Razvoj populacionog biokinetičkog i dinamičkog modela radioaktivnog joda(131I) i primena u optimizaciji doziranja kod pacijenata sa benignim oboljenjem štitaste žlezde

Abstract

Cilj disertacije je bio da se primenom pristupa nelinearnog modelovanja kombinovanih efekata razvije populacioni model intratireoidne biokinetike 131I, kao i dinamički model verovatnoće ishoda terapije 131I kod pacijenata sa benignim oboljenjima štitaste žlezde. Podaci su prikupljeni retrospektivno iz medicinske dokumenatcije, a analiza je izvršena primenom programa NONMEM®. Kroz razvoj populacionog biokinetičkog modela 131I na osnovu rutinskih podataka o merenjima fiksacija 131I u štitastoj žlezdi nakon primene testne doza aktivnosti, ispitan je i kvantifikovan uticaj demografskih i kliničkih karakteristika pacijenata na biokinetiku 131I, kao i njena inter- i intraindividualna varijabilnost. Utvrđeno je da na brzinu preuzimanja 131I u štitastu žlezdu značajno utiču dijagnoza, starost, funkcionalni volumen štitaste žlezde, fT4, lečenje antitireoidnim lekovima, kao i vreme prekida terapije pre primene 131I, dok na brzinu eliminacije utiče starost pacijenta. Na osnovu podataka o ishodu terapije praćenog godinu dana nakon primene terapijske doze 131I razvijen je populacioni dinamički model proporcionalnih šansi. Pokazano je da od ispitivanih mera izloženosti biološki efektivna doza (BED) i apsorbovana doza zračenja statistički značajno bolje korelišu sa verovatnoćom ishodom u odnosu na dozu aktivnosti i maksimalnu brzinu apsorbovane doze, a BED je kao mera sa najnižom vrednošću Akaike informacionog kriterijuma uključena u finalni model. Na verovatnoću ishoda značajno utiče funkcionalni volumen štitaste žlezde. Rezultati ukazuju da bi primena BED formalizma mogla da doprinese personalizaciji terapije kroz individualizaciju vrednosti apsorbovane doze zračenja, kao i da bi ciljne vrednosti BED trebalo definisati za subpopulacije pacijenata sa različitim volumenom štitaste žlezde.

The dissertation aimed to develop a population model of intra-thyroidal 131I biokinetics as well as a dynamic model of the 131I therapy outcome probability in patients with benign thyroid diseases using the nonlinear mixed-effects modelling approach. Data were collected retrospectively from medical records, and the analysis was performed using the NONMEM® software. Through the development of the population biokinetic model 131I, based on routine data on 131I thyroidal uptake measurements after application of a tracer activity dose, the influence of patients' demographic and clinical characteristics on 131I biokinetics, as well as inter-, and intraindividual variability was examined and quantified. The rate of 131I uptake was significantly affected by diagnosis, functional thyroid volume, age, fT4, application of antithyroid drugs, and the therapy discontinuation time before 131I administration, while the rate of elimination was affected by the patient's age. The population dynamic proportional odds model was developed based on the data of therapy outcomes monitored one year after the application of the 131I therapeutic dose. It was found that, among the investigated exposure measures, the biologically effective dose (BED) and absorbed radiation dose correlated significantly better with the probability of an outcome than the activity dose and the maximum absorbed dose rate. As the measure with the lowest value of the Akaike information criterion, BED was included in the final model. The probability of the therapy outcome is significantly affected by the functional thyroid volume. The results indicate that the application of BED formalism could contribute to the personalization of therapy through individualization of the absorbed dose values and that the target values of BED should be defined for subpopulations of patients with different thyroid volumes.