Introduction The aim of this study was to investigate the role of local radiotherapy in the management of epidermal growth factor receptor (EGFR)-mutant non-small cell lung cancers (NSCLCs) treated with EGFR tyrosine kinase inhibitors (TKIs). Materials and Methods Patients with stage IV EGFR-mutant NSCLC treated with radiotherapy concomitant to EGFR TKIs from May 2010 to December 2017 were retrospectively identified. Overall survival (OS) was the primary endpoints of the study. Results A total of 205 patients were enrolled in the study. One hundred eleven patients received one-time single-site radiotherapy (SSR), and 94 patients received multiple-site radiotherapy (MSR). Patients who received MSR had longer OS (median OS, 40.0 months; 95% confidence interval [CI], 29.6 to 50.4) than those who received SSR (median OS, 28.9 months; 95% CI, 24.3 to 33.5; P =0.031). Thoracic radiotherapy was associated with prolonged median OS (41.7 months, 95% CI, 29.0 to 54.4 vs 27.1 months, 95% CI 22.7 to 31.5; log-rank P <0.001). Multivariate analysis confirmed that thoracic radiotherapy was independently associated with improved OS (adjusted hazard ratio [HR], 0.514; 95% CI 32.3% to 81.8%; P =0.005). Conclusion MSR improves survival outcomes in patients with advanced-stage, EGFR-mutant, lung adenocarcinoma, with thoracic radiotherapy having the most significant effect on prognosis.

ObjectiveTo evaluate the incidental coverage dose to the internal mammary nodes (IMN) in patients treated with postmastectomy radiotherapy (PMRT) and its relationship with the treatment plan.Patients and methodsWe retrospectively analyzed 138 patients undergoing PMRT and divided them into three groups: three-dimensional conformal radiotherapy (3D-CRT), field-in-field forward intensity-modulated radiotherapy (F-IMRT), and inverse intensity-modulated radiotherapy (I-IMRT). The IMN were contoured according to the Radiation Therapy Oncology Group consensus and not included in the planning target volume. We analyzed incidental IMN dose coverage and its relationship with the lung and heart.ResultsThe mean dose (Dmean) to the IMN was 32.85 Gy for all patients, and the dose delivered to the IMN showed no differences in 3D-CRT, F-IMRT, and I-IMRT (33.80, 29.65, and 32.95 Gy, respectively). In addition, 10.42%, 2.04%, and 9.76% of patients achieved ≥45 Gy with 3D-CRT, F-IMRT, and I-IMRT, respectively. No differences were evident among the three treatment plans regarding IMN dose in the first three intercostal spaces (ICS1–3). The Dmean, V20, V30, V40, and V50 of ICS2 and ICS3 were superior to those of ICS1 for all three plans. For 3D-CRT, a moderate positive correlation was evident between the Dmean to the IMN and the Dmean to the heart. For F-IMRT and I-IMRT, positive correlations were evident between the Dmean of the IMN and the Dmean and V20 of the lung.ConclusionThe mean incidental dose to the IMN for IMRT (F-IMRT and I-IMRT) and 3D-CRT after modified radical mastectomy was insufficient to treat subclinical disease. A substantial dose was delivered to the IMN in some patients. Higher incidental doses to the IMN were associated with a higher heart mean dose for 3D-CRT and a higher dose to the lung for IMRT. Future prospective studies should further explore subgroups that do not require IMN irradiation.

Objective: To compare differences in setup error assessment and correction between planar kilovolt images and cone beam computed tomography images for external beam partial breast irradiation during free breathing. Methods: Nineteen patients who received external beam partial breast irradiation after breast-conserving surgery were recruited. Interfraction setup error was acquired using planar kilovolt images and cone beam computed tomography. After online setup correction, the residual error was calculated, and the setup error was compared. The residual error and setup margin were quantified for planar kilovolt and cone beam computed tomography images. Results: The largest setup error was observed in the anteroposterior direction for both cone beam computed tomography and planar kilovolt imaging (−1.45 mm, 1.74 mm). The cone beam computed tomography–based setup error (systematic error [Σ]) was less than the planar kilovolt images based on Σ in the anteroposterior direction (–1.2 mm vs 2.00 mm; P = .005), and no significant differences were observed for random error (σ) in 3 dimensions ( P = .948, .376, .314). After online setup correction, cone beam computed tomography significantly reduced the residual setup error compared with planar kilovolt images in the anteroposterior direction (Σ: −0.20 mm vs 0.50 mm, P = .008; σ: 0.45 mm vs 1.34 mm, P = .002). The cone beam computed tomography–based setup margin was smaller than the planar kilovolt image-based setup margin in the anteroposterior direction (−1.39 mm vs 5.57 mm, P = .003; 0.00 mm vs 3.20 mm, P = .003). Conclusions: Discrepancy between the setup errors observed with planar kilovolt and cone beam computed tomography was obvious in the anteroposterior direction. Compared to cone beam computed tomography, the elapsed treatment time was smaller when the initial alignment used kilovolt planar imaging. Whether using planar kilovolt or cone beam computed tomography, residual errors can be reduced to 1.5 mm for external beam partial breast irradiation procedures.