This thesis focuses on the use of Monte Carlo (MC) codes in nuclear medicine for both imaging and absorbed dose (AD) calculation with the aim of a personalized dosimetry evaluation for patients undergoing Peptide Receptor Radionuclide Therapy (PRRT). Molecular RadioTherapy (MRT) exploits tumor-specific radiopharmaceuticals. PRRT is a specific type of MRT in which the pharmaceutical vector is a somatostatin analog peptide. To monitor the efficacy of a cancer therapy in dealing damage to tumors while preserving healthy tissues, a dosimetric evaluation is required. This procedure consists in evaluating the AD to the tissues of interest. In MRT the dosimetric procedure is cumbersome and the involved physical quantities are associated to high relative uncertainties. Thus, MC simulations can be a useful tool. After the bibliographic introductory chapters, the original work is presented. First, a study for the validation of SIMIND MC code to model a Siemens Symbia Intevo Excel SPECT-CT for 99mTc and 177Lu is shown. Phantom experiments using 99mTc and 177Lu were performed to measure spatial resolution, sensitivity, the calibration factor and recovery coefficients. The experimental results were compared with those obtained from MC simulations performed in the same geometries. Results are in good agreement. The following chapter presents the AD results of the FENET 2016, a phase II study applied at University Hospital S. Anna (Ferrara, Italy). Treatment protocol relies on five PRRT cycles with either 177Lu-DOTATOC or a sequence of 177Lu followed by 90Y-DOTATOC spaced two months apart. Tumors and organ-at-risk (OAR) dosimetry were performed on 80 patients using a simplified but patient-tailored dosimetric approach with OLINDA/EXM software. AD per administered activity with 177Lu resulted in 0.6±0.2Gy/GBq [0.1-1.4Gy/GBq] for kidneys, 0.02±0.01Gy/GBq [0.001-0.090 Gy/GBq] for bone marrow and 2.2±1.6Gy/GBq [0.1-9.4Gy/GBq] for tumors. Values obtained for OAR and tumors are in good agreement with data from literature. Then, the AD calculation for 10 patients with the use of Gate MC code is presented. The code was validated using the ICRP 110 standard phantoms to calculate specific absorbed fractions for monoenergetic photons and electrons and S-values for 177Lu. AD to kidneys and tumors was calculated using the patients SPECT and CT images. AD to kidneys are higher than those obtained with OLINDA, with ratios ranging from 0.96 to 1.52 (mean±SD = 1.24±0.17). AD to tumors have a more various behavior with ratios from 0.39 to 1.36 (0.94±0.28). Finally, the images of the same 10 patients were used for the AD calculation to kidneys and tumors with the voxel S-values method. Two softwares were considered: PLANETDose from DOSIsoft company and MIM SurePlan MRT from MIM Software company. AD to kidneys calculated with MIM result all compatible within the associated errors to those calculated with GATE (ratios from 0.83 to 1.14, 0.95±0.09). AD to tumors have a various behavior when comparing GATE with voxel S-values softwares, with ratios from 0.41 to 1.34 (1.01±0.32) for PlanetDose and from 0.58 to 1.07 (0.90±0.15) for MIM. All AD to tumors calculated with GATE and MIM are compatible within the associated errors. Final aim of the work is having a tool (SIMIND) able to simulate accurately the activity distribution inside the patient body, in order to calculate the AD to each organ from the reconstructed activity distribution and finally to compare the AD with the dose calculated from the true distribution (GATE) so as to implement patient-specific dosimetry. Future perspectives of the work will be the implementation of a homemade code for voxel-dosimetry using the voxel S-values approach, so as to overcome the limits of the organ-level dosimetry evaluation and the time-consuming requirements of MC direct calculations.
