PURPOSE:This guideline systematically reviews the evidence for treatment of pancreatic cancer with radiation in the adjuvant, neoadjuvant, definitive, and palliative settings and provides recommendations on indications and technical considerations.METHODS AND MATERIALS:The American Society for Radiation Oncology convened a task force to address 7 key questions focused on radiation therapy, including dose fractionation and treatment volumes, simulation and treatment planning, and prevention of radiation-associated toxicities. Recommendations were based on a systematic literature review and created using a predefined consensus-building methodology and system for grading evidence quality and recommendation strength.RESULTS:The guideline conditionally recommends conventionally fractionated or stereotactic body radiation for neoadjuvant and definitive therapy in certain patients and conventionally fractionated regimens for adjuvant therapy. The task force suggests a range of appropriate dose-fractionation schemes and provides recommendations on target volumes and sequencing of radiation and chemotherapy. Motion management, daily image guidance, use of contrast, and treatment with modulated techniques are all recommended. The task force supported prophylactic antiemetic medication, and patients may also benefit from medications to reduce acid secretion.CONCLUSIONS:The role of radiation in the management of pancreatic cancer is evolving, with many ongoing areas of active investigation. Radiation therapy is likely to become even more important as new systemic therapies are developed and there is increased focus on controlling local disease. It is important that the nuances of available data are discussed with patients and families and that care be coordinated in a multidisciplinary fashion.

Purpose: To assess the efficacy of preoperative positron emission tomography (PET) to stage the ipsilateral hilum in resected non-small cell lung cancer (NSCLC).Methods and materials: All patients who underwent surgery for NSCLC between 1995 and 2008 were evaluated. Patients who underwent preoperative PET imaging at our institution and had hilar nodal samplingwere included. Thosewhose primary tumors extended to the hilumorwho received preoperative chemotherapy or radiation therapy were excluded. All PET studies were interpreted by an attending nuclear medicine radiologist and were scored as positive or negative in the hilum or peribronchial area based on visual analysis alone. A 2-sided Fisher exact test compared patient subgroups.Results: During the time interval, 1558 patients underwent surgery for NSCLC, of whom 484 were eligible for this analysis. The ipsilateral hilum was positive on preoperative PET in 107 patients. The median number of N1 lymph nodes sampled was 4 (range, 1-31). Positive ipsilateralN1 lymph nodes were identified pathologically in 91 patients (19%). Among the 91 patients with involved N1 lymph nodes, 40 were PET positive resulting in a sensitivity of 44%. Among 393 patients without pathologic involvement of hilar lymph nodes, 326 were PET negative resulting in a specificity of 83%. The positive predictive and negative predictive values were 37% and 86%, respectively.Conclusions: Positron emission tomography appears to have limitations in staging the ipsilateral hilar lymph nodes. Invasive sampling is appropriate if treatment would differ based on the nodal status. (C) 2015 American Society for Radiation Oncology. Published by Elsevier Inc. All rights reserved.

We wish to thank Dr Giron and colleagues for their thoughtful comments on our article. 1 Giron J. Lacout A. Marcy P.-Y. Accuracy of positron emission tomography may be improved when combined with postcontrast high-resolution computed tomography scan. In regard to Pepak et al. in: Pract Radiat Oncol.2015: e575-e576 Google Scholar , 2 Pepek J.M. Marks L.B. Berry M.F. Ready N.E. et al. Accuracy of positron emission tomography in identifying hilar (N1) lymph node involvement in non-small cell lung cancer: Implications for stereotactic body radiation therapy. Pract Radiat Oncol. 2014; ([pii: S1879-8500(14)00115-5]) PubMed Google Scholar As the authors pointed out, data from our institution and others have demonstrated that standard positron emission tomography–computed tomography (PET-CT) imaging systems have limitations in staging hilar lymph nodes in non-small cell lung cancer. The causes are almost certainly multifactorial. It was suggested that a contrast-enhanced CT scan combined with PET could improve imaging of the hilum by providing better differentiation of anatomical structures and localization of fluorine-18 fluorodeoxyglucose abnormalities. Though plausible, as far as we are aware, there are no data comparing the accuracy of contrast-enhanced PET-CT to noncontrast PET-CT in this clinical scenario. Many medical centers in the United States, including our own, continue to primarily use noncontrast-enhanced imaging with PET. Further study into the potential diagnostic benefit of contrast enhanced PET-CT would be valuable. Accuracy of positron emission tomography may be improved when combined with postcontrast high-resolution computed tomography scanIn Regard to Pepek et alPractical Radiation OncologyVol. 5Issue 5PreviewIn relation to the article by Pepek and colleagues published in the October 2014 issue,1 we wish to congratulate the authors and make some comments. Full-Text PDF

