Document rpgMaby1w92mYK3mxb7e8bED7

G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS Chemico-Biological Interactions xxx (2010) xxxxxx Contents lists available at ScienceDirect Chemico-Biological Interactions journal homepage: www.elsevier.com/locate/chembioint Mini Review A review of the potential association between childhood leukemia and benzene David Pyatt a,b,, Sean Hays a a Summit Toxicology, L.L.P., United States b University of Colorado, Schools of Public Health and Pharmacy, United States article info Article history: Available online xxx Keywords: Benzene Childhood leukemia Epidemiology Review abstract Chronic exposure to high concentrations of benzene is an established cause of acute myeloid leukemia (AML) in occupationally exposed workers. Based on this association, it is not unreasonable to assume that children could also get AML if they were exposed to comparable levels of benzene. Fortunately, reports of such exposures and subsequent AML development in children are non-existent. However, the question of whether children can develop leukemia at far lower, environmental levels of benzene remains. The existing scientific evidence relevant to this question will be addressed in this review. While positive findings have been reported, the collective literature does not indicate that exposure to environmental levels of benzene is related to an increased risk of childhood leukemia. Our understanding of this important issue would be strengthened by additional studies that accurately characterize exposures as well as differentiate between the various forms of leukemias observed in children. 2010 Elsevier Ireland Ltd. All rights reserved. Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 2. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 2.1. Acute lymphoblastic leukemia (ALL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 2.2. ALL clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 2.3. Acute myeloid leukemia (AML) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 2.4. Infant AML/ALL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 3. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.1. Benzene exposure and childhood leukemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.2. Solvents, hydrocarbons and benzene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.3. Exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.4. Parental exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.5. Household exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.6. Parental occupations involving motor vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.7. Cigarette smoking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.8. Maternal smoking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.9. Paternal smoking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.10. Atmospheric contaminants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.11. Traffic density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 4.12. Proximity to refineries and/or gas stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 5. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 Conflict of interest statement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 Corresponding author at: Summit Toxicology, L.L.P., 1944 Cedaridge Circle, Superior, CO 80027, United States. Tel.: +1 7208903798. E-mail address: dpyatt@summittoxicology.com (D. Pyatt). 0009-2797/$ see front matter 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.cbi.2010.01.002 1. Introduction Leukemias are the most common malignancy in children under 15 years of age. The majority (80%) of childhood leukemia cases are acute lymphoblastic leukemia (ALL), with the remainder being Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 2 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx almost exclusively acute myeloid leukemia (AML). Chronic forms of leukemia, either myeloid or lymphoid are exceedingly rare in children. Acute infant leukemia occurs before the age of 1 and shares phenotypic features with both ALL and AML. At this point, the risk factors for childhood leukemia remain essentially unknown as established etiological factors such as genetic conditions, ionizing radiation and certain chemotherapeutic agents can explain only a small portion of cases. As a result, efforts to identify other potential etiologies remain an active research area. From a biological, clinical and epidemiological point of view, childhood ALL and AML are distinctly different diseases [1]. Nonetheless, many older epidemiology studies on childhood leukemia have combined both types into one etiological group which is strongly influenced by the predominant ALL cases. In contrast, newer studies have begun to segregate out these two main subtypes and are reporting that childhood AML and ALL do not typically share the same risk factors. Combining them into a single group complicates the interpretation of this literature and may even obscure potential etiological relationships. There have been several recent reviews on the etiology and potential risk factors for childhood leukemia. However, none of these reviews focused specifically on benzene. The existing scientific evidence relevant to benzene exposure and the development of childhood leukemia will be addressed in this review. Most epidemiology studies specific to children have included one or more of the following exposure pathways for benzene: (1) direct exposure to the child (via a variety of sources); (2) maternal exposures during pregnancy or while breastfeeding; and (3) exposure to either parent before conception. These pathways are not mutually exclusive and frequently occur together. The timing of exposure (e.g. age of the child) has also been an important consideration. Unfortunately, very few studies contain quantified exposure information and many have no meaningful exposure estimates. The use of various surrogates for exposure such as proximity to gas stations or traffic density makes interpretation difficult, particularly with regard to specific chemical exposures. Parental exposures to benzene or other chemicals in the workplace and its potential effect in the offspring have also been assessed (with varying degrees of scientific rigor) in multiple studies. Cigarette smoke is also a well established source of benzene exposure, so the potential relationship between paternal or maternal smoking during or after pregnancy and the development of childhood leukemia in their offspring has also been included. 2. Background 2.1. Acute lymphoblastic leukemia (ALL) Childhood ALL is a malignant disorder of lymphopoietic stem and progenitor cells that originates in the bone marrow, thymus and lymph nodes. The majority (80%) of ALL cases are B cell derived, with the developmental block in maturation/differentiation occurring at the pre-B or B cell stage of development [2]. Untreated, ALL is rapidly fatal, but with modern therapeutic intervention, the 5 year survival rate is currently greater than 80%. Of these long-term survivors, 85% will not relapse and are essentially cured. Adult onset ALL is quite different from childhood ALL and is far more difficult to treat successfully [2]. ALL is the most common childhood cancer in the United States and occurs at an annual rate of between 3 and 4 cases per 100,000 children under 15 years old [3]. There are approximately 2200 new cases of ALL diagnosed every year in the US [3]. As shown in Fig. 1, ALL has a maximum incidence between 2 and 5 years of age, with a distinct peak occurring at age 34 years [3]. After the peak at 34 years old, the rate of ALL incidence decreases steadily for 4 or 5 years and then remains relatively flat through the rest of childhood [4]. This pronounced peak in early childhood has stimulated many Fig. 1. Age specific childhood ALL and AML incidence rates. Adapted from SEER 2004. investigations yet an explanation for this phenomenon has not been identified. Childhood ALL exhibits profound geographic and ethnic differences in incidence, with as much as a 4-fold difference reported between various countries [5]. This can vary even more for ALL cases observed in children under the age of 4. For example, the difference between the incidence rates in Nigeria (lowest) and Denmark (highest) can vary as much as 30-fold [3]. Observed geographic and ethnic differences could be due to a variety of factors including genetic variability and resulting susceptibility differences as well as lifestyle factors. Accuracy of diagnosis and case ascertainment in different parts of the world could also account for some of the observed differences. Regardless of the underlying cause, ethnic and geographic differences will undoubtedly confound interpretation of any study of ALL and should be considered in the interpretation and analysis of the data. There are several genetic syndromes and abnormalities that increase the risk of childhood ALL. These include Down's syndrome, Type I neurofibromatosis, Fanconi's anemia and others. While this association is very strong, these genetic syndromes only account for a very small percentage (less that 5%) of the total ALL cases [610]. Therefore, the majority of cases have no known etiology. Almost 90% of childhood ALL cases have some type of clonal cytogenetic abnormality and evidence indicates that many of these originate in utero [1115]. A wide assortment of cytogenetic alterations is associated with childhood ALL, however, the most common is the TEL-AML1 fusion gene (observed in 2530% of the cases) [16,17]. Current epidemiology investigations have begun to segregate childhood ALL cases according to specific cytogenetic abnormalities such as TEL and others [18]. To date, the only well established risk factors for childhood ALL are ionizing radiation and specific genetic syndromes [2,3,19,20]. However, the list of postulated risk factors is very long and continues to expand. These hypothesized factors include, but are not limited to, pre-natal exposure to X-rays, chemical exposure (e.g. solvents, pesticides, benzene), vehicle exhaust, non-ionizing radiation, infectious agents, allergies, immunologic isolation, occupational parental exposures (various), maternal diet, maternal age, parental cigarette smoking, maternal alcohol consumption and/or drug use (both recreational and prescription) during pregnancy, population density, socioeconomic status, vitamin deficiencies, birth order, and many others [3]. Although much has been written about these various factors, there is insufficient epidemiologic evidence to support a definitive causal connection with any of these factors and childhood ALL. 2.2. ALL clustering ALL clustering is a topic of considerable scientific debate and public interest and is critical to any discussion of ALL epidemiology. Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx 3 A `cluster' is defined as an aggregation of a relatively uncommon event or disease that is perceived to be occurring more frequently than would be expected based on chance alone [21]. A review of cancer clusters by the Centers for Disease Control (CDC) of 108 clusters spanning a 20-year period was conducted [22]. Over 80% of these `clusters' involved leukemia in children, but none were found to have an underlying environmental cause [22]. Much of the difficulty in investigating `cancer clusters' lies in appropriately defining the geographic and temporal boundaries of the cluster as well as avoiding a variety of other potential sources of bias [23]. Therefore, at this point, with the exception of catastrophic events such as the atomic bombing of Japan, evidence does not support the existence of leukemia clusters in children (or adults) that arise due to a common environmental exposure. 