Abstract
Purpose:
Research suggests better survival among Hispanics with diffuse large B-cell lymphoma (DLBCL) compared to non-Hispanic Whites (NHW); however, less is known about racial/ethnic survival differences in follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL).
Methods:
We identified incident FL and CLL cases diagnosed between 2005–2016 at Montefiore Medical Center in the Bronx, NY. Cox proportional hazards regression assessed the association between race/ethnicity and all-cause mortality among FL and CLL separately.
Results:
Of the 201 FL patients, 39.3% were NHW, 19.4% NHB, and 41.3% Hispanic, with a similar distribution among CLL patients. After adjusting for International Prognostic Index factors, sex, and chemotherapy, Hispanics with FL had lower all-cause mortality compared to NHWs (HR=0.22; 95%CI 0.08–0.63), similar to prior DLBCL findings. All-cause mortality did not differ between NHBs and NHWs for FL or by any race/ethnicity for CLL.
Conclusions:
In our diverse, urban population, we found that Hispanic diagnosed with FL had lower all-cause mortality compared to NHWs. We found no significant difference in all-cause mortality between Hispanics and NHWs diagnosed with CLL. Our study adds to the growing literature on racial and ethnic differences in survival among Hispanics with hematologic malignancies.
Keywords: Follicular lymphoma, chronic lymphocytic leukemia, non-Hodgkin lymphoma, race/ethnicity, clinical research
Introduction
An estimated 102,810 incident cases of non-Hodgkin lymphoma (NHL) will be diagnosed in the United States in 2021, resulting in an estimated 21,940 deaths [1]. Follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL), two indolent subtypes of B-cell NHL, account for approximately 30% of mature B-cell NHL cases [2]. Given the indolent nature of FL, median survival among individuals diagnosed with early stage disease is 19 years, with chemotherapy reserved for patients presenting with symptomatic or advanced disease. Among FL patients diagnosed with late stage disease, there has been found to be no significant difference in mortality between patients who received early chemotherapy and those who were approached with watchful waiting [3]. Similarly, treatment is typically reserved for CLL patients presenting with advanced disease and may be treated with curative intent [4]. Population-based studies suggest lower survival among non-Hispanic Blacks (NHB) with FL and CLL compared to non-Hispanic Whites (NHW) [5]; however, less is known about the survival experience among Hispanics.
We previously reported lower all-cause mortality among Hispanics with diffuse large B-cell lymphoma (DLBCL) compared to NHWs [6] at Montefiore Medical Center (MMC) in the Bronx, NY. MMC is uniquely situated for studying racial/ethnic differences in health outcomes given its largely underserved, racially and ethnically diverse urban patient population. Our objective of this study was to investigate differences in all-cause mortality among Hispanics and NHBs diagnosed with FL or CLL compared to NHWs.
Methods
We identified incident FL and CLL cases in our retrospective cohort, the Hematological Malignancies Cohort at Montefiore Medical Center (HMCMMC) in the Bronx, NY, which has been previously described [6]. In brief, the cohort consists of all incident cases of NHL and multiple myeloma diagnoses at Montefiore Medical Center between 2005–2016 inclusive (n = 3,821). Montefiore’s Electronic Data Warehouse (EDW) was used to extract demographic and clinical data, including serum lactate dehydrogenase (LDH) at diagnosis and hemoglobin at diagnosis. Diagnosis and death data, as well as treatment regimen data, were identified from the Einstein/Montefiore Cancer Registry. Tumor-related data included stage at diagnosis, disease recurrence, and presence of extranodal disease (FL patients only). Clinical and demographic data, including race/ethnicity, flowed from MMC’s electronic medical record system, EPIC, and tumor-related information was from the Einstein/Montefiore Cancer Registry. Deaths were identified through the Einstein/Montefiore Cancer Registry, which utilized data from the National Death Index. Patients were followed through November 2018 with date of last contact defined by last entry into the EDW.
FL (n = 210) and CLL (n = 350) patients were categorized according to the World Health Organization (WHO) 2008 classification of hematologic malignancies using derived International Classification of Diseases for Oncology (ICD-O-3) codes [7,8]. In our study population, FL included follicular lymphoma, NOS (ICD-O-3 96903; n = 75), grade 1 follicular lymphoma (ICD-O-3 96953; n = 55), grade 2 follicular lymphoma (ICD-O-3 96913; n = 69), and grade 3 follicular lymphoma (ICD-O-3 96983; n = 81). CLL included chronic lymphocytic leukemia/small lymphocytic lymphoma (ICD-O-3 98233; n = 460) and malignant lymphoma, small B lymphocytic, NOS (ICD-O-3 96703; n = 15). Patients <18 years (n=4) or who were HIV+ (n=10) were excluded given limited generalizability to the immunocompetent adult population. Race/ethnicity was characterized as NHW, NHB, and Hispanic. Asian and Pacific Islanders (n=5) and American Indian/Alaskan Natives (n=2) were excluded given small sample size. Finally, individuals who self-identified as non-Hispanic/Latino unknown race (n=54) were excluded due to ambiguity in interpretation.
Baseline clinical factors included age at diagnosis, sex, stage at diagnosis, serum hemoglobin, lactate dehydrogenase (LDH), and white blood cell count with differential closest to time of diagnosis, body mass index (BMI), socioeconomic status (SES), time to receipt of chemotherapy, and time to disease progression. SES was characterized using a previously described neighborhood summary score that integrates dimensions of wealth, education, and occupation. Higher summary scores indicate increasing SES [9]. Baseline clinical factors were compared by race/ethnicity for FL and CLL patients independently, using ANOVA for continuous data and the χ2 test for categorical data. Kruskal-Wallis and Fisher exact tests were used when appropriate. Follicular Lymphoma International prognostic index (FLIPI) factors were assessed as categorical variables with age at diagnosis dichotomized at 60 years, stage at diagnosis I/II or III/IV, serum lactate dehydrogenase dichotomized at 225u/L, and serum hemoglobin dichotomized at 12g/dL [10].
All-cause mortality rates were calculated from age at diagnosis to age at death, with living patients censored at age at last contact. Person-years was similarly measured from age at diagnosis to age at death or age at last contact, with age calculated to the nearest day. Race/ethnicity stratified Kaplan-Meier curves were estimated and compared using log-rank tests. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated. Univariate and multivariable Cox proportional hazards models were built to assess the association between race/ethnicity and all-cause mortality among FL and CLL separately. FL and CLL base a priori models included race/ethnicity, age at diagnosis, sex, and receipt of chemotherapy. Our FL base model included additional Follicular Lymphoma International Prognostic Index factors (stage at diagnosis, serum hemoglobin). Additional covariates associated with all-cause mortality in the univariate models (p<0.20) were added using backward stepwise elimination to build our final multivariate model. The proportional hazards assumption was assessed using log-log plots and was found to be supported for all covariates. Confounders (change in beta >15%) were included in the final model. Our final most parsimonious FL model was adjusted for age at diagnosis, sex, stage at diagnosis, serum lactate dehydrogenase, serum hemoglobin, and receipt of chemotherapy. In our FL model, age at diagnosis was dichotomized at 60 years, lactate dehydrogenase at 225u/L, and hemoglobin at 12g/dL in order to adjust for FLIPI-specific prognostic factors [10]. Our final CLL model was adjusted for age at diagnosis, sex, serum lactate dehydrogenase, serum hemoglobin, socioeconomic status and receipt of chemotherapy. In our CLL model, age at diagnosis was dichotomized at 65 years, lactate dehydrogenase at 225u/L, and hemoglobin at 11g/dL to be consistent with CLL staging prognostics [11,12]. Additional variables that were considered in model building but were not included in the final models included lymphocyte-monocyte ratio, preferred language, socioeconomic status (not included in FL model), and extranodal disease status. Patients with missing data for any of the covariates included in our regression models were excluded from analysis, resulting in 146 patients in our multivariate adjusted FL model and 248 patients in our multivariate adjusted CLL model.
