Abstract

Purpose:Multi‐parametric MRI (mp‐MRI) is being introduced in radiotherapy (RT) of prostate cancer, including for tumour delineation in focal boosting strategies. We recently developed an image‐based tumour control probability model, based on cell density distributions derived from apparent diffusion coefficient (ADC) maps. Beyond tumour volume and cell densities, tumour hypoxia is also an important determinant of RT response. Since tissue perfusion from mp‐MRI has been related to hypoxia we have explored the patterns of ADC and perfusion maps, and the relations between them, inside and outside prostate index lesions.Methods:ADC and perfusion maps from 20 prostate cancer patients were used, with the prostate and index lesion delineated by a dedicated uro‐radiologist. To reduce noise, the maps were averaged over a 3×3×3 voxel cube. Associations between different ADC and perfusion histogram parameters within the prostate, inside and outside the index lesion, were evaluated with the Pearson's correlation coefficient. In the voxel‐wise analysis, scatter plots of ADC vs perfusion were analysed for voxels in the prostate, inside and outside of the index lesion, again with the associations quantified with the Pearson's correlation coefficient.Results:Overall ADC was lower inside the index lesion than in the normal prostate as opposed to ktrans that was higher inside the index lesion than outside. In the histogram analysis, the minimum ktrans was significantly correlated with the maximum ADC (Pearson=0.47; p=0.03). At the voxel level, 15 of the 20 cases had a statistically significant inverse correlation between ADC and perfusion inside the index lesion; ten of the cases had a Pearson < −0.4.Conclusion:The minimum value of ktrans across the tumour was correlated to the maximum ADC. However, on the voxel level, the ‘local’ ktrans in the index lesion is inversely (i.e. negatively) correlated to the ‘local’ ADC in most patients.Research agreement with Varian Medical Systems, not related to the work presented in this abstract.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call