SEARCH

SEARCH BY CITATION

Keywords:

  • African American;
  • racial disparities;
  • Hispanic;
  • nonsmall cell lung cancer;
  • socioeconomic status;
  • marital status;
  • survival

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

BACKGROUND.

Racial minorities exhibit poor survival with nonsmall cell lung cancer (NSCLC) that generally is attributed to low socioeconomic status (SES). In this study, the authors investigated the role of SES in this survival disparity among patients with stage I NSCLC.

METHODS.

A case-only analysis was performed on California Cancer Registry (CCR) data (1989–2003). Univariate survival analyses were performed using the Kaplan-Meier method. Multivariate survival analyses were performed using Cox proportional hazards ratios.

RESULTS.

In total, 19,702 incident cases of stage I NSCLC were analyzed. Low SES was identified more commonly in African-American and Hispanic patients and was associated significantly with men, unmarried status, stage IB disease, squamous cell histology, poorly differentiated tumors, fewer surgical resections performed, and less overall treatment received. Reasons for no surgery were associated strongly with low SES and unmarried status but not with race. In multivariate analysis, each incremental improvement in SES quintile was associated with statistically significant decreases in the hazard ratios (HRs) for death (second SES quintile [SES2] vs SES1: HR, 0.91; 95% confidence interval [95% CI], 0.85–0.98; SES3 vs SES1: HR, 0.90; 95% CI, 0.84–0.97; SES4 vs SES1: HR, 0.83; 95% CI, 0.77–0.89; SES5 vs SES1: HR, 0.78; 95% CI, 0.72–0.84; Ptrend < .0001). African-American or Hispanic race was not an independent poor prognostic factor for survival after adjustment for surgery, SES, and marital status.

CONCLUSIONS.

Low SES was an independent poor prognostic factor for survival in patients with stage I NSCLC and was independent of surgery, race, and marital status. Cancer 2008. © 2008 American Cancer Society.

Lung cancer is the leading cause of cancer death in the world.1 Despite advances in treatment, racial minorities with lung cancer in the United States have poorer survival compared with Caucasians.2–6 Explanations for this disparity in survival include advanced stage at diagnosis7 and disparities in the receipt of cancer care,3 such as a decreased likelihood of being offered treatment2, 8, 9 or lower acceptance of curative treatment.10, 11 It also has been demonstrated that low socioeconomic status (SES) leads to poorer cancer outcome and survival,2 and similar explanations have been offered to account for this disparity in outcome, such as advanced stage at diagnosis,7 increased comorbidities, poorer access to specialty care, lower acceptance of treatment, and lower incidence of treatment received.12 However, most racial minority cancer patients are found in the lower SES strata, which may confound these types of analyses.2–4

In this study, we set out to determine how SES affects the survival of racial minority patients with stage I nonsmall cell lung cancer (NSCLC) by using cases from the population-based California Cancer Registry (CCR). We restricted our analysis to patients with stage I NSCLC to limit some of the potential confounding factors, such as delay in the time to diagnosis, which usually results in advanced stage at diagnosis. Second, because surgical resection is the primary curative modality of treatment in early-stage NSCLC, restricting our investigation to stage I NSCLC allowed us to analyze how a single curative treatment modality is affected by SES and other factors. Moreover, because of the racial diversity in California, the CCR contains substantial epidemiological data on African-American, Hispanic, and Asian patients in addition to Caucasian patients. Furthermore, every patient in the CCR had been assigned to 1 of the 5 SES quintiles, whereas these SES variables are not readily available in other population-based databases, such as the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program. The SES variable in the CCR and its relation to race and cancer survival has been used successfully to analyze nasopharyngeal carcinoma,13 cutaneous melanoma,14 and pancreatic cancer.15

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Study Cohort and Diagnostic Codes

We performed a retrospective, case-only analysis of stage I NSCLC cases in the CCR database. The CCR is the largest, contiguous area population-based cancer registry in the world: It collects data on >130,000 new cancer cases per year in California, as described elsewhere,16 and is part of the SEER Program. The state of California legally mandated cancer reporting in 1988, and standardized data-collection and quality-control procedures have been in place ever since. Case reporting is estimated at >98% for the entire state of California.17

Data were obtained on 19,702 incident stage I NSCLC cases between 1989 and 2003 with complete TNM staging data and follow-up data. Tumor site and histology were abstracted as described previously.18, 19 Only patients with histologically or cytologically confirmed NSCLC were included in the analysis. Nonsmall cell histologies were categorized as undifferentiated NSCLC if they were not coded as adenocarcinoma, squamous cell carcinoma, large cell carcinoma, or bronchioloalveolar carcinoma (BAC) or as a metastatic lung lesion from a separate primary tumor, as described previously.18, 19 Ethnicity, marital status, histologic grade, tumor lobar location, tumor size, radiation, and surgical techniques were abstracted by using SEER codes. Chemotherapy given during the first course of therapy and reasons for no surgical treatment were ascertained by using CCR codes. The last date of follow-up was either the date of death or the last date the patient was contacted.

