The influence of comorbid conditions on racial disparities in endometrial cancer survival

Accepted Manuscript

The influence of comorbid conditions on racial disparities in endometrial cancersurvival

Julie J. Ruterbusch, MPH Rouba Ali-Fehmi, MD Sara H. Olson, PhD Shawnita Sealy-Jefferson, PhD Benjamin A. Rybicki, PhD Sharon Hensley-Alford, PhD MohamedA. Elshaikh, MD Arthur R. Gaba, MD Daniel Schultz, MD Adnan R. Munkarah, MDMichele L. Cote, PhD

PII: S0002-9378(14)00609-7

DOI: 10.1016/j.ajog.2014.06.036

Reference: YMOB 9896

To appear in: American Journal of Obstetrics and Gynecology

Received Date: 28 February 2014

Revised Date: 21 May 2014

Accepted Date: 17 June 2014

Please cite this article as: Ruterbusch JJ, Ali-Fehmi R, Olson SH, Sealy-Jefferson S, Rybicki BA,Hensley-Alford S, Elshaikh MA, Gaba AR, Schultz D, Munkarah AR, Cote ML, The influence of comorbidconditions on racial disparities in endometrial cancer survival, American Journal of Obstetrics andGynecology (2014), doi: 10.1016/j.ajog.2014.06.036.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

http://dx.doi.org/10.1016/j.ajog.2014.06.036

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTRuterbusch | page 1

The influence of comorbid conditions on racial disparities in endometrial cancer survival 1

2

Julie J. RUTERBUSCH, MPH1,2

, Rouba ALI-FEHMI, MD1,3

, Sara H. OLSON, PhD4, Shawnita SEALY-3

JEFFERSON, PhD1,5

, Benjamin A. RYBICKI, PhD6, Sharon HENSLEY-ALFORD, PhD

6,9, Mohamed A. 4

ELSHAIKH, MD7, Arthur R. GABA, MD

8, Daniel SCHULTZ, MD

8, Adnan R. MUNKARAH, MD

9, 5

Michele L. COTE, PhD1,2

6

7

Author Affiliations: 8

1. Barbara Ann Karmanos Cancer Institute, Detroit, MI 9

2. Wayne State University, Department of Oncology, Detroit, MI 10

3. Wayne State University, Department of Pathology, Detroit, MI 11

4. Memorial Sloan-Kettering Cancer Center, New York, NY 12

5. Wayne State University, Department of Family Medicine and Public Health Sciences, Detroit, 13

MI 14

6. Henry Ford Health System, Department of Public Health, Detroit, MI 15

7. Henry Ford Health System, Department of Radiation Oncology, Detroit, MI 16

8. Henry Ford Health System, Department of Pathology, Detroit, MI 17

9. Henry Ford Health System, Department of Women’s Health, Detroit, MI 18

19

The authors reports no conflict of interest 20

21

This work was funded in part by the Institute for Populations Studies in Health Assessment, 22

Administration, Services and Economics, a partnership of the Henry Ford Health System and 23

Wayne State University. 24

25

Corresponding Author: 26

Julie J. Ruterbusch 27

4100 John R MM04EP 28

Detroit, MI 48201 29

Phone: 313.578.4242 30

Email: [email protected] 31

32

Word Count 33

Abstract: 242 34

Main Text: 2,080 35

MANUSCRIP

T

ACCEPTED


Condensation: The higher mortality from endometrial cancer in black women cannot be fully 36

explained by known predictors of survival or the higher prevalence of comorbid conditions. 37

38

Sort Title: Race, Comorbidites, and Endometrial Cancer Survival 39

MANUSCRIP

T

ACCEPTED


40

Abstract 41

Objective: There are known disparities in endometrial cancer survival with black women 42

experiencing a greater risk of death compared to white women. The purpose of this 43

investigation was to evaluate the role of comorbid conditions as modifiers of endometrial 44

cancer survival by race. 45

Study Design: 271 black and 356 white women diagnosed with endometrial cancer between 46

