Chimeric Antigen Receptor T-cell (CAR-T) therapy has emerged as a transformative approach in cancer immunotherapy, demonstrating remarkable efficacy in the treatment of hematological malignancies and leading to several FDA-approved therapies. However, its extension to solid tumors, particularly aggressive and immunologically complex malignancies such as glioblastoma (GB), remains a major challenge. Glioblastoma, the most common and lethal primary brain tumor in adults, is characterised by infiltrative growth, pronounced cellular heterogeneity, and a profoundly immunosuppressive tumor microenvironment, which collectively limit the efficacy of immune-based treatments. A critical barrier to the optimisation of CAR-T cell therapy for solid tumors is the inability to monitor the in vivo fate of infused therapeutic cells. Fundamental questions regarding their biodistribution, trafficking to the tumor site, persistence, and correlation with therapeutic response remain largely unanswered. This knowledge gap underscores an urgent need for non-invasive, quantitative imaging technologies capable of tracking the fate of therapeutic cells in real time. Among available imaging modalities, fluorine-19 magnetic resonance imaging (¹⁹F MRI) represents a particularly promising solution for cell tracking applications. Since the human body contains negligible amounts of endogenous fluorine, ¹⁹F MRI provides background-free, "hot-spot" images that allow for direct detection and quantification of ¹⁹F-labelled cells, eliminating the signal ambiguity that hampers conventional proton-based MRI. Furthermore, the ¹⁹F MRI signal intensity is directly proportional to the number of fluorine atoms, enabling absolute cell quantification. Fluorinated probes characterised by magnetically equivalent ¹⁹F nuclei have been developed to maximise signal sensitivity, among which PERFECTA, a recently designed fluorinated molecular probe containing 36 magnetically equivalent fluorine atoms, has shown highly promising properties for ¹⁹F MRI-based cell tracking. PERFECTA can be encapsulated within nanoparticles based on the biodegradable copolymer poly(lactic-co-glycolic acid) (PLGA), widely used in biomedical applications due to its established biocompatibility and tuneable degradation kinetics. In the present work, PERFECTA-loaded PLGA nanoparticles stabilised with sodium cholate (PERFECTA@PLGA-NaC) were developed and characterized with the aim of labelling human CAR-T cells targeting glioblastoma for potential ¹⁹F MRI tracking. The NPs were prepared by emulsification solvent evaporation, and fluorescently labelled with PLGA-rhodamine B to enable their detection. An important part of this work was dedicated to the comprehensive physicochemical characterisation of the nanoparticle formulation. Dynamic Light Scattering (DLS) analysis revealed an average hydrodynamic diameter of approximately 120 nm with a narrow size distribution (PDI < 0.2), indicating a homogeneous nanoparticle dispersion. Zeta Potential measurements confirmed a strongly negative surface charge (−43 mV), attributable to the sodium cholate stabiliser, which provides electrostatic repulsion and prevents aggregation in aqueous media. The encapsulation efficiency of the PERFECTA probe, determined by quantitative ¹⁹F Nuclear Magnetic Resonance (¹⁹F NMR) spectroscopy using trifluoroacetic acid as an external standard, was 41%, yielding a fluorine concentration of 1.8 × 10²⁰ ¹⁹F/mL. The colloidal stability of the nanoparticles in biologically relevant environments was investigated by time-dependent DLS and Zeta Potential measurements in three media: unconditioned T cell culture medium, medium conditioned by CAR-T cells, and serum-free medium. In both protein-containing media, a moderate increase in hydrodynamic diameter was observed within the first hour of incubation, followed by stabilisation at later time points (6 to 24 hours), suggesting rapid adsorption of serum proteins and formation of a protein corona, with subsequent reorganisation and compaction of the adsorbed layer. Crucially, no extensive aggregation or precipitation was observed under any condition tested, confirming that the nanoparticles maintain colloidal stability throughout the labelling incubation period. The protein corona composition was analysed by SDS-PAGE gel electrophoresis, which revealed