2.3. Acute myeloid leukemia (AML) AML is a clonal, hematopoietic malignancy characterized by a block in differentiation and an unregulated proliferation of myeloid progenitor cells [24]. AML itself is not a single disease, but comprises seven (or eight) different subtypes (as classified according to the French-American-British, FAB, classification system) [2,25]. These distinct subtypes of AML are classified according to discrete morphological and cytogenetic criteria, which in turn are used to accurately predict the prognosis as well as select the most appropriate treatment options [25,26]. The new WHO classification system retains some of these morphological characteristics, but places far more emphasis on cytogenetic abnormalities [1]. The same diagnostic criteria are used to classify both adult and childhood cases of AML and the various subtypes appear to posses many of the same cytogenetic and morphologic features, irrespective of the age of the patient [11]. This suggests that similar mechanisms of leukemogenesis likely exist in both children and adults [3,27]. However, the distribution of AML subtypes is somewhat different for childhood cases compared to adults [11]. The reasons for this are not currently understood. AML is the seventh most common pediatric malignancy with about 500 new cases diagnosed each year in the US [24,28,29]. As shown in Fig. 1, there is no clearly defined peak observed for childhood AML [4]. As with ALL, considerable geographic and ethnic variability in incidence and subtype distribution exists [5,24]. For example, acute pro-myelocytic leukemia, a distinctive subtype of AML, comprises about 30% of all pediatric AML cases in Latin countries, but makes up less than 10% in non-Latin countries [24]. While much has been learned regarding etiology and epidemiology, advances in treatment options for childhood AML have progressed at a much slower rate [24]. AML is far less common in children than ALL; however, it has been associated (with varying degrees of scientific rigor) with a variety of environmental or clinical causes. Established causes of AML in adults include certain classes of chemotherapy, ionizing radiation, cigarette smoking and high dose chronic exposure to benzene. Etiological factors that have been well characterized for childhood AML include genetic disorders, ionizing radiation and cytotoxic chemotherapy [30]. Two classes of chemotherapy, alkylating agents and topoisomerase inhibitors, have been clearly established as causes of AML in humans, including children [31]. AML associated with each of these drug classes differ from one another in several important characteristics [31,32]. AML cases induced by therapy with alkylating agents typically present several years (57) after therapy and are frequently characterized by deletions in chromosome 5q and/or 7q [3133]. In contrast, AML cases induced by topoisomerase reactive drugs tend to occur more quickly after therapy (12 years) and are typically characterized by rearrangements involving the MLL gene at 11q23 [24]. ALL is rarely if ever observed as a secondary leukemia in children treated with either class of chemotherapy, indicating that the different types of childhood leukemia likely possess distinct etiological pathways [31,32]. Much has been published regarding other hypothesized risk factors for childhood AML including parental diet, parental smoking, alcohol consumption and drug use (both recreational and prescription) during pregnancy, non-ionizing radiation, environmental exposures to benzene or other chemicals such as butadiene or pesticides, and parental occupations, especially those involving exposure to hydrocarbons or petroleum products [24,34]. 2.4. Infant AML/ALL Infant leukemias are defined as leukemias that arise in the first year of life. AML and ALL occur with approximately equal frequency in this age group, although some cases of infant leukemia possess pathologic features that are consistent with both subtypes [30,35,36]. Epidemiological and molecular genetic studies have confirmed that most infant leukemias arise in utero [35,36]. Many of the cytogenetic abnormalities observed in cases of infant leukemia are actually present at the time of birth [37]. Structural rearrangements of the MLL gene are observed with high incidence in infant leukemia. In fact, over 60% of both infant ALL and infant AML possess balanced reciprocal translocations (11q23) involving the MLL gene [30]. The MLL mutation is also commonly reported in AML cases following treatment with topoisomerase reactive drugs [35,38]. Accordingly, it has been postulated that topoisomerase inhibitors in the diet (which are commonly found in fruits, vegetables, legumes and coffee) may contribute to the development of infant acute leukemia [18,35,38]. Another potential risk factor for infant leukemia that has considerable scientific support is a high birth rate [18,39]. 3. Methods Studies included in this review were all readily available in the published scientific literature (Pubmed, Medline, ISI Web of Knowledge, etc.). Keywords included childhood leukemia, children's leukemia, AML, ALL, infant leukemia, occupational, exposures, solvents, petroleum, hydrocarbons, benzene, painting, parental, maternal, cigarette smoking, passive smoke, secondhand smoke. These were used in a variety of different combinations. Other published review papers were used to identify further relevant publications. All references from each paper were scanned for additional studies. The search procedure was stopped when no further useful study dealing specifically with the question of interest could be found. At the current time, the authors know of no other studies that have direct relevance to this topic than the ones included herein, although studies are being published all the time. An exception was made with cigarette smoking. We collected and evaluated over 50 studies that evaluated parental smoking and the risk of childhood leukemia. As the potential health effects of cigarette smoking were not the primary focus of this paper, we felt that this was representative of the state of the literature and science and did not attempt an exhaustive search on this topic. Many of these studies suffer from the same limitations: small numbers of cases, poorly defined or quantified exposures, and potential exposures with confounding chemicals. These studies are grouped and listed in Tables 19. As most studies contained multiple comparisons, the risk estimates deemed the most relevant to the aims of this review were included. While every study reported values that differ from 1.0 (either elevated or decreased), studies were only considered positive or negative if the exposed groups were statistically significantly different from the controls (p value < 0.05, 95% confidence interval that does not include 1.0). Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 4 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx Table 1 Parental exposures to solvents and hydrocarbons and childhood leukemia (CL) or ALL. Study Exposures Hakulinen et al. [49] Zack et al. [50] Van Steensel-Moll et al. [45] Lowengart et al. [44] Mulder et al. [46] Smulevich et al. [47] Shu et al. [51] Schuz et al. [52] Feingold et al. [53] Freedman et al. [54] Feychting et al. [63] McKinney et al. [55] Infante-Rivard et al. [56] Hydrocarbons (P) Hydrocarbons (P) Chemicals (M) (paints, petroleum products, others) Chemicals (P) (paints, petroleum products, others) Hydrocarbons (M) Hydrocarbons (M) Spray paint (P, occupational) Petroleum (P, occupational) Petroleum products (p) Solvents (M) Solvents (P) Oil products (M) Oil products (P) Petroleum (M) Paints (M) Petroleum (P) Paints (P) Solvents (M) Oil products (m) Paints or lacquers (M) Solvents (P) Oil products (p) Paints or lacquers (P) Hydrocarbons (P) Paints (M) Solvents (P) Solvents (M) Solvents (P) Paints (M) Paints (P) Dermal hydrocarbon (m) Dermal hydrocarbon (p) Solvents (M) Scelo et al. [48] Alderton* et al. [57] Paints (M) Paints (P) Solvents (M) Solvents (P) Paints (M) Petroleum (M) M = maternal, P = paternal, and NA = not available. *Children with Down's syndrome. Disease Leukemia and lymphoma Leukemia and lymphoma ALL ALL ALL ALL CL CL Leukemia and lymphoma CL CL CL CL ALL ALL ALL ALL ALL ALL ALL ALL ALL ALL ALL ALL CL ALL ALL ALL ALL ALL ALL CL ALL ALL ALL ALL ALL ALL ALL Results OR = 0.95 (95% CI = 0.571.6) RR = 0.75, (p = 0.89) RR = 2.4 (95% CI = 1.24.6) RR = 1.2 (95% CI = 0.81.7) No association, data not provided RR = 1.0 (95% CI = 0.61.7) OR = 2.2, (p = 0.03) RR = 1.0 (95% CI = 0.333.06) OR = 9.0 (95% CI = 1.066.1) OR = 3.1 (95% CI = 1.56.3) OR = 1.4 (95% CI = 0.952.1) No association, data not provided OR = 1.3 (95% CI = 0.921.9) OR = 1.2 (95% CI = 0.81.7) OR = 1.1 (95% CI = 0.91.4) OR = 1.0 (95% CI = 0.81.2) OR = 0.9 (95% CI = 0.81.1) OR = 1.1 (95% CI = 0.81.6) OR = 1.3 (95% CI = 0.82.2) OR = 1.8 (95% CI = 1.22.6) OR = 1.0 (95% CI = 0.81.3) OR = 1.1 (95% CI = 0.91.3) OR = 1.1 (95% CI = 0.91.4) OR = 1.3 (95% CI = 0.63.0) OR = 1.1 (95% CI = 0.91.5) OR = 1.25 (95% CI = 0.801.95) OR = 1.00 (95% CI = 0.661.51) OR = 1.05 (95% CI = 0.791.37) No cases OR = 1.22 (95% CI = 0.731.91) OR = 2.16 (95% CI = 1.164.02) OR = 0.96 (95% CI = 0.761.21) OR = 1.00 (95% CI = 0.781.28) OR = 1.19 (95% CI = 0.891.58) OR = 1.44 (95% CI = 1.081.91) OR = 1.08 (95% CI = 0.781.28) OR = 1.17 (95% CI = 0.881.56) OR = 0.92 (95% CI = 0.461.84) OR = 1.03 (95% CI = 0.353.03) Cases 497 158 25 25 37 37 18 8 15 20 70 NA 100 58 169 619 534 151 180 54 66 25 157 30 160 23 27 64 21 14 88 154 172 203 102 158 17 6 4. Results 4.1. Benzene exposure and childhood leukemia There are several recent reviews that provide excellent overviews of childhood leukemia [35,40]. However, environmental exposures, especially those relevant to benzene, have not received a comprehensive discussion in any of these recent reviews. This report provides a more thorough review of the literature deal- ing with potential benzene exposures and the risk of developing childhood ALL or AML. As childhood leukemias are relatively rare diseases, prospective studies are difficult to conduct. Therefore, most of the existing studies are retrospective casecontrol or ecological studies [4]. Retrospective casecontrol studies possess a number of inherent limitations that directly influence the weight that should be afforded the results. For example, the exposure is measured indirectly (if at all), and may be differentially recalled by parents of sick Table 2 Parental exposures to benzene and childhood leukemia (CL) or ALL. Study Exposures Shaw et al. [59] Shu et al. [60] McKinney et al. [61] Feingold et al. [53,59] Kaatsch et al. [62] Shu, et al. [51] Feychting et al. [63] Infante-Rivard et al. [56] Benzene (P) Benzene (P) Benzene (M) Benzene (M) Benzene (P, preconception) Benzene (P, gestation) Benzene (P) Benzene (P) Benzene (M) Benzene (P) Benzene (M) Benzene (P) Benzene (M, during pregnancy) M = maternal, P = paternal, and NA = not available. Disease CL ALL ALL Leukemia and lymphoma Leukemia and lymphoma Leukemia and lymphoma ALL CL CL ALL ALL CL CL Results No association, p = 0.75, OR not provided No association, data not provided OR = 1.3 (95% CI = 0.53.0) OR = 4.00 (95% CI = 0.3117) OR = 5.81 (95% CI = 1.6726.44) OR = 2.98 (95% CI = 0.524.19) OR = 1.6 (95% CI = 0.55.8) No association, data not shown No association, data not shown OR = 1.2 (95% CI = 0.81.6) OR = 0.7 (95% CI = 0.31.6) OR = 1.23 (95% CI = 0.393.85) OR = 1.39 (95% CI = 0.316.25) Cases 205 8 2 12 4 9 NA NA 97 11 3 4 Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx Table 3 Parental exposures to benzene and/or hydrocarbons and childhood AML. Study Exposures Disease Results Shu et al. [60] Buckley et al. [64] Magnani et al. [65] Van Duijn et al. [66] Scelo et al. [48] Alderton* et al. [57] Benzene (P) Benzene (M) Solvent (P) Petroleum (P) Solvent (M) Petroleum (M) Paints (M) Paints (P) Tire production (P) Hydrocarbon (M) Paints (M) Paints (P) Petroleum solvents (M) Petroleum solvents (p) Paints Petroluem AML ANLLL ANLL ANLL ANLL ANLL ANLL ANLL ANLL ANLL AML AML AML AML AML AML No association, data not provided OR = 4.0 (95% CI = 1.88.3) OR = 2.0 (95% CI = 1.23.8) OR = 2.4 (95% CI = 1.34.1) No association, data not provided No association, data not provided OR = 2.2 (95% CI = 0.95.4) OR = 7.0, p = 0.02 OR = 7.6 (95% CI = 1.734.1) OR = 3.3 (95% CI = 0.618.1) OR = 1.37 (95% CI = 0.613.11) OR = 0.39 (95% CI = 0.170.92) OR = 1.64 (95% CI = 0.584.62) OR = 1.94 (95% CI = 0.874.32) OR = 1.41 (95% CI = 0.613.23) OR = 0.73 (95% CI = 0.212.56) M = maternal, P = paternal, NA = not available, and ANLL = acute non-lymphocytic leukemia. *Children with Down's syndrome. 