1990 and 2005 were identified from a large urban integrated health center. A retrospective 47

chart review was conducted to gather information on comorbid conditions, as well as other 48

known demographic and clinical predictors of survival. 49

Results: Black women experienced a higher hazard of death from any cause (HR 1.51, 95% CI 50

1.22-1.87) and from endometrial cancer (HR 2.42, 95% CI 1.63-3.60). After adjusting for known 51

clinical prognostic factors and comorbid conditions, the hazard of death for black women was 52

elevated but no longer statistically significant for overall survival (HR 1.22, 95% CI 0.94-1.57), 53

and hazard of death due to endometrial cancer remained significantly increased (HR 2.27, 95% 54

1.39-3.68). Both black and white women with a history of hypertension experienced a lower 55

hazard of death due to endometrial cancer (HR 0.47, 95% CI 0.23-0.98, HR 0.35, 95% CI 0.19-56

0.67, respectively). 57

Conclusion: The higher prevalence of comorbid conditions among black women does not fully 58

explain the racial disparities seen in endometrial cancer survival. The association between 59

hypertension and a lower hazard of death due to endometrial cancer is intriguing and further 60

investigation into the underlying mechanism is needed. 61

62

Keywords: Endometrial cancer, disease-specific survival, racial disparities, comorbid conditions, 63

hypertension 64

MANUSCRIP

T

ACCEPTED


Introduction 65

Endometrial cancer is the most common malignancy of the female genital tract, and is the 66

fourth leading cancer diagnosed in women, after breast, lung, and colon cancers.1 Incidence of 67

endometrial cancer in the U.S. had been decreasing over the past three decades, but recent 68

data shows a reversal of that trend with a 3.0% annual percentage increase during 2006 to 69

2010 compared with a negative 0.4% annual percentage change during 1997-2006.2 White 70

women are at greater risk of developing endometrial cancer than black women, however black 71

women are more likely to die from this disease. The lower survival rate among black women 72

was identified decades ago and this disparity has persisted over time.3,4

The mortality rate 73

from endometrial cancer in 2006-2010 for black women was nearly twice the mortality rate of 74

white women (7.4 versus 4.0 per 100,000).2 While black women are diagnosed with less 75

favorable histologies, at more advanced stages, and with higher grade tumors than white 76

women,5-8

poorer survival rates are still seen in black women for all stages, grades, and 77

histologic types when compared to their white counterparts.8,9

78

Reasons for these survival differences are likely due to a combination of factors, including 79

differences in socioeconomic resources, environmental and behavioral risk factors, and tumor 80

biology.5,7,10-12

One factor that hasn’t been fully evaluated is the role comorbid conditions play 81

in the racial disparity in endometrial cancer survival. Black women have a higher prevalence of 82

obesity, diabetes, and hypertension.13

While these conditions have been associated with poorer 83

survival rates,14

a recent report using SEER Medicare data illustrated that comorbid conditions 84

do not fully account for the racial disparity seen in endometrial cancer survival in a Medicare 85

MANUSCRIP

T

ACCEPTED


population.15

We sought to further evaluate the relationship between comorbid conditions and 86

the racial disparity in endometrial cancer survival among women of all ages at a single 87

institution. 88

Materials and Methods 89

After IRB approval was obtained, a case-only retrospective analysis of incident endometrial 90

cancer cases (ICD-O3 codes of C54.0-55.9) was conducted. Black and white women who were 91

diagnosed from 1990 to 2005 were identified from the Henry Ford Health System (HFHS) tumor 92

registry. The HFHS is a large, integrated health system in Detroit, Michigan. The HFHS currently 93

consists of 5 hospitals, 36 ambulatory care facilities, and clinics that offer free or low cost care 94

located throughout the metropolitan Detroit area and serves patients of varying levels of 95

socioeconomic and insurance status for both races. Clinical, demographic, risk factor, and 96

survival data were obtained from 3 sources: the Henry Ford Health System database, medical 97

record abstraction, and the Metropolitan Detroit Cancer Surveillance System (MDCSS) registry, 98

which is part of the National Cancer Institute’s Surveillance, Epidemiology and End Results 99

program. The MDCSS ascertainment area encompasses the 3 county (Wayne, Oakland, and 100

