- Email: [email protected]
Toxicity and Outcomes in Patients With and Without Esophageal Stents in Locally Advanced Esophagus Cancer
Poster Viewing E175
Volume 96 Number 2S Supplement 2016 damage, rather than to traditional, acute mechanisms of RT-induced hematologic toxicity.
Abstract 2426; Table 1. Effects of mean liver dose (MLD) on change in platelet count Time Point MLD >[19 GyE (vs. RT Week 4 <19 GyE) 1 Month Post-RT 3 Months Post-RT
Odds Ratio for >[25 K/uL Drop in Platelets (95% CI)
P
1.0 (0.4, 3.0) 1.8 (0.6, 5.1)
0.977 0.270
4.0 (1.4, 11.4)
0.010
Conclusion: Our predictive model could distinguish patients at high risk from patients at low risk for LRR. Similar to thoracic ESCC patients at high risk, LRR primarily involved the upper thorax (supraclavicular/cervical site, mediastinum above the carina, and anastomoses), and this area must be considered in future study designs for radical tri-modality treatment. Author Disclosure: S. Liu: None. S. Anfossi: None. Y. Zheng: None. M. Cai: None. J. Fu: None. B. Qiu: None. H. Yang: None. Q. Liu: None. J. Fu: None. M. Liu: None. J.K. Burks: None. S.H. Lin: None. J. Reuben: None. H. Liu: None.
2428 Author Disclosure: M.A. Dyer: None. E.I. McDonnell: None. B.Y. Yeap: Employee; Massachusetts General Hospital. Consultant; Harvard Clinical Research Institute. Scientific Advisory Board; SISCAPA Assay Technologies. Advisory board; LUNGevity Foundation. Task force; Society for Immunotherapy of Cancer. N. Depauw: None. J.A. Wolfgang: None. J.Y. Wo: None. T.F. DeLaney: Honoraria; UpToDate, Oakstone Medical Publishing. Consultant; Amgen, Gerson Lehman Consulting. E. BenJosef: None. C.H. Crane: Honoraria; EMD Serono, Inc. Consultant; Vertex Pharmaceuticals. A.X. Zhu: None. T.I. Hong: None.
2427 Clinical and Biological Prognostic Factors for Locoregional Recurrence in Patients With Thoracic Esophageal Squamous Cell Carcinoma Treated With Radical 2-field Lymph Node Dissection: Results From Long-term Follow-up S. Liu,1 S. Anfossi,2 Y. Zheng,1 M. Cai,1 J. Fu,1 B. Qiu,1 H. Yang,1 Q. Liu,1 J. Fu,1 M. Liu,1 J.K. Burks,3 S.H. Lin,4 J. Reuben,3 and H. Liu1; 1State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou, China, 2The University of Texas MD Anderson Cancer Center, Houston, TX, 3The University of Texas MD Anderson Cancer Center, Houston, China, 4Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX Purpose/Objective(s): To determine the clinical and biological prognostic factors for locoregional recurrence (LRR) in patients with thoracic esophageal squamous cell carcinoma (ESCC) undergoing radical two-field lymph node resection (2FLD). Materials/Methods: A total of 462 patients diagnosed with thoracic ESCC underwent radical esophagectomy between March 2001 and May 2010 at our Center. Clinical characteristics, CD44 expression, and tumor infiltrating lymphocyte (TIL) levels were evaluated in 198 patients who underwent R0 dissection with long-term follow-up. Partial Cox regression analysis with leave-one-out cross-validation was performed to validate the selected risk factors. Results: With a median follow-up of 54 months (range, 6-152 months), the 5-year local failure-free survival (LFFS) rate of 198 patients was 62.5%. Multivariate analysis revealed that T stage (P Z 0.043), pathological positive tumor above the carina (P Z 0.000), CD44 expression level (P Z 0.045) and TIL level (P Z 0.007) were prognostic factors for LFFS. The Cox model with risk scores estimated from expression data had an area under the curve (AUC) value of 83.6% with SE Z 0.029 (95% CI: 0.78 to 0.89) for the prediction of 5-year LFFS. The best cut-off value for sensitivity and specificity was 11.19 (sensitivity: 83%; specificity: 71%). This cut-off value was used to divide the 198 patients into a high-risk group (sum score 11.19) and a low-risk group (sum score < 11.19) for LRR. Patients at high risk had significantly shorter 5-year LFFS than patients at low risk (25 months vs median not reached; P Z 0.000). The LRR pattern revealed significantly high incidences of recurrent disease at the supraclavicular and cervical sites, mediastinum (above the carina), and anastomosis.
