Purpose: To evaluate the effect of breath-holding on the short-term reproducibility and long-term variability of tumor position during image-guided radiosurgery.Method: Thirteen patients have undergone single-fraction radiosurgery treatments during which the tumor was repeatedly imaged radiographically to observe its position. The imaging data were used to monitor the efficacy of breath-holding and to periodically readjust the alignment of the treatment beam with the tumor. These measurements have allowed the effects of breathing, heartbeat, patient movement, and instrumental uncertainties to be separately identified in the record of tumor position.Results: During inspiration breath-holding, the lung tumor position was reproducible to within 1 mm, on average, in the direction of maximum displacement during regular breathing, and to within 1.8 mm in three dimensions overall. The pancreas tumor position in three dimensions was reproducible to within 2.5 mm on average. Some patients showed a slow, steady drift of tumor position during the extended sequence of breath-holds, which was compensated by periodic retargeting of the treatment beam.Conclusion: Breath-holding can allow the reduction of tumor motion dosimetry margins to 2 turn or less for lung cancer treatments, provided that the treatment system can detect and adapt to long-term variations in the mean tumor position during a lengthy treatment fraction. (C) 2002 Elsevier Science Inc.

PURPOSE:To investigate how respiration influences the motion of lung and pancreas tumors and to relate the observations to treatment procedures intended to improve dose alignment by predicting the moving tumor's position from external breathing indicators. METHODS AND MATERIALS:Breathing characteristics for five healthy subjects were observed by optically tracking the displacement of the chest and abdomen, and by measuring tidal air volume with a spirometer. Fluoroscopic imaging of five radiotherapy patients detected the motion of lung and pancreas tumors synchronously with external breathing indicators. RESULTS:The external and fluoroscopic data showed a wide range of behavior in the normal breathing pattern and its effects on the position of lung and pancreas tumors. This included transient phase shifts between two different external measures of breathing that diminished to zero over a period of minutes, modulated phase shifts between tumor and chest wall motion, and other complex phenomena. CONCLUSIONS:Respiratory compensation strategies that infer tumor position from external breathing signals, including methods of beam gating and dynamic beam tracking, require three-dimensional knowledge of the tumor's motion trajectory as well as the ability to detect and adapt to transient and continuously changing characteristics of respiratory motion during treatment.

Purpose: To study the feasibility of using multileaf collimators and robotically manipulated accelerators to make continuous active adjustments of external radiation beam alignment in response to breathing motion of lung and pancreas tumors.Method and materials: We have studied breathing-related tumor and external anatomy motion fluoroscopically in lung and pancreatic cancer patients, using multiple implanted fiducials to mark the tumor position. The motions have been analyzed for correlation between the internal tumor position and various external breathing measures, including chest strain gauge signals and chest surface displacement. The motions have also been analyzed for predictability ahead by up to one second, to accomodate delay times inherent in realtime control loops.Results: We find a wide range of internal/external breathing motion relationships, including cases of a close steady-state linear correlation between tumor motion and external tracking signal, a linear but nonstationary correlation, and a highly nonlinear correlation. Over a period of 5 – 10 minutes a stationary tapped delay line filter applied to the breathing time series loses predictive ability due to the nonstationary character of the breathing motion. A Kalman filter improves predictability by continuously updating its filter coefficients to the time-changing breathing signal.Conclusions: It appears to be feasible to make realtime beam-alignment corrections for respiratory motion, using external breathing measures to infer tumor position. However, breathing is not strictly periodic and stable. Patients can be expected to exhibit nonstationary motion and time-changing phase relationships involving the tumor motion and the external signals used to tracking breathing during treatment. The internal target position must be monitored episodically during treatment to accomodate these nonstationary trends. These observations are also relevant to respiratory beam-gating techniques that use external optical triggers to track breathing. Purpose: To study the feasibility of using multileaf collimators and robotically manipulated accelerators to make continuous active adjustments of external radiation beam alignment in response to breathing motion of lung and pancreas tumors. Method and materials: We have studied breathing-related tumor and external anatomy motion fluoroscopically in lung and pancreatic cancer patients, using multiple implanted fiducials to mark the tumor position. The motions have been analyzed for correlation between the internal tumor position and various external breathing measures, including chest strain gauge signals and chest surface displacement. The motions have also been analyzed for predictability ahead by up to one second, to accomodate delay times inherent in realtime control loops. Results: We find a wide range of internal/external breathing motion relationships, including cases of a close steady-state linear correlation between tumor motion and external tracking signal, a linear but nonstationary correlation, and a highly nonlinear correlation. Over a period of 5 – 10 minutes a stationary tapped delay line filter applied to the breathing time series loses predictive ability due to the nonstationary character of the breathing motion. A Kalman filter improves predictability by continuously updating its filter coefficients to the time-changing breathing signal. Conclusions: It appears to be feasible to make realtime beam-alignment corrections for respiratory motion, using external breathing measures to infer tumor position. However, breathing is not strictly periodic and stable. Patients can be expected to exhibit nonstationary motion and time-changing phase relationships involving the tumor motion and the external signals used to tracking breathing during treatment. The internal target position must be monitored episodically during treatment to accomodate these nonstationary trends. These observations are also relevant to respiratory beam-gating techniques that use external optical triggers to track breathing.