Macomb) metropolitan Detroit area, where the majority of patients seen at HFHS reside. 101

To standardize data collection, variables were abstracted from the medical record up to five 102

years prior to the endometrial cancer diagnosis. Race information was self-reported and 103

abstracted from the medical record. Comorbid conditions of interest included diabetes, 104

hypertension, obesity (body mass Index (BMI) of 30 kg/m2 or greater), morbid obesity (BMI of 105

40kg/m2 or greater), and a modified Charlson Comorbidity Index (CCI). The CCI is a weighed 106

MANUSCRIP

T

ACCEPTED


score of comorbid conditions shown to predict death16

and was modified in this analysis to 107

exclude diabetes (both complicated and un-complicated), as that condition is an established 108

risk factor for endometrial cancer and thus was evaluated separately. A condition was counted 109

in the CCI if it was diagnosed prior to the endometrial cancer diagnosis. BMI was calculated 110

from height and weight measurements in the medical record that were documented 2-5 years 111

prior to diagnosis, so that weight measures would not be influenced by any weight loss that was 112

part of the disease process. 113

Information on known risk factors for endometrial cancer was also abstracted from the medical 114

record. Smoking history was grouped into 3 categories: never smoker, former smoker, and 115

current smoker. Reported parity was evaluated 2 ways: as a dichotomous variable (nulliparous 116

vs parous) and as a categorical variable (nulliparous, 1 to 3 live birth, and 4 or more live births). 117

Histology and grade information were reviewed by the HFHS pathologists (A.R.G. and D.S.) to 118

standardize categorization. All cases were then re-reviewed by a Wayne State University 119

gynecologic pathologist (R.A.F.) and any discrepancy was resolved by consensus amongst the 3 120

pathologists ( A.R.G, D.S., and R.A.F.). Histologic type was grouped into 4 categories: type I, 121

type II, type III, and other. Type I included endometrioid and mucinous adenocarcinoma 122

histologies, type II included serous, clear cell, and mixed histologies, type III included malignant 123

mixed Mullerian tumors, and the final category included all other histologic types. Vital status 124

information was obtained from the MDCSS database and follow-up information was obtained 125

through the end of 2012. 126

MANUSCRIP

T

ACCEPTED


Differences in age at diagnosis, FIGO stage, grade, histology, comorbid conditions, body mass 127

index (BMI) prior to diagnosis, smoking history, and parity were examined by race. Racial 128

differences in tumor characteristics were further examined stratified by comorbid conditions. 129

The differences in the distribution of these clinical and demographic variables were assessed 130

using chi-square tests. 131

Log-rank tests and Cox proportional hazards models were used to assess the risk of overall and 132

endometrial cancer related deaths. Endometrial cancer related deaths were defined using both 133

the primary and underlying causes of death (ICD-9 codes I79, I82 and ICD-10 code C54) that 134

were recorded on the death certificate and survival time was calculated from the date of 135

biopsy. Race-specific hazard ratios (HR) and 95% confidence intervals (CI) for overall and 136

endometrial cancer related survival, were estimated for each comorbid condition and adjusted 137

for known predictors of survival that were significant in univariate analyses. These predictor 138

variables were (age and year at diagnosis, FIGO stage, grade, histology type, and treatment 139

(surgery, chemotherapy, and radiation therapy). In addition, an interaction term with race and 140

each comorbid condition was calculated. Overall and disease-specific hazard of death was 141

estimated for black women, as compared to white women, and then estimated after 142

adjustment using 2 different models. Model 1 was adjusted for year and age of diagnosis, 143

receipt of surgery, chemotherapy, radiation therapy, and tumor characteristics (stage, grade, 144

and histology). Model 2 was adjusted using all the variables listed for model 1 and comorbid 145

conditions (diabetes, hypertension, obesity, and CCI). These analyses were repeated in the 146

subset of women who were treated surgically. All analyses were performed using SAS 147

statistical software, version 9.2 (SAS Institute Inc., Cary, NC). 148

MANUSCRIP

T

ACCEPTED


Results 149

A total of 627 women, 271 (43%) black women and 356 (57%) white women, were identified for 150

inclusion in this study. Compared to their white counterparts, black women were more likely to 151

have aggressive disease phenotypes with higher proportions of type II/III histologies, higher 152

grade tumors, and to be diagnosed at a later stage of disease (all p<.001). Black women were 153

less likely to receive surgical treatment (17% versus 4%, p<.001) and more likely to receive 154

chemotherapy (p=0.041). Black women were more likely to be diagnosed with hypertension 155