Subgroup analyses were performed for both the univariate and multivariate adjusted FL and CLL models. FL subgroups were examined by receipt of chemotherapy, age at diagnosis dichotomized at 60 years, sex, stage at diagnosis (early vs. advanced), and hemoglobin dichotomized at 12g/dL, consistent with FLIPI cut points. CLL subgroups were examined by receipt of chemotherapy, age at diagnosis dichotomized at 65 years, sex, and hemoglobin dichotomized at 11g/dL consistent with Rai staging.
In order to approximate disease severity, sensitivity analyses with primary endpoints of 5-year mortality, 10-year mortality, and receipt of chemotherapy were performed for the FL and CLL cohorts separately. Additionally, we ran an event-free survival analysis, with a primary endpoint of death or disease recurrence. We also ran a sensitivity analysis adjusting for smoking status, defined as ever versus never smoker. Given the loss of power associated with dichotomizing covariates [13], we ran an additional sensitivity analysis using age at diagnosis, serum hemoglobin, serum LDH, and SES as continuous covariates. Our final sensitivity analysis assessed the association between race/ethnicity and all-cause mortality adjusting for FLIPI score (age at diagnosis greater than 60 years, elevated LDH, hemoglobin less than 12 g/dL, and stage III/IV disease) ranging from 0–4 points. We first ran this using the full cohort, followed by only individuals scoring 0, 3, or 4 points (those whose risk category could not change with addition of extranodal disease site data).
All significance tests were 2-sided, and significance was defined as a p-value <0.05. Statistical analysis was performed using STATA, version 16.1 for Mac (StataCorp LP, College Station, Texas).
Results
Of the 201 individuals diagnosed with FL, 41.3% were Hispanic, 39.3% were NHW, and 19.4% NHB. FL patients were followed over a total of 386.6 person-years among NHWs, 172.2 person-years among NHBs, and 504.5 person-years among Hispanics. Median follow up time among FL patients who died was 3.1 years for NHWs, 3.0 years for NHBs, and 5.8 years for Hispanics (p = 0.27). Among those who survived, median follow up time was similar between NHWs (5.6 years), NHBs (3.8 years), and Hispanics (5.5 years; p = 0.10). FLIPI factors were similar across race/ethnicity, while NHW FL patients were older at time of diagnosis (p = 0.03) and had higher SES compared to NHB and Hispanics (p<0.001; Table 1). Additionally, a greater proportion of NHBs (61.8%) and Hispanics (59.7%) presented with stage III/IV disease, compared to NHWs (50%). Overall, Hispanics with FL had an improved survival probability at 5-years (93.0%) compared to NHWs (77.0%) and NHBs (80.6%, log-rank p = 0.008) (Figure 1). There was an increased risk of all-cause mortality among those who were older than 60 years (HR = 2.59, 95% CI 1.20–5.58) and had low serum hemoglobin (HR = 3.33, 95% CI 1.63–6.70) at time of diagnosis, consistent with FLIPI prognostic scoring (Table 2).
Table 1:
Clinical Characteristics of Follicular Lymphoma (n=201) and Chronic Lymphocytic Leukemia patients (n = 328) Diagnosed at Montefiore Medical Center in the Bronx, NY, Between 2005–2016.
| Follicular Lymphoma | Chronic Lymphocytic Leukemia | |||||||
|---|---|---|---|---|---|---|---|---|
| Characteristic€ | Non-Hispanic White (n = 79) | Non-Hispanic Black (n = 39) | Hispanic (n = 83) | p-value | Non-Hispanic White (n = 164) | Non-Hispanic Black (n = 85) | Hispanic (n = 79) | p-value |
|
| ||||||||
| International Prognostic Index (IPI) Factors | ||||||||
| Age at diagnosis, yr* | 65.3 (54.5, 78.1) | 61.4 (51.1, 69.5) | 57.9 (49.4, 66.0) | 0.03 | 74.1 (66.5, 81.2) | 66.5 (57.8, 75.0) | 67.2 (58.8, 75.4) | 0.002 |
| ≤60 years | 27 (34.2) | 17 (43.6) | 46 (55.4) | 23 (14.0) | 26 (30.6) | 23 (29.1) | ||
| >60 years | 52 (65.8) | 22 (56.4) | 37 (44.6) | 141 (86.0) | 59 (69.4) | 56 (70.9) | ||
| Ann Arbor Stage | 0.38 | |||||||
| Limited (I, II) | 35 (50.0) | 13 (38.2) | 31 (40.3) | --- | --- | --- | ||
| Advanced (III, IV) | 35 (50.0) | 21 (61.8) | 46 (59.7) | --- | --- | --- | ||
| SEER Site | 0.58 | |||||||
| Nodal | 71 (89.9) | 35 (89.7) | 78 (94.0) | |||||
| Extranodal | 8 (10.1) | 4 (10.3) | 5 (6.0) | --- | --- | --- | ||
| Serum Lactate Dehydrogenase* | 209 (179, 253) | 193 (176, 312) | 215 (172, 294) | 0.86 | 207 (177, 261) | 222 (181, 289) | 198 (170, 242) | 0.44 |
| ≤225 u/L | 37 (62.7) | 20 (60.6) | 40 (58.0) | 82 (61.7) | 38 (54.3) | 45 (64.3) | ||
| >225 u/L | 22 (37.3) | 13 (39.4) | 29 (42.0) | 51 (38.4) | 32 (45.7) | 25 (35.7) | ||
| Serum Hemoglobin* | 12.9 (11.7, 14.0) | 12.4 (10.8, 13.9) | 13.2 (11.5, 14.0) | 0.09 | 12.6 (10.9, 14.0) | 11.3 (10.4, 12.4) | 12.5 (11.2, 13.7) | < 0.001 |
| ≥12 g/dL | 56 (73.7) | 19 (52.8) | 55 (67.1) | 99 (65.1) | 26 (32.9) | 47 (60.3) | ||
| <12 g/dL | 20 (26.3) | 17 (47.2) | 27 (32.9) | 53 (38.9) | 53 (67.1) | 31 (39.7) | ||
| Sex | 0.55 | 0.008 | ||||||
| Male | 29 (36.7) | 12 (30.8) | 34 (41.0) | 110 (67.1) | 40 (47.1) | 45 (57.0) | ||
| Female | 50 (63.3) | 27 (69.2) | 49 (59.0) | 54 (32.9) | 45 (52.9) | 34 (43.0) | ||
| Receipt of Chemotherapy+ | 0.20 | 0.02 | ||||||
| Yes | 39 (49.4) | 20 (51.3) | 52 (62.7) | 31 (18.9) | 23 (27.1) | 28 (35.4) | ||
| No | 40 (50.6) | 19 (48.7) | 31 (37.4) | 133 (81.1) | 62 (72.9) | 51 (64.6) | ||
| Receipt of Radiation+ | 0.91 | |||||||
| Yes | 11 (13.9) | 5 (12.8) | 13 (15.7) | --- | --- | --- | ||
| No | 68 (86.1) | 34 (87.2) | 70 (84.3) | |||||
| Socioeconomic Status | < 0.001 | < 0.001 | ||||||
| > Median | 60 (80.0) | 17 (44.7) | 21 (25.3) | 106 (70.2) | 27 (32.5) | 23 (29.5) | ||
| ≤ Median | 15 (20.0) | 21 (55.3) | 62 (74.7) | 45 (29.8) | 56 (67.5) | 55 (70.5) | ||
| Mortality Status | 0.03 | 0.08 | ||||||
| Alive | 60 (76.0) | 32 (82.1) | 76 (91.6) | 127 (77.4) | 55 (64.7) | 60 (76.0) | ||
| Dead | 19 (24.0) | 7 (17.9) | 7 (8.4) | 37 (22.6) | 30 (35.3) | 19 (24.0) | ||
| Follow up Time | ||||||||
| Total Person-Years | 386.6 | 172.2 | 504.5 | 783.2 | 382.4 | 415.7 | ||
| Alive (years)* | 5.6 (2.2, 8.0) | 3.8 (3.1, 6.1) | 5.5 (3.7, 8.7) | .10 | 4.7 (2.6, 8.0) | 4.5 (2,9, 6.8) | 5.0 (3.1, 7.2) | 0.76 |
| Dead (years)* | 3.1 (0.9, 7.4) | 3.0 (1.2, 6.1) | 5.8 (1.1, 7.6) | 0.27 | 1.7 (0.3, 3.8) | 2.3 (0.7, 5.5) | 5.1 (2.1, 7.2) | 0.02 |
Reported as N (%), unless otherwise noted. Missing data Ann Arbor Stage (FL n = 20) on serum lactate dehydrogenase (FL n = 40; CLL n = 55), serum hemoglobin (FL n = 7; CLL n = 19), socioeconomic status (FL n = 5; CLL n = 16).