relatively stable protein profiles across time points, while Fourier Transform Infrared (FTIR) spectroscopy confirmed that the nanoparticle core structure, including the PLGA-sodium cholate interface, remained intact during protein corona formation. The combined multi-technique analysis (DLS, SDS-PAGE, and FTIR) indicated that the dynamic processes occurring during protein corona maturation primarily involve reorganisation of the adsorbed protein layer rather than chemical modification of the nanoparticle constituents. The NP formulation was used for labelling experiments on human B7-H3-targeting CAR-T cells generated for glioblastoma immunotherapy. The labelling protocol was progressively optimised across thirteen independent experiments, evolving through three experimental formats — 24-well plates, 50 mL tubes, and T25 culture flasks — to address challenges related to cell adhesion, working volumes, and cell recovery. A central methodological challenge was the purification of labelled cells from free NPs, which was achieved through Ficoll density gradient centrifugation. While this approach successfully yielded pure samples, as confirmed by clean ¹⁹F NMR spectra showing exclusively cell-associated fluorine signal, it introduced substantial cell losses with overall recovery rates of approximately 2–3%. Despite the limited cell recovery, comprehensive characterisation of labelled CAR-T cells yielded highly encouraging results. Flow cytometry analysis demonstrated that cell viability remained consistently high following the labelling and purification protocol (average 95.0 ± 5.9%), confirming the biocompatibility of the PERFECTA@PLGA-NaC formulation. Rhodamine B fluorescence analysis revealed remarkably efficient and homogeneous NP uptake across the cell population, with 99.1 ± 1.3% of cells showing positive fluorescence, while control cells exhibited negligible background (0.004 ± 0.009%). Quantitative ¹⁹F NMR spectroscopy established an average cellular fluorine loading of 8.08 ± 5.36 × 10¹¹ ¹⁹F atoms per cell at standard labelling conditions (3.13 × 10¹⁹ ¹⁹F/mL), representing a 4–8 fold improvement over previously published CAR-T cell labelling studies using perfluorocarbon-based agents. Exploratory experiments performed at an elevated nanoparticle concentration (approximately 3-fold increase) demonstrated that cellular loading could be further increased to 2.25 × 10¹³ ¹⁹F/cell — a 28-fold enhancement — while maintaining high viability (97.5%), though systematic characterisation was limited by cell recovery challenges. Finally, functional co-culture experiments with U87 glioblastoma neurospheres were performed to verify that NP-labelled CAR-T cells retained their cytotoxic activity. After 3 days of co-culture at a 1:1 effector-to-target ratio, labelled CAR-T cells achieved substantial tumour cell elimination (73% T cells, 23% tumour cells), comparable to unlabelled CAR-T controls (85% T cells, 12% tumour cells), and markedly superior to non-transduced T cells lacking the CAR construct (64% T cells, 32% tumour cells). These results demonstrate that nanoparticle labelling does not fundamentally impair the recognition, binding, and killing mechanisms essential for CAR-T cell therapeutic function, nor does it compromise antigen-driven proliferation capacity. The results obtained in the present work demonstrate the feasibility of using PERFECTA@PLGA-NaC nanoparticles for fluorine-based labelling of CAR-T cells with high efficiency and preserved therapeutic functionality, establishing a promising platform for potential ¹⁹F MRI tracking of adoptive cell therapies. However, important limitations remain, most notably the low cell recovery following Ficoll purification, which prevented the acquisition of ¹⁹F MRI images in this study. Future work should prioritise improving purification efficiency through refined collection techniques or alternative separation strategies to enable recovery of cell numbers adequate for MRI-based validation. Once reliable cell recovery is achieved, phantom imaging studies will be essential to experimentally confirm the detectability predicted by the high fluorine loadings obtained. Additionally, protocol standardisation and in vivo validation in preclinical glioblastoma models will be necessary steps toward clinical translation of this cell tracking platform.