Toxicity and Outcomes in Patients With and Without Esophageal Stents in Locally Advanced Esophagus Cancer S. Francis,1 S. Lloyd,2 A. Orton,2 and C.J. Anker3; 1University of Utah Huntsman Cancer Institute, Salt Lake City, UT, 2Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, 3University of Vermont Cancer Center, Burlington, VT Purpose/Objective(s): Stent placement is often considered to relieve severe dysphagia in patients with locoregionally advanced esophageal cancer. We sought to determine effects of esophageal stenting on acute toxicities and oncologic outcomes in patients treated with radiation therapy (RT). Materials/Methods: Patients treated with RT with curative intent for locoregionally advanced esophageal cancer at our institution were retrospectively analyzed. Chi-squared analysis was used to compare demographic and tumor characteristics between patients with or without esophageal stenting prior to RT. Univariate (UV) and multivariate (MV) logistic regression modeling was used to identify predictors for acute toxicities. UV and MV Cox proportional hazards modeling was used to identify factors associated with survival. In the respective MV regression models, variables that had an association with either toxicity or death, defined as P<0.2 or that changed the odds ratio (OR) or hazard ratio (HR) 10% when removed from analysis were included. Acute toxicities were graded using Common Terminology Criteria for Adverse Events version 4. Results: One hundred twenty-four consecutive patients received RT between 2005 and 2013. Thirty-five had a stent during RT. Median dose was 5040 cGy for all patients. Ninety-seven percent received chemotherapy. Thirty-one percent of stented patients and 47% of no-stent patients underwent esophagectomy. Of those stented, dysphagia improved in 85% as measured 1 week post-stent. Ninety-seven percent of stent patients had T3 tumors compared to 72% of no-stent patients (P Z 0.021). The mean tumor volume was 109 cm3 versus 63 cm3in the stent and no-stent groups, respectively (P < 0.001). Grade 3 acute toxicities were seen in 69% of stent compared to 24% of no-stent patients (P < 0.001), including esophagitis (37% vs 19%, P Z 0.035), dehydration (26% vs 12%, P Z 0.069), and anorexia/weight loss (20% vs 6%, P Z 0.015), respectively. When comparing age, tobacco use, stage, performance status, histology, tumor volume, chemotherapy, or stent, the only significant predictor for acute toxicity on UV (OR 7.06, P<0.001) and MV (OR 8.70, P<0.001) logistic regression was esophageal stenting. Stenting remained the sole predictor of toxicity even after excluding from analysis patients with >grade 1 dysphagia prior to starting CRT (MV OR 20.4, P<0.001). On UV Cox regression, stenting was associated with increased risk of death (HR 1.88, P Z 0.005) but not on MV analysis (HR 1.45, P Z 0.13). T3 tumors (HR 3.40 compared to T1, P Z 0.02) and squamous histology (HR 2.22, P Z 0.035) had an increased death risk on MV analysis, while esophagectomy was associated with a decreased death risk (HR 0.39, P<0.001). Conclusion: Esophageal stenting was associated with significantly increased grade 3 acute toxicities, including esophagitis, anorexia, and weight loss, in patients receiving RT therapy with curative intent. Author Disclosure: S. Francis: None. S. Lloyd: None. A. Orton: None. C.J. Anker: None.