Contemporary randomized trials have demonstrated that radiation therapy combined with chemotherapy and surgery improves survival in both the neoadjuvant and adjuvant treatment of gastroesophageal cancers. Consequently, radiation treatment planning and administration have taken on an added importance to ensure optimal outcomes as well as minimize treatment-related morbidity. This article highlights recent technical advances and considerations for radiation therapy planning for gastroesophageal junction tumors.

Background To examine toxicity and outcomes for patients treated with preoperative chemoradiotherapy (CRT) for gastric cancer. Methods Patients with gastroesophageal (GE) junction (Siewert type II and III) or gastric adenocarcinoma who underwent neoadjuvant CRT followed by planned surgical resection at Duke University between 1987 and 2009 were reviewed. Overall survival (OS), local control (LC) and disease-free survival (DFS) were estimated using the Kaplan-Meier method. Toxicity was graded according to the Common Toxicity Criteria for Adverse Events version 4.0. Results Forty-eight patients were included. Most (73%) had proximal (GE junction, cardia and fundus) tumors. Median radiation therapy dose was 45 Gy. All patients received concurrent chemotherapy. Thirty-six patients (75%) underwent surgery. Pathologic complete response and R0 resection rates were 19% and 86%, respectively. Thirty-day surgical mortality was 6%. At 42 months median follow-up, 3-year actuarial OS was 40%. For patients undergoing surgery, 3-year OS, LC and DFS were 50%, 73% and 41%, respectively. Conclusions Preoperative CRT for gastric cancer is well tolerated with acceptable rates of perioperative morbidity and mortality. In this patient cohort with primarily advanced disease, OS, LC and DFS rates in resected patients are comparable to similarly staged, adjuvantly treated patients in randomized trials. Further study comparing neoadjuvant CRT to standard treatment approaches for gastric cancer is indicated.

Department of Radiation Oncology, Box 3085, Duke University Medical Center, Durham, NC 27710, USA *Author for correspondence: Tel.: +1 919 668 7336; Fax: +1 919 668 7345; czito001@mc.duke.edu The incidence of esophageal cancer has increased dramatically over the last few decades [1,2]. Despite substantial advances in surgery, chemotherapy and radiation therapy, mortality rates for this disease remain high [3]. The optimal treatment approach for patients with locally advanced disease remains controversial. Preoperative chemoradiotherapy is established as a standard of care in rectal cancer [4] and has shown promise in the treatment of other gastrointestinal malignancies. Preoperative chemo radiotherapy offers several potential biological and technical advantages when compared with either surgery alone, or with postoperative therapy delivery, including:

Ampullary carcinoma is a rare malignancy. Despite radical resection, survival rates remain low with high rates of local failure. We performed a single-institution outcomes analysis to define the role of concurrent chemoradiotherapy (CRT) in addition to surgery.

