Advertisement

Journal of Racial and Ethnic Health Disparities

, Volume 6, Issue 1, pp 133–142 | Cite as

Racial, Socioeconomic, and Geographic Disparities in the Receipt, Timing to Initiation, and Duration of Adjuvant Androgen Deprivation Therapy in Men with Prostate Cancer

  • Chi Nguyen
  • David R. Lairson
  • Michael D. Swartz
  • Xianglin L. DuEmail author
Article

Abstract

Objective

This retrospective cohort study aims to examine the receipt, timing to initiation, and duration of androgen deprivation therapy (ADT) in men with prostate cancer by race/ethnicity, socioeconomic status, and geographic location.

Methods

The study population are patients from Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database, who were 66 years or older and newly diagnosed with stage III and IV prostate cancer in 1992–2009 and underwent radiation therapy, where ADT was proven to be highly beneficial and its use was considered as most appropriate (n = 12,170). We use logistic regression to examine the receipt of ADT and linear regression to study factors associated with time to ADT initiation while controlling for baseline characteristics.

Results

Overall, 77% of eligible patients received at least one form of ADT in combination with radiation therapy, of which 12% underwent orchiectomy and the rest received ADT, and 77.2% of non-Hispanic white and 80.7% of Hispanic patients received ADT compared to 73.8% of non-Hispanic black. After adjustment for demographic and tumor characteristics, black men and men of other races are less likely to receive ADT compared to white counterparts (OR = 0.64 and 0.74, respectively). The median time from cancer diagnosis to ADT initiation is 2 months. Once initiated, men received a median of seven drug injections. After controlling for covariates, race/ethnicity and geographic location (SEER areas) are associated with early initiation of therapy. White, Hispanic men and men living in the South initiate ADT earlier.

Conclusion

Significant racial disparities exist in the receipt and use of this highly beneficial therapy, and there are geographic variations in the utilization of this therapy.

Keywords

Prostate cancer Androgen deprivation therapy Treatment Disparities Race/ethnicity 

Introduction

Prostate cancer is the second most common cancer among men worldwide and is the most common type of cancer among men in the USA [1]. American Cancer Society has estimated that about 164,690 new cases will be diagnosed and approximately 29,430 men will die from prostate cancer in 2018 [1]. In the past several decades, advances in treatment and early detection have accounted for significant improvement in survival. The 5-year relative survival rate has increased steadily from 68.3% in 1985 to almost 100% in 2012 [2].

Currently, there are four standard treatment modalities available for patients with prostate cancer: radical prostatectomy, radiation therapy, hormone therapy, and active surveillance. Hormone therapy is also known as androgen deprivation or suppression therapy (ADT). ADT can be achieved by either orchiectomy, a surgical procedure to remove the testicles, or castration drugs named luteinizing hormone-releasing hormone (LHRH) agonists and antagonists. The LHRH drugs approved for prostate cancer treatment in the USA and covered by Medicare include leuprolide goserelin, degarelix, and triptorelin. These drugs help to slow tumor growth by directly reducing androgen levels or preventing LHRH from binding to its receptors and thus indirectly reducing androgen levels.

ADT is recommended to be used in conjunction with radiation therapy among patients with high-risk prostate cancer who were diagnosed with stage III or IV tumors, or patients with recurrent or metastatic cancer [3, 4]. It also can be used among older patients with comorbidity, who are not suitable for aggressive treatment such as surgical treatment. The role of adjuvant ADT combined with radiotherapy among patients with advanced prostate cancer is well defined. A systematic review analyzed data from 10 clinical trials and showed a significant benefit with regard to both overall survival and disease-free survival for the use of ADT combined with radiotherapy compared to radiotherapy alone [5]. Specifically, LHRH drugs are associated with 28% reduction in the risk of death (HR 0.72, 95% confidence interval (CI) 0.60–0.87). They also showed a similar pattern for orchiectomy (HR 0.62, 95% CI 0.36–1.05) but failed to demonstrate statistical significance due to small sample size [5]. The role of ADT is controversial in the intermediate risk group and recommendation is pending, awaiting further evidence of treatment combining ADT with dose-escalated radiation therapy [6]. In contrast, the use of ADT as monotherapy among low-risk prostate cancer patients is associated with an increased risk of both overall and prostate cancer-specific mortality when compared to radical prostatectomy [7, 8]. No recommendation is available for the use of ADT among intermediate- and low-risk prostate cancer patients, or as an adjuvant therapy with radical prostatectomy due to insufficient evidence or no observed benefits [3].

The appropriate time to initiate ADT and the duration of ADT in various prostate cancer settings remain uncertain. When used with radiation therapy, immediate initiation and long duration of ADT appear to be beneficial. The European Organization for Research and Treatment of Cancer (EORTC) trial 30,891 reported a modest but statistically significant increase in overall survival, with median survival for immediate and deferred treatment groups of 7.6 and 6.1 years, respectively [9]. Evidence from a systematic review of trials published from 1950 to 2007 reinforced the benefits of early hormone therapy for patients with locally advanced prostate cancer in reducing all-cause and prostate cancer-specific mortality [10]. The duration of ADT is also a contributing factor. A meta-analysis showed that using ADT for 1 year or longer provided 39% reduction in risk of mortality while duration shorter than 6 months demonstrated only 21% reduction [5]. The national guidelines suggest using a long-term ADT for at least 28–36 months based on findings from two large randomized trials [11, 12].

