Priscilla K. Stumpf

Adjuvant radiotherapy improves overall survival in patients with resected gastric adenocarcinoma: A National Cancer Data Base analysis

For patients with resectable gastric adenocarcinoma, perioperative chemotherapy and adjuvant chemoradiotherapy (CRT) are considered standard options. In the current study, the authors used the National Cancer Data Base to compare overall survival (OS) between these regimens.

Patterns of Care for Locally Advanced Pancreatic Adenocarcinoma Using the National Cancer Database

Objectives The role of radiotherapy (RT) in locally advanced pancreatic cancer (LAPC) is uncertain. This study examines patterns of care and survival outcomes of LAPC undergoing chemotherapy alone versus chemotherapy plus RT (C + RT). Methods The National Cancer Database was queried for nonmetastatic LAPC patients who received chemotherapy alone or C + RT. Results Of the 13,695 patients included, 5306 underwent chemotherapy alone and 4971, C + RT. Use of C + RT declined from 2003 to 2011 (73%–53%), whereas chemotherapy alone increased. Of those receiving RT, rates of intensity-modulated radiotherapy (IMRT) increased (27%–72%), whereas 3-dimensional (3D) RT decreased (73%–28%). Unadjusted 1-year overall survival (OS) was longer for versus chemotherapy (45.6% vs 38.7%), as was 2-year OS (12.9% vs 11.9%) (hazard ratio, 0.88; 0.85–0.91; P < 0.001). Under multivariate analysis, C + RT was associated with improved OS (hazard ratio, 0.84; 0.81–0.87; P < 0.001). On subgroup analysis comparing C + IMRT, C + 3D RT, and chemotherapy alone, 1-year OS was 49.1%, 45.1%, and 38.7%, and 2-year OS was 13.1%, 11.6%, and 11.9% accordingly. Conclusions Utilization of RT in LAPC is decreasing, whereas chemotherapy alone is increasing. Of patients undergoing RT, rates of IMRT are increasing. Whereas C + IMRT appeared to be associated with improved OS compared with chemotherapy alone, 3D RT was not.

Patterns of fractionation for patients with T2N0M0 glottic larynx cancer undergoing definitive radiotherapy in the United States

OBJECTIVES Among patients with T2N0M0 glottic larynx cancer undergoing definitive radiotherapy, recent retrospective and prospective data have suggested improved outcomes with altered fractionation over conventional fractionation (CFxn). We sought to characterize national fractionation patterns and to compare outcomes among them. MATERIALS AND METHODS We queried the National Cancer Database for T2N0M0 squamous cell carcinomas of the glottis diagnosed from 2004-2014 and managed with definitive radiotherapy. Dose-per-fraction and duration of radiotherapy were used to define cohorts undergoing CFxn, hypofractionation (HypoFxn), and hyperfractionation (HyperFxn). Logistic regression was performed to identify predictors of receiving altered fractionation. Cox regression and propensity-score matching (PSM) analyses were used to compare survival between schedules. RESULTS We abstracted 2 006 CFxn patients, 1 166 HypoFxn patients, and 161 HyperFxn patients. Fractionation patterns changed significantly from 2004 to 2014, with use of HyperFxn decreasing from 6.3% to 1.8% and use of HypoFxn increasing from 23.9% to 54.1% (p<0.001). Receipt of altered fractionation was independently associated with later year of diagnosis and higher facility volume. On Cox regression, both HypoFxn (hazard ratio [HR] for mortality 0.84, 95% confidence interval [95%CI] 0.73-0.97) and HyperFxn (HR 0.74, 95%CI 0.56-0.99) were associated with improved survival over CFxn. The survival advantage of each altered fractionation schedule over CFxn was redemonstrated on comparison of PSM groups. CONCLUSION Increasing utilization of HypoFxn for T2N0M0 glottic cancer is driving national practice patterns away from CFxn. Our findings support the use of altered fractionation, particularly HypoFxn, for patients undergoing definitive radiotherapy, although HyperFxn remains understudied in a prospective fashion.

