Abstract
Background
The etiologic role of medical history, lifestyle, family history, and occupational risk factors in sporadic Burkitt lymphoma (BL) is unknown, but epidemiologic and clinical evidence suggests that risk factors may vary by age.
Methods
We investigated risk factors for sporadic BL in 295 cases compared with 21818 controls in a pooled analysis of 18 case–control studies in the International Lymphoma Epidemiology Consortium (InterLymph). Cases were defined to include typical BL or Burkitt-like lymphoma. Odds ratios (ORs) and 95% confidence intervals (CIs) for associations were calculated separately for younger (<50 years) and older (≥50 years) BL using multivariate logistic regression.
Results
Cases included 133 younger BL and 159 older BL (age was missing for three cases) and they were evenly split between typical BL (n = 147) and Burkitt-like lymphoma (n = 148). BL in younger participants was inversely associated with a history of allergy (OR = 0.58; 95% CI = 0.32 to 1.05), and positively associated with a history of eczema among individuals without other atopic conditions (OR = 2.54; 95% CI = 1.20 to 5.40), taller height (OR = 2.17; 95% CI = 1.08 to 4.36), and employment as a cleaner (OR = 3.49; 95% CI = 1.13 to 10.7). BL in older participants was associated with a history of hepatitis C virus seropositivity (OR = 4.19; 95% CI = 1.05 to 16.6) based on three exposed cases. Regardless of age, BL was inversely associated with alcohol consumption and positively associated with height.
Conclusions
Our data suggest that BL in younger and older adults may be etiologically distinct.
Burkitt lymphoma (BL) is an aggressive B-cell non-Hodgkin lymphoma (NHL) that occurs as three histologically indistinguishable subtypes: sporadic, endemic, and immunodeficiency-associated BL (1,2). Sporadic BL occurs in developed countries, accounting for about 30% of all lymphoid malignancies in children and 1%–5% of NHL in adults (3,4). Endemic BL occurs in equatorial Africa and Papua New Guinea as the most common childhood cancer and is associated with Epstein–Barr virus and Plasmodium falciparum infections (5). Immunodeficiency-associated BL is seen in those with HIV infection (6) or a history of solid organ transplantation (7).
In contrast, risk factors for sporadic BL are unknown because no analytical epidemiological studies have examined risk factors other than infectious agents (8,9). Descriptive epidemiological and clinical studies conducted in developed countries point to age-related differences in BL. These include distinct age-specific incidence peaks of BL, one during childhood and the other at 60 years of age, in a study conducted in the Netherlands (10), as well as distinct peaks in adults in two studies conducted in the United States (6,11), and one study conducted using BL data from four continents excluding Africa (12). Clinical studies indicate a superior treatment outcome in pediatric BL patients compared with older BL patients (13). As with other cancers (14,15), the bimodality of BL and the different clinical outcomes by age suggest distinct etiologies of BL diagnosed at different ages.
We investigated the associations between BL and medical history, lifestyle, family history, and occupational risk factors using pooled data from 18 case–control studies conducted in Europe, North America, and Australia as part of the International Lymphoma Epidemiology Consortium (InterLymph) NHL Subtypes Project. Our study is the first analytic epidemiological study to evaluate a broad range of risk factors for BL. The main aim of the study was to identify risk factors for sporadic BL in younger versus older participants (ie, <50 vs ≥50 years old).
Methods
Detailed methods for the InterLymph NHL Subtypes Project are reported elsewhere in this issue. Briefly, de-identified individual-level data from 18 case–control studies were included in this pooled analysis based on the following criteria: 1) case–control design with incident, histologically confirmed cases of BL and 2) availability of individual-level data for several risk factors of interest by December 31, 2011. Participants with serologically confirmed HIV infection or a history of solid organ transplantation were excluded. Because none of the cases was enrolled from equatorial Africa or Papua New Guinea, all cases in this analysis were considered to be sporadic BL. Ethics review committees at the participating institutions approved the studies and participants were enrolled after providing written informed consent.
NHL Subtype Ascertainment and Harmonization
Cases were diagnosed as BL, including a diagnosis of Burkitt-like lymphoma (BLL), according to the Working Formulation (16,17) or the 2001 World Health Organization (WHO) classification for hematologic malignancies (18). The histologic definition of BL is based on observing a diffuse and starry-sky pattern at low magnification under the microscope, a homogenous population of medium-sized lymphoid cells with round nuclei, open chromatin, multiple distinct nucleoli, moderate amounts of basophilic cytoplasm and frequent mitotic figures at high magnification, typical immunohistochemistry (positive for CD10, CD20, and Bcl-6 and negative for Bcl-2, and with high Ki-67 proliferation index), and a simple cytogenetic profile involving c-MYC gene translocation. Cases suggestive of BL but showing a more pleomorphic cytology of the tumor cells were diagnosed as BLL. Not all studies retained tissue samples, so centralized pathology review of the diagnoses was not possible. An interdisciplinary team of pathologists and epidemiologists reviewed the pathology reports from the original studies to ensure that the NHL subtype classification was as accurate and specific as possible using guidelines from the InterLymph Pathology Working Group (19,20).
Risk Factor Ascertainment and Harmonization
Self-reported medical, family, lifestyle, and occupational histories were elicited using a standardized, structured in-person or telephone interview and/or written questionnaire, as reported elsewhere in this issue. The questionnaire data obtained from the contributing studies was harmonized centrally with each individual variable in the dataset reviewed and checked for consistency with related exposure variables.
Statistical Analysis
As the main study objective was to investigate risk factors for BL in younger and older participants, stratified analyses were conducted with cases and controls categorized as younger participants (<50 years) or older participants (≥50 years), based on the median age at BL diagnosis. Demographic and socioeconomic factors were analyzed. Associations of age-specific BL in younger and older participants with risk factors were estimated by calculating pooled odds ratios (ORs) and 95% confidence intervals (CIs) for each exposure category relative to the stated referent group, as well as the appropriate P values, using fixed-effects logistic regression models that adjusted for study, age, sex, and race/ethnicity (ie, the basic adjusted model). Because the controls from most of the studies were frequency matched by age and sex to all cases (ie, all NHL subtypes) in the study rather than just BL, we conducted all analyses using all controls. Sensitivity analyses using a subset of controls from each study that were individually matched by age and sex to BL cases gave essentially similar results (not shown). Associations with nominal multilevel exposures were evaluated by testing for statistical heterogeneity across the variable categories, whereas associations with ordinal multilevel exposures were evaluated by testing for a monotonic statistical trend in the OR per change in variable category using the likelihood ratio test. A two-sided P value less than .05 without adjustment for multiple comparisons indicated a statistically significant association. Individuals with missing data for the exposure variable of interest were excluded from the model when estimating the association. Interstudy heterogeneity among the 18 studies was evaluated using separate logistic regression models performed within each study and the variability of the study-specific coefficients was quantified by the H statistic as described elsewhere in this issue.
The independent association of age-specific BL with variables identified in the basic adjusted model was evaluated by constructing a final logistic regression model including all significantly associated variables and running a forward stepwise multivariate logistic regression until only variables that were statistically significant at P less than .05 were retained (the final model). In contrast to analyses in the basic adjusted model, individuals with missing variable categories were retained in the final model by including a code for the missing category in the variable. We also conducted histology-specific analyses (typical BL or BLL) and with younger and older cases combined for comparison with the age-specific results. Results were reported only when based on data from six or more studies.
