IMPORTANCE Cutaneous squamous cell carcinoma (CSCC) is one of the most common malignant tumors worldwide. There is conflicting evidence regarding the indications for and benefits of adjuvant radiation therapy for advanced CSCC tumors of the head and neck. OBJECTIVE To assess indications for adjuvant radiation therapy in patients with CSCC. DESIGN, SETTING, AND PARTICIPANTS Retrospective analysis of 349 patients with head and neck CSCC treated with primary resection with or without adjuvant radiation therapy at 2 tertiary referral centers from January 1, 2008, to June 30, 2016. MAIN OUTCOMES AND MEASURES Data were compared between treatment groups with a χ 2 analysis. Disease-free survival (DFS) and overall survival (OS) were analyzed using a Kaplan-Meier survival analysis with log-rank test and a Cox proportional hazards multivariate regression. RESULTS A total of 349 patients had tumors that met the inclusion criteria (mean [SD] age, 70 [12] years; age range, 32-94 years; 302 [86.5%] male), and 191 (54.7%) received adjuvant radiation therapy. The 5-year Kaplan-Meier estimates were 59.4% for DFS and 47.4% for OS. Patients with larger, regionally metastatic, poorly differentiated tumors with perineural invasion (PNI) and younger immunosuppressed patients were more likely to receive adjuvant radiation therapy. On Cox proportional hazards multivariate regression, patients with periorbital tumors (hazard ratio [HR], 2.48; 95% CI, 1.00-6.16), PNI (HR, 1.90; 95% CI, 1.12-3.19), or N2 or greater nodal disease (HR, 2.16; 95% CI, 1.13-4.16) had lower DFS. Immunosuppressed patients (HR, 2.17; 95% CI, 1.12-4.17) and those with N2 or greater nodal disease (HR, 2.43; 95% CI, 1.42-4.17) had lower OS. Adjuvant radiation therapy was associated with improved OS for the entire cohort (HR, 0.59; 95% CI, 0.38-0.90). In a subset analysis of tumors with PNI, adjuvant radiation therapy was associated with improved DFS (HR, 0.47; 95% CI, 0.23-0.93) and OS (HR, 0.44; 95% CI, 0.24-0.86). Adjuvant radiation therapy was also associated with improved DFS (HR, 0.36; 95% CI, 0.15-0.84) and OS (HR, 0.30; 95% CI, 0.15-0.61) in patients with regional disease. CONCLUSIONS AND RELEVANCE Among patients with advanced CSCC, receipt of adjuvant radiation therapy was associated with improved survival in those with PNI and regional disease.

Purpose: Increased rates of toxicity have been described following stereotactic body radiotherapy (SBRT) for central lung tumors within 2 cm of the proximal bronchial tree (PBT). Recent studies have defined a new class of "ultracentral" tumors. We report our experience treating ultracentral, central, and paramediastinal tumors with SBRT and compare toxicity, disease control, and survival. Methods and Materials:We reviewed the records of patients with central lung tumors treated with SBRT between September 2009-July 2017. Tumors were classified as central if within 2 cm of the PBT, ultracentral if the planning target volume touched the PBT or esophagus, and paramediastinal if touching mediastinal pleura. Actuarial rates of grade 2+ and 3+ toxicity, local control (LC), and overall survival (OS) were assessed using the Kaplan-Meier method and compared using a log-rank test. Toxicity was scored with CTCAE V4.03. Results:We identified 68 patients with 69 central lung tumors, including 14 ultracentral, 15 paramediastinal, and 39 central tumors. Fifty-three patients were treated for early stage lung cancer and 15 for lung metastases. Prescribed dose ranged from 40-60 Gy over 3-8 fractions. Most patients were treated using five-fractions (83%) followed by eight-fractions (10%). Median follow up was 19.7 months (range: 3.3-78.3). Two-year estimates of LC (89%, 85%, and 93%; p=0.72) and OS (76%, 73%, and 72%; p=0.75) for ultracentral, central, and paramediastinal tumors were similar. Ultracentral tumors had increased risk of grade 2+ toxicity (57.6% vs. 14.2% vs. 7.1%, p= 0.007) at 2 years. One ultracentral patient developed grade 5 respiratory failure. Conclusion:Oncologic outcomes following SBRT for ultracentral, central, and paramediastinal lung tumors were similar, with LC exceeding 85% at 2 years using predominantly 5-fraction schedules. Ultracentral lung tumors were associated with increased risk of toxicity in our patient cohort. Additional studies are needed to minimize toxicity for ultracentral tumors.

