Table 1. Comparison of baseline characteristics of unresected, stage III NSCLC patients treated with systemic therapy, radiotherapy, and concurrent chemoradiotherapy1,2.
| All patients (N = 3,799) |
Systemic therapy only (N = 823) |
Radiotherapy only (N = 1,000) |
Concurrent chemoradiotherapy (N = 1,976) |
P-value3 | |
|---|---|---|---|---|---|
| Characteristics assessed at diagnosis | |||||
| Age at diagnosis, years; mean (SD) [median] | 74.9 ± 6.4 [74.0] | 75.5 ± 6.7 [75.0] | 77.2 ± 7.0 [77.0] | 73.5 ± 5.4 [73.0] | <0.001* |
| Female, N (%) | 1,800 (47.4%) | 408 (49.6%) | 494 (49.4%) | 898 (45.4%) | 0.045* |
| Race/Ethnicity, N (%) | |||||
| White | 3,194 (84.1%) | 680 (82.6%) | 814 (81.4%) | 1,700 (86.0%) | 0.002* |
| Black | 379 (10.0%) | 74 (9.0%) | 138 (13.8%) | 167 (8.5%) | <0.001* |
| Asian | 134 (3.5%) | 43 (5.2%) | 28 (2.8%) | 63 (3.2%) | 0.010* |
| Other4 | 92 (2.4%) | 26 (3.2%) | 20 (2.0%) | 46 (2.3%) | 0.257 |
| Region, N (%) | |||||
| West | 1,236 (32.5%) | 324 (39.4%) | 318 (31.8%) | 594 (30.1%) | <0.001* |
| South | 1,241 (32.7%) | 222 (27.0%) | 354 (35.4%) | 665 (33.7%) | <0.001* |
| Northeast | 818 (21.5%) | 208 (25.3%) | 194 (19.4%) | 416 (21.1%) | 0.008* |
| Midwest | 504 (13.3%) | 69 (8.4%) | 134 (13.4%) | 301 (15.2%) | <0.001* |
| Setting of residence, N (%)5 | |||||
| Big metro | 1,910 (50.3%) | 454 (55.2%) | 515 (51.5%) | 941 (47.6%) | <0.001* |
| Metro | 1,174 (30.9%) | 229 (27.8%) | 307 (30.7%) | 638 (32.3%) | 0.066 |
| Urban | 238 (6.3%) | 63 (7.7%) | 51 (5.1%) | 124 (6.3%) | 0.081 |
| Less Urban | 374 (9.8%) | 57 (6.9%) | 102 (10.2%) | 215 (10.9%) | 0.005* |
| Rural | 103 (2.7%) | 20 (2.4%) | 25 (2.5%) | 58 (2.9%) | 0.673 |
| Primary tumor site, N (%) | |||||
| Main bronchus | 225 (5.9%) | 21 (2.6%) | 68 (6.8%) | 136 (6.9%) | <0.001* |
| Lower lobe | 964 (25.4%) | 248 (30.1%) | 269 (26.9%) | 447 (22.6%) | <0.001* |
| Mid-lobe | 146 (3.8%) | 24 (2.9%) | 34 (3.4%) | 88 (4.5%) | 0.109 |
| Upper lobe | 2,123 (55.9%) | 405 (49.2%) | 553 (55.3%) | 1,165 (59.0%) | <0.001* |
| Other6 | 341 (9.0%) | 125 (15.2%) | 76 (7.6%) | 140 (7.1%) | <0.001* |
| AJCC Stage | |||||
| IIIA | 2,260 (59.5%) | 418 (50.8%) | 632 (63.2%) | 1,210 (61.2%) | <0.001* |
| IIIB | 1,539 (40.5%) | 405 (49.2%) | 368 (36.8%) | 766 (38.8%) | <0.001* |
| Grade | |||||
| Well differentiated | 94 (2.5%) | 31 (3.8%) | 23 (2.3%) | 40 (2.0%) | 0.024* |
| Moderately differentiated | 644 (17.0%) | 130 (15.8%) | 185 (18.5%) | 329 (16.6%) | 0.271 |
| Poorly differentiated | 1,168 (30.7%) | 219 (26.6%) | 307 (30.7%) | 642 (32.5%) | 0.009* |
| Other7 | 1893 (49.8%) | 443 (53.8%) | 485 (48.5%) | 965 (48.8%) | 0.034* |
| Histology (IDC-O-3), N (%)8 | |||||
| Adenocarcinoma (8140) | 1,366 (36.0%) | 408 (49.6%) | 286 (28.6%) | 672 (34.0%) | <0.001* |
| Squamous cell (8070) | 1,608 (42.3%) | 243 (29.5%) | 480 (48.0%) | 885 (44.8%) | <0.001* |
| General NSCLC (8046) | 490 (12.9%) | 90 (10.9%) | 147 (14.7%) | 253 (12.8%) | 0.057 |
| Other | 335 (8.8%) | 82 (10.0%) | 87 (8.7%) | 166 (8.4%) | 0.409 |
| Characteristics assessed during baseline period | |||||
| Index year of first therapeutic regimen, N (%) | |||||
| 2009 | 851 (22.4%) | 242 (29.4%) | 231 (23.1%) | 378 (19.1%) | <0.001* |
| 2010 | 740 (19.5%) | 160 (19.4%) | 200 (20.0%) | 380 (19.2%) | 0.882 |
| 2011 | 743 (19.6%) | 153 (18.6%) | 197 (19.7%) | 393 (19.9%) | 0.726 |
| 2012 | 720 (19.0%) | 126 (15.3%) | 168 (16.8%) | 426 (21.6%) | <0.001* |
| 2013 | 698 (18.4%) | 133 (16.2%) | 186 (18.6%) | 379 (19.2%) | 0.167 |
| Number of unique metastatic sites,9 mean (SD) [median] | 0.8 ± 0.9 [1.0] | 0.9 ± 1.0 [1.0] | 0.7 ± 0.9 [0.0] | 0.9 ± 0.9 [1.0] | <0.001* |
| Charlson Comorbidity Index,10 mean (SD) [median] | 5.5 ± 2.5 [5.0] | 5.8 ± 2.5 [6.0] | 5.6 ± 2.5 [5.0] | 5.3 ± 2.4 [5.0] | <0.001* |
| Comorbidities present during baseline period, N (%) | |||||
| Hypertension | 2,979 (78.4%) | 659 (80.1%) | 808 (80.8%) | 1,512 (76.5%) | 0.012* |
| COPD | 2,871 (75.6%) | 579 (70.4%) | 806 (80.6%) | 1,486 (75.2%) | <0.001* |
| Dyslipidemia | 2,481 (65.3%) | 551 (67.0%) | 589 (58.9%) | 1,341 (67.9%) | <0.001* |