Results
We studied 295 cases (147 typical BL and 148 BLL) and 21818 controls (Table 1). Most came from population-based studies (83.7% of cases and 76.6% of controls) and from studies conducted in North America (62.4% of cases and 51.9% of controls). The median age of all patients with BL was 52 years (interquartile range: 36–63 years) and was similar for typical BL and BLL (data not shown). No differences were noted between the cases and controls with respect to white, non-Hispanic ethnicity (84.4% vs 93.1%) or socioeconomic class distribution (Table 1). The age-stratified analyses were performed by rounding the median age to a more intuitive cut point of 50 years.
Table 1.
Demographical characteristics of cases of sporadic Burkitt lymphoma and controls in the InterLymph NHL Subtypes Project*
| Controls | Cases | |
|---|---|---|
| No. (%)† | No. (%)† | |
| All subjects | 21818 (100) | 295 (100) |
| Age, y | ||
| <30 | 1267 (5.8) | 41 (13.9) |
| 30–39 | 2001 (9.2) | 45 (15.3) |
| 40–49 | 2926 (13.4) | 47 (15.9) |
| 50–59 | 4508 (20.7) | 64 (21.7) |
| 60–69 | 6104 (28.0) | 65 (22.0) |
| 70–79 | 4127 (18.9) | 27 (9.2) |
| ≥80 | 860 (3.9) | 3 (1.0) |
| Sex | ||
| Male | 12809 (58.7) | 207 (70.2) |
| Female | 9009 (41.3) | 88 (29.8) |
| Race/ethnicity | ||
| White, non-Hispanic | 20318 (93.1) | 249 (84.4) |
| Black | 342 (1.6) | 3 (1.0) |
| Asian | 320 (1.5) | 15 (5.1) |
| Hispanic | 356 (1.6) | 21 (7.1) |
| Other/unknown/missing | 482 (2.2) | 7 (2.4) |
| Socioeconomic status | ||
| Low | 8900 (40.8) | 117 (39.7) |
| Medium | 6280 (28.8) | 80 (27.1) |
| High | 6233 (28.6) | 94 (31.9) |
| Other/missing | 405 (1.9) | 4 (1.4) |
| NHL classification | ||
| World Health Organization | 12488 (57.2) | 159 (53.9) |
| Working Formulation | 9330 (42.8) | 136 (46.1) |
| Histological diagnosis | ||
| Burkitt lymphoma | — | 147 (49.8) |
| Burkitt-like lymphoma | — | 148 (50.2) |
| BL anatomic site | ||
| Nodal | — | 120 (40.7) |
| Extranodal | — | 73 (24.7) |
| Systemic | — | 21 (7.1) |
| Not collected | — | 81 (27.5) |
| Study region | ||
| North America | 11323 (51.9) | 184 (62.4) |
| Northern Europe | 5403 (24.8) | 58 (19.7) |
| Southern Europe | 4398 (20.2) | 49 (16.6) |
| Australia | 694 (3.2) | 4 (1.4) |
| Study design | ||
| Population-based | 16707 (76.6) | 247 (83.7) |
| Hospital-based | 5111 (23.4) | 48 (16.3) |
* BL = Burkitt lymphoma; NHL = non-Hodgkin lymphoma.
† The counts do not add up to the total number of cases/controls due to data missing by design or report.
The results from the basic adjusted model showed that BL in younger participants was inversely associated with a history of allergy (OR = 0.52; 95% CI = 0.30 to 0.90) or asthma (OR = 0.35; 95% CI = 0.13 to 0.95; based on four exposed cases). A non-significant inverse association with BL in younger participants was also observed for a history of atopy (OR = 0.68; 95% CI = 0.45 to 1.02), hay fever (OR = 0.60; 95% CI = 0.35 to 1.05), or food allergy (OR = 0.50; 95% CI = 0.17 to 1.50; based on four exposed cases). A history of eczema was unrelated to BL in younger participants. However, when eczema alone without other atopic conditions was considered, it was associated with elevated BL risk (OR = 2.8; 95% CI = 1.35 to 5.86) (Table 2). In contrast to BL in younger participants, a history of allergy, asthma, atopy, hay fever, or eczema was not associated with BL in older participants.
Table 2.
Association of medical and family history with sporadic Burkitt lymphoma in the InterLymph NHL Subtypes Project from the basic model, stratified by age*
| <50 years | ≥50 years | |||||||
|---|---|---|---|---|---|---|---|---|
| Controls | Cases | OR (95% CI)‡ | P | Controls | Cases | OR (95% CI)† | P | |
| No. (%)† | No. (%)† | No. (%)† | No. (%)† | |||||
| Medical history | ||||||||
| Any atopic disorder§ | ||||||||
| No | 3763 (65.4) | 89 (67.4) | 1.00 (referent) | .058 | 10156 (70.3) | 114 (72.2) | 1.00 (referent) | .736 |
| Yes | 1878 (32.6) | 40 (30.3) | 0.68 (0.45 to 1.02) | 3977 (27.5) | 43 (27.2) | 0.94 (0.64 to 1.37) | ||
| Allergy|| | ||||||||
| No | 2490 (64.9) | 64 (70.3) | 1.00 (referent) | .016 | 7082 (71.3) | 64 (68.8) | 1.00 (referent) | .760 |
| Yes | 1093 (28.5) | 21 (23.1) | 0.52 (0.30 to 0.90) | 2231 (22.5) | 23 (24.7) | 1.09 (0.64 to 1.85) | ||
| Food allergy | ||||||||
| No | 3146 (82.0) | 70 (76.9) | 1.00 (referent) | .176 | 7140 (81.7) | 57 (71.3) | 1.00 (referent) | .989 |
| Yes | 274 (7.1) | 4 (4.4) | 0.50 (0.17 to 1.50) | 586 (6.7) | 5 (6.3) | 0.99 (0.38 to 2.59) | ||
| Asthma | ||||||||
| No | 4865 (84.6) | 109 (82.6) | 1.00 (referent) | .015 | 12296 (85.1) | 136 (86.1) | 1.00 (referent) | .745 |
| Yes | 461 (8.0) | 4 (3.0) | 0.35 (0.13 to 0.95) | 1061 (7.3) | 13 (8.2) | 1.11 (0.61 to 2.00) | ||
| Hay fever | ||||||||
| No | 2646 (62.9) | 68 (62.4) | 1.00 (referent) | .062 | 8543 (71.6) | 98 (72.6) | 1.00 (referent) | .228 |
| Yes | 909 (21.6) | 23 (21.1) | 0.60 (0.35 to 1.05) | 1769 (14.8) | 20 (14.8) | 0.71 (0.41 to 1.25) | ||
| Eczema | ||||||||
| No | 3583 (83.7) | 91 (84.3) | 1.00 (referent) | .649 | 9868 (86.9) | 122 (92.4) | 1.00 (referent) | .684 |
| Yes¶ | 516 (12.0) | 12 (11.1) | 1.16 (0.62 to 2.19) | 912 (8.0) | 8 (6.1) | 0.86 (0.41 to 1.80) | ||
| Eczema without other atopic conditions | ||||||||
| No eczema | 3583 (83.7) | 91 (84.3) | 1.00 (referent) | .008 | 9868 (86.9) | 122 (92.4) | 1.00 (referent) | .805 |
| Eczema without other atopic conditions# | 176 (4.1) | 9 (8.3) | 2.82 (1.35 to 5.86) | 402 (3.5) | 4 (3.0) | 1.06 (0.38 to 2.93) | ||