Purpose: To evaluate the accuracy of deep-learning-based auto-segmentation of the superior constrictor, middle constrictor, inferior constrictor, and larynx in comparison with a traditional multi-atlas-based method. Methods and Materials: One hundred and five computed tomography image datasets from 83 head and neck cancer patients were retrospectively collected and the superior constrictor, middle constrictor, inferior constrictor, and larynx were analyzed for deep-learning versus multi-atlas-based segmentation. Eighty-three computed tomography images (40 diagnostic computed tomography and 43 planning computed tomography) were used for training the convolutional neural network, and for atlas-based model training. The remaining 22 computed tomography datasets were used for validation of the atlas-based auto-segmentation versus deep-learning-based auto-segmentation contours, both of which were compared with the corresponding manual contours. Quantitative measures included Dice similarity coefficient, recall, precision, Hausdorff distance, 95th percentile of Hausdorff distance, and mean surface distance. Dosimetric differences between the auto-generated contours and manual contours were evaluated. Subjective evaluation was obtained from 3 clinical observers to blindly score the autosegmented structures based on the percentage of slices that require manual modification. Results: The deep-learning-based auto-segmentation versus atlas-based auto-segmentation results were compared for the superior constrictor, middle constrictor, inferior constrictor, and larynx. The mean Dice similarity coefficient values for the 4 structures were 0.67, 0.60, 0.65, and 0.84 for deep-learning-based auto-segmentation, whereas atlas-based auto-segmentation has Dice similarity coefficient results at 0.45, 0.36, 0.50, and 0.70, respectively. The mean 95th percentile of Hausdorff distance (cm) for the 4 structures were 0.41, 0.57, 0.59, and 0.54 for deep-learning-based auto-segmentation, but 0.78, 0.95, 0.96, and 1.23 for atlas-based auto-segmentation results, respectively. Similar mean dose differences were obtained from the 2 sets of autosegmented contours compared to manual contours. The dose–volume discrepancies and the average modification rates were higher with the atlas-based auto-segmentation contours. Conclusion: Swallowing-related structures are more accurately generated with DL-based versus atlas-based segmentation when compared with manual contours.

Purpose This study aimed to investigate radiomic features extracted from magnetic resonance imaging (MRI) scans performed before and after neoadjuvant chemoradiotherapy (nCRT) in predicting response of locally advanced rectal cancer (LARC). Methods and Materials Thirty-nine patients who underwent nCRT for LARC were included, with 294 radiomic features extracted from MRI that was performed before (pre-CRT) and 6 to 8 weeks after completing nCRT (post-CRT). Based on tumor regression grade (TRG), 26 patients were classified as having a histopathologic good response (GR; TRG 0-1) and 13 as non-GR (TRG 2-3). Tumor downstaging (T-downstaging) occurred in 25 patients. Univariate analyses were performed to assess potential radiomic and delta-radiomic predictors for TRG in pathologic complete response (pCR) versus non-pCR, GR versus non-GR, and T-downstaging. The support vector machine-based multivariate model was used to select the best predictors for TRG and T-downstaging. Results We identified 13 predictive features for pCR versus non-pCR, 14 for GR versus non-GR, and 16 for T-downstaging. Pre-CRT gray-level run length matrix nonuniformity, pre-CRT neighborhood intensity difference matrix (NIDM) texture strength, and post-CRT NIDM busyness predicted all 3 treatment responses. The best predictor for GR versus non-GR was pre-CRT global minimum combined with clinical N stage in the multivariate analysis. The best predictor for T-downstaging was the combination of pre-CRT gray-level co-occurrence matrix correlation, NIDM-texture strength, and gray-level co-occurrence matrix variance. The pre-CRT, post-CRT, and delta radiomic-based models had no significant difference in predicting all 3 responses. Conclusions Pre-CRT MRI, post-CRT MRI, and delta radiomic-based models have the potential to predict tumor response after nCRT in LARC. These data, if validated in larger cohorts, can provide important predictive information to aid in clinical decision making.

No abstract