- Email: [email protected]
Modern Non-operative Treatment Strategies for Patients With Stage II-III Non–small Cell Lung Cancer in the United States
E468
International Journal of Radiation Oncology Biology Physics
prevent premature switching to other, less-effective systemic therapies by maximizing the duration and efficacy of EGFR-TKI treatment in selected patients. Author Disclosure: A.H. Kesarwala: Research Grant; ASTRO. C. Kim: None. J.C. Jones: None. A. Kaushal: None. N. Roper: None. C.D. Hoang: None. E. Szabo: None. M. Connolly: None. E. Padiernos: None. C. Cultraro: None. M. Waris: None. S. Gao: None. S.M. Steinberg: None. J. Khan: None. A. Rajan: None. U. Guha: Research Grant; AstraZeneca.
3109
3108 Metachronous Versus Synchronous Versus Single Tumor Nonesmall Cell Lung Cancers: Is There a Difference in Outcomes? B. Khodayari, R. Rahimian, J. Rahimian, R. Wang, and M.R. Girvigian; Kaiser Permanente Los Angeles Medical Center, Los Angeles, CA Purpose/Objective(s): Patients with early stage (ES) NSCLC are at risk for synchronous or metachronous second primary lung cancers (SPLC). The aim of this study is to investigate the outcomes of patients treated with stereotactic body radiation therapy (SBRT) for synchronous or metachronous tumors versus patients with traditional single tumor ESNSCLC. Materials/Methods: We retrospectively identified 107 cases of single tumor ES-NSCLC treated at our institution from 2008 to 2015. An additional 13 patients with 29 unique tumors were identified to have SPLC; 31% of which had synchronous tumors and 69% had metachronous tumors. Of the SPLCs, 76% were biopsy proven and no patient had two different tumor histologies. SPLC was defined as an isolated parenchymal lung tumor separate from the first primary lung cancer (FPLC) and without associated radiographic evidence of regional or distant metastatic disease on PET scan. Kaplan Meier method with log-rank test was used for overall survival (OS), lung cancer specific survival (LCSS), progression-free survival (PFS, progression defined as development of any recurrence or SPLC), local control (LC), and freedom from distant failure (FFDF) analysis. Results: There was an approximate 1% per year incidence of SPLC, similar to that of historical controls. The median age was 79 [61-89] years old. The median follow-up time for single tumor patients was 26 [6-58] months and for synchronous SPLC patients was 25 [14-40] months. For metachronous tumor patients, median time from FPLC to SPLC was 53 [12-262] months, from SPLC to last follow-up was 14 [12-40] months, and overall median time from FPLC to last follow-up was 71 [17-276] months. In this series, all metachronous FPLCs were managed with surgery. All metachronous SPLCs, synchronous SPLCs, and single tumors were non-operable and treated with SBRT (median dose 50 Gy in 5 fractions). Patients with metachronous tumors demonstrated an OS advantage as compared to synchronous and single tumors (2-year OS 100% vs 80% vs 60%, p Z 0.04). However, this was attributable to metachronous patients having a longer follow-up time, leading to a survival bias with fewer deaths from intercurrent disease; this was demonstrated by the similar outcomes for LCSS (100% vs 80% vs 92%, p Z 0.23). There was no difference in PFS when comparing all tumor groups. The distant pattern of failure between the three groups were similar (Metachronous vs Synchronous vs Single tumor 2-year FFDF: 36% vs 80% vs 68%, p Z 0.60). There was also very good local control of disease after SBRT (2-year LC 100% vs 100% vs 92%, p Z 0.74). Conclusion: Patients with synchronous or metachronous tumors have similar outcomes to traditional single tumor ES-NSCLC patients. These patients demonstrated excellent LC and LCSS after SBRT. Although further investigation in larger cohorts is warranted, this study suggests that carefully selected patients with no clinical or radiographic evidence for advanced stage disease could be considered for SBRT to all solitary primary lung tumors. Author Disclosure: B. Khodayari: None. R. Rahimian: None. J. Rahimian: None. R. Wang: None. M.R. Girvigian: None.