Volume 96 Number 2S Supplement 2016 damage, rather than to traditional, acute mechanisms of RT-induced hematologic toxicity.
Abstract 2426; Table 1. Effects of mean liver dose (MLD) on change in platelet count Time Point MLD >[19 GyE (vs. RT Week 4 <19 GyE) 1 Month Post-RT 3 Months Post-RT
Odds Ratio for >[25 K/uL Drop in Platelets (95% CI)
P
1.0 (0.4, 3.0) 1.8 (0.6, 5.1)
0.977 0.270
4.0 (1.4, 11.4)
0.010
Conclusion: Our predictive model could distinguish patients at high risk from patients at low risk for LRR. Similar to thoracic ESCC patients at high risk, LRR primarily involved the upper thorax (supraclavicular/cervical site, mediastinum above the carina, and anastomoses), and this area must be considered in future study designs for radical tri-modality treatment. Author Disclosure: S. Liu: None. S. Anfossi: None. Y. Zheng: None. M. Cai: None. J. Fu: None. B. Qiu: None. H. Yang: None. Q. Liu: None. J. Fu: None. M. Liu: None. J.K. Burks: None. S.H. Lin: None. J. Reuben: None. H. Liu: None.
2428 Author Disclosure: M.A. Dyer: None. E.I. McDonnell: None. B.Y. Yeap: Employee; Massachusetts General Hospital. Consultant; Harvard Clinical Research Institute. Scientific Advisory Board; SISCAPA Assay Technologies. Advisory board; LUNGevity Foundation. Task force; Society for Immunotherapy of Cancer. N. Depauw: None. J.A. Wolfgang: None. J.Y. Wo: None. T.F. DeLaney: Honoraria; UpToDate, Oakstone Medical Publishing. Consultant; Amgen, Gerson Lehman Consulting. E. BenJosef: None. C.H. Crane: Honoraria; EMD Serono, Inc. Consultant; Vertex Pharmaceuticals. A.X. Zhu: None. T.I. Hong: None.
2427 Clinical and Biological Prognostic Factors for Locoregional Recurrence in Patients With Thoracic Esophageal Squamous Cell Carcinoma Treated With Radical 2-field Lymph Node Dissection: Results From Long-term Follow-up S. Liu,1 S. Anfossi,2 Y. Zheng,1 M. Cai,1 J. Fu,1 B. Qiu,1 H. Yang,1 Q. Liu,1 J. Fu,1 M. Liu,1 J.K. Burks,3 S.H. Lin,4 J. Reuben,3 and H. Liu1; 1State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou, China, 2The University of Texas MD Anderson Cancer Center, Houston, TX, 3The University of Texas MD Anderson Cancer Center, Houston, China, 4Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX Purpose/Objective(s): To determine the clinical and biological prognostic factors for locoregional recurrence (LRR) in patients with thoracic esophageal squamous cell carcinoma (ESCC) undergoing radical two-field lymph node resection (2FLD). Materials/Methods: A total of 462 patients diagnosed with thoracic ESCC underwent radical esophagectomy between March 2001 and May 2010 at our Center. Clinical characteristics, CD44 expression, and tumor infiltrating lymphocyte (TIL) levels were evaluated in 198 patients who underwent R0 dissection with long-term follow-up. Partial Cox regression analysis with leave-one-out cross-validation was performed to validate the selected risk factors. Results: With a median follow-up of 54 months (range, 6-152 months), the 5-year local failure-free survival (LFFS) rate of 198 patients was 62.5%. Multivariate analysis revealed that T stage (P Z 0.043), pathological positive tumor above the carina (P Z 0.000), CD44 expression level (P Z 0.045) and TIL level (P Z 0.007) were prognostic factors for LFFS. The Cox model with risk scores estimated from expression data had an area under the curve (AUC) value of 83.6% with SE Z 0.029 (95% CI: 0.78 to 0.89) for the prediction of 5-year LFFS. The best cut-off value for sensitivity and specificity was 11.19 (sensitivity: 83%; specificity: 71%). This cut-off value was used to divide the 198 patients into a high-risk group (sum score 11.19) and a low-risk group (sum score < 11.19) for LRR. Patients at high risk had significantly shorter 5-year LFFS than patients at low risk (25 months vs median not reached; P Z 0.000). The LRR pattern revealed significantly high incidences of recurrent disease at the supraclavicular and cervical sites, mediastinum (above the carina), and anastomosis.