SES

The measure of SES used in this analysis is a composite measure that was created previously by Yost et al.20 by using CCR and census data. Census files were linked to the CCR file based on the cases' block group of residence at the time of diagnosis. Cases that were not able to be geocoded to a street address (5.5% of cases) were allocated randomly to census block groups within their county of residence. Cases diagnosed during the period from 1989 to 1995 were linked to 1990 Census data, and cases diagnosed from 1996 forward were linked to 2000 Census data. Seven indicator variables that measured education, income, and occupation were combined to create a single SES index using principal-component analysis. “Proportion with a blue-collar job” and “proportion with >16 years in the workforce without a job” were used as the 2 measures of occupation. “Median income” and “proportion <200% of the Poverty Level” were used as measures of income. “Median rent” and “median house value” were used to adjust for cost-of-living differences. A composite “education index,” which weighs the proportion of individuals in a census block group with a given level of education by the number of school years needed to attain that level of education, was used as the only indicator of education. Standardized component scores for the SES index for 20,919 census block groups were sorted and categorized into quintiles, with a value of 1 representing the lowest SES level (SES1) and a value of 5 representing the highest SES level (SES5).16, 20

Statistical Analysis

Comparisons of demographic, clinical, and pathologic variables for patients across various categories were performed by using the Pearson chi-square statistic or the Fisher exact test for nominal variables. Analyses of variance with the Tukey post-hoc test were used for multiple comparisons of continuous variables. Survival curves were constructed with the Kaplan-Meier method and were compared by using the log-rank test. Multivariate Cox regression analyses were used to determine which factors were associated significantly with survival. Statistical significance was assumed for a 2-tailed P value <.05. All statistical analyses were performed using SAS version 9.1 (SAS Institute, Cary, NC).

Ethical Considerations

This research study was approved by the University of California Irvine Institutional Review Board (IRB) (IRB no. 2004-3971).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Patient, Tumor, and Treatment Characteristics

In total, 19,702 patients who were diagnosed with stage I NSCLC from 1989 to 2003 were analyzed. The mean age ay diagnosis was 69 years. The median follow-up was 53 months, and 92.7% of the patients were treated with surgery, chemotherapy, or radiation. The 5-year survival rate and the median overall survival (OS) for the whole cohort were 46.2% and 53 months, respectively.

SES

Data on SES were available on 19,700 patients. The distribution of patient and tumor characteristics among the 5 SES quintiles is shown in Table 1. There were statistically significant more African-American and Hispanic patients in the lowest SES (SES1) quintile (African-American, 18.6%; Hispanic, 13.7%) compared with the highest SES (SES5) quintile (African-American, 2%; Hispanic, 3.4%; P < .0001). Significantly more unmarried patients were in SES1 (49.6%) compared with SES5 (35%; P < .0001), more patients with stage IB NSCLC were in SES1 (57.8%) compared with SES5 (52.2%; P < .0001), more patients with squamous cell carcinoma were in SES1 (34.1%) compared with SES5 (24.9%; P < .0001), and a greater proportion of tumors with poorly differentiated histologic grade was observed in SES1 (47.4%) compared with SES5 (40.8%; P < .0001). There was no difference in intrapulmonary tumor location (P = .085) or age at diagnosis (P = .081) among the 5 SES quintiles.

Table 1. Patients With Stage I Nonsmall Cell Lung Cancer Stratified by Socioeconomic Status Quintiles
VariableSES quartile: no. of patients (%)P
1 (Lowest)2345 (Highest)
  • SES indicates socioeconomic status; BAC, bronchioloalveolar carcinoma; RUL, right upper lobe; RML, right middle lobe; RLL, right lower lobe; LUL, left upper lobe; LLL, left lower lobe; NOS, not otherwise specified; R/L, right or left.

  • *

    Surgery, chemotherapy, or radiation.