Use of Hypofractionated Radiation Therapy With Concurrent Chemotherapy in Inoperable Stage II/ III Nonesmall Cell Lung Cancer E. Kim,1 K.D. Westover,2 M.E. Daly,3 J.D. Murphy,4 and T. Mitin5; 1 Department of Radiation Oncology, Vanderbilt University Medical Center, Nashville, TN, 2University of Texas Southwestern Medical Center, Dallas, TX, 3University of California Davis Comprehensive Cancer Center, Sacramento, CA, 4University of California, San Diego, La Jolla, CA, 5 Oregon Health and Science University, Portland, OR Purpose/Objective(s): Precision hypofractionated radiation therapy (PHRT) without concurrent chemotherapy in inoperable Stage II/III nonsmall cell lung cancer (NSCLC) appears to be safe and effective in early results from randomized trials. PHRT can yield a higher biological effective dose and shorten the total treatment time, leading to more efficient treatment. However, the safety and efficacy of PHRT with concurrent chemotherapy is not well studied. Materials/Methods: We searched the National Cancer Database (NCDB) for Tany N+ M0 NSCLC diagnosed 2004-2013 that received chemotherapy and radiotherapy (RT) in their primary course of treatment. Conventional fractionated chemoradiation (CF-CRT) was defined as 58-80 Gy in 30-45 fractions. Hypofractionated chemoradiation (HF-CRT) was defined as 45-60 Gy in 15 fractions. Patient demographic, diagnosis, and treatment information and overall survival (OS) were compared with descriptive statistics, multivariable logistic regression, and single and multivariable survival analyses. Propensity score matching was used to try to reduce selection bias and confounding variables between the two groups. Results: A total of 126 (0.5%) patients received HF-CRT and 24,286 (99.5%) patients received CF-CRT. Even before 1:1 propensity score matching, the two groups were similar in demographics, Charlson-Deyo comorbidities score, and grade. Before matching, the HF-CRT group was more likely to be treated at academic research facility than CF-CRT group (52% vs 30%, p<0.0001), but after matching, 52% of both groups were treated at an academic/research facility. On univariable and multivariable survival analysis, CF-CRT was associated with better overall survival (OS). On Kaplan-Meier analysis, 1-year and median OS were 68% and 15 months for CF-CRT vs 46% and 10 months for HF-CRT, respectively, with hazard ratio (HR) of 1.9 (95% confidence interval [CI] 1.3-2.7) on MVA. Female gender, later year of diagnosis, living in East/Atlantic region, and unknown grade were also associated with higher OS. Conclusion: Use of HF-CRT has been uncommon in the US in the past decade. Interpretation of outcomes is limited due to the retrospective nature of this study, the unavailability of dosimetric data and the small cohort, despite the use of this national hospital-based database capturing >70% of cancers diagnosed in the US. Nonetheless, this provides valuable information on real-world outcomes of new treatment regimens currently under consideration for trials. While a recently presented randomized phase 3 trial from UTSW showed comparative outcomes between PHRT and CRT, our retrospective study raises hypotheses concerning the safety of combining hypofractionated radiation with concurrent chemotherapy. Carefully designed prospective clinical trials with strict dosimetric guidelines will be required to understand the feasibility of combining PHRT with systemic agents. Author Disclosure: E. Kim: None. K.D. Westover: None. M.E. Daly: Research Grant; American Cancer Society, Department of Defense, NIH/ NCI. J.D. Murphy: None. T. Mitin: Honoraria; Novocure, UpToDate Inc.. Consultant; Novocure. Advisory Board; Novocure. Travel Expenses; Novocure.
