Radiation-induced regional cerebral blood volume (rCBV) changes in normal brain and low-grade astrocytomas: quantification and time and dose-dependent occurrence.

PURPOSE: New tumor-conformal radiation-treatment modalities have been established with the intention to spare normal tissue while maintaining or improving local tumor control. To document radiation-induced changes in normal brain and low-grade astrocytoma we measured regional cerebral blood volumes (rCBV) using a dynamic susceptibility-weighted contrast-enhanced MR technique (DSC-MRI). We attempted to assess pretherapeutic rCBV values and time- and dose-dependent changes following radiotherapy. METHODS AND MATERIALS: For prospective and longitudinal assessment of rCBV in normal brain and low-grade astrocytoma, 25 patients with histologically proven fibrillary astrocytoma (WHO Grade II) were examined before radiotherapy and during follow-up. Based on CT- and MR-data sets in a stereotactic setup, three-dimensional (3D) treatment planning was done. Radiotherapy was delivered using fractionated stereotactic radiotherapy (FSRT) to mean and median total doses of 60.9 and 60 Gy, respectively (range, 55.8-66 Gy). During MR imaging for treatment planning and follow-up examinations, 55 T2-weighted gradient echo images were acquired before, during, and after intravenous contrast bolus injection. The acquired signal-time curves were converted into concentration-time curves. The area under the tissue concentration-time curve was calculated and normalized to an integrated arterial input function. Thus, absolute rCBV values could be calculated.
RESULTS: Pretherapeutic mean rCBV for normal gray (GM) and white brain matter (WM) were 7.2 +/- 2.7 and 3.6 +/- 1.5 mL/100 g tissue, respectively. Mean rCBV for astrocytoma was 6.5 +/- 3.7 mL/100 g tissue. After radiotherapy, rCBV for GM and WM was significantly reduced (p < 0.01) in high-dose areas (40-100% of total dose). A nonsignificant reduction was measured in low-dose areas (up to 40% of total dose). Reduction of rCBV in astrocytomas to a plateau level of 4.6 +/- 0.4 mL/100 g tissue was measured at 6 months after radiotherapy and remained stable in locally controlled tumors.
CONCLUSION: Monitoring of rCBV changes in normal brain and low-grade astrocytoma was feasible using a DSC-MRI technique. The method was able to document radiation effects in low-grade astrocytoma, even if the majority of tumors showed no change in diagnostic MR-imaging. Radiation induced decrease of rCBV in GM and WM was correlated to total dose delivered to a tissue area, with high doses causing a significant decrease. Minor decline of rCBV in GM and WM outside high-dose areas after stereotactic radiotherapy confirms the efficacy to spare normal brain tissue by the use of modern conformal radiotherapy techniques. Nonetheless, a critical minimal dose initiating rCBV changes is yet unknown.