The clinical utility of magnetic resonance imaging in 3-dimensional treatment planning of brain neoplasms.

Results of the clinical experience gained since 1986 in the treatment planning of patients with brain neoplasms through integration of magnetic resonance imaging (MRI) into computerized tomography (CT)-based, three-dimensional treatment planning are presented. Data from MRI can now be fully registered with CT data using appropriate three-dimensional coordinate transformations allowing: (a) display of MRI defined structures on CT images; (b) treatment planning of composite CT-MRI volumes; (c) dose display on either CT or MRI images. Treatment planning with non-coplanar beam arrangements is also facilitated by MRI because of direct acquisition of information in multiple, orthogonal planes. The advantages of this integration of information are especially evident in certain situations, for example, low grade astrocytomas with indistinct CT margins, tumors with margins obscured by bone artifact on CT scan. Target definitions have repeatedly been altered based on MRI detected abnormalities not visualized on CT scans. Regions of gadolinium enhancement on MRI T1-weighted scans can be compared to the contrast-enhancing CT tumor volumes, while abnormalities detected on MRI T2-weighted scans are the counterpart of CT-defined edema. Generally, MRI markedly increased the apparent macroscopic tumor volume from that seen on contrast-CT alone. However, CT tumor information was also necessary as it defined abnormalities not always perceptible with MRI (on average, 19% of composite CT-MRI volume seen on CT only). In all, the integration of MRI data with CT information has been found to be practical, and often necessary, for the three-dimensional treatment of brain neoplasms.

Design of MRI scan protocols for use in 3-D, CT-based treatment planning.

MRI has the potential of providing the radiation therapy treatment planner with new insights into the definition of target and normal tissue volumes to augment CT in 3-D treatment planning. The current speed of MR scan sequences is not sufficient to enable the acquisition of both T1 and T2 weighted images in all three orthogonal planes in a reasonable period of time. Therefore, compromises must be made in the design of protocols specifically for use in radiotherapy planning which: (1) provide enough information to readily enable image registration; (2) preserve the three-dimensionality provided by image acquisition directly in coronal and sagittal planes; (3) yield tissue contrast as well as tumor specificity (where available); but (4) can be completed in a short enough span of time (or with enough checks) that the patient position is not compromised. Protocols designed for use in planning treatment of the brain, head and neck, lung, prostate, cervix, and sarcomas are presented.

Three-dimensional treatment planning of astrocytomas: a dosimetric study of cerebral irradiation.

To demonstrate that 3-dimensional planning is both practical and applicable to the treatment of high-grade astrocytomas, 50 patients over a 2-year period have received cerebral irradiation delivered in focussed, non-axial techniques employing from 2 to 5 beams. Astrocytomas have been planned using rapid, practical incorporation of CT data to define appropriate tumor volumes. Tumor + 3.0 cm and tumor + 1.5 cm volumes have been treated to conventional doses of 4500 cGy and 5940 cGy, respectively, using beam orientations that maximally spared normal remaining parenchyma. Analyses of 3-dimensionally calculated plans have been performed using integral dose-volume histograms (DVH) to help select treatment techniques. Using identical CT-based volumetric data as input for generation of Beam's Eye View (BEV) designed blocks, DVH curves demonstrate dosimetric advantages of non-axial techniques over conventional parallel-opposed orientations. Assessment of the non-axial techniques in selected cases indicates that uniform target volume coverage could be maintained with a typical reduction of 30% in the total amount of brain tissue treated to high dose (95% isodose line).

The influence of lung density corrections on treatment planning for primary breast cancer.

Primary breast cancer is generally treated with opposed radiation beams oriented tangentially with respect to the breast. This technique attempts to minimize the dose to the lung and other normal tissues, while at the same time producing a uniform dose distribution throughout the irradiated breast. Although a part of the lung is always included in the tangential breast fields, the effect of this low density tissue on the dose distribution is rarely taken into account. In the present work, the effect of lung density correction on the dose distribution resulting from tangential breast fields is analyzed. Treatment plans for a series of 34 patients treated for breast cancer have been performed using CT data. To study the effect of density corrections on the tangential field treatment plans for these patients, eight separate treatment plans for each patient have been optimized. For each of four photon energies (60Co, and 4, 6, and 10 MV X rays), treatment plans have been optimized for each patient when density correction is employed, and when unit density is assumed. Four additional dose calculations have been obtained for each patient corresponding to use of the unit density plan, but with density corrections employed in the calculation. The effects that density correction has on the wedge angles used, on the maximum dose ("hot spot") for each of several cross-sectional cuts, on the prescription isodose level which is chosen for each plan, and on homogeneity of the dose distribution over the target volume are all analyzed for the above described plans.