| Eczema with other atopic conditions | 340 (7.9) | 3 (2.8) | 0.41 (0.13 to 1.32) | 510 (4.5) | 4 (3.0) | 0.723 (0.26 to 2.02) | ||
| History of hepatitis C virus infection | ||||||||
| No | 1891 (59.3) | 31 (56.4) | 1.00 (referent) | .430 | 4453 (69.7) | 30 (62.5) | 1.00 (referent) | .066 |
| Yes | 42 (1.3) | 0 (0.0) | — | 109 (1.7) | 3 (6.3) | 4.10 (1.10 to 15.4) | ||
| Blood transfusion | ||||||||
| No | 3062 (89.8) | 71 (86.6) | 1.00 (referent) | .228 | 6091 (70.6) | 78 (75.7) | 1.00 (referent) | .054 |
| Yes | 232 (6.8) | 8 (9.8) | 1.64 (0.77 to 3.50) | 1514 (17.5) | 10 (9.7) | 0.54 (0.27 to 1.06) | ||
| Total number of blood transfusions | ||||||||
| No transfusion | 3062 (89.8) | 71 (86.6) | 1.00 (referent) | .201 | 6091 (70.6) | 78 (75.7) | 1.00 (referent) | .267 |
| 1 transfusion | 163 (4.8) | 4 (4.9) | 1.09 (0.39 to 3.08) | 973 (11.3) | 6 (5.8) | 0.48 (0.21 to 1.13) | ||
| 2 transfusions | 39 (1.1) | 1 (1.2) | 1.61 (0.21 to 12.2) | 292 (3.4) | 2 (1.9) | 0.74 (0.18 to 3.10) | ||
| ≥3 transfusions | 23 (0.7) | 3 (3.7) | 6.46 (1.75 to 23.81) | 189 (2.2) | 2 (1.9) | 0.79 (0.19 to 3.40) | ||
| Number of years from first blood transfusion | ||||||||
| No transfusion | 3062 (89.8) | 71 (86.6) | 1.00 (referent) | .194 | 6091 (70.6) | 78 (75.7) | 1.00 (referent) | .101 |
| <20 years | 134 (3.9) | 6 (7.3) | 2.22 (0.92 to 5.34) | 657 (7.6) | 3 (2.9) | 0.36 (0.11 to 1.16) | ||
| 20–39 years | 85 (2.5) | 1 (1.2) | 0.51 (0.07 to 3.74) | 557 (6.5) | 6 (5.8) | 0.92 (0.39 to 2.17) | ||
| Blood transfusion before 1990 | ||||||||
| No transfusion | 3062 (89.8) | 71 (86.6) | 1.00 (referent) | .117 | 6091 (70.6) | 78 (75.7) | 1.00 (referent) | .068 |
| Transfusion before 1990 | 158 (4.6) | 7 (8.5) | 2.22 (0.98 to 5.04) | 1118 (13.0) | 10 (9.7) | 0.65 (0.33 to 1.28) | ||
| Transfusion year unknown | 11 (0.3) | 1 (1.2) | 1.97 (0.21 to 18.0) | 94 (1.1) | 0 (0.0) | — | ||
| Family history of non-Hodgkin lymphoma | ||||||||
| No | 3057 (91.3) | 68 (77.3) | 1.00 (referent) | .517 | 9501 (91.8) | 111 (91.0) | 1.00 (referent) | .272 |
| Yes | 40 (1.2) | 2 (2.3) | 1.71 (0.37 to 7.91) | 229 (2.2) | 1 (0.8) | 0.38 (0.05 to 2.83) | ||
| Family history of leukemia | ||||||||
| No | 3451 (91.6) | 74 (80.4) | 1.00 (referent) | .088 | 9409 (90.9) | 107 (87.7) | 1.00 (referent) | .644 |
| Yes | 65 (1.7) | 0 (0.0) | — | 321 (3.1) | 5 (4.1) | 1.26 (0.49 to 3.21) | ||
* CI = confidence interval; NHL = non-Hodgkin lymphoma; OR = odds ratio.
† The counts do not add up to the total number of cases/controls due to data missing by design or report.
‡ Adjusted for age, sex, race/ethnicity, and study.
§ Atopic disorders include asthma, eczema, hay fever, or other allergies, excluding drug allergies.
|| History of allergy included plant, food, animal, dust, insect, or mold allergy.
¶ Eczema, with or without other atopic conditions.
# Eczema alone, excluding participants who reported eczema and other atopic conditions.
We observed an elevated risk of BL among older participants who reported of a history of hepatitis C virus (HCV) seropositivity (OR = 4.1; 95% CI = 1.10 to 15.4), based on three exposed cases in the six studies that reported HCV serostatus (Table 2). The relationship between BL and HCV seropositivity in younger participants was indeterminate because none of the 31 younger cases reported a seropositive history. A history of prior blood transfusion (yes vs no) was unrelated to BL in younger participants but was inversely associated with BL risk in older participants. Younger participants who reported receiving three or more transfusions had an elevated risk of BL in (OR = 6.46; 95% CI = 1.75 to 23.8), but this result was based on only three exposed cases and the risk was not increased in the older participants with higher numbers of transfusions (Table 2). Peptic ulcer and hormone use were not associated with BL in younger or older participants (data not shown). The risk of BL in younger or older participants was not associated with a history of NHL or leukemia among first-degree relatives, but the number of cases was small in both groups (Table 2).
With respect to anthropometric factors, the risk of BL doubled in the tallest participants (quartile 4) compared with the shortest ones (quartile 1), both in the younger (OR = 2.14; 95% CI = 1.07 to 4.25) and the older (OR = 1.79; 95% CI = 0.92 to 3.49) age groups (Table 3). BL was not associated with usual weight or body mass index in younger participants, but a non-significant inverse association was observed between BL and body mass index in older participants (P trend = .07). With respect to lifestyle factors, the risks for both younger and older BL were inversely associated with alcohol consumption (Table 3), particularly among older participants. A history of cigarette smoking was not associated with BL in either younger or older participants (Table 3). We observed a non-significantly elevated risk of BL among younger participants who reported the use of light-colored hair dye compared with never use (OR = 3.0; 95% CI = 0.91 to 9.9), but the association with BL among older participants was indeterminate because only one older case was exposed. BL among older participants was not associated with the use of dark hair dyes (OR = 1.29; 95% CI = 0.46 to 3.63) based on nine exposed cases, and among younger participants, only two cases were exposed. BL was not associated with recreational sun exposure in either younger or older participants (Table 3).
Table 3.
