mur F, Pérez-de-Puig I, Ferrer-Ferrer M, Urra X, Justicia C, Chamorro A, Planas AM. Immature monocytes recruited to the ischemic mouse brain differentiate into macrophages with features of alternative activation. Brain Behav Immun. 2016 Mar;53:18-33. doi: 10.1016/j.bbi.2015.08.010) (Objectives 1.i 1.iii 2.ii 4.iv and 4.v).
Analysis of the link between neurodegeneration and neurogenesis
The interplay between these processes was analysed extensively in AD, HD, PD, epilepsy, stroke, and demyelination. In two AD mouse models, prior to plaque deposition, a significant hyperproliferation of DCX+ neural progenitors occurred, however, these cells had a reduced potential to differentiate into neurons (Unger MS, Marschallinger J, Kaindl J, Höfling C, Rossner S, Heneka MT, Van der Linden A, Aigner L. Early Changes in Hippocampal Neurogenesis in Transgenic Mouse Models for Alzheimer's Disease. Mol Neurobiol. 2016 Oct;53(8):5796-806. doi: 10.1007/s12035-016-0018-9). In a transgenic rat model of late onset HD, it was demonstrated that subventricular zone-derived olfactory bulb neurogenesis was reduced and, additionally, a striatal invasion of neuroblasts was observed. An in vivo knock-down of DJ-1, a gene which is known in humans for its loss-of-function mutations which lead to autosomal recessive early-onset PD, impaired adult neurogenesis in the mouse OB. Using a mouse model of mesial temporal lobe epilepsy (MTLE), it was demonstrated that, even in the absence of seizures, overactivation of radial neural stem cells (rNSC) by neuronal hyperexcitation led to accelerated depletion of the rNSC population and premature decline of neurogenesis. Furthermore, seizures induced a dramatic antineurogenic switch of rNSCs which may contribute to hippocampal sclerosis, a pathophysiological feature of MTLE. An in vivo bioluminescence imaging system for non-invasive monitoring of neural stem cells after stroke was developed. For this, conditional LVs (Cre-Flex LVs) encoding Fluc and eGFP were injected into the SVZ of Nestin-Cre transgenic mice to specifically label the endoNSCs. Using this technique, it was shown that nestin+ endoNSCs, originating from the SVZ, responded to stroke injury by increased proliferation, migration towards the infarct region and differentiation into both astrocytes and neurons (P13b). Moreover, it was demonstrated that after stroke an up-regulation of neurogenesis in the healthy hemisphere compensated for the age-dependent decrease of basal neurogenesis. In vivo longitudinal BLI of endoNSCs in the cuprizone model revealed an increased neurogenic potential for the olfactory bulb pathway (P7) (Objectives 2.ii 2.iii 2.iv 4.iii and 6.iv).
Investigation of sex differences
A central goal of INMiND was to investigate the role of sex hormones in neurodegenerative and neuroregenerative processes. Sex-specific and lactation-induced differences were observed in rats in the modulation of adult hippocampal neurogenesis under stress conditions. TGF-β signalling was identified as a highly relevant pathway for the influence of estradiol on neurogenesis by analysing the transcriptome of microglia isolated from adult ovariectomised female mice treated with vehicle or 17-β-estradiol. A protective, anti-inflammatory role of endogenous estrogens was identified only in females in the EAE model (Objective 2.v).
Identification of novel therapeutic strategies
Another central objective of INMiND was to identify possible targets for neuroregenerative therapies. The research towards this goal led to promising results, in particular, to new strategies for the treatment of age-related and Lewy body dementias, AD, stroke, and demyelinating diseases, like MS: Leukotriene receptor signalling was identified as a safe and drugable pathway with a combined mode of action to reduce neuroinflammation and to enhance neurogenesis, particularly in the aged brain. Preclinical in vivo experiments showed that treatment with montelukast, an anti-asthmatic drug acting on leukotriene receptors, reduced microglial activity, enhanced neurogenesis, and restored cognitive function in aged rats (Marschallinger J, Schäffner I, Klein B, Gelfert R, Rivera FJ, Illes S, Grassner L, Janssen M, Rotheneichner P, Schmuckermair C, Coras R, Boccazzi M, Chishty M, Lagler FB, Renic M, Bauer HC, Singewald N, Blümcke I, Bogdahn U, Couillard-Despres S, Lie DC, Abbracchio MP, Aigner L. Structural and functional rejuvenation of the aged brain by an approved anti-asthmatic drug. Nat Commun. 2015 Oct 27;6:8466. doi: 10.1038/ncomms9466). Also in Lewy body dementia, the leukotriene receptor antagonist montelukast restored cognitive deficits, normalised the microglia phenotype and reduced α-synuclein expression. These preclinical results are the basis for future clinical translation of modulating leukotriene receptor signalling for the treatment of dementias, and led to two patent applications in which P3 is an inventor. Pan-PPAR modulation effectively protected APP/PS1 mice from amyloid deposition and cognitive deficits. In comparison to a selective PPARγ-agonist, a novel potent PPAR agonist produced quantitatively superior and qualitatively different therapeutic effects with respect to amyloid plaque burden, insoluble Aβ content and NI at lower exposure rates (Kummer MP, Schwarzenberger R, Sayah-Jeanne S, Dubernet M, Walczak R, Hum DW, Schwartz S, Axt D, Heneka MT. Pan-PPAR modulation effectively protects APP/PS1 mice from amyloid deposition and cognitive deficits. Mol Neurobiol. 2015 Apr;51(2):661-71. doi: 10.1007/s12035-014-8743-4). The consequences of locus coeruleus (LC) degeneration and subsequent noradrenaline (NA) deficiency was assessed in early Alzheimer's disease (AD), and it was shown that acute pharmacological replacement of NA partially restored the observed deficits. CXCR3 promotes plaque formation and behavioural deficits in an AD model. CXCR3-deficient APP/PS1 mice showed reduced plaque burden and Aβ levels, increased microglial uptake of Aβ, microglial morphological activation and reduced plaque association, and reduced concentrations of proinflammatory cytokines. The CX3CL1/CX3CR1 signalling axis was identified as a promising target to prevent microglia-mediated oligodendrocyte death in an in vivo model for MS. In the cuprizone model, it was shown that an increased IL-13 expression in the corpus callosum, either by targeted injection of a lentiviral vector (LV) or by implantation of genetically modified MSCs, reduced demyelination, enhanced survival of oligodendrocytes and modulated inflammation, via M2 polarisation in brain-resident microglia and brain-invading macrophages. CNS-targeted treatment with Oncostatin M, a member of the IL-6 cytokine family, protected against cuprizone-induced demyelination by increasing IL4 expression and inducing a protective microglia phenotype. This study revealed a previously uncharacterised and protective role for Oncostatin M during demyelination, and indicates that OSM is a promising therapeutic candidate to limit CNS damage in demyelinating diseases including MS. In co-culture models, it was shown that the presence of MSCs had an anti-inflammatory effect on astrocytes under pro-inflammatory conditions. However, aging restricted the ability of systemically administered MSCs to promote the generation of oligodendrocytes after a focal demyelinating lesion induced by ethidium bromide injection in the rat cerebellar caudal peduncle. The search for novel neuroprotective strategies led to the identification of chroman-like cyclic prenylflavonids as highly neuroprotective compounds, which moreover promote neuronal differentiation and neurite outgrowth. These results indicate that hop-derived prenylflavonoids might be promising molecules to promote neurogenesis, neuroregeneration and neuroprotection in cases of chronic neurodegenerative diseases or acute CNS lesions (Objectives 1.iii 1.iv 2.iii and 2.iv). WP3 - Synthesis of probes for detection of NIND by PET and SPECT

WP3 focused on the development of new radiotracers for the targets that are of interest for imaging Neuroinflammation (NI) and Neurodegeneration (ND). At the start of the project, we selected seven targets (TSPO, P2X7 receptor, H4 receptor, MMP enzymes, COX2 and MAO-B enzyme, and the CB2 receptor) for which new tracers would be developed (Objective 4.i-ii). Within these tasks, many new tracers have been synthesised and evaluated in a preclinical setting. We have identified highly promising new tracers from those experiments and three of them ([123I]CLINDE, [11C]NE40, [18F]BR35) have been used for the first time in humans, several known tracers like [18F]DPA714, [11C]PBR28, [18F]FEMPA and [18F]GE-180 have been produced for other workpackages and finally three tracers were completely developed within INMIND and are now ready for human application: [18F]deprenyl, [11C]JNJ717 and [11C]SMW139, an excellent achievement in just five years’ time.
With additional funding awarded by the Michael J. Fox foundation, [11C]JNJ717 and [11C]SMW139 will be investigated in a first in humans study in PD patients and age-matched controls to evaluate their use as neuroinflammation PET tracer in PD in a multicentre study involving 7 INMIND partners (P1, P32, P10a, P12a, P13a, P16, P19). Moreover, a grant was obtained from the GSMI fund of Merck Serono for a first in human study with [11C]SMW139 in MS and age-matched controls (P32, collaboration with Prof. dr. H.E. de Vries of the VUmc MS center).
An important achievement of WP3 was the design and synthesis of ligands targeting newly identified biological NI targets (Objectives 3.iii-iv and 4.iii-iv). Together with WP1, WP2, WP4 and WP5 we have selected new targets (receptors, enzymes, Objective 3.iii) that are upregulated in activated microglia, with emphasis on M2 microglia (Objective 3.iv). We have designed new tracers for these targets and the first compounds have been radiolabelled and are currently in preclinical research. This is a major collaborative achievement of INMIND under the lead of WP3.

TSPO. Many new tracers have been developed within the project. Mainly focussed on avoiding the issue with the polymorphism of current TSPO binding tracers and to improve metabolic stability (Objectives 3.i and 4.i). Although the issue with the polymorphism was not solved, new insights were obtained that should lead in the right direction. Improved tracers have been brought forward, like [18F]F-DPA, which performs really well in a longitudinal study in AD mice (P8, P16).

P2X7 receptor. This receptor was postulated to have increased expression in the M2 polarised microglia based on mRNA analyses. However, this hypothesis was rejected, since it was shown by P2 and P32 that the P2X7 receptor is also upregulated in M1 polarised microglia. It might be that PET tracers targeting the P2X7 receptor could be used as general neuroinflammation PET tracers, where the specific signal might be stronger than current neuroinflammation PET tracers, like the TSPO tracers (Objective 4.ii). Many new tracers targeting the P2X7 receptor have been designed and two of them have been made ready for human application. In a MS rat model we have shown that the P2X7 is upregulated at the peak of the disease (M1 related) and declines in the recovery phase (M2 related), thus expression levels seem to be correlated to the pro-inflammatory character of microglia. This was confirmed in post-morten tissue of MS patients (Beaino W, Janssen B, Kooij G, van der Pol SMA, van Het Hof B, van Horssen J, Windhorst AD, de Vries HE. Purinergic receptors P2Y12R and P2X7R: potential targets for PET imaging of microglia phenotypes in multiple sclerosis. J Neuroinflammation. 2017 Dec 22;14(1):259. doi: 10.1186/s12974-017-1034-z).

H4 receptor. The histamine H4 receptor is involved in peripheral inflammation, but is also expressed in the brain, presumably on microglia and on neurons. Two new tracers were designed and synthesised (Objective 4.ii) one of them entered the brain after iv injection, [11C]JNJ7777120. However, this uptake could not be blocked by the pre-treatment with thioperamide, a combined H3 and H4 receptor antagonist. Also in autoradiography studies, binding could be observed, however this binding could not be blocked by several H4 antagonists. Since specific binding of the new H4 receptor PET tracer could not be observed, the development was stopped.

MMP enzymes. The multi-MMP subtype PET tracer [18F]BR351 has been successfully applied in human subjects where increased binding was observed in several neuroinflammation pathologies, reported in WP8 (P1, Objectives 4.ii and 5.i). MMP2 and MMP9 are postulated to be the subtypes that are predominantly involved in inflammation processes, also neuroinflammation. New fluorine-18 labelled tracers targeting MMP2 and MMP9 were designed and evaluated in vivo in rodents. Unfortunately, strong de-fluorination was observed, which leads to high uptake of radioactivity in the bone. This is unacceptable for further use and therefore the development of these tracers has been stopped.

COX2 enzyme. At P4 many COX2 inhibitors have been synthesised, one of the new compounds showed good inhibitory activity for COX2 and was selective over COX1. This new compound was labelled with fluorine-18. The radiochemistry was very challenging, but a proper method was found (Objective 4.ii). This new tracer, [18F]JE186 is currently in preclinical evaluation in rodents. First results have shown that [18F]JE186 does not enter the brain after iv injection.