Receipt of chemotherapy or radiation defined as yes (ever received chemotherapy or radiation as part of treatment) or no (never received chemotherapy or radiation).
Reported as median (interquartile range).
Figure 1:
Survival Curves by Race/Ethnicity Among Follicular Lymphoma (n = 201) & Chronic Lymphocytic Leukemia (n = 328) Patients Diagnosed at Montefiore Medical Center in the Bronx, NY, Between 2005–2016.
Table 2:
Associations Between Race/Ethnicity and All-Cause Mortality Among Follicular Lymphoma & Chronic Lymphocytic Leukemia Patients diagnosed at Montefiore Medical Center in the Bronx NY, between 2005–2016.
| Follicular Lymphoma | Chronic Lymphocytic Leukemia | ||||
|---|---|---|---|---|---|
| HR of All-Cause Mortality (95% CI) | HR of All-Cause Mortality (95% CI)* | HR of All-Cause Mortality (95% CI) | HR of All-Cause Mortality (95% CI)** | ||
|
| |||||
| Race/Ethnicity | Race/Ethnicity | ||||
| Non-Hispanic White | 1.00 (reference) | 1.00 (reference) | Non-Hispanic White | 1.00 (reference) | 1.00 (reference) |
| Non-Hispanic Black | 0.86 (0.06–1.30) | 0.33 (0.08–1.32) | Non-Hispanic Black | 1.64 (1.01–2.65) | 1.45 (0.73–2.89) |
| Hispanic | 0.28 (0.12–0.67) | 0.22 (0.08–0.63) | Hispanic | 1.00 (0.58–1.74) | 1.08 (0.54–2.18) |
| Age at Diagnosis | Age at Diagnosis | ||||
| ≤ 60 years | 1.00 (reference) | 1.00 (reference) | ≤ 65 years | 1.00 (reference) | 1.00 (reference) |
| > 60 years | 2.59 (1.20–5.58) | 1.73 (0.69–4.34) | > 65 years | 3.28 (1.87–5.76) | 3.57 (1.76–7.24) |
| Sex | Sex | ||||
| Female | 1.00 (reference) | 1.00 (reference) | Female | 1.00 (reference) | 1.00 (reference) |
| Male | 1.16 (0.58–2.32) | 1.41 (0.56–3.57) | Male | 1.15 (0.74–1.78) | 1.58 (0.90–2.80) |
| Stage at Diagnosis | Socioeconomic Status | ||||
| Limited (I/II) | 1.00 (reference) | 1.00 (reference) | > Median | 1.00 (reference) | 1.00 (reference) |
| Advanced (III/IV) | 1.44 (0.65–3.18) | 2.00 (0.69–5.74) | ≤ Median | 0.70 (0.45–1.07) | 0.74 (0.41–1.33) |
| Serum Hemoglobin | Serum Hemoglobin | ||||
| ≥ 12 g/dL | 1.00 (reference) | 1.00 (reference) | > 11 g/dL | 1.00 (reference) | 1.00 (reference) |
| < 12 g/dL | 3.33 (1.63–6.70) | 6.15 (2.33–16.25) | ≤ 11 g/dL | 2.51 (1.60–3.95) | 2.66 (1.57–4.52) |
| Serum Lactate Dehydrogenase | Serum Lactate Dehydrogenase | ||||
| ≤ 225 u/L | 1.00 (reference) | 1.00 (reference) | ≤ 225 u/L | 1.00 (reference) | 1.00 (reference) |
| > 225 u/L | 1.13 (0.51–2.50) | 1.46 (0.56–3.78) | > 225 u/L | 1.15 (0.70–1.88) | 0.96 (0.57–1.61) |
| Receipt of Chemotherapy | Receipt of Chemotherapy | ||||
| Yes | 1.00 | 1.00 | Yes | 1.00 (reference) | 1.00 (reference) |
| No | 0.84 (0.42–1.69) | 0.52 (0.16–1.64) | No | 0.95 (0.59–1.53) | 1.17 (0.67–2.05) |
HR – hazard ratio; CI – confidence interval
Multivariate Cox Proportional Hazards model of follicular lymphoma patients adjusted for age at diagnosis, sex, serum lactate dehydrogenase dichotomized at 225u/L and hemoglobin dichotomized at 12 g/dL closest to diagnosis, and receipt of chemotherapy.
Multivariate Cox Proportional Hazards model of chronic lymphocytic leukemia patients adjusted for age at diagnosis, sex, serum lactate dehydrogenase dichotomized at 225u/L and hemoglobin dichotomized at 11 g/dL closest to diagnosis, receipt of chemotherapy, and socioeconomic status dichotomized at median.