Association of anthropometry and lifestyle factors with sporadic Burkitt lymphoma in the InterLymph NHL Subtypes Project from the basic model, stratified by age*
| <50 years | ≥50 years | |||||||
|---|---|---|---|---|---|---|---|---|
| Controls | Cases | OR (95% CI)‡ | P | Controls | Cases | OR (95% CI)‡ | P | |
| N (%)† | N (%)† | N (%)† | N (%)† | |||||
| Usual adult height§ | ||||||||
| Quartile 1 (low) | 816 (17.1) | 12 (12.0) | 1.00 (referent) | .029 | 3039 (26.7) | 19 (20.9) | 1.00 (referent) | .301 |
| Quartile 2 | 979 (20.5) | 14 (14.0) | 1.03 (0.46 to 2.28) | 2693 (23.7) | 22 (24.2) | 1.57 (0.83 to 2.97) | ||
| Quartile 3 | 1174 (24.6) | 21 (21.0) | 1.27 (0.60 to 2.70) | 2703 (23.8) | 24 (26.4) | 1.66 (0.87 to 3.15) | ||
| Quartile 4 (high) | 1612 (33.8) | 47 (47.0) | 2.14 (1.07 to 4.25) | 2528 (22.2) | 24 (26.4) | 1.79 (0.92 to 3.49) | ||
| Usual adult weight|| | ||||||||
| Quartile 1 (low) | 1342 (28.2) | 22 (22.0) | 1.00 (referent) | .294 | 2452 (21.6) | 24 (26.4) | 1.00 (referent) | .636 |
| Quartile 2 | 1122 (23.6) | 23 (23.0) | 1.26 (0.68 to 2.33) | 2555 (22.5) | 22 (24.2) | 0.93 (0.51 to 1.68) | ||
| Quartile 3 | 1102 (23.1) | 29 (29.0) | 1.67 (0.93 to 3.00) | 3032 (26.7) | 19 (20.9) | 0.68 (0.37 to 1.27) | ||
| Quartile 4 (high) | 1015 (21.3) | 20 (20.0) | 1.05 (0.54 to 2.02) | 2924 (25.7) | 24 (26.4) | 0.80 (0.43 to 1.46) | ||
| Usual adult body mass index (kg/m2) | ||||||||
| 15–<18.5 | 105 (2.2) | 1 (1.0) | 0.49 (0.07 to 3.77) | 146 (1.3) | 3 (3.3) | 1.94 (0.55 to 6.89) | ||
| 18.5–<22.5 | 1320 (27.7) | 24 (24.0) | 1.00 (referent) | .71 | 1879 (16.5) | 19 (20.9) | 1.00 (referent) | .07 |
| 22.5–<25 | 1250 (26.2) | 31 (31.0) | 1.31 (0.75 to 2.29) | 2724 (24.0) | 20 (22.0) | 0.68 (0.36 to 1.29) | ||
| 25–<30 | 1434 (30.1) | 27 (27.0) | 0.91 (0.50 to 1.65) | 4415 (38.8) | 28 (30.8) | 0.52 (0.28 to 0.96) | ||
| 30–<35 | 345 (7.2) | 9 (9.0) | 1.11 (0.49 to 2.56) | 1344 (11.8) | 17 (18.7) | 1.08 (0.54 to 2.14) | ||
| 35–50 | 127 (2.7) | 2 (2.0) | 0.73 (0.16 to 3.24) | 455 (4.0) | 2 (2.2) | 0.37 (0.08 to 1.66) | ||
| Alcohol consumption status 2 years prior to diagnosis/interview | ||||||||
| Non-drinker | 1065 (23.4) | 32 (29.9) | 1.00 (referent) | .211 | 3103 (23.4) | 54 (36.0) | 1.00 (referent) | .05 |
| Former drinker | 96 (2.1) | 0 (0.0) | — | 468 (3.5) | 2 (1.3) | 0.48 (0.10 to 2.25) | ||
| Current drinker | 1394 (30.7) | 22 (20.6) | 0.75 (0.36 to 1.57) | 3335 (25.1) | 21 (14.0) | 0.44 (0.21 to 0.91) | ||
| Drinker, status unknown | 1480 (32.6) | 38 (35.5) | 0.68 (0.39 to 1.20) | 3240 (24.4) | 53 (35.3) | 0.65 (0.41 to 1.03) | ||
| Age at first alcohol consumption | ||||||||
| Non-drinker | 1065 (23.4) | 32 (29.9) | 1.00 (referent) | .192 | 3103 (23.4) | 54 (36.0) | 1.00 (referent) | .07 |
| <20 years | 662 (14.6) | 13 (12.1) | 1.26 (0.51 to 3.12) | 1478 (11.1) | 7 (4.7) | 0.38 (0.15 to 0.98) | ||
| 20–29 years | 1045 (23.0) | 11 (10.3) | 0.48 (0.19 to 1.19) | 1801 (13.6) | 15 (10.0) | 0.58 (0.28 to 1.19) | ||
| ≥30 years | 102 (2.2) | 2 (1.9) | 1.10 (0.24 to 5.14) | 654 (4.9) | 3 (2.0) | 0.36 (0.10 to 1.23) | ||
| Drinker, age start unknown | 1161 (25.6) | 34 (31.8) | 0.63 (0.35 to 1.16) | 3110 (23.4) | 51 (34.0) | 0.65 (0.41 to 1.05) | ||
| Duration of alcohol consumption¶ | 6048 (100.0) | 133 (100.0) | 0.98 (0.94 to 1.02) | .353 | 15086 (100.0) | 159 (100.0) | 0.98 (0.96 to 1.00) | .047 |
| Ethanol per week as an adult, consumed from any type of alcoholic beverage# | ||||||||
| Non-drinker | 1065 (23.4) | 32 (29.9) | 1.00 (referent) | .409 | 3103 (23.4) | 54 (36.0) | 1.00 (referent) | .043 |
| Q1 (low) | 730 (16.1) | 22 (20.6) | 0.81 (0.45 to 1.45) | 1508 (11.4) | 24 (16.0) | 0.75 (0.45 to 1.27) | ||
| Q2 | 676 (14.9) | 17 (15.9) | 0.83 (0.44 to 1.58) | 1532 (11.5) | 13 (8.7) | 0.42 (0.22 to 0.80) | ||
| Q3 | 670 (14.8) | 10 (9.3) | 0.56 (0.26 to 1.18) | 1548 (11.7) | 21 (14.0) | 0.69 (0.40 to 1.20) | ||
| Q4 (high) | 683 (15.0) | 10 (9.3) | 0.57 (0.26 to 1.25) | 1682 (12.7) | 17 (11.3) | 0.51 (0.27 to 0.94) | ||
| Drinker, grams unknown | 211 (4.6) | 1 (0.9) | 0.19 (0.02 to 2.42) | 773 (5.8) | 1 (0.7) | 0.16 (0.01 to 2.26) | ||
| Lifetime alcohol consumed | ||||||||
| Non-drinker | 1065 (23.4) | 32 (29.9) | 1.00 (referent) | .604 | 3103 (23.4) | 54 (36.0) | 1.00 (referent) | .025 |
| 1–100 kg | 581 (12.8) | 10 (9.3) | 1.02 (0.40 to 2.55) | 765 (5.8) | 8 (5.3) | 0.90 (0.36 to 2.24) | ||
| 101–200 kg | 226 (5.0) | 2 (1.9) | 0.56 (0.12 to 2.68) | 404 (3.0) | 2 (1.3) | 0.37 (0.08 to 1.67) | ||
| 201–400 kg | 150 (3.3) | 2 (1.9) | 0.83 (0.17 to 4.01) | 497 (3.7) | 2 (1.3) | 0.23 (0.05 to 1.06) | ||
| >400 kg | 116 (2.6) | 2 (1.9) | 1.10 (0.23 to 5.28) | 640 (4.8) | 3 (2.0) | 0.22 (0.06 to 0.86) | ||
| Drinker, consumption unknown | 1897 (41.8) | 44 (41.1) | 0.63 (0.37 to 1.06) | 4737 (35.7) | 61 (40.7) | 0.63 (0.40 to 0.98) | ||
| Cigarette smoking status 2 years prior to diagnosis/interview | ||||||||
| Non-smoker | 2511 (43.2) | 68 (53.1) | 1.00 (referent) | .552 | 5760 (39.8) | 58 (37.7) | 1.00 (referent) | .912 |
| Former smoker | 1075 (18.5) | 19 (14.8) | 0.85 (0.5 to 1.47) | 4777 (33.0) | 55 (35.7) | 0.92 (0.62 to 1.34) | ||
| Current smoker | 1774 (30.5) | 29 (22.7) | 0.72 (0.45 to 1.14) | 2645 (18.3) | 29 (18.8) | 0.88 (0.54 to 1.42) | ||
| Smoker, status unknown | 125 (2.2) | 2 (1.6) | 0.73 (0.16 to 3.37) | 399 (2.8) | 5 (3.2) | 1.21 (0.41 to 3.56) | ||
| Color of hair dye used | ||||||||
| Never hair dye | 426 (13.3) | 6 (9.7) | 1.00 (referent) | .007 | 1318 (17.5) | 7 (12.7) | 1.00 (referent) | .291 |
| Light | 189 (5.9) | 8 (12.9) | 2.99 (0.91 to 9.88) | 690 (9.2) | 1 (1.8) | 0.26 (0.03 to 2.16) | ||
| Dark | 407 (12.7) | 2 (3.2) | 0.35 (0.06 to 1.92) | 1319 (17.5) | 9 (16.4) | 1.29 (0.46 to 3.63) | ||
| Ever hair dye, color unknown | 116 (3.6) | 1 (1.6) | 0.15 (0.01 to 1.89) | 310 (4.1) | 2 (3.6) | 0.57 (0.07 to 4.98) | ||
| Male** | 1995 (62.1) | 44 (71.0) | — | 3716 (49.4) | 35 (63.6) | — | ||
| Recreational sun exposure†† | ||||||||
| Quartile 1 (low) | 409 (16.1) | 7 (13.7) | 1.00 (referent) | .930 | 1425 (22.0) | 12 (28.6) | 1.00 (referent) | .725 |
| Quartile 2 | 654 (25.7) | 13 (25.5) | 1.16 (0.44 to 3.00) | 1324 (20.5) | 7 (16.7) | 0.61 (0.24 to 1.56) | ||
| Quartile 3 | 445 (17.5) | 10 (19.6) | 1.17 (0.43 to 3.19) | 1190 (18.4) | 8 (19) | 0.80 (0.32 to 2.00) | ||
| Quartile 4 (high) | 726 (28.5) | 13 (25.5) | 0.92 (0.36 to 2.39) | 1729 (26.7) | 10 (23.8) | 0.66 (0.28 to 1.57) | ||
* CI = confidence interval; NHL = non-Hodgkin lymphoma; OR = odds ratio.
