Radiobiological modelling of clonogen distribution, hypoxic fraction and tumour size effects on local tumour control of non-small cell lung cancer

Abstract

Published clinical data show that hypoxia in human lung tumours can impede the establishment of optimum local tumour control. However, the overall effect of hypoxia on the tumour control probability (TCP) model is not clear. The focus of this project was to assess the influence of radiobiological parameters (the number of clonogens and the hypoxic fraction), as well as some treatment parameters (i.e., the tumour size), on local tumour control of early stage non-small cell lung cancer (NSCLC). A TCP model, based on LQ cell survival concept combined with the Poisson statistic, was established to predict one, two and three years of local tumour control. This TCP model was created using data from seventeen publications of early-stage NSCLC treated using one of the three radiotherapy modalities: three-dimensional conformal radiation therapy (3D-CRT), continuous hyperfractionated accelerated radiotherapy (CHART) or stereotactic ablative body radiotherapy (SABR). The variations in the TCP with the gross tumour volume (GTV) size, clonogen number and hypoxic fraction were then investigated. This issue was approached by varying the clonogen densities values (between 10̂1 and 10̂7 cm̂3), the GTV volume (20–140 cc) and the hypoxic fraction (20–90%). The optimum values used to compute the TCP model were a clonogen density of 10̂7 cm̂3 and a hypoxic fraction of 20%, which were consistent with the clinical outcome values reported in the literature for NSCLC. This radiobiological model has demonstrated the proof of concept that poor local tumour control is strongly associated with the hypoxic fraction and large tumours.