Recommendations on the practice of calibration, dosimetry, and quality assurance for gamma stereotactic radiosurgery: Report of AAPM Task Group 178.

The American Association of Physicists in Medicine (AAPM) formed Task Group 178 (TG-178) to perform the following tasks: review in-phantom and in-air calibration protocols for gamma stereotactic radiosurgery (GSR), suggest a dose rate calibration protocol that can be successfully utilized with all gamma stereotactic radiosurgery (GSR) devices, and update quality assurance (QA) protocols in TG-42 (AAPM Report 54, 1995) for static GSR devices. The TG-178 report recommends a GSR dose rate calibration formalism and provides tabulated data to implement it for ionization chambers commonly used in GSR dosimetry. The report also describes routine mechanical, dosimetric, and safety checks for GSR devices, and provides treatment process quality assurance recommendations. Sample worksheets, checklists, and practical suggestions regarding some QA procedures are given in appendices. The overall goal of the report is to make recommendations that help standardize GSR physics practices and promote the safe implementation of GSR technologies.

A theoretical investigation of optimal target-dose conformity in gamma knife radiosurgery.

PURPOSE: The purpose of this paper is to suggest guidelines for target-dose conformity in gamma knife stereotactic radiosurgery (GKSRS) by taking into account factors that have been linked to GKSRS complications. We also suggest an explanation for the failure of previous studies to find a correlation between improved conformity index and reduced risk of GKSRS toxicity, where the conformity index, C(S), is defined as the ratio of the prescription volume, V(P), to the target volume, V(T). METHODS: Previous investigations have shown that symptomatic toxicity in GKSRS is correlated with the volume of nontarget tissue receiving the prescription dose, D(P). In this study, we formulated the volume of nontarget tissue, V(NTD), receiving dose D < or = D(P) as a function of the target volume, prescription volume, and prescription dose. We verified the model for D = 12-15 Gy by comparing VNTD calculated from the model versus VNTD calculated directly for 114 tumors in 63 consecutive patients treated at our institution. Once verified, we used this formulation of V(NTD) to calculate the volume of nontarget tissue receiving doses between 12 and 15 Gy from published data reported for patients experiencing varying degrees of GKSRS toxicity. Next, assuming that the VNTD values calculated for those patients who had either no toxicity or mild neurological symptoms in the published study represented safe levels of normal tissue irradiated to the dose in question, we substituted these V(NTD) values into an equation expressing C(S) in terms of V(NTD), V(T), and D(P), and examined how C(S) varied as a function of V(T) and D(P). RESULTS: The R2 value for the correlation between VNTD calculated directly or calculated with the proposed formula for VNTD ranged from 0.98 to 0.99, indicating that the formula accurately models the behavior of the nontarget volume receiving dose D. Applying this formulation of VNTD to historical data suggested that the requirements V(NT15) < or = 2.2 cm3, V(NT14) < or = 2.6 cm3, V(NT13) < or = 3.1 cm3 and V(NT12) < or = 3.8 cm3 minimize the risk of severe complications following GKSRS. Imposing these criteria imply that as the target size increases, delivering a given prescription dose requires increasing target-dose conformity. For tumor sizes >5 cm3 C(S) must be < or = 1.2 to restrict V(NTD) to the values listed above. For very small targets, on the other hand, nearly any reasonable conformity index will lead to acceptable values of V(NTD). These observations may explain why previous investigations failed to show a correlation between improved conformity and decreased toxicity in GKSRS, because in these earlier studies the range of conformity indices represented was not wide enough, in particular C(S) values <1.3 were not represented for large tumors. CONCLUSIONS: Our model suggests that for target volumes > or = 3 cm3, high levels of target-dose conformity (C(S) < 1.3) are required for typical GKSRS prescription doses in order to limit VNTD to levels associated with either no toxicity or mild neurological symptoms in a previous investigation.

Peripheral dose measurement for CyberKnife radiosurgery with upgraded linac shielding.