La terapia con cellule T dotate di recettore chimerico per l'antigene (CAR-T) ha rappresentato una svolta nell'immunoterapia oncologica, dimostrando un'efficacia straordinaria nel trattamento delle neoplasie ematologiche e portando all'approvazione di diverse terapie da parte della FDA. Tuttavia, la sua estensione ai tumori solidi, in particolare a neoplasie aggressive e immunologicamente complesse come il glioblastoma (GB), rimane una sfida considerevole. Il glioblastoma, il tumore cerebrale primario più comune e letale nell'adulto, è caratterizzato da crescita infiltrativa, marcata eterogeneità cellulare e un microambiente tumorale profondamente immunosoppressivo, che nell'insieme limitano l'efficacia dei trattamenti basati sull'immunoterapia. Un ostacolo critico all'ottimizzazione della terapia CAR-T per i tumori solidi è l'impossibilità di monitorare in vivo il destino delle cellule terapeutiche infuse. Domande fondamentali riguardanti la loro biodistribuzione, la migrazione verso il sito tumorale, la persistenza e la correlazione con la risposta terapeutica restano in larga parte senza risposta. Questa lacuna evidenzia un'urgente necessità di tecnologie di imaging non invasive e quantitative capaci di tracciare il destino delle cellule terapeutiche in tempo reale. Tra le modalità di imaging disponibili, la risonanza magnetica al fluoro-19 (¹⁹F MRI) rappresenta una soluzione particolarmente promettente per le applicazioni di tracciamento cellulare. Poiché il corpo umano contiene quantità trascurabili di fluoro endogeno, la ¹⁹F MRI fornisce immagini prive di segnale di fondo (cosiddette "hot-spot"), che consentono la rilevazione diretta e la quantificazione delle cellule marcate con ¹⁹F, eliminando l'ambiguità di segnale che limita la risonanza magnetica convenzionale basata su protoni. Inoltre, l'intensità del segnale ¹⁹F MRI è direttamente proporzionale al numero di atomi di fluoro, permettendo una quantificazione cellulare assoluta. Sonde fluorurate caratterizzate da nuclei ¹⁹F magneticamente equivalenti sono state sviluppate per massimizzare la sensibilità del segnale; tra queste, PERFECTA, una sonda molecolare fluorurata di recente concezione contenente 36 atomi di fluoro magneticamente equivalenti, ha mostrato proprietà altamente promettenti per il tracciamento cellulare tramite ¹⁹F MRI. PERFECTA può essere incapsulata all'interno di nanoparticelle basate sul copolimero biodegradabile acido poli(lattico-co-glicolico) (PLGA), ampiamente utilizzato nelle applicazioni biomediche grazie alla sua comprovata biocompatibilità e alle cinetiche di degradazione modulabili. Nel presente lavoro, nanoparticelle di PLGA caricate con PERFECTA e stabilizzate con sodio colato (PERFECTA@PLGA-NaC) sono state sviluppate e caratterizzate con l'obiettivo di marcare cellule CAR-T umane dirette contro il glioblastoma per un potenziale tracciamento tramite ¹⁹F MRI. Le nanoparticelle sono state preparate mediante emulsificazione ed evaporazione del solvente e marcate fluorescentemente con PLGA-rodamina B per consentirne la rilevazione. Una parte importante del lavoro è stata dedicata alla caratterizzazione fisico-chimica completa della formulazione nanoparticellare. L'analisi tramite Dynamic Light Scattering (DLS) ha evidenziato un diametro idrodinamico medio di circa 120 nm con una distribuzione dimensionale ristretta (PDI < 0,2), indicativa di una dispersione nanoparticellare omogenea. Le misure di Zeta Potenziale hanno confermato una carica superficiale fortemente negativa (−43 mV), attribuibile allo stabilizzante sodio colato, che garantisce repulsione elettrostatica e previene l'aggregazione in mezzo acquoso. L'efficienza di incapsulamento della sonda PERFECTA, determinata tramite spettroscopia ¹⁹F NMR quantitativa utilizzando acido trifluoroacetico come standard esterno, è risultata pari al 41%, con una concentrazione di fluoro di 1,8 × 10²⁰ ¹⁹F/mL. La stabilità colloidale delle nanoparticelle in ambienti biologicamente rilevanti è stata studiata mediante misure DLS e di Zeta Potenziale in funzione del tempo in tre diversi mezzi: terreno di coltura per cellule T non condizionato, terreno condizionato da cellule CAR-T e terreno privo di siero. In entrambi i mezzi contenenti proteine, un moderato aumento del diametro idrodinamico è stato osservato entro la prima ora di incubazione, seguito da stabilizzazione ai tempi successivi (6 e 24 ore), suggerendo un rapido adsorbimento di proteine sieriche e la formazione di una corona proteica, con successiva riorganizzazione e compattamento dello strato adsorbito. In nessuna delle condizioni testate è stata osservata aggregazione estesa o precipitazione, confermando il mantenimento della stabilità colloidale durante l'intero periodo di incubazione. La composizione della corona proteica è stata analizzata mediante elettroforesi SDS-PAGE, che ha rivelato profili proteici relativamente stabili nel tempo, mentre la spettroscopia infrarossa a