3110 Modern Non-operative Treatment Strategies for Patients With Stage II-III Nonesmall Cell Lung Cancer in the United States E. Kim,1 M.E. Daly,2 K.D. Westover,3 J.D. Murphy,4 and T. Mitin5; 1 Department of Radiation Oncology, Vanderbilt University Medical
Volume 99 Number 2S Supplement 2017 Center, Nashville, TN, 2University of California Davis Comprehensive Cancer Center, Sacramento, CA, 3University of Texas Southwestern Medical Center, Dallas, TX, 4University of California, San Diego, La Jolla, CA, 5Oregon Health and Science University, Portland, OR Purpose/Objective(s): Standard of care for inoperable N+ M0 non-small cell lung cancer (NSCLC) is concurrent platinum-based chemoradiation (CRT) and thoracic radiation to 60 Gy in daily fractions. However, many patients with a poor performance status or social limitations may receive radiation monotherapy (RT). The primary aim of this study was to compare characteristics and survival of patients who received standard CRT vs RT alone. Materials/Methods: The national cancer database (NCDB) captures >70% of incident cancer diagnoses in the United States. We selected patients diagnosed with clinical or pathological stage Tany N+ M0 during 2004-2013 who did not undergo definitive surgery. Definitive RT was defined as 58-80 Gy in 30-45 fractions or 45-60 Gy in 15 fractions, starting within 90 days of diagnosis. Definitive CRT was defined as RT with systemic chemotherapy administered within 180 days of diagnosis, without distinction between concurrent and sequential administration. Patient demographic, diagnosis, and treatment information and overall survival (OS) were compared with descriptive statistics, multivariable logistic regression, and single and multivariable survival analyses. Results: After applying selection criteria, 89,195 patients had clinical or pathological stage Tany N+ M0, and of these 3,073 (3.4%) received RT, 24,412 (27%) received CRT, 49,394 (55%) received chemo, and 12,316 (14%) did not receive any treatment. Eleven percent of patients (3,073 out of 27,485) receiving RT did not receive chemo. We used 1:1 propensity score matching to compare RT and CRT groups, focusing on 2,883 patients in each group. RT patients had older age (median 75 vs 70, range 38-90 in both). Baseline and cancer characteristics were comparable except race (RT 85% vs CRT 82% white, pZ0.04), and insurance (RT 70% vs CT 67% Medicare, pZ0.02). In multivariable analysis the groups were comparable except for borderline significance of Medicaid insurance (odds ratio [OR] 1.3, 95% confidence interval [CI] 1.0-1.7) and Charlson-Deyo comorbidity score of 1 (OR 1.1, 95% CI 1.0-1.3) favoring CRT. Kaplan-Meier survival analysis found improved OS associated with CRT (5-year 15%, median 17 months) vs RT (5-year 8%, median 12 months) (p<0.0001), with a hazard ratio (HR) of 0.7 (95%CI 0.7-0.7). Improved survival was also associated with female gender, younger age, insurance other than Medicare/unknown/ none, fewer comorbidities, treatment at an academic/research facility, and earlier stage at diagnosis. Conclusion: Of patients referred for definitive chemoradiation, 11% received RT alone. Definitive CRT was associated with higher OS when compared to RT alone but selection bias may have contributed since patients who received RT alone had more comorbidities and were older than patients who received definitive CRT. Novel approaches are needed to improve the outcomes in patients who are not candidate for systemic chemotherapy with concurrent thoracic RT, such as combination of RT with targeted and immunotherapy agents. Author Disclosure: E. Kim: None. M.E. Daly: Research Grant; American Cancer Society, Department of Defense, NIH/NCI. K.D. Westover: None. J.D. Murphy: None. T. Mitin: Honoraria; Novocure, UpToDate Inc. Consultant; Novocure. Advisory Board; Novocure. Travel Expenses; Novocure.
