Stereotactic conformal radiotherapy for posterior fossa tumours: a modelling study for potential improvement in therapeutic ratio

Abstract

To investigate the optimal technique of stereotactic conformal radiotherapy (SCRT) for posterior fossa tumours. SCRT planning image data sets of four patients with posterior fossa tumours constituted the study material. Based on the operative notes and preoperative imaging, a clinical target volume (CTV) was drawn for each patient. An additional CTV representative of a typical posterior fossa tumour was also drawn on each patient's localisation scans. Therefore a total of eight CTVs differing in their size and location were used for further work. A margin of 5 mm was grown in three dimensions to result in a final planning target volume (PTV). Beam arrangements studied were conformal bilateral parallel pair, a three-field coplanar arrangement with two bilateral and a vertex beam, two three-field non-coplanar techniques and a six-field non-coplanar technique with conformation achieved by micromultileaf collimator (mMLC). Normal structures contoured included normal posterior fossa brain (excluding PTV), brain stem, cochleae, optic apparatus, pituitary-hypothalamic axis (PHA), supratentorial brain and the temporal lobes. Comparative evaluation of plans was done with dose volume histograms (DVH), conformity index (CI) and dose heterogeneity (DH). In all plans, the 95% isodose line covered at least 99% of the PTV with acceptable dose heterogeneity. As compared to the baseline bilateral conformal parallel pair plan, all other plans achieved significantly more sparing of the normal posterior fossa brain at the 95 and 80% dose prescription levels with superior CI (at 95% isodose). The six-field technique resulted in maximum sparing as compared to the bilateral plan with a mean additional sparing of 74% (46.2 cm3) and 55% (33.83) at 95 and 80% dose prescription level, respectively (P<0.0001). Among the three field plans, the technique of bilateral and vertex fields resulted in least doses to cochlea and also irradiated the least volume of brain stem and PHA. Comparison of this technique to the six-field technique did not show any significant difference in sparing of normal structures in the posteriorly placed tumours. In anteriorly placed tumours, the six-field technique was the most optimal irrespective of the size of the target volumes. A six-field plan SCRT achieves the most significant sparing of the normal tissues for localised irradiation in posterior fossa tumours, particularly for anteriorly located tumours. A relatively simpler three-field plan with bilateral and a vertex beam is equally good for tumours that are placed posteriorly and away from the brain stem.