(p<0.001) and to be obese (p<0.001). CCI (p=0.305) and the diagnosis of diabetes (p=0.086) 156

were similar in black and white women with endometrial cancer (Table 1). The significant racial 157

differences seen in histology and grade persisted when examined by either the presence or 158

absence of diabetes, obesity, hypertension, or other comorbid conditions (Table 2). 159

Black women experienced lower overall and disease-specific survival compared to white 160

women (p=0.0001 and p<.0001 respectively, Figure 1a and 1b). The median overall survival for 161

black women was 81 months compared to 148 months for white women. White women 162

diagnosed with diabetes had a higher hazard of overall death (HR: 1.63, 95% CI 1.13-2.35) while 163

black women did not (HR: 1.17, 95% CI 0.80-1.71). Diabetes was not associated with an 164

increased risk of death from endometrial cancer for either racial group (Table 3). Hypertensive 165

black women had a lower hazard of overall death (HR: 0.52, 95% CI 0.34-0.79) while white 166

women with hypertension did not (HR: 0.96, 95% CI 0.70-1.32, pinteraction=0.008). Both black and 167

white women diagnosed with hypertension had a lower hazard of death due to endometrial 168

cancer (white women HR: 0.47, 95% CI 0.23-0.98, black women HR: 0.35, 95% CI 0.19-0.67). 169

MANUSCRIP

T

ACCEPTED


Obesity or morbid obesity was not associated with the overall or disease-specific hazard of 170

death for either racial group. For both white and black women, a score of 2 or more on the CCI 171

was associated with a higher hazard of overall death (HR: 1.71, 95% CI 1.17-2.50, and HR: 2.28, 172

95% CI 1.46-3.56, respectively), but not for death due to endometrial cancer. 173

The hazard of overall death was 51% higher for black women compared to white women and 174

nearly two and half times as high for endometrial cancer related deaths. As seen in Table 4a, 175

adjusting for aggressive disease features attenuated the hazard ratios, but the hazard for 176

overall and disease specific survival remained significantly elevated for black women. After 177

further adjustment for comorbid conditions, the hazard for overall survival remained elevated 178

but was no longer statistically significant and the hazard due to endometrial cancer death 179

remained significantly elevated. When this analysis was limited to women who received 180

surgical treatment (Table 4b), the unadjusted hazard for overall and endometrial cancer related 181

death for black women remained significantly elevated, yet appeared to be slightly smaller in 182

magnitude. The same pattern persisted in this subgroup when adjusting for aggressive disease 183

features and comorbid conditions. 184

Comment 185

In this single institution study, racial differences in survival were seen for both overall and 186

endometrial cancer-related survival, with black women having a higher risk of death. While 187

black women had greater proportions of high grade tumors with aggressive histologic types, 188

adjusting for these differences did not fully explain all of the racial differences seen in survival, 189

nor did adjusting for the presence of comorbid conditions. 190

MANUSCRIP

T

ACCEPTED


Our findings suggest that excess comorbid conditions do not explain the racial differences seen 191

in endometrial cancer survival. This finding supports those seen in an older population-based 192

sample from SEER/Medicare.15

When comparing the impact of individual conditions, 193

interesting similarities and differences were found. Like Olson et al,15

we observed that black 194

women diagnosed with hypertension had improved disease-specific and overall survival. 195