3111 Improved Survival With Postoperative Radiation Therapy in Thymic Carcinoma: A PropensityMatched Analysis of the Surveillance, Epidemiology, and End Results Database Y.J. Lim,1 C. Song,2 and J.S. Kim2; 1Department of Radiation Oncology, Kyung Hee University Medical Center, Seoul, Korea, Republic of (South), 2 Department of Radiation Oncology, Seoul National University Bundang Hospital, Seongnamsi, Korea, Republic of (South) Purpose/Objective(s): Thymic carcinoma is a rare and aggressive malignancy with poor prognosis. Although postoperative radiotherapy
Poster Viewing E469 (PORT) is used for obtaining better locoregional tumor control, its association with survival has not been established. This study evaluated the prognostic impact of PORT in thymic carcinoma. Materials/Methods: We identified patients diagnosed with thymic carcinoma between 2004 and 2013 using the Surveillance, Epidemiology, and End Results (SEER) database. Propensity score (PS) matching with Kaplan-Meier and Cox-regression analyses were used to assess prognosis. Results: In the unmatched population (nZ312), 184 (59%) patients underwent PORT. Overall 5-year survival rates were better with receipt of PORT, both before and after matching (PZ0.012 and 0.007, respectively). After adjusting for related covariates (nZ256), age 63 years (PZ0.023), Masaoka stage III (PZ0.028) and IV (P<0.001), debulking surgery (PZ0.021), and no receipt of PORT (PZ0.013) were independently poor prognostic factors. In subgroup analyses, favorable survival impacts of PORT were observed for Masaoka stage III tumors (hazard ratio [HR] 0.31, 95% confidence interval [CI] 0.15e0.66), tumors sized >6.0 cm (HR 0.48, 95% CI 0.26e0.89), node-negative status (HR 0.58, 95% CI 0.33e1.00), and surgical extent of local excision or partial removal (HR 0.44, 95% CI 0.22e0.86). Conclusion: On SEER analysis, survival benefits of PORT in thymic carcinoma were demonstrated. With strong prognostic associations of Masaoka stage and types of surgery, PORT should be considered for nonmetastatic locally advanced tumors with limited surgical resection. Author Disclosure: Y. Lim: None. C. Song: None. J. Kim: None.
3112 Interfraction Tumor Size Changes During Lung SBRT: What Factors Matter? K. King,1 S. Mayekar,1 G. Marwaha,2 G. Tolekidis,1 G. Cifter,2 and H. Zhen2; 1Rush University Medical Center, Chicago, IL, 2Rush Medical Center, Chicago, IL Purpose/Objective(s): Prior studies have suggested that lung stereotactic body radiation therapy (SBRT) can cause volumetric tumor size changes on setup imaging during treatment, however predictors for such changes have not been well described. With the advent of kilovoltage (kv) cone beam computed tomography (CBCT) scans in the SBRT setup process, visualization of tumor changes are more readily apparent. The purpose of this study was to evaluate the role played by patient, tumor, and dosimetric factors in the interfraction tumor changes seen in our lung SBRT patient population. Materials/Methods: A retrospective chart review was conducted on forty consecutive patients from our IRB-approved lung SBRT database treated between 2014-2016. Each patient was set up under daily stereoscopic x-ray and kv CBCT guidance. A single radiation oncologist retrospectively recontoured the tumor volume on each sequential kv CBCT image to determine volumetric variances. A univariate and multivariate analysis with a linear model were employed in SPSS R 3.3.1. Results: Forty patients with either squamous cell carcinoma (12.5%), adenocarcinoma (52.5%), or metastatic renal cell carcinoma (5%) of the lung were identified. 20 (50%) patients received an SBRT dose of 54 Gy / 3 fxs and 20 (50%) patients received 50-60 Gy / 5 fxs. 8 (20%) of the tumors were centrally located (by the RTOG 0813 definition) while 32 (80%) were peripheral. Six (15%) of the patients had never smoked while 28 (78%) were former smokers and 6 (15%) continued to smoke during their treatment. The median pretreatment neutrophil lymphocyte ratio (NLR) was 3.03 (range: 0.09-67.21) and the median post-treatment NLR was 2.74 (range: 0.17-12.59). Within the 3 fraction group, there was a median decrease in tumor size of 10% between the first and third fractions; within the 5 fraction group, a total decrease in tumor size of 4% was observed between the first and the fifth fractions. On univariate analysis, only location (pZ0.0843), dose (pZ0.0881), and histology (pZ0.0903) correlated with interfraction tumor size reduction. On multivariate analysis, only adenocarcinoma histology (pZ0.0736) and renal cell carcinoma histology (pZ0.0221) remained statistically significant. Conclusion: In this study, we observed interfraction changes in tumor volume in patients treated with SBRT. Of all the patient, tumor, and
International Journal of Radiation Oncology Biology Physics
prevent premature switching to other, less-effective systemic therapies by maximizing the duration and efficacy of EGFR-TKI treatment in selected patients. Author Disclosure: A.H. Kesarwala: Research Grant; ASTRO. C. Kim: None. J.C. Jones: None. A. Kaushal: None. N. Roper: None. C.D. Hoang: None. E. Szabo: None. M. Connolly: None. E. Padiernos: None. C. Cultraro: None. M. Waris: None. S. Gao: None. S.M. Steinberg: None. J. Khan: None. A. Rajan: None. U. Guha: Research Grant; AstraZeneca.
