TO THE EDITOR:

Acute myeloid leukemia (AML) is the most common acute leukemia in adults, with a yearly incidence of 20 000 cases and 10 000 deaths.1 Current risk stratification of AML is driven by cytogenetic and molecular features.2 In addition to these features, environmental factors such as previous cytotoxic therapy, county-level income, and social deprivation index affect survival outcomes in AML.3-5 An analysis by our consortium of 822 patients with newly diagnosed AML across Chicago-area academic centers demonstrated inferior survival among non-Hispanic Black (NHB) and Hispanic patients compared with non-Hispanic White (NHW) patients. A multilevel analysis identified structural racism as the largest mediator of leukemia-related death disparities in NHB and Hispanic patients.6 However, the biologic pathways underpinning this observation remain unknown.

Air pollution disproportionately affects racial-ethnic minorities and has been demonstrated to alter DNA methylation patterns in patients, which in turn influence disease biology.7,8 In patients with nonsmall cell lung cancer, increased exposure to air pollution was associated with higher incidence of TP53 mutations.9 Higher levels of pollution exposure have also been associated with adverse outcomes in multiple myeloma and increased risk of leukemia in children.10,11 Specific exposures implicated in leukemia development/mortality and disparate exposure in historically redlined cities include 1,3-butadiene, benzene, diesel particulate matter (PM), polycyclic aromatic hydrocarbons and polycyclic organic matter (PAHPOM), and PM 2.5 μm (PM 2.5).11-14 Thus, we analyzed exposure to these pollutants by race/ethnicity, in addition to investigating the impact of exposure upon disease characteristics and survival outcomes in AML. All institutions received institutional review board approval for this retrospective study.

In this retrospective study, adult patients with AML diagnosed between 2012 and 2018 from 6 academic cancer centers in the Chicagoland area were included as described in the study of Abraham et al.6 Demographic information, disease characteristics, and outcome measures were collected; risk stratification was done via European LeukemiaNet (ELN) 2010 criteria, given the time period analyzed.15 Patient census tracts were joined with tract-level measures of pollution obtained from the 2014 National Air Toxics Assessment for the following pollutants: 1,3-butadiene, benzene, diesel PM, and PAHPOM.16 Additionally, PM 2.5 exposure was obtained using the 2014 averaged daily census tract measures from the Center for Disease Control National Environmental Public Health Tracking Network.17 

Pollutant exposures by racial/ethnic groups were determined using a Kruskall-Wallis rank sum test and pairwise Wilcoxon rank sum tests. Similarly, pollutant exposure by age of diagnosis was compared using pairwise Wilcoxon rank sum tests. Univariate logistic regressions for each pollutant were performed for 2010 ELN AML risk category and TP53 mutation status. Median overall survival (mOS) among patients in the most and least polluted census tract quartiles for each pollutant were determined, and survival by pollutant quartile was compared using Kaplan-Meier survival analysis. We similarly analyzed pollutant exposures by racial/ethnic groups on a cohort of patients with newly diagnosed AML between the years 2018 and 2022 at The University of Chicago that had cytogenetic and molecular testing performed to risk stratify disease based on the 2022 ELN AML criteria, which incorporates several molecular mutations to better refine risk assessment. Census tract-level pollutant measures from 2014 were rasterized and aggregated to 2022 census tract boundaries by averaging pixels within 2022 census tract polygons. Multivariate multinomial logistic regressions were performed for 2022 ELN risk category and pollutant exposure while controlling for age, gender, and race/ethnicity.

Our multi-institutional data set contained 789 adult patients with AML and pollution data with a median age of diagnosis of 62 years. Sixty percent of patients were NHW, 15.3% were NHB, and 15% were Hispanic. By ELN 2010 classification, 29.6% of our patients had adverse-risk disease, and 8.4% had pathogenic TP53 mutations among those who were tested (Table 1). Exposure to each pollutant differed significantly between racial/ethnic groups (P < .05). NHB and Hispanic patients had significantly greater median exposure to 1,3-butadiene, benzene, diesel PM, PAHPOM, and PM 2.5 (P < .05) than NHW patients. Furthermore, Hispanic patients had significantly greater median exposure to 1,3-butadiene and benzene than NHB patients, whereas NHB patients had significantly greater median exposure to PM 2.5 (P < .05) than Hispanic patients (Table 2; supplemental Tables 3 and 4A-E). In univariate analysis, no pollutant exposure was found to be significantly associated with either ELN 2010 risk categorization or with TP53 mutation status.

Table 1.