MAO-B enzyme. Several new fluorine-18 labelled tracers have been synthesised and evaluated in animals (Objective 4.ii). All these tracers were based on the same lead: deprenyl, which has been labelled successfully with carbon-11 and applied as MAO-B PET tracer. However, fluorine-18 with its half-life of 110 minutes has some clear advantages compared carbon-11. From the series of new tracers, [18F]deprenyl was outperforming the others and this tracer is ready for human applications. Since [18F]deprenyl shows rather strong metabolism, its deuterated isotopologue was also synthesised and evaluated in animals. Indeed, [18F]deprenyl-D2 was more stable in vivo (Nag S, Fazio P, Lehmann L, Kettschau G, Heinrich T, Thiele A, Svedberg M, Amini N, Leesch S, Catafau AM, Hannestad J, Varrone A, Halldin C. In Vivo and In Vitro Characterization of a Novel MAO-B Inhibitor Radioligand, 18F-Labeled Deuterated Fluorodeprenyl. J Nucl Med. 2016 Feb;57(2):315-20. doi: 10.2967/jnumed.115.161083).

CB2 receptor. In the early periods of INMIND, [11C]NE40 was made available for validation in humans (Objective 5.i) however the results were negative: no specific signal was observed. New tracers were designed and evaluated in vivo in rodents (Objective 4.ii). The results do not yet justify a translation to human subjects, however the pre-clinical work is continued.

New targets. In close collaboration with WP1 (P2), WP2 (P3), WP4 (P5) and WP5 (P6), the scientists of WP3 have selected new targets (receptors, enzymes) that might be viable as targets to image the M2 polarised microglia, if appropriate PET tracers would be available (Objectives 3.iii-iv and 4.iii). Based on microarray data 15 genes were selected that showed increased activity in M2 polarisation. These genes encoded for receptors or enzymes and based on a thorough literature search. The P2RY12, P2RY14 and CGRP receptors as well as the enzyme QPCT were identified as new targets for neuroinflammation imaging (Villa et al, Identification of new molecular targets for PET imaging of microglial anti-inflammatory phenotype. Submitted 2018). For QPCT some tracers were already investigated, however showed no brain uptake, therefor this enzyme was discarded. For CGRP, a PET tracer was already known, however [11C]MK4232 was never investigated in neuroinflammation. Unfortunately, no specific binding was observed in a neuroinflammation model. For both P2RY12 and P2RY14, their role in the CNS were unexplored and new tracers were designed and synthesised to explore these receptors as imaging targets. Currently no successful tracers have been synthesised, but the work will continue at P32 with a grant of the Michael J Fox Foundation (P2RY12) and an institutional grant and at P8 in an institutional PhD program (P2RY12 and P2RY14). The P2RY12 receptor has been characterised as a viable imaging target through qPCR studies in several neuroinflammation models and in post-mortem human brain. It was shown that the P2RY12 expression was increased in the recovery phase of the EAE model, typically at 7-11 days after onset of the disease, related to the anti-inflammatory status of microglia (Beaino W, Janssen B, Kooij G, van der Pol SMA, van Het Hof B, van Horssen J, Windhorst AD, de Vries HE. Purinergic receptors P2Y12R and P2X7R: potential targets for PET imaging of microglia phenotypes in multiple sclerosis. J Neuroinflammation. 2017 Dec 22;14(1):259. doi: 10.1186/s12974-017-1034-z). In addition, CSF1R and GPR-84 have been investigated as new targets, but the tracers have not shown specific binding yet. WP4 - Probes for NIND detection by MRI and OI

Several new imaging probes and related preclinical procedures for the in vivo visualisation of neuroinflammation processes have been developed.
The work performed primarily referred to Objective 3, and was organised in two main tasks: one devoted to design procedures for the in vivo assessment of the recruitment of BMD cells by the inflammation sites, and the other one focused on the search for probes for targeting MRI-detectable neuroinflammation biomarkers.
The first activity was carried out by developing innovative probes and procedures for in vitro and in vivo labelling of the phagocytic cells of the immune system.
Among the systems investigated, it is worth mentioning the use of Glucan Particles (GPs) as carriers for imaging agents. Such micro-sized objects have been obtained from a harsh procedure through which Saccaromyces Cerevisiae (i.e. baker’s yeast) have been exhaustively depleted from their lipids, proteins and nuclear material. The remaining hard shell, composed by β-1,3-D-glucan, can be filled up by lipophilic nano/micromaterials (paramagnetic Gd-based complexes, fluorescent dyes, fluorinated compounds) that are formed directly in situ inside the cavity of the hollow particles, using a procedure developed during the project.
GPs are promptly and avidly phagocytised by immune cells through the specific recognition of β-1,3-D-glucan by dectin-1 (D1) and complement receptor 3 (CR3), both expressed on the surface of immune cells.
It has been demonstrated that GPs suitably loaded with imaging agents can provide information on the recruitment of BMD cells either by in vitro (systemic injection of labelled immune cells) or in vivo (systemic injection GPs, successively taken up by circulating immune cells) labelling. One interesting advantage in using these particles for in vivo labelling is represented by their large size that prevent blood extravasation of the free agent, thus making more accurate the assessment of BMD cell recruitment. GPs loaded with MRI and NIRF dyes have been validated on several inflammation mouse models including CCl4-induced liver hepatitis and collagen-induced rheumatoid arthritis (Figueiredo S, Cutrin JC, Rizzitelli S, De Luca E, Moreira JN, Geraldes CF, Aime S, Terreno E. MRI tracking of macrophages labeled with glucan particles entrapping a water insoluble paramagnetic Gd-based agent. Mol Imaging Biol. 2013 Jun;15(3):307-15. doi: 10.1007/s11307-012-0603-x; Garello F, Arena F, Cutrin JC, Esposito G, D'Angeli L , Cesano F, Filippi M, Figueiredo S, Terreno E. Glucan particles loaded with a NIRF agent for imaging monocytes/macrophages recruitment in a mouse model of rheumatoid arthritis. RSC Adv. 2015;5:34078-34087. doi: 10.1039/C5RA00720H) (Objective 4.iv).
Another imaging system that has been tested and successfully validated (brain LPS-induced inflammation and CPZ-EAE models on mice) for in vivo labelling of circulating immune cells is represented by fluorinated nanoemulsions (PFCE-NE) composed by perfluoro-15-crown-5-ether for 19F-MRI detection.
These particles displayed a high uptake efficiency in vitro towards macrophages. Moreover, a tendency of PFCE-NE to polarise cells towards the M2 phenotype was remarked (relevant to Objective 3.iv). When the particles were i.v. injected in healthy animals, a blood clearance time of three days was estimated; however already after three hours the 19F-MRI signal perfectly co-localized with the peripheral blood mononuclear cells.
Concerning the development of improved procedures for in vitro cell labelling, it is worth mentioning a new method based on hypo-osmotic shock that allowed the loading of a large amount of the clinically approved MRI agent gadoteridol into any kind of cell types (including immune cells) without affecting viability and functional status of the systems. Then, the so-labelled cells can be detected by T1w-MRI, thus producing bright spots on the image, an advantageous characteristic over the signal void response obtained with T2 agents (Objective 3.ii).
Further results obtained within this line dealt with the in vivo detection by photoacoustic imaging of primary mouse monocytes pre-labelled in vitro with the clinically approved NIRF agent Indocyanine Green (ICG).
The aim of targeting inflammation biomarkers with new MRI probes has been primarily faced by searching for vascular targets strongly linked to the onset of inflammation processes. The choice has fallen to VCAM-1 receptors, which are part of the pattern of adhesion molecules that are involved in the mechanism by which immune cells are recruited by the inflammation site. A peptide mimicking the ligand for VCAM-1 receptors (VLA-4), obtained by phage display, was reported in literature. The peptide was synthesised and conjugated to a phospholipid in order to make it incorporable in lipid-based nanosized micelles. The incorporation of a suitable amphiphilic paramagnetic Gd-based MRI agent allowed the MRI detection of the micellar system. The nanoagent displayed a good MRI-detectability because of the presence of thousands of imaging units exposed on the nanoparticle surface.
The targeting agent was successfully validated on acute model of both peripheral and brain LPS-induced inflammation mouse models, using micelles incorporating an untargeted scrambled version of the VLA-4-mimicking peptide (Pagoto A, Stefania R, Garello F, Arena F, Digilio G, Aime S, Terreno E. Paramagnetic Phospholipid-Based Micelles Targeting VCAM-1 Receptors for MRI Visualization of Inflammation. Bioconjug Chem. 2016 Aug 17;27(8):1921-30. doi: 10.1021/acs.bioconjchem.6b00308; Garello F, Pagoto A, Arena F, Buffo A, Blasi F, Alberti D, Terreno E. MRI visualization of neuroinflammation using VCAM-1 targeted paramagnetic micelles. Nanomedicine. 2017 Nov 8. pii: S1549-9634(17)30181-8. doi: 10.1016/j.nano.2017.10.002) (Objective 4.iv).
Besides MRI, even a NIRF cyanine-based dye targeting VCAM-1 (Cy5.5-VCAM-1) has been synthesised.
In addition to VCAM-1 targeting, superparamagnetic iron oxide microparticles conjugated with an antibody against ICAM-1 receptors were also developed and tested on a mouse model of brain ischemia.
Work has been also performed for improving the sensitivity detection of MRI agents (Objective 3.ii). To this purpose, several novel amphiphilic ligands for Gd(III) ion have been synthesised and incorporated in conventional (micelles, liposomes) or new nanoparticles (e.g. dendrimersomes) (Filippi M, Martinelli J, Mulas G, Ferraretto M, Teirlinck E, Botta M, Tei L, Terreno E. Dendrimersomes: a new vesicular nano-platform for MR-molecular imaging applications. Chem Commun (Camb). 2014 Apr 4;50(26):3453-6. doi: 10.1039/c3cc49584a).
The much more challenging objective to target intracerebral inflammation biomarkers with MRI probes has been pursued preparing a dual-targeted MRI-detectable micellar nanosystem. One target was represented by phospholipid-conjugated peptide mimicking ApoE in order to confer the nanosystem the ability to cross the blood-brain barrier (BBB) through the low-density lipoprotein receptors (LDLr). The second target, again a phospholipid-based peptide, was chosen to recognise the extracellular inflammation marker TNF-α. The nanosystem was prepared and it will be tested in future studies.
During the project, a MRI agent conjugated with a structural motif targeting TSPO has been synthesised and characterised in vitro.
The synthesis of amphiphilic blocks to be incorporated in lipid-based nanoparticles is a very versatile approach, and it has been also exploited to synthesise new ligands for Zr89 PET detection to be incorporated in lipid-based nanosystems with some potential in dual PET-MRI detection of NIND. WP5 - Validation of new probes for NIND in animal models

