SEARCH

SEARCH BY CITATION

Keywords:

  • cervical cancer;
  • African American;
  • race;
  • disparities;
  • SEER

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURE
  8. REFERENCES

BACKGROUND

The purpose of this study is to examine changes over time in survival for African American (AA) and white women diagnosed with cervical cancer (CC).

METHODS

Surveillance, Epidemiology, and End Results (SEER) Program data from 1985 to 2009 were used for this analysis. Racial differences in survival were evaluated between African American (AA) and white women. Kaplan-Meier and Cox proportional hazards survival methods were used to assess differences in survival by race at 5-year intervals.

RESULTS

The study sample included 23,368 women, including 3886 (16.6%) who were AA and 19,482 (83.4%) who were white. AA women were older (51.4 versus 48.9 years; P < .001) and had a higher rate of regional (38.3% versus 31.8%; P < .001) and distant metastasis (10.7% versus 8.7%; P < .001). AA less frequently received cancer-directed surgery (32.4% versus 46%; P < .001), and more frequently radiotherapy (36.3% versus 26.4%; P < .001). Overall, AA women had a hazard ratio (HR) of 1.41 (95% confidence interval = 1.32-1.51) of cervical cancer (CC) mortality compared with whites. Adjusting for SEER registry, marital status, stage, age, treatment, grade, and histology, AA women had an HR of 1.13 (95% confidence interval = 1.05-1.22) of CC-related mortality. After adjusting for the same variables, there was a significant difference in CC-specific mortality between 1985 to 1989 and 1990 to 1994, but not after 1995.

CONCLUSIONS

After adjusting for race, SEER registry, marital status, stage, age, treatment, grade, and histology, there was a significant difference in CC-specific mortality between 1985 to 1989 and 1990 to 1994, but not after 1995. Cancer 2013;119:3644–3652. © 2013 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURE
  8. REFERENCES

In 2013, there will be an estimated 12,170 new cases of cervical cancer (CC) diagnosed and approximately 4030 cancer-related deaths due to the disease,[1] making it the third most common cancer diagnosis and cause of death among gynecological cancers in the United States, and the second leading cause of death in women aged 20 to 39 years.[1, 2] Effective cervical cytology screening has resulted in a steady decline in the incidence and mortality of CC in the United States.[3] Although this decline has occurred across all racial and ethnic groups, significant disparities in these rates continue to exist.[3-5] Several factors may account for the observed disparity in CC incidence and mortality among African American (AA) compared to white women. These factors include differences in screening and follow-up rates and practices, treatment, behavioral risk factors, and potentially underlying difference in biological characteristics of the tumor.

Studies have examined long-term trends and disparities in US CC mortality according to race. However, the extent to which disparities in CC mortality rates by race have changed over time has not been studied. Observing changes in mortality over time, rather than statically at a cross-sectional moment, may allow us to track progress toward reducing social and geographic disparities in CC mortality. Temporal analysis may also reveal important insights into the differential impact of cancer prevention and CC screening programs among different racial/ethnic minorities. The objective of this study was to examine if trends in death rates for CC between AA and white women with CC have changed over time, while accounting for both biological and treatment related prognostic factors.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURE
  8. REFERENCES

Data from the National Cancer Institute's Surveillance, Epidemiology and End Results (SEER) registry from CC cases diagnosed between 1985 and 2009 were the source for this analysis. Because all information from the SEER database is deidentified, informed consent by the study participants and approval of an ethics committee were unnecessary to perform the analyses in this study. For the current analysis the SEER 9 registries were used, which include Atlanta, Connecticut, Detroit, Hawaii, Iowa, New Mexico, San Francisco-Oakland, Seattle-Puget Sound, and Utah.[6] Eligible women were 18 or older with newly diagnosed, histologically confirmed CC. Women who were not AA or white, women who were diagnosed at autopsy, and women with a previous diagnosis of cancer were excluded.

Variables were coded according to SEER Program criteria. The exposure variable, race, was designated as white and AA. Marital status was categorized as: married, not married and unknown. Geographic location was divided into the 9 SEER locations. Tumor grade was classified as well differentiated, moderately differentiated, poorly differentiated, undifferentiated, or unknown. We used SEER Summary Staging (localized, regional, distant, and unstaged) to categorize the extent of the disease. Histology was categorized as squamous cell carcinoma, adenocarcinoma, adenosquamous or other histologies. The primary treatment modality (radiation, surgery or no treatment) was recorded. The outcome variables included vital status and the time-to-event from the date of diagnosis until death, censoring, or last follow-up, as verified by the SEER program vital status determination. Among deceased persons listed in the SEER Registry, death may have occurred from CC or any other cause of death.