trasformata di Fourier (FTIR) ha confermato che la struttura interna delle nanoparticelle, inclusa l'interfaccia PLGA-sodio colato, si è mantenuta intatta durante la formazione della corona. L'analisi combinata multi-tecnica (DLS, SDS-PAGE e FTIR) ha indicato che i processi dinamici che avvengono durante la maturazione della corona proteica coinvolgono principalmente la riorganizzazione dello strato proteico adsorbito piuttosto che modifiche chimiche dei costituenti nanoparticellari. La formulazione nanoparticellare è stata impiegata per esperimenti di marcatura su cellule CAR-T umane dirette contro l'antigene B7-H3, generate per l'immunoterapia del glioblastoma. Il protocollo di marcatura è stato progressivamente ottimizzato attraverso tredici esperimenti indipendenti, evolvendo in tre formati sperimentali — piastre da 24 pozzetti, tubi da 50 mL e flask T25 — per affrontare le criticità legate all'adesione cellulare, ai volumi di lavoro e al recupero cellulare. Una sfida metodologica centrale è stata la purificazione delle cellule marcate dalle nanoparticelle libere, ottenuta mediante centrifugazione su gradiente di densità con Ficoll. Sebbene questo approccio abbia prodotto campioni puri, come confermato dagli spettri ¹⁹F NMR che mostravano esclusivamente segnale di fluoro associato alle cellule, esso ha introdotto perdite cellulari sostanziali con tassi di recupero complessivi di circa il 2–3%. Nonostante il limitato recupero cellulare, la caratterizzazione completa delle cellule CAR-T marcate ha prodotto risultati molto incoraggianti. L'analisi tramite citometria a flusso ha dimostrato che la vitalità cellulare si è mantenuta costantemente elevata dopo il protocollo di marcatura e purificazione (media 95,0 ± 5,9%), confermando la biocompatibilità della formulazione PERFECTA@PLGA-NaC. L'analisi della fluorescenza della rodamina B ha rivelato un'internalizzazione delle nanoparticelle notevolmente efficiente e omogenea nella popolazione cellulare, con il 99,1 ± 1,3% delle cellule positive alla fluorescenza, mentre le cellule di controllo hanno mostrato un fondo trascurabile (0,004 ± 0,009%). La spettroscopia ¹⁹F NMR quantitativa ha stabilito un carico medio di fluoro cellulare di 8,08 ± 5,36 × 10¹¹ atomi di ¹⁹F per cellula alle condizioni standard di marcatura (3,13 × 10¹⁹ ¹⁹F/mL), rappresentando un miglioramento di 4–8 volte rispetto a studi precedentemente pubblicati sulla marcatura di cellule CAR-T con agenti a base di perfluorocarburi. Esperimenti esplorativi condotti a concentrazione elevata di nanoparticelle (incremento di circa 3 volte) hanno dimostrato che il carico cellulare può essere ulteriormente aumentato fino a 2,25 × 10¹³ ¹⁹F/cellula — un aumento di 28 volte — mantenendo un'elevata vitalità (97,5%), sebbene la caratterizzazione sistematica sia stata limitata dalle difficoltà di recupero cellulare. Infine, sono stati condotti esperimenti funzionali di co-coltura con neurosfere di glioblastoma U87 per verificare che le cellule CAR-T marcate con nanoparticelle conservassero la propria attività citotossica. Dopo 3 giorni di co-coltura con rapporto effettore-bersaglio 1:1, le cellule CAR-T marcate hanno raggiunto una sostanziale eliminazione delle cellule tumorali (73% cellule T, 23% cellule tumorali), comparabile ai controlli CAR-T non marcati (85% cellule T, 12% cellule tumorali) e nettamente superiore rispetto alle cellule T non trasdotte prive del costrutto CAR (64% cellule T, 32% cellule tumorali). Questi risultati dimostrano che la marcatura con nanoparticelle non compromette fondamentalmente i meccanismi di riconoscimento, legame ed eliminazione essenziali per la funzione terapeutica delle cellule CAR-T, né la loro capacità di proliferazione antigene-dipendente. I risultati ottenuti nel presente lavoro dimostrano la fattibilità dell'utilizzo delle nanoparticelle PERFECTA@PLGA-NaC per la marcatura a base di fluoro di cellule CAR-T con elevata efficienza e funzionalità terapeutica preservata, stabilendo una piattaforma promettente per il potenziale tracciamento delle terapie cellulari adottive tramite ¹⁹F MRI. Tuttavia, permangono limitazioni importanti, in primo luogo il basso recupero cellulare dopo purificazione con Ficoll, che ha impedito l'acquisizione di immagini ¹⁹F MRI in questo studio. I lavori futuri dovranno prioritizzare il miglioramento dell'efficienza di purificazione attraverso tecniche di raccolta perfezionate o strategie di separazione alternative, per consentire il recupero di numeri cellulari adeguati alla validazione mediante MRI. Una volta ottenuto un recupero cellulare affidabile, studi di imaging su phantom saranno essenziali per confermare sperimentalmente la rilevabilità predetta dagli elevati carichi di fluoro ottenuti. Inoltre, la standardizzazione del protocollo e la validazione in vivo in modelli preclinici di glioblastoma rappresenteranno i passi necessari verso la traslazione clinica di questa piattaforma di tracciamento cellulare.
Design and characterization of fluorinated nanoparticles for CAR-T cell labelling and 19F MRI tracking in glioblastoma therapy
Lobello, Raffaella
2025/2026
Abstract
Chimeric Antigen Receptor T-cell (CAR-T) therapy has emerged as a transformative approach in cancer immunotherapy, demonstrating remarkable efficacy in the treatment of hematological malignancies and leading to several FDA-approved therapies. However, its extension to solid tumors, particularly aggressive and immunologically complex malignancies such as glioblastoma (GB), remains a major challenge. Glioblastoma, the most common and lethal primary brain tumor in adults, is characterised by infiltrative growth, pronounced cellular heterogeneity, and a profoundly immunosuppressive tumor microenvironment, which collectively limit the efficacy of immune-based treatments. A critical barrier to the optimisation of CAR-T cell therapy for solid tumors is the inability to monitor the in vivo fate of infused therapeutic cells. Fundamental questions regarding their biodistribution, trafficking to the tumor site, persistence, and correlation with therapeutic response remain largely unanswered. This knowledge gap underscores an urgent need for non-invasive, quantitative imaging technologies capable of tracking the fate of therapeutic cells in real time. Among available imaging modalities, fluorine-19 magnetic resonance imaging (¹⁹F MRI) represents a particularly promising solution for cell tracking applications. Since the human body contains negligible amounts of endogenous fluorine, ¹⁹F MRI provides background-free, "hot-spot" images that allow for direct detection and quantification of ¹⁹F-labelled cells, eliminating the signal ambiguity that hampers conventional proton-based MRI. Furthermore, the ¹⁹F MRI signal intensity is directly proportional to the number of fluorine atoms, enabling absolute cell quantification. Fluorinated probes characterised by magnetically equivalent ¹⁹F nuclei have been developed to maximise signal sensitivity, among which PERFECTA, a recently designed fluorinated molecular probe containing 36 magnetically equivalent fluorine atoms, has shown highly promising properties for ¹⁹F MRI-based cell tracking. PERFECTA can be encapsulated within nanoparticles based on the biodegradable copolymer poly(lactic-co-glycolic acid) (PLGA), widely used in biomedical applications due to its established biocompatibility and tuneable degradation kinetics. In the present work, PERFECTA-loaded PLGA nanoparticles stabilised with sodium cholate (PERFECTA@PLGA-NaC) were developed and characterized with the aim of labelling human CAR-T cells targeting glioblastoma for potential ¹⁹F MRI tracking. The NPs were prepared by emulsification solvent evaporation, and fluorescently labelled with PLGA-rhodamine B to enable their detection. An important part of this work was dedicated to the comprehensive physicochemical characterisation of the nanoparticle formulation. Dynamic Light Scattering (DLS) analysis revealed an average hydrodynamic diameter of approximately 120 nm with a narrow size distribution (PDI < 0.2), indicating a homogeneous nanoparticle dispersion. Zeta Potential measurements confirmed a strongly negative surface charge (−43 mV), attributable to the sodium cholate stabiliser, which provides electrostatic repulsion and prevents aggregation in aqueous media. The encapsulation efficiency of the PERFECTA probe, determined by quantitative ¹⁹F Nuclear Magnetic Resonance (¹⁹F NMR) spectroscopy using trifluoroacetic acid as an external standard, was 41%, yielding a fluorine concentration of 1.8 × 10²⁰ ¹⁹F/mL. The colloidal stability of the nanoparticles in biologically relevant environments was investigated by time-dependent DLS and Zeta Potential measurements in three media: unconditioned T cell culture medium, medium conditioned by CAR-T cells, and serum-free medium. In both protein-containing media, a moderate increase in hydrodynamic diameter was observed within the first hour of incubation, followed by stabilisation at later time points (6 to 24 hours), suggesting rapid adsorption of serum proteins and formation of a protein corona, with subsequent reorganisation and compaction of the adsorbed layer. Crucially, no extensive aggregation or precipitation was observed under any condition tested, confirming that the nanoparticles maintain colloidal stability throughout the labelling incubation period. The protein corona composition was analysed by SDS-PAGE gel electrophoresis, which revealed relatively stable protein profiles across time points, while Fourier Transform Infrared (FTIR) spectroscopy confirmed that