Convection-Enhanced Delivery (CED) has been recently introduced as a promising surgical technique to bypass the blood-brain barrier and inject a chemotherapeutic agent directly in the brain tissue. Although this technique was expected to be effective, especially against recurrent tumors, the clinical trials did not achieve the desired results in terms of life expectancy for the patients. A major impairment to progress is given by the fact that the cancerous areas are usually not reached by a sufficiently high concentration of drug. Indeed, since the brain is an anisotropic and heterogeneous porous medium, for the clinicians it is very difficult to set the infusion in the best way possible and often the drug misses the target area. To tackle this issue, researchers have worked on predictive numerical models that can offer the surgeons a simulation environment to test different infusion settings. Despite these models are extremely valuable, their predictive capability is still not sufficiently accurate thus preventing their use in standard clinical practice. In the present contribution, we develop an extensive study that starts from a detailed analysis of the brain microstructure, with particular emphasis on the white matter (WM) permeability, and finishes with the integration of the acquired information in a CED predictive model at the macroscale. This computational model exploits a relatively new imaging technique, namely the Neurite Orientation Dispersion and Density Imaging, to incorporate a more advanced and comprehensive description of the brain microstructure. We demonstrate the relevance of the work by showing the impact on the predicted drug distribution, which differs significantly from the state-of-the-art model in terms of distribution shape, concentration profile and infusion linear penetration length.
La procedura CED (Convection-Enhanced Delivery) è stata recentemente introdotta come una promettente tecnica chirurgica per bypassare la barriera emato-encefalica e iniettare un agente chemioterapico direttamente nel tessuto cerebrale. Sebbene ci si aspettasse che questa tecnica fosse efficace, soprattutto contro i tumori ricorrenti, gli studi clinici non hanno ottenuto i risultati desiderati in termini di aspettativa di vita dei pazienti. Un grave ostacolo al progresso è dato dal fatto che le aree cancerose non sono solitamente raggiunte da una concentrazione di farmaco sufficientemente elevata. Infatti, poiché il cervello è un mezzo poroso anisotropo ed eterogeneo, per i medici è molto difficile impostare i parametri di infusione nel miglior modo possibile e spesso il farmaco manca l'area desiderata. Per affrontare questo problema, i ricercatori hanno lavorato allo sviluppo di modelli numerici predittivi in grado di offrire ai chirurghi un ambiente di simulazione per testare diversi scenari di infusione. Nonostante questi modelli siano estremamente complessi, la loro capacità predittiva non è ancora sufficientemente accurata, impedendo così il loro utilizzo nella pratica clinica standard. La presente dissertazione sviluppa un ampio studio che parte da un'analisi dettagliata della microstruttura cerebrale, con particolare attenzione alla permeabilità della materia bianca, e termina con l'integrazione delle informazioni acquisite in un modello predittivo CED alla macroscala. Questo modello computazionale sfrutta una tecnica di imaging relativamente nuova, ovvero la Neurite Orientation Dispersion and Density Imaging, per incorporare una descrizione più avanzata e completa della microstruttura cerebrale. Si dimostra la rilevanza del lavoro svolto, mostrando l'impatto sulla distribuzione prevista del farmaco, che differisce significativamente dai modelli presenti nello stato dell'arte in termini di volume di distribuzione, profilo di concentrazione e profondità di penetrazione dell'infusione.
A combined experimental and numerical approach towards a comprehensive drug delivery model
VIDOTTO, MARCO
Abstract
Convection-Enhanced Delivery (CED) has been recently introduced as a promising surgical technique to bypass the blood-brain barrier and inject a chemotherapeutic agent directly in the brain tissue. Although this technique was expected to be effective, especially against recurrent tumors, the clinical trials did not achieve the desired results in terms of life expectancy for the patients. A major impairment to progress is given by the fact that the cancerous areas are usually not reached by a sufficiently high concentration of drug. Indeed, since the brain is an anisotropic and heterogeneous porous medium, for the clinicians it is very difficult to set the infusion in the best way possible and often the drug misses the target area. To tackle this issue, researchers have worked on predictive numerical models that can offer the surgeons a simulation environment to test different infusion settings. Despite these models are extremely valuable, their predictive capability is still not sufficiently accurate thus preventing their use in standard clinical practice. In the present contribution, we develop an extensive study that starts from a detailed analysis of the brain microstructure, with particular emphasis on the white matter (WM) permeability, and finishes with the integration of the acquired information in a CED predictive model at the macroscale. This computational model exploits a relatively new imaging technique, namely the Neurite Orientation Dispersion and Density Imaging, to incorporate a more advanced and comprehensive description of the brain microstructure. We demonstrate the relevance of the work by showing the impact on the predicted drug distribution, which differs significantly from the state-of-the-art model in terms of distribution shape, concentration profile and infusion linear penetration length.File | Dimensione | Formato | |
---|---|---|---|
Doctoral_Thesis_Marco_Vidotto_final.pdf
Open Access dal 10/06/2021
Descrizione: Tesi di dottorato Marco Vidotto
Dimensione
9.03 MB
Formato
Adobe PDF
|
9.03 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/10589/169159