† The counts do not add up to the total number of cases/controls due to data missing by design or report.
‡ Adjusted for age, sex, race/ethnicity, and study.
§ Height quartile cutoffs are: 172, 177.8, and 182cm for Q1-2, Q2-3, and Q3-4, respectively, for males, and 159, 163.0, and 168cm, respectively, for females.
|| Weight quartile cutoffs are: 72.6, 80, and 89kg for Q1-2, Q2-3, and Q3-4, respectively, for males, and 58.1, 65, and 74.8, respectively, for females.
¶ Adjusted for age, sex, race/ethnicity, and study.
# Alcohol quartile cutoff points: 44.4, 110.7, and 252g, respectively, for Q1-2, Q2-3, and Q3-4, respectively.
** Males were not included in the analysis of use of hair dye.
†† Recreational sun exposure (hours per week, study-specific quartiles available upon request).
Among 31 occupations evaluated, we observed an increased risk of BL among younger participants who reported employment as a cleaner (OR = 2.6; 95% CI = 1.01 to 6.83), especially a char worker cleaner (OR = 3.8; 95% CI = 1.26 to 11.3) based on four exposed cases (Table 4). A char worker cleaner is a person employed to clean and maintain the interiors of buildings. Occupation as a cleaner was not associated with BL among older participants nor was any other occupation associated with BL in younger or older participants (data not shown).
Table 4.
Association of cleaning-related occupations with sporadic Burkitt lymphoma in the InterLymph NHL Subtypes Project from the basic model, stratified by age*
| <50 years | ≥50 years | |||||||
|---|---|---|---|---|---|---|---|---|
| Controls | Cases | Controls | Cases | |||||
| N (%)† | N (%)† | OR (95% CI)‡ | P | N (%)† | N (%)† | OR (95% CI)‡ | P | |
| Cleaners | ||||||||
| No | 3026 (88.5) | 59 (85.5) | 1.00 (referent) | .077 | 6721 (94.5) | 54 (96.4) | 1.00 (referent) | .863 |
| Yes | 137 (4.0) | 5 (7.2) | 2.62 (1.01 to 6.83) | 337 (4.7) | 2 (3.6) | 0.88 (0.21 to 3.70) | ||
| Duration of employment as cleaner | ||||||||
| Never | 3026 (88.5) | 59 (85.5) | 1.00 (referent) | .158 | 6721 (94.5) | 54 (96.4) | 1.00 (referent) | .051 |
| 1–10 years | 109 (3.2) | 5 (7.2) | 3.27 (1.24 to 8.60) | 188 (2.6) | 0 (0.0) | — | ||
| >10 years | 23 (0.7) | 0 (0.0) | — | 143 (2.0) | 1 (1.8) | 1.14 (0.15 to 8.48) | ||
| Yes, missing duration | 5 (0.1) | 0 (0.0) | — | 6 (0.1) | 1 (1.8) | – (3.69 to infinity) | ||
| Charworker cleaner or related work§ | ||||||||
| No | 3071 (89.8) | 60 (87.0) | 1.00 (referent) | .040 | 6473 (96.2) | 52 (100.0) | 1.00 (referent) | .124 |
| Yes | 86 (2.5) | 4 (5.8) | 3.76 (1.26 to 11.3) | 203 (3.0) | 0 (0.0) | — | ||
| Duration of employment as charworker cleaner | ||||||||
| Never | 3071 (89.8) | 60 (87.0) | 1.00 (referent) | .095 | 6473 (96.2) | 52 (100.0) | 1.00 (referent) | .500 |
| 1–10 years | 64 (1.9) | 4 (5.8) | 5.00 (1.65 to 15.21) | 103 (1.5) | 0 (0.0) | — | ||
| >10 years | 18 (0.5) | 0 (0.0) | — | 99 (1.5) | 0 (0.0) | — | ||
| Yes, missing duration | 4 (0.1) | 0 (0.0) | — | 1 (0.0) | 0 (0.0) | — | ||
* CI = confidence interval; NHL = non-Hodgkin lymphoma; OR = odds ratio.
† The counts do not add up to the total number of cases/controls due to data missing by design or report.
‡ OR (95% CI): adjusted for age, sex, race/ethnicity, and study.
§ Charworker cleaner is a person who is employed to clean and maintain the interiors of buildings.
In the final model, BL in younger participants was associated with a history of allergy (OR = 0.58; 95% CI = 0.32 to 1.05), eczema without other atopic conditions (OR = 2.54; 95% CI = 1.20 to 5.40), taller height (OR = 2.17; 95% CI = 1.08 to 4.36), the use of light-colored hair dye (OR = 2.89; 95% CI = 0.84 to 9.94), and working in a cleaning occupation (OR = 3.49; 95% CI = 1.13 to 10.7) (Table 5). BL in older participants was independently associated with a history of HCV infection (OR = 4.19; 95% CI = 1.05 to 16.6), usual adult body mass index (P trend = .049), and consumption of at least one drink of alcohol per month (OR = 0.63; 95% CI = 0.40 to 0.98).
Table 5.
