Consistencies and inconsistencies underlying the quantitative assessment of leukemia risk from benzene exposure

Exposure to benzene at work and the risk of leukemia: a systematic review and meta-analysis

BACKGROUND

A substantial number of epidemiologic studies have provided estimates of the relation between exposure to benzene at work and the risk of leukemia, but the results have been heterogeneous. To bridge this gap in knowledge, we synthesized the existing epidemiologic evidence on the relation between occupational exposure to benzene and the risk of leukemia, including all types combined and the four main subgroups acute myeloid leukemia (AML), acute lymphocytic leukemia (ALL), chronic lymphocytic leukemia (CLL), and chronic myeloid leukemia (CML).

METHODS

A systematic literature review was carried out using two databases 'Medline' and 'Embase' from 1950 through to July 2009. We selected articles which provided information that can be used to estimate the relation between benzene exposure and cancer risk (effect size).

RESULTS

In total 15 studies were identified in the search, providing 16 effect estimates for the main analysis. The summary effect size for any leukemia from the fixed-effects model was 1.40 (95% CI, 1.23-1.57), but the study-specific estimates were strongly heterogeneous (I2 = 56.5%, Q stat = 34.47, p = 0.003). The random-effects model yielded a summary- effect size estimate of 1.72 (95% CI, 1.37-2.17). Effect estimates from 9 studies were based on cumulative exposures. In these studies the risk of leukemia increased with a dose-response pattern with a summary-effect estimate of 1.64 (95% CI, 1.13-2.39) for low (< 40 ppm-years), 1.90 (95% CI, 1.26-2.89) for medium (40-99.9 ppm-years), and 2.62 (95% CI, 1.57-4.39) for high exposure category (> 100 ppm-years). In a meta-regression, the trend was statistically significant (P = 0.015). Use of cumulative exposure eliminated heterogeneity. The risk of AML also increased from low (1.94, 95% CI, 0.95-3.95), medium (2.32, 95% CI, 0.91-5.94) to high exposure category (3.20, 95% CI, 1.09-9.45), but the trend was not statistically significant.

CONCLUSIONS

Our study provides consistent evidence that exposure to benzene at work increases the risk of leukemia with a dose-response pattern. There was some evidence of an increased risk of AML and CLL. The meta-analysis indicated a lack of association between benzene exposure and the risk of CML.

Cancer mortality in workers exposed to dieldrin and aldrin: over 50 years of follow up

OBJECTIVE

Dieldrin and aldrin, pesticides widely used until the 1970s, have been under suspicion of being carcinogenic. In this study, overall and cause-specific mortality was assessed in a cohort of 570 employees occupationally exposed to the pesticides dieldrin and aldrin to investigate the long-term health effects, in particular carcinogenic effects.

METHODS

All of the employees worked in the production plants between January 1954 and January 1970 and were followed for cause-specific mortality until 30 April 2006. Based on dieldrin levels in blood samples taken from 343 workers during the exposure period, the total intake of dieldrin was estimated for each individual subjects in the cohort. The estimated total intake ranged from 11 to 7,755 mg of dieldrin, with an average of 737 mg.

RESULTS

Two hundred and twenty-six workers had died before 30 April 2006 compared with an expected number of 327.3, giving a standardized mortality ratio (SMR) of 69.0 (95% confidence interval (CI): 60.3-78.7). Overall cancer mortality was also significantly lower than expected (SMR: 76.4, 95% CI: 60.8-94.9). Also, none of the specific cancer sites showed a significant excess mortality and no association between exposure level and cancer mortality was found.

CONCLUSION

The results from this study support findings from other epidemiological and recent animal studies concluding that dieldrin and aldrin are not likely human carcinogens.

A 56-year mortality follow-up of Texas petroleum refinery and chemical employees, 1948-2003

OBJECTIVE

To further investigate the mortality risk of employees who worked in the petroleum refinery industry, we updated an earlier investigation by extending the mortality follow-up by an additional 14 years through 2003.

METHODS

The cohort consisted of 10,621 employees with an average follow-up of 34 years. We used the standardized mortality ratio (SMR) adjusted for age, race, and calendar years as a measure of risk.

RESULTS

Overall mortality (SMR=0.77, 95% confidence interval [CI], 0.74-0.79), all cancer mortality (SMR=0.87, 95% CI=0.82-0.93), and most cause-specific mortalities for the total study population were lower than or similar to that of the population of Harris County, Texas. This study did not show a significant increase in leukemia in the total population or in any of the subgroups. The only statistically significant excess of mortality found in this study was an increase in mesothelioma among maintenance employees; the SMR was 4.78 (95% CI=2.54-8.17) among employees who worked for a minimum of one year and was 7.51 (95% CI=3.75-13.45) among those with 10 or more years of employment and 20 or more years of latency.

