Abstract

Background

Racial disparity has been investigated in a number of cancers; however, there remains a comparative paucity of data in Hodgkin's lymphoma (HL).

Patients and methods

We examined time-, age-, and gender-specific incidence, disease characteristics, and survival across and within races for adolescent/adult HL (age 10–79 years) diagnosed during 1992–2007 in the SEER 13 registries.

Results

A total of 15 662 HL cases were identified [11 211 non-Hispanic whites, 2067 Hispanics, 1662 blacks, and 722 Asian/Pacific Islanders (A/PI)]. Similar to whites, A/PIs had bimodal age-specific incidence, while blacks and Hispanics did not. Further, HL was significantly more common in Hispanics versus whites age >65 years (7.0/1 × 106 versus 4.5/1 × 106, respectively, P <0.01). By place of birth, US-born Hispanics and A/PIs age 20–39 years had higher incidence of HL versus their foreign-born counterparts (P <0.05), however, rates converged age >40 years. Interestingly, from 1992–1997 to 2003–2007, A/PI incidence rates increased >50% (P <0.001). Moreover, this increase was restricted to US-born A/PI. We also identified a number of disease-related differences based on race. Finally, 5-, 10-, and 15-year overall survival rates were inferior for blacks and Hispanics compared with whites (P <0.005 and P <0.001, respectively) and A/PI (P <0.018 and P <0.001, respectively). These differences persisted on multivariate analysis.

Conclusion

Collectively, we identified multiple racial disparities, including survival, in adolescent/adult HL.

introduction

Racial disparity has been investigated in a number of cancers [1–10]. Many studies have noted significant differences in incidence rates, patient and disease-related characteristics, and/or survival based on race. However, population-based survival analyses of racial disparities have been reported mostly in solid tumors (e.g. breast, colon, prostate, and lung) with recent reports in non-Hodgkin's lymphoma (HL) [6] and multiple myeloma [10]. A comparative paucity of data are available in adult HL.

In pediatric HL, Metzger et al. [11] found that disease-related features and clinical characteristics did not differ between races (white and black children). Furthermore, black children with HL had lower event-free survival (EFS) compared with white children, while both populations had similar 5-year overall survival (OS). Studies examining racial disparities in adult HL including age-specific HL incidence patterns have been reported, however they have been smaller scope (statewide), did not encompass all races (i.e. including Hispanics and Asian/Pacific Islanders (A/PI)), and/or did not include analyses by place of birth [12, 13].

The overall incidence of HL varies greatly throughout the world. Pathogenesis of this geographic discrepancy of HL incidence is not known; however, environmental and lifestyle factors have been theorized as potential factors [14]. A British Columbia study of Chinese immigrants to Canada [15] and a San Francisco (California) analysis[16] reported that the incidence of HL among Asian immigrants was lower than expected from the background populations, although the rates of Canadian- or California-born Asians were higher than foreign born. However, minimal recent data are available regarding age-specific incidence rates for US- versus foreign-born A/PI including any potential differences in disease-related characteristics or survival.

The goals of the current project were to examine differences in HL age-specific incidence patterns, patient and disease-related characteristics, and survival across all major races within the United States including whites, Hispanics, blacks, and A/PIs. Furthermore, we investigated potential differences within Hispanics and A/PIs based on birthplace (i.e. US versus foreign born).

patients and methods

data sources

Incidence data and survival data were obtained from the US National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) program. The 13 registries in SEER-13 account for ∼14% of the US population, including 12% of the white population, 12% of black population, 35% of A/PI population, and 22% of the Hispanic population (see supplemental Methods, available at Annals of Oncology online) [17]. The current report examined the incidence of all HL diagnosed between 1 January 1992 and 31 December 2007. Native American and Alaskan natives were excluded from this analysis because the number of cases were too small for meaningful analyses [17]. Our analysis utilized data from the November 2009 SEER data submission, which was released April 2010 [18].

population estimates

The SEER program also links cases to the vintage 2008 intercensal population estimates, provided by the US Census Bureau. Population estimates are matched to cases by county, age group, race group, Hispanic/non-Hispanic group, gender, and year of diagnosis. While birthplace can be identified in the SEER registries, the corresponding population estimates are not readily available. Therefore, we multiplied SEER's population estimates by the proportion of the population that were US and foreign born in the 2000 Census, 5% Public Use Microdata Sample (PUMS) files [19], while taking into account gender, age, race, and geographic area of each subpopulation listed in the SEER population files. In addition, while SEER population files identified population estimates by year, there were no data sources to provide estimates of the proportion of the population that were US or foreign born for each year. We used the 5% sample of the Census 2000 to provide these proportions at the midpoint (year 2000) of the study period of interest (i.e. 1992–2007).

Census PUMS files provide the population estimates of Public Use Microdata Areas, which are areas that are not equivalent to a complete group of counties in the SEER registry areas. Thus, we excluded the Rural Georgia registry and Grays Harbor, a county in the Seattle-Pudget Sound registry, from the analysis that examined incidence of HL by birthplace.

statistical analysis

Age-specific incidence rates were computed for 10-year age groups ranging from age 10 to 79 years for each race group [non-Hispanics whites (referred to as whites), Hispanic whites (referred to as Hispanics), blacks, and A/PIs] and rates were all presented in terms of cases per 100 000 population. Characteristics for the patient population were also examined by Pearson's χ2 statistic. Incidence rates and 95% confidence intervals (CI) were calculated according to the algorithms provided by SEER [20]. Age-specific incidence rates were computed for 10-year age groups ranging from age 10 to 79 years. We also calculated incidence rates for major histological subtypes including mixed cellularity (MC) and nodular sclerosis (NS). The ‘Other’ group included lymphocyte-rich HL, lymphocyte-depleted HL, and nodular lymphocyte predominant HL. Incidence rates were compared by sex for each of four racial groups (non-Hispanics whites, Hispanics, blacks, and A/PIs). Incidence rates were also examined according to three periods 1992–1997, 1998–2002, and 2003–2007. Furthermore, we investigated incidence rates for Hispanic and A/PIs according to birthplace (i.e. US and foreign born).