All patients26403674438544424559 
Ethnic origin     <.0001
 Caucasian1576 (59.7)2767 (75.3)3573 (81.5)3738 (84.2)3989 (87.5) 
 African-American490 (18.6)317 (8.6)268 (6.1)168 (3.8)92 (2) 
 Hispanic362 (13.7)303 (8.3)261 (6)200 (4.5)153 (3.4) 
 Non-Chinese Asian140 (5.3)213 (5.8)196 (4.5)221 (5)185 (4.1) 
 Chinese64 (2.4)63 (1.7)71 (1.6)108 (2.4)122 (2.7) 
 Other8 (0.3)11 (0.3)16 (0.4)7 (0.2)18 (0.4) 
Age, y     .0812
 0–3922 (0.84)20 (0.53)20 (0.46)20 (0.45)26 (0.57) 
 40–4998 (3.7)131 (3.5)140 (3.2)142 (3.2)141 (3.1) 
 50–59384 (14.5)511 (13.9)628 (14.3)638 (14.4)653 (14.3) 
 60–69865 (32.8)1262 (34.4)1325 (30.2)1441 (32.4)1394 (30.6) 
 70–79949 (35.9)1319 (35.9)1715 (39.1)1641 (36.9)1765 (38.7) 
 80–89322 (12.2)431 (11.7)557 (12.7)560 (12.6)580 (12.7) 
Sex     .0001
 Men1429 (54.1)1932 (52.6)2239 (51.1)2239 (50.4)2229 (48.9) 
 Women1211 (45.8)1742 (47.4)2146 (48.9)2203 (49.6)2330 (51.1) 
Marital status     <.0001
 Single353 (13.3)370 (10.1)358 (8.2)388 (8.7)338 (7.4) 
 Married1274 (48.3)1967 (53.5)2543 (58)2659 (59.9)2893 (63.5) 
 Separated37 (1.4)35 (1)43 (1)25 (0.6)29 (0.6) 
 Divorced353 (13.4)464 (12.6)487 (11.1)408 (9.2)377 (8.3) 
 Widowed569 (21.6)774 (21.1)887 (20.2)894 (20.1)853 (18.7) 
 Unknown54 (2.1)64 (1.7)67 (1.5)68 (1.5)69 (1.5) 
Tumor stage     <.0001
 T11113 (42.2)1723 (46.9)2035 (46.4)2104 (47.4)2181 (47.8) 
 T21527 (57.8)1951 (53.1)2350 (53.6)2338 (52.6)2378 (52.2) 
Histology     <.0001
 Adenocarcinoma1017 (38.5)1519 (41.3)1786 (40.7)1959 (44.1)2079 (45.6) 
 Squamous cell carcinoma900 (34.1)1076 (29.3)1340 (30.6)1139 (25.6)1135 (24.9) 
 Large cell carcinoma170 (6.4)247 (6.7)265 (6)265 (6)217 (4.8) 
 BAC240 (9.1)387 (10.5)493 (11.2)588 (13.2)689 (15.1) 
 Undifferentiated313 (11.9)445 (12.1)501 (11.4)491 (11.1)439 (9.6) 
Histologic grade     <.0001
 Well differentiated235 (8.9)351 (9.6)452 (10.3)529 (11.9)595 (13.1) 
 Moderately differentiated780 (29.5)1160 (31.6)1416 (32.3)1429 (32.2)1481 (32.5) 
 Poorly differentiated1027 (38.9)1337 (36.4)1579 (36)1581 (35.6)1586 (34.8) 
 Undifferentiated127 (4.8)193 (5.3)238 (5.4)219 (4.9)225 (4.9) 
 Unknown471 (17.8)633 (17.2)700 (16)684 (15.4)672 (14.7) 
Tumor location     .0848
 RUL943 (35.7)1222 (33.3)1528 (34.6)1507 (33.9)1643 (36.1) 
 RML119 (4.5)178 (4.9)201 (4.6)235 (5.3)222 (4.9) 
 RLL409 (15.5)589 (16)687 (15.7)712 (16)731 (16) 
 LUL715 (27.1)1008 (27.4)1191 (27.2)1192 (26.8)1168 (25.6) 
 LLL318 (12.1)504 (13.7)581 (13.3)637 (14.3)596 (13.1) 
 R/L main bronchus/carina/hilus43 (1.6)42 (1.1)59 (1.4)34 (0.76)52 (1.1) 
 NOS/unknown93 (3.5)131 (3.6)138 (3.2)125 (2.8)147 (3.2) 
Surgery     <.0001
 No663 (25.1)750 (20.4)830 (18.9)676 (15.2)596 (13.1) 
 Local1 (0.04)10 (0.27)16 (0.36)10 (0.23)10 (0.22) 
 Wedge/segmentectomy262 (9.9)391 (10.6)521 (11.9)538 (12.1)506 (11.1) 
 Lobectomy1581 (59.9)2349 (63.9)2813 (64.2)3014 (67.9)3261 (71.5) 
 Pneumonectomy124 (4.7)167 (4.6)197 (4.5)198 (4.5)178 (3.9) 
 NOS/unknown9 (0.35)7 (0.19)8 (0.18)6 (0.14)8 (0.18) 
Radiation treatment     <.0001
 Yes438 (16.6)607 (16.5)643 (147)559 (12.6)532 (11.7) 
 No2202 (83.4)3067 (83.5)3742 (85.3)3883 (87.4)4027 (88.3) 
Chemotherapy     .0001
 Yes196 (7.4)252 (6.9)234 (5.3)251 (5.7)242 (5.3) 
 No2407 (91.2)3384 (92.1)4115 (93.8)4162 (93.7)4282 (93.9) 
 Unknown37 (1.4)38 (1.0)36 (0.8)29 (0.7)35 (0.8) 
Any treatment*     <.0001
 Yes2302 (87.2)3376 (91.9)4042 (92.2)4178 (94.1)4318 (94.7) 
 No321 (12.2)284 (7.7)333 (7.6)258 (5.8)236 (5.2) 
 Unknown17 (0.6)14 (0.4)10 (0.2)6 (0.1)5 (0.1) 

Surgical Treatment

There were significantly fewer surgical resections in SES1 (74.9%) than in SES5 (86.9%; P < .0001). Fewer African-American patients underwent surgical treatment (76.4%) compared with Hispanic (79.7%), non-Chinese Asian (80.2%), Chinese (80.4%), and Caucasian (83.1%) patients. A greater proportion of married patients (85.4%) underwent surgical treatment compared with unmarried patients (78.0%; P < .0001). For patients who underwent surgical resection, lobectomy accounted for 80.4% of all surgical procedures performed. The proportions of lobectomy among patients who underwent surgical treatment were similar across all 5 SES quintiles (SES1, 80%; SES2, 80.3%; SES3, 79.1%; SES4, 80%; and SES5, 82.3%). Similar proportions of lobectomy were reported among Caucasian (80.1%), African-American (79.7%), Hispanic (80.6%), Chinese (83.6%) and non-Chinese Asian (85.6%) patients. Lobectomy also was undergone in similar proportions of married patients (81.4%) and unmarried patients (79.0%) among those who underwent surgical resection.