Patient demographics and disease characteristics

Patient demographics2012-2018 multi-institution cohort (N = 789)2018-2022 The University of Chicago cohort (N = 330)
Age at AML diagnosis, y n = 788 n = 330 
Median age (range) 62 (18-95) 69 (18-92) 
Age 18-59, n (%) 350 (44) 78 (24) 
Age 60-74, n (%) 313 (40) 160 (48) 
Age 75+, n (%) 125 (16) 92 (28) 
Sex, n (%)   
Male 427 (54) 192 (58) 
Female 362 (46) 138 (42) 
Race/ethnicity, n (%) n = 789 n = 302 
NHW 473 (60) 212 (70) 
NHB 121 (15) 54 (18) 
Hispanic 118 (15) 22 (7) 
Other 77 (10) 14 (5) 
ELN 2022 categorization, n (%)  n = 330 
Favorable  57 (17) 
Intermediate  66 (20) 
Adverse  207 (63) 
ELN 2010 categorization, n (%) n = 750  
Favorable 112 (15)  
Intermediate 1 280 (37)  
Intermediate 2 136 (18)  
Adverse 222 (30)  
TP53 mutation status tested n = 512 n = 330 
Pathogenic mutation, n (%) 43 (8) 74 (22) 
No pathogenic mutation, n (%) 469 (92) 256 (78) 
Patient demographics2012-2018 multi-institution cohort (N = 789)2018-2022 The University of Chicago cohort (N = 330)
Age at AML diagnosis, y n = 788 n = 330 
Median age (range) 62 (18-95) 69 (18-92) 
Age 18-59, n (%) 350 (44) 78 (24) 
Age 60-74, n (%) 313 (40) 160 (48) 
Age 75+, n (%) 125 (16) 92 (28) 
Sex, n (%)   
Male 427 (54) 192 (58) 
Female 362 (46) 138 (42) 
Race/ethnicity, n (%) n = 789 n = 302 
NHW 473 (60) 212 (70) 
NHB 121 (15) 54 (18) 
Hispanic 118 (15) 22 (7) 
Other 77 (10) 14 (5) 
ELN 2022 categorization, n (%)  n = 330 
Favorable  57 (17) 
Intermediate  66 (20) 
Adverse  207 (63) 
ELN 2010 categorization, n (%) n = 750  
Favorable 112 (15)  
Intermediate 1 280 (37)  
Intermediate 2 136 (18)  
Adverse 222 (30)  
TP53 mutation status tested n = 512 n = 330 
Pathogenic mutation, n (%) 43 (8) 74 (22) 
No pathogenic mutation, n (%) 469 (92) 256 (78) 
Table 2.

Median pollutant exposure by race/ethnicity

PollutantNHWNHBHispanic
2012-2018 multi-institution cohort 
1,3-butadiene (μg/m30.047888 0.05674  0.0627515,  
Benzene (μg/m30.564339 0.622442  0.672733,  
Diesel PM (μg/m30.657941 0.832843  0.9168055  
PAHPOM (μg/m30.002269 0.00261  0.002719  
PM 2.5 (μg/m312.52878 13.13635,  13.01355  
2018-2022 The University of Chicago cohort 
Benzene (μg/m30.4895228 0.6238213  0.5801873  
Diesel PM (μg/m30.5302405 0.8536716  0.7287631  
PAHPOM (μg/m30.002017242 0.002570282  0.002401643  
PM 2.5 (μg/m312.44476 13.22909,  12.72864  
PollutantNHWNHBHispanic
2012-2018 multi-institution cohort 
1,3-butadiene (μg/m30.047888 0.05674  0.0627515,  
Benzene (μg/m30.564339 0.622442  0.672733,  
Diesel PM (μg/m30.657941 0.832843  0.9168055  
PAHPOM (μg/m30.002269 0.00261  0.002719  
PM 2.5 (μg/m312.52878 13.13635,  13.01355  
2018-2022 The University of Chicago cohort 
Benzene (μg/m30.4895228 0.6238213  0.5801873  
Diesel PM (μg/m30.5302405 0.8536716  0.7287631  
PAHPOM (μg/m30.002017242 0.002570282  0.002401643  
PM 2.5 (μg/m312.44476 13.22909,  12.72864  

Median exposure significantly higher than NHW (P < .05).

Median exposure significantly higher than NHB (P < .05).

Median exposure significantly higher than Hispanic (P < .05).

mOS from AML diagnosis was compared between census tracts with the lowest quartile of pollutant exposure (Q1) and the highest quartile of exposure (Q4) for each pollutant. Kaplan-Meier survival was also compared among these quartiles. For 1,3-butadiene, mOS was 2.52 years in Q1 compared with 2.96 years in Q4 (P = .75). For benzene, mOS was 2.62 in Q1 vs 3.00 years in Q4 (P = .38). For diesel PM, mOS was 2.56 years in Q1 vs 2.96 years in Q4 (P = .56). For PAHPOM, mOS was 2.76 years in Q1 vs 2.96 years in Q4 (P = .31). Finally, mOS was 2.92 years in Q1 vs 2.70 years in Q4 (P = .32; supplemental Table 1; supplemental Figure 1).