Finally, we constructed life tables to show 5-, 10-, and 15-year survival rates by actuarial methods. Survival curves were constructed using the Kaplan–Meier method and compared with the log-rank test. Z-test was used to test the equivalence of relative survival at a given time. Furthermore, a multivariate Cox proportional hazards model was computed to estimate hazard ratios (HRs) and 95% CIs for the association of race with survival adjusting for age, sex, and stage. All statistics were computed using the STATA, MP 11.1 (StataCorp LP Copyright 2009) and National Cancer Institute SEER*Stat software, version 6.6.2.

results

incidence of HL according to race, age, and gender

Between 1992 and 2007, 16 783 cases of HL were reported to the SEER 13 registries. HL cases between the age of 10–79 years of white, black, Hispanic, or A/PI racial descent were included resulting in a final study cohort of 15 662 subjects. Table 1 describes the characteristics of patient populations by race. Interestingly, the mean age of HL among whites was significantly older compared with all other races; mean age was calculated versus median due to the prominent bimodal pattern of several races. Racial disparity in incidence is further illustrated through age-specific analyses (Figure 1). Whites showed a bimodal age-incidence pattern, while blacks had a much less clear bimodal distribution. A/PIs exhibited a bimodal pattern, although at all ages, A/PI have the lowest incidence rates of any race/ethnic group. Notably, incidence of HL in Hispanics was distinctly not bimodal with a small increase at age 20–29 years (2.3/100 000) followed by an exponential-like rise with peak incidence at age 70–79 years (5.1/1 × 106) (Figure 1D). Overall, whites had the highest incidence rates at age <65 years, while Hispanics had the highest incidence >65 years compared with all other races (P <0.05).

Table 1.

Subject and disease-related characteristics based on race

 White (n = 11211)
 
Black (n = 1662)
 
A/PI (n = 722)
 
Hispanic (n = 2067)
 
P
 
 Count Count Count Count W versus B W versus A W versus H 
Gender            
 Male 6096 54 865 52 386 53 1193 58 0.075 0.663 0.005 
 Female 5115 46 797 48 336 47 874 42    
Age            
 Mean 40  36.9  37.7  38.0  <0.001 0.002 <0.001 
 Interquartile range   26–52   25–46   23–50    23–51     
Histology            
 Nodular sclerosis 7284 65 919 55 461 64 1148 56 <0.001 0.788 <0.001 
 Mixed cellularity 1541 14 254 15 105 15 402 19    
 Other 2386 21 489 29 156 22 517 25    
Site            
 Extranodal 290 36 26 43 0.308 0.100 0.176 
Stage            
 I 2526 23 385 23 128 18 377 18 <0.001 0.012 <0.001 
 II 4503 40 548 33 320 44 690 33    
 III 2059 18 325 20 130 18 398 19    
 IV 1681 15 337 20 119 16 477 23    
 NA 442 67 25 125    
B symptoms            
 No 4337 39 533 32 323 45 714 35 <0.001 0.201 <0.001 
 Yes 3825 34 615 37 255 35 903 44    
 NA 3049 27 514 31 144 20 450 22    
 White (n = 11211)
 
Black (n = 1662)
 
A/PI (n = 722)
 
Hispanic (n = 2067)
 
P
 
 Count Count Count Count W versus B W versus A W versus H 
Gender            
 Male 6096 54 865 52 386 53 1193 58 0.075 0.663 0.005 
 Female 5115 46 797 48 336 47 874 42    
Age            
 Mean 40  36.9  37.7  38.0  <0.001 0.002 <0.001 
 Interquartile range   26–52   25–46   23–50    23–51     
Histology            
 Nodular sclerosis 7284 65 919 55 461 64 1148 56 <0.001 0.788 <0.001 
 Mixed cellularity 1541 14 254 15 105 15 402 19    
 Other 2386 21 489 29 156 22 517 25    
Site            
 Extranodal 290 36 26 43 0.308 0.100 0.176 
Stage            
 I 2526 23 385 23 128 18 377 18 <0.001 0.012 <0.001 
 II 4503 40 548 33 320 44 690 33    
 III 2059 18 325 20 130 18 398 19    
 IV 1681 15 337 20 119 16 477 23    
 NA 442 67 25 125    
B symptoms            
 No 4337 39 533 32 323 45 714 35 <0.001 0.201 <0.001 
 Yes 3825 34 615 37 255 35 903 44    
 NA 3049 27 514 31 144 20 450 22    

A/PI, Asian/Pacific Islander; B, black; H, Hispanic; NA, not available; W, white.

Figure 1

Age-specific incidence of Hodgkin's lymphoma by race. Data shown are age-specific incidence rates for 10-year age groups ranging from age 10 to 79 years for each race (non-Hispanics whites (referred to as: whites), Hispanic whites (referred to as: Hispanics), blacks, and A/PIs). Rates are presented in terms of cases per 100 000 population. (A) Whites showed a continued bimodal age-incidence pattern, while (B) blacks had a much less clear bimodal distribution. (C) A/PIs exhibited a bimodal pattern and have the lowest incidence rates of any race/ethnic group. (D) Age-specific incidence in Hispanics was distinctly not bimodal with a small increase at age 20–29 followed by an exponential-like rise in incidence. A/PI, Asian/Pacific Islander.