WP5 aimed at testing in animal models imaging probes for NI in ND diseases as a step for preclinical validation prior to human studies. To achieve this objective a number of animal models of NI and ND were developed. Intense work between WP1, WP2, WP3 and WP5 was carried out with the purpose of identifying molecular targets for ND that could be used for imaging and therapeutics. Successful translation of potential molecular targets to imaging probes suitable to investigate the human brain imposes severe constrains that had to be taken into account. First, the adequacy of the choice of biological targets must be validated in animal models of NI/ND by showing sufficient upregulation. Second, the biological target must be suitable for binding to an exogenous ligand that must be labelled with a PET radioactive isotope. This implies that the target is accessible, drugs able to bind this target are known or can be developed, and the radiochemistry for PET radiotracer synthesis is feasible. Third, the radiotracer will have to reach the brain after systemic administration. Fourth, the radiotracer must be specific for the target and has to provide an appropriate signal-to-noise ratio. Finally, we aimed at identifying targets that were relevant to humans and therefore the findings in animals require validation in human material. This last step is critical to avoid probe development efforts not relevant for humans. To fulfill this requirement, targets identified in rodents have been validated in human cells and post-mortem human tissue. Solving the problem taking into account this strong constrain needs a concerted approach by interdisciplinary scientists. This is a main task that we have undertaken in a transversal manner across WPs in this project. While WP1/2 has focused on the identification of putative markers of microglia activation mostly using cell cultures challenged with different compounds turning microglia into a classical (M1) or alternative (M2) phenotype and WP3 has concentrated in developing new tracers for markers of microglia activation, WP5 has worked on the study of markers of microglia activation in animal models of human neurological diseases and has tested a range or PET tracers to image NI in several animal models of brain diseases.
Microglia cells can become activated into different phenotypes depending on the specific environment. Microglia, as well as macrophages, can be activated with pro-inflammatory stimuli to a classical pro-inflammatory or M1 phenotype, whereas exposure to certain cytokines, such as IL4, induces an alternative activation or M2 phenotype that is associated to repair functions. Between these poles, many different activation states can arise depending on the specific cytokines used to activate microglia in vitro. Beyond the typical M1/M2 phenotypes, different activation states can occur in vivo under neurodegenerative conditions and relevant markers that can serve the purpose of targets for imaging probes must be identified since the brain environment changes in different directions depending on the type of disease and with disease evolution. We addressed Objective 5 of this project by modeling a range of neuroinflammatory and neuropathological conditions in collaboration with WP6. We aimed at identifying the type(s) of microglial activation and neuroinflammatory hallmarks in the various disease conditions and disease stages. In the first instance, we achieved experimental conditions inducing M1 microglial activation by administration of bacterial lipopolysacharide (LPS) into the brain, and conditions inducing M2 activation by intracerebral administration of the prototypic M2 cytokine IL4. These experimental models were suitable to induce the expression of M1 and M2 markers in vivo for the purpose of testing the binding of different PET tracers to the brain under these artificial conditions. We studied the phenotype of microglia under different states of activation and identified the expression of activation markers like certain prostaglandin E receptors and K+ channels. The second step was to investigate in animal models of brain diseases the phenotype of microglial activation, the contribution of infiltrating leukocytes, and the binding of the various PET probes. Animal models for brain diseases have been set including transgenic models of Alzheimer’s disease, models of Huntington’s and Parkinson’s diseases, stroke models, and demyelinization models to study multiple sclerosis. The latter tasks were proven to be far more complex due to mixed microglia/macrophage phenotypes under disease conditions and the dynamic evolution of the inflammatory responses with disease progression. However, we achieved the identification of the longitudinal expression of several M1 and M2 markers in various disease models. Although this marker expression does not guarantee that in vivo microglial phenotypes conform to the typical M1/M2 phenotypes, it informs on main traits that microglial cells acquire during disease progression and identifies the upregulation of potential target molecules that can be used for the development of imaging probes and test the binding of available tracers to specific microglia phenotypes in vivo.
The main PET tracer available to image neuroinflammation at the beginning of this project was (R)-[11C]PK11195, a tracer that binds to TSPO. To achieve Objective 5i.a several new TSPO tracers were tested in various experimental models of brain diseases and neuroinflammatory models. Results have been achieved with the TSPO tracers [18F]DPA-714, [18F]GE-180, [18F]VC701, and [11C]PBR28 (e.g. Takkinen JS, López-Picón FR, Al Majidi R, Eskola O, Krzyczmonik A, Keller T, Löyttyniemi E, Solin O, Rinne JO, Haaparanta-Solin M.Brain energy metabolism and neuroinflammation in ageing APP/PS1-21 mice using longitudinal 18F-FDG and 18F-DPA-714 PET imaging. J Cereb Blood Flow Metab. 2017 Aug;37(8):2870-2882. doi: 10.1177/0271678X16677990 and Dickens AM, Vainio S, Marjamäki P, Johansson J, Lehtiniemi P, Rokka J, Rinne J, Solin O, Haaparanta-Solin M, Jones PA, Trigg W, Anthony DC, Airas L. Detection of microglial activation in an acute model of neuroinflammation using PET and radiotracers 11C-(R)-PK11195 and 18F-GE-180. J Nucl Med. 2014 Mar;55(3):466-72. doi: 10.2967/jnumed.113.125625). In addition, novel TSPO tracers with improved performance were developed, like e.g. [18F]F-DPA. Also, studies have been carried out with the SPECT TSPO tracer [123I]CLINDE. The TPSO ligand DPA-714 labelled with 18F was the tracer most widely used across laboratories for the advantages it offers compared to (R)-[11C]PK11195, due to the properties of the compound and the fact that 18F has a considerably longer half-live than 11C facilitating experimental studies. The performance of [18F]DPA-714 was compared to the classic TSPO tracer (R)-[11C]PK11195. The former showed comparatively better signal-to-noise ratio and better performance than the latter. The adequacy of this tracer over (R)-[11C]PK11195 to image neuroinflammation in the experimental settings has been demonstrated. Regarding new tracers targeting molecules other than TSPO, the newly developed radioligand [11C]A-836339 for the cannabinoid subtype 2 receptor (CB2-R) was investigated in vivo in an experimental model of neuroinflammation but the results did not show increased PET binding of the radiotracers. Although increases in CB2-R were reported in the literature in mouse AD models, we showed decreased in vivo availability of the CB2-R [11C]NE4 tracer in AD patients, with no relationship to Aβ plaques. Given that the inflammatory reaction includes increased expression and activation of matrix metalloproteinases (MMPs), we studied a MMP radiotracer, [18F]BR-351, which shows increased binding but does not show precisely the same dynamics than the microglia reaction (Zinnhardt B, Viel T, Wachsmuth L, Vrachimis A, Wagner S, Breyholz HJ, Faust A, Hermann S, Kopka K, Faber C, Dollé F, Pappata S, Planas AM, Tavitian B, Schäfers M, Sorokin LM, Kuhlmann MT, Jacobs AH. Multimodal imaging reveals temporal and spatial microglia and matrix metalloproteinase activity after experimental stroke. J Cereb Blood Flow Metab. 2015 Nov;35(11):1711-21. doi: 10.1038/jcbfm.2015.149). Initial studies with the new COX-2 ligand [11C]VA426 have shown that it successfully binds to inflammed tissues in the periphery but it was not good to show brain inflammation. Purinergic receptors were also identified as putative targets for imaging NI. We performed studies with [11C]JNJ717, which is a novel tracer for the human P2X7 receptor that can be a potential target for imaging neuroinflammation, and also developed tracers for P2YR12. However, the current tracers must be improved to facilitate crossing the blood-brain barrier (BBB) and to provide an adequate signal for PET imaging. Beyond tracers aimed to assess the microglial activation, we also imaged the astroglial reaction with the specific tracer [11C]-deuterium-L-deprenyl in Tg murine AD models. Overall, we have tested in animal models a range of novel PET tracers that have open the scope for imaging neuroinflammation with PET (see Annex 3.1.1; Table 2, Table 3, Table 4 and Table 5).
We related the binding of PET tracers in the brain under the experimental disease conditions to the molecular and neuropathological hallmarks of the disease conditions (Gargiulo S, Anzilotti S, Coda AR, Gramanzini M, Greco A, Panico M, Vinciguerra A, Zannetti A, Vicidomini C, Dollé F, Pignataro G, Quarantelli M, Annunziato L, Brunetti A, Salvatore M, Pappatà S. Imaging of brain TSPO expression in a mouse model of amyotrophic lateral sclerosis with 18F-DPA-714 and micro-PET/CT. Eur J Nucl Med Mol Imaging. 2016 Jul;43(7):1348-59. doi: 10.1007/s00259-016-3311-y) to achieve Objective 5i.b. Tracer uptake was studied in relation to BBB breakdown as assessed with MRI to ensure that binding truly reflects microglial activation. [18F]DPA-714 uptake does not seem to be related to BBB breakdown. This result was validated in autoradiography and immunohistochemical studies in post-mortem tissue examining the cellular expression of TSPO using markers of microglia cells and macrophages. The cellular correlates of imaging results with TSPO PET tracers show that the PET signal corresponds to strong TSPO expression in microglia. The glial reactions as imaged by PET have been compared in vivo with disease hallmarks, e.g. amyloid-β deposition also imaged by PET with [11C]PIB, and with tracers potentially suitable to mark ND (e.g. the sigma-1 receptor ligand [18F]IAM6067), or by SPECT using ligands of the dopamine transporter (DAT) in PD models, and with immunohistochemical studies in the post-mortem tissue. We showed that [11C]PK11195 binding increased in mice deficient in TREM2, a gene with mutations that confer risk for AD and susceptibility for PD. Imaging findings have been related to findings in histological and immunohistochemical studies in the post-mortem tissue. We compared neuroinflammation, as assessed by PET imaging, with features of neurodegenerations by combining in vivo and ex vivo studies in various disease conditions. Besides PET, we used other imaging modalities, like in vivo bioluminiscene imaging, to image neuropathological hallmarks, as for instance alpha-synuclein oligomerization in the mouse brain. Also, MRI, including MR spectroscopy and resting state fMRI, allowed the assessment of metabolic alterations and functional connectivity in mouse models of ND diseases (Orije J, Kara F, Guglielmetti C, Praet J, Van der Linden A, Ponsaerts P, Verhoye M. Longitudinal monitoring of metabolic alterations in cuprizone mouse model of multiple sclerosis using 1H-magnetic resonance spectroscopy. NeuroImage. 2015 Jul 1;114:128-35. doi: 10.1016/j.neuroimage.2015.04.012).
To improve the knowledge on how NI processes influence the course of ND (Objective 5.ii) we investigated the effects of anti-inflammatory treatments like Montelukast, which is a leukotriene receptor antagonist and a high affinity agonist of α7nAChRs, amongst other strategies. Histological (immunohistochemistry, immunofluorescence, histology) and molecular (gene expression) correlates of imaging findings have been established (Objective 5.ia) as well as their response to treatment. Imaging studies at early time points during disease development in genetic or chemical models of diseases have been carried out together with their molecular/histological correlates correlates (Belloli S, Pannese M, Buonsanti C, Maiorino C, Di Grigoli G, Carpinelli A, Monterisi C, Moresco RM, Panina-Bordignon P. Early upregulation of 18-kDa translocator protein in response to acute neurodegenerative damage in TREM2-deficient mice. Neurobiol Aging. 2017 May;53:159-168. doi: 10.1016/j.neurobiolaging.2017.01.010) (Objective 5.iib). Importantly, we carried out longitudinal imaging studies (Objective 5ii.c) for instance in Tg AD rodent models or MS models, where the inflammatory response has been characterised by examining gene expression of pro- and anti-inflammatory markers. We investigated how different interventions (Objective 5ii.d) e.g. by inducing additional inflammation (infection) or by strategies targeting inflammation (e.g. viral-induced over-expression of IL13, anti-inflammatory drug treatments) affected the functional neurological outcome in various disease models (e.g. Rapic S, Backes H, Viel T, Kummer MP, Monfared P, Neumaier B, Vollmar S, Hoehn M, Van der Linden A, Heneka MT, Jacobs AH. Imaging microglial activation and glucose consumption in a mouse model of Alzheimer's disease. Neurobiol Aging. 2013 Jan;34(1):351-4. doi: 10.1016/j.neurobiolaging.2012.04.016). We demonstrated beneficial effects of anti-inflammatory treatments in MS and PD models. For new imaging-guided therapies (Objectives 5.iv and 6), advances have been made for instance by showing that the PET tracer [18F]DPA-714 is sensitive to treatments that reduce the inflammatory reaction and therefore it seems to be suitable to monitor the effects of treatments (see also WP6 summary below). Furthermore, we showed that established therapies, like fingolimod for MS reduced brain binding of the TSPO ligand [18]F-GE180 in a mouse model of MS, and the antitumor drug erufosine reduced [18F]DPA-714 binding in a rat 9L intracranial glioma model.
The extensive studies on neuroinflammation in the various diseases models have provided valuable information on how imaging findings correlate with the disease pathophysiology (Objective 5.iii). The brain inflammatory response under disease conditions is complex and evolves with disease progression. Comparative studies in various Tg AD murine models has shown that the relationship between amyloid-β deposition and microgliosis is to some extent model-dependent and region-dependent, suggesting that the aggressiveness of the disease might confer different degrees of Aβ deposition and inflammation. For instance, in the APP/PS1-21 mouse model of AD, cortical expression of TSPO, TNF-α and IL1β mRNA increased progressively from 3 months of age to 9 months. In addition, the expression of the M2 marker arginase-1 mRNA is also increased at 9 months, indicating the co-existence of M1 and M2 markers in this AD model. In the TgF344-AD rat model that was studied from 6 to 18 months of age, IL1β mRNA, rather than TSPO and TNF-α mRNAs, increased in the cortex at 18 months, whereas arginase-1 mRNA did not appear to increase in the cortex in this model at the time points studied. Notably, the expression of the microglial marker P2RY12 increased in both AD models. P2RY12 also increased in PD and stroke models. Efforts in collaborative studies between WP1/2/3/5 identified P2RY12 as a candidate for imaging microglial activation by PET and WP3 developed novel radiotracers targeting P2RY12. Further studies investigating the adequacy of targeting P2RY12 to image microglial activation in ND are guaranteed. WP6 - In vivo monitoring of NI, ND and endoNSC interactions in animal models and of the effect of M1/M2 conversion modulators in animal models