The authors investigated the peripheral dose reduction for CyberKnife radiosurgery treatments after the installation of a linac shielding upgrade. As in a previous investigation, the authors considered two treatment plans, one for a hypothetical target in the brain and another for a target in the thorax, delivered to an anthropomorphic phantom. The results of the prior investigation showed that the CyberKnife delivered significantly higher peripheral doses than comparable model C Gamma Knife or IMRT treatments. Current measurements, after the linac shielding upgrade, demonstrate that the additional shielding decreased the peripheral dose, expressed as a percentage of the delivered monitor units (MU), by a maximum of 59%. The dose reduction was greatest for cranial-caudal distances from the field edge less than 30 cm, and at these distances, the CyberKnife peripheral dose, expressed as a percentage of the delivered MU, is now comparable to that measured for the other treatment modalities in our previous investigation. For distances between 30 and 70 cm from the field edge, the additional shielding reduced the peripheral dose by between 20% and 55%. At these distances, the CyberKnife peripheral dose remains higher than doses measured in our previous study for the model C Gamma Knife and IMRT.

Use of hybrid shots in planning Perfexion Gamma Knife treatments for lesions close to critical structures.

OBJECT: The authors investigated the use of different collimator values in different sectors (hybrid shots) when treating patients with lesions close to critical structures with the Perfexion model Gamma Knife. METHODS: Twelve patients with various tumors (6 with a pituitary tumor, 3 with vestibular schwannoma, 2 with meningioma, and 1 with metastatic lesion) that were within 4 mm of the brainstem, optic nerve, pituitary stalk, or cochlea were considered. All patients were treated at the authors' institution between June 2007 and March 2008. The patients' treatments were replanned in 2 different ways. In the first plan, hybrid shots were used such that the steepest dose gradient was aligned with the junction between the target and the critical structure(s). This was accomplished by placing low-value collimators in appropriate sectors. In the second plan, no hybrid shots were used. Sector blocking (either manual or dynamic) was required for all plans to reduce the critical structure doses to acceptable levels. Prescribed doses ranged from 12 to 30 Gy at the periphery of the target. The plans in each pair were designed to be equally conformal in terms of both target coverage (as measured by the Paddick conformity index) and critical structure sparing. RESULTS: The average number of shots required was roughly the same using either planning technique (16.7 vs 16.6 shots with and without hybrids). However, for all patients, the number of blocked sectors required to protect critical areas was larger when hybrid shots were not used. On average, nearly twice as many blocked sectors (14.8 vs 7.0) were required for the plans that did not use hybrid shots. The number of high-value collimators used in each plan was also evaluated. For small targets ( 1 cm(3)), for which 16 mm was considered a high value for the collimator, hybrid plans used an average of 1.4 times as many 16-mm sectors as did the plans without hybrids (10.7 vs 7.7 sectors). Decreasing the number of blocked sectors and increasing the number of high-value collimator sectors led to use of shorter beam-on times. Beam-on times were 1-39% higher (average 17%) when hybrid shots were not allowed. The average beam-on time for plans with and without hybrid shots was 67.4 versus 78.4 minutes. CONCLUSIONS: The judicious use of hybrid shots in patients for whom the target is close to a critical structure is an efficient way to achieve conformal treatments while minimizing the beam-on time. The reduction in beam-on time with hybrid shots is attributed to a reduced use of blocked sectors and an increased number of high-value collimator sectors.

Anatomic landmarks versus fiducials for volume-staged gamma knife radiosurgery for large arteriovenous malformations.

PURPOSE: The purpose of this investigation was to compare the accuracy of using internal anatomic landmarks instead of surgically implanted fiducials in the image registration process for volume-staged gamma knife (GK) radiosurgery for large arteriovenous malformations. METHODS AND MATERIALS: We studied 9 patients who had undergone 10 staged GK sessions for large arteriovenous malformations. Each patient had fiducials surgically implanted in the outer table of the skull at the first GK treatment. These markers were imaged on orthogonal radiographs, which were scanned into the GK planning system. For the same patients, 8-10 pairs of internal landmarks were retrospectively identified on the three-dimensional time-of-flight magnetic resonance imaging studies that had been obtained for treatment. The coordinate transformation between the stereotactic frame space for subsequent treatment sessions was then determined by point matching, using four surgically embedded fiducials and then using four pairs of internal anatomic landmarks. In both cases, the transformation was ascertained by minimizing the chi-square difference between the actual and the transformed coordinates. Both transformations were then evaluated using the remaining four to six pairs of internal landmarks as the test points. RESULTS: Averaged over all treatment sessions, the root mean square discrepancy between the coordinates of the transformed and actual test points was 1.2 +/- 0.2 mm using internal landmarks and 1.7 +/- 0.4 mm using the surgically implanted fiducials. CONCLUSION: The results of this study have shown that using internal landmarks to determine the coordinate transformation between subsequent magnetic resonance imaging scans for volume-staged GK arteriovenous malformation treatment sessions is as accurate as using surgically implanted fiducials and avoids an invasive procedure.