Hypertensive white women also had improved disease-specific survival, but not improved 196

overall survival. The confirmation of this finding is intriguing. There were no differences seen 197

in stage, grade, or histology between normotensive and hypertensive women. 198

One possible explanation for the association between hypertension and improved survival is 199

the purported beneficial role of drugs commonly used in the treatment of hypertension and 200

cancer survival. A recent review by McMenamin et al concluded that angiotensin-converting 201

enzyme inhibitors and angiotensin receptor blockers may be associated with improved 202

outcomes in cancer patients.17

However, there were inconsistent results across studies and 203

none of the studies included in the review were of endometrial cancer. There have been some 204

smaller studies that suggest these drugs may have a role in survival among other gynecological 205

cancers. In particular, the use of statins and beta-blockers has been shown to be independent 206

prognostic factors in survival of women with epithelial ovarian cancer.18,19

207

The higher proportion of aggressive tumors among black women persisted when examined by 208

the presence, or absence of, comorbid conditions. Unlike other cancer sites, where it has been 209

suggested that comorbid conditions are associated with aggressive tumor types,20-22

it is 210

MANUSCRIP

T

ACCEPTED


unlikely that the higher prevalence of diabetes, hypertension, or obesity seen among black 211

women accounts for higher proportion of aggressive endometrial tumors. 212

There are limitations to consider when interpreting the results of this analysis. All of the data 213

sources were retrospective, and though our medical record review lends confidence to the 214

presence of the comorbid conditions, we did not capture information on disease severity or 215

medication use. These factors are likely to be important confounders of the effect of 216

comorbidity and survival. This single institution study may not represent the larger population 217

but does include a substantial population of black women from a diverse metropolitan region. 218

The results of this study are supported by data gathered from multiple sources. We leveraged 219

the strengths of existing registry data and supplemented known data gaps with medical record 220

abstraction and expert pathology review. These results add to the body of literature that 221

illustrates racial disparities in endometrial cancer survival between white and black women that 222

are not fully explained by known risk factors. Further work is needed to elucidate the 223

underlying factors of this disparity, and to explore the potential mechanisms between 224

hypertension and endometrial cancer survival. 225

Acknowledgements: 226

This work was funded in part by the Institute for Populations Studies in Health Assessment, 227

Administration, Services and Economics, a partnership of the Henry Ford Health System and 228

Wayne State University. 229

230

Figure Title 231

Figure 1. Endometrial cancer survival by race; overall survival (1A) and disease specific survival 232

(1B). 233

MANUSCRIP

T

ACCEPTED


References 234

1. Siegel R, Ma J, Zou Z, Jemal A (2014) Cancer statistics, 2014. CA Cancer J Clin. 2014 Jan-Feb; 235

64(1):9-29. doi:10.3322/caac.21208.2. Howlader N NA, Krapcho M, Garshell J, Neyman N, 236

Altekruse SF, Kosary CL, Yu M, Ruhl J, Tatalovich Z, Cho H, Mariotto A, Lewis DR, Chen HS, Feuer 237

EJ, Cronin KA (eds). SEER Cancer Statistics Review, 1975-2010, National Cancer Institute. 238

Bethesda, MD, http://seer.cancer.gov/csr/1975_2010/, based on November 2012 SEER data 239

submission, posted to the SEER web site, April 2013. 240

3. Hill HA, Eley JW, Harlan LC, Greenberg RS, Barrett RJ, 2nd, Chen VW (1996) Racial differences 241

in endometrial cancer survival: the black/white cancer survival study. Obstetrics and gynecology 242

88 (6):919-926 243

4. Yap OW, Matthews RP (2006) Racial and ethnic disparities in cancers of the uterine corpus. 244

Journal of the National Medical Association 98 (12):1930-1933 245

5. Long B, Liu FW, Bristow RE (2013) Disparities in uterine cancer epidemiology, treatment, and 246

survival among African Americans in the United States. Gynecologic oncology 130 (3):652-659. 247

doi:10.1016/j.ygyno.2013.05.020 248

6. Setiawan VW, Pike MC, Kolonel LN, Nomura AM, Goodman MT, Henderson BE (2007) 249

Racial/ethnic differences in endometrial cancer risk: the multiethnic cohort study. American 250

journal of epidemiology 165 (3):262-270. doi:10.1093/aje/kwk010 251

7. Randall TC, Armstrong K (2003) Differences in treatment and outcome between African-252

American and white women with endometrial cancer. Journal of clinical oncology : official 253

journal of the American Society of Clinical Oncology 21 (22):4200-4206. 254

doi:10.1200/JCO.2003.01.218 255

8. Hicks ML, Phillips JL, Parham G, Andrews N, Jones WB, Shingleton HM, Menck HR (1998) The 256

National Cancer Data Base report on endometrial carcinoma in African-American women. 257