The work performed in WP6 focussed on the characterisation of animal models for ND, work performed in close collaboration with WP5. The involvement of NI in ND has been investigated in animal models of MS, PD, AD, ALS, and HD, using various in vivo imaging tools. The main focus has been on studying dynamics of microglial activation and related NI and ND, but also neuroregenerative processes, in different NIND animal models (Objectives 2 and 3). NI imaging findings acquired in various known and newly developed NIND animal models were correlated with histopathological and imaging findings related to other disease-specific hallmarks (amyloidosis, astrogliosis, BBB permeability, cell density, functional and structural connectivity, etc) (Objectives 3 and 4). Several neuroprotective and -regenerative strategies have been evaluated and mainly PET and/or MRI have been used as imaging read-outs to monitor treatment response in animal models (Objective 6).
More specifically, available PET tracers to image NI in different animal models of ND have been used (reviewed e.g. in: Gargiulo S, Coda AR, Panico M, Gramanzini M, Moresco RM, Chalon S, Pappatà S. Molecular imaging of neuroinflammation in preclinical rodent models using positron emission tomography. Q J Nucl Med Mol Imaging 2017;61:60-75. doi: 10.23736/S1824-4785.16.02948-4) and promising results have been achieved with the known TSPO tracers [18F]DPA-714, [18F]GE-180, [18F]VC70, and [11C]PBR28. In addition, newly developed PET tracers targeting TSPO have been tested and compared with the known tracers (e.g. [18F]F-DPA) (Objectives 3 and 4.i). [11C]-deuterium-L-deprenyl ([11C]DED) has been used to asses astrocytosis and [18F]MK-9470, [18F]JNJ42259152 as well as [18F]FMPEP-d2 to image type-1 cannabinoid receptors (Objectives 3 and 4.ii-v). [123I]Ioflupane SPECT has been used to determine nigrostriatal degeneration. [18F]-flutemetamol employed for imaging Aβ depositions in AD and a novel tau tracer ([18F]-S-THK5117) has been tested as well (Objectives 4.ii-v). In the context of being able to monitor ND non-invasively in vivo, a new radiotracer ([18F]IAM6067) for sigma-1 receptor was developed (Objectives 4.ii-v). These radioligands have proven useful in various NIND animal models for characterisation as well as treatment monitoring.
The activation of microglia and/or macrophages has been analysed in various in vivo models of MS (EAE-PLP, CPZ, EAE, and CPZ-EAE models) (Objectives 3 and 4.i). Microglial reactivity has been studied during de- and remyelination using different markers for immunohistochemistry (Objectives 3 and 4.i). The expression of TSPO has been compared with the distribution of markers for M1- and M2-polarised microglial activation. Moreover, the interaction between ND, NI and neuroregeneration has been investigated in vivo, using different MRI techniques, localized proton magnetic resonance spectroscopy (1H-MRS) and diffusion kurtosis imaging (DKI) (Objectives 4.iv-v). NI and macrophage activation could be visualised successfully by hyperpolarized 13C-MRSI (Objectives 4.iv-v). The imaging of macrophage infiltration into the brain by MRI after labelling with USPIOs and other types of MR detectable nanoparticles has been optimised (Objectives 4.iv-v). The impact of NI on neuroregeneration has been reported in the CPZ model (Objectives 2.iii-iv).
In PD models, ND has been studied using [123I]Ioflupane-SPECT and [18F]FECT PET (Objectives 4.ii-v). Microglia activation has been assessed by [18F]DPA-714 PET (Objectives 3 and 4.i) and a good correlation between in vivo and IHC results was observed. ND (progressive dopaminergic dysfunction) and NI (microglia activation) have been studied in known and newly developed robust models of PD (e.g. injection of adeno-associated viral vectors (rAAV2/7) encoding α-synuclein). A possible role for the metabotropic glutamate receptor type 5 (mGluR5) in both PD and LID pathology (6-OHDA rat model) has been confirmed using [18F]FPEB PET and 1H-MRS (Objectives 4.ii-iv). Non-invasive visualisation of endoNSC proliferation and migration (neuroregeneration) has been studied employing BLI, however no significant changes could be picked up (Objectives 2.iii-iv). Additionally, it has been shown that degeneration in the nigrostriatal system leads to acute microgliosis, without compromising neuroblast generation in the SVZ (Objectives 2.iii-iv and 4.ii-v).
Analyses of NI and ND in AD models progressed significantly. During the first and second year, NI was studied using [18F]DPA-714 PET (Objectives 3 and 4.i). Moreover, [11C]DED was used to visualise astrocytosis indicating the relevance of NI in AD (Objectives 3 and 4.ii-v). A longitudinal in vivo PET study with [18F]FDG and [18F]DPA-714 demonstrated decreased energy metabolism and increased inflammation in the brains of APP/PS1-21 mice compared to WT mice (Takkinen JS, López-Picón FR, Al Majidi R, Eskola O, Krzyczmonik A, Keller T, Löyttyniemi E, Solin O, Rinne JO, Haaparanta-Solin M. Brain energy metabolism and neuroinflammation in ageing APP/PS1-21 mice using longitudinal 18F-FDG and 18F-DPA-714 PET imaging. J Cereb Blood Flow Metab. 2017 Aug;37(8):2870-2882. doi: 10.1177/0271678X16677990). In collaboration with WP2, adult hippocampal neurogenesis has been examined in two different AD mouse models using immunohistochemical fluorescence staining (IHC) and the rate of neurogenesis was quantitatively analysed (Unger MS, Marschallinger J, Kaindl J, Höfling C, Rossner S, Heneka MT, Van der Linden A, Aigner L. Early Changes in Hippocampal Neurogenesis in Transgenic Mouse Models for Alzheimer's Disease. Mol Neurobiol. 2016 Oct;53(8):5796-806. doi: 10.1007/s12035-016-0018-9.). In addition, evidence has been found that astrocytosis occurs early, preceding Aβ plaque deposition. MRI techniques have been used to study microstructural integrity and functional connectivity as read-outs for ND and synaptic integrity (Objectives 4.iv-vi). rsfMRI revealed impaired functional connectivity in several cortical regions with disease progression. Early stage functional connectivity deficits could be detected prior to amyloid plaque deposition and were related to soluble amyloid content (Objectives 4.iv-vi) (Shah D, Praet J, Latif-Hernandez A, Höfling C, Anckaerts C, Bard F, Morawski M, Detrez JR, Prinsen E, Villa A, De Vos WH, Maggi A, D’Hooge R, Balschun D, Rossner S, Verhoye M, Van der Linden A. Early pathological amyloid causes hypersynchrony of BOLD resting-state networks in transgenic mice and provides an early therapeutic window before amyloid plaque deposition. Alzheimers Dement. 2016 Sep;12(9):964-76. doi: 10.1016/j.jalz.2016.03.010). Additionally, it has been shown that reproductive axis hormones and their deregulation lead to changes in brain functional networks during healthy, accelerated and pathological aging as well as hypothalamic inflammation in an AD mouse model (Objectives 2.v and 4.iv-vi). A new Tg model of AD (TgF344) was distributed to 8 partners (P4, P6, P7, P8, P13, P16, P20, P21) by P10b. These labs subsequently demonstrated rsfMRI alterations in brain regions affected by AD pathology and consistent with previous observations in AD patients (Objectives 4.iv-vi). NI was also demonstrated to be increased in TgF344-AD rats vs WT by in vivo PET with [18F]DPA-714 (P10b) and ex vivo autoradiography in male and female rats by P4 (Objectives 4.iv-vi). Tau pathology was confirmed in vivo (P16) and ex vivo (external collaboration with UCL) in this model (Objectives 4.iv-vi). Furthermore, P10b in collaborations with P3, P4 and P6 demonstrated that recurrent peripheral inflammation had a negative impact on AD pathology and behaviour by accelerating NI.
Studies using ALS models have been successfully evaluating NI using [18F]DPA-714 PET as well as MRI (Objectives 3 and 4.ii-v). 1H-MRS has been used to detect (pre-)symptomatic changes in several brain metabolites (Glx, GABA, creatine, and NAA) in the SOD1G93A model of ALS (Objectives 4.iv-v). In addition, [11C]JNJ717, a newly developed radioligand targeting the P2X7 receptor, has been applied to characterise this model (Objectives 4.ii and 4.iv).
Finally, animal model studies have been performed focussing on HD. Microglial activation markers (e.g. TSPO) have been assessed in this animal model (Objectives 3 and 4.i). Early regional dysfunctions in endocannabinoid and phosphodiesterase 10A signalling have been found, which may play a role in the progression of the disease (Objectives 4.ii-v). Moreover, the involvement of endoNCS has been assessed using MRI and immunohistochemistry (Objectives 2.iii-iv and 4.ii-v). A reduction in SVZ-derived olfactory bulb neurogenesis, accompanied by striatal invasion of neuroblasts has been observed in a transgenic rat model of late onset HD (TgHD rats) (Objectives 2.iii-iv and 4.ii-v). Studies on the migration of neural progenitor cells and the use of diffusion kurtosis imaging for quantifying the alterations in the brain in the YAC128 mouse model for HD have been successfully completed (Objectives 2.iii-iv and 4.ii-v). In addition, alterations in the metabotropic glutamate subtype 5 receptors (mGluR5) in relation to behavioural changes has been investigated in a rat model of HD, using the PET tracer [18F]FPEB (Objectives 4.ii-iv).
One important focus of WP6 has been to study the effect of potential treatment strategies by using known and newly developed probes and imaging modalities (Objective 6). Several therapeutic studies were initiated. Neuroprotective and -regenerative strategies have been evaluated using PET, SPECT and/or MRI as imaging read-outs to assess the effect of therapeutics and molecular modulators of M1/M2 conversion on NI, ND and endoNSC migration/multiplication in the different NIND animal models.
The targeted expression of IL13 (by gene therapy or intracerebral transplantation of IL13-expressing mesenchymal stem cells) led to reduced microgliosis and demyelination, which has been visualised using in vivo MRI (T2-weighted MRI, MTC imaging), gene expression analysis and (immuno)histochemical methods in the CPZ-model of MS (Objectives 6.i 6.iii and 6.iv) (Guglielmetti C, Le Blon D, Santermans E, Salas-Perdomo A, Daans J, De Vocht N, Shah D, Hoornaert C, Praet J, Peerlings J, Kara F, Bigot C, Mai Z, Goossens H, Hens N, Hendrix S, Verhoye M, Planas AM, Berneman Z, van der Linden A, Ponsaerts P. Interleukin-13 immune gene therapy prevents CNS inflammation and demyelination via alternative activation of microglia and macrophages. Glia. 2016 Dec;64(12):2181-2200. doi: 10.1002/glia.23053).
In the chronic focal DTH-EAE lesion rat model (MS), FTY720 (sphingosine-1-phosphate receptor modulator - Fingolimod) reduced the microglial activation surrounding the chronic DTH-EAE lesion, and this reduction can be visualised using PET imaging (Objectives 6.i 6.iii and 6.iv). Using the TSPO PET tracer [18F]DPA-714, a pilot study of the effects of SKI-II, a selective non-lipid inhibitor of sphingosine kinase, showed a significantly higher clinical score in the treated vs. placebo group, though the PET signal indicated no difference in labelling in a EAE-PLP mouse model of MS (Objectives 6.i 6.iii and 6.iv).
The TSPO tracer [18F]GE-180 has been used to monitor immunomodulatory treatment with (i) anti-VLA-4 and (ii) BG-12 in focal MS mouse models (Objectives 6.i 6.iii and 6.iv).
The same tracer has been used to show a decreasing trend in several regions of interest by dynamic PET/CT after treatment with the monoacylglycerol lipase inhibitor JZL184 in APPSwe/PSEN1dE9 mice (AD) (Objectives 6.i 6.iii and 6.iv).
Treatment with a leukotriene receptor inhibitor in PD mice has resulted in improved learning and memory. In an α-synuclein-based rat model of PD, treatment with the immunophilin ligand FK506 (Tacrolimus) significantly reduced NI and dopaminergic ND (Objective 6.iv). The effect of estrogen deprivation and replacement after induction of ND has been investigated in a mouse model of PD. Estrogen deprivation had a deleterious impact, which was counteracted by its replacement (Objective 6.iv).
Early stage anti-Aβ immunotherapy prevented an impairment of neuronal function in an AD model as shown in longitudinal rsfMRI studies (Blockx I, Einstein S, Guns PJ, Van Audekerke J, Guglielmetti C, Zago W, Roose D, Verhoye M, Van der Linden A, Bard F. Monitoring Blood-Brain Barrier Integrity Following Amyloid-β Immunotherapy Using Gadolinium-Enhanced MRI in a PDAPP Mouse Model. J Alzheimers Dis. 2016 Sep 6;54(2):723-35. doi: 10.3233/JAD-160023). Thus, rsfMRI can be used as a non-invasive read-out of early preclinical AD pathology. MRI has also been used to study the transient BBB disruption following Aβ immunotherapy in an AD mouse model (Objectives 6.i 6.ii and 6.iv). WP7 - Quantification of imaging biomarkers