Studies suggest racial and socioeconomic disparities in prostate cancer treatment and overall survival. Black men have higher incidence, more aggressive tumors, less aggressive treatment, and greater mortality compared to white men [13]. Black and Hispanic men are less likely to receive treatment compared to white men, even when they have more aggressive disease [14, 15]. Moreover, racial disparities are often in men who most likely benefited from treatment and had moderate- to high-risk tumors and life expectancy of more than 10 years [16]. Under-treatment is also reported among men from lower socioeconomic groups [17]. Disparity in treatment is a contributing factor for observed variations in survival and outcomes. Other possible factors include biological factors, patient beliefs, preferences, and behaviors [13]. Although one study examined racial differences in time to ADT initiation for metastatic prostate cancer, little is known about racial disparities in ADT initiation and duration, especially among men for whom ADT is highly appropriate [18].

This study examines the receipt, timing to initiation, and duration of ADT among patients who are candidates for this therapy by race/ethnicity, socioeconomic status, and geographic area. Our analysis focuses on patients diagnosed with high-risk (defined as stage III or IV) prostate cancer treated with radiation therapy, where ADT is proven highly beneficial, and its use is the standard of care. We hypothesize that black men are less likely to initiate ADT, show longer median time between diagnosis and receipt of ADT, and haveshorter duration of ADT use compared to white counterparts. We hypothesize substantial socioeconomic and geographic variations where those in higher socioeconomic status (SES) areas are more likely to have a shorter median initiation time and longer duration of treatment.

Patients and Methods

Data Sources

Data for the study are extracted from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. The SEER program collects data on clinical characteristics, demographics, and causes of death for cancer cases throughout the USA. The SEER data consist of 17 population-based cancer registries, including seven metropolitan or rural areas (San Francisco/Oakland, Detroit, Atlanta, Seattle-Puget Sound, Rural Georgia, Los Angeles, San Jose-Monterey) and eight states (Connecticut, Iowa, New Mexico, Utah, Hawaii, Kentucky, Louisiana, New Jersey), plus Greater California and Greater Georgia. The SEER data covered approximately 14% of the total US population in the period from 1992 to 1999 and then expanded coverage to about 28% of the total population in 2000. The population in SEER areas is comparable to the general US population in terms of poverty and education, but is more urban and has a higher proportion of foreign-born people.

Medicare is a federally funded program that provides health insurance for people older than 65 years, people with end-stage renal disease, and disabled people. Medicare enrollees have part A coverage, including hospital, skilled-nursing facility, hospice, and some home health care. About 96% of part A beneficiaries pay a monthly premium to enroll in part B, which covered physician and outpatient services. Medicare part C refers to HMO enrollment and part D provides prescription drug coverage. Overall, the Medicare database covers 96% of the total population older than 65 years old. The Committee for the Protection of Human Subjects at the University of Texas Health Science Center at Houston approved this study.

Study Design and Study Population

This is a retrospective cohort study. We assessed 354,701 men from the SEER-Medicare linked database, who were 66 years or older and diagnosed with prostate cancer between January 1992 and December 2009 for eligibility. Patients who were 65 years old at cancer diagnosis are excluded, as their medical claims 1 year before cancer diagnosis were not available in the Medicare data to assess baseline comorbidity. Patients who had prostate cancer identified by autopsy or in the death certificate, died within 6 months of diagnosis, had concurrent secondary cancer, had race/ethnicity or month of diagnosis missing, or received any form of hormone therapy before cancer diagnosis are omitted from the analysis (Fig. 1). We also exclude men who are not continuously enrolled in both part A and part B, or enroll in an HMO 1 year before and 1 year after diagnosis because their claims may not be processed through Medicare. The final cohort includes 12,170 patients with prostate cancer stage III or IV treated with radiation therapy. We conduct descriptive analyses and logistic regression to examine the disparities in receipt of ADT. The time to ADT initiation and duration of therapy are examined among 9373 patients who received at least one dose of LHRH agonists or orchiectomy after cancer diagnosis. Figure 1 is a flow chart showing our study exclusion and inclusion criteria.
Fig. 1

Flow chart of study population

Outcome Measures

The outcome measures are receipt of ADT, time from cancer diagnosis to ADT initiation, duration of ADT, and number of LHRH drug injections that patients received. A patient is classified as receiving ADT if they had a Medicare procedure code for LHRH drugs or orchiectomy (Table 1). The services from the outpatient bill are captured from Healthcare Common Procedure Coding System (HCPCS) codes for ADT injection and Current Procedural Terminology (CPT) and International Classification of Diseases ninth revision for procedures (ICD-9 procedure) codes for orchiectomy. Time to ADT initiation is calculated between diagnosis date and the first time the patient received ADT. Duration of LHRH agonists is identified as the last date a patient received LHRH injection minus the date a patient initiated the therapy. The number of LHRH agonist doses is determined by summarizing total LHRH claims after removing duplicate claims on the same day.
Table 1

Codes for defining treatment and procedures

Procedures

HCPCS

CPT code

ICD-9 procedure

ICD-9 diagnosis

Revenue centers

Leuprolide injection

J9217, J9218, J9219, J9150

    

Goserelin injection

J9202

    

Triptorelin pamoate

J3315

    

Degarelix

J9155

    

Orchiechtomy

 

54,520, 54,521, 54,522, 54,530, 54,535, 54,690, 49,510

62.3, 62.4, 62.41, 62.42

  

Radical prostatectomy

 

55,810–55,815, 55,840–55,845

60.5

  

Radiation therapy

77,401–77,499, 77,520–77,523, 77,750–77,799, G0256–G0261

 

92.21–92.29

V58.0, V66.1, V67.1

0330, 0333

Other treatments, including radiation and radical prostatectomy, are also ascertained from Medicare HCPCS, CPT, ICD-9 procedure, ICD-9 diagnosis, and revenue center codes [19]. Table 1 presents codes for defining treatments and procedures that patients received.