Independent risk factors for younger and older sporadic Burkitt lymphoma in the InterLymph NHL Subtypes Project from the final model*
| OR (95% CI) | P | |
|---|---|---|
| <50 years | ||
| Allergy† | ||
| No | 1.00 (referent) | .065 |
| Yes | 0.58 (0.32 to 10.5) | |
| Eczema without other atopic conditions | ||
| No eczema | 1.00 | |
| Eczema with no other atopic conditions | 2.54 (1.20 to 5.40) | |
| Eczema with other atopic conditions | 0.51 (0.15 to 1.74) | .041 |
| Usual adult height‡ | ||
| Quartile 1 (low) | 1.00 (referent) | .028 |
| Quartile 2 | 1.03 (0.46 to 2.30) | |
| Quartile 3 | 1.31 (0.62 to 2.80) | |
| Quartile 4 | 2.17 (1.08 to 4.36) | |
| Color of hair dye used | ||
| Never hair dye | 1.00 (referent) | .014 |
| Light | 2.89 (0.84 to 9.94) | |
| Dark | 0.37 (0.06 to 2.06) | |
| Charworker cleaner or related work§ | ||
| No | 1.00 (referent) | .054 |
| Yes | 3.49 (1.13 to 10.7) | |
| ≥50 years | ||
| Hepatitis C virus infection | ||
| Negative | 1.00 (referent) | .070 |
| Positive | 4.19 (1.05 to 16.61) | |
| Usual adult body mass index (kg/m2) | ||
| 15–<18.5 | 1.00 (referent) | .049 |
| 18.5–<22.5 | 0.55 (0.15 to 1.97) | |
| 22.5–<25 | 0.37 (0.10 to 1.32) | |
| 25–<30 | 0.27 (0.08 to 0.98) | |
| 30–<35 | 0.56 (0.15 to 2.08) | |
| 35–50 | 0.16 (0.02 to 1.05) | |
| History of alcohol consumption | ||
| Non-drinker | 1.00 (referent) | .001 |
| Drinker (at least 1 drink per month) | 0.63 (0.40 to 0.98) | |
* Table reports results from the final stepwise regression analysis models. CI = confidence interval; NHL = non-Hodgkin lymphoma; OR = odds ratio.
† History of allergy includes participants who reported eczema plus other atopic conditions.
‡ Height quartile cutoffs are: 172, 177.8, and 182cm for Q1-2, Q2-3, and Q3-4, respectively, for males, and 159, 163.0, and 168cm, respectively, for females.
§ Charworker cleaner is a person who is employed to clean and maintain the interior of buildings.
The associations with BL found in the age-specific analyses were attenuated or became statistically non-significant when the histological subtypes of BL and BLL were analyzed individually or with the younger and older cases combined (data not shown). There was no evidence of significant interstudy heterogeneity across the studies for the associations found in this study (data not shown).
Discussion
Our study, which is the first to evaluate a broad range of noninfectious risk factors for BL, showed some differences and similarities between BL in younger and older participants. Specifically, we observed significant inverse associations between BL in younger participants and a history of allergy, and significant positive associations with a history of eczema alone and working in a cleaning occupation, especially as a char worker cleaner. Conversely, among older participants, we observed a positive association with a history of HCV infection based on three exposed cases. Among both younger and older participants, BL was positively associated with height and inversely associated with alcohol consumption. These associations were attenuated or not statistically significant when the histological subtypes of typical BL and BLL were analyzed individually or when younger and older cases were combined. Our results provide epidemiological support for the hypothesis that younger and older adult sporadic BL may be distinct biologically or etiologically (11,12).
Our results agree with recent epidemiologic and clinical studies that reported multiple incidence peaks and different outcomes by age (6,10–13,15,21), all suggesting the possibility of etiologically and clinically distinct BL entities at different ages. Because no single parameter is a gold standard for BL diagnosis, and because the diagnosis of BL may be difficult, particularly in older patients (22), diagnostic uncertainty may be a factor in our results. Expert hematopathologists agree on the diagnosis of BL in 80%–90% of pediatric cases, but in only 26%–48% of adult cases (22), in part because the classification of older adult BL has been in some flux during the period of the study (1). Thus, some cases previously diagnosed as BLL might now be reclassified as BL or as a gray-zone lymphoma with features intermediate between BL and diffuse large B-cell lymphoma (4). Gray-zone lymphoma, which is characterized by a pleomorphic tumor cell morphology and complex cytogenetics, sometimes including a combination of c-MYC and BCL2 or BCL6 gene rearrangements (referred to as “double hit” lymphoma) (23), is overrepresented in older patients and has an extremely poor prognosis with standard treatment (24). Further studies are needed to confirm the age-related BL entities and to elucidate their molecular abnormalities.
The significant inverse association between younger BL and a history of allergy mirrors the associations that have been reported for diffuse large B-cell lymphoma and follicular lymphoma (25). Allergic diseases appear to be protective against certain NHL subtypes, but the mechanism is unknown. Assuming these results are valid, protective effects could be mediated via immunological hypersensitivity (26) or antitumor cytotoxicity (27). We found a significant positive association between a history of HCV infection and BL only in older participants, based on three cases of BL with HCV infection. This result agrees with the fivefold risk for BL reported among elderly patients in US Medicare data who had two or more insurance claims for HCV infection (28). Although HCV and HIV coinfection are common, particularly in intravenous drug users, coinfection with HIV is unlikely to explain our results because serologically confirmed HIV-positive cases were excluded in the cases in our study. Because HCV infection may increase risk for some NHL subtypes (29), perhaps by inducing chronic immune activation (30), further studies are needed to investigate the potential role of HCV infection in older adult BL.
BL in both younger and older participants was positively associated with height. This association has also been reported for other cancers (31,32) and might be indicative of the role of early-life exposures that influence height and also impact risk for cancer, such as insulin-like growth factors. Insulin-like growth factors are correlated with adult height and are thought to modulate cell proliferation and/or turnover (33,34). Our finding that the risk for BL was decreased in both younger and older participants who consumed alcohol is in accord with a previous report using much of the same data (35), but it is unexplained. In contrast to the previous reports of a positive association between use of dark hair dye and NHL risk (36–40), we found a positive association between the use of light-colored hair dye and BL only in younger participants. These results are conflicting and highlight the difficulty of evaluating hair dye use in rare cancers such as BL. A history of working in cleaning occupations, especially as a char worker cleaner, was positively associated with BL in younger participants. This result is similar to the associations reported for all NHL (41) and may point to a role for chemicals that are found in products used for cleaning.
The strengths of our pooled analysis include having relatively large case numbers from geographically diverse areas and centrally harmonized data classification and analysis procedures. Because up to one-third of BL cases may be HIV-associated, exclusion of HIV infected cases enabled us to focus our interpretation on adult sporadic BL. The limitations include possible selection and recall biases common to case–control studies, and possible diagnostic errors related to the use of older lymphoma classification systems such as the Working Formulation (16,17) and the 2001 WHO classification for hematologic malignancies (42). Lack of access to archival tissues and slides precluded a central pathology review to confirm and update the diagnoses (1). Lack of tumor Epstein–Barr virus status or molecular characterization also precluded us from evaluating whether Epstein–Barr virus or molecular abnormalities are correlated with the epidemiological clues for distinct BL biology and/or etiology (43). Other limitations include multiple significance tests, which increases the number of expected false-positive findings, and missing data, for example, for HCV infection status, which could have reduced the power to detect weak associations.
To summarize, we found both similar and contrasting age-specific patterns of associations for BL with medical history, anthropometric, lifestyle, and occupational risk factors in younger and older participants. These results support the hypothesis that BL in younger and older patients may be etiologically distinct.
Funding
This pooled analysis was supported by the Intramural Research Program of the National Cancer Institute/National Institutes of Health and National Cancer Institute/National Institutes of Health (R01 CA14690, U01 CA118444, and R01 CA92153-S1).
InterLymph annual meetings during 2010–2013 were supported by the Epidemiology and Genomics Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute/National Institutes of Health (2010–2013); Lymphoma Coalition (2010–2013); National Institutes of Health Office of Rare Diseases Research (2010); National Cancer Institute/National Institutes of Health (R13 CA159842 01) (2011); University of Cagliari, Provincial Administration of Cagliari, Banca di Credito Sardo, and Consorzio Industriale Sardo, Italy (2011); Intramural Research Program of the National Cancer Institute/National Institutes of Health (2012); and Faculté de Médecine de Dijon, Institut de Veille Sanitaire, Registre des hémopathies malignes de Côte d’Or, INSERM, Institut National du Cancer, Université de Bourgogne, Groupe Ouest Est d’Etude des Leucémies et Autres Maladies du Sang (GOELAMS), l’Institut Bergonié, The Lymphoma Study Association (LYSA), Registre Régional des Hémopathies de Basse Normandie, and the City of Dijon, France (2013). Meeting space at the 2013 Annual Meeting of the American Association for Cancer Research (AACR) was provided by the Molecular Epidemiology Group (MEG) of the AACR.