Questa tesi riguarda l’utilizzo dei codici Monte Carlo (MC) in medicina nucleare, sia per l’imaging che per il calcolo della dose assorbita (AD), con lo scopo di sviluppare una dosimetria personalizzata per i pazienti della terapia PRRT. La terapia radiometabolica (MRT) utilizza dei radiofarmaci specifici per i tumori. Del materiale radioattivo viene legato ad un farmaco cell-targeting che si attacca ad una cellula tumorale e rilascia una dose localizzata. La PRRT è un particolare tipo di MRT in cui il farmaco vettore è un analogo della somatostatina. Per monitorare l’efficacia di un trattamento tumorale nel danneggiare i tessuti malati e preservare quelli sani, si effettua una valutazione dosimetrica. Questa procedura consiste nel valutare l’AD ai tessuti di interesse; in MRT è particolarmente complessa, e le grandezze fisiche sono caratterizzate da incertezze elevate. Per questo, le simulazioni MC possono essere utili. Dopo dei paragrafi bibliografici introduttivi, si presenta il lavoro originale. Per primo, lo studio per la validazione del codice MC SIMIND nel modellare una Symbia Intevo Excel SPECT-CT con 99mTc and 177Lu. Degli esperimenti con un fantoccio sono stati effettuati per misurare la risoluzione spaziale, la sensibilità, il fattore di calibrazione ed i recovery coefficients. I risultati sperimentali sono stati confrontati con le simulazioni MC, risultando compatibili. Il capitolo successivo presenta i risultati di AD del FENET 2016, uno studio di fase II usato all’Ospedale Universitario di S.Anna (Ferrara). Il protocollo prevede 5 somministrazioni o tutte di 177Lu-DOTATOC o una sequenza di 177Lu intervallata da 90Y-DOTATOC a distanza di due mesi. La dosimetria è stata effettuata su 80 pazienti per i tumori e gli organi a rischio (OAR) usando una procedura personalizzata e semplificata con il software OLINDA/EXM. L’AD con il 177Lu è risultata in 0.6±0.2Gy/GBq [0.1-1.4Gy/GBq] per i reni, 0.02±0.01Gy/GBq [0.001-0.090 Gy/GBq] per il midollo e 2.2±1.6Gy/GBq [0.1-9.4Gy/GBq] per i tumori. I valori ottenuti sono compatibili con i dati di letteratura. Quindi, viene presentato il calcolo dell’AD per 10 dei pazienti con il Gate MC. Il codice è stato validato usando i fantocci standard ICRP 110 calcolando le specific absorbed fractions per fotoni ed elettroni monoenergetici e gli Svalues per il 177Lu. L’AD è stata quindi calcolata per reni e tumori usando le immagini SPECT e CT dei pazienti. Le AD per i reni risultano maggiori di quelle ottenute con OLINDA, con rapporti che vanno da 0.96 a 1.52 (media±SD = 1.24±0.17). L’AD per i tumori ha un comportamento più variabile, con rapporti da 0.39 a 1.36 (0.94±0.28). Infine, le immagini dei medesimi 10 pazienti sono state utilizzate per calcolare l’AD a reni e tumori con il metodo dei voxel S-values. Sono stati utilizzati due software: PLANETDose della DOSIsoft e MIM SurePlan MRT della MIM. L’AD ai reni calcolate con MIM sono rirultate tutte compatibili con quelle di GATE (rapporti tra 0.83 to 1.14, 0.95±0.09). L’AD ai tumori ha un comportamento più vario nel confronto tra GATE e i software di voxel S-values, con rapporti tra 0.41 e 1.34 (1.01±0.32) per PlanetDose e tra 0.58 e 1.07 (0.90±0.15) per MIM. Tutte le AD per i tumori calcolate con GATE e MIM sono compatibili entro gli errori. Obiettivo finale del lavoro è avere uno strumento (SIMIND) in grado di simulare accuratamente la distribuzione di attività all’interno del corpo del paziente, con l’obiettivo di calcolare l’AD ad ogni organo tramite l’attività ricostruita e confrontare l’AD con quella ottenuta dalla reale distribuzione di attività (tramite GATE) così da implementare una dosimetria personalizzata. Prospettive future del lavoro sono lo sviluppo di un codice per la voxel-dosimetry usando l’approccio voxel S-values, così da superare i limiti imposti dalla dosimetria a livello di organo e la time-consuming dosimetria con il calcolo diretto via MC.