Statistical Analysis

Racial differences in the distribution of demographic, clinical, and treatment characteristics were compared using chi-square tests. Student t tests and analysis of variance (ANOVA) were used to assess the significance of differences in the mean values of continuous variables. We used the Kaplan-Meier method to estimate survival curves in order to compare observed survival between AA and white women for a given period of diagnosis. Survival curves were constructed to show all-cause mortality within the first 5 years of diagnosis for each race within each cohort, although the hazard ratios (HR) and resulting P values were calculated using all available data through last date of follow-up at the end of 2009, not only the first 5 years after diagnosis.[7] Unadjusted all-cause mortality and disease-specific mortality analyses were performed to assess the survival of AA compared with whites within diagnosis years 1985-1989, 1990-1994, 1995-1999, 2000-2004, and 2005-2009, and then for all cohorts combined. Cox proportional hazards models were used to calculate adjusted racial group HRs and their 95% confidence intervals (CI) to assess the importance of race as an independent predictor of survival after adjusting for the following prognostic factors: SEER site, age, marital status, stage, treatment, histology, grade. Five-year diagnosis cohort was also included as a categorical variable when the entire population was analyzed.

All statistical tests were 2-sided and differences were considered statistically significant at P < .05. We used R software, version 2.15.2, and the package survival version 2.36-14 for all statistical analyses.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURE
  8. REFERENCES

The SEER data set included 129,070 women who were diagnosed with reproductive cancer within the SEER 9 registries from 1985 to 2009, of whom 28,346 were diagnosed with invasive CC. A total of 4978 patients were excluded from the final analysis: 3175 cases because of race (not white or AA), 22 cases because of age, 1397 cases who had a prior malignancy, 372 cases that were not microscopically confirmed, and 12 cases were excluded because diagnosis was made at autopsy. The final study group consisted of 23,368 women, including 19,482 (83.4%) whites and 3886 (16.6%) AA.

Table 1 summarizes the demographic and clinical characteristics of the study population. The mean age at diagnosis was 51.4 years for AA and 48.9 years for whites (P < .001). AA were less frequently married at the time of diagnosis (25.8% versus 48.8%; P < .001). AA had a higher rate of regional (38.2% versus 31.9; P < .001) and distant metastasis (10.7% versus 8.7; P < .001); and a lower rate of well-differentiated cancers (4.8% versus 8.6%; P < .001). AA had a higher rate of squamous cell carcinoma (77.1% versus 69.0%; P < .001), and a lower rate of adenocarcinoma (9.3% versus 18.2%; P < .001). AA had surgery less frequently (53.3% versus 66.3%; P < .001), and radiotherapy more frequently (57.2% versus 46.7%; P < .001).

Table 1. Demographic and Clinical Characteristics of the Study Population: SEER Program, 1985-2009
CharacteristicWhiteAfrican AmericanP
  1. Abbreviation: SEER, Surveillance, Epidemiology, and End Results.

N (%)19,482 (83.37%)3886 (16.63%) 
Age at diagnosis, y   
Mean (standard deviation)48.92 (16.06)51.41 (16.22)<.001
Median (interquartile range)46 (36-60)49 (39-63) 
Marital status at diagnosis, N (%)  <.001
Unmarried8836 (45.4%)2611 (67.2%) 
Married9512 (48.8%)1001 (25.8%) 
Unknown1134 (5.8%)274 (7.1%) 
SEER registry, N (%)  <.001
Metropolitan Atlanta1615 (8.3%)1113 (28.6%) 
Connecticut2886 (14.8%)406 (10.4%) 
Metropolitan Detroit3088 (15.9%)1617 (41.6%) 
Hawaii371 (1.9%)18 (0.5%) 
Iowa2832 (14.5%)64 (1.6%) 
New Mexico1675 (8.6%)30 (0.8%) 
San-Francisco–Oakland2547 (13.1%)500 (12.9%) 
Seattle (Puget Sound)3065 (15.7%)124 (3.2%) 
Utah1403 (7.2%)14 (0.4%) 
Stage, N (%)  <.001
Localized10,556 (54.2%)1712 (44.1%) 
Regional6210 (31.9%)1488 (38.3%) 
Distant1693 (8.7%)414 (10.7%) 
Unstaged1023 (5.3%)272 (7.0%) 
Grade, N (%)  <.001
I1681 (8.6%)186 (4.8%) 
II4903 (25.2%)901 (23.2%) 
III4806 (24.7%)1121 (28.8%) 
IV470 (2.4%)91 (2.3%) 
Unknown7622 (39.1%)1587 (40.8%) 
Histology, N (%)  <.001
Adenocarcinoma3555 (18.2%)363 (9.3%) 
Adenosquamous793 (4.1%)149 (3.8%) 
Squamous13,443 (69.0%)2995 (77.1%) 
Other1691 (8.7%)379 (9.8%) 
Treatment, N (%)  <.001
None1398 (7.2%)402 (10.3%) 
Radiation5152 (26.4%)1412 (36.3%) 
Surgery8969 (46.0%)1260 (32.4%) 
Surgery and Radiation3963 (20.3%)812 (20.9%) 

In the crude models for both all-cause mortality and cancer-specific mortality, AA had a significantly increased overall and disease-specific hazard of death compared with white women (Table 2). The overall HR for AA women was 1.53 (95% CI = 1.46-1.61), and the disease-specific HR was 1.41 (95% CI = 1.32-1.51). Figure 1 shows Kaplan-Meier survival curves for disease-specific mortality for whites and AA women diagnosed with CC by 5-year diagnosis cohort and overall. Table 2 presents the risk of mortality for AA compared with whites, stratified by 5-year diagnosis cohorts. AA women had a higher HR of all cause mortality and CC related mortality for all the 5-year diagnosis cohorts. Over the entire study period, after adjusting for race, SEER registry, marital status, stage, age, treatment (surgery versus radiotherapy versus surgery and radiotherapy versus none), grade, histology, and 5-year diagnosis cohort, AA race remained significantly associated with an increased overall hazard of death (HR = 1.17; 95% CI = 1.11-1.23), and CC-related mortality compared to whites (HR = 1.13, 95% CI = 1.05-1.22). After adjusting for the same variables there was a significant difference in CC-specific mortality between AA and whites between 1985-1989 and 1990-1994, but not after 1995.