Figure 1

Age-specific incidence of Hodgkin's lymphoma by race. Data shown are age-specific incidence rates for 10-year age groups ranging from age 10 to 79 years for each race (non-Hispanics whites (referred to as: whites), Hispanic whites (referred to as: Hispanics), blacks, and A/PIs). Rates are presented in terms of cases per 100 000 population. (A) Whites showed a continued bimodal age-incidence pattern, while (B) blacks had a much less clear bimodal distribution. (C) A/PIs exhibited a bimodal pattern and have the lowest incidence rates of any race/ethnic group. (D) Age-specific incidence in Hispanics was distinctly not bimodal with a small increase at age 20–29 followed by an exponential-like rise in incidence. A/PI, Asian/Pacific Islander.

According to gender, there were comparatively more males diagnosed with HL among Hispanics compared with whites (P = 0.005), while there was a trend of more females diagnosed with HL for blacks versus whites (P = 0.075) (Figure 2). Within all races, women had a significantly lower age-adjusted incidence rates than men (P <0.001) (Table 2); however, the male excess did not occur until after age 40–49 years. Within the white HL cohort, men and women had similar age specific rates until age 40, while after age 40, the incidence rate decreased for females compared with males; this difference persisted for all ages thereafter (P <0.05) (supplemental Table S1, available at Annals of Oncology online). Figure 2D shows that Hispanic females had a slight bimodal age pattern, although the lowest incidence was experienced at age 40–49 years rather than age 50–59 years compared with other races (P <0.05). Further, the incidence of HL in Hispanic males rose exponentially, and after the age of 40, male incidence was significantly higher than females (P <0.05).

Table 2.

Age-adjusted Hodgkin's lymphoma incidence rates based on race and gender

 White (95% CI) Black (95% CI) A/PI (95% CI) Hispanic (95% CI) 
Female incidence rates 3.44 (3.35–3.54) 2.72 (2.54–2.92) 1.15 (1.03–1.28) 2.06 (1.92–2.22) 
Male incidence rates 4.07 (3.97–4.17) 3.49 (3.25–3.73) 1.47 (1.33–1.63) 2.99 (2.80–3.19) 
Female versus male incidence ratio 0.85 (0.81–0.88) 0.78 (0.67–0.91) 0.69 (0.63–0.76) 0.69 (0.63–0.76) 
P <0.001 <0.001 0.001 <0.001 
 White (95% CI) Black (95% CI) A/PI (95% CI) Hispanic (95% CI) 
Female incidence rates 3.44 (3.35–3.54) 2.72 (2.54–2.92) 1.15 (1.03–1.28) 2.06 (1.92–2.22) 
Male incidence rates 4.07 (3.97–4.17) 3.49 (3.25–3.73) 1.47 (1.33–1.63) 2.99 (2.80–3.19) 
Female versus male incidence ratio 0.85 (0.81–0.88) 0.78 (0.67–0.91) 0.69 (0.63–0.76) 0.69 (0.63–0.76) 
P <0.001 <0.001 0.001 <0.001 

A/PI, Asian/Pacific Islander; CI, confidence interval.

Figure 2

Age-specific incidence of HL by race and gender. (A) HL in whites was more common in males versus females, while both genders exhibited a bimodal incidence pattern. (B) Black men had a rise in incidence at age 20--29 years followed by a relatively flat rate through age 70--79 years. (C) A/PIs had similar male to female ratios as whites as well as an apparent bimodal age distribution, while among (D) Hispanics, incidence rates rose exponentially; after the age of 40 years, the incidence of HL was significantly higher in Hispanic males versus females (P <0.05). A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma.

Figure 2

Age-specific incidence of HL by race and gender. (A) HL in whites was more common in males versus females, while both genders exhibited a bimodal incidence pattern. (B) Black men had a rise in incidence at age 20--29 years followed by a relatively flat rate through age 70--79 years. (C) A/PIs had similar male to female ratios as whites as well as an apparent bimodal age distribution, while among (D) Hispanics, incidence rates rose exponentially; after the age of 40 years, the incidence of HL was significantly higher in Hispanic males versus females (P <0.05). A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma.

patient and disease characteristics according to race

Table 1 denotes patient and disease-related characteristics based on race. Interestingly, early-stage HL (i.e. stage I/II) was more prevalent among whites (63%) and A/PI (62%) subjects versus blacks (56%) and Hispanics (52%) (P <0.0001). Overall (all races), men presented more commonly with advanced-stage disease compared with women (40% versus 30%, respectively, P <0.0001) (Table 3). This may in part explain the discrepancy in stage for Hispanics given the increased frequency of male gender compared with whites and A/PIs as described before. The gender difference would not explain, however, the early/advanced-stage discrepancy for blacks since there was a trend of higher prevalence of female HL versus Whites (Table 3). The gender-specific distribution differences in stage were most notable for black men versus women (stage I/II: black men 50% versus black women 62%, P <0.001) and white men versus women (stage I/II: white men 59% versus white women 67%, P <0.001).

Table 3.

