Mitigating errors in external respiratory surrogate-based models of tumor position.

PURPOSE: To investigate the effect of tumor site, measurement precision, tumor-surrogate correlation, training data selection, model design, and interpatient and interfraction variations on the accuracy of external marker-based models of tumor position. METHODS AND MATERIALS: Cyberknife Synchrony system log files comprising synchronously acquired positions of external markers and the tumor from 167 treatment fractions were analyzed. The accuracy of Synchrony, ordinary-least-squares regression, and partial-least-squares regression models for predicting the tumor position from the external markers was evaluated. The quantity and timing of the data used to build the predictive model were varied. The effects of tumor-surrogate correlation and the precision in both the tumor and the external surrogate position measurements were explored by adding noise to the data. RESULTS: The tumor position prediction errors increased during the duration of a fraction. Increasing the training data quantities did not always lead to more accurate models. Adding uncorrelated noise to the external marker-based inputs degraded the tumor-surrogate correlation models by 16% for partial-least-squares and 57% for ordinary-least-squares. External marker and tumor position measurement errors led to tumor position prediction changes 0.3-3.6 times the magnitude of the measurement errors, varying widely with model algorithm. The tumor position prediction errors were significantly associated with the patient index but not with the fraction index or tumor site. Partial-least-squares was as accurate as Synchrony and more accurate than ordinary-least-squares. CONCLUSIONS: The accuracy of surrogate-based inferential models of tumor position was affected by all the investigated factors, except for the tumor site and fraction index.

Inferring positions of tumor and nodes in Stage III lung cancer from multiple anatomical surrogates using four-dimensional computed tomography.

PURPOSE: To investigate the feasibility of modeling Stage III lung cancer tumor and node positions from anatomical surrogates. METHODS AND MATERIALS: To localize their centroids, the primary tumor and lymph nodes from 16 Stage III lung cancer patients were contoured in 10 equal-phase planning four-dimensional (4D) computed tomography (CT) image sets. The centroids of anatomical respiratory surrogates (carina, xyphoid, nipples, mid-sternum) in each image set were also localized. The correlations between target and surrogate positions were determined, and ordinary least-squares (OLS) and partial least-squares (PLS) regression models based on a subset of respiratory phases (three to eight randomly selected) were created to predict the target positions in the remaining images. The three-phase image sets that provided the best predictive information were used to create models based on either the carina alone or all surrogates. RESULTS: The surrogate most correlated with target motion varied widely. Depending on the number of phases used to build the models, mean OLS and PLS errors were 1.0 to 1.4 mm and 0.8 to 1.0 mm, respectively. Models trained on the 0%, 40%, and 80% respiration phases had mean (+/- standard deviation) PLS errors of 0.8 +/- 0.5 mm and 1.1 +/- 1.1 mm for models based on all surrogates and carina alone, respectively. For target coordinates with motion >5 mm, the mean three-phase PLS error based on all surrogates was 1.1 mm. CONCLUSIONS: Our results establish the feasibility of inferring primary tumor and nodal motion from anatomical surrogates in 4D CT scans of Stage III lung cancer. Using inferential modeling to decrease the processing time of 4D CT scans may facilitate incorporation of patient-specific treatment margins.

Investigation of motion sickness and inertial stability on a moving couch for intra-fraction motion compensation.

BACKGROUND. Respiration-induced tumor motion compensation using a treatment couch requires moving the patient at non-trivial speeds. The purpose of this work was to investigate motion sickness and stability of the patient's external surface due to a moving couch with respiration-comparable velocities and accelerations. MATERIAL AND METHODS. A couch was designed to move with a peak-peak displacement of 5 cm and 1 cm in the S-I and A-P directions, respectively, and a period of 3.6 s. Fifty patients completed a 16-question motion sickness assessment questionnaire (MSAQ) prior to, during, and after the study. Seven optical reflectors affixed to the abdomen of each patient were monitored by infrared cameras. The relationship between reflector positions under stationary and moving conditions was evaluated to assess the stability of the patient's external surface. RESULTS AND DISCUSSION. Among the 4800 responses, 95% were 1 (no discomfort) of 9, and there were no scores of 6 or higher. Mild discomfort (scores of 4-5) was similar during couch motion and before couch motion (p = 0.39). Mild discomfort was less common after couch motion (p = 0.039) than before or during couch movement. There was a near 1:1 correspondence between marker-pair regression coefficients and phase offset values during couch-stationary and couch-moving conditions. Our results show that patients do not suffer motion sickness or external surface instability on a moving couch.

