Vernon Smith

THERAPEUTIC IRRADIATION AND BRAIN INJURY

Abstract This is a review and reanalysis of the literature on adverse effects of therapeutic irradiation on the brain. Reactions have been grouped and considered according to time of appearance. The emphasis of the analysis is on delayed reactions, especially those that occur from a few months to several years after irradiation. Over 100 such cases are reported in the literature. Eighty cases were identified in which the patient was given a single course of radiation therapy and in which reasonable estimates of the time-dose-fractionation regimen could be made. Eleven of these patients had been irradiated for cancer of the skin, 17 for other extracranial tumors, 20 for either a pituitary tumor or a craniopharyngioma and 32 for primary brain tumors. All dose specifications were converted into equivalent megavoltage rads. The data were analyzed in terms of total dose, overall treatment time and number of treatment fractions. The data were also analyzed in terms of time-dose-fractionation (TDF) and nominal standard dose (NSD). NSD calculations were done according to the usual Ellis formula and also according to a modification in which the N exponent was −0.44 and the T exponent −0.06. When total dose was plotted against number of fractions, a line with slope 0.44 fitted well with the lower dose limits at which brain necrosis has been reported. Also discussed were acute radiation reactions, early delayed radiation reactions, somnolence and leukoencephalopathy post-irradiation/chemotherapy and combined effects of radiation and chemotherapy.

Temperature distribution measurements in two‐dimensional NMR imaging

This paper represents a preliminary study of the effects of regional temperature distribution in two-dimensional nuclear magnetic resonance (NMR) T1 imaging. It is found, as expected, that variations in local temperature appear as variations in the corresponding T1 image. The potential use of NMR T1 imaging in temperature measurements is evaluated in the case of water and blood samples. Using containers where the temperature could be either known or directly controlled with reasonable accuracy, images are obtained with samples having at least two regions at different temperatures. As expected, T1 is found to vary linearly with 1/T over the range of 0 degrees C to about 40 degrees C for blood. The potential use of T1 imaging in hyperthermia applications is also discussed.

Radiation pneumonitis: A new approach to the derivation of time‐dose factors

A series of 51 patients treated with 92 separate lung fields for metastatic pulmonary disease between 1958 and 1971 is reviewed. The treatment data are converted into nominal single doses (NSD) and a newly derived formula for estimated single doses (ED). The 5% incidence level of pneumonitis without dactinomycin is 770 rets (NSD) and 510 rets (ED) and with dactinomycin, 520 rets (NSD) and 450 rets (ED). A safe treatment regimen for avoiding radiation pneumonitis is 1500 rads in 10 fractions with dactinomycin and 2500 rads in 20 fractions without dactinomycin.

Radiosurgery for brain metastases: Relationship of dose and pattern of enhancement to local control

PURPOSE This study aimed to analyze dose, initial pattern of enhancement, and other factors associated with freedom from progression (FFP) of brain metastases after radiosurgery (RS). METHODS AND MATERIALS All brain metastases treated with gamma-knife RS at the University of California, San Francisco, from 1991 to 1994 were reviewed. Evaluable lesions were those with follow-up magnetic resonance or computed tomographic imaging. Actuarial FFP was calculated using the Kaplan-Meier method, measuring FFP from the date of RS to the first imaging study showing tumor progression. Controlled lesions were censored at the time of the last imaging study. Multivariate analyses were performed using a stepwise Cox proportional hazards model. RESULTS Of 261 lesions treated in 119 patients, 219 lesions in 100 patients were evaluable. Major histologies included adenocarcinoma (86 lesions), melanoma (77), renal cell carcinoma (21), and carcinoma not otherwise specified (17). The median prescribed RS dose was 18.5 Gy (range, 10-22) and the median tumor volume was 1.3 ml (range, 0.02-30.9). The initial pattern of contrast enhancement was homogeneous in 68% of lesions, heterogeneous in 12%, and ring-enhancing in 19%. The actuarial FFP was 82% at 6 months and 77% at 1 year for all lesions, and 93 and 90%, respectively, for 145 lesions receiving > or = 18 Gy. Multivariate analysis showed that longer FFP was significantly associated with higher prescribed RS dose, a homogeneous pattern of contrast enhancement, and a longer interval between primary diagnosis and RS. Adjusted for these factors, adenocarcinomas had longer FFP than melanomas. No significant differences in FFP were noted among lesions undergoing RS for recurrence after prior radiotherapy (119 lesions), RS alone as initial treatment (45), or RS boost (55). CONCLUSION A minimum prescribed radiosurgical dose > or = 18 Gy yields excellent local control of brain metastases. The influence of pattern of enhancement on local control, a new finding in this retrospective analysis, needs to be confirmed.