3109
3108 Metachronous Versus Synchronous Versus Single Tumor Nonesmall Cell Lung Cancers: Is There a Difference in Outcomes? B. Khodayari, R. Rahimian, J. Rahimian, R. Wang, and M.R. Girvigian; Kaiser Permanente Los Angeles Medical Center, Los Angeles, CA Purpose/Objective(s): Patients with early stage (ES) NSCLC are at risk for synchronous or metachronous second primary lung cancers (SPLC). The aim of this study is to investigate the outcomes of patients treated with stereotactic body radiation therapy (SBRT) for synchronous or metachronous tumors versus patients with traditional single tumor ESNSCLC. Materials/Methods: We retrospectively identified 107 cases of single tumor ES-NSCLC treated at our institution from 2008 to 2015. An additional 13 patients with 29 unique tumors were identified to have SPLC; 31% of which had synchronous tumors and 69% had metachronous tumors. Of the SPLCs, 76% were biopsy proven and no patient had two different tumor histologies. SPLC was defined as an isolated parenchymal lung tumor separate from the first primary lung cancer (FPLC) and without associated radiographic evidence of regional or distant metastatic disease on PET scan. Kaplan Meier method with log-rank test was used for overall survival (OS), lung cancer specific survival (LCSS), progression-free survival (PFS, progression defined as development of any recurrence or SPLC), local control (LC), and freedom from distant failure (FFDF) analysis. Results: There was an approximate 1% per year incidence of SPLC, similar to that of historical controls. The median age was 79 [61-89] years old. The median follow-up time for single tumor patients was 26 [6-58] months and for synchronous SPLC patients was 25 [14-40] months. For metachronous tumor patients, median time from FPLC to SPLC was 53 [12-262] months, from SPLC to last follow-up was 14 [12-40] months, and overall median time from FPLC to last follow-up was 71 [17-276] months. In this series, all metachronous FPLCs were managed with surgery. All metachronous SPLCs, synchronous SPLCs, and single tumors were non-operable and treated with SBRT (median dose 50 Gy in 5 fractions). Patients with metachronous tumors demonstrated an OS advantage as compared to synchronous and single tumors (2-year OS 100% vs 80% vs 60%, p Z 0.04). However, this was attributable to metachronous patients having a longer follow-up time, leading to a survival bias with fewer deaths from intercurrent disease; this was demonstrated by the similar outcomes for LCSS (100% vs 80% vs 92%, p Z 0.23). There was no difference in PFS when comparing all tumor groups. The distant pattern of failure between the three groups were similar (Metachronous vs Synchronous vs Single tumor 2-year FFDF: 36% vs 80% vs 68%, p Z 0.60). There was also very good local control of disease after SBRT (2-year LC 100% vs 100% vs 92%, p Z 0.74). Conclusion: Patients with synchronous or metachronous tumors have similar outcomes to traditional single tumor ES-NSCLC patients. These patients demonstrated excellent LC and LCSS after SBRT. Although further investigation in larger cohorts is warranted, this study suggests that carefully selected patients with no clinical or radiographic evidence for advanced stage disease could be considered for SBRT to all solitary primary lung tumors. Author Disclosure: B. Khodayari: None. R. Rahimian: None. J. Rahimian: None. R. Wang: None. M.R. Girvigian: None.
