Modern radiation therapy techniques are capable to deliver dose within millimeter or even submillimeter precision. Highly accurate knowledge about position and shape of anatomical and pathological structures is a prerequisite to exploit the potential of these techniques. However, in current clinical practice treatment planning is still based on 3D CT imaging; movements of anatomical and pathological structures are not represented. Therefore uncertainties arising are accounted for by introducing safety margins (fig. 1). Margins are usually defined with respect to observations based on appropriate patient populations. This causes problems for the individual patient: On the one hand safety margins could be too large; this would increase the volume of healthy tissue to be irradiated which in turn would increase the likelihood of treatment related complications. On the other hand margins could be too small; this would limit the chances of successful treatment. Furthermore, for certain radiation delivery techniques such as intensity modulated radiotherapy the interplay of segmentation (i.e. application of small radiation fields of only a few monitor units) and target motion raises the risk of cold and hot spots (areas of high or low dose when compared to the planned dose).

The problems gain in importance with increasing motion amplitudes. Consequently, breathing motion - which causes motion amplitudes of up to several centimetres (fig. 2) - states a profound problem within the field of radiation therapy of thoracic tumors. By means of 4D CT data sets of lung tumor patients our project aims to answer fundamental questions within the field of radiation therapy of thoracic tumors. As a starting point we analyze breathing dynamics; related questions are here: What kind of motion patterns of anatomical and pathological structures exist? Do current safety margin concepts used in clinical practice cover the motion patterns? Is there a chance to improve the safety margin concepts by our observations?

In the next step we study the impact of breathing motion on the dose distribution applied to the patient; these analyses are done within a computer based simulation study. Different radiation delivery techniques are simulated such as 3D conformal radiation therapy and intensity modulated radiotherapy. For the different techniques the advantage of gated radiation delivery over ungated radiation delivery is examined.

This project is funded by the Deutsche Krebshilfe (DKH-Nr. 107899).