Les progrès en imagerie fonctionnelle ouvrent de nouvelles perspectives dans la délimitation des volumes cibles en radiothérapie. Il est envisagé d’adapter les doses d’irradiation sur l’activité tumorale et de réaliser une escalade de dose. Nos objectifs étaient caractériser la détectabilité en TEP, en quantifiant les incertitudes de contour du GTV, de mettre en place la géométrie appropriée, d’évaluer l’impact dosimétrique de ce nouveau protocole et de vérifier la bonne délivrance de cette irradiation. 3 fantômes originaux et 2 fantômes virtuels avec les 3 niveaux de dose recherchés ont été spécialement conçus. Le diamètre de 1cm pour le niveau 3 a pu être atteint. Une escalade de 20Gy a été possible. L’impact dosimétrique sur deux cas réels était satisfaisant. Les tolérances cliniques ont été respectées. Toute la chaîne de traitement a été évaluée et validée. Cependant de telles doses doivent être évaluées avec précaution avant d’être prescrites et des progrès sont attendus en imagerie / The progress in functional imaging opens new perspectives in the delineation of target volumes in radiotherapy. We can intend to adapt the irradiation doses on the tumor activity and to perform a dose escalation. Our objectives were (i) to characterize the TEP thresholding, by quantifying the uncertainties of the target volume contour, (ii) to set up the geometry suited, (iii) to estimate the dosimetric impact of this new protocol and (iv) to verify that dosimetry is perfectly distributed. 3 original phantoms and 2 virtual phantoms containing 3 dose levels were specially created. The diameter of 1cm for the 3rd level was able to be reached. A dose escalation of 20Gy was possible. The dosimetric impact on two real cases was suitable. The clinical tolerances were respected. So all the treatment process was estimated and validated. However such doses should be carefully estimated before being prescribed clinically and progress is also expected in imaging

<p>In this thesis, a model of tissue oxygenation is presented, that takes into account the heterogeneous nature of tumor vasculature. Even though the model is rather simple, the resulting oxygen distributions agree very well with clinically observed oxygen distributions for most tumors and healthy normal tissues. The model shows that the vascular density may not describe the oxygenation of a tissue sufficiently well, unless the heterogeneity of the vascular system is taken into account. Based on the oxygen distributions from the tissue model, the associated radiation response at low and high doses can be determined. </p><p>The radiation response of heterogeneous tumors should preferably be described by two clonogen compartments, one resistant and one sensitive, dominating the response at high and low radiation doses, respectively. Furthermore, each compartment should be characterized by the effective radiation resistance and the effective clonogen number. The resistant-sensitive model of radiation response has been analyzed in great detail. It accurately describes the response of severely heterogeneous tumors, both at low and high doses and LET values. The effective response parameters are given as integrals, averaged over the whole spectrum of radiation resistance. The parameters can also be determined from clinically established dose-response relations. </p><p>The main properties of the dose-response relation for a generally heterogeneous tumor is described in some detail. The normalized dose-response gradient has been generalized to take heterogeneities in both dose delivery and radiation response into account. This quantity is important for accurate treatment plan optimization using intensity modulated radiation therapy for individual patients. </p>

In this thesis, a model of tissue oxygenation is presented, that takes into account the heterogeneous nature of tumor vasculature. Even though the model is rather simple, the resulting oxygen distributions agree very well with clinically observed oxygen distributions for most tumors and healthy normal tissues. The model shows that the vascular density may not describe the oxygenation of a tissue sufficiently well, unless the heterogeneity of the vascular system is taken into account. Based on the oxygen distributions from the tissue model, the associated radiation response at low and high doses can be determined. The radiation response of heterogeneous tumors should preferably be described by two clonogen compartments, one resistant and one sensitive, dominating the response at high and low radiation doses, respectively. Furthermore, each compartment should be characterized by the effective radiation resistance and the effective clonogen number. The resistant-sensitive model of radiation response has been analyzed in great detail. It accurately describes the response of severely heterogeneous tumors, both at low and high doses and LET values. The effective response parameters are given as integrals, averaged over the whole spectrum of radiation resistance. The parameters can also be determined from clinically established dose-response relations. The main properties of the dose-response relation for a generally heterogeneous tumor is described in some detail. The normalized dose-response gradient has been generalized to take heterogeneities in both dose delivery and radiation response into account. This quantity is important for accurate treatment plan optimization using intensity modulated radiation therapy for individual patients.