Table 2. Adjusted and Unadjusted Cox Proportional Hazards Models for All-Cause and Cervical Cancer–Specific Mortality, Comparing Whites to African Americans, by 5-Year Diagnosis Cohort, and Overall. SEER Program, 1985-2009.
Year of Diagnosis All-Cause Mortality, HRCervical Cancer Mortality, HR
NUnadjustedAdjustedUnadjustedAdjusted
  1. Abbreviations: HR, hazard ratio; SEER, Surveillance, Epidemiology, and End Results.

  2. Adjusted model includes race, SEER registry, marital status, stage, age as continuous variable, treatment, grade, and histology. For all years, 5-year diagnosis cohort was also included in the model. Unadjusted model only includes race.

1985-198950271.66 (1.52-1.81)1.12 (1.02-1.24)1.65 (1.45-1.88)1.17 (1.01-1.36)
1990-199451561.53 (1.38-1.69)1.32 (1.18-1.48)1.41 (1.22-1.62)1.23 (1.05-1.44)
1995-199949021.48 (1.33-1.65)1.16 (1.03-1.31)1.28 (1.11-1.49)1.06 (0.89-1.25)
2000-200442901.39 (1.23-1.57)1.06 (0.93-1.22)1.28 (1.09-1.50)1.01 (0.85-1.21)
2005-200939931.55 (1.32-1.82)1.13 (0.94-1.36)1.41 (1.16-1.71)1.12 (0.90-1.40)
All Years23,3681.53 (1.46-1.61)1.17 (1.11-1.23)1.41 (1.32-1.51)1.13 (1.05-1.22)
image

Figure 1. Kaplan-Meier survival curves show disease-specific mortality for whites and African American (AA) women diagnosed with cervical cancer, by 5-year diagnosis cohort and overall. HR indicates hazard ratio.

Download figure to PowerPoint

Survival analysis using the Cox proportional hazards model identified an independent association of race, time period of diagnosis, older age, SEER registry site, advanced stage, grade, and type of treatment, with higher mortality (Table 3). The strongest quantitative predictor of death was stage at the time of diagnosis, with more advanced stages having higher mortality rates. Surgical therapy, radiation therapy, and combination therapy were all protective compared with patients who did not receive therapy. Marriage was protective against all-cause mortality, but not CC-specific mortality. Histology was not a significant factor. The hazard of disease-specific mortality decreased over time in the periods of 2000-2004 and 2005-2009 compared to 1985-1989. Table 4 summarizes the effect of the same factors on mortality in patients with CC by period of diagnosis.

Table 3. Effect of Various Factors on Mortality in Patients With Cervical Cancer, SEER Program, 1985-2009.
CharacteristicAll-Cause Mortality, HRCervical Cancer Mortality, HR
  1. Abbreviations: HR, hazard ratio; Ref, reference; SEER, Surveillance, Epidemiology, and End Results.

Year of diagnosis  
1985-1989Ref.Ref.
1990-19941.07 (1.01-1.13)1.08 (1.00-1.17)
1995-19991.04 (0.98-1.11)1.02 (0.94-1.11)
2000-20040.95 (0.89-1.02)0.89 (0.82-0.97)
2005-20090.93 (0.86-1.01)0.88 (0.79-0.97)
Race  
WhiteRef.Ref.
African American1.17 (1.11-1.23)1.13 (1.05-1.22)
Age at diagnosis  
1-year increase1.03 (1.03-1.03)1.01 (1.00-1.01)
Marital status at diagnosis  
UnmarriedRef.Ref.
Married0.86 (0.82-0.90)0.94 (0.89-1.00)
Unknown0.82 (0.74-0.90)0.77 (0.67-0.89)
SEER registry  
Metropolitan AtlantaRef.Ref.
Connecticut1.02 (0.94-1.10)0.98 (0.88-1.10)
Metropolitan Detroit1.09 (1.02-1.17)1.04 (0.94-1.15)
Hawaii1.40 (1.18-1.66)1.46 (1.16-1.84)
Iowa0.99 (0.91-1.08)1.04 (0.92-1.16)
New Mexico1.09 (0.99-1.20)1.05 (0.92-1.20)
San-Francisco–Oakland1.03 (0.94-1.11)1.06 (0.94-1.18)
Seattle (Puget Sound)0.98 (0.90-1.07)1.04 (0.92-1.17)
Utah1.11 (1.00-1.24)1.08 (0.93-1.26)
Stage  
LocalizedRef.Ref.
Regional1.96 (1.85-2.08)3.21 (2.93-3.52)
Distant6.26 (5.84-6.71)10.79 (9.75-11.95)
Unstaged1.35 (1.23-1.49)1.67 (1.43-1.94)
Grade  
IRef.Ref.
II1.17 (1.06-1.29)1.36 (1.17-1.58)
III1.37 (1.24-1.50)1.69 (1.46-1.96)
IV1.58 (1.37-1.83)1.73 (1.41-2.12)
Unknown1.03 (0.93-1.13)1.08 (0.93-1.25)
Histology  
AdenocarcinomaRef.Ref.
Adenosquamous1.04 (0.93-1.16)1.13 (0.98-1.30)
Squamous0.97 (0.91-1.03)0.95 (0.87-1.03)
Other1.03 (0.94-1.12)0.98 (0.87-1.11)
Treatment  
NoneRef.Ref.
Surgery0.23 (0.21-0.25)0.14 (0.12-0.16)
Radiation0.64 (0.59-0.69)0.61 (0.55-0.67)
Surgery and radiation0.40 (0.36-0.43)0.33 (0.29-0.36)
Table 4. Effect of Various Factors on Mortality in Patients With Cervical Cancer by Period of Diagnosis. SEER Program, 1985-2009.
Characteristic1985-19891990-19941995-19992000-20042005-2009
  1. Abbreviations: Ref, reference; SEER: Surveillance, Epidemiology, and End Results.