Neural network dose models for knowledge‐based planning in pancreatic SBRT

Purpose: Stereotactic body radiation therapy (SBRT) for pancreatic cancer requires a skillful approach to deliver ablative doses to the tumor while limiting dose to the highly sensitive duodenum, stomach, and small bowel. Here, we develop knowledge‐based artificial neural network dose models (ANN‐DMs) to predict dose distributions that would be approved by experienced physicians. Methods: Arc‐based SBRT treatment plans for 43 pancreatic cancer patients were planned, delivering 30–33 Gy in five fractions. Treatments were overseen by one of two physicians with individual treatment approaches, with variations in prescribed dose, target volume delineation, and primary organs at risk. Using dose distributions calculated by a commercial treatment planning system (TPS), physician‐approved treatment plans were used to train ANN‐DMs that could predict physician‐approved dose distributions based on a set of geometric parameters (vary from voxel to voxel) and plan parameters (constant across all voxels for a given patient). Patient datasets were randomly allocated, with two‐thirds used for training, and one‐third used for validation. Differences between TPS and ANN‐DM dose distributions were used to evaluate model performance. ANN‐DM design, including neural network structure and parameter choices, was evaluated to optimize dose model performance. Results: Remarkable improvements in ANN‐DM accuracy (i.e., from > 30% to < 5% mean absolute dose error, relative to the prescribed dose) were achieved by training separate dose models for the treatment style of each physician. Increased neural network complexity (i.e., more layers, more neurons per layer) did not improve dose model accuracy. Mean dose errors were less than 5% at all distances from the PTV, and mean absolute dose errors were on the order of 5%, but no more than 10%. Dose–volume histogram errors (in cm3) demonstrated good model performance above 25 Gy, but much larger errors were seen at lower doses. Conclusions: ANN‐DM dose distributions showed excellent overall agreement with TPS dose distributions, and accuracy was substantially improved when each physicians treatment approach was taken into account by training their own dedicated models. In this manner, one could feasibly train ANN‐DMs that could predict the dose distribution desired by a given physician for a given treatment site.

Reply to Tumor localization may change the type of adjuvant treatment in gastric cancer

We read the article entitled “Adjuvant Radiotherapy Improves Overall Survival in Patients With Resected Gastric Adenocarcinoma: A National Cancer Data Base Analysis” with great interest. However, we think that some points must be clarified. Complications of radiotherapy affecting the surrounding organs are severe, although recent irradiation technologies support the role of radiotherapy in gastric cancer. Therefore, this therapy may be performed safely, but radiation oncologists must be familiar with the proper techniques for the delivery of upper abdominal radiation. Anatomically, gastric cancer may be classified as proximal (gastroesophageal junction/cardia) or distal. Differentially expressed genes may distinguish each subtype from the other. The presence of molecular heterogeneity has been shown through the existence of subtypes, which have been described on the basis of the anatomic site. A molecular classification defines 4 major genomic subtypes of gastric cancer: Epstein-Barr virus–infected tumors, tumors with microsatellite instability, genomically stable tumors, and chromosomally unstable tumors. For the primary tumor location, chromosomally unstable tumors have been found more frequently at the gastroesophageal junction and in the cardia, whereas EpsteinBarr virus–positive tumors have been found more frequently in the fundus and body. Responses to chemotherapy and radiotherapy differ for the genomic subtypes. In conclusion, resectable gastric cancer still represents a challenge for surgeons and medical/radiation oncologists. More well-designed studies are needed to investigate the optimal complementary strategy. Radiotherapy is not a completely safe treatment method. Anatomic localization is one of the most important factors determining the response to treatment. Stumpf et al have made a significant contribution to our knowledge of these aspects, and their results should be carefully interpreted and regarded with great caution. The anatomic localization of tumors must be included in the results of studies.