5 Cases 11 57 53 NA NA 15 7 2 3 24 22 16 31 13 5 Table 4 Children's exposure to solvents (home use) and childhood leukemia. Study Exposures Disease Mulder et al. [46] Lowengart et al. [44] Buckley et al. [64] Freedman et al. [54] Alderton* et al. [57] Petroleum products (p) 15 Spray paint (M) Spray paint (P) Petroleum (M) Petroleum (P) Petroleum products Artwork (solvents) Furniture stripping Mechanics Paints Paints Leukemia and lymphoma CL CL CL CL ANLL ALL ALL ALL ALL AML M = maternal, P = paternal, and NA = not available. *Children with Down's syndrome. Results OR = 8.0 (95% CI = 2.2129.1) RR = 1.3, p = 0.30 RR = 1.0, p = 0.50 RR = 0.8, p = 0.27 RR = 0.8, p = 0.20 OR = 1.8 (95% CI = 0.74.3) OR = 4.1 (95% CI = 1.115.1) OR = 1.0 (95% CI = 0.427.0) OR = 1.5 (95% CI = 0.82.7) OR = 1.39 (95% CI = 0.643.04) OR = 3.36 (95% CI = 0.3929.3) Cases 4 2 13 23 57 11 8 19 20 3 vs. healthy children (recall bias). It is also possible for the results to be biased if certain segments of the population do not respond or are under-represented in the study or control cohorts [4]. Thus some positive findings may be due to these confounding factors and may not represent true causal relationships [5]. Additionally, some positive findings would be expected simply as chance occurrences, depending on the number of studies and comparisons that have been made [21]. Ecological studies are specifically vulnerable to the so-called `ecological fallacy' where the alleged relationship may have little or no relevance to the individuals within the actual study. On the other hand, small but meaningful increases in risk may not be detectable in standard epidemiologic analysis and combining childhood ALL and AML together for statistical analysis could mask potential etiologic relationships with AML. As a result of these limitations, the current body of literature on childhood leukemia is useful for hypothesis generation and discussion, but is not suitably robust to clearly identity etiological risk factors. 4.2. Solvents, hydrocarbons and benzene Occupational and/or environmental exposures to solvents have been widely studied as a potential etiological risk factor in the development of childhood leukemia [4,41]. As a chemical class, the term `solvent' or `organic solvent' encompasses a wide assortment of chemical compounds, ranging from complex mixtures such as gasoline to simple, single chain alcohols such as ethanol. These Table 5 Parental occupations related to motor vehicles and childhood leukemia. Study Exposures Hakulinen et al. [49] Kwa et al. [69] Hemminki et al. [70] Gold et al. [71] Vianna et al. [67] Van Steensel-Moll et al. [45] Shu et al. [60] Buckley et al. [64] Magnani et al. [65] Roman et al. [72] Perez-Saldivar et al. [73] Drivers (P) Mechanics and attendants (P) Drivers (P) Drivers (P) Drivers, mechanics and attendants (P) Mechanics and attendants (P) Mechanics, attendants (P) Transportation (M) Transportation (P) Mechanic (P) Driver (P) Drivers (P) Drivers (P) Mechanics M = maternal, P = paternal, and NA = not available. Disease Leukemia and lymphoma Leukemia and lymphoma Leukemia and lymphoma CL CL Infant leukemia CL CL CL ANLL ALL Leukemia and lymphoma CL CL Results RR = 1.06 (95% CI = 0.63-1.8) No association, data not shown No association, data not shown RR = 1.5, p < 0.10 No association, data not shown RR = 2.5 (95% CI = NA) RR = 0.8 (95% CI = 0.41.5) RR = 1.4 (95% CI = 0.45.0) RR = 1.2 (95% CI = 0.62.3) RR = 3.5, p = 0.02 No association, data not shown RR = 0.6 (95% CI = 0.12.0) RR = 1.77 (95% CI = 0.398.00) RR = 2.81(95% CI = 0.4816.43) Cases NA 21 28 NA 43 65 5 24 14 NA 3 12 4 Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 6 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx Table 6 Maternal smoking and childhood leukemia. Study Van Steensel-Moll et al. [45] Buckley et al. [76] Stjernfeldt et al. [77] McKinney et al. [81] Magnani et al. [65] Exposures Smoking Smoking Smoking Smoking Smoking John et al. [82] Urquhart et al. [83] Stjernfeldt et al. [139] Pershagen et al. [84] Lymphomas Severson et al. [140] Van Duijn et al. [66] Klebanoff et al. [78] Cnattingius et al. [88] Shu et al. [89] Sorahan et al. [102] Smoking Smoking Smoking Smoking RR = 1.04 (95% CI = 0.71-1.52) Smoking Smoking Smoking Smoking Smoking Smoking Smoking Kaatsch et al. [62] Brondum et al. [90] Boffetta et al. [98] Infante-Rivard et al. [141] Pang et al. [91] Smoking Smoking Smoking Smoking Smoking Smoking (pregnancy) Mucci et al. [96] Clavel et al. [97] Chang et al. [79] Smoking Smoking Smoking Smoking Trivers et al. [92] Menegaux et al. [80] Smoking (marijuana) During pregnancy Smoking Rudant et al. [99] MacArthur et al. [100] Smoking Smoking Smoking, pregnancy NA = not available. Disease CL ALL ALL CL AML ALL ALL CL ALL Leukemias 129 AML ANLL CL AML AML ALL ALL AML CL AML ALL CL ALL ALL AML ALL AML CL ALL AML AML AML ALL AML ALL AML ALL AML Results RR = 1.0 (95% CI = 0.71.3) RR = 0.9 (95% CI = 0.71.1) RR = 2.07, p = 0.01 RR = 0.6 (95% CI = 0.41.0) RR = 2.0 (95% CI = 0.82.7) RR = 0.7 (95% CI = 0.51.1) RR = 1.9 (95% CI = 0.9-4.1) RR = 1.0 (95% CI = 0.31.4) RR = 2.2 (95% CI = 1.14.5) RR = 1.2 (95% CI = 0.771.86) OR = 0.6 (95% CI = 0.31.2) RR = 0.82 (95% CI = 0.312.11) RR = 2.4 (95% CI = 0.96.5) RR = 0.69 (95% CI = 0.85.78) RR = 0.48 (95% CI = 0.121.90) RR = 1.2 (95% CI = 1.01.5) RR = 1.2 (95% CI = 0.81.7) OR = 0.48 (0.092.65) RR = 0.73 (95% CI = 0.31.07) RR = 1.04 (95% CI = 0.871.25) RR = 1.05 (95% CI = 0.821.34) RR = 1.2 (95% CI = 0.82.0) RR = 0.89 (95% CI = 0.771.03) RR = 0.76 (95% CI = 0.541.07) RR = 0.84 (95% CI = 0.611.15) RR = 2.20 (95% CI = 1.004.93) RR = 0.9 (95% CI = 0.42.0) RR = 1.01 (95% CI = 0.921.11) RR = 1.01 (95% CI = 0.951.07) RR = 0.89 (95% CI = 0.661.19) RR = 0.43 (95% CI = 0.230.8) RR = 1.5 (95% CI = 0.92.5) RR = 0.7 (95% CI = 0.22.6) RR = 1.0 (95% CI = 0.61.4) RR = 0.9 (95% CI = 0.42.2) RR = 0.98 (95% CI = 0.51.89) RR = 0.4 (95% CI = 0.043.80) Cases 519 749 171 10 NA 14 128 49 22 NA NA NA NA 367 115 2 Meta-analysis 43 1449 249 44 9 41 3 38 6 22 2 different compounds possess significantly different toxicological profiles, so studies that report findings (positive or negative) associated with `solvent' exposures are difficult to interpret. Another class of chemicals that has received considerable attention with regard to a potential association with childhood leukemia are hydrocarbons, particularly petroleum derived hydrocarbons. From a toxicological point of view, hydrocarbons are also quite diverse, although benzene is widely considered to be most important [42,43]. Table 7 Paternal smoking and childhood leukemia. Study Magnani et al. [65] John et al. [82] Shu et al. [89] Ji et al. [101] Sorahan et al. [102] Brondum et al. [90] Infante-Rivard et al. [141] Roman et al. [72] Chang et al. [79] Pang et al. [91] Rudant et al. [99] MacArthur et al. [100] Lee et al. [103] NA = not available. Exposures Smoking Smoking Smoking Smoking Smoking Smoking Smoking Smoking Smoking Smoking Smoking (M and P) Smoking (M and P) Passive Smoking Smoking Smoking Smoking, pregnancy Smoking Disease ALL AML ALL ALL AML CL ALL ALL ALL AML ALL CL AML ALL ALL ALL AML ALL AML ALL AML CL Results RR = 0.9 (95% CI = 0.61.5) RR = 0.9 (95% CI = 0.32.1) RR = 1.7 (95% CI = 0.73.8) RR = 1.51 (95% CI = 0.822.77) RR = 1.29 (95% CI = 0.443.74) RR = 1.9 (95% CI = 0.84.6) RR = 1.1 (95% CI = 0.91.3) RR = 1.0 (95% CI = 0.71.3) RR = 1.06 (95% CI = 0.881.26) RR = 0.81 (95% CI = 0.581.14) RR = 1.0 (95% CI = 0.71.3) RR = 0.50 (95% CI = 0.21.2) RR = 3.84 (95% CI = 1.0414.17) RR = 1.32 (95% CI = 0.862.04) RR = 0.41 (95% CI = 0.170.97) RR = 1.04 (95% CI = 0.911.18) RR = 1.07 (95% CI = 0.81.43) RR = 1.7 (95% CI = 1.32.1) RR = 1.7 (95% CI = 1.02.9) RR = 1.12 (95% CI = 0.761.32) RR = 0.99 (95% CI = 0.362.71) RR = 2.0 (95% CI = 1.23.4) Cases 142 22 NA NA NA 166 367 115 NA NA 167 9 16 74 11 1375 230 168 26 86 11 76 Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx Table 8 Traffic density (or related measures) and childhood leukemia. Study Exposures Savitz et al. [109] Swaen et al. [111] Alexander et al. [112] Nordlinder and Jarvhome [113] Feychting et al. [114] Harrison et al. [119] Pearson et al. [110] Raaschou-Nielsen et al. [115] Langholz et al. [120] Reynolds et al. [116] Visser et al. [121] Steffen et al. [122] Crosignani et al. [124] Reynolds et al. [118] Von Behren et al. [123] Weng et al. [125] Traffic density Gasoline consumption Car ownership Traffic density Traffic density Traffic density Traffic density Traffic density Benzene levels Traffic density Traffic density Traffic density Traffic density Traffic density Traffic density Traffic density Benzene levels Traffic density Traffic density Traffic density, NO2 levels M = maternal, P = paternal, and NA = not available. Disease CL Leukemia ALL AML ALL CL CL CL CL CL CL ALL AML ALL CL CL CL CL ALL CL Results RR = 4.7 (95% CI = 1.613.5) No association, data not shown RR = 1.11 (95% CI = 0.831.5) Positive association, p = 0.05 No association, data not shown No association, data not shown RR = 1.16 (95% CI = 0.741.72) RR = 8.28 (95% CI = 2.0932.8) RR = 0.95 (95% CI = 0.821.09) RR = 0.8 (95% CI = 0.51.3) RR = 1.4 (95% CI = 0.92.3) RR = 1.14 (95% CI = 0.941.39) RR = 1.04 (95% CI = 0.691.58) RR = 2.63 (95% CI = 0.856.14) RR = 1.3 (95% CI = 0.53.2) RR = 2.09 (95% CI = 0.855.12) RR = 3.91 (95% CI = 1.3611.27) RR = 0.92 (95% CI = 0.731.15) RR = 1.24 (95% CI = 0.742.08) RR = 2.29 (95% CI = 1.443.64) 7 Cases 8 NA 97 89 301 39 24 8 NA NA NA 187 31 5 11 8 7 155 52 117 Accordingly, benzene has been the subject of increased attention in the search for environmental causes of childhood leukemia. 4.3. Exposures There are a variety of relevant exposure pathways for consideration in assessing the potential relationship between environmental benzene exposure and childhood leukemia. These include maternal exposures which could have occurred prior to conception, during pregnancy or while breastfeeding the infant. Paternal exposures are also of interest, including those that occur prior to conception as well as post-natally. There is also the possibility that direct benzene exposures to the infant or child from the environment might occur. These various pathways are not mutually exclusive and could conceivably occur together. 