A multi-institutional analysis of complication outcomes after arteriovenous malformation radiosurgery

PURPOSE To better understand radiation complications of arteriovenous malformation (AVM) radiosurgery and factors affecting their resolution. METHODS AND MATERIALS AVM patients (102/1255) who developed neurological sequelae after radiosurgery were studied. The median AVM marginal dose (Dmin) was 19 Gy (range: 10-35). The median volume was 5.7 cc (range: 0.26-143). Median follow-up was 34 months (range: 9-140). RESULTS Complications consisted of 80/102 patients with evidence of radiation injury to the brain parenchyma (7 also with cranial nerve deficits, 12 also with seizures, 5 with cyst formation), 12/102 patients with isolated cranial neuropathies, and 10/102 patients with only new or worsened seizures. Severity was classified as minimal in 39 patients, mild in 40, disabling in 21, and fatal in 2 patients. Symptoms resolved completely in 42 patients for an actuarial resolution rate of 54% +/- 7% at 3 years post-onset. Multivariate analysis identified significantly greater symptom resolution in patients with no prior history of hemorrhage (p = 0.01, 66% vs. 41%), and in patients with symptoms of minimal severity: headache or seizure as the only sequelae of radiosurgery (p < 0.0001, 88% vs. 34%). CONCLUSION Late sequelae of radiosurgery manifest in varied ways. Further long-term studies of these problems are needed that take into account symptom severity and prior hemorrhage history.

Five year results of linac radiosurgery for arteriovenous malformations: outcome for large AVMS

PURPOSE For radiosurgery of large arteriovenous malformations (AVMs), the optimal relationship of dose and volume to obliteration, complications, and hemorrhage is not well defined. Multivariate analysis was performed to assess the relationship of multiple AVM and treatment factors to the outcome of AVMs significantly larger than previously reported in the literature. METHODS AND MATERIALS 73 patients with intracranial AVMs underwent LINAC radiosurgery. Over 50% of the AVMs were larger than 3 cm in diameter and the median and mean treatment volumes were 8.4 cc and 15.3 cc, respectively (range 0.4-143.4 cc). Minimum AVM treatment doses varied between 1000-2200 cGy (median: 1600 cGy). RESULTS The obliteration rates for treatment volumes < 4 cc, 4-13.9 cc, and > or = 14 cc were 67%, 58%, and 23%, respectively. AVM obliteration was significantly associated with higher minimum treatment dose and negatively associated with a history of prior embolization with particulate materials. No AVM receiving < 1400 cGy was obliterated. The incidence of post-radiosurgical imaging abnormalities and clinical complications rose with increasing treatment volume. For treatment volumes > 14 cc receiving > or = 1600 cGy, the incidence of post-radiosurgical MRI T2 abnormalities was 72% and the incidence of radiation necrosis requiring resection was 22%. The rate of post-radiosurgical hemorrhage was 2.7% per person-year for AVMs with treatment volumes < 14 cc and 7.5% per person-year for AVMs > or = 14 cc. CONCLUSION As AVM size increases, the dose-volume range for the optimal balance between successful obliteration and the risk of complications and post-radiosurgical hemorrhage narrows.