When adjusting for age at diagnosis, sex and receipt of chemotherapy, we found that Hispanics experienced a lower risk of all-cause mortality compared to NHWs (HR = 0.32, 95% CI 0.13–0.78) and no significant difference in mortality between NHBs and NHWs (HR = 1.00, 95% CI 0.41–2.43). After further adjusting for FLIPI factors, we continued to find that Hispanics had lower all-cause mortality compared to NHWs (HR=0.22; 95% CI 0.08–0.63) (Table 2). Our findings persisted after estimating separate models by age at diagnosis (age ≤60 years HR=0.31, 95% CI 0.07–1.96; age >60 years HR=0.10; 95% CI 0.02–0.55), sex (male HR=0.22; 95% CI 0.05–0.94; female HR=0.18; 95% CI 0.03–1.08), stage at diagnosis (I/II HR=0.16; 95% CI 0.02–1.58; III/IV HR=0.19; 95% CI 0.05–0.71), lactate dehydrogenase (≤225 u/L HR=0.11; 95% CI 0.02–0.57; >225 u/L HR=0.26; 95% CI 0.04–1.5), and receipt of chemotherapy (received chemotherapy HR=0.31; 95% CI 0.09–1.08; no chemotherapy HR=0.06; 95% CI 0.004–0.89) (Table 3). We found no difference in all-cause mortality between NHBs and NHWs overall (HR=0.33; 95% CI 0.08–1.32) (Table 2) or in our subgroup analyses (Supplemental Table 1).
Table 3:
Associations Between Hispanics versus Non-Hispanic Whites with Follicular Lymphoma and All-Cause Mortality, Diagnosed at Montefiore Medical Center in the Bronx NY, between 2005–2016, by Prognostic Factors & Treatment.
| Follicular Lymphoma+ | |||||||
|---|---|---|---|---|---|---|---|
| Total (N) | Died (N) | Person-Years | HR of All-Cause Mortality (95% CI) for Hispanics Compared to Non-Hispanic Whites | p-value | Multivariate Adjusted HR of All-Cause Mortality (95% CI) for Hispanics Compared to Non-Hispanic Whites* | p-value | |
|
| |||||||
| All Patients | 146 | 21 | 808.1 | 0.28 (0.12–0.67) | 0.004 | 0.22 (0.08–0.63) | 0.005 |
| Received Chemotherapy | 96 | 15 | 578.0 | 0.35 (0.12–1.04) | 0.06 | 0.31 (0.09–1.08) | 0.07 |
| No Chemotherapy | 50 | 6 | 230.1 | 0.22 (0.05–1.00) | 0.05 | 0.06 (0.004–0.89) | 0.04 |
| Age ≤ 60 years | 71 | 8 | 427.8 | 0.35 (0.08–1.56) | 0.17 | 0.31 (0.07–1.96) | 0.24 |
| Age > 60 years | 75 | 13 | 380.3 | 0.30 (0.10–0.91) | 0.03 | 0.10 (0.02–0.55) | 0.008 |
| Female | 86 | 12 | 471,4 | 0.43 (0.13–1.43) | 0.17 | 0.18 (0.03–1.08) | 0.06 |
| Male | 75 | 14 | 408.9 | 0.16 (0.04–0.57) | 0.005 | 0.22 (0.05–0.94) | 0.04 |
| Stage I/II | 58 | 6 | 286.3 | 0.34 (0.07–1.72) | 0.19 | 0.16 (0.02–1.58) | 0.12 |
| Stage III/IV | 88 | 15 | 521.9 | 0.25 (0.09–0.72) | 0.01 | 0.19 (0.05–0.71) | 0.01 |
| Hemoglobin ≤ 12g/dL | 49 | 14 | 265.7 | 0.26 (0.08–0.81) | 0.02 | 0.13 (0.03–0.69) | 0.02 |
| Hemoglobin > 12g/dL | 97 | 7 | 542.4 | 0.09 (0.01–0.72) | 0.02 | 0.09 (0.01–0.81) | 0.03 |
HR – hazard ratio; CI – confidence interval
Reference group includes non-Hispanic Whites, adjusting for non-Hispanic Black race (not shown).
Multivariate Cox Proportional Hazards model of association between all-cause mortality and race/ethnicity in follicular lymphoma patients adjusted for age at diagnosis dichotomized at 60 years, sex, stage at diagnosis, serum lactate dehydrogenase dichotomized at 225u/L and hemoglobin dichotomized at 12 g/dL closest to diagnosis, and receipt of chemotherapy.
Sample size refers to multivariate models. For unadjusted models, all patients (n = 201, died = 33, person-years = 1,063.3), received chemotherapy (n = 111, died =19, person-years = 645.9), did not received chemotherapy (n = 90, died = 14, person-years = 417.4), age ≤ 60 years at diagnosis (n = 90, died = 9, person-years = 523.6), age > 60 years at diagnosis (n = 111, died = 24, person-years = 539.7), females (n = 126, died = 19, person-years = 654.4), males (n = 75, died = 14, person-years = 408.9), stage I/II at diagnosis (n = 79, died = 9, person-years = 383.0), stage III/IV at diagnosis (n = 102, died = 20, person-years = 592.6), serum hemoglobin <12 g/dL closest to diagnosis (n = 64, died = 18, person-years = 313.6), and hemoglobin ≥12 g/dL closest to diagnosis (n = 130, died = 13, person-years = 724.7).
We found similar results in our sensitivity analysis comparing racial and ethnic differences in 5-year and 10-year mortality. Compared to NHWs with FL, we found that Hispanics had lower 5-year (HR = 0.08, 95% CI 0.02–0.43) and 10-year mortality (HR = 0.22, 95% CI 0.08–0.63). Among NHBs with FL, there continued to be no significant difference in 5-year (HR = 0.29, 95% CI 0.06–1.30) or 10-year mortality (HR = 0.33, 95% CI 0.08–1.32) compared to NHWs (Supplemental Table 2). In order to further approximate differences in disease severity in our FL cohort, we ran an additional sensitivity analysis comparing receipt of chemotherapy by race/ethnicity. In our unadjusted model, we found no significant difference in receipt of chemotherapy by race/ethnicity. After adjusting for age at diagnosis, sex, advanced stage, serum LDH, serum hemoglobin, and SES, we found that non-Hispanic Blacks with FL (OR = 0.16, 95% CI 0.04–0.61), but not Hispanics (OR = 0.34, 95% CI 0.11–1.09), were less likely to receive chemotherapy compared to non-Hispanic Whites. We found that Hispanics had improved event-free survival compared to NHWs (HR = 0.44, 95% CI 0.20–0.98). There was no difference in event-free survival between NHBs and NHWs (HR = 0.73, 95% CI 0.24–2.23). In our model adjusting for age, serum hemoglobin and LDH as continuous covariates, we continued to find that Hispanics had lower all-cause mortality compared to non-Hispanic Whites (HR = 0.22; 95% CI 0.07–0.69) and no significant difference in all-cause mortality between non-Hispanic Blacks and non-Hispanic Whites (HR = 0.45; 95% CI 0.11–1.82). We found similar results when adjusting for FLIPI score (Hispanic HR = 0.29, 95% CI 0.11–0.80). Additionally, when excluding individuals with a score of 1 or 2 (whose risk category might change with addition of extranodal disease site), we continued to find suggestion of an association between all-cause mortality in Hispanics compared to non-Hispanic Whites (HR = 0.31, 95% CI 0.08–1.12), as well as between increased FLIPI score all-cause mortality (HR = 1.56, 95% CI 0.88–2.77).