Feasibility of small animal cranial irradiation with the microRT system.

PURPOSE: To develop and validate methods for small-animal CNS radiotherapy using the microRT system. MATERIALS AND METHODS: A custom head immobilizer was designed and built to integrate with a pre-existing microRT animal couch. The Delrin couch-immobilizer assembly, compatible with multiple imaging modalities (CT, microCT, microMR, microPET, microSPECT, optical), was first imaged via CT in order to verify the safety and reproducibility of the immobilization method. Once verified, the subject animals were CT-scanned while positioned within the couch-immobilizer assembly for treatment planning purposes. The resultant images were then imported into CERR, an in-house-developed research treatment planning system, and registered to the microRTP treatment planning space using rigid registration. The targeted brain was then contoured and conformal radiotherapy plans were constructed for two separate studies: (1) a whole-brain irradiation comprised of two lateral beams at the 90 degree and 270 degree microRT treatment positions and (2) a hemispheric (left-brain) irradiation comprised of a single A-P vertex beam at the 0 degree microRT treatment position. During treatment, subject animals (n=48) were positioned to the CERR-generated treatment coordinates using the three-axis microRT motor positioning system and were irradiated using a clinical Ir-192 high-dose-rate remote after-loading system. The radiation treatment course consisted of 5 Gy fractions, 3 days per week. 90% of the subjects received a total dose of 30 Gy and 10% received a dose of 60 Gy. RESULTS: Image analysis verified the safety and reproducibility of the immobilizer. CT scans generated from repeated reloading and repositioning of the same subject animal in the couch-immobilizer assembly were fused to a baseline CT. The resultant analysis revealed a 0.09 mm average, center-of-mass translocation and negligible volumetric error in the contoured, murine brain. The experimental use of the head immobilizer added 0.1 mm to microRT spatial uncertainty along each axis. Overall, the total spatial uncertainty for the prescribed treatments was +/-0.3 mm in all three axes, a 0.2 mm functional improvement over the original version of microRT. Subject tolerance was good, with minimal observed side effects and a low procedure-induced mortality rate. Throughput was high, with average treatment times of 7.72 and 3.13 min/animal for the whole-brain and hemispheric plans, respectively (dependent on source strength). CONCLUSIONS: The method described exhibits conformality more in line with the size differential between human and animal patients than provided by previous prevalent approaches. Using pretreatment imaging and microRT-specific treatment planning, our method can deliver an accurate, conformal dose distribution to the targeted murine brain (or a subregion of the brain) while minimizing excess dose to the surrounding tissue. Thus, preclinical animal studies assessing the radiotherapeutic response of both normal and malignant CNS tissue to complex dose distributions, which closer resemble human-type radiotherapy, are better enabled. The procedural and mechanistic framework for this method logically provides for future adaptation into other murine target organs or regions.

Efficient use of continuous, real-time prostate localization.

Recent technological advances make it possible to monitor prostate movement during radiation delivery. Using previously published data from 35 patients who underwent continuous localization during prostate cancer treatment, we simulated various interventions to identify the radiation-gating and patient-repositioning strategies that least prolonged the time to complete the daily treatment. Acceptable response protocols were those that resulted in at least 95% of patients' prostates remaining within the planning margins at least 95% of the time. Gating and repositioning were not necessary for margins of 7 or 10 mm because of the rarity of excursions at these margins. However, intervention was routinely necessary for margins of 3 and 5 mm. In simulated interventions for which the therapist could reposition the treatment couch without entering the room, the most time-efficient response protocol was to reposition the couch immediately after the prostate position was outside the treatment margins. In simulations in which the therapist had to enter the room to reposition the couch, overall treatment time could be reduced and accuracy could be increased by manually gating treatment for 11 and 21 s for 3- and 5-mm margins, respectively, before interrupting treatment to reposition the treatment couch.