An additional cohort of patients diagnosed with AML at The University of Chicago between 2018 and 2022 who had testing available for ELN 2022 risk stratification were also analyzed. A total of 330 patients were identified; patient and disease characteristics are summarized in Table 1. The median age of diagnosis was 69 years; 70% of patients were NHW, 18% were NHB, and 7% were Hispanic. By ELN 2022 criteria, 17% of patients had favorable-risk disease, 20% had intermediate-risk disease, and 63% had adverse-risk disease. When analyzing the same pollutant exposures previously mentioned, both Hispanic and NHB patients had significantly higher median exposure to all 5 pollutants than NHW patients. NHB patients also had significantly higher median exposure to PM 2.5 than Hispanic patients (Table 2; supplemental Tables 5 and 6A-E). We performed a multivariate multinominal logistic regression to analyze the impact of individual pollutant exposures on 2022 ELN risk criteria when controlling for age, gender, and race (supplemental Table 2A-E). Only PAHPOM exposure was found to be associated with a significantly higher log odds of intermediate-risk and adverse-risk disease than favorable-risk disease by ELN 2022 criteria (P < .001).

Our previous analysis supported observations that racial and socioeconomic disparities have a significant impact on survival outcomes in AML.6 Here, we confirm that racial-ethnic minorities with AML are disproportionately exposed to air pollution also. Although pollutant exposure did not appear to directly affect survival outcomes, exposure to PAHPOM may be associated with nonfavorable-risk disease by ELN22 classification. The incorporation of secondary-like mutations into risk stratification may explain the ability to capture this association.18 There are multiple avenues of industrial PAHPOM emissions including waste incineration, metal production, refinery exhaust, and operation of vehicles.19 The primary mechanism of carcinogenesis is the formation of polycyclic aromatic hydrocarbons-DNA adducts, which subsequently leads to genotoxicity.20 Limitations of our study include the source of our air pollutant data, in which misclassification of pollutants may exist and affect the ability to detect true associations. In addition, our study only looked at outdoor air pollutant exposure and not household air pollution or occupational exposures, which may be better indicators of exposure at an individual level.21 Nonetheless, further investigation into the impact of pollution upon disease biology, respiratory comorbidities, infectious complications, and survival outcomes is merited.

Contribution: A.P. and A.A.P. designed the study plan, performed data analysis, and wrote the manuscript; G.R. performed data analysis and reviewed/revised the manuscript; and all other authors collected data and reviewed/revised the manuscript.

Conflict-of-interest disclosure: S.B.T. reports honoraria from Jazz and Sanofi, and advisory board fees from Bristol Myers Squibb. J.K.A. reports consultancy fees from AbbVie, Aptitude Health, Astellas Pharma, BioSight, bluebird bio, Curio, Daiichi Sankyo, Dark Blue Therapeutics, Gilead, Kura Oncology, Kymera, Stemline Therapeutics, Treadwell Therapeutics, Rigel, and Syros, and research funding from AbbVie, Agios, ALX Oncology, Amgen, Amphivena, Aprea AB, Aptose Biosciences, Astellas Pharma, BioSight, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Cyclacel Pharmaceuticals, Fujifilm, ImmunoGen, Kartos Therapeutics, Kura Oncology, Loxo, and Pfizer. W.S. reports adviser fees from Kura, Servier, Newave, and Asofarma. A.A.P. reports honoraria from AbbVie, Bristol Myers Squibb, and Sobi, and research funding from Pfizer, Kronos Bio, and Sumitomo. The remaining authors declare no competing financial interests.

Correspondence: Anand A. Patel, Section of Hematology/Oncology, The University of Chicago, 5841 S Maryland Ave MC 2115, Chicago, IL 60637; email: anand.patel@bsd.uchicago.edu.