INMiND is a translational neuroinflammation project, ranging from target identification, through synthesis of imaging biomarkers, to clinical proof of concept studies. WP7 is concerned with quantification of imaging biomarkers of neuroinflammation, i.e. translation of image information into quantitative values of the underlying neuroinflammatory process. The work within WP7 addresses Objectives 4 and 5 and can be divided into standardisation (Task 7.1) development of tracer kinetic models (Task 7.2) and development of quantitative procedures for small animal studies (Task 7.3). Standardisation is needed to obtain comparable results from different institutes, thereby facilitating multi-centre trials. Development of tracer kinetic models is needed to extract quantitative data on the underlying neuroinflammatory process from confounding processes such as perfusion and non-specific binding. Tracer kinetic modelling is needed for both human and small animal studies but, due to the much smaller size, special attention needs to be paid to measuring radioactivity concentrations in small animals. In general, the objectives set out at the beginning of the project have been achieved, thereby facilitating the interpretation of preclinical and clinical neuroinflammation imaging studies.
Standardisation heavily depends on the willingness of the various groups to adjust scanning protocols in accordance with a jointly agreed approach. Therefore, initially, several (tele)- conferences were organised to obtain consensus on the standardisation strategy. This was achieved quite rapidly for clinical studies. A major step was the development of the modified supervised cluster analysis (SCA) algorithm for data-driven identification of the reference tissue (Yaqub M, van Berckel BNM, Schuitemaker A, Hinz R, Turkheimer FE, Tomasi G, Lammertsma AA, Boellaard R. Optimization of supervised cluster analysis for extracting reference tissue input curves in (R)-[11C]PK11195 brain PET studies. J Cereb Blood Flow Metab. 2012 Aug;32(8):1600-8. doi: 10.1038/jcbfm.2012.59). This is important as most neuroinflammation tracers can be analysed using a reference tissue approach, but variability in the final results heavily depends on variability in the definition of the reference tissue. Next, the SCA algorithm was distributed to all partners in the consortium, so that all data could be analysed in a similar fashion. Finally, a database of normal human (R)-[11C]PK11195 data was generated including already existing data originating from various partners. As data acquisition of these older studies had not been harmonised, the question was whether this would work or whether harmonisation would be essential. In the end, data could be combined, but a difference remained between standard and so-called high-resolution scanners. On the other hand, using the database, it was possible to identify an age effect in TSPO binding, which could not be derived from the individual studies contributing to the database. It is clear that, for future studies, improved results can be obtained by harmonising scanners and acquisition protocols prior to collecting data. Nevertheless, this effort illustrates the potential of this approach and could pave the way for the generation of European imaging databases for imaging. Making use of such databases would reduce costs and radiation exposure of imaging studies (e.g. by using normal controls from a database rather than scanning a new cohort).
Standardisation took longer for preclinical studies, not only because of the larger number of different scanner types and tracers used, but also because of animal related factors, such as animal model, anaesthesia and other experimental conditions. Nevertheless, a protocol for standardising animal models of neuroinflammation was distributed amongst partners. This, in turn, led to cross-validation and standardisation of preclinical studies between groups using the same animal model and neuroinflammation tracer. In addition, within the consortium a preferred blood withdrawal system for preclinical studies was identified.
Tracer kinetic models were developed for all new ligands developed within the INMiND consortium, enabling quantitative studies using these ligands (e.g. Golla SS, Boellaard R, Oikonen V, Hoffmann A, van Berckel BNM, Windhorst AD, Virta J, Haaparanta-Solin M, Luoto P, Savisto N, Solin O, Valencia R, Thiele A, Eriksson J, Schuit RC, Lammertsma AA, Rinne JO. Quantification of [18F]DPA-714 binding in the human brain: initial studies in healthy controls and Alzheimer's disease patients. J Cereb Blood Flow Metab. 2015 May;35(5):766-72. doi: 10.1038/jcbfm.2014.261) (Objectives 4.i-ii and 5.i). For some of these ligands also test-retest data have been acquired to assess test-retest variability (e.g. Collste K, Forsberg A, Varrone A, Amini N, Aeinehband S, Yakushev I, Halldin C, Farde L, Cervenka S. Test–retest reproducibility of [11C]PBR28 binding to TSPO in healthy control subjects. Eur J Nucl Med Mol Imaging. 2016 Jan;43(1):173-83. doi: 10.1007/s00259-015-3149-8). In particular tracer kinetic models, ready for human use, were developed for the following tracers:
• [18F]DPA-714
• [11C]PBR28
• [11C]NE40
• [123I]CLINDE
• [18F]FEMPA
• [11C]GMOM
In addition, two different P2X7 ligands became available in the final year of the project. Pilot studies in normal controls, PD and MS patients are ongoing, and initial tracer kinetic models have already been developed. Nevertheless, further evaluation of these models will be carried out in an ongoing multicentre study in which both ligands are compared with each other. These ligands are:
• [11C]JNJ717
• [11C]SMW139
Finally, tracer kinetic models for various amyloid and tau tracers were developed. This is important, so that in future studies the degree of neuroinflammation can be investigated in vivo in relation to amyloid and/or tau load (Objective 5.i). For the two most widely used tracers from the list above tracer kinetic models were adapted to allow for quantitative small animal studies:
• [18F]DPA-714
• [11C]PBR28
In addition to the development of kinetic models for quantitative analysis of tracer uptake, also more general tools for quantification were developed. These methods included:
• Various clustering methods were evaluated both for other tracers than (R)-[11C]PK11195 and for small animal studies In particular, SCA was modified and implemented for routine clinical and preclinical [18F]DPA-714 studies.
• Several methods to correct for partial volume effects were developed and implemented for both human and small animal scans. These methods were combined with methods to improve signal-to-noise ratios, so that higher resolution images could be obtained without a significant increase in noise levels. This is important, as neuroinflammatory signals, especially in healthy subjects, are relatively small.
• A number of model selection criteria exist to identify the optimal model from a series of potential models. These model selection criteria themselves were evaluated on data from several tracers and it was shown that all criteria investigated selected the same optimal model.
• An unexpected finding was that whole blood to plasma ratios changed after the samples had been withdrawn, thereby potentially compromising blood input data. Therefore, the stability of blood samples was characterised and for various TSPO ligands guidelines were established how to handle blood samples.
• Although working models for TSPO tracers had been developed and validated, it has been questioned whether binding sites are present in the vascular wall, which could explain the relatively high background levels. Therefore, models that incorporate a vascular component were developed and validated for [11C]PBR28 and [18F]DPA-714.
• Although the focus of methodological developments was on tracer kinetic models for PET ligands, also a number of MRI based biomarkers for neuroinflammation were developed and validated.
Quantitative PET imaging studies in small laboratory animals are more difficult than corresponding studies in humans, partly because of the small size of the animals compared with the spatial resolution of a PET scanner (even in case of dedicated small animal scanners) and partly because quantification requires a metabolite corrected arterial plasma input function. Sampling methods used in humans cannot be applied to small animals simply because of the limited amount of blood that is available. Therefore, an important methodological aim was to develop a method to measure the arterial input function for small animals (rats). Methods were developed (e,g, Ory D, Postnov A, Koole M, Celen S, de Laat B, Verbruggen A, Van Laere K, Bormans G, Casteels C. Quantification of TSPO overexpression in a rat model of local neuroinflammation induced by intracerebral injection of LPS by the use of [18F]DPA-714 PET. Eur J Nucl Med Mol Imaging. 2016 Jan;43(1):163-72. doi: 10.1007/s00259-015-3172-9) and throughout the lifetime of INMiND they were constantly modified and improved. As mentioned above, ultimately a preferred blood withdrawal system for preclinical studies was identified and thereby also the procedure to obtain arterial input functions.
One solution, circumventing the need for measuring the arterial input function, is the use of a reference tissue model. This is only valid if a reference tissue exists that is devoid of the binding site under study. This appears to be the case for TSPO ligands, where the cerebellum can be used as reference tissue, at least for quantifying neuroinflammation in neurodegenerative diseases. The main problem in applying this approach to small animal studies is that a structure without specific binding is prone to spill-over of activity from surrounding brain areas with specific binding, which is more likely to be the case when structures become smaller. Again, developing methods to measure reference tissue input functions has been an ongoing process with constant modifications and improvements. As mentioned above, however, in the end it was possible to modify and implement SCA for preclinical [18F]DPA-714 studies.
Another source of variability is the exact definition of volumes of interest (VOI). For human applications, many VOI templates exist. For rat studies, however, there are no existing templates and usually VOI definition is based on an anatomical atlas. To facilitate quantitative PET studies of the rat brain, a dedicated rat brain VOI template appeared to be extremely useful. In addition, general use of such a template would also be an important step in data analysis standardisation. Therefore, based on high resolution MRI data available within the consortium and after consultation of all partners concerning relevant brain areas, a rat brain VOI template was generated and implemented in the (commercial) software of one of the small animal scanner vendors (Mediso), participating SME in INMiND (P26). In addition, the software was made available to all partners in the consortium even if they did not have a Mediso PET/CT or PET/MRI scanner. A similar approach was followed for a general partial volume correction method. WP8 - Clinical studies