Race     
WhiteRef.Ref.Ref.Ref.Ref.
African American1.16 (1.00-1.35)1.24 (1.06-1.45)1.06 (0.89-1.25)1.01 (0.85-1.21)1.15 (0.92-1.43)
Age at diagnosis     
1-year increase1.00 (1.00-1.01)1.01 (1.00-1.01)1.01 (1.00-1.01)1.01 (1.01-1.01)1.01 (1.01-1.02)
Marital status at diagnosis     
UnmarriedRef.Ref.   
Married0.94 (0.83-1.06)1.04 (0.92-1.17)0.90 (0.79-1.02)0.87 (0.75-0.99)0.86 (0.72-1.02)
Unknown0.85 (0.66-1.10)0.69 (0.51-0.93)0.72 (0.51-1.02)0.89 (0.64-1.24)0.61 (0.40-0.92)
SEER registry     
Metropolitan AtlantaRef.Ref.Ref.Ref.Ref.
Connecticut0.95 (0.76-1.20)1.23 (0.97-1.56)0.90 (0.71-1.14)0.95 (0.74-1.21)0.92 (0.66-1.29)
Metropolitan Detroit1.00 (0.81-1.23)1.38 (1.12-1.72)0.91 (0.73-1.13)0.98 (0.78-1.22)0.98 (0.74-1.30)
Hawaii1.27 (0.82-1.96)2.04 (1.34-3.12)1.51 (0.87-2.64)1.39 (0.82-2.35)1.14 (0.46-2.85)
Iowa0.98 (0.77-1.25)1.27 (1.00-1.62)1.04 (0.81-1.33)0.97 (0.75-1.26)0.92 (0.65-1.30)
New Mexico0.97 (0.72-1.29)1.22 (0.92-1.62)0.94 (0.70-1.26)0.84 (0.62-1.14)1.53 (1.08-2.18)
San Francisco–Oakland1.13 (0.90-1.42)1.33 (1.05-1.67)0.97 (0.76-1.24)1.00 (0.77-1.30)0.76 (0.53-1.09)
Seattle (Puget Sound)1.02 (0.80-1.30)1.24 (0.97-1.59)1.09 (0.85-1.40)0.94 (0.72-1.22)0.95 (0.69-1.33)
Utah1.34 (0.97-1.87)1.26 (0.92-1.73)1.00 (0.73-1.37)0.79 (0.55-1.12)0.94 (0.61-1.43)
Stage     
LocalizedRef.Ref.Ref.Ref.Ref.
Regional3.24 (2.72-3.87)3.10 (2.59-3.71)3.50 (2.87-4.27)3.09 (2.46-3.89)2.72 (1.99-3.72)
Distant11.06 (9.07-13.49)9.48 (7.70-11.68)11.59 (9.23-14.54)10.13 (7.94-12.93)10.99 (7.98-15.11)
Unstaged1.75 (1.33-2.31)1.52 (1.14-2.03)2.22 (1.62-3.05)1.61 (1.04-2.51)2.22 (1.30-3.78)
Grade     
IRef.Ref.Ref.Ref.Ref.
II1.32 (0.97-1.79)1.01 (0.76-1.33)2.11 (1.47-3.04)1.42 (1.01-1.99)1.34 (0.86-2.09)
III1.58 (1.17-2.14)1.42 (1.08-1.86)2.28 (1.59-3.28)1.69 (1.21-2.38)1.93 (1.24-2.98)
IV2.19 (1.43-3.37)1.17 (0.79-1.74)2.80 (1.77-4.43)2.15 (1.37-3.38)1.22 (0.60-2.47)
Unknown1.08 (0.80-1.46)0.80 (0.61-1.06)1.51 (1.04-2.18)1.15 (0.81-1.63)1.10 (0.70-1.72)
Histology     
AdenocarcinomaRef.Ref.Ref.Ref.Ref.
Adenosquamous1.18 (0.87-1.61)0.98 (0.73-1.33)1.39 (1.02-1.89)0.92 (0.66-1.29)1.10 (0.74-1.64)
Squamous0.94 (0.79-1.13)0.94 (0.79-1.12)0.96 (0.79-1.17)0.89 (0.74-1.07)0.90 (0.72-1.14)
Other0.72 (0.54-0.96)1.03 (0.79-1.34)0.96 (0.73-1.26)1.03 (0.79-1.33)0.92 (0.66-1.27)
Treatment     
NoneRef.Ref.Ref.Ref.Ref.
Surgery0.66 (0.54-0.81)0.63 (0.51-0.78)0.93 (0.73-1.20)0.52 (0.42-0.65)0.46 (0.37-0.58)
Radiation0.15 (0.11-0.20)0.14 (0.11-0.19)0.20 (0.15-0.27)0.11 (0.08-0.16)0.12 (0.08-0.18)
Surgery and Radiation0.42 (0.33-0.53)0.35 (0.27-0.44)0.47 (0.36-0.61)0.23 (0.18-0.29)0.25 (0.19-0.34)