CONCLUSIONS

After more than half a century of follow-up, employees at this facility continue to show more favorable mortality outcomes than the general local population. Overall, no statistically significant increase of leukemia or of any of the specific cell types was found. The increased mesothelioma is likely related to past exposure to asbestos.

Modeling nonlinear dose-response relationships in epidemiologic studies: statistical approaches and practical challenges

Non-linear dose response relationships pose statistical challenges for their discovery. Even when an initial linear approximation is followed by other approaches, the results may be misleading and, possibly, preclude altogether the discovery of the nonlinear relationship under investigation. We review a variety of straightforward statistical approaches for detecting nonlinear relationships and discuss several factors that hinder their detection. Our specific context is that of epidemiologic studies of exposure-outcome associations and we focus on threshold and J-effect dose response relationships. The examples presented reveal that no single approach is universally appropriate; rather, these (and possibly other) nonlinearities require for their discovery a variety of both graphical and numeric techniques.

Review of the literature on benzene exposure and leukemia subtypes

The epidemiologic literature on benzene exposure and leukemia in the MEDLINE and TOXNET databases was examined through October 2004 using the keywords "benzene", "leukemia" and "adverse health effects". This search was complemented by reviewing the reference lists from extant literature reviews and criteria documents on benzene. Published studies were characterized according to the type of industry studied and design, exposure assessment, disease classification, and control for confounding variables. Study design consisted of either cohort studies or case-control studies, which were further categorized into population-based and nested case-control studies. Disease classification considered the source of diagnostic information, whether there was clinical confirmation from medical records or histopathological, morphological and/or cytogenetic reviews, and as to whether the International Classification of Diseases (ICD) or the French-American-British (FAB) schemes were used (no studies used the Revised European-American Lymphoma (REAL) classification scheme). Nine cohort and 13 case-control studies met inclusion criteria for this review. High and significant acute myeloid leukemia risks with positive dose response relationships were identified across study designs, particularly in the "well-conducted" cohort studies and especially in more highly exposed workers in rubber, shoe, and paint industries. Risks for chronic lymphocytic leukemia (CLL) tended to show elevations in nested case-control studies, with possible dose response relationships in at least two of the three studies. However, cohort studies on CLL show no such risks. Data for chronic myeloid leukemia and acute lymphocytic leukemia are sparse and inconclusive.

Health Watch exposure estimates: do they underestimate benzene exposure?

A nested case-control study found that the excess of leukemia, identified among the male members of the Health Watch cohort, was associated with benzene exposure. Exposure had been retrospectively estimated for each individual occupational history using an algorithm in a relational database. Benzene exposure measurements, supplied by Australian petroleum companies, were used to estimate exposure for specific tasks. The tasks carried out within each job, the products handled, and the technology used, were identified from structured interviews with contemporary colleagues. More than half of the subjects started work after 1965 and had an average exposure period of 20 years. Exposure was low; nearly 85% of the cumulative exposure estimates were at or below 10 ppm-years. Matched analyses showed that leukemia risk increased with increasing cumulative benzene exposures and with increasing exposure intensity of the highest-exposed job. Non-Hodgkin lymphoma and multiple myeloma were not associated with benzene exposure. A reanalysis reported here, showed that for the 7 leukemia case-sets with greater than 16 ppm-years cumulative exposure, the odds ratio was 51.9 (5.6-477) when compared to the 2 lowest exposed categories combined to form a new reference category. The addition of occasional high exposures, e.g. as a result of spillages, increased exposure for 25% of subjects but for most, the increase was less than 5% of total exposure. The addition of these exposures reduced the odds ratios. Cumulative exposures did not range as high as those in comparable studies; however, the recent nature of the cohort and local handling practices can explain these differences.

Effect of follow-up time on risk estimates: a longitudinal examination of the relative risks of leukemia and multiple myeloma in a rubber hydrochloride cohort

BACKGROUND

Choice of follow-up time for an occupational cohort can influence risk estimates. We examined the effects of follow-up time on relative risk estimates for leukemia and multiple myeloma in a cohort of 1,845 rubber hydrochloride workers.

MATERIALS AND METHODS

We generated standardized mortality ratios (SMRs) for yearly follow-ups, beginning each study in 1940 and increasing study end dates from 1950 through 1996. We used Cox proportional hazards modeling to explore the effects of follow-up time on the exposure-response relationship.

RESULTS

The SMR for leukemia rose to 13.55 in 1961 and fell nearly monotonically to 2.47 by 1996. Cox modeling suggested interaction between cumulative exposure and time since exposure. A longer time to peak risk was seen for multiple myeloma.

CONCLUSIONS

Because summary risk estimates change with follow-up time, exposure limits set using these estimates may not adequately protect workers. Consideration of appropriate follow-up time and use of more complex temporal models are critical to the risk assessment process.