Reasons for No Cancer-directed Surgery

Reasons for not undergoing surgery were abstracted from CCR codes. In total, 3518 (17.9%) patients did not undergo any surgical treatments. The most common reason was “not part of the planned first course of treatment” (54.9%). The second most common reason was “because of patient risk factors, such as comorbid conditions, advanced age, etc” (21.5%). The third most common reason was “refusal by patient, patient's family, or guardian” (9.9%). Patients who did not undergo surgery were stratified according to those 3 most common reasons into the percentages of total patients within each SES quintile, individual race, and marital status and are listed in Table 2.

Table 2. Patients Who Did Not Receive Surgery Expressed as the Percentage of Total Patients Within Each Socioeconomic Status Quintile, Individual Race, and Marital Status Categorized by 3 Main Reasons for No Surgery at the Primary Site
VariableReasons for no surgery at primary site, %
Not part of the planned first course of treatmentContradictions because of patient risk factors (comorbid conditions, advanced age, etc)Refused by patient, family member, or guardian and is noted in patient's record
  • SES indicates socioeconomic status.

  • *

    Unmarried: (single, separated, divorced, widowed).

SES quintile
 113.14.93.2
 210.44.62.6
 310.54.21.6
 48.73.31.2
 57.92.81
Race
 African American13.23.92.9
 Hispanic11.142
 Caucasian9.33.91.5
 Chinese10.13.54
 Non-Chinese Asian113.23.2
Marital status
 Married83.31.3
 Unmarried*12.14.52.4

Univariate Survival Analyses

Patients who had stage IA disease had significantly better 5-year survival estimates compared with patients who had stage IB disease (53.2% vs 40.2%; P < 0001). Patients with BAC had the best 5-year survival (61.9%) followed by patients with adenocarcinoma (50.3%), patients with large cell carcinoma (42.8%), patients with squamous cell carcinoma (39.2%), and patients who had tumors with undifferentiated histology (33%; P < .0001). For a comparison of OS by histologic grade, patients who had tumors with well differentiated histology had the best 5-year survival (61.0%), followed by patients who had tumors with moderately differentiated (51.5%), poorly differentiated (43.8%), and undifferentiated (43.8%) histology (P < .0001). Patients with tumors located in the upper lobe had better 5-year survival estimates than patients with tumors located in the lower lobe (right upper lobe, 50.1%; left upper lobe, 47.1%; right middle lobe, 42.7%; right lower lobe, 43.4%; and left lower lobe, 43.5%; P < .0001). Patients who underwent surgery had significantly improved 5-year survival estimates compared with patients who did not undergo surgery (54.3% vs 7.9%; P < .0001). Patients who received at least 1 treatment modality (surgery, chemotherapy, or radiation) survived significantly longer than patients who did not receive any treatment (49.2% vs 6.7%; P < .0001).

Women had an improved 5-year survival rate compared with men (50.9% vs 41.8%; P < .0001). The 5-year survival rate and the median OS were 46.4% and 53 months, respectively, for Caucasians; 41.4% and 44 months, respectively, for African-Americans; 44.7% and 48 months, respectively, for Hispanics; 50.2% and 61 months, respectively, for Chinese; and 49.4% and 59 months, respectively, for non-Chinese Asians (P < .0001).

The 5-year survival rate and median OS of patients with stage I NSCLC decreased progressively and significantly from the highest SES quintile to the lowest SES quintile (SES5: 52.3% and 65 months, respectively; SES4: 49% and 58 months, respectively; SES3: 44.6% and 50 months, respectively; SES2: 42.8% and 48 months, respectively; SES1: 38.5% and 39 months, respectively; P < .0001) (Fig. 1). Married patients had significantly improved 5-year survival estimates compared with unmarried patients (single, separated, divorce, widowed; 49.2% vs 42.1%; P < .0001) (Fig. 2).

thumbnail image

Figure 1. Kaplan-Meier survival curves for patients with stage I nonsmall cell lung cancer stratified by socioeconomic status (SES) quintiles.

Download figure to PowerPoint

thumbnail image

Figure 2. Kaplan-Meier survival curves for patients with stage I nonsmall cell lung cancer stratified by marital status.

Download figure to PowerPoint

Multivariate Survival Analysis

Advanced age at diagnosis, male sex, unmarried status, low SES, stage IB (T2 tumor stage) disease, squamous cell carcinoma, poorly differentiated histologic grade, nonupper lobe location, and nonsurgical intervention were factors that were associated independently with increased mortality risk in multivariate analysis (Table 3). Each sequential increase in the SES quintile resulted in a statistically significant decrease in the hazard ratio (HR) of death: (SES2 vs SES1: HR, 0.91; 95% confidence interval [95% CI], 0.85–0.98; SES3 vs SES1: HR, 0.90; 95% CI, 0.84–0.97; SES4 vs SES1: HR, 0.83; 95% CI, 0.77–0.89; SES5 vs SES1: HR, 0.79; 95% CI, 0.72–0.84; Ptrend < .0001).