Changes in disease characteristics and patterns of care over time were analyzed (Table 5). Among white women, there was a shift toward a more advanced stage at diagnosis (8.2% in 1985-1989 versus 11.3% in 2005-2009; P < .001), and lower rate of localized disease (53.3% in 1985-1989 versus 50.6% in 2005-2009; P = .01). Among AA, there was not a significant difference in the rate of advanced stage at diagnosis (13.7% in 1985-1989 versus 12.8% in 2005-2009; P = .6). However, there was a shift toward a more localized disease at diagnosis (38.0% in 1985-1989 versus 44.3% in 2005-2009; P = .01). Squamous cell carcinoma was the most common histologic variant, although the rate of cervical adenocarcinoma increased more in whites (14.9% in 1985-1999 versus 23.5% in 2005-2009; P < .001) compared to AA (7.6% in 1985-1999 versus 11.0% in 2005-2009; P = .029). In whites, the rates of surgery remained relatively constant over time (60.6% in 1985-1989 versus 62.1% in 2005-2009; P = .195), in contrast to AA, were the rate of surgery appeared to increased over time (45.8% in 1985-1989 versus 49.2% in 2005-2009; P = .214), although not statistically significant.

Table 5. Patients With Cervical Cancer Stratified by Race and Date of Diagnosis, SEER Program, 1985-2009.
 1985-1989 1990-1994 1995-1999 2000-2004 2005-2009 
CharacteristicN = 5027 (21.5%) N = 5156 (22%) N = 4902 (20.9%) N = 4290 (18.3%) N = 3993 (17%) 
RaceWhiteAAPWhiteAAPWhiteAAPWhiteAAPWhiteAAP
  1. Abbreviations: AA, African American; SEER, Surveillance, Epidemiology, and End Results.