4.4. Parental exposures There have been numerous attempts to establish a relationship between parental exposures to solvents and hydrocarbons in gen- eral or benzene specifically and childhood leukemia [3]. The data obtained from these various investigations is extremely inconsistent from study to study. Further, all of these studies suffer from the same limitations: small numbers of cases, poorly defined or quantified exposures, and potential exposures with confounding chemicals. The studies that evaluated either `childhood leukemia' or ALL are listed in Table 1. There have been a few small casecontrol studies that have reported a link between excesses of childhood leukemia and parental occupational exposure to hydrocarbons. Lowengart et al. [44] reported an association with paternal but not maternal use of spray paint. No other category such as exposure to petroleum products was elevated. Van Steensel et al. [45] reported an elevated risk from maternal, but not paternal exposures to the broadly defined `chemical' group. These authors did not report increased risks associated with either maternal or paternal exposures to hydrocarbons. Mulder and Drijver [46] reported also significantly elevated risk for `hematopoietic malignancies' in young people (less than 40) associated with `petroleum products' and `pesticides'. However, this study does not specifically apply to children and appropriate Table 9 Proximity to industrial facilities and childhood leukemia. Study Hearey et al. [127] Knox et al. [128] Lyons et al. [131] Sans et al. [132] Knox and Gilman [138] Wilkinson et al. [130] Exposures Refineries and petrochemical plants Various facilities Petrochemical plant Petrochemical plant Various facilities, many chemicals Refineries Harrison et al. [119] Reynolds et al. [117] Hurtig et al. [134] Steffen et al. [122] Weng et al. [135] Whitworth et al. [137] Petrol stations Hazardous air pollutants Oil fields in Ecuador Oil fields in Ecuador Repair garages and gas stations Repair garages and gas stations Petrochemical plant Air pollutants, modeled benzene Weng et al. [136] Brosselin et al. [133] Petrol stations Repair garages and gas stations Repair garages and gas stations Intrauterine exposures Intrauterine exposures M = maternal, P = paternal, and NA = not available. Disease CL CL CL CL CL ALL AML CL CL ALL AML ALL AML CL ALL AML CL ALL AML ALL AML Results No association, data not shown No data provided No association, p = 0.32 RR = 0.94 (95% CI = 0.113.39) No data provided RR = 1.08 (95% CI = 0.621.76) RR = 0.66 (95% CI = 0.082.38) RR = 1.48 (95% CI = 0.652.93) RR = 1.21 (95% CI = 1.031.42) RR = 2.56 (95% CI = 1.25.47) RR = 2.56 (95% CI = 0.778.5) RR = 3.6 (95% CI = 1.39.9) RR = 7.7 (95% CI = 1.734.3) RR = 1.75 (95% CI = 1.003.06) RR = 1.24 (95% CI = 0.921.66) RR = 2.02 (95% CI = 1.033.96) RR = 1.95 (95% CI = 1.472.59) RR = 1.8 (95% CI = 0.93.5) RR = 0.8 (95% CI = 0.16.2) RR = 6.0 (95% CI = 1.819.5) RR = 8.3 (95% CI = 1.449.4) Cases NA NA 10 2 NA 16 2 24 277 20 8 13 4 116 124 31 312 14 1 9 2 Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 8 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx disease specificity was lacking. Smulevich et al. [47] reported an association with maternal exposure to solvents, but not oil products. In this study, paternal exposure to solvents or oil products was not associated with an increased risk [47]. Scelo et al. [48], reported that paternal, but not maternal exposure to paints resulted in an increased risk of ALL. Neither maternal of paternal exposure to solvents were related to an increased risk [48]. A larger body of literature did not report excesses in the risk of childhood leukemia or ALL associated with parental exposures to hydrocarbons. These include Hakulinen et al. [49], Zack et al. [50], Smulevich et al. [47], and Shu et al. [51]. Additionally, Schuz et al. [52] published a large pooled analysis of 1138 ALL cases and various parental occupational exposures. Neither maternal or paternal exposure to solvents or oil products were associated with an excess risk of ALL, although maternal exposure to paints was elevated [52]. Feingold et al. [53] reported no association between maternal exposure to benzene and ALL [53]. Freedman et al. [54] reported that paternal exposure to hydrocarbons was not associated with an increased risk of ALL. McKinney et al. [55] evaluated several sources of parental exposures and childhood leukemia risks including paints, solvents and hydrocarbons. All were negative, with the exception of maternal dermal exposure to hydrocarbons [55]. Infante-Rivard et al. [56] did not report an excess of ALL risk associated with maternal exposure to solvents, which was consistent with findings described by Scelo et al. [48]. Alderton et al. [57] did not observe a positive association with maternal exposure to either paints or petroleum in children with Down's syndrome. This was an interesting finding as children with Down's syndrome have a greatly increased risk of developing acute leukemia [58]. Some studies specifically evaluated parental exposure to benzene (in a non-quantitative manner) and the potential association with childhood leukemia (Table 2). Shaw et al. [59] did not find an association between childhood leukemia and paternal exposures to benzene. Maternal occupation was not considered. Shu et al. [60] was also negative for both maternal and paternal exposures to benzene. Mckinney et al. [61] conducted a casecontrol study to evaluate both maternal and paternal exposures to benzene. Benzene exposure was not identified as a risk factor for maternal exposures, but paternal exposures to benzene during the pre-conception period was significantly associated [61]. Feingold et al. [53] conducted a casecontrol study of childhood ALL and parental occupation. These investigators reported no association between maternal exposure to benzene and ALL [53]. Kaatsch et al. [62] conducted a casecontrol study of childhood leukemia. Though the majority of the leukemias were ALL, there were 147 cases of childhood AML included in the analysis (but not analyzed separately). No association between parental exposure to benzene and childhood leukemia was found [62]. Shu et al. [51] conducted a large casecontrol study (1842 cases of ALL) and found no association between paternal or maternal exposure to benzene or `petroleum products' (including during pregnancy) [51]. Feychting et al. [63] also reported that paternal exposure to benzene was not associated with an increased childhood leukemia risk. More recently, Infante-Rivard, et al. [56] conducted a large casecontrol study of childhood ALL and maternal exposure to various solvents (790 ALL cases). There was no excess of ALL associated with benzene exposure during the 2 years before birth or during pregnancy [56]. There have been at least 4 studies that specifically addressed the risks of AML and parental occupational exposures, including benzene (Table 3). Shu et al. [60] evaluated the relationship between parental occupational exposures and childhood leukemia in Shanghai, China. A variety of potential etiological agents were investigated, including benzene. Paternal occupational exposure was not associated with either form of acute childhood leukemia (ALL or AML). However, maternal exposure to benzene and gasoline was positively associated with ANLL [60]. This was based on 11 cases of ANLL. No attempt was made to characterize direct chemical exposures that could have occurred in the home or the general environment or potential exposures to confounding chemicals in the workplace. In contrast, Buckley and Robinson [64] reported a significant elevation in the rates of childhood AML associated with `solvent' and `petroleum' exposure to the father, but these associations were not observed with maternal exposures. These investigators also report that the paternal, but not maternal, occupations of painting and mechanic work were associated with a significantly elevated risk of childhood AML [64]. Magnani et al. [65] reported a highly significant association with childhood AML with paternal occupational exposures in tire production. However, this was based on only two cases [65]. Van Duijn et al. [66] did not observe a relationship between maternal exposures to hydrocarbons and childhood ANLL. Alderton et al. [57] did not observe a positive association with childhood AML associated with maternal exposure to either paints or petroleum in children with Down's syndrome. More recently, Scelo et al. [48] evaluated several types of exposures and did not report elevated risk estimates with maternal exposures to paints or petroleum solvents. Moreover, paternal exposure to paints resulted in a statistically significant decrease in the risk of childhood AML. Paternal exposure to petroleum solvents was not associated positively or negatively with childhood AML [48]. Considered collectively, there is no consistent support in the scientific and medical literature that parental exposure to solvents, petroleum-based hydrocarbons, benzene or benzene containing products is a causative factor in the development of childhood ALL or AML. Some of the studies reported elevated risks associated with maternal but not paternal exposures, while other studies reported opposite findings. In either case, the majority of the studies were negative and as mentioned earlier, all suffer from a long list of limitations. The three largest studies all failed to find an association between parental benzene or hydrocarbon exposure and children's ALL or AML [51,52,56]. 4.5. Household exposures There have been a few studies that have attempted to evaluate the risk of childhood leukemia associated with household exposures to solvents (Table 4). Most of these exposures were associated with hobbies such as painting or model building. One study reported an increase in ALL risk associated with artwork (solvents?) in the home [54]. However, furniture stripping and mechanical work was not associated [54]. Lowengart et al. [44] did not report an increase in childhood leukemia risk associated with spray painting in the home. This was consistent with Alderton et al. [57] that also found no increases in either childhood ALL or AML associated with home use of paints or petroleum products. 4.6. Parental occupations involving motor vehicles There have been at least twelve studies of childhood leukemia and parental (mostly paternal) employment in occupations related to motor vehicles (Table 5). These workers include drivers, mechanics, gasoline station attendants and others that would have exposure to exhaust emissions, fuel, oil and other petroleum products and potentially cleaning solvents. Of these, two reported significant excesses in childhood leukemia [64,67]. Vianna et al. [67] reported an increase in acute infant leukemia associated with paternal occupation as a mechanic or gas station attendant. Buckley and Robinson [64] reported a significant excess in childhood AML associated with paternal occupation in the category of `mechanic'. In contrast, the remaining ten studies reported no increase in risk associated with drivers, mechanics, gas station attendants or transportation workers [45,49,61,64,65,6773]. Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx 9 4.7. Cigarette smoking There is a robust literature that has attempted to characterize the potential risks to children associated with parental smoking during pregnancy or after birth (passive smoke). Many of these studies have specifically addressed childhood ALL and/or AML as an endpoint. The majority of these studies do not support the position that parental smoking or living in a smoking household results in an increased risk of either form of childhood leukemia. In fact, many studies provide evidence of a decreased risk of childhood leukemia associated with maternal or paternal smoking (Tables 6 and 7). The literature on the effect of maternal or paternal smoking on childhood leukemia rates is highly relevant as cigarette smoke contains a variety of hazardous substances, including butadiene and benzene. In fact, it is also generally agreed upon that in smoking households, cigarette smoke is the largest non-occupational contributor of benzene (and likely butadiene) to those individuals [74,75]. Therefore, if low level exposure to these chemicals can indeed result in an increased risk of developing childhood ALL/AML, then the smoking literature should reflect this. Studies have been segregated according to maternal or paternal smoking. 