Race/Ethnicity, SES, and Geographic Location

The exposure variables include race/ethnicity, SES, and geographic location. Race is extracted from the SEER database (n = 12,107). Those with unknown or unspecified race from SEER data (n = 63) are then categorized by using Medicare race/ethnicity data. A patient’s race is categorized as non-Hispanic white (hereafter referred to as white), non-Hispanic black (hereafter referred to as black), Hispanic, and others. Other races include American Indians, Alaska Native, Asian, Pacific Islander, and patients with unspecified race/ethnicity.

SES is evaluated based on the median household income for census tract using the 1990 Census Bureau survey for patients diagnosed from 1992 to 1999 and the 2000 Census Bureau for those diagnosed after 2000. The median income is then categorized into quartiles.

Geographic location is identified from SEER areas, which are classified as Midwest, North East, South, and West. Midwest contains Detroit and Iowa; North East includes Connecticut and New Jersey; South region includes Georgia, Kentucky, and Louisiana; and West region includes San Francisco, San Jose, Los Angeles, Greater California, Hawaii, New Mexico, and Seattle.

Covariates

The other treatments patients received, patient demographics, and tumor characteristics are evaluated. Potential confounders include age at diagnosis, marital status, urban or rural area, comorbidity, prostate cancer stage, tumor grade, and other treatments (surgery and radiation) that patients received, which were explored in previous studies [20-24]. Age at diagnosis is categorized into 5-year increments. Marital status is defined as currently married at time of diagnosis, not married, or unknown. Urban area is defined as metropolitan counties and non-metropolitan counties with a population of 20,000 or more, and rural contained non-metropolitan counties with a population less than 20,000.

Charlson comorbidity at baseline is determined from Medicare inpatient and outpatient claims within 12 months before diagnosis. We use the algorithm recommended by NCI, which assessed 14 conditions including acute myocardial infarction, history of myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, chronic obstructive pulmonary disease, dementia, paralysis, any diabetes, chronic renal failure, any liver disease, ulcers, rheumatologic disease, and AIDS [20]. The comorbidity index is then weighted and categorized as 0, 1, 2, and ≥ 3. Tumor stage is identified using the American Joint Committee on Cancer (AJCC) stage third edition for cases diagnosed before 2003 and derived AJCC stage sixth edition for cases diagnosed in 2004 and after. Cancer stage is categorized into stages III or IV. Tumor grade is categorized as “well differentiated,” “moderately differentiated,” “poorly differentiated,” and “undifferentiated or unknown.”

Prostate cancer treatment is classified as radiation or the combination of radical prostatectomy and radiation therapy. Hormone therapy is referred as ADT by either orchiectomy or LHRH agonists.

Statistical Analysis

We examine the baseline characteristics of study population across racial/ethnical groups using the chi-squared test. We also use the chi-squared test to compare the receipt of ADT with respect to patient’s race/ethnicity, SES, and SEER areas. Fisher’s exact test is used if the large-sample assumption was violated for the chi-squared test. Multivariable logistic regression is used to explore the relationship between receipt of ADT and patients’ race/ethnicity, SES, and SEER areas, while controlling for relevant covariates. The precision of odds ratio (OR) estimates is evaluated by calculating the corresponding 95% CIs.

The distributions of time to ADT initiation, duration of ADT, and number of LHRH injections are explored using the Kruskal-Wallis test. We use linear regression models to examine the factors associated with time to ADT initiation and duration of ADT while controlling for baseline characteristics. We model the time to events, specifically, time from prostate cancer diagnosis to the first LHRH drug injection or orchiectomy surgery and time from first LHRH injection to the last one. The outcomes on time to ADT initiation and duration of ADT are transformed on log scale to satisfy the normality assumption in linear regression models. A p < 0.05 is considered significant for all analyses. SAS software version 9.4 (SAS Institute, Cary, NC) is utilized for all analyses [25].

Results

Baseline Characteristics

We identified 12,170 eligible men diagnosed with stage III or IV prostate cancer from 1991 to 2009 who underwent radiation therapy for inclusion into the study. Table 2 shows types of treatment, demographic characteristics, and tumor characteristics with respect to patient race and ethnicity. Substantial differences by race/ethnicity are observed in all baseline characteristics (p < 0.01). A significantly higher proportion of black and Hispanic men are 66–69 years old (40.2 and 34.9%, respectively). Black men are more likely to be unmarried at the time they were diagnosed with prostate cancer (41.7% compared to 16.5 to 25.7% in other racial/ethnic groups). A significantly higher proportion of black and Hispanic men are among the lowest-income group compared to white men or other races (59.5% for black and 39.2% for Hispanic men vs. 19.1–20.7% for other races and white men, respectively). High coexisting comorbidity (Charlson index of 2 or higher) is more common among black and Hispanic patients.
Table 2

Demographic and clinical characteristics of the study patients by race/ethnicity (n = 12,170)

Variables

Non-Hispanic white (n = 9857)

Non-Hispanic black (n = 1043)

Hispanic (n = 657)

