Molecular Imaging Biomarkers and Theragnosis Lab
Molecular imaging represents a novel and growing medical discipline that enables the visualization, characterization, and quantification of biologic processes taking place at the cellular and molecular levels within living subjects. It is the only genuinely theragnostic medical discipline that integrates diagnosis, through the efficient detection of disease-specific molecules, and treatment, with therapies targeted by such disease-specific molecules. Our mission is to develop quantitative molecular imaging biomarkers and new theragnosis approaches for clinical and translational research, with a particular focus on the brain.
Research Lines
PET BIOMARKERS FOR EARLY DIAGNOSIS OF NEURODEGENERATIVE DISEASES
We develop brain PET/MR imaging biomarkers for early diagnosis of patients with neurodegenerative diseases, with a special focus on Alzheimer patients in pre-symptomatic stages.
PET BIOMARKERS AND THERAGNOSIS IN ALZHEIMER DISEASE
We are aimed at developing a new generation of radiopharmaceuticals and PET biomarkers to advance toward more targeted and personalized treatments in Alzheimer disease. Our strategy is based on the radiolabelling of monoclonal antibodies, nanobodies and fragments, that are being tested in animal models of Alzheimer disease.
PET BIOMARKERS AND RADIOTHERAGNOSIS IN BRAIN TUMOURS
The potential of molecular imaging lies in the fact that detected disease-specific molecules can be used both as diagnostic and targets for radionuclide therapies. We are aimed at developing new radiotheragnostic PET and SPECT radiopharmaceuticals for the treatment of brain tumours based on monoclonal antibodies specifically designed for BBB overcoming.
STANDARIZATION AND PET QUANTIFICATION METHODS
We use an original approach based on realistic in silico PET images using Monte Carlo techniques, which can be efficiently used for the validation of PET biomarkers using ground-truth reference values. We have made available for the community a web-based platform for the realistic simulation of PET studies (sim-pet.org). In terms of PET quantification methods, we are currently focused on Total-body PET scanners. This revolutionary technology allows acquire PET images of the whole-body, from head to toe all at the same time, enabling, for the first time in history, the generation of whole-body dynamic PET imaging studies.
Facilities and transfer
PRECLINICAL IMAGING FACILITY
Our group manages a preclinical imaging facility aimed at developing preclinical imaging biomarkers and radiolabelling procedures. First, our aim is to provide imaging endpoints (PET, SPECT, MRI and CT), which are directly translated to clinical trials, that can be used for key decision making throughout drug development. Second, we provide pharmacokinetic and biodistribution information of radiolabelled drugs using PET and SPECT, revealing information about target engagement and proof of mechanism. Our group has extensive experience implementing preclinical PET and SPECT quantification approaches and investigating main factors affecting imaging biomarkers, such as scan protocol design, physical effects, instrumentation, tomographic reconstruction, quantification method and tracer kinetics in different areas.
NEUROCLOUD (QUBIOTECH)
Our brain PET quantification methods are currently available for clinical use in a cloud-based platform (neurocloud.es), after technology transfer to the spin-off company Qubiotech Health Intelligence.
Members
Selected publications
Biodistribution and pharmacokinetics of [89Zr]-anti-VEGF mAbs using PET in glioblastoma rat models.
Increased Medial Temporal Tau Positron Emission Tomography Uptake in the Absence of Amyloid-β Positivity.
Quantitative PET tracking of intra-articularly administered 89Zr-peptide-decorated nanoemulsions.
Impact of spill-in counts from off-target regions on [18F]Flortaucipir PET quantification
Dose-response assessment of cerebral P-glycoprotein inhibition in vivo with [18F]MC225 and PET
18F-florbetapir PET as a marker of myelin integrity across the Alzheimer's disease spectrum.
Head-to-head comparison of (R)-[11C]verapamil and [18F]MC225 in non-human primates, tracers for measuring P-glycoprotein function.
SimPET-An open online platform for the Monte Carlo simulation of realistic brain PET data. Validation for 18 F-FDG scans.
[18F]-FMISO PET/MRI Imaging Shows Ischemic Tissue around Hematoma in Intracerebral Hemorrhage
Selected Results
PET BIOMARKERS FOR EARLY DIAGNOSIS OF NEURODEGENERATIVE DISEASES
PET BIOMARKERS FOR ALZHEIMER DISEASE
We have explored implications in the design of predictive models based on MR volumetric biomarkers (Moscoso et al. 2019 Neuroimage Clin) and investigated the ability of episodic memory to track cognitive changes in mild cognitive impairment patients (Moscoso et al. 2019 Neurobiol Aging). We carried out the first study showing that MRI-based white-matter hyperintensities are associated with future amyloid accumulation in cognitively normal elderly subjects. Our findings provided cortical regions in which amyloid accumulation is significantly associated to higher white-matter hyperintensities (Moscoso et al. Neuroimage 2020). Furthermore, we proposed a new biomarker of myelin integrity across the Alzheimer’s disease spectrum from Amyloid-PET images (Moscoso et al. EJNMMI 2022). We also evaluated longitudinally Tau-PET studies from amyloid-negative patients, showing that the tau accumulation is comparably slower compared to amyloid-positive patients, and tau uptake remains restricted to the temporal lobe (Costoya-Sanchez et al. JAMA Neurology 2023).
PET AND THERAGNOSIS IN ALZHEIMER DISEASE
We are aimed at developing new radiopharmaceuticals and PET biomarkers to advance toward more targeted and personalized treatments in Alzheimer disease. Our strategy is based on the radiolabelling of monoclonal antibodies and fragments, that are being tested in animal models of Alzheimer disease.
On the other hand, we have collaborated in the development of PET biomarkers for measuring P-glycoprotein function at the BBB. Our group participated in the extensive preclinical validation of a new PET radiotracer, consisting of dose-response studies in healthy rats (Garcia-Varela et al. J Controlled Release 2022), PET studies in non-human primates (Garcia-Varela et al. EJNNMMI 2021), inhibition and induction studies (Garcia-Varela et al. Mol Pharm 2021) and test-retest evaluation (Garcia-Varela et al. ACS Chem Neurosci 2020).
PET AND RADIOTHERAGNOSIS IN BRAIN TUMOURS
The potential of molecular imaging lies in the fact that detected disease-specific molecules can be used both as diagnostic and targets for radionuclide therapies. We are aimed at developing new theragnostic radiopharmaceuticals for brain tumours based on monoclonal antibodies (anti-VEGF, anti-EGFRvIII and anti-HER3) specifically designed for BBB overcoming, encapsulated in nanocarriers or other strategies. To date, anti-VEGF PET studies were conducted in a glioblastoma rat model to evaluate the ability of anti-VEGF antibodies to overcome the BBB (Rey-Bretal et al. EJNMMI 2021). We found that bevacizumab and aflibercept can adequately target the VEGF expressed but tumor VEGF expression is earlier detected by bevacizumab (Garcia-Varela et al. Int J Pharm 2024).
STANDARIZATION AND PET QUANTIFICATION METHODS
BRAIN PET/SPECT QUANTIFICATION METHODS.
In Alzheimer, we have developed and implemented different quantification pipelines for FDG-PET, Amyloid-PET and Tau-PET. In particular, we investigated the impact of intensity normalization scaling factors on FDG-PET quantification, showing they provide remarkable bias in the detected hypometabolism and serious concerns in terms of false positives (Lopez-Gonzalez et al. Neuroimage 2020). Furthermore, we investigated the impact of spill-in counts from white matter on the calculation of SUVRs using Florbetapir-PET (Lopez-Gonzalez et al. EJNMMI Physics 2020), and spill-in counts from off-target regions on the calculation of SUVRs using Flortaucipir-PET (Lopez-Gonzalez et al. Neuroimage 2022).
WHOLE-BODY PET/CT QUANTIFICATION METHODS.
We evaluated the effect of image artifacts in Choline-PET quantification of prostate cancer (Silva-Rodriguez et al. Phys Med Biol 2016) and we have implemented texture analysis methods for predicting the immunohistochemical expression and the subtype of the breast cancer from FDG-PET images (Moscoso et al. EJNMMI 2018). Furthermore, we have investigated the correlations between texture analysis with immunohistochemical factors, cancer subtype in overall survival in breast and lung cancer (Piñeiro-Fiel et al. Eur Radiol 2021) (Piñeiro-Fiel et al. Diagnostics 2021).
PRECLINICAL PET QUANTIFICATION METHODS.
We have studied preclinical brain FDG-PET images under multiple conditions such as benzodiazepines administration, prolonged fasting, non-fasting, eye activation or low room-temperature (Silva-Rodriguez et al. Nucl Med and Biology 2016). Furthermore, we have developed a double arterial-venous shunt strategy for measuring real-time arterial input functions in brain FDG-PET studies. Furthermore, we have implemented image-derived input function (IDIF) approaches based on the segmentation of the vena cava and the myocardium (Rey-Bretal et al. Mol Imaging 2020).
SIMULATION AND PET IMAGE RECONSTRUCTION.
In terms of image standardization, we use an original approach based on realistic in silico PET images using Monte Carlo techniques, which can be efficiently used for the validation of PET biomarkers using ground-truth reference values. To this end, we have made available for the community a web-based platform for the realistic simulation of brain PET studies based on the use of open-source SimSET and STIR packages (sim-pet.org), presenting the key features of the platform and validated PET scanners (Paredes-Pacheco et al. Med Phys 2021). On the other hand, we have also modelled preclinical PET scanners using Monte Carlo simulation and NEMA standards (Popota et al. IEEE Trans Nucl Med 2012). we have implemented tomographic reconstruction methods for PET systems that included advanced point spread function modelling (Aguiar et al. Med Phys 2010) and scatter correction methods (Tsoumpas et al. IEEE MIC 2004) (Tsoumpas et al. IEEE MIC 2005) (Polycarpou et al. Ann Nucl Med 2011). We have implemented reconstruction methods for SPECT systems (El Bitar et al. Phys Med Biol 2014)(Aguiar et al. Phys Med Biol 2014), which are currently implemented in a portable device for preclinical SPECT imaging (Aguiar et al. 2014 J Instrum) (Silva-Rodriguez et al. 2015 Phys Med). Part of our contributions are available for the community in the Open-Source project STIR version 2 (Thielemans et al. Phys Med Biol 2012), accessible via stir.sourceforge.net
Alliances