4.8. Maternal smoking Van Steensel-Moll et al. [45] published one of the first studies to evaluate the potential risk of ALL associated with maternal smoking. In this study, no association was reported [45]. Buckley et al. [76] reported that the risk of ALL in children from mothers who smoked more than 10 cigarettes/day while pregnant was not elevated. Stjernfeldt et al. [77] published one of the few positive studies and reported an increased risk of ALL associated with maternal smoking. Klebanoff et al. [78] conducted one of the largest studies conducted on this topic and evaluated the risk of maternal smoking in 54,000 children. In contrast to Stjernfeldt et al. [77], the risk of childhood leukemia associated with maternal smoking was not elevated in the larger Klebanoff study [78]. Chang et al. [79] found no association with maternal smoking and ALL risk even if the smoking overlapped with the breastfeeding period. Menegaux et al. [80] was also negative for maternal smoking during pregnancy or breastfeeding. Many smaller studies also reported negative findings [66,72,8184]. Several studies specifically evaluated childhood AML risk associated with maternal smoking. This is particularly relevant as cigarette smoking is an established risk factor for AML in adults [34,8587]. Cnattingius et al. [88] did not report a significant excess in AML rates associated with maternal smoking. In contrast, Shu et al. [89] actually reported a statistically significant deficit in AML rates if the mother smoked during pregnancy. Additionally, these authors reported a statistically significant decrease in ALL if the mother smoked during the nursing period [89]. In 1999, Brondum et al. [90] found no elevation in childhood ALL or AML associated with maternal smoking. In fact, in the cohort of women who smoked for longer than 20 years, there was also a statistically significant decrease in ALL rates reported [90]. A significant decrease in AML rates was also reported by Pang et al. [91] and Trivers et al. [92], although the latter was marijuana smoke. Other studies of parental use of marijuana yielded inconsistent results [9395]. A large cohort of 1,440,542 Swedish children was evaluated by Mucci et al. [96] for a potential link to childhood ALL/AML with maternal smoking. In this study, maternal smoking was also found to be associated with a statistically significant deficient of ALL. On the other hand, there was evidence, particularly among very heavy smokers, of an increase in AML [96]. Clavel et al. [97] reported no association with childhood leukemia and maternal smoking. The exception was an elevated risk observed with smoking and a subset of children that were found to possess genetic polymorphisms in certain metabolically important genes (CYP1A1 and GSTM1) [97]. This interaction with polymorphisms was also observed with the consumption of coffee during pregnancy; in fact, the associated risk for coffee consumption was 3-fold higher than the smoking risk [97]. NCI investigators conducted a meta-analysis and reported no elevated risk associated with maternal smoking and all hematopoietic malignancies or for acute leukemias (all combined) [98]. These findings were supported by most, but not all recent studies [80,99,100]. 4.9. Paternal smoking There were fewer studies in the literature that evaluated the potential role of paternal smoking and the risk on childhood leukemia. However, the majority of these studies were consistent with the maternal smoking literature (Table 7). Magnani, et al. [65] reported that paternal smoking did not result in an increased risk of childhood ALL or AML. These findings were corroborated by multiple smaller studies [82,89,90,101,102]. In contrast, Chang et al. [79] reported a marginally significant increase in AML risk associated with paternal smoking. Paternal smoking was not associated with an increase risk in ALL; however, when both parents smoked the risk for ALL was also elevated [79]. Rudant et al. [99] reported an increase in ALL, but not AML associated with paternal smoking (the AML was of borderline significance). Recently, MacArthur et al. [100] found no association with paternal smoking and the risk of childhood ALL, but Lee et al. [103] did find a significant excess in ALL associated with certain genetic polymorphisms. One study, Chang et al. [79] explicitly addressed passive smoke and childhood leukemia risk. In this study, a statistically significantly decrease in ALL risk was found associated with children's exposure to passive cigarette smoke [79]. Taken collectively, there is good evidence in the open scientific and medical literature that cigarette (or marijuana) smoke is not a risk factor for childhood ALL or AML. This includes both maternal and paternal smoking as well as when the child was exposed to passive cigarette smoke. In fact, the largest study to date actually reports a statistically significant deficit in ALL incidences in children with smoking mothers. Cigarette smoke contains a variety of carcinogenic substances including benzene and butadiene. On average, there are 48 g benzene and 38.5 g butadiene per cigarette [104]. It has been shown that cigarette smoke is the largest contributor of benzene in non-occupational environments [74,75,105,106]. Therefore, if low level exposure to benzene and/or butadiene in the environment actually increased the risk for childhood leukemia, this association should have been evident in the smoking literature, particularly for AML, since cigarette smoking is an established risk factor for adult AML. This does not appear to be the case and provides direct evidence that low, environmentally relevant concentrations of benzene and/or butadiene do not appreciably increase the risk of developing either subtype of childhood leukemia. Recently, IARC has confirmed that cigarette smoking is causally related to adult AML, but not to childhood leukemia [107]. 4.10. Atmospheric contaminants Another source of chemical exposure explored as a potential cause of childhood leukemia includes atmospheric contaminants such as benzene and butadiene. Unfortunately, the majority of studies that have evaluated this potential association have used indirect measures such as traffic density, car ownership, gasoline consumption or proximity to industrial sources as surrogate measures in lieu of quantitative exposure information (data). Without reliable exposure estimates, a great deal of uncertainty and potential bias is introduced into the actual analysis, particularly when it comes to exposureresponse relationships and/or attempting to establish a Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 10 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx definitive association between a given health effect and a specific chemical exposure. 4.11. Traffic density One of the more common exposure surrogates for air pollution found in the literature is traffic density. These studies are essentially based on the assumption that a linear relationship exists between traffic density and the levels of air contaminants found in vehicle exhaust and emissions and thus exposure to children. While not conceptually incorrect, there are many other variables potentially associated with traffic density that could, without careful evaluation, confound the interpretation of the data. For example, there have been reports of an increased risk in childhood leukemia associated with increased population density (which may or may not have anything to do with the number of cars) [108]. The results of these `ecological' type investigations are briefly discussed below (Table 8). The first study to attempt to use traffic density as a surrogate for exposure was Savitz and Feingold [109]. These authors reported an increase in childhood leukemia rates associated with car density, with higher risk estimates seen for areas with more cars. The odds ratio for childhood leukemia was significantly elevated for areas with greater than 10,000 vehicles per day [109]. In a follow-up study, Pearson et al. [110] also report a significant association between distance-weighted traffic density and childhood leukemia. However, only the `greater than 20,000 cars/day' category was elevated. No elevation was observed in the `15,00019,999' cars/day categories. The lack of a meaningful dose response could either indicate a threshold (above 19,999 cars/day) or alternatively it could be influenced by confounding variables [110]. Swaen and Slangen [111] conducted a study attempting to correlate childhood leukemia and gasoline consumption in a variety of European communities. This is an even more indirect measure of potential exposure to vehicle emissions and corresponding chemicals. However, they did not observe a relationship with childhood leukemia [111]. In a somewhat similar study, Alexander et al. [112] reported that there was no increase in risk of childhood ALL associated with car ownership. Nordlinder and Jarvholm [113] conducted a small ecological study in Sweden and reported a modest increase in childhood AML associated with car density but did not observe this same increase for ALL. The reported incidence for childhood AML was slightly elevated for 59 cars/km2 when compared to less than 5 cars/km2. There was no further increase in AML incidence at 1019 cars/km2 or >20 cars/km2. The authors speculate that the increase in AML risk might be related to benzene in the petrol or engine exhaust but do not offer an explanation as to the lack of an exposure response relationship [113]. In a related study, Feychting et al. [114] evaluated the risk of childhood leukemia associated with NO2 concentrations which were used as a surrogate for exposure, including benzene. These authors did not observe a statistically significant increase in the risk of childhood leukemia in areas with higher NO2 concentrations. In a similar Danish study, RaaschouNielsen et al. [115] also reported no change in childhood leukemia rates associated with increasing benzene or NO2 concentrations caused by increased traffic density. This study also reported modeled estimates of benzene exposure at the homes of the subjects [115]. Reynolds et al. [116] reported that traffic density was highly correlated with butadiene and benzene air concentrations. However, the OR for children living in the 90th percentile for traffic density was not elevated for childhood leukemia. The subtypes of children leukemia were broken out and there was no evidence of increased ALL or AML risk associated with traffic density [116]. These same authors published a follow-up paper reporting a modest increase in childhood leukemia associated with `hazardous air pollutants' (HAP). There was no exposure information provided for specific chemical constituents. However, based on data from their previous study, butadiene and benzene do not appear to be driving the observed risk [117]. These authors have also published a very large study that reported no relationship between traffic density and children's leukemia [118]. At least 5 other smaller studies corroborated the negative findings of Reynolds et al. [116,115,119123]. Crosignani et al. [124] attempted to associate the rates of childhood leukemia to traffic density and modeled benzene air concentrations. These investigators reported that children with the highest average benzene exposure (>10 g benzene/m3) had a significantly elevated risk of developing childhood leukemia compared to those with the lowest benzene concentrations. As a result, these authors suggest that traffic emissions may be involved in the etiology of childhood leukemia [124]. Weng et al. [125] also reported a significant excess in the risk of childhood leukemia associated with traffic density in Taiwan. While there are some positive findings associated with traffic density, the published data is too inconsistent to support the conclusion that vehicle emissions play a role in the development of childhood AML or ALL. Moreover, the rates of childhood ALL have been slightly increasing over the past 3 decades, while an opposite trend in emissions from automobiles has occurred [126]. If vehicle emissions were really an etiological factor in childhood leukemia, there should be clear differences between urban and rural populations or in cities with increased air pollution (e.g. Mexico City). These relationships have not been reported, which provides some crude evidence against a role for vehicle emissions in the development of childhood leukemia. However, the positive studies are suggestive and additional investigation may be warranted. 4.12. Proximity to refineries and/or gas stations Another indirect approach for attempting to evaluate chemical exposures and childhood leukemia risk is using proximity to potential sources as a surrogate for exposure. This approach contains many potential confounding variables and the results, whether positive or negative, should be interpreted cautiously. In most of these studies, actual air measurements of benzene were not reported or used. The premise of these studies is that children are being exposed to air emissions from various facilities and that chemical concentrations should be inversely proportional to the distance from the source. These studies are briefly summarized below (Table 7). Hearey et al. [127] evaluated a portion of the San Francisco Bay area with a high percentage of chemical and petroleum facilities compared with other parts of the area. There was no association with leukemia or any other cancer in children or adults associated with living near these petrochemical industrial sites. They concluded that having a residence near various petrochemical industries is not associated with an increased cancer risk including childhood leukemia [127]. Knox [128] reported on 22,458 children in the UK from 1953 to 1980. In this ecological study, `childhood leukemias' were reported to be geographically associated with industrial atmospheric effluents. These findings, as well as others by this same author, have been strongly challenged on methodological grounds [128,129]. Knox has published a related series of reports, using approximately the same methodology and attempting to relate cancer risk in children to a variety of industrial sources of air pollution. These studies are consistently positive, although no attempt to separate out specific types of childhood cancer or chemicals is made. In contrast, Wilkinson et al. [130] carried out a casecontrol study in the UK. There was no evidence of a change in the incidence rates of any type of hematopoietic malignancy in children, includ- Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx 11 ing AML and ALL, associated with proximity to large industrial oil complexes/refineries in Great Britain [130]. Lyons et al. [131] did not observe a significant excess of leukemias in young people or children associated with proximity to the British Petroleum (BP) Chemical Plant in South Wales. A companion study by Sans et al. [132] confirmed these findings and reported no decline in childhood leukemia rates (mortality or incidence) associated with increasing distance from this facility. Steffan et al. [122] reported that proximity to repair garages/gas stations in France resulted in an increased risk of both childhood ALL and AML. In this study, the exposure period that was the most important for risk was during childhood, not in utero [122]. Although the authors speculate benzene might be emitted from these facilities, no air data or actual measurements were presented. This study has been updated and expanded and the earlier results have been confirmed [133]. In contrast to their earlier study, intrauterine exposures were the most important [133]. Additionally, in a very similar study, Harrison et al. [119] found no evidence of an association with childhood leukemia and proximity to petrol stations. A small study was conducted in Ecuador and reported elevated risks of ALL associated with proximity to oil exploration areas [134]. Recently, a study was published that attempted to correlate air pollution levels with the proportion of individuals in a neighborhood employed by the petrochemical industry. In this study, there was a reported trend in childhood leukemia rates associated with this indirect surrogate for exposure [135]. A subsequent analysis by these authors also reported a significant excess in childhood leukemia rates associated with the highest density of petrol stations, again used as an indirect surrogate for exposure data [136]. The lack of exposure data makes it hard to interpret these findings. Whitworth et al. [137] attempted to correlate modeled benzene and 1,3-butadiene air concentrations to childhood leukemia in Houston Texas. There was no association observed between benzene exposures and ALL, but a modest increase was reported for AML [137]. As can be seen, there is considerable inconsistency in studies attempting to relate proximity to industrial installations and childhood leukemia risk. This is likely due to the uncertainty associated with such an indirect method of estimating exposures. Additionally, there are many quantitative studies that have demonstrated that industrial facilities such as refineries and petrochemical facilities actually contribute very little to an individual's non-occupational benzene exposure [106]. This is difficult to reconcile with suggestions that observed increases in childhood leukemia risk are related to fugitive emissions and illustrates the importance in collecting appropriate exposure data. 5. Summary High dose, occupational exposures to benzene is an established cause of adult AML, raising the question of whether children can develop leukemia at far lower, environmental exposures to benzene. This review has evaluated and compiled the studies that address (either directly or indirectly) whether environmental benzene exposure is an etiologic risk factor for childhood leukemia. Childhood leukemia is not a single disease, but is comprised of two major subtypes of leukemia, ALL and AML. These two subtypes are distinct diseases that do not share the same etiologies or risk factors [1]. Approximately 80% of all childhood leukemia cases are ALL. Therefore, epidemiology studies that combine the various types of leukemia under the rubric of childhood leukemia will necessarily be strongly influenced by the ALL cases and may actually mask potential relationships between exposures and childhood AML. Many older studies combined the two main types if childhood leukemia into one group. However, more recent studies recognize this important distinction and have started to conduct epidemiologic analysis on the unique subtypes of childhood leukemia. This type of study is far more informative. Another widespread problem with the literature evaluated in this review is the absence of any meaningful quantitative exposure information. Lacking critical quantitative exposure data, it becomes impossible to clearly identify specific exposures that may be responsible for the observed effects. In almost every category there are some positive results. However, additional `ecological' studies are unlikely to provide epidemiologic clarity of this issue and are of limited value (in these authors' opinion). Rigorous quantification of exposure and better disease classification are needed in all future investigations of this issue. Potential environmental risk factors for childhood leukemia (all types combined) including parental exposures to chemicals (that may or may not contain benzene), potential exposure to atmospheric pollutants to children, and residential proximity to industrial facilities or other potential chemical sources have been assessed in dozens of epidemiology studies. There is no consistent pattern of evidence relating any of these factors to an increased risk of childhood ALL or AML. Cigarette smoke contains substantial amounts of butadiene and benzene and smoking represents a significant contributor to an individual's non-occupational exposure to these chemicals. If low level exposure to butadiene or benzene was associated with childhood leukemia, then the smoking literature should have reflected that association. This does not appear to be the case as the vast majority of studies does not indicate that cigarette smoke is an etiologic risk factor for childhood AML or ALL. This was also the conclusion reached by IARC. Therefore, at this point, there is insufficient epidemiologic support for an association or causal connection between environmental benzene exposure (however that is defined) and the development of childhood ALL or AML. A formal meta-analysis of this data could help elucidate more subtle associations between environmental exposures and either type of childhood leukemia if they truly exist. Conflict of interest statement The American Petroleum Institute provided partial funding for this project. They have a long-standing interest in benzene toxicity and epidemiology. References [1] S. Swerdlow, E. Campo, N. Harris, E. Jaffe, S. Pileri, H. Stein, J. Thiele, J.W. Vardiman (Eds.), Pathology & Genetics Tumours of Haematopoietic and Lymphoid Tissues, International Agency for Research on Cancer, Lyon, 2008. [2] J.H. Jandl (Ed.), Blood, Textbook of Hematology, Little, Brown & Company, Boston, 1997. [3] P. Buffler, M. Kwan, P. Reynolds, K. Urayama, Environmental and genetic risk factors for childhood leukemia: appraising the evidence, Cancer Invest. 1 (2005) 6075. [4] M. Belson, B. Kingsley, A. Holmes, Risk factors for acute leukemia in children: a review, Environ. Health Perspect. 115 (1) (2007) 138145. [5] R. McNally, L. Parker, Environmental factors and childhood acute leukemia and lymphomas, Leukemia Lymphoma 47 (4) (2006) 583598. [6] J. Little, Epidemiology of Childhood Cancer, IARC Scientific Publication 149, International Agency for Research on Cancer, 1999. [7] S.A. Narod, Genetic epidemiology of childhood cancer, Biochim. Biophys. Acta: Rev. Cancer 1288 (3) (1996) F141F150. [8] S.A. Narod, C. Stiller, G.M. Lenoir, An estimate of the inheritable fraction of childhood cancer, Br. J. Cancer 63 (6) (1991) 993999. [9] M.S. Watson, A.J. Carroll, J.J. Shuster, C.P. Steuber, M.J. Borowitz, F.G. Behm, D.J. Pullen, V.J. Land, Trisomy-21 in childhood acutre lymphoblastic leukemia--a pediatric-oncology group study, Blood 82 (10) (1993) 30983102. [10] L. Robinson, M. Nesbitt, H. Sather, C. Level, N. Shahidi, D. Hammond, Down Syndrome and acute leukemia in children: a 10-year retrospective survey from Children's Cancer Study Group, J. Pediatrics 105 (2) (1984) 235242. [11] C. Pui, M. Schrappe, R. Ribeiro, C. Neimeyer, Childhood and adolescent lymphoid and myeloid leukemia, Hematology (Am Soc of Hematology Ed Program) (2004) 118145. Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 12 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx [12] J. Wiemels, Z. Xiao, P. Buffler, A. Mala, X. Ma, B. Dicks, M. Smith, L. Zhang, J. Feusner, J. Wiencke, K. Pritchard-Jones, H. Kempski, M. Greaves, In utero origin ay t(8;21) AML1-ETO translocations in childhood acute myeloid leukemia, Blood 99 (2002) 38013805. [13] J.L. Wiemels, G. Cazzaniga, M. Daniotti, O.B. Eden, G.M. Addison, G. Masera, V. Saha, A. Biondi, M.F. Greaves, Prenatal origin of acute lymphoblastic leukaemia in children, Lancet 354 (9189) (1999) 14991503. [14] M. Greaves, Pre-natal origins of childhood leukemia, Rev. Clin. Exp. Hematol. 7 (3) (2003) 233245. [15] M.F. Greaves, J. Wiemels, Origins of chromosome translocations in childhood leukaemia, Nat. Rev. Cancer 3 (9) (2003) 639649. [16] J.E. Rubnitz, C.H. Pui, J.R. Downing, The role of TEL fusion genes in pediatric leukemias, Leukemia 13 (1) (1999) 613. [17] S.A. Shurtleff, A. Buijs, F.G. Behm, J.E. Rubnitz, S.C. Raimondi, M.L. Hancock, G.C.F. Chan, C.H. Pui, G. Grosveld, J.R. Downing, TEL/AML1 fusion resulting from a cryptic t(12;21) is the most common genetic lesion in pediatric ALL and defines a subgroup of patients with an excellent prognosis, Leukemia 9 (12) (1995) 19851989. [18] R.L. Tower, L.G. Spector, The epidemiology of childhood leukemia with a focus on birth weight and diet, Crit. Rev. Clin. Lab. Sci. 44 (3) (2007) 203242. [19] D.L. Preston, S. Kusumi, M. Tomonaga, S. Izumi, E. Ron, A. Kuramoto, N. Kamada, H. Dohy, T. Matsui, H. Nonaka, D.E. Thompson, M. Soda, K. Mabuchi, Cancer incidence in atomic bomb survivors. Part III. Leukemia, lymphoma and multiple myeloma, 19501987, Radiat. Res. 137 (1994) S68S97. [20] R. Doll, R. Wakeford, Risk of childhood cancer from fetal irradiation, Br. J. Radiol. 70 (1997) 130139. [21] R.R. Monson (Ed.), Occupational Epidemiology, CRC Press, Inc., Boca Raton, 1990. [22] G. Caldwell, Twenty-two years of cancer cluster investigations at the Centers for Disease Control, Am. J. Epidemiol. 132 (Suppl. 1) (1990) S1S202. [23] S. Olsen, M. Martuzzi, Cluster analysis and disease mapping--why, when and how? A step by step guide, Br. Med. J. 313 (7061) (1996) 863866. [24] P Langmuir, R. Aplenc, B. Lange, Acute myeloid leukaemia in children, Best Pract. Res. Clin. Haematol. 14 (1) (2001) 7793. [25] J.P. Greer, J. Johnson, G.M. Rodgers, F. Paraskevas, B. Glader (Eds.), Wintrobe's Clinical Hematology, Lippincott Williams & Wilkins, 2008. [26] J.W. Vardiman, N.L. Harris, R.D. Brunning, The World Health Organization (WHO) classification of the myeloid neoplasms, Blood 100 (7) (2002) 22922302. [27] T. Leblanc, R. Berger, Molecular cytogenetics of childhood acute myelogenous leukemias, Eur. J. Haematol. 59 (1997) 113. [28] S. Bhatia, J. Neglia, Epidemiology of childhood acute myelogenous leukemia, J. Pediatr. Hematol. Oncol. 17 (2) (1995) 94100. [29] T. Lightfoot, Aetiology of childhood leukemia, Bioelectromagnetics Suppl. 7 (2005) S5S11. [30] F. Giles, A. Keating, A. Goldstone, I. Avivi, C. Willman, H. Kantarjian, Acute myeloid leukemia, Hematology (Am. Soc. Hematology Educ. Program) (2002) 73110. [31] J. Pedersen-Bjergaard, M.K. Andersen, D.H. Christiansen, C. Nerlov, Genetic pathways in therapy-related myelodysplasia and acute myeloid leukemia, Blood 99 (6) (2002) 19091912. [32] J. Pedersen-Bjergaard, M.K. Andersen, M.T. Andersen, D.H. Christiansen, Genetics of therapy-related myelodysplasia and acute myeloid leukemia, Leukemia 22 (2) (2008) 240248. [33] G. Leone, L. Pagano, D. Ben-Yehuda, M. Voso, Therapy-related leukemia and myelodysplasia: susceptibility and incidence, Haematologica 92 (2007) 13891398. [34] D. Sandler, J.A. Ross, Epidemiology of acute leukemia in children and adults, Semin. Oncol. 24 (1) (1997) 316. [35] J.A. Ross, J.D. Potter, L.L. Robison, Infant leukemia, topoisomerase II inhibitors, and the MLL gene, J. Natl. Cancer Inst. 86 (22) (1994) 16781680. [36] J.A Ross, S.M. Davies, J.D. Potter, L.L. Robison, Epidemiology of childhood leukemia, with a focus on infants, Epidemiol. Rev. 16 (2) (1994) 243272. [37] M. Greaves, A natural history for pediatric acute leukemia, Blood 82 (1993) 10431051. [38] F.E. Alexander, S.L. Patheal, A. Biondi, S. Brandalise, M.E. Cabrera, L.C. Chan, Z. Chen, G. Cimino, J.C. Cordoba, L.J. Gu, H. Hussein, E. Ishii, A.M. Kamel, S. Labra, I.Q. Magalhaes, S. Mizutani, E. Petridou, M.P. De Oliveira, P. Yuen, J.L. Wiemels, M.F. Greaves, Transplacental chemical exposure and risk of infant leukemia with MLL gene fusion, Cancer Res. 61 (6) (2001) 25422546. [39] S. Koifman, F M.S. Pombo-de-Oliveira, L. Brazilian Collaborative Study Group of infant acute, high birth weight as an important risk factor for infant leukemia, Br J Cancer 98 (3) (2008) 664667. [40] T. Lightfoot, E. Roman, Causes of childhood leukemia and lymphoma, TAP 199 (2004) 104117. [41] X. Shu, J. Perentesis, W. Wen, J. Buckley, E. Boyle, J. Ross, L. Robison, Parental exposure to medications and hydrcarbons and ras mutations in children with acute lymphoblastic leukemia: a report from the children's oncology group, Cancer Epidemiol. Biomarkers Prev. 13 (7) (2004) 12301235. [42] R.A. Rinsky, A.B. Smith, R.W. Hornung, T.G. Filloon, R.J. Young, A.H. Okun, P.J. Landrigan, Benzene, leukemia: an epidemiologic risk assessment, N. Engl. J. Med. 316 (17) (1987) 10441050. [43] USEPA, IRIS, Integrated Risk Information System: Benzene, 2000, http://cfpub.epa.gov/ncea/iris/index.cfm. [44] R. Lowengart, J. Peters, C. Cicioni, Childhood leukemia and parents occupational and home exposure, J. Natl. Cancer Inst. 70 (1987) 3946. [45] H.A. Van Steensel-Moll, H.A. Valkenburg, G.E. Van Zanen, Childhood leukemia and parental occupation, Am. J. Epidemiol. 121 (1985) 216224. [46] Y. Mulder, M. Drijver, I. Kreis, Casecontrol study on the association between a cluster of childhood haematopoietic malignancies and local environmental factors in AAlsmeer, The Netherlands, J. Epidemiol. Community Health 48 (1994) 161165. [47] V. Smulevich, L. Solionova, S. Belyakova, Parental occupation and other factors and cancer risk in children. II. Occupational factors, Int. J. Cancer 83 (1999) 718722. [48] G. Scelo, C. Metayer, L.P. Zhang, J.L. Wiemels, M.C. Aldrich, S. Selvin, S. Month, M.T. Smith, P.A. Buffler, Household exposure to paint and petroleum solvents, chromosomal translocations, and the risk of childhood leukemia, Environ. Health Perspect. 117 (1) (2009) 133139. [49] T Hakulinen, T. Salonen, L. Teppo, Cancer in the offspring of fathers in hydrocarbon-related occupations, Br. J. Prev. Soc. Med. 30 (1976) 138140. [50] M. Zack, S. Cannon, D. Loyd, C. Heath, J. Falletta, B. Jones, J. Housworth, S. Crowley, Cancer in children of parents exposed to hydrocarbon-related industries and occupations, Am. J. Epidemiol. 111 (3) (1980) 329336. [51] X. Shu, P. Stewart, W. Wen, D. Han, J. Potter, J. Buckley, E. Heineman, L. Robinson, Parental occupational exposure to hydrocarbons and risk of acute lymphocytic leukemia in offspring, Cancer Epidemiol. Biomarkers Prev. 8 (1999) 783791. [52] J. Schuz, U. Kaletsch, R. Meinert, P. Kaatsch, J. Michaelis, Risk of childhood leukemia and parental self-reported occupational exposure to chemcials, dusts andf fumes: results from pooled analyses of German population-based casecontrol studies, Cancer Epidemiol. Biomarkers Prev. 9 (2000) 835838. [53] L. Feingold, D. Savitz, E. John, Use of a job-exposure matrix to evaluate parental occupation and childhood cancer, Cancer Causes Control 3 (2) (1992) 161169. [54] D.M. Freedman, P. Stewart, R.A. Kleinerman, S. Wacholder, E.E. Hatch, R.E. Tarone, L.L. Robison, M.S. Linet, Household solvent exposures and childhood acute lymphoblastic leukemia, Am. J. Public Health 91 (4) (2001) 564567. [55] P. McKinney, N. Fera, D. Stockton, Parental occupation at periconception: findings from the United Kingdom Childhool Cancer Study, Occup. Environ. Med. 60 (2003) 901909. [56] C. Infante-Rivard, J. Siemiatycki, R. Lakhani, L. Nadon, Maternal exposure to occupational solvents and childhood leukemia, Environ. Health Perspect. 113 (6) (2005) 787792. [57] L.E. Alderton, L.G. Spector, C.K. Blair, M. Roesler, A.F. Olshan, L.L. Robison, J.A. Ross, Child and maternal household chemical exposure and the risk of acute leukemia in children with Down's syndrome: A report from the children's oncology group, Am. J. Epidemiol. 164 (3) (2006) 212221. [58] A. Zipursky, P. Thorner, E. Harven, Myelodysplasis and acute megakaryoblastic leukemia in Down's Syndrome, Leukemia Res. 18 (1994) 163171. [59] G. Shaw, R. Lavey, R. Jackson, D. Austin, Association of childhood leukemia with maternal age, birth order, and parental occupation: a casecontrol study, Am. J. Epidemiol. 119 (5) (1984) 788795. [60] X.O. Shu, Y.T. Gao, L.A. Brinton, M.S. Linet, J.T. Tu, W. Zheng, J. Fraumeni, A population-based casecontrol study of childhood leukemia in Shanghai, Cancer 62 (1988) 635644. [61] P.A. McKinney, F.E. Alexander, R.A. Cartwright, L. Parker, Parental occupations of children with leukaemia in west Cumbria, north Humberside, and Gateshead, Br. Med. J. 302 (1991) 681687. [62] P Kaatsch, U. Kaletsch, R. Meinert, A. Miesner, M. Hoisl, J. Schuz, J. Michaelis, German case control study on childhood leukemia--basic considerations, methodology and summary of the results, Klin. Padiatr. 210 (1998) 185191. [63] M. Feychting, N. PLato, G. Nise, A. Ahlbom, Paternal occupational exposures and childhood cancer, Environ. Health Perspect. 109 (2) (2001) 193196. [64] J. Buckley, L. Robinson, Occupatinal exposures of parents of children with acute nonlymphocytic leukemia: a report form the Children's Cancer Study Group, Cancer Res 49 (1989) 40304037. [65] C. Magnani, G. Pastore, L. Luzzatto, B. Terracini, Parental occupation and other environmental factors in the etiology of leukemias and non-Hodgkin's lymphomas in childhood: a casecontrol study, Tumori 76 (5) (1990) 413419. [66] C.M. Van Duijn, A. Vansteenselmoll, J.W.W. Coebergh, G.E. Vanzanen, Risk-factors for childhood acute nonlymphocytic leukemia- An association with maternal alcohol-consumption during pregnancy, Cancer Epidemiol. Biomarkers Prev. 3 (6) (1994) 457460. [67] N.J. Vianna, B. Kovasznay, A. Polan, C. Ju, Infant leukemia and paternal exposure to motor vehicle exhaust fumes, J. Occup. Med. 26 (9) (1984) 679682. [68] J. Fabia, T. Thuy, Occupation of fathers at the time of birth of children dying of malignant diseases, Br. J. Prev. Soc. Med. 28 (1974) 98100. [69] S.L. Kwa, L.J. Fine, The association between parental occupation and childhood malignancy, J. Occup. Med. 22 (12) (1980) 792794. [70] K. Hemminki, I. Saloniemi, T. Salonen, T. Partanen, H. Vainio, Childhood cancer and parental occupation in Finland, J. Epidemiol. Community Health 35 (1) (1981) 1115. [71] E.B. Gold, M.D. Diener, M. Szklo, Parental occupations and cancer in children--a casecontrol study and review of the methodologic issues, J. Occup. Med. 24 (8) (1982) 578584. [72] E. Roman, A. Watson, V. Beral, S. Buckle, D. Bull, K. Baker, H. Ryder, C. Barton, Casecontrol study of leukaemia and non-Hodgkin's lymphoma among children aged 04 years living in west Berkshire and north Hampshire health districts, Br. Med. J. 306 (6878) (1993) 615621. [73] M.L. Perez-Saldivar, M.C. Ortega-Alvarez, A. Fajardo-Gutierrez, R.M Bernaldez-Rios, d.L.A. Del Campo-Martinez, A. Medina-Sanson, M.A. Palomo- Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx 13 Colli, R. Paredes-Aguilera, A. Martinez-Avalos, V.H.M. Borja-Aburto, d.J. Rodriguez-Rivera, V.M. Vargas-Garcia, J. Zarco-Contreras, J. Flores-Lujano, J.M. Mejia-Arangure, Father's occupational exposure to carcinogenic agents and childhood acute leukemia: a new method to assess exposure (a casecontrol study), BMC Cancer 8 (7) (2008), Published online: 10.1186/1471-2407r-r1188-1187. [74] L. Wallace, Environmental exposure to benzene: an update, Environ. Health Perspect. 104 (Suppl. 6) (1996) 11291136. [75] L. Wallace, E. Pellizzari, T.D. Hartwell, R. Perritt, R. Ziegenfus, Exposures to benzene and other volatile compounds from active and passive smoking, Arch. Environ. Health 42 (5) (1987) 272279. [76] J. Buckley, W. Hobbie, K. Ruccione, H. Sather, W. Woods, G. Hammond, Maternal smoking during pregnancy and the risk of childhood cancer, Lancet 2 (1986) 519520. [77] M. Stjernfeldt, K. Berglund, J. Lindsten, J. Ludvigsson, Maternal smoking during pregancy and risk of childhood cancer, Lancet 14 (June) (1986) 13501352. [78] M. Klebanoff, J. Clemens, J. Read, Maternal smoking during pregnancy and childhood cancer, Am. J. Epidemiol. 144 (11) (1996) 10281033. [79] J. Chang, S. Selvin, V. Crouse, P. Buffler, Parental smoking and the risk of childhood leukemia, Am. J. Epidemiol. 163 (12) (2006) 10911100. [80] F. Menegaux, M. Ripert, D. Hemon, J. Clavel, Maternal alcohol and coffee drinking, parental smoking and childhood leukaemia: a French population-based casecontrol study, Paediatr. Perinat. Epidemiol. 21 (4) (2007) 293299. [81] P.A McKinney, C.A. Stiller, Maternal smoking during pregnancy and the risk of childhood-cancer, Lancet 2 (8505) (1986) 5191519. [82] E. John, D. Savitz, D. Sandler, Prenatal exposure to parents smoking and childhood cancer, Am. J. Epidemiol. 133 (2) (1991) 123132. [83] J.D. Urquhart, R.J. Black, M.J. Muirhead, L. Sharp, M. Maxwell, O.B. Eden, D.A. Jones, Case-control study of leukaemia and non-Hodgkin's lymphoma in children in Caithness near the Dounreay nuclear installation, Br. Med. J. 302 (1991) 687692. [84] G. Pershagen, A. Ericson, P. Otterbladolausson, Maternal smoking in pregnancy--Does it increase the risk of childhood-cancer? Int. J. Epidemiol. 