Among the 328 CLL patients, 50.0% were NHW, 25.9% NHB, and 24.1% Hispanic (Table 1). Median follow up time was similar by race/ethnicity among those who were alive at the end of the study (NHW = 4.7 years, NHB = 4.5 years, Hispanics = 5.0 years; p = 0.76) and was longest among Hispanics who had died (NHW = 1.7 years, NHB = 2.3 years, Hispanics = 5.1 years; p = 0.02). Total person-years of follow up was greatest among NHWs (783 person-years), followed by Hispanics (415.7 person-years) and NHBs (382.4 person-years). NHW CLL patients tended to be older at time of diagnosis (p=0.002) and had a higher SES compared to NHB and Hispanics (p < 0.001). The majority of NHW (67.1%) and Hispanics (57.0%) were male, compared to 47.1% of NHB CLL patients (p=0.008). Although NHBs presented at a younger age (66.5 years) with a lower average hemoglobin (11.3 g/dL, p <0.001), a greater proportion of NHWs received chemotherapy (81.1%), compared to 72.9% of NHBs and 64.6% of Hispanics (p = 0.02).
Similar to our FL patient cohort, we found that older age (HR = 3.28, 95% CI 1.87–5.76) and low serum hemoglobin (HR = 2.51, 95% CI 1.60–3.95) at time of diagnosis were associated with increased all-cause mortality in our CLL patient population. After adjusting for age, sex, lactate dehydrogenase, hemoglobin, and receipt of chemotherapy, we found no difference in survival between Hispanic and NHW CLL patients (adjusted HR = 1.08; 95% CI 0.54–2.18) (Table 4). Among only who received chemotherapy, we also found no survival difference between Hispanic and NHW CLL patients (adjusted HR = 1.24; 95% CI 0.37–4.22) (Table 4). We found no significant difference in all-cause mortality between NHBs and NHWs overall (HR=1.45; 95% CI 0.73–2.89) or in our subgroup analyses (Supplemental Table 3).
Table 4:
Associations Between Hispanics versus Non-Hispanic Whites with Chronic Lymphocytic Leukemia and All-Cause Mortality Diagnosed at Montefiore Medical Center in the Bronx NY, between 2005–2016 by Prognostic Factors & Treatment.
| Chronic Lymphocytic Leukemia+ | |||||||
|---|---|---|---|---|---|---|---|
| Total (N) | Died (N) | Person-Years | HR of All-Cause Mortality (95% CI) for Hispanics Compared to Non-Hispanic Whites | p-value | Multivariate Adjusted HR of All-Cause Mortality (95% CI) for Hispanics Compared to Non-Hispanic Whites* | p-value | |
|
| |||||||
| All Patients | 248 | 62 | 1,237.0 | 1.00 (0.58–1.74) | 0.99 | 1.11 (0.55–2.22) | 0.77 |
| Received Chemotherapy | 71 | 21 | 396.0 | 1.14 (0.38–3.40) | 0.81 | 1.24 (0.37–4.22) | 0.73 |
| No Chemotherapy | 177 | 41 | 841.1 | 0.99 (0.51–1.94) | 0.98 | 0.99 (0.41–2.36) | 0.98 |
| Age ≤ 65 years | 76 | 11 | 482.1 | 2.19 (0.52–9.21) | 0.29 | 1.93 (0.23–15.90) | 0.54 |
| Age > 65 years | 172 | 51 | 754.9 | 1.07 (0.57–2.00) | 0.83 | 1.03 (0.48–2.21) | 0.94 |
| Female | 104 | 24 | 553.6 | 1.20 (0.50–2.87) | 0.68 | 0.87 (0.23–3.31) | 0.84 |
| Male | 144 | 38 | 683.4 | 0.89 (0.43–1.86) | 0.76 | 0.85 (0.35–2.05) | 0.71 |
| Hemoglobin ≤ 11g/dL | 76 | 29 | 292.0 | 0.50 (0.14–1.81) | 0.30 | 0.42 (0.10–1.72) | 0.23 |
| Hemoglobin > 11g/dL | 172 | 33 | 945.0 | 1.07 (0.55–2.09) | 0.84 | 1.45 (0.63–3.34) | 0.39 |
HR – hazard ratio; CI – confidence interval
Reference group includes non-Hispanic Whites, adjusting for non-Hispanic Black race (not shown).
Multivariate Cox Proportional Hazards model of chronic lymphocytic leukemia patients adjusted for age at diagnosis, sex, serum lactate dehydrogenase dichotomized at 225u/L and hemoglobin dichotomized at 11 g/dL closest to diagnosis, receipt of chemotherapy, and socioeconomic status dichotomized at median.
Sample size refers to multivariate models. For unadjusted models, all patients (n = 328, died = 86, person-years = 1,581.3), received chemotherapy (n = 82, died =23, person-years = 443.3), did not received chemotherapy (n = 246, died = 63, person-years = 1,138.0), age ≤ 65 years at diagnosis (n = 105, died = 15, person-years = 644.0), age > 65 years at diagnosis (n = 223, died = 71, person-years = 937.3), females (n = 133, died = 35, person-years = 684.7), males (n = 195, died = 51, person-years = 896.6), serum hemoglobin ≤11 g/dL closest to diagnosis (n = 92, died = 33, person-years = 334.1), and hemoglobin >11 g/dL closest to diagnosis (n = 217, died = 49, person-years = 1,168.0).
In our sensitivity analyses focused on 5-year and 10-year mortality, we continued to find similar mortality rates by race/ethnicity. In our full CLL cohort, we found that Hispanics had similar 5-year (HR = 0.94, 95% CI 0.39–2.30) and 10-year (HR = 1.09, 95% CI 0.54–2.20) mortality compared to NHWs. Among those who received chemotherapy, Hispanics and NHWs continued to have similar 5-year (HR = 1.09, 95% CI 0.16–7.63) and 10-year (HR = 1.37, 95% CI 0.41–4.60) mortality rates. Similarly, NHBs had similar 5-year (HR = 1.51, 95% CI 0.68–3.35) and 10-year mortality rates compared to NHWs. Among those who received chemotherapy, there continued to be similar 5-year (HR = 4.09, 95% CI 0.80–20.97) and 10-year (HR = 1.89, 95% CI 0.55–6.57) mortality rates (Supplemental Table 4). Among CLL patients, Hispanics were more likely to receive chemotherapy compared to NHWs (OR = 2.36, 95% CI 1.29–4.31), but there was no significant difference between NHBs and NHWs. After adjusting for age at diagnosis, sex, serum LDH, serum hemoglobin and SES, however, we no longer found a difference in receipt of chemotherapy by race or ethnicity (Hispanic OR = 1.44, 95% CI 0.63–3.27; NHB OR = 0.90, 95% CI 0.40–2.04). Event-free survival was similar by race and ethnicity (Hispanic HR = 1.09, 95% CI 0.56–2.14; NHB HR = 1.58, 95% CI 0.82–3.04). In our analysis using continuous covariates, we found no significant association between all-cause mortality and Hispanic ethnicity (HR = 1.23, 95% CI 0.60–2.53) or NHB race (HR = 1.90, 95% CI 0.94–3.84) compared to NHWs.
In the current study, we found suggestion of an association between history of smoking and all-cause mortality among our FL (HR = 1.53, 95% CI 0.73–3.20) and CLL (HR = 1.15, 95% CI 0.68–1.95) patient cohorts, although the association was not statistically significant. After further adjusting for smoking status in our FL analysis, we continued to find lower all-cause mortality among Hispanics compared to non-Hispanic Whites (HR = 0.23, 95% CI 0.07–0.77). There continued to be no association between race/ethnicity and all-cause mortality after further adjusting for smoking status in our full CLL cohort (Hispanic HR = 0.62, 95% CI 0.23–1.68; NHB HR = 1.70, 95% CI 0.77–3.77), as well as among only those who received chemotherapy (Hispanic HR = 0.51, 95% CI 0.11–2.40; NHB HR = 1.82, 95% CI 0.45–7.29).