Peripheral doses in CyberKnife radiosurgery.

The purpose of this work is to measure the dose outside the treatment field for conformal CyberKnife treatments, to compare the results to those obtained for similar treatments delivered with gamma knife or intensity-modulated radiation therapy (IMRT), and to investigate the sources of peripheral dose in CyberKnife radiosurgery. CyberKnife treatment plans were developed for two hypothetical lesions in an anthropomorphic phantom, one in the thorax and another in the brain, and measurements were made with LiF thermoluminescent dosimeters (TLD-100 capsules) placed within the phantom at various depths and distances from the irradiated volume. For the brain lesion, gamma knife and 6-MV IMRT treatment plans were also developed, and peripheral doses were measured at the same locations as for the CyberKnife plan. The relative contribution to the CyberKnife peripheral dose from inferior- or superior-oblique beams entering or exiting through the body, internally scattered radiation, and leakage radiation was assessed through additional experiments using the single-isocenter option of the CyberKnife treatment-planning program with different size collimators. CyberKnife peripheral doses (in cGy) ranged from 0.16 to 0.041% (+/- 0.003%) of the delivered number of monitor units (MU) at distances between 18 and 71 cm from the field edge. These values are two to five times larger than those measured for the comparable gamma knife brain treatment, and up to a factor of four times larger those measured in the IMRT experiment. Our results indicate that the CyberKnife peripheral dose is due largely to leakage radiation, however at distances less than 40 cm from the field edge, entrance, or exit dose from inferior- or superior-oblique beams can also contribute significantly. For distances larger than 40 cm from the field edge, the CyberKnife peripheral dose is directly related to the number of MU delivered, since leakage radiation is the dominant component.

Comparison of intensity-modulated radiosurgery with gamma knife radiosurgery for challenging skull base lesions.

PURPOSE: To quantitatively compare intensity-modulated radiosurgery (IMRS) using 3-mm mini-multileaf collimation with gamma knife radiosurgery (GKRS) plans for irregularly shaped skull base lesions in direct proximity to organs at risk (OAR). METHODS AND MATERIALS: Ten challenging skull base lesions originally treated with GKRS were selected for comparison with IMRS using inverse treatment planning and 3-mm mini-multileaf collimation operating in step-and-shoot delivery mode. The lesions ranged in volume from 1.6 to 32.2 cm(3) and were treated with 9-20 GK isocenters (mean 13.2). The IMRS plans were designed with the intent to, at minimum, match the GKRS plans with regard to OAR sparing and target coverage. For each case, IMRS plans were generated using 9 coplanar, 11 equally spaced noncoplanar, and 11 OAR-avoidant noncoplanar beams; the best of these approaches with respect to target conformality, sparing of OAR, and maintaining coverage was selected for comparison with the original GKRS plan. RESULTS: Assuming no patient motion or setup error, IMRS provided comparable target coverage and sparing of OAR and an improved conformity index at the prescription isodose contour but sometimes less conformity at lower isodose contours compared with the actual GKRS plan. All IMRS plans produced less target dose heterogeneity and shorter estimated treatment times compared with the GKRS plans. CONCLUSION: Compared with GKRS for complex skull base lesions, IMRS plans using a 3-mm mini-multileaf collimator achieved comparable or sometimes improved target coverage, conformity, and critical structure sparing with shorter estimated treatment times.

Checking monitor unit calculations using coordinate transformations to calculate off-axis distances in the collimator frame of reference.

An important part of treatment-planning QA is to check Monitor Units (MUs) calculated by treatment-planning programs. This is generally straightforward, unless the central axis is blocked. One way to check MUs in this case is to select a reference point in the open portion of the field and use the off-axis distance (OAD), as well as other relevant data, to verify the dose. If wedges are employed in the treatment, the OAD must be specified in the collimator frame-of-reference because one must know where the calculation point is with respect to the wedge to calculate the dose correctly. The purpose of this paper is to describe a method of calculating the OAD in the collimator frame-of-reference using the system of coordinate transformations described by Siddon