Dose conformity of gamma knife radiosurgery and risk factors for complications

PURPOSE To quantitatively evaluate dose conformity achieved using Gamma Knife radiosurgery, compare results with those reported in the literature, and evaluate risk factors for complications. METHODS AND MATERIALS All lesions treated at our institution with Gamma Knife radiosurgery from May 1993 (when volume criteria were routinely recorded) through December 1998 were reviewed. Lesions were excluded from analysis for reasons listed below. Conformity index (the ratio of prescription volume to target volume) was calculated for all evaluable lesions and for lesions comparable to those reported in the literature on conformity of linac radiosurgery. Univariate Cox regression models were used to test for associations between treatment parameters and toxicity. RESULTS Of 1612 targets treated in 874 patients, 274 were excluded, most commonly for unavailability of individual prescription volume data because two or more lesions were included within the same dose matrix (176 lesions), intentional partial coverage for staged treatment of large arteriovenous malformations (AVMs) (33 lesions), and missing target volume data (26 lesions). The median conformity indices were 1.67 for all 1338 evaluable lesions and 1.40-1.43 for lesions comparable to two linac radiosurgery series that reported conformity indices of 1.8 and 2.7, respectively. Among all 651 patients evaluable for complications, there were one Grade 5, eight Grade 4, and 27 Grade 3 complications. Increased risk of toxicity was associated with larger target volume, maximum lesion diameter, prescription volume, or volume of nontarget tissue within the prescription volume. CONCLUSIONS Gamma Knife radiosurgery achieves much more conformal dose distributions than those reported for conventional linac radiosurgery and somewhat more conformal dose distributions than sophisticated linac radiosurgery techniques. Larger target, nontarget, or prescription volumes are associated with increased risk of toxicity.

Comparison of radiosurgery treatment modalities based on physical dose distributions

PURPOSE As a means of selecting the optimal stereotactic radiosurgery (SRS) treatment modality, a comparison of physical dose distributions to defined targets and nontarget brain tissue has been made for a group of test cases selected to represent a range of treatment-planning situations from small, nearly spherical volumes to large irregular volumes. METHODS AND MATERIALS Plans were developed for each case using photon beams from the Leksell Gamma Unit (LGU), multiarc bremsstrahlung photon beams from a linear accelerator (linac) and proton beams, with the objective of encompassing the target as closely as possible with the prescription isodose line, and minimizing dosage to normal tissue within the bounds of standard clinical practice. Dose-volume histograms (DVHs) were calculated for target and for nontarget brain tissue and compared for the various modalities. RESULTS In general, protons delivered less dosage to normal brain than other modalities for large and peripheral lesions and LGU plans were more successful at conforming to highly irregular shapes than conventional linac plans. CONCLUSIONS Differences were observed to depend on treatment modality, target characteristics (shape, size and location), and the amount of effort expended on treatment planning and the time allotted for treatment implementation.