The contribution of benzene to smoking-induced leukemia

Cigarette smoking is associated with an increased risk of leukemia; benzene, an established leukemogen, is present in cigarette smoke. By combining epidemiologic data on the health effects of smoking with risk assessment techniques for low-dose extrapolation, we assessed the proportion of smoking-induced total leukemia and acute myeloid leukemia (AML) attributable to the benzene in cigarette smoke. We fit both linear and quadratic models to data from two benzene-exposed occupational cohorts to estimate the leukemogenic potency of benzene. Using multiple-decrement life tables, we calculated lifetime risks of total leukemia and AML deaths for never, light, and heavy smokers. We repeated these calculations, removing the effect of benzene in cigarettes based on the estimated potencies. From these life tables we determined smoking-attributable risks and benzene-attributable risks. The ratio of the latter to the former constitutes the proportion of smoking-induced cases attributable to benzene. Based on linear potency models, the benzene in cigarette smoke contributed from 8 to 48% of smoking-induced total leukemia deaths [95% upper confidence limit (UCL), 20-66%], and from 12 to 58% of smoking-induced AML deaths (95% UCL, 19-121%). The inclusion of a quadratic term yielded results that were comparable; however, potency models with only quadratic terms resulted in much lower attributable fractions--all < 1%. Thus, benzene is estimated to be responsible for approximately one-tenth to one-half of smoking-induced total leukemia mortality and up to three-fifths of smoking-related AML mortality. In contrast to theoretical arguments that linear models substantially overestimate low-dose risk, linear extrapolations from empirical data over a dose range of 10- to 100-fold resulted in plausible predictions.

A case-control study to investigate the risk of leukaemia associated with exposure to benzene in petroleum marketing and distribution workers in the United Kingdom

OBJECTIVES

To investigate the risk of leukaemia in workers in the petroleum distribution industry who were exposed to low levels of benzene.

METHODS

From the cohort of distribution workers, 91 cases were identified as having leukaemia on either a death certificate or on cancer registration. These cases were compared with controls (four per case) randomly selected from the cohort, who were from the same company as the respective case, matched for age, and alive and under follow up at the time of case occurrence. Work histories were collected for the cases and controls, together with information about the terminals at which they had worked, fuel compositions, and occupational hygiene measurements of benzene. These data were used to derive quantitative estimates of personal exposure to benzene. Odds ratios (OR) were calculated conditional on the matching, to identify those variables in the study which were associated with risk of leukaemia. Examination of the potential effects of confounding and other variables was carried out with conditional logistic regression. Analyses were carried out for all leukaemia and separately for acute lymphoblastic, chronic lymphocytic, acute myeloid and monocytic, and chronic myeloid leukaemias.

RESULTS

There was no significant increase in the overall risk of all leukaemias with higher cumulative exposure to benzene or with intensity of exposure, but risk was consistently doubled in subjects employed in the industry for > 10 years. Acute lymphoblastic leukaemia tended to occur in workers employed after 1950, who started work after the age of 30, worked for a short duration, and experienced low cumulative exposure with few peaks. The ORs did not increase with increasing cumulative exposure. The risk of chronic lymphocytic leukaemia seemed to be related most closely to duration of employment and the highest risk occurred in white collar workers with long service. These workers had only background levels of benzene exposure. There was no evidence of an association of risk with any exposure variables, and no evidence of an increasing risk with increasing cumulative exposure, mean intensity, or maximum intensity of exposure. The patterns of risk for acute myeloid and monocytic leukaemia were different from those of the lymphoid subgroups, in which duration of employment was the variable most closely related to risk. Risk was increased to an OR of 2.8 (95% confidence interval (95% CI) 0.8 to 9.4) for a cumulative exposure between 4.5 and 45 ppm-years compared with < 0.45 ppm-years. For mean intensity between 0.2 and 0.4 ppm an OR of 2.8 (95% CI 0.9 to 8.5) was found compared with < 0.02 ppm. Risk did not increase with cumulative exposure, maximum intensity, or mean intensity of exposure when treated as continuous variables. Cases of acute myeloid and monocytic leukaemia were more often classified as having peaked exposures than controls, and when variables characterising peaks, particularly daily and weekly peaks, were included in the analysis these tended to dominate the other exposure variables. However, because of the small numbers it is not possible to distinguish the relative influence of peaked and unpeaked exposures on risk of acute myeloid and monocytic leukaemia. There was no evidence of an increased risk of chronic myeloid leukaemia with increases in cumulative exposure, maximum intensity, mean intensity, and duration of employment, either as continuous or categorical variables. Analyses exploring the sensitivity of the results to the source and quality of the work histories showed similar patterns in general. However, no increases in ORs for categories of cumulative exposure were found for acute myeloid and monocytic leukaemia in the data set which included work histories obtained from personnel records still in existence, although numbers were reduced. Analyses excluding the last five and 10 years of exposure showed a tendency for ORs to reduce for chronic lymphocytic leukaemia and chronic myeloid leukaemia, and to increase for acute myeloid and monocytic leukaemia. Limitations of the study include uncertainties and gaps in the information collected, and small numbers in subcategories of exposure which can lead to wide CIs around the risk estimates and poor fit of the mathematical models.

