This record is currently in review state, the data hasn’t been validated yet.
Modeling a dose-response relationship for thyroid cancer at radiotherapy dose levels
- Master Thesis
A general discrete formalism for generating models which can be used to estimate the excess absolute risk (EAR) of secondary cancer development after fractionated radiation is presented. The mathematical formulation describes the evolution of the number of original cells, repopulated cells and the resulting mutated cells with cancer potential and is presented as a set of undefined, but constrained functions. The first main assumption made is that one can evaluate the change in the population of cells on the temporal scale of the treatment by considering alternating phases of irradiation and repopulation. The second main assumption used is that the number of distinctly mutated cells with cancer potential is proportional to the EAR of developing a secondary cancer solely due to therapeutic radiation. In the second part, a model is created via the previously mentioned formalism, which is based on a linear quadratic model of cell survival with the same parameters for all types of cells. The repopulation term is designed such that the rate is proportional to the number of stem cells with a cut-off at full repopulation. The number of mutations (that are potentially carcinogenic) resulting from one fractionation is set to be proportional to the administered dose in the respective fraction. The resulting model is determined by four parameters, namely the well known linear quadratic coefficients, α, β, the low dose linear-no-threshold parameter, μ, that plays the role of a scaling factor, and a repopulation parameter, o, that contains information about the ratio of stem cells to differentiated cells and the average time required for cellular division. The new model along with the one developed by Schneider are applied to a set of EAR(D) data for the case of thyroid cancer obtained from the Childhood Cancer Survivor Study. The best sets of parameters for both models were found for the clinically relevant constraint of the linear quadratic coefficients of α/β=1 Gy. In the case of the new model the parameters were: μ=1.63 (/10000 PY/Gy), α=0.020 Gy-1, β=0.020 Gy-2 and o=0.047 and for the other model the best set was: μ=1.92 (/10000 PY/Gy), α=0.017 Gy-1, β=0.017 Gy-2 and ε≈0. The mutation parameters were larger, but comparable in value to μ=0.60 /10000 PY/Gy derived from the A-bomb survivors data. Finally the two models based on the previously presented sets of parameters were applied for estimating the treatment specific EAR for developing secondary cancer among Hodgkin’s disease treated patients. The estimates were 7.2/10000 PY for the model by Schneider and 6.4/10000 PY for the model developed during the thesis work. Both values overestimate the EAR reported in the literature of 3.8/10000 PY. Show more
External linksSearch via SFX
Subjectradiotherapy; second cancers; second malignancies
Organisational unit02010 - Dep. Physik / Dep. of Physics
MoreShow all metadata