Purpose/Objective(s)18-fluorine fluorodeoxyglucose positron emission tomography (PET) is increasingly utilized to stage non-small cell lung cancer (NSCLC). Several studies have shown that the maximum standardized uptake value (SUVmax) of the primary tumor is prognostic for overall survival, though this remains controversial. This analysis evaluated whether SUVmax of the primary tumor independently predicts for regional lymph node involvement (pN1-N2) and for subsequent local/regional recurrence (LRR).Materials/MethodsAll patients with resected Stage I-IIIA NSCLC between 1995 and 2008 who did not receive any preoperative therapy and had a preoperative PET with a reported SUVmax of the primary tumor were evaluated from a single institution database. Those without any lymph node sampling were excluded when evaluating factors associated with regional lymph node involvement. Patients who received postoperative radiation therapy or had positive surgical margins were excluded when evaluating factors associated with LRR. Disease recurrence at the surgical margin or within ipsilateral hilar and/or mediastinal lymph nodes was considered a LRR. Disease recurrence in the contralateral hilum, supraclavicular fossae, ipsilateral lung parenchyma, or elsewhere was defined as a distant metastasis. Univariate and multivariate analyses, using the median SUVmax of our cohort as a threshold (SUVmax 9.7) and SUVmax as a continuous variable, were performed to evaluate independent risk factors of regional lymph node involvement and LRR.ResultsAmong 1558 patients undergoing surgery during the time interval, 236 were evaluable for risk of regional lymph node involvement and 234 for LRR risk. As a continuous variable, SUVmax was not associated with a higher risk of regional lymph node involvement on univariate (HR 0.99, p = 0.93) or multivariate (HR 0.95, p = 0.06) analysis. Similarly, SUVmax as a continuous variable was not associated with LRR on univariate (HR 1.03, p = 0.12) or multivariate (HR 1.02, p = 0.55) analysis. When using median SUVmax as a threshold, SUVmax was not predictive for regional lymph node involvement on univariate analysis (p = 0.39). Although SUVmax > 9.7 was associated with higher LRR risk on univariate analysis (20% vs 11%, p = 0.047), it was not significant on multivariate analysis (HR 0.60, p = 0.29).ConclusionsOn multivariate analysis, SUVmax of the primary tumor did not predict for regional lymph node involvement or LRR in resected NSCLC. Purpose/Objective(s)18-fluorine fluorodeoxyglucose positron emission tomography (PET) is increasingly utilized to stage non-small cell lung cancer (NSCLC). Several studies have shown that the maximum standardized uptake value (SUVmax) of the primary tumor is prognostic for overall survival, though this remains controversial. This analysis evaluated whether SUVmax of the primary tumor independently predicts for regional lymph node involvement (pN1-N2) and for subsequent local/regional recurrence (LRR). 18-fluorine fluorodeoxyglucose positron emission tomography (PET) is increasingly utilized to stage non-small cell lung cancer (NSCLC). Several studies have shown that the maximum standardized uptake value (SUVmax) of the primary tumor is prognostic for overall survival, though this remains controversial. This analysis evaluated whether SUVmax of the primary tumor independently predicts for regional lymph node involvement (pN1-N2) and for subsequent local/regional recurrence (LRR). Materials/MethodsAll patients with resected Stage I-IIIA NSCLC between 1995 and 2008 who did not receive any preoperative therapy and had a preoperative PET with a reported SUVmax of the primary tumor were evaluated from a single institution database. Those without any lymph node sampling were excluded when evaluating factors associated with regional lymph node involvement. Patients who received postoperative radiation therapy or had positive surgical margins were excluded when evaluating factors associated with LRR. Disease recurrence at the surgical margin or within ipsilateral hilar and/or mediastinal lymph nodes was considered a LRR. Disease recurrence in the contralateral hilum, supraclavicular fossae, ipsilateral lung parenchyma, or elsewhere was defined as a distant metastasis. Univariate and multivariate analyses, using the median SUVmax of our cohort as a threshold (SUVmax 9.7) and SUVmax as a continuous variable, were performed to evaluate independent risk factors of regional lymph node involvement and LRR. All patients with resected Stage I-IIIA NSCLC between 1995 and 2008 who did not receive any preoperative therapy and had a preoperative PET with a reported SUVmax of the primary tumor were evaluated from a single institution database. Those without any lymph node sampling were excluded when evaluating factors associated with regional lymph node involvement. Patients who received postoperative radiation therapy or had positive surgical margins were excluded when evaluating factors associated with LRR. Disease recurrence at the surgical margin or within ipsilateral hilar and/or mediastinal lymph nodes was considered a LRR. Disease recurrence in the contralateral hilum, supraclavicular fossae, ipsilateral lung parenchyma, or elsewhere was defined as a distant metastasis. Univariate and multivariate analyses, using the median SUVmax of our cohort as a threshold (SUVmax 9.7) and SUVmax as a continuous variable, were performed to evaluate independent risk factors of regional lymph node involvement and LRR. ResultsAmong 1558 patients undergoing surgery during the time interval, 236 were evaluable for risk of regional lymph node involvement and 234 for LRR risk. As a continuous variable, SUVmax was not associated with a higher risk of regional lymph node involvement on univariate (HR 0.99, p = 0.93) or multivariate (HR 0.95, p = 0.06) analysis. Similarly, SUVmax as a continuous variable was not associated with LRR on univariate (HR 1.03, p = 0.12) or multivariate (HR 1.02, p = 0.55) analysis. When using median SUVmax as a threshold, SUVmax was not predictive for regional lymph node involvement on univariate analysis (p = 0.39). Although SUVmax > 9.7 was associated with higher LRR risk on univariate analysis (20% vs 11%, p = 0.047), it was not significant on multivariate analysis (HR 0.60, p = 0.29). Among 1558 patients undergoing surgery during the time interval, 236 were evaluable for risk of regional lymph node involvement and 234 for LRR risk. As a continuous variable, SUVmax was not associated with a higher risk of regional lymph node involvement on univariate (HR 0.99, p = 0.93) or multivariate (HR 0.95, p = 0.06) analysis. Similarly, SUVmax as a continuous variable was not associated with LRR on univariate (HR 1.03, p = 0.12) or multivariate (HR 1.02, p = 0.55) analysis. When using median SUVmax as a threshold, SUVmax was not predictive for regional lymph node involvement on univariate analysis (p = 0.39). Although SUVmax > 9.7 was associated with higher LRR risk on univariate analysis (20% vs 11%, p = 0.047), it was not significant on multivariate analysis (HR 0.60, p = 0.29). ConclusionsOn multivariate analysis, SUVmax of the primary tumor did not predict for regional lymph node involvement or LRR in resected NSCLC. On multivariate analysis, SUVmax of the primary tumor did not predict for regional lymph node involvement or LRR in resected NSCLC.