21 (1) (1992) 15. [85] D.P. Sandler, D.L. Shore, J.R. Anderson, F.R. Davey, D. Arthur, R.J. Mayer, R.T. Silver, R.B. Weiss, J.O. Moore, C.A. Schiffer, D.H. Wurster-Hill, O.R. McIntyre, C.D. Bloomfield, Cigarette smoking and risk of acute leukemia: Associations with morphology and cytogenetic abnormalities in bone marrow, J. Natl. Cancer Inst. 85 (1993) 19942003. [86] A.V. Moorman, Smoking and the risk of myeloid leukemia in cytogenetic subgroups, Br. J. Cancer 86 (2002) 6062. [87] X. Thomas, Y. Chelghoum, Cigarette smoking acute leukemia, Leukemia Lymphoma 45 (6) (2004) 11031109. [88] S. Cnattingius, M. Zack, A. Ekbom, J. Gunnarskog, M. Linet, H. Adami, Prenatal and neonatal risk factors for childhood myeloid leukemia, Cancer Epidemiol. Biomarkers Prev. 4 (1995) 441445. [89] X. Shu, J. Ross, G. Reaman, Are parental alcohol consumption and cigarette smoking associated with risk of infant leukemia? A report for the Children's Cancer Group, J. Natl. Cancer Inst. 88 (1996) 2431. [90] J. Brondum, X.O. Shu, M. Steinbuch, R.K. Severson, J.D. Potter, L.L. Robison, Parental cigarette smoking and the risk of acute leukemia in children, Cancer 85 (1999) 13801388. [91] D. Pang, R. McNally, J. Birch, Parental smoking and childhood cancer: results from the United Kingdom Cancer Study, Br. J. Cancer 88 (2003) 373381. [92] K.F. Trivers, A.C. Mertens, J.A. Ross, M. Steinbuch, A.F. Olshan, L.L. Robison, Parental marijuana use and risk of childhood acute myeloid leukaemia: a report from the Children's Cancer Group (United States and Canada), Paediatr. Perinat. Epidemiol. 20 (2) (2006) 110118. [93] W.Q. Wen, X.O. Shu, M. Steinbuch, R.K. Severson, G.H. Reaman, J.D. Buckley, L.L. Robison, Paternal military service and risk for childhood leukemia in offspring, Am. J. Epidemiol. 151 (3) (2000) 231240. [94] W.Q. Wen, X.O. Shu, J.D. Potter, R.K. Severson, J.D. Buckley, G.H. Reaman, L.L. Robison, Parental medication use and risk of childhood acute lymphoblastic leukemia, Cancer 95 (8) (2002) 17861794. [95] M. Hashibe, K. Straif, D.P. Tashkin, H. Morgenstern, S. Greenland, Z.F. Zhang, Epidemiologic review of marijuana use and cancer risk, Alcohol 35 (3) (2005) 265275. [96] L. Mucci, F. Granath, S. Cnattingius, Maternal smoking and childhood leukemia and lymphoma risk among 1,440,542 Swedish children, Cancer Epidemiol. Biomarkers Prev. 13 (9) (2004) 15281533. [97] J. Clavel, S. Bellec, S. Rebouissou, M. Loriot, Childhood leukemia, polymorphisms of metabolism enzyme genes and interactions with maternal smoking, coffee and alcohol consumption during pregnancy, Eur. J. Cancer Prev. 14 (2005) 531540. [98] P. Boffetta, Risk of childhood cancer after exposure to passive smoke, EHP 108 (1) (2000) 7382. [99] J. Rudant, F. Menegaux, G. Leverger, A. Baruchel, A. Lambilliotte, Y. Bertrand, C. Patte, H. Pacquement, C. Verite, A. Robert, G. Michel, G. Margueritte, V. Gandemer, D. Hemon, J. Clavel, Childhood hematopoietic malignancies and parental use of tobacco and alcohol: the ESCALE study (SFCE), Cancer Causes Control 19 (10) (2008) 12771290. [100] A.C. MacArthur, M.L. McBride, J.J. Spinelli, S. Tamaro, R.P. Gallagher, G. Theriault, Risk of childhood leukemia associated with parental smoking and alcohol consumption prior to conception and during pregnancy: the crossCanada childhood leukemia study, Cancer Causes Control 19 (3) (2008) 283 295. [101] B. Ji, X. Shu, M. Linet, W. Zheng, S. Wacholder, Y. Gao, D. Ying, Paternal cigarette smoking and the risk of childhood cancer among offspring of non-smoking mothers, JNCI 89 (1997) 238244. [102] T. Sorahan, R. Lancashire, M. Hultan, I. Peck, A. Stewart, Childhood cancer and parentla use of tobacco: deaths from 19531955, Br. J. Cancer 75 (1997) 134136. [103] K.M. Lee, M.H. Ward, S. Han, H.S. Ahn, H.J. Kang, H.S. Choi, H.Y. Shin, H.H. Koo, J.J. Seo, J.E. Choi, Y.O. Ahn, D. Kang, Paternal smoking, genetic polymorphisms in CYP1A1 and childhood leukemia risk, Leukemia Res. 33 (2) (2009) 250 258. [104] T Adam, S. Mitschke, T. Streibel, R. Baker, R. Zimmermann, Quantitative puff by puff resolved characterization of selected toxic compounds in cigarette mainstream smoke, Chem. Res. Toxicol. 19 (2006) 511520. [105] A. McNabola, B. Broderick, P. Johnston, L. Gill, Effects of the smoking ban on benzene and 1,3 butadiene levels in pubs in Dublin, J. Environ. Sci. Health, Part A 41 (2006) 799810. [106] L.A. Wallace, Major sources of benzene exposure, Environ. Health Perspect. 82 (1989) 165169. [107] IARC, A review of human carcinogens: lifestyle factors, Monographs on the evaluation of carcinogenic risks to humans, IARC 100(E) (2009). [108] C. Muirhead, Childhood leukemia in metropolitan regions in the United States: a possible relation to population density, Cancer Causes Controls 6 (1995) 383388. [109] D. Savitz, L. Feingold, Association of childhood cancer with residential traffic density, Scand. J. Work. Envior. Health 15 (1989) 360363. [110] R. Pearson, H. Wachtel, K. Ebi, Distance weighted traffic density in proximity to a home is a risk factor for leukemia and other childhood cancers, J. Air Waste Manage. 50 (2000) 175180. [111] G. Swaen, J. Slangen, Gasoline consumption and leukemia mortality and morbidity in 19 European countries: an ecological study, Int. Arch. Occup. Environ. Health 67 (1995) 8593. [112] F. Alexander, D. Leon, R. Cartwright, Isolation, car ownership and small area variation in incidence of acute lymphoblastic leukemia in children, Paeditric Perinatal Epidemiol. 10 (1996) 411417. [113] R Nordlinder, B. Jarvholm, Environmental exposure to gasoline and leukemia in children and young adults--an ecology study, Int. Arch. Occup. Environ. Health 70 (1997) 5760. [114] M. Feychting, D. Svensson, A. Ahlbom, Exposure to motor vehicle exhaust and childhood cancer, Scand. J. Work Environ. Health 24 (1) (1998) 811. [115] O. Raaschou-Nielsen, O. Hertel, B.L. Thomsen, J.H. Olsen, Air pollution from traffic at the residence of children with cancer, Am. J. Epidemiol. 153 (5) (2001) 433443. [116] P. Reynolds, J. Behren, R. Gunier, D. Goldberg, A. Hertz, D. Smith, Traffic patterns and childhood cancer incidence rates in California, United States, Cancer Causes Controls 13 (2002) 665673. [117] P. Reynolds, J. Behren, R. Gunier, D. Goldberg, A. Hertz, D. Smith, Childhood cancer incidence rates and hazardous air pollutants in California: an exploratory analysis, EHP 111 (4) (2003) 663668. [118] P. Reynolds, J. Von Behren, R.B. Gunier, D.E. Goldberg, A. Hertz, Residential exposure to traffic in California and childhood cancer, Epidemiology 15 (1) (2004) 612. [119] R. Harrison, P. Leung, L. Somerville, R. Smith, E. Gilman, Analysis of incidence of childhood cancer in the West Midlands of the United Kingdom in relation to proximity to main roads and petrol stations, Occup. Environ. Med. 56 (1999) 774780. [120] B. Langholz, K. Ebi, D. Thomas, J. Peters, S. London, Traffic density and the risk of childhood leukemia in a Los Angeles casecontrol study, Ann. Epidemiol. 12 (2002) 482487. [121] O. Visser, J.H. van Wijnen, F.E. van Leeuwen, Residential traffic density and cancer incidence in Amsterdam, 19891997, Cancer Causes Control 15 (4) (2004) 331339. [122] C. Steffen, M. Auclerc, A. Auvrignon, A. Baruchel, K. Kebaili, J. Clavel, Acute childhood leukemia and environmental exposure to potential sources of benzene and other hydrocarbons; a casecontrol study, Occup. Environ. Med. 61 (2004) 773778. [123] J. Von Behren, P. Reynolds, R.B. Gunier, R.P. Rull, A. Hertz, K.Y. Urayama, D. Kronish, P.A. Buffler, Residential traffic density and childhood leukemia risk, Cancer Epidemiol. Biomarkers Prev. 17 (9) (2008) 22982301. [124] P. Crosignani, A. Tittarelli, A. Borgini, F. Berrino, Childhood leukemia and road traffic: a population-based casecontrol study, Int. J. Cancer 108 (2004) 596599. [125] H.-H. Weng, S.-S. Tsai, C.-C. Chen, H.-F. Chiu, T.-N. Wu, C.-Y. Yang, Childhood leukemia development and correlation with traffic air pollution in Taiwan using nitrogen dioxide as an air pollutant marker, J. Toxicol. Environ. Health A 71 (7) (2008) 434438. [126] M. Linet, L. Ries, M. Smith, R. Tarone, S. Devesa, Cancer surveillance series: recent trends in childhood cancer incidence and mortaility in the United States, J. Natl. Cancer Inst. 91 (2) (1999) 10511058. [127] C. Hearey, U. Hans, A. Slegelaub, M. Ho, H. Salomon, R. Cella, Lack of an association between cancer incidence and residence near petrochemical industry in the San Fransisco Bay area, JNCI 64 (6) (1980) 12951299. [128] E. Knox, Leukemia clusters in childhood: geographical analysis in Britain, J. Epidemiol. Community Health 48 (1994) 369376. [129] J. Bithell, G. Draper, Apparent association between benzene and childhood leukemia: methodological doubts concerning a report by Knox, J. Epidemiol. Community Health 49 (1995) 437439. Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002 G Model CBI-6106; No. of Pages 14 14 ARTICLE IN PRESS D. Pyatt, S. Hays / Chemico-Biological Interactions xxx (2010) xxxxxx [130] P. Wilkinson, B. Thakrar, P. Walls, M. Landon, S. Falconer, C. Grundy, P. Elliot, Lymphohaematopoietic malignacy around all industrial complexes that include major oil refineries in Great Britain, Occup. Environ. Med. 56 (1999) 577580. [131] R. Lyons, S. Monaghan, M. Heaven, B. Littlepage, T. Vincent, G. Draper, Incidence of leukemia and lymphoma in young people in the vicinity of the petrochemical plant at Baglan Bay, South Wales, 19741991, Occup. Environ. Med. 52 (1995) 225228. [132] S. Sans, P. Elliot, I. Kleinschmidt, H. Dolk, Cancer incidence and mortality near the Baglan Bay petrochemcial works, South Wales, Occup. Environ. Med. 52 (217224) (1995). [133] P. Brosselin, J. Rudant, L. Orsi, G. Leverger, A. Baruchel, Y. Bertrand, B. Nelken, A. Robert, G. Michel, G. Margueritte, Y. Perel, F. Mechinaud, P. Bordigoni, D. Hemon, J. Clavel, Acute childhood leukaemia and residence next to petrol stations and automotive repair garages: the ESCALE study (SFCE), Occup. Environ. Med. 66 (9) (2009) 598606. [134] A.-K. Hurtig, M. Sebastian, Incidence of childhood leukemia and oil exploration in the Amazon Basin of Ecuador, Int. J. Occup. Environ. Health 10 (2004) 245250. [135] H.-H. Weng, S.-S. Tsai, H.-F. Chiu, T.-N. Wu, C.-Y. Yang, Association of childhood leukemia with residential exposure to petrochemical air pollution in Taiwan, Inhal. Toxicol. 20 (1) (2008) 3136. [136] H.-H. Weng, S.-S. Tsai, H.-F. Chiu, T.-N. Wu, C.-Y. Yang, Childhood leukemia and traffic air pollution in Taiwan: petrol station density as an indicator, J. Toxicol. Environ. Health A 72 (2) (2009) 8387. [137] K.W. Whitworth, E. Symanski, A.L. Coker, Childhood lymphohematopoietic cancer incidence and hazardous air pollutants in southeast Texas, 19952004, Environ. Health Perspect. 116 (11) (2008) 15761580. [138] E.G. Knox, E.A. Gilman, Hazard proximities of childhood cancers in Great Britain from 195380, J. Epidemiol. Community Health 51 (2) (1997) 151 159. [139] M. Stjernfeldt, K. Berglund, J. Lindsten, J. Ludvigsson, Maternal smoking and irradiation during pregancy as risk-factors for child leukemia, Cancer Detect. Prev. 16 (2) (1992) 129135. [140] R. Severson, J.B. Woods, Cigarette smoking and alcohol consumption by parents of children with acute myeloid leukemia: an analysis within morphologic subgroups: a report from the Children's Cancer Group, Cancer Epidemiol. Biomarkers Prev. 2 (1993) 433439. [141] C. Infante-Rivard, M. Krajinovic, D. Labuda, D. Sinnett, Parental smoking, CYP1A1 genetic polymorphisms and childhood leukemia (Quebec, Canada), Cancer Causes Control 11 (6) (2000) 547553. Please cite this article in press as: D. Pyatt, S. Hays, A review of the potential association between childhood leukemia and benzene, Chem. Biol. Interact. (2010), doi:10.1016/j.cbi.2010.01.002