Toxicity and Outcomes in Patients With and Without Esophageal Stents in Locally Advanced Esophagus Cancer S. Francis,1 S. Lloyd,2 A. Orton,2 and C.J. Anker3; 1University of Utah Huntsman Cancer Institute, Salt Lake City, UT, 2Huntsman Cancer Institute, University of Utah, Salt Lake City, UT, 3University of Vermont Cancer Center, Burlington, VT Purpose/Objective(s): Stent placement is often considered to relieve severe dysphagia in patients with locoregionally advanced esophageal cancer. We sought to determine effects of esophageal stenting on acute toxicities and oncologic outcomes in patients treated with radiation therapy (RT). Materials/Methods: Patients treated with RT with curative intent for locoregionally advanced esophageal cancer at our institution were retrospectively analyzed. Chi-squared analysis was used to compare demographic and tumor characteristics between patients with or without esophageal stenting prior to RT. Univariate (UV) and multivariate (MV) logistic regression modeling was used to identify predictors for acute toxicities. UV and MV Cox proportional hazards modeling was used to identify factors associated with survival. In the respective MV regression models, variables that had an association with either toxicity or death, defined as P<0.2 or that changed the odds ratio (OR) or hazard ratio (HR) 10% when removed from analysis were included. Acute toxicities were graded using Common Terminology Criteria for Adverse Events version 4. Results: One hundred twenty-four consecutive patients received RT between 2005 and 2013. Thirty-five had a stent during RT. Median dose was 5040 cGy for all patients. Ninety-seven percent received chemotherapy. Thirty-one percent of stented patients and 47% of no-stent patients underwent esophagectomy. Of those stented, dysphagia improved in 85% as measured 1 week post-stent. Ninety-seven percent of stent patients had T3 tumors compared to 72% of no-stent patients (P Z 0.021). The mean tumor volume was 109 cm3 versus 63 cm3in the stent and no-stent groups, respectively (P < 0.001). Grade 3 acute toxicities were seen in 69% of stent compared to 24% of no-stent patients (P < 0.001), including esophagitis (37% vs 19%, P Z 0.035), dehydration (26% vs 12%, P Z 0.069), and anorexia/weight loss (20% vs 6%, P Z 0.015), respectively. When comparing age, tobacco use, stage, performance status, histology, tumor volume, chemotherapy, or stent, the only significant predictor for acute toxicity on UV (OR 7.06, P<0.001) and MV (OR 8.70, P<0.001) logistic regression was esophageal stenting. Stenting remained the sole predictor of toxicity even after excluding from analysis patients with >grade 1 dysphagia prior to starting CRT (MV OR 20.4, P<0.001). On UV Cox regression, stenting was associated with increased risk of death (HR 1.88, P Z 0.005) but not on MV analysis (HR 1.45, P Z 0.13). T3 tumors (HR 3.40 compared to T1, P Z 0.02) and squamous histology (HR 2.22, P Z 0.035) had an increased death risk on MV analysis, while esophagectomy was associated with a decreased death risk (HR 0.39, P<0.001). Conclusion: Esophageal stenting was associated with significantly increased grade 3 acute toxicities, including esophagitis, anorexia, and weight loss, in patients receiving RT therapy with curative intent. Author Disclosure: S. Francis: None. S. Lloyd: None. A. Orton: None. C.J. Anker: None.