Original Article
Influence of Conformal Radiotherapy Technique on Survival After Chemoradiotherapy for Patients With Stage III Non-Small Cell Lung Cancer in the National Cancer Data Base David J. Sher, MD, MPH1; Matthew Koshy, MD2; Michael J. Liptay, MD3; and Mary Jo Fidler, MD4
BACKGROUND: Definitive chemoradiotherapy is a core treatment modality for patients with stage III non-small cell lung cancer (NSCLC). Although radiotherapy (RT) technologies have advanced dramatically, to the authors’ knowledge relatively little is known regarding the importance of irradiation technique on outcome, particularly given the competing risk of distant metastasis. The National Cancer Data Base was used to determine predictors of overall survival (OS) in patients with AJCC stage III NSCLC who were treated with chemoradiotherapy, focusing on the importance of conformal RT (CRT). METHODS: Patients with stage III NSCLC who were treated with chemoradiotherapy between 2003 and 2005 in the National Cancer Data Base were included. RT technique was defined as conventional, 3-dimensional–conformal, or intensity-modulated RT (IMRT), the latter 2 combined as CRT. Cox proportional hazards regression was performed for univariable and multivariable analyses of OS. RESULTS: The median, 3-year, and 5-year survival outcomes for the 13,292 patients were 12.9 months, 19%, and 11%, respectively. The 3-year and 5-year survival probabilities of patients receiving CRT versus no CRT were 22% versus 19% and 14% versus 11%, respectively (P 1500 CoC-accredited institutions, and thus the NCDB includes > 70% of patients newly diagnosed with cancer. Although case completeness has not been assessed in patients with lung cancer in this database, Mallin et al recently showed that the NCDB captured 87% and 83% of chemotherapy claims among patients with breast and colorectal cancer, respectively, and 86% and 84%, respectively, of RT claims among these same patients, suggesting quite acceptable capture of relevant patient treatment information.10 Cohort Definition
Patients diagnosed with stage III NSCLC between 2003 and 2005 and treated with definitive chemoradiotherapy were eligible for inclusion in this cohort; patients with small cell and carcinoid tumors were excluded. This time window was used to ensure that all patients were assigned a comorbidity score and RT modality. The NCDB “analytic stage” was used, which defaulted to clinical stage in cases without pathologic staging studies (eg, mediastinoscopy). Figure 1 details the inclusion=exclusion criteria and the cohort selection. Treatment Assignment
Patients were required to have initiated RT and chemotherapy within 3 weeks of each other to signify concurrent therapy. Patients without specified start dates were excluded. Determination of Predictor Variables Figure 1. Flowchart describing the composition of the patient cohort is shown. NSCLC indicates non-small cell lung cancer; CoC, Commission on Cancer; RT, radiotherapy.
The National Cancer Data Base (NCDB) is a cancer registry of patients treated at hospitals accredited by the American College of Surgeons Commission on Cancer (CoC); it contains diverse clinical, demographic, and socioeconomic data. To help define the role of technology in Cancer
July 1, 2014
Predictors were divided into clinical, time=geographic, socioeconomic, and institutional. Several of the ordinal variables were categorized by the NCDB (ie, educational level, income in the patient zip code) as quartiles relative to the US population, and the other continuous variables were categorized as quartiles within the data set (eg, age, distance from treatment center). RT modality was reported for the initial RT course as well as a boost course. The modality was defined as either intensity-modulated RT (IMRT), 3D-CRT, or a beam energy (eg, 6- to 10-megavolt photons). Patients 2061
Original Article TABLE 1. Patient and Disease Characteristics
TABLE 1. Continued Characteristics
Characteristics Clinical factors Median age at diagnosis (IQR), y Sex Male Female Histology Squamous cell carcinoma Nonsquamous cell carcinoma AJCC Stage IIIA IIIB III (not otherwise specified) Clinical T classification 1 2 3 4 0, X Clinical N classification 1 2 3 0, X Comorbidity score 0 1 2 RT modality Conventional RT 3D-CRT IMRT Geographic factors Location of facility Atlantic/Northeast South/Southeast Midwest/Great Lakes West/Mountain/Pacific Socioeconomic factors Race/ethnicity White Nonwhite Income $46,000 $35,000-$45,999 $30,000-$34,999 $30,000 Unknown Insurance Private insurance or managed care Medicare Medicaid Other Uninsured Unknown Institutional factors Facility type Academic/research program Comprehensive community cancer program Community cancer program Other Yearly patient volume 8.67 cases/y
2062
All Patients No. (%)
All Patients No. (%)
65 (58–72) 8009 (60%) 5283 (40%) 8473 (63.8%) 4819 (36.2%) 5465 (41.1%) 7154 (53.8%) 673 (5.1%) 1204 3614 2377 4963 1134
(9.1%) (27.2%) (17.9%) (37.3%) (8.5%)
827 7427 2689 2249
(6.2%) (55.9%) (20.2%) (17.7%)
9413 (70.8%) 2996 (22.5%) 883 (6.7%) 11,725 (88.2%) 1330 (10%) 237 (1.8%)
2428 4293 4140 2431
(18.3%) (32.3%) (31.1%) (18.3%)
11,263 (84.7%) 2029 (15.3%) 3886 3812 2740 2229 625
(29.2%) (28.7%) (20.6%) (16.8%) (4.7%)
4570 6786 834 227 586 289
(34.4%) (51%) (6.3%) (1.7%) (4.4%) (2.2%)
3373 7290 2428 201
(25.4%) (54.9%) (18.3%) (1.5%)
3188 3425 3254 3425
(24%) (25.8%) (24.4%) (25.8%)
High treatment volume