Discussion
In our diverse population, Hispanics diagnosed with FL had lower all-cause mortality compared to NHWs. Similar to our previous report among Hispanic diffuse large B-cell lymphoma (DLBCL) patients [6], our findings persisted after adjusting for clinical factors, as well as stratifying by prognostic factors and receipt of chemotherapy. In contrast, there was no significant difference in all-cause mortality between Hispanics and NHWs diagnosed with CLL. To the best of our knowledge, this is the first study to show better survival among Hispanics with FL compared to NHWs. Overall, these findings contribute key insights into our understanding of racial and ethnic differences in B-cell lymphoma survival.
The growing racial and ethnic diversity in the Bronx population, with approximately 54% identifying as Hispanic/Latino [14], as well as across the United States, understates the need to better understand health outcomes in Hispanics/Latinos. Prior research has found longer progression-free survival among Hispanics with FL compared to NHWs [15]. In our study population, Hispanics and NHBs tended to be younger at time of diagnosis, but with a greater proportion of advanced disease (59.7% and 61.8%) compared to NHWs (50%). These findings were consistent with national data suggesting earlier age at diagnosis among NHBs compared to NHWs [5,16]. Despite the greater proportion of advanced disease among Hispanics, our findings were consistent among Hispanics diagnosed with early stage (HR = 0.16; 95% CI 0.02–1.58) and late stage disease (HR = 0.19; 95% CI 0.05–0.71). This suggests that differences in stage at diagnosis do not explain the survival difference between Hispanics and NHWs.
While NHWs in our population tended to have a higher median SES, overall poverty rates in the Bronx are more than double national rates (27.4% versus 13.1%) [14]. Given that racial/ethnic disparities in health outcomes may be partly attributed to differences in SES [17], the high poverty rates across race and ethnicity in the Bronx may provide a potential explanation as to why we found similar all-cause mortality rates between NHBs and NHWs. In the current study however, there was no evidence that lower SES resulted in delayed time to receipt of chemotherapy among our FL (1.21 vs. 1.16 years, p = 0.92) or CLL patients (3.00 vs 3.75 years, p = 0.40). This finding may also be attributable to the access to care available through Montefiore Medical Center, an academic, safety-net hospital that provides high-quality care regardless of ability to afford care. Given that poverty rates were higher among Hispanic FL patients compared to NHWs, we believe SES does not fully explain the observed survival disparities.
While more research is needed on genetic differences in disease susceptibility and prognosis, data suggest that genetic variants associated with disease risk may vary by NHL subtype and race/ethnicity [18]. Given the distinct pathophysiology of FL and CLL, it is plausible that differences in genetic susceptibility to disease progression may explain the observed survival disparities. For example, genetic variation in the class II major histocompatibility complex has been associated with risk of FL, but not DLBCL or CLL [19]. Furthermore, 30–40% of FL cases progress to transformed FL, an aggressive malignancy with poor prognosis. Transformed FL is associated with genetic mutations resulting in deregulation of cell-cycle progression and DNA-damage response [20]. Not much is known about racial/ethnic differences in risk of FL transformation and its implications on overall survival.
Current treatment regimens are based on clinical trials in which Hispanics and NHBs are underrepresented, potentially influencing survival outcomes. Key clinical trials assessing the efficacy of Rituximab maintenance [21] and Rituximab plus Lenalidomide [22] did not describe the racial/ethnic composition of their study populations, potentially limiting the generalizability of findings. For example, a large study comparing racial/ethnic differences in NHL survival found that not receiving first-course chemotherapy was a significant mortality risk factor in NHBs and Hispanics, but provided a protective effect in Asians and Pacific Islanders [23], suggesting a potential difference in response to chemotherapy by race/ethnicity. In our current study, we attempt to provide an improved understanding of racial/ethnic differences in all-cause mortality among individuals diagnosed with FL and CLL. However, given the relatively limited size of our cohort, and its unique setting (i.e., poorest urban county in the United States coupled with a major medical center that can operate as a safety-net institution), our findings may not be broadly generalizable to the United States. Additional research is needed, particularly among rural populations.
Information on FL- and CLL-specific risk factors, including family history of hematologic malignancies, certain autoimmune diseases and environmental exposures to organic compounds, may provide insight into differences in age at diagnosis between NHWs, and NHBs and Hispanics [24,25]. National data suggest higher rates of exposure to endocrine-disrupting chemicals, including pesticides and flame retardants, among NHBs and Hispanics resulting in racial/ethnic disparities in disease burden [26]. Additionally, prior research suggests that residential benzene exposure (i.e., hazardous waste sites) may be associated with increased CLL risk and that this risk is greater in NHBs compared to NHWs [27]. In New York City, residential air benzene concentration and pesticide exposure, another potential risk factor for FL and CLL [28], have been strongly associated with poverty and low median household income. Given that a greater proportion of NHBs and Hispanics in our study had low SES, it is possible that poverty-associated environmental exposures may have influenced younger age at diagnosis among Hispanics and NHBs.
Less is known, however, about the association between occupational exposures and disease mortality. A history of cigarette smoking has been found to be associated with inferior overall survival among FL patients [29]. In the current study, we found suggestion of an association between history of smoking and all-cause mortality among our FL and CLL patient cohorts, although the association was not statistically significant. After further adjusting for smoking status in our FL analysis, we continued to find lower all-cause mortality among Hispanics compared to non-Hispanic Whites. There continued to be no association between race/ethnicity and all-cause mortality after further adjusting for smoking status in our full CLL cohort, as well as among only those who received chemotherapy. These findings must be interpreted with caution, however, given the incomplete nature of EHR smoking status data. Our study was limited, however, by a lack of data on residential and occupational exposures that may have contributed to disparities in disease prognosis.
Our study was strengthened by the racial and ethnic diversity of our patient population, providing an approximately equal distribution of NHWs and Hispanics. Despite the diversity of our study, we may have been underpowered by the limited sample size in our FL (79 NHW, 39 NHB, 83 Hispanic patients) and CLL (164 NHW, 85 NHB, 79 Hispanic patients) populations. The lack of observed disparities in CLL patient mortality may also have been attributable to the indolent nature of the disease and mortality being due to other causes. Additionally, we observed greater healthcare utilization among Hispanics and NHBs with CLL compared to NHWs, as defined by proportion of patients receiving chemotherapy at MMC. This may reflect MMC’s ability as a safety net hospital to provide care for patients regardless of insurance status. Our subgroup analyses, in particular, were further limited due to small sample size and, as such, should be interpreted cautiously for general trends. Our study was further strengthened by the utilization of multiple data sources (Montefiore Medical Center EMR, Montefiore/Einstein Cancer Registry, and National Death Index), reducing the potential for missing data and improving overall patient follow up. However, we had limited ability to assess potential loss to follow up. Additionally, diagnosis and treatment data were extracted from the EMR, which may potentially lead to fewer errors than if the information had been extracted from claims data [30].