Use of Hypofractionated Radiation Therapy With Concurrent Chemotherapy in Inoperable Stage II/ III Nonesmall Cell Lung Cancer E. Kim,1 K.D. Westover,2 M.E. Daly,3 J.D. Murphy,4 and T. Mitin5; 1 Department of Radiation Oncology, Vanderbilt University Medical Center, Nashville, TN, 2University of Texas Southwestern Medical Center, Dallas, TX, 3University of California Davis Comprehensive Cancer Center, Sacramento, CA, 4University of California, San Diego, La Jolla, CA, 5 Oregon Health and Science University, Portland, OR Purpose/Objective(s): Precision hypofractionated radiation therapy (PHRT) without concurrent chemotherapy in inoperable Stage II/III nonsmall cell lung cancer (NSCLC) appears to be safe and effective in early results from randomized trials. PHRT can yield a higher biological effective dose and shorten the total treatment time, leading to more efficient treatment. However, the safety and efficacy of PHRT with concurrent chemotherapy is not well studied. Materials/Methods: We searched the National Cancer Database (NCDB) for Tany N+ M0 NSCLC diagnosed 2004-2013 that received chemotherapy and radiotherapy (RT) in their primary course of treatment. Conventional fractionated chemoradiation (CF-CRT) was defined as 58-80 Gy in 30-45 fractions. Hypofractionated chemoradiation (HF-CRT) was defined as 45-60 Gy in 15 fractions. Patient demographic, diagnosis, and treatment information and overall survival (OS) were compared with descriptive statistics, multivariable logistic regression, and single and multivariable survival analyses. Propensity score matching was used to try to reduce selection bias and confounding variables between the two groups. Results: A total of 126 (0.5%) patients received HF-CRT and 24,286 (99.5%) patients received CF-CRT. Even before 1:1 propensity score matching, the two groups were similar in demographics, Charlson-Deyo comorbidities score, and grade. Before matching, the HF-CRT group was more likely to be treated at academic research facility than CF-CRT group (52% vs 30%, p<0.0001), but after matching, 52% of both groups were treated at an academic/research facility. On univariable and multivariable survival analysis, CF-CRT was associated with better overall survival (OS). On Kaplan-Meier analysis, 1-year and median OS were 68% and 15 months for CF-CRT vs 46% and 10 months for HF-CRT, respectively, with hazard ratio (HR) of 1.9 (95% confidence interval [CI] 1.3-2.7) on MVA. Female gender, later year of diagnosis, living in East/Atlantic region, and unknown grade were also associated with higher OS. Conclusion: Use of HF-CRT has been uncommon in the US in the past decade. Interpretation of outcomes is limited due to the retrospective nature of this study, the unavailability of dosimetric data and the small cohort, despite the use of this national hospital-based database capturing >70% of cancers diagnosed in the US. Nonetheless, this provides valuable information on real-world outcomes of new treatment regimens currently under consideration for trials. While a recently presented randomized phase 3 trial from UTSW showed comparative outcomes between PHRT and CRT, our retrospective study raises hypotheses concerning the safety of combining hypofractionated radiation with concurrent chemotherapy. Carefully designed prospective clinical trials with strict dosimetric guidelines will be required to understand the feasibility of combining PHRT with systemic agents. Author Disclosure: E. Kim: None. K.D. Westover: None. M.E. Daly: Research Grant; American Cancer Society, Department of Defense, NIH/ NCI. J.D. Murphy: None. T. Mitin: Honoraria; Novocure, UpToDate Inc.. Consultant; Novocure. Advisory Board; Novocure. Travel Expenses; Novocure.