Radiosurgery for malignant meningioma: results in 22 patients

The purpose of this paper was to note a potential source of error in magnetic resonance (MR) imaging. Magnetic resonance images were acquired for stereotactic planning for GKS of a vestibular schwannoma in a female patient. The images were acquired using three-dimensional sequence, which has been shown to produce minimal distortion effects. The images were transferred to the planning workstation, but the coronal images were rejected. By examination of the raw data and reconstruction of sagittal images through the localizer side plate, it was clearly seen that the image of the square localizer system was grossly distorted. The patient was returned to the MR imager for further studies and a metal clasp on her brassiere was identified as the cause of the distortion.A-60-year-old man with medically intractable left-sided maxillary division trigeminal neuralgia had severe cardiac disease, was dependent on an internal defibrillator and could not undergo magnetic resonance imaging. The patient was successfully treated using computerized tomography (CT) cisternography and gamma knife radiosurgery. The patient was pain free 2 months after GKS. Contrast cisternography with CT scanning is an excellent alternative imaging modality for the treatment of patients with intractable trigeminal neuralgia who are unable to undergo MR imaging.The authors describe acute deterioration in facial and acoustic neuropathies following radiosurgery for acoustic neuromas. In May 1995, a 26-year-old man, who had no evidence of neurofibromatosis Type 2, was treated with gamma knife radiosurgery (GKS; maximum dose 20 Gy and margin dose 14 Gy) for a right-sided intracanalicular acoustic tumor. Two days after the treatment, he developed headache, vomiting, right-sided facial weakness, tinnitus, and right hearing loss. There was a deterioration of facial nerve function and hearing function from pretreatment values. The facial function worsened from House-Brackmann Grade 1 to 3. Hearing deteriorated from Grade 1 to 5. Magnetic resonance (MR) images, obtained at the same time revealed an obvious decrease in contrast enhancement of the tumor without any change in tumor size or peritumoral edema. Facial nerve function improved gradually and increased to House-Brackmann Grade 2 by 8 months post-GKS. The tumor has been unchanged in size for 5 years, and facial nerve function has also been maintained at Grade 2 with unchanged deafness. This is the first detailed report of immediate facial neuropathy after GKS for acoustic neuroma and MR imaging revealing early possibly toxic changes. Potential explanations for this phenomenon are presented.In clinical follow-up studies after radiosurgery, imaging modalities such as computerized tomography (CT) and magnetic resonance (MR) imaging are used. Accurate determination of the residual lesion volume is necessary for realistic assessment of the effects of treatment. Usually, the diameters rather than the volume of the lesion are measured. To determine the lesion volume without using stereotactically defined images, the software program VOLUMESERIES has been developed. VOLUMESERIES is a personal computer-based image analysis tool. Acquired DICOM CT scans and MR image series can be visualized. The region of interest is contoured with the help of the mouse, and then the system calculates the volume of the contoured region and the total volume is given in cubic centimeters. The defined volume is also displayed in reconstructed sagittal and coronal slices. In addition, distance measurements can be performed to measure tumor extent. The accuracy of VOLUMESERIES was checked against stereotactically defined images in the Leksell GammaPlan treatment planning program. A discrepancy in target volumes of approximately 8% was observed between the two methods. This discrepancy is of lesser interest because the method is used to determine the course of the target volume over time, rather than the absolute volume. Moreover, it could be shown that the method was more sensitive than the tumor diameter measurements currently in use. VOLUMESERIES appears to be a valuable tool for assessing residual lesion volume on follow-up images after gamma knife radiosurgery while avoiding the need for stereotactic definition.This study was conducted to evaluate the geometric distortion of angiographic images created from a commonly used digital x-ray imaging system and the performance of a commercially available distortion-correction computer program. A 12 x 12 x 12-cm wood phantom was constructed. Lead shots, 2 mm in diameter, were attached to the surfaces of the phantom. The phantom was then placed inside the angiographic localizer. Cut films (frontal and lateral analog films) of the phantom were obtained. The films were analyzed using GammaPlan target series 4.12. The same procedure was repeated with a digital x-ray imaging system equipped with a computer program to correct the geometric distortion. The distortion of the two sets of digital images was evaluated using the coordinates of the lead shots from the cut films as references. The coordinates of all lead shots obtained from digital images and corrected by the computer program coincided within 0.5 mm of those obtained from cut films. The average difference is 0.28 mm with a standard deviation of 0.01 mm. On the other hand, the coordinates obtained from digital images with and without correction can differ by as much as 3.4 mm. The average difference is 1.53 mm, with a standard deviation of 0.67 mm. The investigated computer program can reduce the geometric distortion of digital images from a commonly used x-ray imaging system to less than 0.5 mm. Therefore, they are suitable for the localization of arteriovenous malformations and other vascular targets in gamma knife radiosurgery.

Tumor and normal tissue response to irradiation in vivo: variation with decreasing dose rates

Continuous irradiation in vivo at dose rates ranging form 0.54 to 274 rads per minute was performed in mice using the EMT6 tumor, the gut clone system, and the bone-marrow CFU system. A progressive increase in D0 and decrease in ñ were seen with decreasing dose rates in the EMT6 tumor and the gut clone system. The D0 showed little change in the bone-marrow CFU system. These findings are related to the D2-D1 values obtained from split-dose experiments at conventional dose rates. The results do not fit the various mathematical models proposed for correlating the effects of continuous low-dose-rate irradiation and acute exposure.