1.
Siegel
RL
,
Miller
KD
,
Fuchs
HE
,
Jemal
A
.
Cancer statistics, 2021
.
CA Cancer J Clin
.
2021
;
71
(
1
):
7
-
33
.
2.
Döhner
H
,
Wei
AH
,
Appelbaum
FR
, et al
.
Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN
.
Blood
.
2022
;
140
(
12
):
1345
-
1377
.
3.
McNerney
ME
,
Godley
LA
,
Le Beau
MM
.
Therapy-related myeloid neoplasms: when genetics and environment collide
.
Nat Rev Cancer
.
2017
;
17
(
9
):
513
-
527
.
4.
Master
S
,
Munker
R
,
Shi
Z
,
Mills
G
,
Shi
R
.
Insurance status and other non-biological factors predict outcomes in acute myelogenous leukemia: analysis of data from the national cancer database
.
Anticancer Res
.
2016
;
36
(
9
):
4915
-
4921
.
5.
Rebechi
M
,
Kohlschmidt
J
,
Mrózek
K
, et al
.
Association of social deprivation with survival in younger adult patients with AML: an alliance study
.
Blood Adv
.
2023
;
7
(
15
):
4019
-
4023
.
6.
Abraham
IE
,
Rauscher
GH
,
Patel
AA
, et al
.
Structural racism is a mediator of disparities in acute myeloid leukemia outcomes
.
Blood
.
2022
;
139
(
14
):
2212
-
2226
.
7.
Tessum
CW
,
Paolella
DA
,
Chambliss
SE
,
Apte
JS
,
Hill
JD
,
Marshall
JD
.
PM2.5 polluters disproportionately and systemically affect people of color in the United States
.
Sci Adv
.
2021
;
7
(
18
):
eabf4491
.
8.
Rider
CF
,
Carlsten
C
.
Air pollution and DNA methylation: effects of exposure in humans
.
Clin Epigenetics
.
2019
;
11
(
1
):
131
.
9.
Erhunmwunsee
L
,
Wing
SE
,
Shen
J
, et al
.
The Association between polluted neighborhoods and TP53-mutated non-small cell lung cancer
.
Cancer Epidemiol Biomarkers Prev
.
2021
;
30
(
8
):
1498
-
1505
.
10.
Wysota
M
,
Jacobs
S
,
Pradhan
K
, et al
.
Exposure to fine particulate matter (PM2.5) is associated with worse multiple myeloma outcomes in a large, urban, multi-ethnic cohort
.
Blood
.
2023
;
142
(
suppl 1
):
531
.
11.
Gao
Y
,
Zhang
Y
,
Kamijima
M
, et al
.
Quantitative assessments of indoor air pollution and the risk of childhood acute leukemia in Shanghai
.
Environ Pollut
.
2014
;
187
:
81
-
89
.
12.
Lane
HM
,
Morello-Frosch
R
,
Marshall
JD
,
Apte
JS
.
Historical redlining is associated with present-day air pollution disparities in U.S. cities
.
Environ Sci Technol Lett
.
2022
;
9
(
4
):
345
-
350
.
13.
Dehghani
M
,
Keshtgar
L
,
Javaheri
MR
, et al
.
The effects of air pollutants on the mortality rate of lung cancer and leukemia
.
Mol Med Rep
.
2017
;
15
(
5
):
3390
-
3397
.
14.
Taj
T
,
Poulsen
AH
,
Ketzel
M
, et al
.
Exposure to PM2.5 constituents and risk of adult leukemia in Denmark: a population-based case-control study
.
Environ Res
.
2021
;
196
:
110418
.
15.
Döhner
H
,
Estey
EH
,
Amadori
S
, et al
.
Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet
.
Blood
.
2010
;
115
(
3
):
453
-
474
.
16.
US Environmental Protection Agency
.
2014 NATA: assessment results
. Accessed 15 February 2024. https://www.epa.gov/national-air-toxics-assessment/2014-nata-assessment-results.
17.
Centers for Disease Control and Prevention
.
Daily census tract-level PM2.5 concentrations, 2011-2014
. Accessed 15 February 2024. https://data.cdc.gov/Environmental-Health-Toxicology/Daily-Census-Tract-Level-PM2-5-Concentrations-2011/fcqm-xrf4/about_data.
18.
Lindsley
RC
,
Mar
BG
,
Mazzola
E
, et al
.
Acute myeloid leukemia ontogeny is defined by distinct somatic mutations
.
Blood
.
2015
;
125
(
9
):
1367
-
1376
.
19.
Patel
AB
,
Shaikh
S
,
Jain
KR
,
Desai
C
,
Madamwar
D
.
Polycyclic aromatic hydrocarbons: sources, toxicity, and remediation approaches
.
Front Microbiol
.
2020
;
11
:
562813
.
20.
Moorthy
B
,
Chu
C
,
Carlin
DJ
.
Polycyclic aromatic hydrocarbons: from metabolism to lung cancer
.
Toxicol Sci
.
2015
;
145
(
1
):
5
-
15
.
21.
Chan
IYS
,
Liu
AMM
.
Effects of neighborhood building density, height, greenspace, and cleanliness on indoor environment and health of building occupants
.
Build Environ
.
2018
;
145
:
213
-
222
.

Author notes

Data are available on request from the corresponding author, Anand A. Patel (anand.patel@bsd.uchicago.edu).

The full-text version of this article contains a data supplement.

Supplemental data