3110 Modern Non-operative Treatment Strategies for Patients With Stage II-III Nonesmall Cell Lung Cancer in the United States E. Kim,1 M.E. Daly,2 K.D. Westover,3 J.D. Murphy,4 and T. Mitin5; 1 Department of Radiation Oncology, Vanderbilt University Medical
Volume 99 Number 2S Supplement 2017 Center, Nashville, TN, 2University of California Davis Comprehensive Cancer Center, Sacramento, CA, 3University of Texas Southwestern Medical Center, Dallas, TX, 4University of California, San Diego, La Jolla, CA, 5Oregon Health and Science University, Portland, OR Purpose/Objective(s): Standard of care for inoperable N+ M0 non-small cell lung cancer (NSCLC) is concurrent platinum-based chemoradiation (CRT) and thoracic radiation to 60 Gy in daily fractions. However, many patients with a poor performance status or social limitations may receive radiation monotherapy (RT). The primary aim of this study was to compare characteristics and survival of patients who received standard CRT vs RT alone. Materials/Methods: The national cancer database (NCDB) captures >70% of incident cancer diagnoses in the United States. We selected patients diagnosed with clinical or pathological stage Tany N+ M0 during 2004-2013 who did not undergo definitive surgery. Definitive RT was defined as 58-80 Gy in 30-45 fractions or 45-60 Gy in 15 fractions, starting within 90 days of diagnosis. Definitive CRT was defined as RT with systemic chemotherapy administered within 180 days of diagnosis, without distinction between concurrent and sequential administration. Patient demographic, diagnosis, and treatment information and overall survival (OS) were compared with descriptive statistics, multivariable logistic regression, and single and multivariable survival analyses. Results: After applying selection criteria, 89,195 patients had clinical or pathological stage Tany N+ M0, and of these 3,073 (3.4%) received RT, 24,412 (27%) received CRT, 49,394 (55%) received chemo, and 12,316 (14%) did not receive any treatment. Eleven percent of patients (3,073 out of 27,485) receiving RT did not receive chemo. We used 1:1 propensity score matching to compare RT and CRT groups, focusing on 2,883 patients in each group. RT patients had older age (median 75 vs 70, range 38-90 in both). Baseline and cancer characteristics were comparable except race (RT 85% vs CRT 82% white, pZ0.04), and insurance (RT 70% vs CT 67% Medicare, pZ0.02). In multivariable analysis the groups were comparable except for borderline significance of Medicaid insurance (odds ratio [OR] 1.3, 95% confidence interval [CI] 1.0-1.7) and Charlson-Deyo comorbidity score of 1 (OR 1.1, 95% CI 1.0-1.3) favoring CRT. Kaplan-Meier survival analysis found improved OS associated with CRT (5-year 15%, median 17 months) vs RT (5-year 8%, median 12 months) (p<0.0001), with a hazard ratio (HR) of 0.7 (95%CI 0.7-0.7). Improved survival was also associated with female gender, younger age, insurance other than Medicare/unknown/ none, fewer comorbidities, treatment at an academic/research facility, and earlier stage at diagnosis. Conclusion: Of patients referred for definitive chemoradiation, 11% received RT alone. Definitive CRT was associated with higher OS when compared to RT alone but selection bias may have contributed since patients who received RT alone had more comorbidities and were older than patients who received definitive CRT. Novel approaches are needed to improve the outcomes in patients who are not candidate for systemic chemotherapy with concurrent thoracic RT, such as combination of RT with targeted and immunotherapy agents. Author Disclosure: E. Kim: None. M.E. Daly: Research Grant; American Cancer Society, Department of Defense, NIH/NCI. K.D. Westover: None. J.D. Murphy: None. T. Mitin: Honoraria; Novocure, UpToDate Inc. Consultant; Novocure. Advisory Board; Novocure. Travel Expenses; Novocure.