Cancer 83 (12):2629-2637 258

9. Wright JD, Fiorelli J, Schiff PB, Burke WM, Kansler AL, Cohen CJ, Herzog TJ (2009) Racial 259

disparities for uterine corpus tumors: changes in clinical characteristics and treatment over 260

time. Cancer 115 (6):1276-1285. doi:10.1002/cncr.24160 261

10. Madison T, Schottenfeld D, James SA, Schwartz AG, Gruber SB (2004) Endometrial cancer: 262

socioeconomic status and racial/ethnic differences in stage at diagnosis, treatment, and 263

survival. American journal of public health 94 (12):2104-2111 264

11. Sherman ME, Devesa SS (2003) Analysis of racial differences in incidence, survival, and 265

mortality for malignant tumors of the uterine corpus. Cancer 98 (1):176-186. 266

doi:10.1002/cncr.11484 267

12. Allard JE, Maxwell GL (2009) Race disparities between black and white women in the 268

incidence, treatment, and prognosis of endometrial cancer. Cancer control : journal of the 269

Moffitt Cancer Center 16 (1):53-56 270

13. Fryar CD, Hirsch R, Eberhardt MS, Yoon SS, Wright JD (2010) Hypertension, high serum total 271

cholesterol, and diabetes: racial and ethnic prevalence differences in U.S. adults, 1999-2006. 272

NCHS data brief (36):1-8 273

14. Nicholas Z, Hu N, Ying J, Soisson P, Dodson M, Gaffney DK (2012) Impact of Comorbid 274

Conditions on Survival in Endometrial Cancer. American journal of clinical oncology. 275

doi:10.1097/COC.0b013e318277d5f4 276

MANUSCRIP

T

ACCEPTED


15. Olson SH, Atoria CL, Cote ML, Cook LS, Rastogi R, Soslow RA, Brown CL, Elkin EB (2012) The 277

impact of race and comorbidity on survival in endometrial cancer. Cancer epidemiology, 278

biomarkers & prevention : a publication of the American Association for Cancer Research, 279

cosponsored by the American Society of Preventive Oncology 21 (5):753-760. 280

doi:10.1158/1055-9965.EPI-11-0735 281

16. Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying 282

prognostic comorbidity in longitudinal studies: development and validation. Journal of chronic 283

diseases 40 (5):373-383 284

17. Mc Menamin UC, Murray LJ, Cantwell MM, Hughes CM (2012) Angiotensin-converting 285

enzyme inhibitors and angiotensin receptor blockers in cancer progression and survival: a 286

systematic review. Cancer causes & control : CCC 23 (2):221-230. doi:10.1007/s10552-011-287

9881-x 288

18. Diaz ES, Karlan BY, Li AJ (2012) Impact of beta blockers on epithelial ovarian cancer survival. 289

Gynecologic oncology 127 (2):375-378. doi:10.1016/j.ygyno.2012.07.102 290

19. Elmore RG, Ioffe Y, Scoles DR, Karlan BY, Li AJ (2008) Impact of statin therapy on survival in 291

epithelial ovarian cancer. Gynecologic oncology 111 (1):102-105 292

20. Maiti B, Kundranda MN, Spiro TP, Daw HA (2010) The association of metabolic syndrome 293

with triple-negative breast cancer. Breast cancer research and treatment 121 (2):479-483. 294

doi:10.1007/s10549-009-0591-y 295

21. Kheterpal E, Sammon JD, Diaz M, Bhandari A, Trinh QD, Pokala N, Sharma P, Menon M, 296

Agarwal PK (2013) Effect of metabolic syndrome on pathologic features of prostate cancer. 297

Urologic oncology 31 (7):1054-1059. doi:10.1016/j.urolonc.2011.12.012 298

22. Moses KA, Utuama OA, Goodman M, Issa MM (2012) The association of diabetes and 299

positive prostate biopsy in a US veteran population. Prostate cancer and prostatic diseases 15 300

(1):70-74. doi:10.1038/pcan.2011.40 301

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1: Select clinical, demographic, and treatment variables of women diagnosed with endometrial

cancer from a single institution by race.