the nanoparticle core structure, including the PLGA-sodium cholate interface, remained intact during protein corona formation. The combined multi-technique analysis (DLS, SDS-PAGE, and FTIR) indicated that the dynamic processes occurring during protein corona maturation primarily involve reorganisation of the adsorbed protein layer rather than chemical modification of the nanoparticle constituents. The NP formulation was used for labelling experiments on human B7-H3-targeting CAR-T cells generated for glioblastoma immunotherapy. The labelling protocol was progressively optimised across thirteen independent experiments, evolving through three experimental formats — 24-well plates, 50 mL tubes, and T25 culture flasks — to address challenges related to cell adhesion, working volumes, and cell recovery. A central methodological challenge was the purification of labelled cells from free NPs, which was achieved through Ficoll density gradient centrifugation. While this approach successfully yielded pure samples, as confirmed by clean ¹⁹F NMR spectra showing exclusively cell-associated fluorine signal, it introduced substantial cell losses with overall recovery rates of approximately 2–3%. Despite the limited cell recovery, comprehensive characterisation of labelled CAR-T cells yielded highly encouraging results. Flow cytometry analysis demonstrated that cell viability remained consistently high following the labelling and purification protocol (average 95.0 ± 5.9%), confirming the biocompatibility of the PERFECTA@PLGA-NaC formulation. Rhodamine B fluorescence analysis revealed remarkably efficient and homogeneous NP uptake across the cell population, with 99.1 ± 1.3% of cells showing positive fluorescence, while control cells exhibited negligible background (0.004 ± 0.009%). Quantitative ¹⁹F NMR spectroscopy established an average cellular fluorine loading of 8.08 ± 5.36 × 10¹¹ ¹⁹F atoms per cell at standard labelling conditions (3.13 × 10¹⁹ ¹⁹F/mL), representing a 4–8 fold improvement over previously published CAR-T cell labelling studies using perfluorocarbon-based agents. Exploratory experiments performed at an elevated nanoparticle concentration (approximately 3-fold increase) demonstrated that cellular loading could be further increased to 2.25 × 10¹³ ¹⁹F/cell — a 28-fold enhancement — while maintaining high viability (97.5%), though systematic characterisation was limited by cell recovery challenges. Finally, functional co-culture experiments with U87 glioblastoma neurospheres were performed to verify that NP-labelled CAR-T cells retained their cytotoxic activity. After 3 days of co-culture at a 1:1 effector-to-target ratio, labelled CAR-T cells achieved substantial tumour cell elimination (73% T cells, 23% tumour cells), comparable to unlabelled CAR-T controls (85% T cells, 12% tumour cells), and markedly superior to non-transduced T cells lacking the CAR construct (64% T cells, 32% tumour cells). These results demonstrate that nanoparticle labelling does not fundamentally impair the recognition, binding, and killing mechanisms essential for CAR-T cell therapeutic function, nor does it compromise antigen-driven proliferation capacity. The results obtained in the present work demonstrate the feasibility of using PERFECTA@PLGA-NaC nanoparticles for fluorine-based labelling of CAR-T cells with high efficiency and preserved therapeutic functionality, establishing a promising platform for potential ¹⁹F MRI tracking of adoptive cell therapies. However, important limitations remain, most notably the low cell recovery following Ficoll purification, which prevented the acquisition of ¹⁹F MRI images in this study. Future work should prioritise improving purification efficiency through refined collection techniques or alternative separation strategies to enable recovery of cell numbers adequate for MRI-based validation. Once reliable cell recovery is achieved, phantom imaging studies will be essential to experimentally confirm the detectability predicted by the high fluorine loadings obtained. Additionally, protocol standardisation and in vivo validation in preclinical glioblastoma models will be necessary steps toward clinical translation of this cell tracking platform.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_03_Lobello_Executive Summary.pdf
accessibile in internet per tutti
Descrizione: Executive Summary
Dimensione
1.25 MB
Formato
Adobe PDF
|
1.25 MB | Adobe PDF | Visualizza/Apri |
|
2026_03_Lobello.pdf
accessibile in internet per tutti
Descrizione: Thesis text
Dimensione
7.48 MB
Formato
Adobe PDF
|
7.48 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/253315