  2. Asterisk (*) iindicates P values not statistically significant.

 4182 (83.1%)845 (16.8%) 4354 (84.4%)802 (15.5%) 4058 (82.7%)844 (17.2%) 3552 (82.8%)738 (17.2%) 3336 (83.6%)657 (16.5%) 
Mean age, y49.553.1*48.850.6.00448.150.7*49.151.3*49.051.0.002
Marital status               
Unmarried1768 (42.3%)521 (61.7%)*1941 (44.6%)527 (65.7%)*1878 (46.3%)588 (69.7%)*1682 (47.4%)505 (68.4%)*1567 (47.0%)470 (71.5%)*
Married2201 (52.6%)246 (29.1%) 2149 (49.4%)207 (25.8%) 1962 (48.3%)207 (24.5%) 1651 (46.5%)198 (26.8%) 1549 (46.4%)143 (21.8%) 
Unknown213 (5.1%)78 (9.2%) 264 (6.1%)68 (8.5%) 218 (5.4%)49 (5.8%) 219 (6.2%)35 (4.7%) 220 (6.6%)44 (6.7%) 
Stage (%)               
Localized2230 (53.3%)321 (38.0%)*2429 (55.8%)352 (43.9%)*2330 (57.4%)418 (49.5%)*1880 (52.9%)330 (44.7%)*1687 (50.6%)291 (44.3%).034
Regional1319 (31.5%)342 (40.5%) 1328 (30.5%)291 (36.3%) 1205 (29.7%)282 (33.4%) 1176 (33.1%)311 (42.1%) 1182 (35.4%)262 (39.9%) 
Distant343 (8.2%)116 (13.7%) 310 (7.1%)72 (9.0%) 282 (6.9%)66 (7.8%) 382 (10.8%)76 (10.3%) 376 (11.3%)84 (12.8%) 
Unstaged290 (6.9%)66 (7.8%) 287 (6.6%)87 (10.8%) 241 (5.9%)78 (9.2%) 114 (3.2%)21 (2.8%) 91 (2.7%)20 (3.0%) 
Grade (%)               
I283 (6.8%)44 (5.2%)*310 (7.1%)32 (4.0%).007383 (9.4%)40 (4.7%)*347 (9.8%)42 (5.7%)*358 (10.7%)28 (4.3%)*
II848 (20.3%)166 (19.6%) 961 (22.1%)174 (21.7%) 1025 (25.3%)207 (24.5%) 1029 (29.0%)186 (25.2%) 1040 (31.2%)168 (25.6%) 
III907 (21.7%)222 (26.3%) 1069 (24.6%)221 (27.6%) 1019 (25.1%)217 (25.7%) 905 (25.5%)238 (32.2%) 906 (27.2%)223 (33.9%) 
IV77 (1.8%)18 (2.1%) 113 (2.6%)27 (3.4%) 104 (2.6%)17 (2.0%) 91 (2.6%)20 (2.7%) 85 (2.5%)9 (1.4%) 
Unknown2067 (49.4%)395 (46.7%) 1901 (43.7%)348 (43.4%) 1527 (37.6%)363 (43.0%) 1180 (33.2%)252 (34.1%) 947 (28.4%)229 (34.9%) 
Histology (%)               
Adenocarcinoma624 (14.9%)64 (7.6%)*671 (15.4%)80 (10.0%)*736 (18.1%)74 (8.8%)*739 (20.8%)73 (9.9%)*785 (23.5%)72 (11.0%)*
Adenosquamous140 (3.3%)41 (4.9%) 202 (4.6%)32 (4.0%) 177 (4.4%)26 (3.1%) 152 (4.3%)32 (4.3%) 122 (3.7%)18 (2.7%) 
Squamous3072 (73.5%)660 (78.1%) 3120 (71.7%)601 (74.9%) 2781 (68.5%)659 (78.1%) 2344 (66.0%)571 (77.4%) 2126 (63.7%)504 (76.7%) 
Other346 (8.3%)80 (9.5%) 361 (8.3%)89 (11.1%) 364 (9.0%)85 (10.1%) 317 (8.9%)62 (8.4%) 303 (9.1%)63 (9.6%) 
Treatment (%)               
None312 (7.5%)91 (10.8%)*304 (7.0%)88 (11.0%)*252 (6.2%)81 (9.6%)*249 (7.0%)61 (8.3%)*281 (8.4%)81 (12.3%)*
Surgery1821 (43.5%)224 (26.5%) 2095 (48.1%)285 (35.5%) 2062 (50.8%)310 (36.7%) 1577 (44.4%)229 (31.0%) 1414 (42.4%)212 (32.3%) 
Radiation1335 (31.9%)367 (43.4%) 1052 (24.2%)264 (32.9%) 861 (21.2%)248 (29.4%) 921 (25.9%)280 (37.9%) 983 (29.5%)253 (38.5%) 
Surgery and radiation714 (17.1%)163 (19.3%) 903 (20.7%)165 (20.6%) 883 (21.8%)205 (24.3%) 805 (22.7%)168 (22.8%) 658 (19.7%)111 (16.9%) 

Conclusions

In the present analysis, AA women presented overall with higher disease stage and had a higher prevalence of adverse prognostic indicators compared with white women. Although we found that after adjusting for standard treatment and other confounders, there were no differences in CC-specific survival for AA compared with whites after 1995, when looking at the survival for the whole group, AA women had worse overall and disease-specific survival. Prior studies have shown that CC incidence and death rates vary considerably among racial and ethnic groups. Siegel et al[8] showed that the CC death rate in the United States during the years of 2003 to 2007 was higher for AA (4.4 per 100,000) compared to whites (2.2 per 100,000). Patel et al,[9] while evaluating the racial/ethnic differences in survival after diagnosis with invasive CC, after adjusting for tumor characteristics and treatment reported that AA were at 19% increased risk of death (HR = 1.19, 95% CI = 1.06-1.33). In a study from the South Carolina Central Cancer Registry from 1996-2006, the authors noted significant differences in survival rates for AA and white patients after matching on several factors and adjusting for lifestyle confounders.[10]

In this investigation, AAs were less likely than whites to be diagnosed with a cancer at a localized stage. This finding is consistent with findings from previously published studies.[5, 9, 11-14] The reason for the higher proportion of advanced stage disease we observed among AA women is not clear. Studies have shown that, in part, racial/ethnic differences in disease stage at diagnosis are a result of the underutilization of screening among racial/ethnic minorities.[10] In fact, in a study of women with CC treated in an equal access, military health care system, where the impact of sociodemographic biases should be diminished, race was not an independent predictor of survival.[15] Several studies have reported that medically uninsured women have lower cancer screening rates and often present at later stages of disease.[16] Among Medicare patients with CC, those enrolled in health maintenance organizations are less likely than fee-for-service enrollees to be diagnosed with late-stage disease.[17] In addition, Hiatt et al[18] found that the strongest predictors of cancer screening were having private health insurance and frequent use of medical services.