3111 Improved Survival With Postoperative Radiation Therapy in Thymic Carcinoma: A PropensityMatched Analysis of the Surveillance, Epidemiology, and End Results Database Y.J. Lim,1 C. Song,2 and J.S. Kim2; 1Department of Radiation Oncology, Kyung Hee University Medical Center, Seoul, Korea, Republic of (South), 2 Department of Radiation Oncology, Seoul National University Bundang Hospital, Seongnamsi, Korea, Republic of (South) Purpose/Objective(s): Thymic carcinoma is a rare and aggressive malignancy with poor prognosis. Although postoperative radiotherapy
Poster Viewing E469 (PORT) is used for obtaining better locoregional tumor control, its association with survival has not been established. This study evaluated the prognostic impact of PORT in thymic carcinoma. Materials/Methods: We identified patients diagnosed with thymic carcinoma between 2004 and 2013 using the Surveillance, Epidemiology, and End Results (SEER) database. Propensity score (PS) matching with Kaplan-Meier and Cox-regression analyses were used to assess prognosis. Results: In the unmatched population (nZ312), 184 (59%) patients underwent PORT. Overall 5-year survival rates were better with receipt of PORT, both before and after matching (PZ0.012 and 0.007, respectively). After adjusting for related covariates (nZ256), age 63 years (PZ0.023), Masaoka stage III (PZ0.028) and IV (P<0.001), debulking surgery (PZ0.021), and no receipt of PORT (PZ0.013) were independently poor prognostic factors. In subgroup analyses, favorable survival impacts of PORT were observed for Masaoka stage III tumors (hazard ratio [HR] 0.31, 95% confidence interval [CI] 0.15e0.66), tumors sized >6.0 cm (HR 0.48, 95% CI 0.26e0.89), node-negative status (HR 0.58, 95% CI 0.33e1.00), and surgical extent of local excision or partial removal (HR 0.44, 95% CI 0.22e0.86). Conclusion: On SEER analysis, survival benefits of PORT in thymic carcinoma were demonstrated. With strong prognostic associations of Masaoka stage and types of surgery, PORT should be considered for nonmetastatic locally advanced tumors with limited surgical resection. Author Disclosure: Y. Lim: None. C. Song: None. J. Kim: None.
3112 Interfraction Tumor Size Changes During Lung SBRT: What Factors Matter? K. King,1 S. Mayekar,1 G. Marwaha,2 G. Tolekidis,1 G. Cifter,2 and H. Zhen2; 1Rush University Medical Center, Chicago, IL, 2Rush Medical Center, Chicago, IL Purpose/Objective(s): Prior studies have suggested that lung stereotactic body radiation therapy (SBRT) can cause volumetric tumor size changes on setup imaging during treatment, however predictors for such changes have not been well described. With the advent of kilovoltage (kv) cone beam computed tomography (CBCT) scans in the SBRT setup process, visualization of tumor changes are more readily apparent. The purpose of this study was to evaluate the role played by patient, tumor, and dosimetric factors in the interfraction tumor changes seen in our lung SBRT patient population. Materials/Methods: A retrospective chart review was conducted on forty consecutive patients from our IRB-approved lung SBRT database treated between 2014-2016. Each patient was set up under daily stereoscopic x-ray and kv CBCT guidance. A single radiation oncologist retrospectively recontoured the tumor volume on each sequential kv CBCT image to determine volumetric variances. A univariate and multivariate analysis with a linear model were employed in SPSS R 3.3.1. Results: Forty patients with either squamous cell carcinoma (12.5%), adenocarcinoma (52.5%), or metastatic renal cell carcinoma (5%) of the lung were identified. 20 (50%) patients received an SBRT dose of 54 Gy / 3 fxs and 20 (50%) patients received 50-60 Gy / 5 fxs. 8 (20%) of the tumors were centrally located (by the RTOG 0813 definition) while 32 (80%) were peripheral. Six (15%) of the patients had never smoked while 28 (78%) were former smokers and 6 (15%) continued to smoke during their treatment. The median pretreatment neutrophil lymphocyte ratio (NLR) was 3.03 (range: 0.09-67.21) and the median post-treatment NLR was 2.74 (range: 0.17-12.59). Within the 3 fraction group, there was a median decrease in tumor size of 10% between the first and third fractions; within the 5 fraction group, a total decrease in tumor size of 4% was observed between the first and the fifth fractions. On univariate analysis, only location (pZ0.0843), dose (pZ0.0881), and histology (pZ0.0903) correlated with interfraction tumor size reduction. On multivariate analysis, only adenocarcinoma histology (pZ0.0736) and renal cell carcinoma histology (pZ0.0221) remained statistically significant. Conclusion: In this study, we observed interfraction changes in tumor volume in patients treated with SBRT. Of all the patient, tumor, and