Individual studies were supported by: the Canadian Institutes for Health Research (CIHR), Canadian Cancer Society, and Michael Smith Foundation for Health Research (British Columbia); Intramural Research Program of the National Cancer Institute/National Institutes of Health (Iowa/Minnesota); National Cancer Institute/National Institutes of Health (N01-CP-ES-11027) (Kansas); National Cancer Institute/National Institutes of Health (R01 CA50850) (Los Angeles); National Cancer Institute/National Institutes of Health (R01 CA92153 and P50 CA97274), Lymphoma Research Foundation (164738), and the Henry J. Predolin Foundation (Mayo Clinic); Intramural Research Program of the National Cancer Institute/National Institutes of Health and Public Health Service (contracts N01-PC-65064, N01-PC-67008, N01-PC-67009, N01-PC-67010, and N02-PC-71105) (NCI-SEER); National Cancer Institute/National Institutes of Health (R01CA100555 and R03CA132153) and American Institute for Cancer Research (99B083) (Nebraska [newer]); National Cancer Institute/National Institutes of Health (N01-CP-95618) and State of Nebraska Department of Health (LB-506) (Nebraska [older]); National Cancer Institute/National Institutes of Health (R01CA45614, RO1CA154643-01A1, and R01CA104682) (UCSF1); National Cancer Institute/National Institutes of Health (CA143947, CA150037, R01CA087014, R01CA104682, RO1CA122663, and RO1CA154643-01A1) (UCSF2); National Heart Lung and Blood Institute/National Institutes of Health (hematology training grant award T32 HL007152), National Center for Research Resources/National Institutes of Health (UL 1 RR024160), and National Cancer Institute/National Institutes of Health (K23 CA102216 and P50 CA130805) (University of Rochester); National Cancer Institute/National Institutes of Health (CA62006 and CA165923) (Yale); Association pour la Recherche contre le Cancer, Fondation de France, AFSSET, and a donation from Faberge employees (Engela); European Commission (QLK4-CT-2000-00422 and FOOD-CT-2006-023103), Spanish Ministry of Health (CIBERESP, PI11/01810, RCESP C03/09, RTICESP C03/10, and RTIC RD06/0020/0095), Rio Hortega (CM13/00232), Agència de Gestió d’Ajuts Universitaris i de Recerca–Generalitat de Catalunya (Catalonian Government, 2009SGR1026), National Institutes of Health (contract NO1-CO-12400), Italian Ministry of Education, University and Research (PRIN 2007 prot. 2007WEJLZB, PRIN 2009 prot. 20092ZELR2), Italian Association for Cancer Research (IG grant 11855/2011), Federal Office for Radiation Protection (StSch4261 and StSch4420), José Carreras Leukemia Foundation (DJCLS-R04/08), German Federal Ministry for Education and Research (BMBF-01-EO-1303), Health Research Board, Ireland, Cancer Research, Ireland, and Czech Republic MH CZ - DRO (MMCI, 00209805) (EpiLymph); National Cancer Institute/National Institutes of Health (CA51086), European Community (Europe Against Cancer Programme), and Italian Alliance Against Cancer (Lega Italiana per la Lotta contro i Tumori) (Italy, multicenter); Italian Association for Cancer Research (Italy, Aviano-Milan); Italian Association for Cancer Research (Italy, Aviano-Naples); Swedish Cancer Society (2009/659), Stockholm County Council (20110209), Strategic Research Program in Epidemiology at Karolinska Institut, Swedish Cancer Society (02 6661), Danish Cancer Research Foundation, Lundbeck Foundation (R19-A2364), Danish Cancer Society (DP 08-155), National Cancer Institute/National Institutes of Health (5R01 CA69669-02), and Plan Denmark (SCALE); Leukaemia & Lymphoma Research, UK; and Australian National Health and Medical Research Council (ID990920), Cancer Council NSW, and University of Sydney Faculty of Medicine (New South Wales).
We thank the following individuals for their substantial contributions to this project: Aaron D. Norman, Dennis P. Robinson, and Priya Ramar (Mayo Clinic College of Medicine) for their work at the InterLymph Data Coordinating Center in organizing, collating, harmonizing, and documenting of the data from the participating studies in the InterLymph Consortium; Michael Spriggs, Peter Hui, and Bill Wheeler (Information Management Services, Inc) for their programming support; and Noelle Richa Siegfried and Emily Smith (RTI International) for project coordination.
References
- 1. Leoncini L, Raphael M, Stein H, Harris NL, Jaffe ES, Kluin PM. Burkitt lymphoma. In: Swerdlow SH, Campo E, Harris NL, et al., eds. WHO Classification of Tumours Haematopoetic and Lymphoid Tissues. 4th ed. Lyon, France: IARC Press; 2008:262–264. [Google Scholar]
- 2. Wright DH. What is Burkitt’s lymphoma and when is it endemic? Blood. 1999;93(2):758. [PubMed] [Google Scholar]
- 3. Linch DC. Burkitt lymphoma in adults. Br J Haematol. 2012;156(6):693–703 [DOI] [PubMed] [Google Scholar]
- 4. Jaffe ES, Pittaluga S. Aggressive B-cell lymphomas: a review of new and old entities in the WHO classification. Hematology Am Soc Hematol Educ Program. 2011;2011:506–514 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Rochford R, Cannon MJ, Moormann AM. Endemic Burkitt’s lymphoma: a polymicrobial disease? Nat Rev Microbiol. 2005;3(2):182–187 [DOI] [PubMed] [Google Scholar]
- 6. Guech-Ongey M, Simard EP, Anderson WF, et al. AIDS-related Burkitt lymphoma in the United States: what do age and CD4 lymphocyte patterns tell us about etiology and/or biology? Blood. 2010;116(25):5600–5604 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mbulaiteye SM, Clarke CA, Morton LM, et al. Burkitt lymphoma risk in U.S. solid organ transplant recipients. Am J Hematol. 2013;88(4):245–250 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Roman E, Smith AG. Epidemiology of lymphomas. Histopathology. 2011;58(1):4–14 [DOI] [PubMed] [Google Scholar]
- 9. Thorley-Lawson DA, Allday MJ. The curious case of the tumour virus: 50 years of Burkitt’s lymphoma. Nat Rev Microbiol. 2008;6(12):913–924 [DOI] [PubMed] [Google Scholar]
- 10. Boerma EG, van Imhoff GW, Appel IM, Veeger NJ, Kluin PM, Kluin-Nelemans JC. Gender and age-related differences in Burkitt lymphoma–epidemiological and clinical data from The Netherlands. Eur J Cancer. 2004;40(18):2781–2787 [DOI] [PubMed] [Google Scholar]
- 11. Mbulaiteye SM, Anderson WF, Bhatia K, Rosenberg PS, Linet MS, Devesa SS. Trimodal age-specific incidence patterns for Burkitt lymphoma in the United States, 1973-2005. Int J Cancer. 2010;126(7):1732–1739 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Mbulaiteye SM, Anderson WF, Ferlay J, et al. Pediatric, elderly, and emerging adult-onset peaks in Burkitt’s lymphoma incidence diagnosed in four continents, excluding Africa. Am J Hematol. 2012;87(6):573–578 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kelly JL, Toothaker SR, Ciminello L, et al. Outcomes of patients with Burkitt lymphoma older than age 40 treated with intensive chemotherapeutic regimens. Clin Lymphoma Myeloma. 2009;9(4):307–310 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Grimley PM, Matsumo R, Rosenberg PS, Henson DE, Schwartz AM, Anderson WF. Qualitative age interactions between low and high grade serous ovarian carcinoma. Cancer Epidemiol Biomarkers Prev. 2009;18(8):2256–2261. [DOI] [PubMed] [Google Scholar]