Table 3. Cox Proportional Hazards Model for Overall Survival in Patients With Stage I Nonsmall Cell Lung Cancer
VariableHR95% CIP
  • HR indicates hazard ratio; 95% CI, 95% confidence interval; SES, socioeconomic status; BAC, bronchioalveolar carcinoma; LLL, left lower lobe; LUL, left upper lobe; RUL, right upper lobe; RML, right middle lobe; RLL, right lower lobe; NOS, not otherwise specified.

  • *

    Analyzed as a continuous variable.

Ethnic origin
 Caucasian1.00  
 African-American1.0110.928–1.101.8077
 Hispanic1.0270.940–1.122.5535
 Chinese0.8220.705–0.959.0129
 Non-Chinese Asian0.8130.731–0.905.0002
 Other0.6710.427–1.054.0836
Age at diagnosis*1.0311.029–1.034<.0001
Sex
 Men1.00  
 Women0.7580.725–0.793<.0001
Marital status
 Married1.00  
 Unmarried1.1751.122–1.229<.0001
 Unknown1.0970.920–1.309.3007
SES quintile
 1 (Lowest)1.00  
 20.9140.850–0.984.0165
 30.8990.837–0.966.0036
 40.8290.771–0.891<.0001
 5 (Highest)0.7780.723–0.838<.0001
Histology
 Adenocarcinoma1.00  
 Squamous cell carcinoma1.0721.020–1.127.0060
 Large cell carcinoma0.9070.823–1.000.0492
 BAC0.8400.772–0.915<.0001
 Undifferentiated0.9670.892–1.049.4206
Tumor location
 LLL1.00  
 LUL0.8790.826–0.936<.0001
 RUL0.8620.811–0.916<.0001
 RML0.9820.883–1.092.7355
 RLL0.9660.900–1.037.3398
Tumor stage
 T21.00  
 T10.7590.726–0.793<.0001
Histologic grade*1.1191.084–1.155<.0001
Surgery
 No1.00  
 Local0.5640.381–0.833.0041
 Segmentectomy0.3720.339–0.408<.0001
 Lobectomy0.2780.257–0.301<.0001
 Pneumonectomy0.3540.314–0.399<.0001
 NOS0.4910.316–0.764.0016
Radiation  .9378
 No1.00  
 Yes1.0030.929–1.083 
Chemotherapy  .0618
 No1.00  
 Yes1.0920.996–1.197 

Separate multivariate survival analyses were conducted to determine the relevant effect of race on the survival of patients with stage I NSCLC (Table 4). African-American patients (HR, 1.19; 95% CI, 1.09–1.29) and Hispanic patients (HR, 1.11; 95% CI, 1.01–1.21) had an increased risk of death when adjustments for surgery, SES, and marital status were omitted from the multivariate analysis. African-American race (HR, 1.12; 95% CI, 1.03–1.21), but not Hispanic race (HR, 1.08; 95% CI, 0.99–1.18), continued to have an increased risk of death when surgery was included but SES and marital status were omitted in the multivariate analysis (Table 4). African-American race no longer was associated significantly with an increased risk of death when either SES (HR, 1.04; 95% CI, 0.95–1.13) or marital status (HR, 1.08; 95% CI, 0.99–1.17) was included in the multivariate analysis.

Table 4. Multivariate Survival Analysis and Analysis for Effect Modification in Patients With Stage I Nonsmall Cell Lung Cancer by Ethnicity Using Cox Proportional Hazards Models*
VariableHR (95% CI)
Model 1Model 2Model 3Model 4Model 5
  • HR indicates hazard ratio; 95% CI, 95% confidence interval; NOS, not otherwise specified; SES, socioeconomic status.

  • *

    Each model included adjustment for age at diagnosis, sex, histology, histologic grade, and tumor lobar location.

  • Unknown marital status was included in the multivariate analysis but is not shown.

  • Unmarried means single, separated, divorced, or widowed.