White (N=356) Black (N=271)

N % N % p value*

Age at Diagnosis

0.204

<50 49 14% 32 12%

50-59 71 20% 39 14%

60-69 112 31% 99 37%

70+ 124 35% 101 37%

FIGO Stage

<.001

I 246 69% 142 52%

II 40 11% 32 12%

III 40 11% 37 14%

IV 19 5% 45 17%

Unknown 11 3% 15 6%

Grade

<.001

I 189 53% 79 29%

II 67 19% 35 13%

III 91 26% 135 50%

Unknown 9 3% 22 8%

Histology Group

<.001

Type I 260 73% 148 55%

Type II 67 19% 84 31%

Type III 24 7% 33 12%

Other 5 1% 6 2%

Modified Charlson CoMorbidity Score

0.305

None 208 58% 140 52%

1 64 18% 55 20%

2+ 84 24% 73 27%

Unknown 0 0% 3 1%

Hypertension

<.001

No 151 42% 65 24%

Yes 205 58% 206 76%

Diabetes

0.086

No 275 77% 193 71%

Yes 81 23% 78 29%

BMI

<.001

<25 66 19% 31 11%

25-29.9 91 26% 38 14%

30-39.9 108 30% 119 44%

40+ 76 21% 74 27%

Unknown 15 4% 9 3%

Smoking History

0.590

Non-Smoker 223 63% 182 67%

Current 47 13% 30 11%

Former 75 21% 55 20%

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Unknown 11 3% 4 1%

Nulliparity

0.004

No 264 74% 218 80%

Yes 85 24% 38 14%

Unknown 7 2% 15 6%

Parity

<.001

None 85 24% 38 14%

1-3 186 52% 126 46%

4+ 77 22% 92 34%

Unknown 8 2% 15 6%

Surgery

<.001

No 14 4% 45 17%

Yes 342 96% 226 83%

Radiation Therapy

0.686

No 225 63% 167 62%

Yes 131 37% 104 38%

Chemotherapy

0.041

No 286 80% 199 73%

Yes 70 20% 72 27%

Vital Status <.001

Alive 164 46% 100 37%

Deceased – Endometrial Cancer 43 12% 66 24%

Deceased – Other Reason 149 42% 105 39%

* p value calculation does not include unknown values

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 2: Racial differences in the distribution of clinical characteristics by the presence of comorbid conditions

Diabetes Obesity

Absent Present Absent Present

White Black p value White Black p value White Black p value White Black p value

Total 275 193 81 78 157 69 184 193

Histology Group <.001 0.051 0.001 0.006

Type I 75% 56% 73% 56% 74% 50% 73% 58%

Type II 20% 31% 16% 32% 20% 32% 19% 32%

Type III 6% 13% 11% 12% 6% 18% 8% 10%

Grade <.001 0.007 <.001 <.001

I 55% 31% 53% 33% 54% 22% 58% 36%

II 19% 14% 19% 13% 19% 13% 19% 15%

III 26% 55% 28% 53% 27% 65% 24% 49%

FIGO Stage <.001 0.040 0.012 0.004

I 71% 57% 71% 51% 74% 52% 69% 60%

II 13% 12% 8% 14% 8% 15% 15% 12%

III 11% 14% 13% 15% 12% 21% 12% 11%

IV 5% 16% 8% 21% 6% 12% 5% 17%

Hypertension Other Comorbid Conditions

Absent Present Absent Present

White Black p value White Black p value White Black p value White Black p value

Total 151 65 205 206 208 140 148 128

Histology Group 0.116 0.001 0.003 0.001

Type I 77% 63% 72% 54% 75% 58% 73% 52%

Type II 16% 26% 22% 34% 19% 36% 19% 28%

Type III 7% 11% 6% 13% 6% 7% 8% 19%

Grade 0.004 <.001 0.003 <.001

I 53% 31% 56% 32% 48% 35% 64% 28%

II 21% 19% 18% 13% 21% 16% 16% 12%

III 26% 50% 26% 56% 31% 49% 20% 60%

FIGO Stage 0.001 0.003 0.196 <.001

I 72% 54% 71% 56% 71% 64% 72% 47%

II 12% 10% 12% 13% 12% 12% 10% 14%

III 10% 10% 13% 16% 10% 11% 13% 19%

IV 6% 26% 5% 15% 6% 13% 5% 21%

*percentages may not total to 100 due to rounding

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 3: Impact of comorbid conditions on survival in women diagnosed with endometrial cancer.

White women Black women

HR* 95% CI p value HR* 95% CI p value p interaction

Death from Any Cause

Diabetes 1.63 1.13-2.35 0.009

1.17 0.80-1.71 0.429 0.302

Hypertension 0.96 0.70-1.32 0.817

0.52 0.34-0.79 0.003 0.008

Obesity (BMI 30+) 1.04 0.75-1.43 0.820

1.04 0.69-1.56 0.865 0.670

Morbid Obesity (BMI 40+) 1.33 0.86-2.06 0.198 1.07 0.70-1.63 0.771 0.901

Charlson CoMorb Score

0.991

1 1.89 1.24-2.90 0.003

1.07 0.66-1.72 0.795

2+ 1.71 1.17-2.50 0.006

2.28 1.46-3.56 <.001

Death from Endometrial Cancer

Diabetes 1.00 0.40-2.49 0.999 0.71 0.36-1.37 0.302 0.978

Hypertension 0.47 0.23-0.98 0.044 0.35 0.19-0.67 0.001 0.740

Obesity (BMI 30+) 0.89 0.44-1.79 0.739 0.81 0.44-1.51 0.510 0.723

Morbid Obesity (BMI 40+) 0.72 0.24-2.15 0.557 1.04 0.53-2.02 0.913 0.760

Charlson CoMorb Score 0.312

1 2.74 1.16-6.48 0.022 1.10 0.51-2.36 0.806

2+ 0.92 0.35-2.41 0.864 1.97 0.98-3.98 0.059

*Adjusted for year and age at diagnosis, FIGO stage, grade, histology type, receipt of surgery, chemotherapy and

radiation therapy.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 4: Hazard of death for black women diagnosed with endometrial cancer compared to white women at a

single institution.

4a. All Cases

Unadjusted Adjusted (model 1) Adjusted (model 2)

HR for black

race p value

HR for black

race p value

HR for black

race p value

Overall 1.51 (1.22-1.87) <.001 1.29 (1.02-1.63) 0.036 1.22 (0.94-1.57) 0.134

Disease-specific 2.42 (1.63-3.60) <.001 2.11 (1.35-3.30) 0.001 2.27 (1.39-3.68) 0.001

4b. Surgically treated cases

Unadjusted Adjusted (model 1) Adjusted (model 2)

HR for black

race p value

HR for black

race p value

HR for black

race p value

Overall 1.39 (1.11-1.75) 0.005 1.28 (1.00-1.64) 0.051 1.26 (0.97-1.64) 0.082

Disease-specific 2.15 (1.39-3.32) <.001 2.03 (1.27-3.24) 0.003 2.19 (1.32-3.64) 0.002

Model 1 – Adjusted for year and age at diagnosis, FIGO stage, grade, histology type, receipt of chemotherapy, and

radiation therapy. In table 4a, model also adjusted for surgery.

Model 2 – Adjusted for year and age at diagnosis, FIGO stage, grade, histology type, receipt of chemotherapy and

radiation therapy, diabetes, hypertension, obesity, and Charlson comorbid index. In table 4a, model also adjusted

for surgery.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Figure 1. Endometrial cancer survival by race; overall survival (1a) and disease specific survival (1b).

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Documents

The influence of comorbid conditions on racial disparities in endometrial cancer survival