- 15. Macmahons B. Epidemiological evidence of the nature of Hodgkin’s disease. Cancer. 1957;10(5):1045–1054 [DOI] [PubMed] [Google Scholar]
- 16. National Cancer Institute sponsored study of classifications of non-Hodgkin’s lymphomas: summary and description of a working formulation for clinical usage. The Non-Hodgkin’s Lymphoma Pathologic Classification Project. Cancer. 1982;49(10):2112–2135 [DOI] [PubMed] [Google Scholar]
- 17. Percy C, O’Conor G, Ries LG, Jaffe ES. Non-Hodgkin’s lymphomas. Application of the International Classification of Diseases for Oncology (ICD-O) to the Working Formulation. Cancer. 1984;54(7):1435–1438 [DOI] [PubMed] [Google Scholar]
- 18. Jaffe ES, Diebold J, Harris NL, Muller-Hermelink HK, Flandrin G, Vardiman JW. Burkitt’s lymphoma: a single disease with multiple variants. The World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues. Blood. 1999;93(3):1124. [PubMed] [Google Scholar]
- 19. Morton LM, Turner JJ, Cerhan JR, et al. Proposed classification of lymphoid neoplasms for epidemiologic research from the International Lymphoma Epidemiology Consortium (InterLymph). Blood. 2007;110(2): 695–708 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Turner JJ, Morton LM, Linet MS, et al. InterLymph hierarchical classification of lymphoid neoplasms for epidemiologic research based on the WHO classification (2008): update and future directions. Blood. 2010;116(20):e90–e98 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Jarrett RF. Viruses and Hodgkin’s lymphoma. Ann Oncol. 2002;13(suppl 1):23–29 [DOI] [PubMed] [Google Scholar]
- 22. Haralambieva E, Boerma EJ, van Imhoff GW, et al. Clinical, immunophenotypic, and genetic analysis of adult lymphomas with morphologic features of Burkitt lymphoma. Am J Surg Pathol. 2005;29(8):1086–1094 [PubMed] [Google Scholar]
- 23. Hummel M, Bentink S, Berger H, et al. ; Molecular Mechanisms in Malignant Lymphomas Network Project of the Deutsche Krebshilfe. A biologic definition of Burkitt’s lymphoma from transcriptional and genomic profiling. N Engl J Med. 2006;354(23):2419–2430 [DOI] [PubMed] [Google Scholar]
- 24. Friedberg JW. Double-hit diffuse large B-cell lymphoma. J Clin Oncol. 2012;30(28):3439–3443 [DOI] [PubMed] [Google Scholar]
- 25. Vajdic CM, Falster MO, de Sanjose S, et al. Atopic disease and risk of non-Hodgkin lymphoma: an InterLymph pooled analysis. Cancer Res. 2009;69(16):6482–6489 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Jensen-Jarolim E, Achatz G, Turner MC, et al. AllergoOncology: the role of IgE-mediated allergy in cancer. Allergy. 2008;63(10):1255–1266 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Strik MC, de Koning PJ, Kleijmeer MJ, et al. Human mast cells produce and release the cytotoxic lymphocyte associated protease granzyme B upon activation. Mol Immunol. 2007;44(14):3462–3472 [DOI] [PubMed] [Google Scholar]
- 28. Anderson LA, Pfeiffer R, Warren JL, et al. Hematopoietic malignancies associated with viral and alcoholic hepatitis. Cancer Epidemiol Biomarkers Prev. 2008;17(11):3069–3075 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. de Sanjose S, Benavente Y, Vajdic CM, et al. Hepatitis C and non-Hodgkin lymphoma among 4784 cases and 6269 controls from the International Lymphoma Epidemiology Consortium. Clin Gastroenterol Hepatol. 2008;6(4):451–458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Marcucci F, Mele A. Hepatitis viruses and non-Hodgkin lymphoma: epidemiology, mechanisms of tumorigenesis, and therapeutic opportunities. Blood. 2011;117(6):1792–1798 [DOI] [PubMed] [Google Scholar]
- 31. Green J, Cairns BJ, Casabonne D, Wright FL, Reeves G, Beral V; Million Women Study collaborators. Height and cancer incidence in the Million Women Study: prospective cohort, and meta-analysis of prospective studies of height and total cancer risk. Lancet Oncol. 2011;12(8):785–794 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Walter RB, Brasky TM, Buckley SA, Potter JD, White E. Height as an explanatory factor for sex differences in human cancer. J Natl Cancer Inst. 2013;105(12):860–868 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Albanes D, Winick M. Are cell number and cell proliferation risk factors for cancer? J Natl Cancer Inst. 1988;80(10):772–774 [DOI] [PubMed] [Google Scholar]
- 34. Gunnell D, Okasha M, Smith GD, Oliver SE, Sandhu J, Holly JM. Height, leg length, and cancer risk: a systematic review. Epidemiol Rev. 2001;23(2):313–342 [DOI] [PubMed] [Google Scholar]
- 35. Morton LM, Zheng T, Holford TR, et al. ; InterLymph Consortium. Alcohol consumption and risk of non-Hodgkin lymphoma: a pooled analysis. Lancet Oncol. 2005;6(7):469–476 [DOI] [PubMed] [Google Scholar]
- 36. de Sanjosé S, Benavente Y, Nieters A, et al. Association between personal use of hair dyes and lymphoid neoplasms in Europe. Am J Epidemiol. 2006;164(1):47–55 [DOI] [PubMed] [Google Scholar]
- 37. Morton LM, Bernstein L, Wang SS, et al. Hair dye use, genetic variation in N-acetyltransferase 1 (NAT1) and 2 (NAT2), and risk of non-Hodgkin lymphoma. Carcinogenesis. 2007;28(8):1759–1764 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Zhang Y, de Sanjosé S, Bracci PM, et al. Personal use of hair dye and the risk of certain subtypes of non-Hodgkin lymphoma. Am J Epidemiol. 2008;167(11):1321–1331 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Zhang Y, Holford TR, Leaderer B, et al. Hair-coloring product use and risk of non-Hodgkin’s lymphoma: a population-based case-control study in Connecticut. Am J Epidemiol. 2004;159(2):148–154 [DOI] [PubMed] [Google Scholar]
- 40. Zahm SH, Weisenburger DD, Babbitt PA, Saal RC, Vaught JB, Blair A. Use of hair coloring products and the risk of lymphoma, multiple myeloma, and chronic lymphocytic leukemia. Am J Public Health. 1992;82(7):990–997 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Karami S, Bassig B, Stewart PA, et al. Occupational trichloroethylene exposure and risk of lymphatic and haematopoietic cancers: a meta-analysis. Occup Environ Med. 2013;70(8):591–599 [DOI] [PubMed] [Google Scholar]
- 42. Non-Hodgkin Lymphoma. In: Fritz A, Percy C, Jack A, eds. International Classification of Diseases for Oncology (ICD-O-3) 3rd ed. Geneva, Switzerland: World Health Organization; 2000:85–86 [Google Scholar]
- 43. Mbulaiteye SM, Pullarkat ST, Nathwani BN, et al. Epstein-Barr virus patterns in US Burkitt lymphoma tumors from the SEER residual tissue repository during 1979–2009. APMIS. 2014;122(1):5–15 [DOI] [PMC free article] [PubMed] [Google Scholar]