Most studies would suggest that racial disparities in cancer treatment occur in the area of receipt of definitive primary therapy, conservative surgery, adjuvant therapy, and follow-up.[19, 20] Reports have demonstrated disparities among racial/ethnic minorities in the receipt of CC treatment affecting the assignment of clinical staging, and the receipt of surgical treatment, intracavitary radiation therapy (ICRT), and definitive treatment. [13, 21-23] Based on SEER data, AA when compared to whites, were less likely to be treated with surgery alone, more likely to receive no treatment or to have radiation therapy alone for their CC. This difference was seen after adjusting for age and stage of disease.[13] Patel et al.,[9] found that more AA patients had radiation therapy of any type as part of the first course of cancer-directed treatment compared with the other racial/ethnic groups. In a study using SEER data, del Carmen et al.[24] reported racial differences in the management of women with Stage IA2 CC with older minority women less likely treated by hysterectomy and more likely to be treated by fertility-sparing, less definitive procedures. Proposed explanations for these disparities in treatment include structural barriers, the existence of other medical comorbidities, patient's choice to decline recommended treatment, and physician's bias in making treatment recommendations.[20, 22, 23]

The temporal trend analysis in the present study showed that after adjusting for SEER registry, marital status, stage, age, treatment, grade and histology, there was a significant difference in CC-specific mortality between AA and whites between 1985-1989 and 1990-1994, but after 1995 there was not a difference in CC-specific mortality. The findings that AA were older, less likely to present with localized stage, and also less likely to receive surgical treatment in the earlier cohorts, may be a reflection of more aggressive targeting by screening programs among white women when compared to AA women in the cohorts before 1995. These findings also suggest that screening programs in minorities probably improved in the 1990s, when AA women were more often diagnosed with localized disease and receive more frequently surgical treatment, compared to earlier cohorts. However, despite these efforts, the significant difference in the rate of early stage and surgical management between the groups persists, even in the last 10 years of this analysis. In addition, we also found that the rate of cervical adenocarcinoma increased more in whites compared to AA. Prior studies suggest that cytologic screening has been shown to effectively detect squamous cell carcinoma in early stages, whereas adenocarcinomas have been reported to be less detectable by screening.[25, 26] These differences in trends with regards histology could also reflect differences in screening or biological characteristics of the tumor. Targeted screening has been shown to be successful for minority women. In cooperation with the NCI-funded community-based Cancer Control Programs, the Indian Health Service sponsored screening programs more aggressively targeting American Indian women older than 60 years.[27] Through these efforts, the incidence rate of CC decreased by 66% among American Indian women in New Mexico, with a significant concomitant shift also seen toward earlier diagnosis of CC. In an era where immigrants and ethnic minorities increasingly comprise a growing segment of the CC burden in the United States, it is incumbent upon the medical and public health community to overcome socioeconomic and cultural barriers to provide adequate care for patients with CC, and develop effective programs for prevention and early detection of CC.[28, 29]

We found that mortality declined in the cohorts of 2000-2004 and 2005-2009 compared to earlier cohorts. In 1999 the National Cancer Center released a clinical alert to practicing oncologists, based on significant improvement in both disease-free survival (DFS) and overall survival (OS) when cisplatin-based chemotherapy was administered during radiation for various stages of cervical cancer.[30] The mortality declined in the cohorts of 2000-2004 and 2005-2009 suggests an impact of chemoradiation in survival in a population-based analysis. The effect of the addition of cisplatin when used as a radiation sensitizer may be of higher magnitude, as observed in this study, for women presenting with more advanced stage disease that will necessitate radiation therapy as opposed to curative extirpative treatment. However, whether the superior survival observed in earlier cohorts is a direct function of chemoradiation itself or due to other factors cannot be answered from these data.

This study includes a large population-based cohort of patients with up to 25 years of follow-up, selected from SEER 9 areas that account for approximately 9.4% of the US population. However, there are several limitations to this current study that must be considered in interpreting the data. First, our results could be biased if the accuracy of assessments of cause of death varies by race or across the 5-year cohorts. In addition, this data set was only able to capture race based on classification by a health care provider; therefore, we are limited by an imperfect assessment of race. Additionally, data collected by SEER represent a nonhomogeneous population and, thus, outcomes may be affected by differences in treatment protocols, health care access, health behavior attitudes, regional customs, socioeconomic status, or environmental exposures.[31] Finally, our study may have benefited from adjustment for comorbid conditions, health insurance, or socioeconomic status.