The contemporary treatment of anal cancer is combined-modality therapy with radiation therapy, fluorouracil, and mitomycin. This therapy results in long-term disease-free survival and sphincter preservation in the majority of patients. Tempering these positive results is the high rate of treatment-related morbidity associated with chemoradiation therapy for anal cancer. The use of intensity-modulated radiation therapy (IMRT) has the potential to reduce acute and chronic treatment-related toxicity, minimize treatment breaks, and potentially improve disease-related outcomes by permitting radiation dose escalation in selected cases.

Introduction: To evaluate how well the tumor, node, metastasis (TNM) 6 and TNM 7 staging systems predict rates of local/regional recurrence (LRR) after surgery alone for non-small cell lung cancer.Methods: All patients who underwent surgery for non-small cell lung cancer at Duke between 1995 and 2005 were reviewed. Those undergoing sublobar resections, with positive margins or involvement of the chest wall, or those who received any chemotherapy or radiation therapy (RT) were excluded. Disease recurrence at the surgical margin, or within ipsilateral hilar and/or mediastinal lymph nodes, was considered as a LRR. Stage was assigned based on both TNM 6 and TNM 7. Rates of LRR were estimated using the Kaplan-Meier method. A Cox regression analysis evaluated the hazard ratio of LRR by stage within TNM 6 and TNM 7.Results: A total of 709 patients were eligible for the analysis. Median follow-up was 32 months. For all patients, the 5-year actuarial risk of LRR was 23%. Conversion from TNM 6 to TNM 7 resulted in 21% stage migration (upstaging in 13%; downstaging in 8%). Five-year rates of LRR for stages IA, IB, IIA, IIB, and IIIA disease using TNM 6 were 16%, 26%, 43%, 35%, and 40%, respectively. Using TNM 7, corresponding rates were 16%, 23%, 37%, 39%, and 30%, respectively. The hazard ratios for LRR were statistically different for IA and IB in both TNM 6 and 7 but were also different for IB and IIA in TNM 7.Conclusions: LRR risk increases monotonically for stages IA to IIB in the new TNM 7 system. This information might be valuable when designing future studies of postoperative RT.