For optimal radiotherapy treatment of tumors in the abdomen and thorax, the respiratory motion of some abdominal and thoracic organs should be compensated for. We have studied the feasibility of tracking tumor volume and adapting the radiation beam alignment in order to compensate for the respiratory motion of these organs in real time. A linear correlation has been observed between internal markers implanted in a patient's pancreas and external markers attached to the patient's chest. It has also been found that optical tracking of the external markers produces a signal that is in phase with spirometry, allowing the two measurements of breathing to be used interchangeably. We conclude that either optical tracking or spirometry can be used to infer the position of the tumor volume in real time for radiotherapy application.

Purpose: Recent data suggests that adjuvant locoregional radiotherapy may improve survival in women with node positive breast cancer.Concerns have been raised, however, regarding toxicities associated with comprehensive locoregional treatment including the chestwall (CW).supraclavicular fossa (SCF), and internal mammary nodes (IMN).This study compares seven commonly used comprehensive treatment techniques using NTCP predictions for pneumonitis and ischemic heart disease (IHD) and DVH analyses for normal and target tissues.Methods: Twenty representative left-sided post-mastectomy cases were selected.All had contrast enhanced CT scans with clinical delineation of standard field borders using radio-opaque catheters.CW, IMN (interspaces one through three), lung and heart structures were defined in three dimensions.Seven commonly used techniques were implemented for each case, assuming a prescription of 50 Gy in 25 fractions.All techniques included an AP field to the SCF.Techniques were standard tangents to CW only; electron fields to IMN and CW: cobalt fields using an AP IMN field and lateral tangents to CW; reverse hockey stick (RHS) with an AP IMN electron held and lateral AP/PA photon fields to CW; two variations of angled and mixed photon/electron beams (30/70 and 20/80) to IMN with lateral tangents to CW; and partly wide tangent fields (PWTF) to IMN and CW.One hundred forty dose distributions were calculated and nearly 6000 data points were tabulated.NTCP mode1 metrics (Lyman-Kutcher-Burma" NTCP method for lung and Relative Seriality for lung and heart) were used to quantify the risks of pneumonitis and IHD, while dose-volume metrics were used to assess target coverage (CW and IMN) and normal tissue doses.Mean values.standard deviations (SD), and repeated measures analysis of variance (an expansion of a pair-comparison t-test) were used to rank techniques. Results:The twenty cases showed a wide spectrum of body habitus characteristics.The CW was treated most uniformly by standard tangents [mean dose (Dmean) 49.6 Gy (SD 18 Gy)], p = 0.0003, although the other techniques (except the cobalt and electron plans) provided excellent coverage as well.IMN coverage was superior for both the 30/70 photon/electron mix [Dmean 50.0 Gy (3.8 Gy)], p < 0.0001, and PWTF [Dmean 49.3 Gy (3.6 Gy)], p < 0.0001 Pneumonitis predictions for standard tangents [NTCP 0.4% (O.S%)], cobalt [NTCP 0.8% (lo%)], and PWTF [NTCP 2.3% (2.5%)] were lower than values for the 20/80 mix [NTCP 6.0% (8.7%) and 30/70 mix [NTCP 6.8% (114%)], p = 0.0078.The NTCP for IHD was signiticantly lower for PWTF versus all other techniques, with [NTCP 0.0% (O.l%)] for PWTF versus values ranging from [NTCP 0.5% (1 I%)] for RHS to [NTCP 4.4% ( 17%)] for cobalt, p < 0.000 1 Of interest, the NTCP for IHD [ 11% (1 S%)] was significantly greater using standard tangents compared to the NTCP for PWTF.p = 0.0041.Conclusion: No one technique studied combines the best CW and IMN coverage with minimal lung and heart complication probabilities.The choice of technique should be based upon clinical discretion and the technical expertise available to implement these complex plans.Of the techniques studied, our analysis supports PWTF as the most appropriate technique for irradiation of the CW and IMN.

A water beam imaging system (WBIS) was developed and used to verify dose distributions for intensity modulated radiotherapy (IMRT) using dynamic multileaf collimator (MLC). The WBIS consists of a water container, a scintillator screen, a CCD camera, and a portable personal computer. The scintillation image is captured by the CCD camera. The pixel value in this image indicates the dose value in the scintillation screen. The verification is performed by comparing the WBIS image achieved from the measurement with dose distribution from the IMRT plan. Because of light scattering in the WBIS the image is blurred. An iterative reconstruction algorithm is proposed to remove the blurring effect of light scattering. From the measured image of a 10 cm /spl times/10 cm X-ray beam and the simulation result of the dose distribution using the Monte Carlo method, the blur function can be achieved. Based on this function, the proposed algorithm is applied to reconstruct the true dose distribution for an IMRT plan from the measured WBIS image. The reconstructed dose distributions are compared with Monte Carlo simulation results. Reasonable agreement can be observed from the comparison. The proposed approach makes it possible to carry out real-time quality assurance tasks for IMRT dose verification.

Plan comparison is more and more important since the development of new modalities of radiation therapy such as conformal therapy, IMRT, radiosurgery, etc. However, plan comparison is not so simple. The traditional isodose comparison method is too subjective, the dose volume histogram analysis is not good enough for the case of crossing curves or for multiple organs. The uncomplicated tumor control probability is not practical for routine clinical application. In this study, we have developed the practical plan comparison method which uses pseudo-biologic plan scoring.