The clinical WP8 has centred its work around Objective 5 which has been (i) the translation of new imaging paradigms for microglial activation into patients with various ND (AD, PD, HD, ALS) and MS with and without immunomodulatory therapies through (i) improved technical development, (ii) the improvement of knowledge on how NI processes influence the course of ND in humans, and how we may be able to therapeutically intervene with these processes, (iii) improved knowledge on disease pathophysiology and (iv) the development of new, innovative imaging-guided therapy paradigms (theranostics) as well as the related subobjectives.
A large number of studies in neurodegenerative diseases have been performed with second-generation TSPO tracers, particularly with [18F]DPA-714 and [11C]PBR28, including longitudinal studies and testretest studies. These tracers are now well-characterised and we have now a good knowledge of the advantages and disadvantages in a clinical context (Objective 5.ia).
Not only novel PET markers for microglial activation have been evaluated and widely used but also a SPECT tracer for microglial activation ([123I]CLINDE) now has been characterised bringing TSPO imaging closer to the clinical arena due to the wider availability of SPECT in comparison to PET (Feng L, Svarer C, Thomsen G, de Nijs R, Larsen VA, Jensen P, Adamsen D, Dyssegaard A, Fischer W, Meden P, Krieger D, Møller K, Knudsen GM, Pinborg LH. In vivo quantification of cerebral translocator protein binding in humans using 6-chloro-2-(4'-123I-iodophenyl)-3-(N,N-diethyl)-imidazo[1,2-a]pyridine-3-acetamid e SPECT. J Nucl Med 2014; 55(12): 1966-72. doi: 10.2967/jnumed.114.143727).
Other markers of neuroinflammation and especially astroglial activation has been examined in genetic forms of Alzheimer’s disease providing further insight into the very early pathophysiological changes in this disorder (Rodriguez-Vieitez E, Saint-Aubert L, Carter SF, Almkvist O, Farid K, Scholl M, Chiotis K, Thordardottir S, Graff C, Wall A, Långström B, Nordberg A. Diverging longitudinal changes in astrocytosis and amyloid PET in autosomal dominant Alzheimer's disease. Brain 2016; 139(Pt 3): 922-36. doi: 10.1093/brain/awv404) (Objective 5.ia).
A subgroup within the consortium has been successful in obtaining external funding from the Michael J Fox Foundation to evaluate and compare clinically 2, within WP3 newly developed, PET markers for the microglial P2X7 receptor in Parkinson’s disease extending and continuing the original INMiND work.
A large longitudinal study using [18F]DPA-714 and amyloid PET in early Alzheimer’s disease within the consortium has allowed important insights about the potential role of microglial activation at different stages of the disease and test-retest studies in using [11C]PBR28 suggest that there might be a diurnal variability in microglial activation (Hamelin L, Lagarde J, Dorothee G, Leroy C, Labit M, Comley RA, de Souza LC, Corne H, Dauphinot L, Bertoux M, Dubois B, Gervais P, Colliot O, Potier MC, Bottlaender M, Sarazin M; Clinical IMABio3 team. Early and protective microglial activation in Alzheimer's disease: a prospective study using 18F-DPA-714 PET imaging. Brain 2016; 139(Pt 4): 1252-64. doi: 10.1093/brain/aww017) (Objectives 5.ib and 5.iic).
Several proof of principle studies in neurodegenerative disorders as well as in non-neurodegenerative disorders (particularly stroke, MS and psychosis) have been conducted for the duration of the consortium further elucidating the role of neuroinflammatory changes in these diseases. Particular emphasis has been put on examining the role of microglial activation in genetically determined forms of Alzheimer’s disease as well as Huntington’s disease demonstrating that microglia activation is part of the disease process even in the presymptomatic phase of these disorders (Objectives 5.iib and c, 5.iii).
In conjunction with WP9 we have performed the first immunomodulatory trial in Alzheimer’s disease, monitoring response in vivo and we refer to the details in the WP9 section (Objective 5.iic) below.
We have conducted the first study on TSPO expression in human gliomas that combined
(R)-[11C]PK11195 PET-imaging and tissue assessment of TSPO expression and assessed the extent and distribution of TSPO in gliomas using a large cohort of cases where we compared in vivo and in vitro results of the same patients. Experiments using glioma cell lines have also been conducted to investigate the function of TSPO in tumour cells.
A study combining PET-imaging and analysis of post-mortem brains of subjects with low grade gliomas to investigate microglial activation in the contralateral brain has also been concluded and demonstrated neuronal and synaptic loss away from the tumour epicentre (Su Z, Roncaroli F, Durrenberger PF, Coope DJ, Karabatsou K, Hinz R, Thompson G, Turkheimer FE, Janczar K, Du Plessis D, Brodbelt A, Jackson A, Gerhard A, Herholz K. The 18-kDa mitochondrial translocator protein in human gliomas: an 11C-(R)PK11195 PET imaging and neuropathology study. J Nucl Med 2015; 56(4): 512-7. doi: 10.2967/jnumed.114.151621) (Objective 5.iv).
For a detailed overview of the studies performed per tracer, see Annex 3.1.1; Table 2. WP9 - Imaging-guided immune-modulation in patients with MCI

Within this WP it was foreseen to (i) implement a clinical study in 46 amyloid-positive patients with amnestic MCI, who have a high likely-hood for conversion into manifest AD and to (ii) non-invasively follow within this patient group the effects of immunotherapy (anti-TNF-α agent Certolizumab pegol, Cimzia; UCB) using the microglial activation (second generation TSPO) PET marker [18F]GE-180 and after patient selection with the amyloid marker [18F]Flutametamol (both PET ligands from GE Healthcare) and to (iii) relate the imaging markers with the clinical outcome at the end of a 12 month treatment period (Objectives 5 and 6). At the beginning of the project there had been a secured agreement with both companies for free supply of the drug and the imaging ligands. However, during the course of the project first GE Healthcare (at month 20 of the project) and later also UCB (month 36) withdraw their commitments for supply of these agents resulting in the search for alternative compounds and funding and in several reiterations to obtain the necessary study approvals. After careful re-evaluations, plans were adjusted to employ the gold standard TSPO marker (R)-[11C]PK11195 for treatment evaluation, the amyloid marker [18F]florbetapir for patient selection and the TNF-α inhibitor Etanercept (Pfizer) as immunomodulatory drug, the supply of which was secured from the Alzheimer’s Society UK (AS UK) and the Alzheimer’s Drug Discovery Foundation (ADDF). These drawbacks resulted in a severe delay in patient recruitment with randomisation of the first study patient only at the very end of the forth project period (February 2016) and a total of 13 patients randomised at the end of the fifth project period and this despite the fact that in order to support and expedite the recruitment process a second recruitment site besides Southampton had been implemented and steps had been undertaken to increase the recruitment area around Southampton. Based on the recruitment rates during the fifth project period a no-cost extension of 28 months (for recruitment and treatment of all foreseen 46 patients) or at least of 12 months (for completion of the 12 months follow up in those 13 patients recruited by the end of project period 5) was requested. Agreement was given for a 12 months extension and so these 13 subjects were followed to completion. Randomisation of these 13 patients led to two treatment groups (n= 6 and 7) that were similar with respect to demographic details. Compliance to study medication was high (median 94% both groups) over the twelve months trial period. The overall study completion was 77% (10/13) with no statistical difference in the completion rates between those allocated etanercept and those allocated placebo. There were 38 (36%) adverse events in the etanercept group and 67 (64%) in the placebo group. No new adverse events were identified that were out of keeping with the drugs known safety profile. None of the clinical outcomes at 12 months were statistically significant between treatment groups. Following randomisation, inactivated serum TNFα was higher in the treatment compared with the placebo group at all time points but there were no significant differences between other serum inflammatory markers between the treatment and placebo groups. Of the 13 subjects randomised to treatment allocation a final quantification of amyloid resulted in one subject being reassessed as being negative for amyloid. Of the 12 subjects remaining 11 patients underwent a repeat (R)-[11C]PK11195 scan at 12 months. 10 subjects also had a repeat amyloid PET scan, eight also a repeat MRI scan. Binding potential values have been obtained for all (R)-[11C]PK11195 scans as well as uptake ratios for all amyloid PET scans. Currently we are conducting statistical analysis to compare the differences between the placebo and treatment group as well as the correlation analysis of (R)-[11C]PK11195 binding and peripheral inflammatory markers. We expect the analysis concluded by the end of May 2018 and submission of the manuscript summarising the main results by July 2018. WP12 - Training

The aim of WP12 was to ensure a high level education in the field of INMiND-related research subjects within, throughout and across the INMiND consortium. To do so, a huge effort was done during the five years of the project to (i) create, update and manage the INMiND training courses (ii) settle and update the training rules (iii) implement, adjust and coordinate the annual training offers thanks to the organisation of 19 conference calls (iv) organise the training sessions at the training sites and (v) prepare the 5 annual training offers.
Over the 5 past years, in total 473 participants, of which 134 INMiND members, attended 31 training courses organised by the consortium. All in all, 189 students took part in the INMiND training courses.
Considering the assessment done by the participants, we can say that the INMiND training courses were really useful and appreciated. Indeed, 97% of the participants considered the learning objectives were reached. Moreover, 62% of the trainees applied the acquired knowledge immediately after the course.
Courses were very successful with good interaction between participants and teachers. Participants were enthusiastic about the organisation of the courses, the quality of the teaching staff, the possibilities to perform hands-on exercises on the systems, the different modalities assessed in the courses, the interactive lectures, the multi-cultural and multi-disciplinary environment, the small group sessions and the visits of core-facilities.

Overview of the total number of courses and participants per project period:
• Period 1: 4 training courses, attended by 53 participants
• Period 2: 8 training courses, attended by 135 participants
• Period 3: 7 training courses, attended by 86 participants
• Period 4: 8 training courses, attended by 142 participants
• Period 5: 4 training courses, attended by 57 participants
• In total: 31 training courses, attended by 473 participants.
For a detailed description of each training course, please see Annex 3.2.1; Table 7.

In order to boost and promote the mobility and exchange of personnel within the consortium, an inquiry has been circulated and collected all offers and requests for specific needs/skills among students and/or experienced researchers leading to the exchange of 14 students and researchers over the 5 years of the INMiND project.
The embedding of INMiND training activities in the scope of related European educational programmes was achieved through (i) the co-organisation and promotion of the annual winter conferences (TOPIM – hot TOPics in molecular IMaging) of the European Society for Molecular Imaging (ESMI) since 2013, (ii) the co-operation with European and international societies (ESMI, EANM, SNMMI, WMIS) to set up the curriculum on ‘Preclinical Imaging in Small Laboratory Animals – PRIMA’, (iii) the embedding of the INMiND trainings in the curriculum of the European Molecular Imaging Doctoral School – EMIDS launched in 2012, (iv) the promotion and support ensured by the WP Training of the French Infrastructure France Life Imaging (FLI) leading to the financial support of 24 PhD students, and (v) the contribution to the sustainability of the training actions with submission of 2 European Training Networks project proposals (H2020/MSCA) coordinated by 2 INMiND partners in 2017. Indeed, the INMiND training office launched a survey in July 2016 in order to inventory the partners willing to take part in European Training Network actions, as well as the potential topics and industrial partners. Based on the outcome of this survey, P8 and P13a implemented 2 ETN projects in the scope of the Marie Skłodowska-Curie Actions and in which they act as coordinator. P8 coordinated the TRAiNSGlioma proposal and P13a the proposal called PET-IQ in which another 5 INMiND academic partners from 5 European countries are involved (P1, P32/VUmc, P10a, P11, P14). Both proposals were submitted during the January 2017 call of the Marie Skłodowska-Curie Actions. Unfortunately, the review of both proposals was not favourable enough for funding (88% and 87% respectively, required score > 93%). The TRAINSGlioma proposal was adjusted to the comments of the reviewers and resubmitted under the name eTrain2Brain during the January 2018 call (review pending), whereas P13b plans to resubmit the PET-IQ proposal for the 2019 call.
Potential Impact: Strategic impact

The INMiND project combined basic research to increase knowledge of biological processes and mechanisms involved in specific disease situations (identification of new molecular targets involved in microglia-mediated neuroinflammation for diagnostic and therapeutic purposes; theranostics) and transported this knowledge into clinical applications including disease control and treatment (recruitment of patients for therapy at early disease stages based on imaging; translational personalized medicine).
As INMiND focussed on neurodegenerative diseases (e.g. AD, PD), the results obtained will have an impact not only on one of the “major health-related societal challenges”, such as brain-related diseases, but also on the “socio-economic dimension” in our aging European population.
Due to the true translational nature of the INMiND research and the concerted collaborations of the INMiND partners, the specific impacts of the project are on
1. increased knowledge on brain dysfunction related to NIND
2. endurable European research efforts: innovation in the area by participation of clinical centres, research-intensive SMEs, industry (pharmaceutical, imaging instrumentation, contrast media) and other stakeholders (charity and patient organisations); and training of a new generation of imaging scientists
3. efficient deployment of health care system Impact on understanding brain dysfunction and on brain-related research within the field of NIND

The major impact of the multidisciplinary work performed within the INMiND consortium consists in the generation of a complete array of innovative, translational tools for the study of the dynamics of neuroinflammation in relation to the onset and progression of the most common neurodegenerative diseases (one of the “major health-related societal challenges”) paving the way for earlier and better diagnosis and therapy of these disorders and for the understanding of their gender-selective manifestation.
By revealing the molecular mechanisms of microglia activation and the application of high-end imaging technologies to a variety of animal models available in the consortium, the INMiND project shed novel light on the involvement of neuroinflammation in the onset and progression of neurodegenerative diseases and provided highly innovative and powerful strategies to test the therapeutic potential of immunomodulation both on a preclinical level and in patients at the earliest detectable disease state (patients with PIB-positive MCI). In addition, the INMiND consortium identified new molecular targets specific for the neuroprotective M2 microglia phenotype and designed several M2-specific PET radioligands that are being validated for in vivo application.
Due to the translational nature of the INMiND research, the project results substantiated the feasibility of image-guided patient selection at very early disease stages, where promising and innovative therapies still have the potential to preserve neuronal function and, hence, patient well-being.
Moreover, INMiND illuminated the dynamic concert of molecular mechanisms involved in microglia-directed neuroinflammatory responses in neurodegeneration and neuro-regeneration, and showed that these mechanisms also play a role in other common brain-related diseases (e.g. stroke and glioma). This makes that the established paradigms of image-guided diagnosis and treatment (‘imaging theranostics’) are translatable to other diseases, where imaging biomarkers play a role in diagnosis, prognosis and early assessment of therapy efficiency, multiplicating the outreach of the project beyond NIND.
The knowledge gathered by INMiND on the basic mechanisms of NI, ND and NR, on how to design, implement and value image-guided theranostic strategies for NIND and beyond, and on how to set new standards and guidelines on the diagnosis and management of NIND will have a direct and long-term impact within the ERA. All INMiND partners contributed significantly to share and implement the newly generated knowledge with national, European and world-wide research communities (e.g. ESMI, SMI, AMI, EANM, ESR, SNM, SNMMI, ENS, EFNS, AAN, SRS, and many others), with the general public, stakeholders and policy makers. For the details, see Paragraph 4.2 Section A. Impact on sustainable European research efforts

In line with the EU’s Lisbon Strategy “to become the most dynamic competitive knowledge-based economy in the world”, the INMiND consortium gathered together key players from 12 European countries, with many of them being world leaders in their respective field, for the execution of cutting-edge translational and reverse-translational research in the field of NIND. The INMiND project integrated 6 research-intensive SMEs with complementary expertise in the generation of mouse models (P22), design of tracers and contrast media (P23/24), searching for drug targets in the CNS (P25/28) and improving imaging systems (P26). These SMEs substantially contributed to the success of the project and in turn benefited from the collaborations the academic partners offered.
The INMiND project expected to improve the performance of these individual companies as an EU-based company and allow them to better compete at a global level and secure jobs. At the end of the project the individual SMEs appraised their participation in the consortium as follows:

P22/TOP - Transgenic Operative Products SRL: Beneficiary P22 had to withdraw from the INMiND consortium due to liquidation of the company. The liquidation process started on September 3rd 2013. The effective date of termination was February 28th 2014.