99 Background: To examine acute toxicity and outcomes for patients treated with preoperative chemoradiotherapy (CRT) for gastric cancer. Methods: Patients with gastroesophageal (GE) junction (Siewert type II and III) or stomach adenocarcinoma who underwent curative intent CRT followed by planned surgical resection at Duke University between 1987 and 2009 were reviewed. Tumors were staged according to AJCC 6 th edition. Local recurrence was defined as radiographic or biopsy- proven disease within the radiation treatment field. Overall survival (OS), local control (LC) and disease-free survival (DFS) were estimated using the Kaplan-Meier method. Toxicity was graded according to CTCAE v4.0. Results: Forty-eight patients (60% stage III, 8% stage IV) were included. Most (73%) had proximal (GE junction, cardia and fundus) tumors. Thirty-five percent had signet ring histology, 52% had poorly differentiated tumors and 10% had linitis plastica. Median age was 60 years and median RT dose was 45 Gy. All patients received concurrent chemotherapy (CT) with 40 (83%) receiving 5-FU-based CT. Rates of acute > grade 2 hematologic and non-hematologic toxicity were 38% and 10%, respectively. Six patients (13%) required treatment break and two (4%) were unable to complete the prescribed treatment course. Thirty-six patients (75%) underwent surgery. Patients did not undergo surgery due to distant metastases at laparotomy or restaging (n=9), patient refusal (n=2) or poor performance status (n=1). Pathologic complete response and R0 resection rates were 19% and 86%, respectively. Thirty-day surgical mortality was 6%. At 42 months median follow-up, 3-year actuarial OS for all patients was 40%. For those undergoing surgery, 3-year OS, LC and DFS were 50%, 73% and 41%, respectively. Conclusions: Preoperative CRT for gastric cancer is reasonably well tolerated with acceptable rates of perioperative morbidity and mortality. In this patient cohort with advanced disease, LC, DFS and OS rates in resected patients are comparable to similarly staged, adjuvantly treated historic controls. Further study comparing neoadjuvant CRT to standard treatment approaches for gastric cancer is indicated. No significant financial relationships to disclose.

Purpose/Objective(s)Multiple investigations have evaluated the accuracy of 18-fluorine fluorodeoxyglucose positron emission tomography (PET) to stage the mediastinum, but few studies have specifically studied its ability to stage the hilum. The purpose of this analysis is to determine whether preoperative PET accurately predicts for pathologic hilar lymph node involvement.Materials/MethodsAll patients who underwent surgery for NSCLC between 1995 and 2008 were evaluated. Patients who underwent preoperative chemotherapy or chemoradiotherapy were excluded. Patients who did not undergo preoperative PET imaging at our institution, those who had primary tumors that extended to the hilum, and those with no hilar nodal sampling were also excluded. All PET studies were interpreted by an attending nuclear medicine radiologist and were scored as positive or negative in the hilum/peribronchial area based on visual analysis alone. A two-sided Fisher's exact test compared patient subgroups.ResultsOne thousand five hundred fifty-eight patients underwent surgery for NSCLC, of whom 679 (44%) had preoperative PET imaging. After excluding those whose primary tumors extended to the hilum (n = 44) and those with no hilar nodal sampling (n = 130), 505 patients were included in the analysis, 92 (18%) of whom had positive ipsilateral hilar lymph nodes. The median number of hilar lymph nodes sampled was four (range, 1-31). The ipsilateral hilum was positive on preoperative PET in 108 patients. Among the 92 patients with involved N1 lymph nodes, 40 were PET positive resulting in a sensitivity of 43%. Among 413 pN0 patients, 345 were PET negative resulting in a specificity of 84%. The positive predictive and negative predictive values were 37% and 87%, respectively. The sensitivity was not different when comparing patients with one involved N1 lymph node (22/51, 43%) vs. more than one involved N1 lymph node (18/41, 44%, p = 1). Sensitivity (51% vs. 37%, p = 0.21) and specificity (82% vs. 85%, p = 0.50) were not different between right-sided and left-sided lung tumors, respectively.ConclusionsThe sensitivity, specificity, positive predictive value, and negative predictive value of PET for staging the ipsilateral hilum was 43%, 84%, 37%, and 87%, respectively. These numbers are generally inferior to those reported for mediastinal PET staging. This analysis suggests that PET may be insufficient to adequately evaluate the hilum, which has potential implications for elective nodal irradiation and stereotactic body radiation therapy. Purpose/Objective(s)Multiple investigations have evaluated the accuracy of 18-fluorine fluorodeoxyglucose positron emission tomography (PET) to stage the mediastinum, but few studies have specifically studied its ability to stage the hilum. The purpose of this analysis is to determine whether preoperative PET accurately predicts for pathologic hilar lymph node involvement. Multiple investigations have evaluated the accuracy of 18-fluorine fluorodeoxyglucose positron emission tomography (PET) to stage the mediastinum, but few studies have specifically studied its ability to stage the hilum. The purpose of this analysis is to determine whether preoperative PET accurately predicts for pathologic hilar lymph node involvement. Materials/MethodsAll patients who underwent surgery for NSCLC between 1995 and 2008 were evaluated. Patients who underwent preoperative chemotherapy or chemoradiotherapy were excluded. Patients who did not undergo preoperative PET imaging at our institution, those who had primary tumors that extended to the hilum, and those with no hilar nodal sampling were also excluded. All PET studies were interpreted by an attending nuclear medicine radiologist and were scored as positive or negative in the hilum/peribronchial area based on visual analysis alone. A two-sided Fisher's exact test compared patient subgroups. All patients who underwent surgery for NSCLC between 1995 and 2008 were evaluated. Patients who underwent preoperative chemotherapy or chemoradiotherapy were excluded. Patients who did not undergo preoperative PET imaging at our institution, those who had primary tumors that extended to the hilum, and those with no hilar nodal sampling were also excluded. All PET studies were interpreted by an attending nuclear medicine radiologist and were scored as positive or negative in the hilum/peribronchial area based on visual analysis alone. A two-sided Fisher's exact test compared patient subgroups. ResultsOne thousand five hundred fifty-eight patients underwent surgery for NSCLC, of whom 679 (44%) had preoperative PET imaging. After excluding those whose primary tumors extended to the hilum (n = 44) and those with no hilar nodal sampling (n = 130), 505 patients were included in the analysis, 92 (18%) of whom had positive ipsilateral hilar lymph nodes. The median number of hilar lymph nodes sampled was four (range, 1-31). The ipsilateral hilum was positive on preoperative PET in 108 patients. Among the 92 patients with involved N1 lymph nodes, 40 were PET positive resulting in a sensitivity of 43%. Among 413 pN0 patients, 345 were PET negative resulting in a specificity of 84%. The positive predictive and negative predictive values were 37% and 87%, respectively. The sensitivity was not different when comparing patients with one involved N1 lymph node (22/51, 43%) vs. more than one involved N1 lymph node (18/41, 44%, p = 1). Sensitivity (51% vs. 37%, p = 0.21) and specificity (82% vs. 85%, p = 0.50) were not different between right-sided and left-sided lung tumors, respectively. One thousand five hundred fifty-eight patients underwent surgery for NSCLC, of whom 679 (44%) had preoperative PET imaging. After excluding those whose primary tumors extended to the hilum (n = 44) and those with no hilar nodal sampling (n = 130), 505 patients were included in the analysis, 92 (18%) of whom had positive ipsilateral hilar lymph nodes. The median number of hilar lymph nodes sampled was four (range, 1-31). The ipsilateral hilum was positive on preoperative PET in 108 patients. Among the 92 patients with involved N1 lymph nodes, 40 were PET positive resulting in a sensitivity of 43%. Among 413 pN0 patients, 345 were PET negative resulting in a specificity of 84%. The positive predictive and negative predictive values were 37% and 87%, respectively. The sensitivity was not different when comparing patients with one involved N1 lymph node (22/51, 43%) vs. more than one involved N1 lymph node (18/41, 44%, p = 1). Sensitivity (51% vs. 37%, p = 0.21) and specificity (82% vs. 85%, p = 0.50) were not different between right-sided and left-sided lung tumors, respectively. ConclusionsThe sensitivity, specificity, positive predictive value, and negative predictive value of PET for staging the ipsilateral hilum was 43%, 84%, 37%, and 87%, respectively. These numbers are generally inferior to those reported for mediastinal PET staging. This analysis suggests that PET may be insufficient to adequately evaluate the hilum, which has potential implications for elective nodal irradiation and stereotactic body radiation therapy. The sensitivity, specificity, positive predictive value, and negative predictive value of PET for staging the ipsilateral hilum was 43%, 84%, 37%, and 87%, respectively. These numbers are generally inferior to those reported for mediastinal PET staging. This analysis suggests that PET may be insufficient to adequately evaluate the hilum, which has potential implications for elective nodal irradiation and stereotactic body radiation therapy.