Stage based on race and gender

Stage White
 
P Black
 
P A/PI
 
P Hispanic
 
P 
 Male (n = 6096)
 
Female (n = 5115)
 
 Male (n = 865)
 
Female (n = 797)
 
 Male (n = 386)
 
Female (n = 336)
 
 Male (n = 1193)
 
Female (n = 874)
 
 
 Count Count  Count Count  Count Count  Count Count  
1463 24 1063 21 <0.001 192 22 193 24 <0.001 69 18 59 18 <0.001 218 18 159 18 <0.001 
II 2137 35 2366 46  243 28 305 38  147 38 173 51  329 28 361 41  
III 1247 20 812 16  173 20 152 19  76 20 54 16  255 21 143 16  
IV 1021 17 660 13  223 26 114 14  82 21 37 11  319 27 158 18  
Not staged 228 214  34 33  12 13  72 53  
Stage White
 
P Black
 
P A/PI
 
P Hispanic
 
P 
 Male (n = 6096)
 
Female (n = 5115)
 
 Male (n = 865)
 
Female (n = 797)
 
 Male (n = 386)
 
Female (n = 336)
 
 Male (n = 1193)
 
Female (n = 874)
 
 
 Count Count  Count Count  Count Count  Count Count  
1463 24 1063 21 <0.001 192 22 193 24 <0.001 69 18 59 18 <0.001 218 18 159 18 <0.001 
II 2137 35 2366 46  243 28 305 38  147 38 173 51  329 28 361 41  
III 1247 20 812 16  173 20 152 19  76 20 54 16  255 21 143 16  
IV 1021 17 660 13  223 26 114 14  82 21 37 11  319 27 158 18  
Not staged 228 214  34 33  12 13  72 53  

P values compare differences in stage for male versus female within each race.

A/PI, Asian/Pacific Islander.

Racial differences were also noted with blacks and Hispanics having more frequent presence of B symptoms at diagnosis (Table 1). Interestingly, the presence of B symptoms followed a similar pattern as stage with men (all races) presenting more frequently with presence of B symptoms compared with women (all races) (39% and 32%, respectively, P <0.0001). This gender-specific disparity was notable within all races (supplemental Table S2, available at Annals of Oncology online). Differences in the presence of extranodal disease were not different across races (data not shown), however, the overall frequency of extranodal disease was low.

histologic subtype

The HL histologic tumor subtypes (cell type) also differed by race (Figure 3). Across races, Hispanics and blacks presented significantly less frequently with NS HL compared with whites (Table 1, P <0.001). Among Hispanics, the incidence of NS HL increased only slightly during age 20–29 years, decreased during age 30–59, and then rose slightly again after age 60. The incidence of all HL subtypes within the Hispanic population had an exponential increase after the age of 40 years, primarily due to the rapid rise in incidence of MC HL as well as an increased incidence in ‘other’ subtypes [age 20–29 years: 0.4 (95% CI 0.3–0.5), age 50–59 years: 1.0 (95% CI 0.8–1.3), age 70–79 years: 2.5 (95% CI 1.9–3.2)]. Gender-specific histologic differences are noted in supplemental Table S3 (available at Annals of Oncology online).

Figure 3

Age-specific incidence of HL by histology and race. Within the (A) white HL cohort, the young adult rise was mostly due to NS subtype, while the decreased incidence of NS in older whites was accompanied by an increased incidence of MC. Similar to whites, the young adult rise in incidence of (B) black and (C) A/PI populations were predominantly due to NS HL; the incidence of NS fell steadily after age 30 years. (D) Hispanics (and blacks) presented significantly less frequently with NS HL compared with whites, both presenting more commonly with MC subtype. A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma; MC, mixed cellularity; NS, nodular sclerosis.

Figure 3

Age-specific incidence of HL by histology and race. Within the (A) white HL cohort, the young adult rise was mostly due to NS subtype, while the decreased incidence of NS in older whites was accompanied by an increased incidence of MC. Similar to whites, the young adult rise in incidence of (B) black and (C) A/PI populations were predominantly due to NS HL; the incidence of NS fell steadily after age 30 years. (D) Hispanics (and blacks) presented significantly less frequently with NS HL compared with whites, both presenting more commonly with MC subtype. A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma; MC, mixed cellularity; NS, nodular sclerosis.

changes in HL incidence patterns from 1992 to 2007

Period-specific patterns of HL incidence based on race presented in Figure 4. During the 16-year period from 1992 to 2007, age-specific incidence patterns did not vary within races, except for A/PI. Within A/PI's, the incidence rates of HL successively increased within each period [1992–1997: 1.0 (95% CI 0.9–1.2), 1998–2002: 1.3 (95% CI 1.1–1.5), and 2003–2007: 1.6 (95% CI 1.4–1.7)] (supplemental Table S4, available at Annals of Oncology online). Moreover, among young A/PI adults aged 20–29 years, the incidence doubled from 1.5/100 000 (95% CI 1.2–2.0) during the years of 1992–1997 to an incidence of 3.2/100 000 (95% CI 2.6–3.8) in the latter years 2003–2007 (P <0.05).

Figure 4

Period changes in age-specific incidence of HL by race. (A–D) During the 16-year period from 1992 through 2007, age-specific incidence patterns did not vary within races except for A/PI. Within the A/PI HL population, incidence rates successively increased within each period. A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma.

Figure 4

Period changes in age-specific incidence of HL by race. (A–D) During the 16-year period from 1992 through 2007, age-specific incidence patterns did not vary within races except for A/PI. Within the A/PI HL population, incidence rates successively increased within each period. A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma.

incidence and HL characteristics by place of birth

Within Hispanic and A/PI populations, place of birth (i.e. US born versus foreign born) was a significant predictor in age-specific incidence. US-born A/PI age 20–49 years had a higher incidence of HL than foreign-born A/PIs (Figure 5A, P <0.05) with incidence of both populations converging after age 50 years. Similarly, US-born Hispanic population age 20–39 years had a significantly higher incidence of HL than foreign-born Hispanics of the same age (Figure 5B, P <0.05); and the incidences of both races converged after age 40 years. Interestingly, the relative increased incidence of HL in younger US-born Hispanics and A/PI populations consisted predominantly of NS subtype (supplemental Figure S1, available at Annals of Oncology online). In terms of period-specific HL incidence patterns based on birthplace, there was a trend toward significance for increased incidence of US-born A/PI aged 20–29 years [1992–1997: 1.9/100 000 (95% CI 1.2–2.8), 1998–2002: 2.2/100 000 (95% CI 1.4–3.3), and 2003–2007: 3.6/100 000 (95% CI 2.6–5.0)], while there were no differences among other age groups. Among Hispanics, there were no changes in HL incidence for both US- and foreign-born populations (Figure 5E and F).