Ethnic origin
 Caucasian1.001.001.001.001.00
 African-American1.186 (1.092–1.289)1.116 (1.027–1.213)1.077 (0.990–1.172)1.038 (0.953–1.131)1.011 (0.928–1.101)
 Hispanic1.105 (1.013–1.206)1.082 (0.991–1.181)1.081 (0.990–1.179)1.024 (0.937–1.119)1.027 (0.940–1.122)
 Chinese0.845 (0.724–0.985)0.823 (0.706–0.959)0.842 (0.722–0.981)0.803 (0.689–0.937)0.822 (0.705–0.959)
 Non-Chinese Asian0.875 (0.786–0.973)0.814 (0.731–0.906)0.827 (0.743–0.921)0.800 (0.718–0.890)0.813 (0.731–0.905)
 Other0.732 (0.466–1.148)0.692 (0.441–1.086)0.670 (0.426–1.053)0.693 (0.441–1.088)0.671( 0.427–1.054)
Radiation
 No1.001.001.001.001.00
 Yes1.896 (1.781–2.020)1.003 (0.929–1.083)1.004 (0.930–1.084)1.002 (0.928–1.082)1.003 (0.929–1.083)
Chemotherapy
 No1.001.001.001.001.00
 Yes1.215 (1.108–1.333)1.094 (0.998–1.200)1.101 (1.005–1.207)1.085 (0.989–1.190)1.092 (0.996–1.197)
Surgery
 No1.001.001.001.00
 Local0.552 (0.373–0.815)0.560 (0.379–0.828)0.555 (0.376–0.821)0.563 (0.381–0.832)
 Wedge/segmentectomy0.363 (0.331–0.399)0.365 (0.333–0.401)0.371 (0.338–0.407)0.372 (0.339–0.408)
 Lobectomy0.270 (0.249–0.292)0.273 (0.252–0.295)0.275 (0.255–0.298)0.278 (0.257–0.301)
 Pneumonectomy0.346 (0.307–0.390)0.349 (0.310–0.393)0.352 (0.312–0.397)0.354 (0.314–0.399)
 NOS0.480 (0.309–0.747)0.488 (0.313–0.759)0.482 (0.311–0.752)0.491 (0.315–0.763)
Marital status
 Married1.001.00
 Unmarried1.192 (1.139–1.247)1.175 (1.122–1.229)
SES quintile
 1 (Lowest)1.001.00
 20.912 (0.847–0.981)0.914 (0.850–0.984)
 30.888 (0.826–0.954)0.899 (0.837–0.966)
 40.818 (0.761–0.880)0.829 (0.771–0.891)
 5 (Highest)0.763 (0.709–0.821)0.778 (0.723–0.838)

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

In this study, we determined that low SES was associated independently with a poor prognosis for patients with stage I NSCLC in our Cox proportional hazards analysis in addition to male sex, unmarried status, stage IB disease, squamous cell histology, poorly differentiated tumor grade, and fewer surgical interventions. Significantly more African-American and Hispanic patients were in the lower SES quintiles compared with Caucasian patients. Blackstock et al.,21 in a retrospective analysis of patients with advanced NSCLC who were enrolled in Cancer and Leukemia Group B trials, reported that African-American patients were more likely to be unmarried, to be disabled or unemployed, to be receiving Medicaid, and to have a lower median income, and they had an unadjusted 1-year survival rate that was lower than non-African-American patients. However, the effect of race on survival disappeared when the analysis was adjusted for poor performance status and weight loss ≥5%. The racial distribution of patients with NSCLC among the 5 SES quintiles likely reflects the California population in general rather than any specific attributions to NSCLC in particular.

Our observation that low SES is an independent poor prognostic factor in patients with stage I NSCLC is consistent with prior reports demonstrating that low SES played an important role in the survival outcome of African-American2, 22 and Hispanic6 patients with stage I NSCLC independent of the effects of curative surgical resection. Greenwald et al.22 studied 5189 patients with stage I NSCLC and observed that African-American race no longer was a significant adverse prognostic factor compared with Caucasian race in determining OS outcome after factoring in median income and surgery in the Cox proportional hazards model. Bach et al.2 used the Medicare database and observed that low median income (but not race) was a poor prognostic factor in patients with stage I and II NSCLC who underwent surgical resection in an adjusted analysis.

Hispanic race was an independent poor prognostic factor in the current study when the analysis was not adjusted for surgical resection, SES, and marital status. Our findings were similar to those of Wisnivesky et al.,6 who reported that Hispanic patients with stage I NSCLC had lower median income, were older age at diagnosis, were less likely to undergo surgical resection, presented with more stage IB disease, and had a poorer survival outcome compared with Caucasian patients. However, after adjusting for surgery and stage, there was no survival difference between Hispanic and Caucasian patients in that study, whereas per capita income remained an independent prognostic factor.

We analyzed the 3 most commonly stated reasons why patients did not receive surgery and stratified the results by race, SES, and marital status. African-American patients with lung cancer reportedly are less likely to be offered surgical treatment2, 3, 9, 22 and are less likely to accept treatment,9, 10 which may explain the disparity in survival observed between African-American and Caucasian patients with lung cancer. All 3 reasons for “no surgery at primary site” were associated more closely with SES and marital status than with race. Patients with low SES or unmarried patients were less likely to undergo surgery because of comorbidities, more likely to refuse treatment, and more likely not to be offered surgery as the planned course of treatment (Table 2). Comorbidity was identified previously as an independent poor prognostic factor in patients with stage I NSCLC who underwent surgical resection.2, 23–25 Battafarano et al.23 reported that, in patients with stage I NSCLC who underwent surgical resection, the relative risk (RR) of death as a function of comorbidity was 2 times higher for patients with moderate or severe comorbidity and 40% higher for patients with mild comorbidity compared with patients who had no comorbidity. Indeed, in the current study, we observed that a progressively greater proportion of patients in the lower SES quintiles did not undergo surgery because of comorbidities (Table 2). Thus, the effect of SES on the survival of patients with early-stage NSCLC may be caused in part by increased comorbidities in patients with low SES precluding curative surgical treatment for NSCLC. It is important to point out that the proportions of lobectomies performed were similar across race, SES quintiles, and marital status among patients who underwent surgical treatment. Other factors that are associated with low SES or unmarried status and are not identified easily in the CCR database likely contribute greatly to the poor survival of patients with low SES or unmarried status.