7072 Background: The International Association for the Study of Lung Cancer (IASLC) recently recommended changes to the TNM classification of lung cancers. For early-stage disease, increasing tumor size and nodal status are of primary prognostic relevance for overall survival. However, the ability of the old and the new staging system to predict the risk of local/regional recurrence (LRR) has not been evaluated. Therefore, we examined whether the proposed changes in AJCC 7 would refine the risk of LRR in patients who undergo surgery for early-stage NSCLC. Methods: The medical records and pertinent radiographs for all patients who underwent surgery for early-stage (N0-N1) NSCLC at Duke between 1995 and 2005 were reviewed. Patients undergoing suboptimal surgery (sublobar resections, positive margins), or those who received any preoperative therapy or postoperative radiation therapy were excluded. Disease recurrence at the surgical margin, ipsilateral hilum, and/or mediastinum was considered a LRR. Stage was assigned based on both AJCC 6 and AJCC 7. Actuarial rates of LRR were estimated using the Kaplan-Meier method. Results: Of 975 patients who underwent surgery during the time interval, 741 were eligible for this analysis. Surgery consisted of lobectomy in 91% (n=672) and pneumonectomy in 9% (n=69) of patients. 82% (n=605) of patients were pathologic N0 (pN0) and 18% (n=136) were pN1. Adjuvant chemotherapy was administered to 7% (n=52). Median follow-up was 35 months. Conversion from AJCC 6 to AJCC 7 resulted in 25% stage migration (upstaging in 15%; downstaging in 10%). For all patients, the 5-year actuarial risk of LRR was 20% (95% CI 17%-24%). 5-year rates of LRR for stage IA, IB, IIA, and IIB disease using AJCC 6 were 12%, 26%, 40%, and 27%, respectively. Using AJCC 7, corresponding rates were 12%, 21%, 34%, and 38%, respectively. Conclusions: The risk of LRR increases monotonically with stage in the new AJCC 7 system, but not for the older AJCC 6 system. Thus, the newer staging system is better in predicting the risk of LRR. This information will be valuable when designing future studies of postoperative RT. No significant financial relationships to disclose.

Radiation therapy (RT) is established as the primary treatment of squamous cell carcinoma of the anus. Multiple randomized trials have shown that combined modality therapy with RT, 5-fluorouracil, and mitomycin-C results in high rates of local control, disease-free survival, and sphincter preservation. However, treatment-related toxicity using conventional radiation approaches remains high and may compromise therapeutic efficacy because of prolonged treatment breaks and inability to deliver adequate radiation dose. Recent developments, including the use of PET for staging, radiation planning, and response assessment, and advanced RT planning using intensity-modulated radiation therapy (IMRT), may decrease acute and late treatment-related toxicity, provide high-dose target conformality, and permit safe radiation dose escalation. This article reviews the basic principles of IMRT and highlights current literature on these recent advances and the application of new RT techniques.