Figure 5

Age-specific incidence and period changes by race and birthplace. (A) US-born A/PIs age 20–49 years had a higher incidence of HL than foreign-born A/PI of the same age (P <0.05) with incidence converging after age 50 years. (B) US-born Hispanics age 20–39 years also had a significantly higher incidence of HL than foreign-born Hispanics of the same age (P <0.05). (C–D) In terms of period-specific HL incidence patterns, there was a trend toward increased incidence of US-born A/PI age 20–29 years, while there were no differences among other age groups. (E–F) Among Hispanics, there were no differences in HL incidence for either US- and foreign-born populations. A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma.

Figure 5

Age-specific incidence and period changes by race and birthplace. (A) US-born A/PIs age 20–49 years had a higher incidence of HL than foreign-born A/PI of the same age (P <0.05) with incidence converging after age 50 years. (B) US-born Hispanics age 20–39 years also had a significantly higher incidence of HL than foreign-born Hispanics of the same age (P <0.05). (C–D) In terms of period-specific HL incidence patterns, there was a trend toward increased incidence of US-born A/PI age 20–29 years, while there were no differences among other age groups. (E–F) Among Hispanics, there were no differences in HL incidence for either US- and foreign-born populations. A/PI, Asian/Pacific Islander; HL, Hodgkin's lymphoma.

We analyzed disease characteristics and incidence patterns for A/PI's and Hispanics according to birthplace (Table 4). Significantly, more foreign-born Hispanics with HL were male compared with US-born Hispanics (64% versus 55%, respectively, P <0.003). Foreign-born Hispanics had significantly less frequent incidence of NS compared with US-born Hispanics (47% versus 60%, respectively, P <0.001); this was in part due to significantly more frequent MC subtype for foreign-born versus US-born Hispanics (25% versus 15%, respectively, P <0.0001). Foreign-born A/PI had less frequent NS and more common MC compared with US-born A/PI, however, these differences were not significant.

Table 4.

Hispanic and A/PI disease-related characteristics based on birthplace

 A/PI
 
PUS versus foreign born Hispanic
 
P US versus foreign born 
 US born
 
Foreign born
 
Unknown
 
US born
 
Foreign born
 
Unknown
 
 Count Count Count Count Count Count 
Gender               
 Male 120 50 148 58 118 52 0.052 412 55 401 64 380 55 <0.001 
 Female 121 50 105 42 110 48  341 45 222 36 311 45  
Age               
 Mean 34.7  43.0  34.9 <0.001 35.7 41.2 37.6 <0.001 
 Interquartile range  21–44   28–57   23–44   21–50  27–55  23–50  
Histology               
 Nodular sclerosis 161 67 154 61 146 64 0.253 454 60 292 47 402 58 0.001 
 Mixed cellularity 30 12 44 17 31 14  130 17 156 25 116 17  
 Other HL 50 21 55 22 51 22  169 22 175 28 173 25  
Site               
 Extranodal 10 0.921 19 16 0.099 
Stage at dx               
 I 44 18 47 19 37 16 0.761 129 17 92 15 156 23 0.007 
 II 100 41 101 40 119 52  269 36 189 30 232 34  
 III 44 18 48 19 38 17  145 19 119 19 134 19  
 IV 50 21 41 16 28 12  165 22 184 30 128 19  
 NA 16  45 39 41  
B symptoms               
 No 115 48 97 38 111 49 0.202 247 33 195 31 272 39 0.194 
 Yes 87 36 95 38 73 32  342 45 317 51 244 35  
 NA 39 16 61 24 44 19  164 22 111 18 175 25  
 A/PI
 
PUS versus foreign born Hispanic
 
P US versus foreign born 
 US born
 
Foreign born
 
Unknown
 
US born
 
Foreign born
 
Unknown
 
 Count Count Count Count Count Count 
Gender               
 Male 120 50 148 58 118 52 0.052 412 55 401 64 380 55 <0.001 
 Female 121 50 105 42 110 48  341 45 222 36 311 45  
Age               
 Mean 34.7  43.0  34.9 <0.001 35.7 41.2 37.6 <0.001 
 Interquartile range  21–44   28–57   23–44   21–50  27–55  23–50  
Histology               
 Nodular sclerosis 161 67 154 61 146 64 0.253 454 60 292 47 402 58 0.001 
 Mixed cellularity 30 12 44 17 31 14  130 17 156 25 116 17  
 Other HL 50 21 55 22 51 22  169 22 175 28 173 25  
Site               
 Extranodal 10 0.921 19 16 0.099 
Stage at dx               
 I 44 18 47 19 37 16 0.761 129 17 92 15 156 23 0.007 
 II 100 41 101 40 119 52  269 36 189 30 232 34  
 III 44 18 48 19 38 17  145 19 119 19 134 19  
 IV 50 21 41 16 28 12  165 22 184 30 128 19  
 NA 16  45 39 41  
B symptoms               
 No 115 48 97 38 111 49 0.202 247 33 195 31 272 39 0.194 
 Yes 87 36 95 38 73 32  342 45 317 51 244 35  
 NA 39 16 61 24 44 19  164 22 111 18 175 25  

A/PI, Asian/Pacific Islander; dx, diagnosis; HL, Hodgkin's lymphoma; NA, not available.

survival

We analyzed OS according to race. Kaplan–Meier analysis showed that Hispanics and blacks had significantly inferior OS compared with whites and A/PIs (Figure 6). The 5-, 10-, and 15-year OS for whites were 82%, 75%, and 69%, respectively; for A/PIs 81%, 72%, and 69%, respectively; for Blacks 76%, 68%, and 62%, respectively; and for Hispanics 75%, 65%, and 61%, respectively. On multivariate Cox regression adjusting for age, gender, and stage, blacks and Hispanics had a persistent increased risk of death (inferior OS) compared with whites [Asians versus whites: HR 1.05 (95% CI 0.89–1.23); blacks versus whites: HR 1.23 (95% CI 1.02–1.47); Hispanics versus whites: HR 1.69 (95% CI 1.28–2.17)]. When analyzing place of birth, there were no OS differences comparing US- versus foreign-born A/PIs (P = 0.96) or Hispanics (P = 0.40).

Figure 6

Survival of Hodgkin's lymphoma according to race. Overall survival (OS) by Kaplan–Meier analysis for whites, Hispanics, A/PI's, and Hispanics. Associated inlet compares OS across races.

Figure 6

Survival of Hodgkin's lymphoma according to race. Overall survival (OS) by Kaplan–Meier analysis for whites, Hispanics, A/PI's, and Hispanics. Associated inlet compares OS across races.

discussion

To our knowledge, this report represents the largest and most comprehensive analysis examining the impact of race (white, black, Hispanic, and A/PI) on age- and gender-specific HL incidence as well as patient/disease-related characteristics, place-of-birth, and survival.

The age-specific bimodal incidence of HL is well documented; however, most of these reports consisted of a vast majority white population [21, 22]. We confirmed a bimodal distribution for whites, as well as A/PIs, in this analysis, although this was more prominent for men compared with women. Interestingly, after the age of 30 years, black men had a rather constant incidence rate, while black women had the suggestion of a bimodal pattern. Hispanics on the other hand had a small incidence peak at age 20–29 years followed by an exponential rise in HL incidence thereafter. Furthermore, after age 65 years, the incidence of HL was highest in Hispanics compared with all other races. Clinicopathologic racial differences included less common frequency of NS histology, more frequent presence of B symptoms, and more common advanced-stage disease for Hispanics and blacks compared with whites. The etiology of these differences is not known, although several factors have been hypothesized.

Socioeconomic status (SES) has been shown to impact incidence with individuals living in higher SES having higher risk of HL. Several California state-based studies showed that SES predicted HL incidence [12, 23, 24]. Underscoring the etiologic complexity of HL, Clarke et al. [23] showed that SES differences varied by age, race, histology, and gender. Other groups have shown that Epstein–Barr virus (EBV)-related HL is most commonly associated with MC histology subtype [25, 26]. The significant rise in HL incidence noted here in US-born A/PI and Hispanics is unlikely related to EBV since the increase consisted primarily of subjects aged 15–45 years and with NS histology.

We also identified significant racial disparities in survival. In a California statewide database study of EBV+ HL, Keegan et al. [24] showed that blacks and Hispanics had increased risk of HL-related death for patients age 15–44 years as well as for patients age >44 years compared with whites. Moreover, the survival disparity persisted for blacks and Hispanics after adjustment for SES. In another study, blacks had an increased risk of death (HR 1.54) compared with whites (Hispanics were not included in this analysis) [13].

There are several potential explanations of why cancer-related survival may differ based on race. This includes disease biology, host pharmacogenetics [27], delay in therapy [28], hospital factors [29], access to health care [30], and/or differences in SES as noted before [31–34]. In a SWOG study, Hershman et al. [5] showed that in part due to lower white blood cell counts, black women were more likely to experience treatment delay or early discontinuation of breast cancer therapy compared with white women. However, after adjustment of these and other factors, inferior survival for black women persisted. In a pediatric cancer study, Armstrong et al. [2] reported that blacks had an inferior survival compared with whites, Hispanics, and A/PI's and that this was primarily due to significantly higher mortality from nonrecurrence/nonexternal causes [2]. The survival difference emerged mainly >20 to 25 years of follow-up indicating that racial disparities in long-term utilization of health care may have been a factor. However, this would not explain the survival rates noted herein as the inferior survival for blacks and Hispanics emerged within the first few years of follow-up. The increased incidence of advanced-stage HL for blacks and Hispanics compared with whites and A/PI may be a factor in the survival disparity. Keegan et al. [24] showed that younger and older California blacks and Hispanics with HL stage III/IV disease had inferior survival rates compared with California whites with stage III/IV HL.

The current analysis consisted of US HL subjects and the findings may not be applicable to other countries, especially ones with less prominent racial diversity. However, we believe the findings herein are applicable to the overall epidemiology of HL especially given the differences in incidence and disease characteristics within Hispanics and A/PI's based on birthplace. While genetic influences may in part explain the global differences in HL incidence, differences in US versus foreign born support acculturation with an impact of lifestyle and environmental factors.

A shortcoming of the current analysis is the lack of tumor (or serum)-specific analyses as well as the absence of information on access to care and therapy received. In addition, birthplace information was missing for a significant minority of cases, especially for young adults. While studies have linked a number of patient factors to incompleteness of registry birthplace data [35, 36], these studies also found that registry birthplace data were highly accurate when they were recorded. Similar to the current report, numerous studies have used birthplace data from the SEER registries to identify disparities in cancer occurrence and for exploring cancer etiology [37–39]. Furthermore, similar to other analyses [40], our study uniquely linked the rich SEER database to estimates of the population that are immigrant in order to elucidate incidence patterns of US- and foreign-born individuals. Our study provides evidence that predictors of incidence and survival, such as race, as well as birthplace, are important factors that should be included in future studies especially when analyzing potential environmental/lifestyle and biological differences in disease. An additional strength of our study includes the large sample size and the population-based design that included broad geographic and racial diversity.

In summary, we identified a number of prominent racial disparities among a large population-based HL SEER database analysis. This included marked differences in age-specific incidence rates as well as varying clinical characteristics and disease-related factors (e.g. tumor histology, B symptoms, and stage) based on race. Furthermore, we identified a recent increase in the incidence of HL among young adult, US-born A/PIs, while divergent OS rates were noted based on race. Continued examination of racial disparities in adolescent/adult HL, especially the impact on survival, is needed.

disclosure

The authors declare no conflict of interest.

acknowledgements

Author contributions: AME designed research, carried out research, analyzed data, and wrote the paper. MA carried out research, analyzed data, and wrote the paper. BA analyzed data and wrote the paper. BCHC designed research, carried out research, analyzed data, and wrote the paper.

Presented at the 11th International Conference on Malignant Lymphoma; June 2011, Lugano, Switzerland (oral presentation).

references

1
Albain
KS
Unger
JM
Crowley
JJ
, et al.  . 
Racial disparities in cancer survival among randomized clinical trials patients of the Southwest Oncology Group
J Natl Cancer Inst
 , 
2009
, vol. 
101
 (pg. 
984
-
992
)
2
Armstrong
GT
Pan
Z
Ness
KK
, et al.  . 
Temporal trends in cause-specific late mortality among 5-year survivors of childhood cancer
J Clin Oncol
 , 
2010
, vol. 
28
 (pg. 
1224
-
1231
)
3
Blackstock
AW
Herndon
JE
II
Paskett
ED
, et al.  . 
Outcomes among African-American/non-African-American patients with advanced non-small-cell lung carcinoma: report from the Cancer and Leukemia Group B
J Natl Cancer Inst
 , 
2002
, vol. 
94
 (pg. 
284
-
290
)
4
Godley
PA
Schenck
AP
Amamoo
MA
, et al.  . 
Racial differences in mortality among Medicare recipients after treatment for localized prostate cancer
J Natl Cancer Inst
 , 
2003
, vol. 
95
 (pg. 
1702
-
1710
)
5
Hershman
DL
Unger
JM
Barlow
WE
, et al.  . 
Treatment quality and outcomes of African American versus white breast cancer patients: retrospective analysis of Southwest Oncology studies S8814/S8897
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
2157
-
2162
)
6
Koshiol
J
Lam
TK
Gridley
G
, et al.  . 
Racial differences in chronic immune stimulatory conditions and risk of non-Hodgkin's lymphoma in Veterans from the United States
J Clin Oncol
 , 
2011
, vol. 
29
 (pg. 
378
-
385
)
7
Menashe
I
Anderson
WF
Jatoi
I
Rosenberg
PS
Underlying causes of the black-white racial disparity in breast cancer mortality: a population-based analysis
J Natl Cancer Inst
 , 
2009
, vol. 
101
 (pg. 
993
-
1000
)
8
Pollock
BH
DeBaun
MR
Camitta
BM
, et al.  . 
Racial differences in the survival of childhood B-precursor acute lymphoblastic leukemia: a Pediatric Oncology Group Study
J Clin Oncol
 , 
2000
, vol. 
18
 (pg. 
813
-
823
)
9
Tammemagi
CM
Nerenz
D
Neslund-Dudas
C
, et al.  . 
Comorbidity and survival disparities among black and white patients with breast cancer
JAMA
 , 
2005
, vol. 
294
 (pg. 
1765
-
1772
)
10
Waxman
AJ
Mink
PJ
Devesa
SS
, et al.  . 
Racial disparities in incidence and outcome in multiple myeloma: a population-based study
Blood
 , 
2010
, vol. 
116
 (pg. 
5501
-
5506
)
11
Metzger
ML
Castellino
SM
Hudson
MM
, et al.  . 
Effect of race on the outcome of pediatric patients with Hodgkin's lymphoma
J Clin Oncol
 , 
2008
, vol. 
26
 (pg. 
1282
-
1288
)
12
Glaser
SL
Gulley
ML
Clarke
CA
, et al.  . 
Racial/ethnic variation in EBV-positive classical Hodgkin lymphoma in California populations
Int J Cancer
 , 
2008
, vol. 
123
 (pg. 
1499
-
1507
)
13
Shenoy
P
Maggioncalda
A
Malik
N
Flowers
CR
Incidence patterns and outcomes for Hodgkin lymphoma patients in the United States
Adv Hematol
 , 
2011
, vol. 
2011
 pg. 
725219
 
14
Landgren
O
Caporaso
NE
New aspects in descriptive, etiologic, and molecular epidemiology of Hodgkin's lymphoma
Hematol Oncol Clin North Am
 , 
2007
, vol. 
21
 (pg. 
825
-
840
)
15
Au
WY
Gascoyne
RD
Gallagher
RE
, et al.  . 
Hodgkin's lymphoma in Chinese migrants to British Columbia: a 25-year survey
Ann Oncol
 , 
2004
, vol. 
15
 (pg. 
626
-
630
)
16
Glaser
SL
Hsu
JL
Hodgkin's disease in Asians: incidence patterns and risk factors in population-based data
Leuk Res
 , 
2002
, vol. 
26
 (pg. 
261
-
269
)
17
 
Number of Persons by Race and Hispanic Ethnicity for SEER Participants (2000 Census Data). Inhttp://seer.cancer.gov/registries/data.html (22 July 2010, date last accessed)
18
 
Surveillance, Epidemiology, and End Results (SEER) Program. Research Data (1973--2007), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2010, based on the November 2009 submission.www.seer.cancer.gov (22 July 2010, date last accessed)
19
 
Ruggles S, Alexander JT, Genadek K, et al. Integrated Public Use Microdata Series: Version 5.0 [Machine-readable database]. Minneapolis, MN: University of Minnesota, 2010
20
 
Surveillance Research Program, National Cancer Institute SEER*Stat software version 6.6.2. Help System: Rate Algorithms. http://seer.cancer.gov/seerstat/508_WebHelp/seerstat.htm#Rate_Algorithms.htm. (22 July 2010, date last accessed)
21
Glaser
SL
Swartz
WG
Time trends in Hodgkin's disease incidence. The role of diagnostic accuracy
Cancer
 , 
1990
, vol. 
66
 (pg. 
2196
-
2204
)
22
Hjalgrim
H
Askling
J
Pukkala
E
, et al.  . 
Incidence of Hodgkin's disease in Nordic countries
Lancet
 , 
2001
, vol. 
358
 (pg. 
297
-
298
)
23
Clarke
CA
Glaser
SL
Keegan
TH
Stroup
A
Neighborhood socioeconomic status and Hodgkin's lymphoma incidence in California
Cancer Epidemiol Biomarkers Prev
 , 
2005
, vol. 
14
 (pg. 
1441
-
1447
)
24
Keegan
TH
Clarke
CA
Chang
ET
, et al.  . 
Disparities in survival after Hodgkin lymphoma: a population-based study
Cancer Causes Control
 , 
2009
, vol. 
20
 (pg. 
1881
-
1892
)
25
Jarrett
RF
Krajewski
AS
Angus
B
, et al.  . 
The Scotland and Newcastle epidemiological study of Hodgkin's disease: impact of histopathological review and EBV status on incidence estimates
J Clin Pathol
 , 
2003
, vol. 
56
 (pg. 
811
-
816
)
26
Diepstra
A
van Imhoff
GW
Schaapveld
M
, et al.  . 
Latent Epstein-Barr virus infection of tumor cells in classical Hodgkin's lymphoma predicts adverse outcome in older adult patients
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
3815
-
3821
)
27
Sanoff
HK
Sargent
DJ
Green
EM
, et al.  . 
Racial differences in advanced colorectal cancer outcomes and pharmacogenetics: a subgroup analysis of a large randomized clinical trial
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
4109
-
4115
)
28
Fedewa
SA
Ward
EM
Stewart
AK
Edge
SB
Delays in adjuvant chemotherapy treatment among patients with breast cancer are more likely in African American and Hispanic populations: a national cohort study 2004-2006
J Clin Oncol
 , 
2010
, vol. 
28
 (pg. 
4135
-
4141
)
29
Breslin
TM
Morris
AM
Gu
N
, et al.  . 
Hospital factors and racial disparities in mortality after surgery for breast and colon cancer
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
3945
-
3950
)
30
Bickell
NA
Wang
JJ
Oluwole
S
, et al.  . 
Missed opportunities: racial disparities in adjuvant breast cancer treatment
J Clin Oncol
 , 
2006
, vol. 
24
 (pg. 
1357
-
1362
)
31
Baker
KS
Davies
SM
Majhail
NS
, et al.  . 
Race and socioeconomic status influence outcomes of unrelated donor hematopoietic cell transplantation
Biol Blood Marrow Transplant
 , 
2009
, vol. 
15
 (pg. 
1543
-
1554
)
32
Byrne
MM
Halman
LJ
Koniaris
LG
, et al.  . 
Effects of poverty and race on outcomes in acute myeloid leukemia
Am J Clin Oncol
 , 
2010
, vol. 
34
 (pg. 
297
-
304
)
33
Kent
EE
Morris
RA
Largent
JA
, et al.  . 
Socioeconomic impacts on survival differ by race/ethnicity among adolescents and young adults with non-Hodgkin's lymphoma
J Cancer Epidemiol
 , 
2010
, vol. 
2010
 pg. 
824691
 
34
Wang
M
Burau
KD
Fang
S
, et al.  . 
Ethnic variations in diagnosis, treatment, socioeconomic status, and survival in a large population-based cohort of elderly patients with non-Hodgkin lymphoma
Cancer
 , 
2008
, vol. 
113
 (pg. 
3231
-
3241
)
35
Gomez
SL
Glaser
SL
Kelsey
JL
Lee
MM
Bias in completeness of birthplace data for Asian groups in a population-based cancer registry (United States)
Cancer Causes Control
 , 
2004
, vol. 
15
 (pg. 
243
-
253
)
36
Gomez
SL
Glaser
SL
Quality of cancer registry birthplace data for Hispanics living in the United States
Cancer Causes Control
 , 
2005
, vol. 
16
 (pg. 
713
-
723
)
37
Clarke
CA
Glaser
SL
Gomez
SL
, et al.  . 
Lymphoid malignancies in US Asians: incidence rate differences by birthplace and acculturation
Cancer Epidemiol Biomarkers Prev
 , 
2011
, vol. 
20
 (pg. 
1064
-
1077
)
38
Herrinton
LJ
Stanford
JL
Schwartz
SM
Weiss
NS
Ovarian cancer incidence among Asian migrants to the United States and their descendants
J Natl Cancer Inst
 , 
1994
, vol. 
86
 (pg. 
1336
-
1339
)
39
Flood
DM
Weiss
NS
Cook
LS
, et al.  . 
Colorectal cancer incidence in Asian migrants to the United States and their descendants
Cancer Causes Control
 , 
2000
, vol. 
11
 (pg. 
403
-
411
)
40
Lin
SS
O'Malley
CD
Clarke
CA
Le
GM
Birthplace and survival among Asian women diagnosed with breast cancer in cancer registry data: the impact of selection bias
Int J Epidemiol
 , 
2002
, vol. 
31
 (pg. 
511
-
513
author reply 513