P23/BV-Cyclotron: For the BV Cyclotron VU it is very important to stay strongly connected to the academic groups that are active in the development of new PET tracers. The BV Cyclotron VU has had the opportunity within INMiND to learn about new tracers in neuroinflammation, has participated in two particular projects related to tracer development (MMP and the H4 receptor) and has strengthened its network in the EU. Because of the experiences learned within INMiND, BV Cyclotron VU has gained knowledge in processes to take new tracers from concept towards human applications where normally BV Cyclotron VU has experience in processes to up-scale tracer production towards multi-customer batches. This complementary knowledge will benefit the BV Cyclotron VU in the future.

P24/Cage: CAGE Chemicals, through the participation to the project INMIND, had the opportunity to improve the collaboration with P5 (Università degli Studi di Torino), leading to the development of novel imaging probes precursors. Moreover, the project represented a unique opportunity to meet international experts in this field regularly and to start new collaborations with them. In this context, the support of the INMIND project allowed CAGE Chemicals to allot additional personnel, resources and time to the research in the field of imaging probes, providing the project partners with original probes and corresponding precursors and to receive from them valuable feedbacks on their performances.

P25/Pharmidex: The benefits for Pharmidex participating in INMiND has been tremendous in terms of giving us an insight into the application of the significant collective research with very dedicated scientists on the molecular mechanisms that link neuroinflammation with neurodegeneration. We have greatly treasured and profited from the great diversity of the unique scientific minds across Europe within this collaborative project.
Pharmidex has benefited from not only participating heavily in the various scientific Work Packages, but has also taken the opportunity to send some of its staff to the numerous INMiND Training Courses across the 13 training sites. This dissemination improved our awareness, knowledge and technical methods at the numerous consortium European sites. We have been extremely grateful for this unique aspect within this grant, especially assisting in the training and potential career progression of our junior Pharmidex staff.
Pharmidex has also profited from the numerous scientific publications that have been published in related peer-reviewed high-impact journals. This has enabled us to contribute significantly to new knowledge generated within the research communities and has increased impressively our scientific presence as an SME company.
We have been extremely grateful to work in this unique and exceptional multi-disciplinary consortium. It has indeed been an honour for Pharmidex, as a predominately neuroscience and CNS company, to participate in such a consortium that will have direct and lasting impact within the industry of how to design, validate, implement and further progress NIND theranostic strategies.
We have helped in the goal of INMiND to identify and advance fundamental knowledge in the novel biological targets of neuroinflammation for both diagnostic/therapeutic purposes, translating this knowledge into clinical application that will have a lasting legacy in this scientific arena. We hope to transfer the ideas and technology learnt from the INMiND grant to other future European projects/grants/services. We expect that the increased profile for Pharmidex will improve us, an EU based SME Company, to better compete at a global level with the advantages of increased revenue generation activities, helping with our diversification of our portfolio, increased revenue generation activities and with this to help to create more high quality jobs within the EU community.

P26/Mediso: As part of the INMiND consortium Mediso mainly benefited from the large-scale collaboration of partners with different scientific background. All of the basic tools and algorithms developed during the project are built upon our standard image processing software library, paving the way for future commercialization. New image registration types, new filters and new volumetric measurement tools have been added to the image processing software library alongside with the MRI based rat brain atlas which was developed based on specification given by experienced academic INMiND partners. All newly developed tools have been implemented and integrated into our multi-modality post processing software, the InterView FUSION. Following successful testing by project partners, these are now available for future commercialization.

P27/Neurosearch: Due to major restructuration processes at NeuroSearch during the first project year the company left the consortium with the effective date of February 28th, 2013. All tasks assigned to P27 were taken over by P28 (Saniona) being founded as a management buy-out from NeuroSearch. Saniona entered the consortium on March 1st, 2013.

P28/Saniona: With the contribution from INMIND Saniona were able to hire a microglia expert and to establish a collaboration with NRU, University of Copenhagen, concerning characterization of expression of potassium channels in human microglia cells from epilepsy patients. Saniona had already for several years been heavily involved in the role of potassium channels in autoimmune and neurological diseases, and the collaboration allowed us to specifically validate these potential drug targets in human microglia. A cornerstone conclusion was that Ca-activated K channels, in particular KCa3.1 and KCa1.1 were much more dominating in human microglia than had previously been deduced based on rodent studies, whereas voltage-dependent potassium channels such as Kv1.3 that is very important in animal models, were completely absent in humans. In the longer perspective, this work has accentuated our focus on KCa channels as a target class, but also raised a more general concern about the translatability of animal models in drug development. We have recently started exploration of the role of microglia in pain, and found very encouraging effects of KCa3.1 inhibitors in both neuropatic and inflammatory rat models of pain (unpublished data). The results obtained have thus significantly influenced our scientific and business strategy.

Apart from its research and innovation activities, INMiND also aimed to serve a trans-national (European) educational programme for young researchers. Nine INMiND partners (P1/2/4/5/6/7/8(30)/12a/14) offered an extended and intensified training programme on INMiND topics of research (activities organised within WP12). The INMiND training courses fostered and strengthened the trans-national exchange of students and researchers, attracted students not only from Europe but from all over the world and served (i) to increase the competitiveness for excellent educational and research opportunities in Europe and (ii) as magnet for excellent students from North America, India and Asia to support “brain gain” in Europe.
For more details on the INMiND training courses, see WP12 summary (see and the training paragraph in on main dissemination activities.
During and especially towards the end of the INMiND project several measures were started to ensure the sustainability of the research collaborations initiated within INMIND involving EC initiatives, industry, private funding, charity or patient organisations.

WP9 trial ‘Imaging-guided immune-modulation in patients with MCI’ together with AS UK and ADDF: Initially within this randomised placebo-controlled trial the use of the anti-TNF-α agent Cimzia as study drug was foreseen with free supply use of this compound and placebo by the company UCB. However, after the withdrawal of the secured agreement by UCB additional funding of the trial became necessary as the EC funds within the INMiND project were not enough to cover these extra expenses. Additional external funding was searched and finally secured from the Alzheimer’s Society UK (AS UK, and the Alzheimer’s Drug Discovery Foundation (ADDF, Both foundations have the mission to accelerate the discovery of drugs to prevent, treat, and cure Alzheimer's, a mission in line with the INMiND mission, and agreed to jointly pay for the provision of Etanercept (a TNF-α inhibitor, Pfizer) with placebo and randomisation by ACE pharmaceuticals. Only because of these additional funds would it have been possible to recruit, treat, image and analyse all planned study patients and could the trial be initiated.

First in human study of [11C]SMW139 together with Merck Serono: Clinical application of one of the two P2X7 receptor-targeted radioligands developed within INMiND WP3, namely [11C]SMW139, has been made possible by an award obtained from the Grant for Multiple Sclerosis Innovation (GSMI) fund of Merck Serono. Prof. Windhorst (P9/31/P32) in collaboration with Prof. H.E. de Vries of the VUmc MS Center were one of the four award winners of the third GMSI in 2015 for a first in human study with [11C]SMW139 in MS and age matched controls.

PRI-PD project together with the Michael J. Fox Foundation: Furthermore, led by Prof. Jacobs (P1) and Prof. Windhorst (P9/31/32) seven INMiND partners (P1, P9/31/32, P10a, P12a, P13a, P16 and P19) were invited by the Michael J. Fox Foundation to submit a research proposal to continue investigations of both P2X7 receptor-targeted radioligands developed within WP3 ([11C]JNJ717 and [11C]SMW139) in a multi-centre study of PD patients. In spring 2016 MJFF decided to fund this project, entitled PRI-PD (P2X7 receptor-based microglia PET imaging in Parkinson’s Disease), with about $935.000 for two years (10/2016-09/2018).

D-CogPlast - Identification of dietary modulators of cognitive ageing and brain plasticity and proof-of-concept of efficacy for preventing/reversing cognitive decline: The D-CogPlast consortium, funded under the JPI HDHL joint action "Nutrition and Cognitive Function" (NutriCog, and headed by Sandrine Thuret (King’s College London, UK) will examine the role of dietary bioactives affecting brain plasticity for cognitive ageing processes. Furthermore, different age groups will be under study to investigate a sensitive population responsive to dietary interventions. Exposure to stress and genetic predispositions will be considered in addition. D-CogPlast consists of 6 partners from 5 countries (UK, AU, FR, NL ES) including INMIND partner P3. D-CogPlast is a translational project across nutrition, epidemiology, and neurosciences, ranging from studies in rodent models to human data. Innovative methods will be employed, such as food metabolomics, in-vitro approaches to study brain plasticity in humans and cognitive ageing models in rodents. Wihtin the project, P3 analyses the activity of serum from the study cohort on microglia phagocytosis. It is funded with 1 Mio Euro for 36 months.

IMBI project within the EU Joint Programme – Neurodegenerative Disease Research: Beginning of 2016, eleven INMiND partner institutions (P1, P4, P9/31/32, P10a, P11, P12a, P12b, P13a, P14, P15 and P16) together with some other institutions joint efforts to submit a proposal under the scope of the JPND call for “Working Groups for Harmonisation and Alignment in Brain Imaging Methods for ND”. The working group led by Prof. Perani (P11) and Prof. Jacobs (P1) and titled “Framework for Innovative Multi-tracer molecular Brain Imaging (IMBI) to enable multi-centre trials and image evaluation in (early) neurodegenerative disease” received a positive quote and is being funded by JPND with about 50.000€ for a duration of 9 months (02-11/2017) (

Marie Skłodowska-Curie Actions: The INMiND management office supported the training office launching a survey in July 2016 in order to inventory the partners willing to take part in European Training Networks actions, as well as the potential topics and industrial partners. Based on the outcome of this survey, P8 and P13a implemented 2 ETN projects in the scope of the Marie Skłodowska-Curie Actions and in which they act as coordinator. P8 coordinated the TRAiNSGlioma proposal and P13a the proposal called PET-IQ in which another 5 INMiND academic partners from 5 European countries are involved (P1, P9/31/32, P10a, P11, P14). Both proposals were submitted during the January 2017 call of the Marie Skłodowska-Curie Actions. Unfortunately, the review of both proposals was not favourable enough for funding (88% and 87% respectively, required score > 93%). The TRAINSGlioma proposal was adjusted to the comments of the reviewers and resubmitted under the name eTrain2Brain during the January 2018 call (review pending), whereas P13b plans to resubmit the PET-IQ proposal for the 2019 call.

TripleP project, granted by the Michael J. Fox Foundation: Prof. Windhorst (P9/31/32), in collaboration with Dr. D.J. Vugts, Dr. B. Drukarch and Prof. H.W. Berendse obtained a grant in 2016 ($200.000) to continue the work on P2RY12 PET tracers development, which was initiated within INMiND. They will design and synthesise new P2RY12 PET tracers and evaluate them with preclinical PET/CT/MR.