We also identified unmarried status is an independent poor prognostic factor in the survival of patients with stage I NSCLC. Significantly more unmarried patients were in the lower SES quintile (Table 1), and unmarried patients were less likely to be offered surgery, less likely to undergo surgery because of comorbidities, and less likely to accept surgery. Similarly, Goodwin et al.26 and Greenberg et al.27 also reported that unmarried patients were less likely to undergo treatment. Several groups26, 28 reported that unmarried patients with lung cancer had a higher risk of death, whereas other studies did not observe this difference.27, 29 Improving social and psychological support may be beneficial to the survival of patients with very-early-stage NSCLC.

A limitation of the CCR is that smoking history is not captured and, thus, the prognostic significance of smoking cannot be incorporated into the multivariate analysis. Smoking generally is associated with a poorer outcome even in stage I NSCLC.30, 31 It has been demonstrated that the incidence of smoking increases with lower SES,32 and smoking also predicts adverse comorbidity.33 Moreover, performance status and medical comorbidities were not captured by the current analysis of CCR data. Other limitations of this report are the retrospective nature of the analysis, lack of a central pathology review, and lack of a standardized staging protocol and treatment pathway.

Finally, although African-American and Hispanic race were not independent poor prognostic factors in stage I NSCLC, members of these ethnic groups made up a disproportionately large percentage of patients in the lower SES quintiles. Moreover, slightly more African-American patients were not offered surgical treatment (Table 2). Therefore, measures directed at increasing curative surgical resection rates, such as better patient education, improved physician-patient interaction and trust34 increased physician willingness to offer treatment, and optimization of the general health of the patient, likely would alleviate some of the survival detriment associated with lower SES. Efforts to provide social and psychological support for unmarried patients to undergo surgery probably will improve survival in early-stage lung cancer. Because lung cancer is the leading cause of death from cancer in the United States, increasing the proportion of patients with early-stage disease that undergo surgical resection will improve lung cancer survival in this country. This is especially important if screening for lung cancer is to be implemented widely in the future, leading to the detection of more early-stage disease that is amenable to surgery with subsequent excellent survivorship.35