Sviluppo di un Sistema di Dosimetria Personalizzato per la PRRT
LONGO, Lorenzo
2023
Abstract
This thesis focuses on the use of Monte Carlo (MC) codes in nuclear medicine for both imaging and absorbed dose (AD) calculation with the aim of a personalized dosimetry evaluation for patients undergoing Peptide Receptor Radionuclide Therapy (PRRT). Molecular RadioTherapy (MRT) exploits tumor-specific radiopharmaceuticals. PRRT is a specific type of MRT in which the pharmaceutical vector is a somatostatin analog peptide. To monitor the efficacy of a cancer therapy in dealing damage to tumors while preserving healthy tissues, a dosimetric evaluation is required. This procedure consists in evaluating the AD to the tissues of interest. In MRT the dosimetric procedure is cumbersome and the involved physical quantities are associated to high relative uncertainties. Thus, MC simulations can be a useful tool. After the bibliographic introductory chapters, the original work is presented. First, a study for the validation of SIMIND MC code to model a Siemens Symbia Intevo Excel SPECT-CT for 99mTc and 177Lu is shown. Phantom experiments using 99mTc and 177Lu were performed to measure spatial resolution, sensitivity, the calibration factor and recovery coefficients. The experimental results were compared with those obtained from MC simulations performed in the same geometries. Results are in good agreement. The following chapter presents the AD results of the FENET 2016, a phase II study applied at University Hospital S. Anna (Ferrara, Italy). Treatment protocol relies on five PRRT cycles with either 177Lu-DOTATOC or a sequence of 177Lu followed by 90Y-DOTATOC spaced two months apart. Tumors and organ-at-risk (OAR) dosimetry were performed on 80 patients using a simplified but patient-tailored dosimetric approach with OLINDA/EXM software. AD per administered activity with 177Lu resulted in 0.6±0.2Gy/GBq [0.1-1.4Gy/GBq] for kidneys, 0.02±0.01Gy/GBq [0.001-0.090 Gy/GBq] for bone marrow and 2.2±1.6Gy/GBq [0.1-9.4Gy/GBq] for tumors. Values obtained for OAR and tumors are in good agreement with data from literature. Then, the AD calculation for 10 patients with the use of Gate MC code is presented. The code was validated using the ICRP 110 standard phantoms to calculate specific absorbed fractions for monoenergetic photons and electrons and S-values for 177Lu. AD to kidneys and tumors was calculated using the patients SPECT and CT images. AD to kidneys are higher than those obtained with OLINDA, with ratios ranging from 0.96 to 1.52 (mean±SD = 1.24±0.17). AD to tumors have a more various behavior with ratios from 0.39 to 1.36 (0.94±0.28). Finally, the images of the same 10 patients were used for the AD calculation to kidneys and tumors with the voxel S-values method. Two softwares were considered: PLANETDose from DOSIsoft company and MIM SurePlan MRT from MIM Software company. AD to kidneys calculated with MIM result all compatible within the associated errors to those calculated with GATE (ratios from 0.83 to 1.14, 0.95±0.09). AD to tumors have a various behavior when comparing GATE with voxel S-values softwares, with ratios from 0.41 to 1.34 (1.01±0.32) for PlanetDose and from 0.58 to 1.07 (0.90±0.15) for MIM. All AD to tumors calculated with GATE and MIM are compatible within the associated errors. Final aim of the work is having a tool (SIMIND) able to simulate accurately the activity distribution inside the patient body, in order to calculate the AD to each organ from the reconstructed activity distribution and finally to compare the AD with the dose calculated from the true distribution (GATE) so as to implement patient-specific dosimetry. Future perspectives of the work will be the implementation of a homemade code for voxel-dosimetry using the voxel S-values approach, so as to overcome the limits of the organ-level dosimetry evaluation and the time-consuming requirements of MC direct calculations.File | Dimensione | Formato | |
---|---|---|---|
Tesi Longo Lorenzo.pdf
accesso aperto
Descrizione: Tesi
Tipologia:
Tesi di dottorato
Dimensione
9.16 MB
Formato
Adobe PDF
|
9.16 MB | Adobe PDF | Visualizza/Apri |
I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.