In conclusion, analysis of population-based SEER data indicates significant survival differences by race for women with invasive CC, even after adjusting for SEER registry, marital status, stage, age, treatment, grade and histology. The difference in survival is likely the end result of multiple complex factors including screening, diagnosis, access to care, quality of that care, and treatment disparities, as well as other cultural and social issues. The existence of racial differences in CC demands attention, and the persistence of those disparities over the past 3 decades underscores the charge to dissipate these disparities a priority. The dissolution of health care disparities in the CC spectrum, from screening to treatment and subsequent follow-up, will require the implementation of accessible, culturally competent screening, vaccination, and treatment programs, as well as further research efforts to better understand the factors that contribute to the access and delivery of equitable care across all racial and ethnic groups.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. FUNDING SOURCES
  7. CONFLICT OF INTEREST DISCLOSURE
  8. REFERENCES
  • 1
    Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin. 2013;63:1130.
  • 2
    Kohler BA, Ward E, McCarthy BJ, et al. Annual report to the nation on the status of cancer, 1975-2007, featuring tumors of the brain and other nervous system. J Natl Cancer Inst. 2011;103:714736.
  • 3
    Howlader N, Noone A, Krapcho M, et al. editors. SEER Cancer Statistics Review, 1975-2008. Bethesda, MD: National Cancer Institute; 2011.
  • 4
    Scarinci IC, Garcia FA, Kobetz E, Partridge EE, Brandt HM, Bell MC, et al. Cervical cancer prevention: new tools and old barriers. Cancer. 2010;116:25312542.
  • 5
    Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin. 2005;55:1030.
  • 6
    Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) Research Data (1973-2009), National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2012, based on the November 2011 submission.
  • 7
    Terplan M, Schluterman N, McNamara EJ, Tracy JK, et al. Have racial disparities in ovarian cancer increased over time? An analysis of SEER data. Gynecol Oncol. 2012;125:1924.
  • 8
    Siegel R, Ward E, Brawley O, et al. Cancer statistics, 2012: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin. 2012;61:212236.
  • 9
    Patel DA, Barnholtz-Sloan JS, Patel MK, et al. A population-based study of racial and ethnic differences in survival among women with invasive cervical cancer: analysis of Surveillance, Epidemiology, and End Results data. Gynecol Oncol. 2005;97:550558.
  • 10
    Adams SA, Fleming A, Brandt HM, et al. Racial disparities in cervical cancer mortality in an African American and European American cohort in South Carolina. J S C Med Assoc. 2009;105:237244.
  • 11
    Del Carmen MG, Avila-Wallace M. Effect of health care disparities on screening. Clin Obstet Gynecol. 2013;56:6575.
  • 12
    Howell EA, Chen YT, Concato J. Differences in cervical cancer mortality among Black and White women. Obstet Gynecol. 1999;94:509515.
  • 13
    Brooks SE, Baquet CR, Gardner JF, et al. cervical cancer—the impact of clinical presentation, health and race on survival. J Assoc Acad Minor Phys. 2000;11:5559.
  • 14
    Hicks ML, Yap OW, Matthews R, Parham G. Disparities in cervical cancer screening, treatment and outcomes. Ethn Dis. 2006;16(2 suppl 3):S363-6.
  • 15
    Farley JH, Hines JF, Taylor RR, et al. Equal care ensures equal survival for African American women with cervical carcinoma. Cancer. 2001;91:869873.
  • 16
    Ferrante JM, Gonzalez EC, Roetzheim RG, et al. Clinical and demographic predictors of late-stage cervical cancer. Arch Fam Med. 2000;9:439445.
  • 17
    Riley GF, Potosky AL, Lubitz JD, Brown ML. Stage of cancer at diagnosis for Medicare HMO and fee-for-service enrollees. Am J Public Health. 1994;84:15981604.
  • 18
    Hiatt RA, Pasick RJ, Stewart S, et al. Community-based cancer screening for underserved women: design and baseline findings from the Breast and Cervical Cancer Intervention Study. Prev Med. 2001;33:1 90203.
  • 19
    Dignam JJ. Differences in breast cancer prognosis among African American and Caucasian women. CA Cancer J Clin. 2000;50:5064.
  • 20
    Shavers VL, Brown ML. Racial and ethnic disparities in the receipt of cancer treatment. J Natl Cancer Inst. 2002;94:334357.
  • 21
    Russell AH, Shingleton HM, Jones WB, et al. Trends in the use of radiation and chemotherapy in the initial management of patients with carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys. 1998;40:605613.
  • 22
    Merrill RM, Merrill AV, Mayer LS. Factors associated with no surgery or radiation therapy for invasive cervical cancer in Black and White women. Ethn Dis. 2000;10:248256.
  • 23
    Mundt AJ, Connell PP, Campbell T, Hwang JH, Rotmensch J, Waggoner S. Race and clinical outcome in patients with carcinoma of the uterine cervix treated with radiation therapy. Gynecol Oncol. 1998;71:151158.
  • 24
    del Carmen MG, Montz FJ, Bristow RE, Bovicelli A, Cornelison T, Trimble E. Ethnic differences in patterns of care of stage 1A(1) and stage 1A(2) cervical cancer: a SEER database study. Gynecol Oncol. 1999;75:113117.
  • 25
    Mitchell H, Medley G, Gordon I, Giles G. Cervical cytology reported as negative and risk of adenocarcinoma of the cervix: no strong evidence of benefit. Br J Cancer. 1995;71:894897.
  • 26
    Zappa M, Visioli CB, Ciatto S, Iossa A, Paci E, Sasieni P. Lower protection of cytological screening for adenocarcinomas and shorter protection for younger women: the results of a case-control study in Florence. Br J Cancer. 2004;4:17841786.
  • 27
    Chao A, Thomas B, Scott JW. Decreasing rates of cervical cancer among American Indians and Hispanics in New Mexico (United States). Cancer Causes Control. 1996;7:205213.
  • 28
    Downs LS, Smith JS, Scarinci I, et al. The disparity of cervical cancer in diverse populations. Gynecol Oncol. 2008;109:S22S30.
  • 29
    Johnson CE, Mues KE, Mayne SL, Kiblawi AN. Cervical cancer screening among immigrants and ethnic minorities: a systematic review using the Health Belief Model. J Low Genit Tract Dis. 2008;12:232241.
  • 30
    National Cancer Institute: NCI Clinical Announcement. Bethesda, MD, United States Department of Health and Human Services, Public Health Service, National Institutes of Health, February 1999.
  • 31
    Cornelison TL, Trimble EL, Kosary CL. SEER data, corpus uteri cancer: treatment trends versus survival for FIGO stage II, 1988-1994. Gynecol Oncol. 1999;74:350355.