Neuroscience Campus Amsterdam (NCA) grant: Prof. Windhorst (P9/31/32), in collaboration with Dr. D.J. Vugts and Prof. H.E. de Vries obtained a grant of the NCA (€100.000) in 2015 to study M1/M2 polarisation of microglia with P2X7 and P2RY12 PET/CT/MR in the rat EAE model and investigate the effect of PPAR agonists on modulation microglia toward M2 polarisation in vivo.

Montelukast trial together with a Biotech company and a non-profit organisation: As follow up on the very successful Montelukast studies performed by P3 within INMiND, which resulted in a Nature Communications publication (Marschallinger et al. Nat Commun. 2015 Oct 27;6:8466. doi: 10.1038/ncomms9466) this partner teamed up with IntelGenx, a Canadian Biotech company, for a reformulation/repurposing project for Montelukast and Alzheimer’s Disease. The company developed a product out of it and P3 positively passed a Phase I clinical trial in summer 2016. P3 is now in the process of discussion with The Consortium of Canadian Centres for Clinical Cognitive Research (C5R) to start a clinical phase IIa study in Canada within the next months. Health Canada has already approved the study and it is already listed in under NCT03402503. C5R is a not-for-profit research network that facilitates collaboration and partnerships between pharmaceutical companies and Canadian dementia researchers. C5R research sites conduct clinical trials in the desire to research and develop treatments for patients with mild cognitive impairment, Alzheimer’s Disease as well as other forms of dementia, making C5R a perfect partner for INMiND researchers ( In addition, negotiations with IMI EPAD European prevention of Alzheimer’s dementia consortium, and the ADDF foundation are ongoing for further funding and supporting of this project.

Tau project with Avid: In a multi-centre study among 2 INMiND partners (P11 and P16), the tau tracer [18F]AV-1451 will be tested in genetic preseninil 1 cases (both patients and siblings). This study will be supported by the company AVID who will supply the tracer precursor for free. The mission of Avid, a wholly owned subsidiary of Eli Lilly and Company (NYSE:LLY), perfectly fits within the INMiND goal: to develop new molecular imaging agents capable of changing the medical management of significant, chronic human diseases by identifying the first stages of pathological change, potentially assisting in earlier diagnosis, and better management and development of new therapies ( The project is about to start. The Clinical Agreement between Eli Lilly and Company (NYSE:LLY), and the San Raffaele University, Milan Italy (P11), respectively the University of Turku Finland (P16) has been obtained. The approval by the Italian Commission (AIFA) for the use of the tau tracer [18F]AV-1451 is in the due course and respective permissions from local authorities in Finland are to be applied. Local funding for this project has been secured.
IMI2 – Call 14: In April 2018 two INMiND partner institutions (P1, P9/31/32) decided to expand their collaboration and to prepare, together with other research groups, a proposal covering Topic 2 “Non-invasive clinical molecular imaging of immune cells”. This initiative is headed by Prof. Windhorst (P9/31/32). Impact on management of NINDs, health care cost and society

Although clinical neuroimaging technologies are cost-intensive themselves, we expect that our paradigms will influence healthcare costs as image-guided patient selection and efficacy approval will select individual patients for appropriate therapy paradigms (individualized patient care). Even more importantly, image-guided patient selection and follow-up will have great impact on the development, implementation and in vivo analysis of new therapeutic compounds in Phase I-III clinical trials. This has been demonstrated in the past by P18 in the detailed analysis of the influence of new dopaminergic drugs on disease progression in patients with Parkinson’s disease (Whone et al. 2003) and by P16/P18 in the demonstration that passive immunization with bapineuzumab, a humanised anti-amyloid- β monoclonal antibody, achieves a significant reduction in cortical fibrillar amyloid-β load as measured by PIB-PET in patients with Alzheimer’s disease (Rinne et al. Lancet Neurol 2010; 9: 363-72). We expect that the image-guided immunomodulatory trial started within INMiND WP9 by P1/P10a/P17 will confirm the feasibility of this paradigm. Furthermore, the (R)-[11C]PK11195 normals database implemented within the project will reduce health care costs and also radiation exposure to subjects in future imaging studies as it will be possible to use the normal controls data from the database rather than having to scan a new cohort.

Especially in our aging population, the implementation of new therapy paradigms aiming to delay dependency will have important impact on health care and costs. As detailed in the World Alzheimer Report 2010 (Wimo & Prince 2010; the “cost of dementia will continue to increase at an alarming rate”:
• approximately 0.5% of the world’s total population live with dementia;
• the total worldwide costs of dementia for social and medical care have been estimated to 604 billion US$ in 2010, with 70% of costs occurring in Western Europe and North America (if dementia were a company, it would be the world’s largest by annual revenue);
• there are 35.6 million people affected in 2010 (9.9 million in Europe), increasing to 65.7 million by 2030 and 115.5 million by 2050;
• costs for informal (42%) and social care (42%) exceed the costs of medical care (16%);
• for Europe, costs in 2010 were 103.6 (informal), 98.8 (social), and 36.3 (medical) billion US$, respectively.

The image-guided therapeutic strategies for NIND developed and implemented within INMiND and especially applicable at very early disease stages will have a great impact on disease course and quality of life (QoL) of our elderly population. Although costs in the medical care sector will increase due to implementation of new drugs on the market and imaging technologies, the better, earlier and personalised management of costly neuroinflammatory and subsequent neurodegenerative diseases will result in a delay of disease progression and have its greatest impact on (i) the disease course, (ii) patient and family well-being (QoL) and, hence, (iii) reduction of informal and social costs. Main dissemination activities and exploitation of results

The INMiND consortium integrated multiple NIND-related disciplines with extraordinary excellence in specific applications. All core technologies and applications of INMiND functioned as leverage (i) to create new standards and procedures and (ii) in the commercialization of new compounds and products in conjunction with SMEs and industry at the European and international level, respectively. Moreover, the INMiND training sites attracted early and advanced research scientists from all over the world for high-level training in NIND-related specialties.

The specific dissemination achievements included:
• new molecular targets involved in microglial function and neuroinflammation in vivo
• new targeted probes for assessing microglial function and neuroinflammation in vivo
• new targeted therapeutics to control microglial function and neuroinflammation in vivo
• new animal models to study microglial function in the context of NIND
• new image-guided theranostic protocols in animals and in patients
• improved guidelines for early disease detection and disease progression of NIND
• improved guidelines for diagnosis and management of patients with NIND

The quality of the work performed within INMiND is attested by the following tables (Annex 3.1.1):
• Table 2: overview of all PET tracer studies
• Table 3: overview of all MR readout studies
• Table 4: overview of all in vitro treatment studies
• Table 5: overview of all in vivo treatment studies Dissemination and Communication

Coordination of the INMiND dissemination activities has been guaranteed by P14 of the STB in close collaboration with the SMB and P1 (WP11). The activities aimed at making all possible users aware of INMiND project output (e.g. research results; novel mouse models; novel probes, compounds and CNS targeted molecules; novel theranostic protocols; conferences; training courses; publications) in a coherent and consistent way.

Dissemination was ensured through
• research publications in peer-reviewed journals, review articles and book chapters
• organisation and coorganisation (e.g. together with ESMI, WMIC) of national and international research meetings and conferences, summer/winter schools
• poster, oral presentations and educational talks of INMiND foreground at national and internal meetings, conferences, symposia, seminars, schools, etc related to NIND research, brain related disease, neuroscience, molecular imaging, and chemistry (e.g. organised by ESMI/SMI/AMI; EFNS/ENS/AAN; EANM/SNM, ESR, SRS, etc.)
• lectures to the general public
• advertisements, press and web releases, radio- and television broadcasts
• the INMiND website ( (public domain) and intranet (protected): to facilitate the flow of information within the INMiND consortium, to the general public, stakeholders and policy makers
• training activities, exchange of personnel

The target addresses of dissemination activities were
• neurologists in general practice and at clinical departments
• scientists within the ERA and world wide
• SMEs and industry promoting related technologies
• international, national and regional authorities and policy makers in the EU countries
• general public.

Any dissemination activities and publications in the project acknowledged European Commission's FP7 funding, as defined by clause of the Consortium Agreement.

The quality of the communication, dissemination and exploitation of the INMiND foreground within and beyond the INMiND consortium is attested by the following tables included in Paragraph 4.2:
• list of INMiND publications and dissemination activities (see Paragraph 4.2 Section A)
• list of IP developed (see Annex 3.1.2; Table 6)
• list of exploitable foreground and patents (see Paragraph 4.2 Section B) Training

Specialized training on NIND-specific core technologies and subjects has been performed on INMiND training sites as depicted in the WP12 section. The training sites, which also included one SME site (P25), were selected on the basis of their level of development in a key technology, their previous training experience (5th and 6th FW; Marie Curie; NoE), and their commitment towards the sharing of their know-how that they have developed in the past to the highest scientific level. They agreed to create the conditions (staff, facilities) necessary to welcome, teach and train other scientists and students from INMiND partners and from the general research community on their specific competence.

All the INMiND training activities served to
• exponentially disseminate INMiND research ideas and applications throughout the ERA
• enhance the communication within and outside the research community, e.g. to industry and public

The quality of the INMiND trainings and the INMiND training sites is attested by the overview table of all trainings organised during the course of the project (see Annex 3.2.1; Table 7).

Moreover, all academic INMiND partners took their local teaching duties at heart and provided supervision and support for the successful completion of the thesis of 41 students (24 females, 17 males) within INMiND-related topics during the project period. For another student his thesis defense was planned within the next 6-9 months after project end. For details, see Annex 3.2.2; Table 8. Exchange and mobility of personnel, exchange of material and know-how

With regards to appropriate resource identification within the consortium, the available equipment, tools, and facilities from the partner laboratory (e.g. animal models, scanners, tracers, ...) were identified and registered in the INMiND intranet. The centralized management of this data base identified duplications, synergies and missing entities and actively helped partners to find and share the facilities, tools or equipment needed for their projects.
Furthermore, the mobility of personnel around different institutions was supported for rapid exchange of common knowledge, know-how and expertise, especially within the scope of translating basic knowledge into clinical application with the many different techniques involved. This was ensured by providing relevant information regarding (i) training courses, (ii) facilities for individual “hands-on” training, and (iii) student and post-doctoral fellowships and open positions, on the INMiND website.

The quality and success of the exchange and mobility of researchers is attested by the exchange of personnel table (see Annex 3.2.3; Table 9). Integration of SMEs

The INMiND consortium was successful to attract 6 highly qualified, research-intensive SMEs with complementary fields of expertise in mouse technology (P22), probe design (P23/24), neuropharmacology (P25/28), and imaging technology (P26). These SMEs are instrumental for dissemination and exploitation of new knowledge, models, compounds, technologies generated within INMiND under the lead of P25 and the STB.
Moreover, SMEs participated at annual meetings of INMiND, ESMI, and other meetings where INMiND members participate to ensure dedicated interactions about NIND-related concepts and ideas between researchers and industry representatives. For the details, see Paragraph 4.2 Section A. Management of common knowledge and IPR

New molecular targets, mouse models, probes, compounds, CNS active molecules, tests, imaging procedures, quantification protocols, and theranostic paradigms developed within INMiND are based on intellectual properties belonging to the developers. The exploitation of foreground and technologies generated by INMiND has been managed with the support of the STB under the lead of P4 and P13b together with the SMB and the PC. The STB was the central organisation for handling questions regarding intellectual property and for coordinating the knowledge management activities of the INMiND consortium. A coherent joint management of the INMiND knowledge arising from the jointly executed research ensured an appropriate use and dissemination of INMiND results in compliance with the general arrangements stipulated in the contract with the European Commission.
With regards to individual and jointly owned foreground being generated by INMiND, within the consortium agreement (based on the proposal of the DESCA group and according to industrial standards) the partners stipulated general and specific arrangements on ownership, confidentiality, protection, access rights, use and dissemination, and commercialization also in line with the terms of the EC contract and European guidelines.

The quality and success of the foreground and IPR management is attested by the following tables:
• list of IP developed (see Annex 3.1.2; Table 6)
• list of exploitable foreground and patents (see Paragraph 4.2 Section B)
List of Websites:
Project Coordinator
Prof. Dr. med. Andreas H. Jacobs
European Institute for Molecular Imaging (EIMI)
Westfälische Wilhelms-Universität Münster (WWU)
Waldeyerstr. 15
D-48149 Münster

Tel: +49 251 83 49300
Project website address: