Harkenrider, Matthew M.

Abstract Background Markerless tumor tracking (MTT) using single‐energy (SE) kilovoltage (kV) imaging has been proposed as a technique for lung tumor motion management. However, bony structures can obscure the tumor and make tracking challenging. Dual‐energy (DE) subtraction imaging can suppress the bone and therefore enhance tumor visibility, potentially improving tracking accuracy. Purpose To develop a machine learning model that predicts the reliability of MTT for SE and DE imaging using phantom and patient data. Methods Images were collected at high and low energy (120/60 kV) using fast‐kV switching on dynamic thorax phantom with 5–15 diameter targets and 20 lung cancer patients, including one patient with two tumor locations. DE images were then generated off‐line using weighted logarithmic subtraction. Separately, a template‐based tracking algorithm was used to track the tumor on both phantom and patient images. For phantom data, the ground truth (GT) was defined by the cos 4 waveform programmed into the motion controller. Since the GT was unknown for patient data, it was estimated using a Kalman filter and further refined using visual inspection. The tracking success rate (TSR) was subsequently calculated. Based on TSR, patients were stratified into three tracking groups: good, moderate, and poor. A logistic regression (LR) model was implemented with tracking features, including match score, peak to side‐peak ratio (PSR), x ‐ and y ‐velocity. To address the distribution differences between phantom and patient data, a domain indicator and interaction features were also incorporated. The model was trained on phantom and 17 patient data using leave‐one‐out cross‐validation, and its performance was evaluated on three held‐out patients, one from each tracking group. Separate LR models were developed for each imaging modality, resulting in SE‐LR and DE‐LR models. For each held‐out patient, the confusion matrix, sensitivity, and specificity were reported. Results For both SE‐LR and DE‐LR models, PSR has the largest coefficient value, indicating that it is the most influential factor when determining the reliability of tracking. At a specificity of 95%, the DE‐LR model achieved significantly higher sensitivity than the SE‐LR model (0.908 vs. 0.731, p < 0.01 for phantom data and 0.657 vs. 0.589, p < 0.01 for patient data). In the three held‐out patients, the DE‐LR model outperformed the SE‐LR model across all metrics. The largest gains were found in the patient with good tracking quality. Conclusions The proposed LR models successfully predicted the reliability of SE‐ and DE‐MTT. Across phantom and patient datasets, the DE‐LR model consistently demonstrated improved performance compared with the SE‐LR model, suggesting that DE imaging may be beneficial for MTT applications.

BACKGROUNDStereotactic body radiotherapy (SBRT) is the standard of care for patients with nonoperative, early‐stage non–small cell lung cancer (NSCLC) measuring < 5 cm, but its use among patients with tumors measuring ≥5 cm is considerably less defined, with the existing literature limited to small, single‐institution reports. The current multi‐institutional study reported outcomes evaluating the largest such population reported to date.METHODSClinical/treatment characteristics, outcomes, toxicities, and patterns of failure were assessed in patients with primary NSCLC measuring ≥5 cm without evidence of distant/lymph node metastasis who underwent SBRT using ≤5 fractions. Statistics included Kaplan‐Meier survival analyses and univariate/multivariate Cox proportional hazards models.RESULTSA total of 92 patients treated from 2004 through 2016 were analyzed from 12 institutions. The median follow‐up was 12 months (15 months in survivors). The median age and tumor size among the patients were 73 years (range, 50‐95 years) and 5.4 cm (range, 5.0‐7.5 cm), respectively. The median dose/fractionation was 50 Gray/5 fractions. The actuarial local control rates at 1 year and 2 years were 95.7% and 73.2%, respectively. The disease‐free survival rate was 72.1% and 53.5%, respectively, at 1 year and 2 years. The 1‐year and 2‐year disease‐specific survival rates were 95.5% and 78.6%, respectively. The median, 1‐year, and 2‐year overall survival rates were 21.4 months, 76.2%, and 46.4%, respectively. On multivariate analysis, lung cancer history and pre‐SBRT positron emission tomography maximum standardized uptake value were found to be associated with overall survival. Posttreatment failures were most commonly distant (33% of all disease recurrences), followed by local (26%) and those occurring elsewhere in the lung (23%). Three patients had isolated local failures. Grade 3 to 4 toxicities included 1 case (1%) and 4 cases (4%) of grade 3 dermatitis and radiation pneumonitis, respectively (toxicities were graded according to the Common Terminology Criteria for Adverse Events [version 4.0]). Grades 2 to 5 radiation pneumonitis occurred in 11% of patients. One patient with a tumor measuring 7.5 cm and a smoking history of 150 pack‐years died of radiation pneumonitis.CONCLUSIONSThe results of the current study, which is the largest study of patients with NSCLC measuring ≥5 cm reported to date, indicate that SBRT is a safe and efficacious option. Cancer 2017;123:688–696. © 2016 American Cancer Society.