A key limitation of our study was insufficient data on cause-specific mortality and comorbidities, which would provide a more nuanced understanding of disparities in health outcomes among patients diagnosed with FL and CLL. Incorporation of cause-specific mortality and comorbidity burden in future studies, such as with the Charlson Comorbidity Index, may help further characterize our findings. Additionally, we were unable to distinguish between chemotherapy regimen or grade of disease among our FL patients, limiting our findings given that grade 3 FL has previously been associated with distinct prognostic implications [3]. In our sensitivity analyses assessing 5-year and 10-year mortality, however, we found a similar survival benefit among Hispanics compared to NHWs suggesting that the survival advantage persists among individuals with aggressive disease (Supplemental Table 1). Additionally, the current study was unable to account for Rai stage in our CLL patient cohort, limiting the generalizability of our findings and also providing a potential source of bias. However, our subgroup analysis including only patients who received chemotherapy – likely those with advanced disease (Rai stage ≥3) –continued to show no significant association between race/ethnicity and all-cause mortality.
In this study, we attempt to highlight key racial/ethnic differences in all-cause mortality among individuals diagnosed with FL and CLL, however, residual confounding from measured and unmeasured confounders, such as with the use of dichotomized clinical and demographic characteristics, as well as unmeasured patient comorbidities, cannot be ruled out. In particular, national registry data suggest worse 5-year survival among NHBs with Stage IV CLL compared to NHWs, but no racial/ethnic differences in survival among individuals with Stage I, II, or III disease [5]. It is possible the lack of observed differences in CLL outcome by race/ethnicity may be due to our inability to adjust for disease stage and Rai prognostic factors, resulting in residual confounding. Our findings among Hispanics with FL are consistent with research highlighting survival differences in diffuse large B-cell lymphoma [6], an aggressive lymphoma, and non-small cell lung cancer in the Bronx [31], further highlighting the importance of understanding health outcomes Hispanics in urban centers with equitable access to care. Further research that incorporates additional clinical markers (Rai stage, serum beta-2 microglobulin) is needed to better understand racial/ethnic differences in CLL-specific mortality.
In our diverse, urban population, we found that Hispanics with FL had lower all-cause mortality compared to NHWs. In contrast, we found no significant difference in all-cause mortality between Hispanics and NHWs diagnosed with CLL. Our study adds to the growing literature on racial and ethnic differences in survival among Hispanics with hematologic malignancies. Further research is needed to identify potential differences in genetic markers, socioeconomic status, or treatment response by race/ethnicity that may be underlying this survival differential.
Supplementary Material
Funding:
This work was supported by a grant from the National Institutes of Health/National Center for Advancing Translational Science (NCATS) Einstein-Montefiore CTSA (Grant Number UL1TR001073).
Footnotes
Declarations:
Ethics approval: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This study was approved by the Institutional Review Board at Albert Einstein College of Medicine (Number 2018–9853).
Consent to participate: N/A
Consent for publication: N/A
Code availability: N/A
Conflicts of interest: The authors declare no conflicts of interest.
Availability of data and material:
Available upon reasonable request.
References:
- 1.American Cancer Society (2021) Cancer Facts & Figures 2021. American Cancer Society, Atlanta [Google Scholar]
- 2.Swerdlow SH, Campo E, Pileri SA, Harris NL, Stein H, Siebert R, Advani R, Ghielmini M, Salles GA, Zelenetz AD, Jaffe ES (2016) The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood 127 (20):2375–2390. doi: 10.1182/blood-2016-01-643569 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Freedman A, Jacobsen E (2020) Follicular lymphoma: 2020 update on diagnosis and management. Am J Hematol 95 (3):316–327. doi: 10.1002/ajh.25696 [DOI] [PubMed] [Google Scholar]
- 4.Hallek M (2019) Chronic lymphocytic leukemia: 2020 update on diagnosis, risk stratification and treatment. Am J Hematol 94 (11):1266–1287. doi: 10.1002/ajh.25595 [DOI] [PubMed] [Google Scholar]
- 5.Li Y, Wang Y, Wang Z, Yi D, Ma S (2015) Racial differences in three major NHL subtypes: descriptive epidemiology. Cancer Epidemiol 39 (1):8–13. doi: 10.1016/j.canep.2014.12.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Blansky D, Fazzari M, Mantzaris I, Rohan T, Hosgood HD (2020) Racial and ethnic differences in diffuse large B-cell lymphoma survival among an underserved, urban population. Leuk Lymphoma:1–9. doi: 10.1080/10428194.2020.1839656 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Campo E, Swerdlow SH, Harris NL, Pileri S, Stein H, Jaffe ES (2011) The 2008 WHO classification of lymphoid neoplasms and beyond: evolving concepts and practical applications. Blood 117 (19):5019–5032. doi: 10.1182/blood-2011-01-293050 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jaffe ES (2009) The 2008 WHO classification of lymphomas: implications for clinical practice and translational research. Hematology Am Soc Hematol Educ Program:523–531. doi: 10.1182/asheducation-2009.1.523 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Diez Roux AV, Merkin SS, Arnett D, Chambless L, Massing M, Nieto FJ, Sorlie P, Szklo M, Tyroler HA, Watson RL (2001) Neighborhood of residence and incidence of coronary heart disease. N Engl J Med 345 (2):99–106. doi: 10.1056/NEJM200107123450205 [DOI] [PubMed] [Google Scholar]
- 10.Solal-Celigny P, Roy P, Colombat P, White J, Armitage JO, Arranz-Saez R, Au WY, Bellei M, Brice P, Caballero D, Coiffier B, Conde-Garcia E, Doyen C, Federico M, Fisher RI, Garcia-Conde JF, Guglielmi C, Hagenbeek A, Haioun C, LeBlanc M, Lister AT, Lopez-Guillermo A, McLaughlin P, Milpied N, Morel P, Mounier N, Proctor SJ, Rohatiner A, Smith P, Soubeyran P, Tilly H, Vitolo U, Zinzani PL, Zucca E, Montserrat E (2004) Follicular lymphoma international prognostic index. Blood 104 (5):1258–1265. doi: 10.1182/blood-2003-12-4434 [DOI] [PubMed] [Google Scholar]
- 11.Rai KR, Sawitsky A, Cronkite EP, Chanana AD, Levy RN, Pasternack BS (1975) Clinical staging of chronic lymphocytic leukemia. Blood 46 (2):219–2341139039 [Google Scholar]
- 12.International CLLIPIwg (2016) An international prognostic index for patients with chronic lymphocytic leukaemia (CLL-IPI): a meta-analysis of individual patient data. Lancet Oncol 17 (6):779–790. doi: 10.1016/S1470-2045(16)30029-8 [DOI] [PubMed] [Google Scholar]
