Quantitative helical dynamic contrast enhanced computed tomography assessment of the spatial variation in whole tumour blood volume with radiotherapy in lung cancer

Abstract 

We aim to assess the spatial distribution of blood volume (BV) in whole lung tumours in patients undergoing radiotherapy using helical dynamic contrast enhanced computed tomography (DCE-CT), and to determine whether conventional single level, or whole tumour measurements is more representative of the vascular effects of radiotherapy. Following ethical approval and informed consent, 15 patients with histologically proven non-small cell lung cancer underwent paired helical DCE-CT studies at baseline to assess repeatability, and after two fractions of radiotherapy (9Gy total dose). Tumour BV was calculated for individual contiguous 10mm axial slices, and for the entire tumour volume on a pixel-per-pixel basis. Baseline tumour BV was heterogeneous varying by 15.33%±17.11 between adjacent 10mm axial slices. Within subject coefficient of variation was 36.72% with conventional single tumour level evaluation, and 13.62% with whole tumour measurements. Following radiotherapy, one patient had an increase in BV greater than baseline variation (derived from the 95% limits of change) using single level evaluation; in contrast, seven patients had an increase in BV when the whole tumour was assessed. As a group, following radiotherapy, mean BV increased by 17.27% (paired t-test, p=0.20) with single level evaluation and 19.26% (p=0.049) with whole tumour assessment. Tumour BV measured using DCE-CT is spatially heterogeneous. Given the slice-by-slice variation in blood volume, our results demonstrate that whole tumour DCE-CT measurements are more repeatable, and may be a better predictor of vascular changes following therapy, compared to conventional single tumour level evaluations.

Keywords: Non-small cell lung cancer, Functional imaging, Perfusion computed tomography, Dynamic contrast enhanced computed tomography, Radiotherapy, Tumour angiogenesis, Measurement reproducibility