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    Kamangar F,Dores GM,Anderson WF. Patterns of cancer incidence, mortality, and prevalence across 5 continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol. 2006; 24: 21372150.
  • 2
    Bach PB,Cramer LD,Warren JL,Begg CB. Racial differences in the treatment of early stage lung cancer. N Engl J Med. 1999; 341: 11981205.
  • 3
    Clegg LX,Li FP,Hankey BF,Chu E,Edwards BK. Cancer survival among US whites and minorities: a SEER (Surveillance, Epidemiology and End Results) Program population-based study. Arch Intern Med. 2002; 162: 19851993.
  • 4
    Shavers VL,Brown ML. Racial and ethnic disparities in the receipt of cancer treatment. J Natl Cancer Inst. 2002; 94: 334357.
  • 5
    Bach PB,Schrag D,Brawley OW,Galaznik A,Yakren S,Begg CB. Survival of blacks and whites after a cancer diagnosis. JAMA. 2002; 287: 21062113.
  • 6
    Wisnivesky JP,McGinn T,Henschke C,Hebert P,Iannuzzi MC,Halm EA. Ethnic disparities in the treatment of stage I non-small cell lung cancer. Am J Respir Crit Care Med. 2005; 11581163.
  • 7
    Schwatrz KL,Crossley-May H,Vigneau FD,Brown K,Banerjee M. Race, socioeconomic status and stage at diagnosis for 5 common malignancies. Cancer Causes Control. 2003; 14: 761766.
  • 8
    Potosky AL,Saxman S,Wallace RB,Lynch CF. Population variations in the initial treatment of non-small-cell lung cancer. J Clin Oncol. 2004; 22: 32613268.
  • 9
    Lathan CS,Neville BA,Earle CC. The effect of race on invasive staging and surgery in non-small-cell lung cancer. J Clin Oncol. 2006; 24: 413418.
  • 10
    McCann J,Artinian V,Duhaime L,Lewis JWJr,Kvale PA,DiGiovine B. Evaluation of the causes for racial disparity in surgical treatment of early stage lung cancer. Chest. 2005; 128: 34403446.
  • 11
    Margolis ML,Christie JD,Silvestri GA,Kaiser L,Santiago S,Hansen-Flaschen J. Racial differences pertaining to a belief about lung cancer surgery: results of a multicenter survey. Ann Intern Med. 2003; 139: 558563.
  • 12
    Woods LM,Rachet B,Coleman MP. Origins of socio-economic inequalities in cancer survival: a review. Ann Oncol. 2006; 17: 519.
  • 13
    Ou SI,Zell JA,Ziogas A,Anton-Culver H. Epidemiology of nasopharyngeal carcinoma (NPC) in California: survival advantage of Chinese WHO type 1 NPC patients resulted in Asian ethnicity as an independent and favorable prognostic factor of survival for NPC patients as a whole and for patients with the WHO histologic type 1. J Clin Oncol. 2006; 24(18S): 283S. Abstract 5514.
  • 14
    Zell JA,Cinar P,Mobasher M,Ziogas A,Meyskens FL,Anton-Culver H. The effects of race, socioeconomic status, and treatment on survival for invasive cutaneous melanoma cases. J Clin Oncol. 2008; 26: 6675.
  • 15
    Zell JA,Rhee JM,Ziogas A,Lipkin SM,Anton-Culver H. Race, socioeconomic status, treatment, and survival time among pancreatic cancer cases in California. Cancer Epidemiol Biomarkers Prev. 2007; 16: 546552.
  • 16
    Parikh-Patel A,Bates JH,Campleman S. Colorectal cancer stage at diagnosis by socioeconomic and urban/rural status in California, 1988–2000. Cancer. 2006; 107( 5 suppl): 11891195.
  • 17
    California Cancer Registry. How complete are California Cancer Registry data? Available from: http://www.ccrcal.org/questions.html, 2006 Accessed on May 5, 2006.
  • 18
    Zell JA,Ou S-HI,Ziogas A,Anton-Culver H. Epidemiology of bronchioloalveolar carcinoma: improvement in survival after release of the 1999 WHO classification of lung tumors. J Clin Oncol. 2005; 23: 83968405.
  • 19
    Zell JA,Ou S-HI,Ziogas A,Anton-Culver H. Long-term survival difference for bronchiolo-alveolar carcinoma patients with ipsilateral intrapulmonary metastasis at diagnosis. Ann Oncol. 2006; 17: 12551262.
  • 20
    Yost K,Perkins C,Cohen R,Morris C,Wright W. Socioeconomic status and breast cancer incidence in California for different race/ethnic groups. Cancer Causes Control. 2001; 12: 703711.
  • 21
    Blackstock AW,Herndon JE,Paskett ED, et al. Outcomes among African-American/non-African-American patients with advanced non-small-cell lung carcinoma: report from the Cancer and Leukemia Group B. J Natl Cancer Inst. 2002; 94: 284290.
  • 22
    Greenwald HP,Polissar NL,Borgatta EF,McCorkle R,Goodman G. Social factors, treatment, and survival in early stage non-small cell lung cancer. Am J Public Health. 1998; 8: 16811684.
  • 23
    Battafarano RJ,Piccirillo JF,Meyers BF, et al. Impact of comorbidity on survival after surgical resection in patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg. 2002; 123: 280287.
  • 24
    Moro-Sibilot D,Aubert A,Diab S, et al. Comorbidities and Charlson score in resected stage I nonsmall cell lung cancer. Eur Respir J. 2005; 26: 480486.
  • 25
    Firat S,Bousamra M,Gore E,Byhardt RW. Comorbidity and KPS are independent prognostic factors in stage I non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2002; 52: 10471057.
  • 26
    Goodwin JS,Hunt WC,Key CR,Samet JM. The effect of marital status on stage, treatment, and survival of cancer patients. JAMA. 1987; 258: 31253130.
  • 27
    Greenberg ER,Chute CG,Stukel T, et al. Social and economic factors in the choice of lung cancer treatment. A population-based study in 2 rural states. N Engl J Med. 1988; 318: 612617.
  • 28
    Ganz PA,Lee JJ,Siau J. Quality of life assessment. An independent prognostic variable for survival in lung cancer. Cancer. 1991; 67: 31313135.
  • 29
    Saito-Nakaya K,Nakaya N,Fujimori M, et al. Marital status, social support and survival after curative resection in non-small-cell lung cancer. Cancer Sci. 2006; 97: 206213.
  • 30
    Fujisawa T,Lizasa T,Saitoh Y, et al. Smoking before surgery predicts poor long-term survival in patients with stage I non-small cell lung carcinomas. J Clin Oncol. 1999; 17: 20862091.
  • 31
    Wu YC,Lin CF,Hsu WH,Huang BS,Huang MH,Wang LS. Long term results of pathological stage I non-small cell lung cancer: validation of using the number of totally removed lymph nodes as a staging control. Eur J Cardiothorac Surg. 2003; 24: 9941001.
  • 32
    Barbeau EM,Krieger N,Soobader MJ. Working class matters: socioeconomic disadvantage, race/ethnicity, gender, and smoking in NHIS 2000. Am J Public Health. 2004; 94: 269278.
  • 33
    Tammemagi CM,Neslund-Dudas C,Simoff M,Kvale P. In lung cancer patients, age, race-ethnicity, gender and smoking predict adverse comorbidity, which in turns predicts treatment and survival. J Clin Epidemiol. 2004; 57: 597609.
  • 34
    Gordon HS,Street RLJr,Sharf BF,Kelly A,Souchek J. Racial differences in trust and lung cancer patients' perceptions of physician communication. J Clin Oncol. 2006; 24: 904909.
  • 35
    The International Early Lung Cancer Action Program Investigators. Survival of patients with stage I lung cancer detected on CT scanning. N Engl J Med. 2006; 355: 17631771.