Othersa (n = 613)

n

%

n

%

n

%

n

%

Age

 66–69

3127

31.7

419

40.2

229

34.9

164

26.8

 70–74

3488

35.4

340

32.6

236

35.9

199

32.5

 75–79

1904

19.3

186

17.8

117

17.8

159

25.9

 80+

1338

13.6

98

9.4

75

11.4

91

14.9

Marital status

 Unmarried

1845

18.7

435

41.7

169

25.7

101

16.5

 Married

7568

76.8

553

53.0

464

70.6

492

80.3

 Unknown

444

4.5

55

5.3

24

3.7

20

3.3

SES census tract median income

 First low income

2025

20.7

619

59.5

255

39.2

106

19.1

 Second

2481

25.4

226

21.7

168

25.8

131

23.7

 Third

2586

26.5

122

11.7

150

23.0

144

26.0

 Fourth high income

2680

27.4

73

7.0

78

12.0

173

31.2

Comorbidity

 0

7799

79.1

745

71.4

481

73.2

455

74.2

 1

1440

14.6

192

18.4

106

16.1

109

17.8

 2

394

4.0

73

7.0

36

5.5

33

5.4

 ≥ 3

224

2.3

33

3.2

34

5.2

16

2.6

Prostate cancer stage

 Stage III

5412

54.9

441

42.3

300

45.7

281

45.8

 Stage IV

4445

45.1

602

57.7

357

54.3

332

54.2

Tumor grade

 Well or moderately differentiated

3365

34.2

311

29.8

226

34.4

165

27.0

 Poorly differentiated

5762

58.5

623

59.7

390

59.4

397

64.8

 Undifferentiated or unknown

730

7.4

109

10.5

41

6.2

51

8.3

Types of treatment

 Radiation

926

9.4

166

15.9

50

7.6

82

13.4

 Radiation + surgery

1326

13.5

107

10.3

77

11.7

63

10.3

 ADT + radiation

5714

58.0

661

63.4

416

63.3

373

60.9

 ADT + radiation + surgery

1891

19.2

109

10.5

114

17.4

95

15.5

All p values were < 0.01

aOther races included American Indians, Alaska Native, Asian, Pacific Islander, and patients with unspecified race/ethnicity

Tumor clinical characteristics also varied across racial/ethnic groups. Black men report the largest proportion with stage IV at diagnosis (57.7%), followed by Hispanic and men of other races (approximately 54%). Poorly differentiated tumor grade accounts for the highest percentage among men of other races (64.8%).

Among a total 12,170 patients, 9373 (77%) received at least one form of hormone in combination with radiation therapy. Table 2 shows the aggressive combination of surgery combined with radiation and hormone therapy account for the highest fraction among white men (19.2% compared to 10.5–17.4% in other three racial groups).

Receipt of ADT

We conduct descriptive analysis and logistic regression to examine the disparity in receipt of ADT. The analysis included all 12,170 eligible men, who were candidates for the therapy. Table 3 describes the receipt of hormone therapy among eligible patients by race/ethnicity, SES, and SEER areas. Among 12,170 eligible patients, 9373 (77%) received hormone therapy. Significant differences are detected in race/ethnicity and in SEER areas (p values < 0.05). Significantly higher proportions of eligible men who received any form of hormone therapy are Hispanic (80.7%) and live in the North East region (80.3%). Compared to other race/ethnicity groups, black men are less likely to receive ADT (73.8%). The use of adjuvant hormone therapy does not vary across socioeconomic groups (p = 0.64).
Table 3

Receipt of ADT stratified by race/ethnicity, SES, and SEER areas (n = 12,170)

 

No. of patients with advanced prostate cancer and underwent radiation (n = 12,170)

Patients underwent orchiectomy (n = 1,468)

Patients received LHRH drugs (n = 8,371)

Patients who received any form of ADT (n = 9,373)

n

%

p

%

p

%

p

Race

 Non-Hispanic white

9,857

11.8

0.2

69.2

< 0.01

77.2

< 0.01

 Non-Hispanic black

1,043

13.6

 

63.5

 

73.8

 

 Hispanic

657

12.9

 

72.5

 

80.7

 

 Others

613

13.4

 

67.2

 

76.4

 

SES (income)

 First low income

3,005

20.6

< 0.01

61.7

< 0.01

76.3

0.64

 Second

3,006

12.7

 

68.8

 

77.6

 

 Third

3,002

9.5

 

70.6

 

76.7

 

 Fourth high income

3,004

5.5

 

73.6

 

77.0

 

SEER areas

 Midwest

2,608

16.0

< 0.01

65.3

< 0.01

76.5

< 0.01

 North East

1,837

8.6

 

74.2

 

80.3

 

 South

1,383

8.5

 

70.0

 

76.2

 

 West

6,342

12.2

 

68.4

 

76.5

 

Of the patients, 466 received both orchiectomy and LHRH drugs

Regarding the type of hormone therapy, men with low income and who lived in the Midwest are more likely to choose orchiectomy. LHRH drugs are more common among Hispanic men, men with high income, and men who resided in the North East of the country.

Table 4 describes the associations between receipt of ADT and patients’ race/ethnicity, SES, and SEER areas while controlling for demographic and clinical characteristics. Race/ethnicity is significantly associated with the receipt of ADT among prostate cancer patients who underwent radiation therapy (p < 0.01). Compared to white men, black men and men of other races are less likely to receive hormone therapy (OR = 0.64 and 0.74, respectively). SES and SEER area are not related to whether eligible patients receive ADT or not. Other factors related to not initiating ADT appropriately are the combination of radiation and surgery (OR = 0.3, 95% CI 0.27–0.33) and living in a rural area (OR = 0.8, 95% CI 0.67–0.95). Tumor stage IV and moderately or poorly differentiated tumor grade appear to significantly increase the likelihood of receiving ADT. In fact, patients with cancer stage IV are 2.40 times more likely to receive ADT than those diagnosed with stage III (OR = 2.40, 95% CI 2.16–2.67). Poor tumor grade is associated with ADT initiation (OR = 2.09, 95% CI 1.55–2.83 for moderately differentiated tumor grade and OR = 4.58, 95% CI 3.37–6.22 for poorly differentiated tumor grade).
Table 4

The associations between receipt of ADT and race/ethnicity, SES, and SEER areas, controlling for demographic and clinical characteristics (n = 12,170)

Variables

Receipt of ADT

OR

95% CI

p value

Race

 Non-Hispanic white

1.00

< 0.01

 Non-Hispanic black

0.64

0.54

0.76

 

 Hispanic

1.13

0.90

1.41

 

 Others

0.74

0.59

0.93

 

SES (income)

 First low income

1.00

0.85

 Second

1.05

0.92

1.20

 

 Third

1.05

0.91

1.21

 

 Fourth high income

1.01

0.87

1.18

 

SEER areas

 Midwest

1.00

0.57

 North East

1.12

0.95

1.33

 

 South

1.00

0.84

1.19

 

 West

1.05

0.92

1.19

 

Radiation and surgery

 Radiation

1.00

< 0.01

 Radiation + surgery

0.30

0.27

0.33

 

Age

 66–69

1.00

0.14

 70–74

0.96

0.87

1.07

 

 75–79

1.11

0.96

1.28

 

 80+

1.13

0.94

1.37

 

Marital status

 Unmarried

1.00

0.20

 Married

1.11

0.99

1.26

 

 Unknown

1.03

0.80

1.32

 

Year of diagnosis

1.01

1.00

1.02

0.32

Urban/rural

 Urban

1.00

0.01

 Rural

0.80

0.67

0.95

 

Comorbidity

 0

1.00

0.73

 1

0.93

0.82

1.06

 

 2

1.00

0.78

1.27

 

 ≥ 3

0.94

0.68

1.30

 

Prostate cancer stage

 Stage III

1.00

< 0.01

 Stage IV

2.40

2.16

2.67

 

Tumor grade

 Well differentiated

1.00

< 0.01

 Moderately differentiated

2.09

1.55

2.83

 

 Poorly differentiated

4.58

3.37

6.22

 

 Undifferentiated or unknown

3.19

2.22

4.60

 

Time to Initiation and Duration of ADT

Overall, the median time from cancer diagnosis to ADT initiation is 2 months with interquartile range from 1.1 to 7.0 months. Once ADT is initiated, men receive 19.3 months (median) with interquartile range from 6.7 to 43.5 months. Patients receive a median of seven LHRH injections. Table 5 presents the univariate analysis, and Table 6 summarizes the results from the linear regression models examining time to ADT initiation and ADT duration across race/ethnicity, SES, and SEER areas.
Table 5

Time from cancer diagnosis to ADT initiation, duration, and number of LHRH doses among patients with advanced prostate cancer and who underwent adjuvant ADT and radiation (median and IQR) (n = 9373)

Variables

Time to initiation (IQR, 1.1–7.0 months)

Duration of ADT (IQR, 6.7–43.5 months)

Number of doses (IQR, 3–13 doses)

Median

IQR

p value

Median

IQR

p value

Median

IQR

p value

Race

 Non-Hispanic white (n = 7605)

2.0

1.1–7.7

0.75

19.4

6.7–43.7

0.92

7

3–13

0.78

 Non-Hispanic black (n = 770)

2.2

1.1–6.0

 

19.7

6.7–41.3

 

7

3–13

 

 Hispanic (n = 530)

2.00

1.2–5.5

 

17.5

7.4–40.9

 

7

3–14

 

 Others (n = 468)

1.9

1.1–5.7

 

18.8

6.2–43.4

 

8

3–13

 

SES (income quartile)

 First low income

2.0

1.1–11.7

0.42

20.9

6.6–49.3

< 0.01

8

3–16

< 0.01

 Second

2.0

1.1–6.5

 

19.5

7.3–41.1

 

7

3–13

 

 Third

2.0

1.2–6.4

 

20.1

7.5–44.0

 

7

3–13

 

 Fourth high income

2.0

1.2–5.7

 

16.7

6.2–36.8

 

6

3–11

 

SEER areas

 Midwest (n = 1994)

2.1

1.1–12.9

< 0.01

23.4

8.0–50.8

< 0.01

8

3–15

< 0.01

 North East (n = 1475)

2.0

1.2–5.2

 

18.1

7.4–38.5

 

7

3–12

 

 South (n = 1054)

1.8

1.1–4.7

 

18.6

7.8–36.0

 

6

3–10

 

 West (n = 4850)

2.0

1.1–7.5

 

18.4

6.1–43.7

 

7

3–13

 

Kruskal-Wallis test is used as the time to initiation and duration of ADT are not normally distributed

Table 6

Linear regression models examining the time from diagnosis to ADT initiation and time from the first ADT injection to the last one among patients with advanced disease and who underwent ADT + radiation

Variables

Time to initiation of ADT

Duration of ADT

 

Estimates

Standard error

p value

Estimates

Standard error

p value

Race

 Non-Hispanic white

0.00

 

0.00

 

 Non-Hispanic black

0.11

0.06

0.06

− 0.03

0.05

0.57

 Hispanic

0.04

0.06

0.52

0.00

0.06

0.99

 Others

0.15

0.07

0.03

0.10

0.07

0.12

SES (income quartile)

 First low income

0.00

 

0.00

 

 Second

− 0.02

0.04

0.67

0.01

0.04

0.81

 Third

− 0.03

0.05

0.53

0.09

0.04

0.06

 Fourth high income

0.02

0.05

0.66

− 0.02

0.05

0.64

SEER areas

 South

0.00

  

 

 Midwest

0.14

0.04

0.05

0.00

0.05

1.00

 North East

0.12

0.05

0.01

− 0.06

0.05

0.28

 West

0.02

0.03

0.73

− 0.06

0.05

0.21

The variables patient’s age, year of diagnosis, urban/rural, comorbidity, cancer stage, and tumor grade are controlled in the models. Log transformation is applied to satisfy the normality assumption

The univariate analysis shows that time to initiation and duration of ADT do not vary across race/ethnicity groups. Substantial differences are observed in SES and SEER areas (p < 0.05). More specifically, men with income in the highest quartile report the shortest duration of ADT and the fewest doses. Southern men are more likely to start ADT sooner, and men in the Midwest underwent a longer period of hormone therapy.

Using linear regression models, we examine two outcomes, including the time from diagnosis to ADT initiation (days) and time from the first ADT injection to the last one (duration of ADT in days). Both outcomes are log-transformed, and results are presented in Table 6. After controlling for demographic and clinical baseline characteristics, race and SEER area are significantly associated with time to ADT initiation. Men living in the North East and Midwest start ADT later than those living in the South (p = 0.01 and 0.05, respectively). White men and Hispanic men initiate ADT earlier than black men and those of other races. SES is not related to time to ADT initiation while controlling for covariates. All three factors are not statistically associated with duration of ADT.

Discussion

ADT use in our study population shows unique patterns across racial/ethnic groups and SEER areas after controlling for baseline characteristics. Overall, 77% of high-risk prostate cancer patients receive hormone therapy, but still, 23% of them do not initiate the therapy. Race/ethnicity is associated with the initiation of ADT, where black men and men of other races are least likely to initiate ADT compared to white and Hispanic men. People living in the North East appear to start adjuvant hormone therapy earlier than those living in other parts of the country. Overall, men initiate ADT 2 months (median) after diagnosis and complete 19 months of therapy with a median of seven doses.

Racial disparities in receipt of ADT are consistent with previous studies [18, 21-24], though the magnitudes differ. It is previously reported that black men are less likely than white men to receive ADT after diagnosis [18, 21]. The relative inconsistency across studies is attributable to the differences in sample characteristics. Holmes et al. examined the use of ADT among all prostate cancer patients and included those with low-risk tumors, and reported 24% initiation of ADT in black men relative to 27% in Caucasians men [21]. Carson et al. evaluate patients with metastatic cancer only and describe slightly lower percentages of men who initiated ADT compared to our findings [18]. Specifically, 61.2% of black men and 74.5% of white men received ADT [18]. Both studies used SEER-Medicare linked database. Our study focus on using ADT as adjuvant therapy combined with radiation therapy in patients with stage III or IV prostate cancer. Although the study cohort may not be representative of the general elderly population, the study examine a group of patients, where ADT was considered as the most appropriate therapy [26] and the survival benefits are well reported in clinical trials [27-30]. Though ADT use differ among racial groups, previous studies confirmed that the receipt of hormone therapy did not contribute to racial differences in survival [21, 22]. It is unclear why blacks are less likely to receive ADT. A prospective study may be necessary to identify the patient barriers to care. Out data exclude the contribution of SES in ADT disparity. Other potential reasons include service availability, patient knowledge, preference, or concern about hormone-related side effects.

Another feature that distinguished our study from previous published research is the examination of disparities in SES and SEER areas. There is little evidence about the contribution of demographic factors in the consideration for early or deferred hormone therapy. After controlling for baseline characteristics, SES and SEER area are not associated with the receipt of ADT, but SEER area is associated with timing from cancer diagnosis to ADT initiation. More specifically, men with advanced disease in the North East and Midwest SEER areas start adjuvant hormone therapy significantly later than those in the South region. Although reasons for geographic variations are not clear in our study and in other studies [23, 24], the results may reflect provider behavior and/or patient choice. Although SEER area identifies the patient’s place of residence, it does not identify the area of the treating provider who may be outside the SEER area. The advantages of early treatment with hormone therapy varied in clinical trials, strongly depending on cancer stage. A systematic review reported that immediate treatment with hormone therapy among patients with locally advanced prostate cancer would generate important reductions in mortality and disease progression [10]. More specifically, delaying hormone therapy among men with high-risk disease would increase the risk of metastasis and perhaps the risk of mortality.

The duration of hormone therapy is controversial. Intermittent hormone therapy is proposed to ameliorate side effects of long-term therapy. If adopted, this will affect duration of therapy. In our study, we examine the time between the first and last LHRH injections and the number of doses, which would provide a good picture of how long and how much ADT patients received. Though SES and SEER area are found to be associated with duration of ADT in our bivariate analysis, these associations are diminished in the stratified regression models.

The strength of this study is its use of the SEER-Medicare linked database to assess the racial disparity in receipt and utilization of hormone therapy. SEER-Medicare is a large population-based database that is well suited to study patient access to care. The large sample size also allow for sub-group analyses among patients who benefited the most from hormone therapy. Demographic information from SEER and Medicare data including race/ethnicity, SES, and SEER area are considered reliable and valid. However, a limitation of our study is the inability to evaluate oral intake of hormone therapy. In this study, ADT is defined as either orchiectomy or LHRH drug injections in outpatient settings. The use of oral forms of hormone therapy is not billed or covered by Medicare, and thus cannot be included in the analysis. Medicare covers the cost of surgical castration and LHRH injections, but may not cover expensive androgen blockade agents (CBA) or antiandrogens. Another limitation is the use of median income for census tracts as a proxy of SES. The income for census tracts is not able to capture the change of patients’ SES over time. It is a measure at the group level and may not be as precise as individual level measure of socioeconomic status; hence, it is vulnerable to ecological fallacy bias.

In conclusion, this study confirms substantial racial disparities in adjuvant hormone therapy for men with advanced prostate cancer. The data also indicate that SES is not associated with early initiation and duration of therapy after controlling for relevant covariates. Patients in the South of the country start therapy earlier than men in the North East and Midwest regions. Additional research should examine the reasons fewer black men receive hormone therapy, including access to care and treatment services, patient choice or knowledge, and service availability. Further study of providers is necessary to evaluate if geographic differences in hormone therapy are related to provider characteristics and preferences or to patient factors.

Notes

Acknowledgements

We acknowledge the efforts of the National Cancer Institute; Center for Medicare and Medicaid Services; Information Management Services, Inc.; and the Surveillance, Epidemiology, and End Results Program tumor registries in the creation of this database. The interpretation and reporting of these data are the sole responsibilities of the authors.

Funding Information

This research is supported in part by the Cancer Prevention Research Institute of Texas (grant nos. RP130051 and RP170668).

Compliance with Ethical Standards

Conflict of Interest

The authors declare that there are no conflicts of interest.

References

  1. 1.
    American Cancer Society. Cancer facts & figures 2018. Atlanta: American Cancer Society; 2018.Google Scholar
  2. 2.
    Howlader N, N.A., Krapcho M, Miller D, Bishop K, Altekruse SF, Kosary CL, Yu M, Ruhl J, Tatalovich Z, Mariotto A, Lewis DR, Chen HS, Feuer EJ, Cronin KA, SEER Cancer Statistics Review, 1975-2013. National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/csr/1975_2013/, based on November 2015 SEER data submission, posted to the SEER web site, April 2016., 2015.
  3. 3.
    Golabek T, Belsey J, Drewa T, Kołodziej A, Skoneczna I, Milecki P, et al. Evidence-based recommendations on androgen deprivation therapy for localized and advanced prostate cancer. Cent European J Urol. 2016;69(2):131–8.Google Scholar
  4. 4.
    Shahinian VB, Kuo YF, Gilbert SM. Reimbursement policy and androgen-deprivation therapy for prostate cancer. N Engl J Med. 2010;363(19):1822–32.CrossRefGoogle Scholar
  5. 5.
    Sasse AD, Sasse E, Carvalho AM, Macedo LT. Androgenic suppression combined with radiotherapy for the treatment of prostate adenocarcinoma: a systematic review. BMC Cancer. 2012;12:54.CrossRefGoogle Scholar
  6. 6.
    Kauffmann G, Liauw SL. The use of hormonal therapy to augment radiation therapy in prostate Cancer: an update. Curr Urol Rep. 2017;18(7):50.CrossRefGoogle Scholar
  7. 7.
    Holmes L Jr, et al. Effectiveness of androgen deprivation therapy in prolonging survival of older men treated for locoregional prostate cancer. Prostate Cancer Prostatic Dis. 2007;10(4):388–95.CrossRefGoogle Scholar
  8. 8.
    Liu J, Shi L, Sartor O, Culbertson R. Androgen-deprivation therapy versus radical prostatectomy as monotherapy among clinically localized prostate cancer patients. Onco Targets Ther. 2013;6:725–32.CrossRefGoogle Scholar
  9. 9.
    Studer UE, Whelan P, Albrecht W, Casselman J, de Reijke T, Hauri D, et al. Immediate or deferred androgen deprivation for patients with prostate cancer not suitable for local treatment with curative intent: European Organisation for Research and Treatment of Cancer (EORTC) trial 30891. J Clin Oncol. 2006;24(12):1868–76.Google Scholar
  10. 10.
    Prezioso D, Iacono F, Romeo G, Ruffo A, Russo N, Illiano E. Early versus delayed hormonal treatment in locally advanced or asymptomatic metastatic prostatic cancer patient dilemma. World J Urol. 2014;32(3):661–7.CrossRefGoogle Scholar
  11. 11.
    Horwitz EM, Bae K, Hanks GE, Porter A, Grignon DJ, Brereton HD, et al. Ten-year follow-up of radiation therapy oncology group protocol 92-02: a phase III trial of the duration of elective androgen deprivation in locally advanced prostate cancer. J Clin Oncol. 2008;26(15):2497–504.Google Scholar
  12. 12.
    Bolla M, de Reijke TM, van Tienhoven G, van den Bergh A, Oddens J, Poortmans PM, et al. Duration of androgen suppression in the treatment of prostate cancer. N Engl J Med. 2009;360(24):2516–27.Google Scholar
  13. 13.
    Barocas DA, Penson DF. Racial variation in the pattern and quality of care for prostate cancer in the USA: mind the gap. BJU Int. 2010;106(3):322–8.CrossRefGoogle Scholar
  14. 14.
    Moses KA, Orom H, Brasel A, Gaddy J, Underwood W III. Racial/ethnic disparity in treatment for prostate cancer: does cancer severity matter? Urology. 2017;99:76–83.CrossRefGoogle Scholar
  15. 15.
    Hoffman RM, Harlan LC, Klabunde CN, Gilliland FD, Stephenson RA, Hunt WC, et al. Racial differences in initial treatment for clinically localized prostate cancer. Results from the prostate cancer outcomes study. J Gen Intern Med. 2003;18(10):845–53.Google Scholar
  16. 16.
    Presley CJ, Raldow AC, Cramer LD, Soulos PR, Long JB, Yu JB, et al. A new approach to understanding racial disparities in prostate cancer treatment. J Geriatr Oncol. 2013;4(1):1–8.Google Scholar
  17. 17.
    Lyratzopoulos G, Barbiere JM, Greenberg DC, Wright KA, Neal DE. Population based time trends and socioeconomic variation in use of radiotherapy and radical surgery for prostate cancer in a UK region: continuous survey. BMJ. 2010;340:c1928.CrossRefGoogle Scholar
  18. 18.
    Carson AP, Howard DL, Carpenter WR, Taylor YJ, Peacock S, Schenck AP, et al. Trends and racial differences in the use of androgen deprivation therapy for metastatic prostate cancer. J Pain Symptom Manag. 2010;39(5):872–81.Google Scholar
  19. 19.
    Shahinian VB, Kuo YF, Freeman JL, Goodwin JS. Risk of fracture after androgen deprivation for prostate cancer. N Engl J Med. 2005;352(2):154–64.CrossRefGoogle Scholar
  20. 20.
    Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. J Clin Epidemiol. 2000;53(12):1258–67.CrossRefGoogle Scholar
  21. 21.
    Holmes L Jr, Chan W, Jiang Z, Ward D, Essien EJ, du XL. Impact of androgen deprivation therapy on racial/ethnic disparities in the survival of older men treated for locoregional prostate cancer. Cancer Control. 2009;16(2):176–85.CrossRefGoogle Scholar
  22. 22.
    Keating NL, O’Malley AJ, McNaughton-Collins M, Oh WK, Smith MR. Use of androgen deprivation therapy for metastatic prostate cancer in older men. BJU Int. 2008;101(9):1077–83.CrossRefGoogle Scholar
  23. 23.
    Mettlin CJ, Murphy GP, Cunningham MP, Menck HR. The National Cancer Data Base report on race, age, and region variations in prostate cancer treatment. Cancer. 1997;80(7):1261–6.CrossRefGoogle Scholar
  24. 24.
    Harlan LC, Potosky A, Gilliland FD, Hoffman R, Albertsen PC, Hamilton AS, et al. Factors associated with initial therapy for clinically localized prostate cancer: prostate cancer outcomes study. J Natl Cancer Inst. 2001;93(24):1864–71.Google Scholar
  25. 25.
    Kleinbaum DG, Klein M. Survival analysis: a self-learning text. Third ed; 2011.Google Scholar
  26. 26.
    National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology (NCCN Guidelines) Prostate Cancer Version 1. 2016. at http://www.nccn.org/professionals/physician_gls/pdf/prostate.pdf, accessed on 2/2017.
  27. 27.
    Jones CU, Hunt D, McGowan DG, Amin MB, Chetner MP, Bruner DW, et al. Radiotherapy and short-term androgen deprivation for localized prostate cancer. N Engl J Med. 2011;365(2):107–18.Google Scholar
  28. 28.
    Bolla M, van Tienhoven G, Warde P, Dubois JB, Mirimanoff RO, Storme G, et al. External irradiation with or without long-term androgen suppression for prostate cancer with high metastatic risk: 10-year results of an EORTC randomised study. Lancet Oncol. 2010;11(11):1066–73.Google Scholar
  29. 29.
    Mottet N, van Damme J, Loulidi S, Russel C, Leitenberger A, Wolff JM, et al. Intermittent hormonal therapy in the treatment of metastatic prostate cancer: a randomized trial. BJU Int. 2012;110(9):1262–9.Google Scholar
  30. 30.
    Warde P, Mason M, Ding K, Kirkbride P, Brundage M, Cowan R, et al. Combined androgen deprivation therapy and radiation therapy for locally advanced prostate cancer: a randomised, phase 3 trial. Lancet. 2011;378(9809):2104–11.Google Scholar

Copyright information

© W. Montague Cobb-NMA Health Institute 2018

Authors and Affiliations

  • Chi Nguyen
    • 1
  • David R. Lairson
    • 2
  • Michael D. Swartz
    • 3
  • Xianglin L. Du
    • 1
    Email author
  1. 1.Department of Epidemiology, Human Genetics, and Environmental Science, University of Texas School of Public HealthUniversity of Texas Health Science Center in HoustonHoustonUSA
  2. 2.Department of Management Policy and Community Health, School of Public HealthUniversity of Texas Health Science Center in HoustonHoustonUSA
  3. 3.Department of Biostatistics and Data Science, School of Public HealthUniversity of Texas Health Science Center in HoustonHoustonUSA

Personalised recommendations