| Ischemic heart disease | 1,580 (41.6%) | 337 (40.9%) | 457 (45.7%) | 786 (39.8%) | 0.008* |
| Diabetes | 1,279 (33.7%) | 300 (36.5%) | 350 (35.0%) | 629 (31.8%) | 0.036* |
| Cerebrovascular disease | 908 (23.9%) | 193 (23.5%) | 273 (27.3%) | 442 (22.4%) | 0.011* |
| Heart failure | 716 (18.8%) | 174 (21.1%) | 273 (27.3%) | 269 (13.6%) | <0.001* |
| Chronic kidney disease | 452 (11.9%) | 107 (13.0%) | 161 (16.1%) | 184 (9.3%) | <0.001* |
| No comorbidities | 94 (2.5%) | 14 (1.7%) | 22 (2.2%) | 58 (2.9%) | 0.129 |
| Tumor marker tests during baseline period11 | |||||
| Patients with ≥1 test, N (%) | 239 (6.3%) | 84 (10.2%) | 37 (3.7%) | 118 (6.0%) | <0.001* |
| Predicted performance status, N (%)12 | |||||
| Good | 2,631 (57.9%) | 554 (57.5%) | 530 (42.5%) | 1,547 (66.3%) | <0.001* |
| Poor | 1,913 (42.1%) | 409 (42.5%) | 717 (57.5%) | 787 (33.7%) | <0.001* |
* Significant at the 5% level
Abbreviations:
NSCLC: non-small cell lung cancer; SD: standard deviation
[1] The baseline period is defined as the six months prior to initiation of the first therapeutic regimen for NSCLC. Initiation of the first therapeutic regimen was required to occur within 90 days after the initial NSCLC diagnosis.
[2] Patients who were determined to have only received systemic therapy had claims for targeted therapy or chemotherapy within 90 days of their initial NSCLC diagnosis, and no claims for radiotherapy during the same period. Patients who were determined to have only received radiotherapy had radiotherapy claims within 90 days of their initial NSCLC diagnosis, and no claims for targeted therapy or chemotherapy during the same period. Concurrent chemoradiotherapy (cCRT) patients had claims for radiotherapy along with either targeted therapy or chemotherapy, within 90 days of their initial NSCLC diagnosis.
[3] Kruskal-Wallis tests were used to compare categorical variables. Analysis of variance (ANOVA) tests were used to compare continuous variables.
[4] Other category includes Hispanic, Native American, Unknown and races classified as “Other” by SEER-Medicare data.
[5] Setting of residence classifications are based on Rural-Urban Continuum Codes that distinguish metropolitan (metro) counties by the population size of their metro area, and nonmetropolitan counties by degree of urbanization and adjacency to a metro or rural areas.
[6] Other category includes primary tumor site of “Overlap” and “Lung, unspecified”.
[7] Other category includes “Undifferentiated” or “Unknown” grade.
[8] Only the three most frequently observed histology types in the study population are reported in this table.
[9] Metastatic sites were evaluated based on ICD-9 codes. Diagnoses of other malignancies were considered to be metastases since all patients in the study population were required to only have 1 primary tumor (NSCLC). Metastatic sites included but were not limited to diagnosis codes for lymphatic and hematopoietic tissue cancer, respiratory cancer (excluding cancer of the lungs), bone and bone marrow cancer, brain and spinal cord cancer, tissue cancer, and endocrine cancer among others.
[10] CCI was calculated based on the method described in Quan et al. (2005). Source: Quan H, Sundararajan V, Halfon P et al. Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Medical Care 2005;43:1130–1139.
[11] Tumor marker tests during baseline period include EGFR gene, KRAS gene, BRAF gene, molecular cytogenetics (e.g., FISH for anaplastic lymphoma kinase or ROS-1 gene arrangement), morphometric analysis, tumor immunohistochemistry (e.g., programmed death-ligand 1, programmed cell death protein 1), multiple gene panel, lung cancer panel, and proteomic testing panel.
[12] Predicted performance status was calculated using age at diagnosis, COPD status, number of inpatient stays, any outpatient visit, number of ED visits, any DME claim, and any prescription drug dispensing during the six month baseline period, based on the method described in Salloum et al. (2011). Good predicted performance status was assigned to patients with ≥ 70% probability of having an Eastern Cooperative Oncology Group (ECOG) score of 0–2 or a Karnofsky Performance Scale (KPS) of 100–60.