Intensity-modulated radiation therapy (IMRT) has the potential to reduce toxicities associated with chemoradiotherapy in the treatment of anal cancer. This study reports the results of using IMRT in the treatment of anal cancer.Records of patients with anal malignancies treated with IMRT at Duke University were reviewed. Acute toxicity was graded using the NCI CTCAEv3.0 scale. Overall survival (OS), metastasis-free survival (MFS), local-regional control (LRC) and colostomy-free survival (CFS) were calculated using the Kaplan-Meier method.Forty-seven patients with anal malignancy (89% canal, 11% perianal skin) were treated with IMRT between August 2006 and September 2008. Median follow-up was 14 months (19 months for SCC patients). Median radiation dose was 54 Gy. Eight patients (18%) required treatment breaks lasting a median of 5 days (range, 2-7 days). Toxicity rates were as follows: Grade 4: leukopenia (7%), thrombocytopenia (2%); Grade 3: leukopenia (18%), diarrhea (9%), and anemia (4%); Grade 2: skin (93%), diarrhea (24%), and leukopenia (24%). The 2-year actuarial overall OS, MFS, LRC, and CFS rates were 85%, 78%, 90% and 82%, respectively. For SCC patients, the 2-year OS, MFS, LRC, and CFS rates were 100%, 100%, 95%, and 91%, respectively.IMRT-based chemoradiotherapy for anal cancer results in significant reductions in normal tissue dose and acute toxicities versus historic controls treated without IMRT, leading to reduced rates of toxicity-related treatment interruption. Early disease-related outcomes seem encouraging. IMRT is emerging as a standard therapy for anal cancer.

Randomized trials have demonstrated a survival benefit with either postoperative chemoradiotherapy (CRT) or perioperative chemotherapy (CT) for resectable gastric adenocarcinoma, albeit with high rates of treatment-related toxicity. Neoadjuvant CRT has potential advantages compared to adjuvant approaches and improves outcomes in esophageal and rectal malignancies. We examined acute toxicity and disease-related outcomes for patients treated with neoadjuvant CRT for gastric cancer at our institution. All patients with gastroesophageal (GE) junction (Siewert type II and III) or stomach adenocarcinoma who underwent curative intent CRT followed by planned surgical resection at Duke University between 1987 and 2009 were reviewed. Tumors were clinically staged according to AJCC 6th edition. Local recurrence was defined as radiographic or biopsy-proven disease within the radiation treatment field. Actuarial rates of overall survival (OS), local control (LC) and disease-free survival (DFS) were estimated using the Kaplan-Meier method. Toxicity was graded according to CTCAE v4.0. Forty-eight patients (2% stage IB, 19% stage II, 60% stage III, 8% stage IV, 10% incompletely staged) were included. Most (73%) had proximal (GE junction, cardia and fundus) tumors. Thirty-eight percent had signet ring histology, 52% had poorly differentiated tumors, and 10% had linitis plastica. Median age was 60 years (range, 28-79) and median RT dose was 45 Gy. All patients received concurrent CT with forty (83%) receiving fluoropyrimidine-based CT. Six patients (13%) required treatment break and two (4%) were unable to complete the prescribed treatment course. Rates of acute > grade 2 hematologic and non-hematologic toxicity were 38% and 10%, respectively. Thirty-six patients (75%) underwent surgery. Patients did not receive surgery due to distant metastases noted at laparotomy or restaging (n = 9), patient refusal (n = 2) or poor performance status (n = 1). Pathologic complete response and R0 resection rates were 19% and 86%, respectively. Thirty-day surgical mortality was 6%. At a median follow-up of 42 months, 3-year actuarial OS for the entire cohort was 40%. For patients undergoing surgery, 3-year OS, LC, and DFS were 50%, 73% and 41%, respectively. Neoadjuvant CRT for gastric adenocarcinoma is reasonably well tolerated with acceptable rates of perioperative morbidity and mortality. In this group of patients with advanced disease, LC, DFS and OS rates in resected patients are comparable to similarly staged, adjuvantly treated historic controls. Further study comparing neoadjuvant CRT to standard treatment approaches for gastric cancer is indicated.