- 13.Altman DG, Royston P (2006) The cost of dichotomising continuous variables. BMJ 332 (7549):1080. doi: 10.1136/bmj.332.7549.1080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.U.S. Census Bureau (2019) QuickFacts Bronx County (Bronx Burough) New York. U.S. Census Bureau, [Google Scholar]
- 15.Nabhan C, Byrtek M, Taylor MD, Friedberg JW, Cerhan JR, Hainsworth JD, Miller TP, Hirata J, Link BK, Flowers CR (2012) Racial differences in presentation and management of follicular non-Hodgkin lymphoma in the United States: report from the National LymphoCare Study. Cancer 118 (19):4842–4850. doi: 10.1002/cncr.27513 [DOI] [PubMed] [Google Scholar]
- 16.Shenoy PJ, Malik N, Sinha R, Nooka A, Nastoupil LJ, Smith M, Flowers CR (2011) Racial differences in the presentation and outcomes of chronic lymphocytic leukemia and variants in the United States. Clin Lymphoma Myeloma Leuk 11 (6):498–506. doi: 10.1016/j.clml.2011.07.002 [DOI] [PubMed] [Google Scholar]
- 17.Tao L, Foran JM, Clarke CA, Gomez SL, Keegan TH (2014) Socioeconomic disparities in mortality after diffuse large B-cell lymphoma in the modern treatment era. Blood 123 (23):3553–3562. doi: 10.1182/blood-2013-07-517110 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Skibola CF, Bracci PM, Nieters A, Brooks-Wilson A, de Sanjose S, Hughes AM, Cerhan JR, Skibola DR, Purdue M, Kane E, Lan Q, Foretova L, Schenk M, Spinelli JJ, Slager SL, De Roos AJ, Smith MT, Roman E, Cozen W, Boffetta P, Kricker A, Zheng T, Lightfoot T, Cocco P, Benavente Y, Zhang Y, Hartge P, Linet MS, Becker N, Brennan P, Zhang L, Armstrong B, Smith A, Shiao R, Novak AJ, Maynadie M, Chanock SJ, Staines A, Holford TR, Holly EA, Rothman N, Wang SS (2010) Tumor necrosis factor (TNF) and lymphotoxin-alpha (LTA) polymorphisms and risk of non-Hodgkin lymphoma in the InterLymph Consortium. Am J Epidemiol 171 (3):267–276. doi: 10.1093/aje/kwp383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Skibola CF, Bracci PM, Halperin E, Conde L, Craig DW, Agana L, Iyadurai K, Becker N, Brooks-Wilson A, Curry JD, Spinelli JJ, Holly EA, Riby J, Zhang L, Nieters A, Smith MT, Brown KM (2009) Genetic variants at 6p21.33 are associated with susceptibility to follicular lymphoma. Nat Genet 41 (8):873–875. doi: 10.1038/ng.419 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pasqualucci L, Khiabanian H, Fangazio M, Vasishtha M, Messina M, Holmes AB, Ouillette P, Trifonov V, Rossi D, Tabbo F, Ponzoni M, Chadburn A, Murty VV, Bhagat G, Gaidano G, Inghirami G, Malek SN, Rabadan R, Dalla-Favera R (2014) Genetics of follicular lymphoma transformation. Cell Rep 6 (1):130–140. doi: 10.1016/j.celrep.2013.12.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Salles G, Seymour JF, Offner F, Lopez-Guillermo A, Belada D, Xerri L, Feugier P, Bouabdallah R, Catalano JV, Brice P, Caballero D, Haioun C, Pedersen LM, Delmer A, Simpson D, Leppa S, Soubeyran P, Hagenbeek A, Casasnovas O, Intragumtornchai T, Ferme C, da Silva MG, Sebban C, Lister A, Estell JA, Milone G, Sonet A, Mendila M, Coiffier B, Tilly H (2011) Rituximab maintenance for 2 years in patients with high tumour burden follicular lymphoma responding to rituximab plus chemotherapy (PRIMA): a phase 3, randomised controlled trial. Lancet 377 (9759):42–51. doi: 10.1016/S0140-6736(10)62175-7 [DOI] [PubMed] [Google Scholar]
- 22.Morschhauser F, Fowler NH, Feugier P, Bouabdallah R, Tilly H, Palomba ML, Fruchart C, Libby EN, Casasnovas RO, Flinn IW, Haioun C, Maisonneuve H, Ysebaert L, Bartlett NL, Bouabdallah K, Brice P, Ribrag V, Daguindau N, Le Gouill S, Pica GM, Martin Garcia-Sancho A, Lopez-Guillermo A, Larouche JF, Ando K, Gomes da Silva M, Andre M, Zachee P, Sehn LH, Tobinai K, Cartron G, Liu D, Wang J, Xerri L, Salles GA, Investigators RT (2018) Rituximab plus Lenalidomide in Advanced Untreated Follicular Lymphoma. N Engl J Med 379 (10):934–947. doi: 10.1056/NEJMoa1805104 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kent EE, Breen N, Lewis DR, de Moor JS, Smith AW, Seibel NL (2015) US trends in survival disparities among adolescents and young adults with non-Hodgkin lymphoma. Cancer Causes Control 26 (8):1153–1162. doi: 10.1007/s10552-015-0609-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Cerhan JR (2020) Epidemiology of Follicular Lymphoma. Hematol Oncol Clin North Am 34 (4):631–646. doi: 10.1016/j.hoc.2020.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Slager SL, Benavente Y, Blair A, Vermeulen R, Cerhan JR, Costantini AS, Monnereau A, Nieters A, Clavel J, Call TG, Maynadie M, Lan Q, Clarke CA, Lightfoot T, Norman AD, Sampson JN, Casabonne D, Cocco P, de Sanjose S (2014) Medical history, lifestyle, family history, and occupational risk factors for chronic lymphocytic leukemia/small lymphocytic lymphoma: the InterLymph Non-Hodgkin Lymphoma Subtypes Project. J Natl Cancer Inst Monogr 2014 (48):41–51. doi: 10.1093/jncimonographs/lgu001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Attina TM, Malits J, Naidu M, Trasande L (2019) Racial/ethnic disparities in disease burden and costs related to exposure to endocrine-disrupting chemicals in the United States: an exploratory analysis. J Clin Epidemiol 108:34–43. doi: 10.1016/j.jclinepi.2018.11.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Boberg E, Lessner L, Carpenter DO (2011) The role of residence near hazardous waste sites containing benzene in the development of hematologic cancers in upstate New York. Int J Occup Med Environ Health 24 (4):327–338. doi: 10.2478/s13382-011-0037-8 [DOI] [PubMed] [Google Scholar]
- 28.Kachuri L, Beane Freeman LE, Spinelli JJ, Blair A, Pahwa M, Koutros S, Hoar Zahm S, Cantor KP, Weisenburger DD, Pahwa P, Dosman JA, McLaughlin JR, Demers PA, Harris SA (2020) Insecticide use and risk of non-Hodgkin lymphoma subtypes: A subset meta-analysis of the North American Pooled Project. Int J Cancer 147 (12):3370–3383. doi: 10.1002/ijc.33164 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ollberding NJ, Evens AM, Aschebrook-Kilfoy B, Caces DB, Weisenburger DD, Smith SM, Chiu BC (2013) Pre-diagnosis cigarette smoking and overall survival in non-Hodgkin lymphoma. Br J Haematol 163 (3):352–356. doi: 10.1111/bjh.12512 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Devoe JE, Gold R, McIntire P, Puro J, Chauvie S, Gallia CA (2011) Electronic health records vs Medicaid claims: completeness of diabetes preventive care data in community health centers. Ann Fam Med 9 (4):351–358. doi: 10.1370/afm.1279 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Klugman M, Xue X, Ginsberg M, Cheng H, Rohan T, Hosgood HD 3rd, (2020) Hispanics/Latinos in the Bronx Have Improved Survival in Non-Small Cell Lung Cancer Compared with Non-Hispanic Whites. J Racial Ethn Health Disparities 7 (2):316–326. doi: 10.1007/s40615-019-00660-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Available upon reasonable request.