Influence of Conformal Radiotherapy Technique on Survival After Chemoradiotherapy for Patients With Stage III Non-Small Cell Lung Cancer in the National Cancer Data Base David J. Sher, MD, MPH1; Matthew Koshy, MD2; Michael J. Liptay, MD3; and Mary Jo Fidler, MD4
BACKGROUND: Definitive chemoradiotherapy is a core treatment modality for patients with stage III non-small cell lung cancer (NSCLC). Although radiotherapy (RT) technologies have advanced dramatically, to the authors’ knowledge relatively little is known regarding the importance of irradiation technique on outcome, particularly given the competing risk of distant metastasis. The National Cancer Data Base was used to determine predictors of overall survival (OS) in patients with AJCC stage III NSCLC who were treated with chemoradiotherapy, focusing on the importance of conformal RT (CRT). METHODS: Patients with stage III NSCLC who were treated with chemoradiotherapy between 2003 and 2005 in the National Cancer Data Base were included. RT technique was defined as conventional, 3-dimensional–conformal, or intensity-modulated RT (IMRT), the latter 2 combined as CRT. Cox proportional hazards regression was performed for univariable and multivariable analyses of OS. RESULTS: The median, 3-year, and 5-year survival outcomes for the 13,292 patients were 12.9 months, 19%, and 11%, respectively. The 3-year and 5-year survival probabilities of patients receiving CRT versus no CRT were 22% versus 19% and 14% versus 11%, respectively (P 1500 CoC-accredited institutions, and thus the NCDB includes > 70% of patients newly diagnosed with cancer. Although case completeness has not been assessed in patients with lung cancer in this database, Mallin et al recently showed that the NCDB captured 87% and 83% of chemotherapy claims among patients with breast and colorectal cancer, respectively, and 86% and 84%, respectively, of RT claims among these same patients, suggesting quite acceptable capture of relevant patient treatment information.10 Cohort Definition
Patients diagnosed with stage III NSCLC between 2003 and 2005 and treated with definitive chemoradiotherapy were eligible for inclusion in this cohort; patients with small cell and carcinoid tumors were excluded. This time window was used to ensure that all patients were assigned a comorbidity score and RT modality. The NCDB “analytic stage” was used, which defaulted to clinical stage in cases without pathologic staging studies (eg, mediastinoscopy). Figure 1 details the inclusion=exclusion criteria and the cohort selection. Treatment Assignment
Patients were required to have initiated RT and chemotherapy within 3 weeks of each other to signify concurrent therapy. Patients without specified start dates were excluded. Determination of Predictor Variables Figure 1. Flowchart describing the composition of the patient cohort is shown. NSCLC indicates non-small cell lung cancer; CoC, Commission on Cancer; RT, radiotherapy.
The National Cancer Data Base (NCDB) is a cancer registry of patients treated at hospitals accredited by the American College of Surgeons Commission on Cancer (CoC); it contains diverse clinical, demographic, and socioeconomic data. To help define the role of technology in Cancer
July 1, 2014
Predictors were divided into clinical, time=geographic, socioeconomic, and institutional. Several of the ordinal variables were categorized by the NCDB (ie, educational level, income in the patient zip code) as quartiles relative to the US population, and the other continuous variables were categorized as quartiles within the data set (eg, age, distance from treatment center). RT modality was reported for the initial RT course as well as a boost course. The modality was defined as either intensity-modulated RT (IMRT), 3D-CRT, or a beam energy (eg, 6- to 10-megavolt photons). Patients 2061
Original Article TABLE 1. Patient and Disease Characteristics
TABLE 1. Continued Characteristics
Characteristics Clinical factors Median age at diagnosis (IQR), y Sex Male Female Histology Squamous cell carcinoma Nonsquamous cell carcinoma AJCC Stage IIIA IIIB III (not otherwise specified) Clinical T classification 1 2 3 4 0, X Clinical N classification 1 2 3 0, X Comorbidity score 0 1 2 RT modality Conventional RT 3D-CRT IMRT Geographic factors Location of facility Atlantic/Northeast South/Southeast Midwest/Great Lakes West/Mountain/Pacific Socioeconomic factors Race/ethnicity White Nonwhite Income $46,000 $35,000-$45,999 $30,000-$34,999 $30,000 Unknown Insurance Private insurance or managed care Medicare Medicaid Other Uninsured Unknown Institutional factors Facility type Academic/research program Comprehensive community cancer program Community cancer program Other Yearly patient volume 8.67 cases/y
2062
All Patients No. (%)
All Patients No. (%)
65 (58–72) 8009 (60%) 5283 (40%) 8473 (63.8%) 4819 (36.2%) 5465 (41.1%) 7154 (53.8%) 673 (5.1%) 1204 3614 2377 4963 1134
(9.1%) (27.2%) (17.9%) (37.3%) (8.5%)
827 7427 2689 2249
(6.2%) (55.9%) (20.2%) (17.7%)
9413 (70.8%) 2996 (22.5%) 883 (6.7%) 11,725 (88.2%) 1330 (10%) 237 (1.8%)
2428 4293 4140 2431
(18.3%) (32.3%) (31.1%) (18.3%)
11,263 (84.7%) 2029 (15.3%) 3886 3812 2740 2229 625
(29.2%) (28.7%) (20.6%) (16.8%) (4.7%)
4570 6786 834 227 586 289
(34.4%) (51%) (6.3%) (1.7%) (4.4%) (2.2%)
3373 7290 2428 201
(25.4%) (54.9%) (18.3%) (1.5%)
3188 3425 3254 3425
(24%) (25.8%) (24.4%) (25.8%)
High treatment volume