CONCLUSIONS

There is no evidence in this study of an association between exposure to benzene and lymphoid leukaemia, either acute or chronic. There is some suggestion of a relation between exposure to benzene and myeloid leukaemia, in particular for acute myeloid and monocytic leukaemia. Peaked exposures seemed to be experienced for this disease. However, in view of the limitations of the study, doubt remains as to whether the risk of acute myeloid and monocytic leukaemia is increased by cumulative exposures of < 45 ppm-years. Further work is recommended to review the work histories and redefine their quality, to explore the discrepancies between results for categorical and continuous variables, and to develop ranges around the expose estimates to enable further sensitivity analyses to be carried out.

Does benzene cause multiple myeloma? An analysis of the published case-control literature

Two case series and two epidemiological studies in the 1970s and 1980s suggested that benzene exposure might be a risk factor for multiple myeloma. An analysis has now been conducted of the published population-based and hospital-based case-control studies published through mid-1995 that permit examination of the relationship between multiple myeloma and benzene exposure or surrogates for benzene exposure. No increased association was found between multiple myeloma and benzene exposure or exposure to chemical groups that included benzene. The odds ratios from these analyses approximated 1.0. Exposures to petroleum products and employment in petroleum-related occupations did not appear to be risk factors for multiple myeloma. Cigarette smoking, as a surrogate of benzene exposure, was not found to be associated with myeloma, while some studies of products of combustion described as "engine exhaust" did show a significant association with multiple myeloma. In toto, the population-based and hospital-based case-control literature indicated that benzene exposure was not a likely causal factor for multiple myeloma.

Hairy cell leukaemia and occupational exposure to benzene

OBJECTIVES

The role of occupational exposures in hairy cell leukaemia (HCL) was investigated through a multicentre, hospital based, case-control study. This paper analyses the role of exposure to benzene in HCL.

METHODS

A population of 226 male cases of HCL and 425 matched controls were included in the study. Benzene exposure was evaluated by expert review of the detailed data on occupational exposures generated by case-control interviews.

RESULTS

No association was found between HCL and employment in a job exposed to benzene (odds ratio (OR) 0.9 (95% confidence interval (95% CI) 0.6-1.3)). The sample included 125 subjects, 34 cases (15%), and 91 controls (21%) who had been exposed to benzene, as individually assessed by the experts, for at least one hour a month during one of their jobs. Benzene exposure was not associated with a risk of HCL (OR 0.8 (0.5-1.2)). No trend towards an increase in OR was detected for increasing exposures, the percentage of work time involving exposure to > 1 ppm, or the duration of exposure. No findings suggested a particular risk period, when the OR associated with the time since first or last exposure, or since the end of exposure, were examined.

CONCLUSIONS

In conclusion, with the low exposures prevalent in the sample, the study did not show any association between benzene exposure and HCL.

Associations between disease and occupation: hypotheses generated from the National Mortality Followback Survey

This study uses the National Mortality Followback Survey of 1986 to identify the top five Sentinel Health Events Occupational [SHE(O)s], the five leading causes of death, and to ascertain the primary occupations and industries associated with these. We found that, as expected, cardiovascular diseases were four of the five leading causes of death overall. In addition, the SHE(O) responsible for most deaths was cancer of the trachea, bronchus, and lung, followed by renal failure, bladder cancer, myeloid leukemia, and liver cancer. We employed proportionate mortality ratios to analyze the relationship between industry and occupation and category of mortality. In brief, we validated findings by other researchers; for example, farmers were at lower risk of cancer of the trachea, bronchus, and lung, and workers in eating/drinking places had excess risk of liver cancer. We also hypothesize other relationships, such as between motor vehicle dealers and bladder cancer.

On cancer risk estimation of urban air pollution

The usefulness of data from various sources for a cancer risk estimation of urban air pollution is discussed. Considering the irreversibility of initiations, a multiplicative model is preferred for solid tumors. As has been concluded for exposure to ionizing radiation, the multiplicative model, in comparison with the additive model, predicts a relatively larger number of cases at high ages, with enhanced underestimation of risks by short follow-up times in disease-epidemiological studies. For related reasons, the extrapolation of risk from animal tests on the basis of daily absorbed dose per kilogram body weight or per square meter surface area without considering differences in life span may lead to an underestimation, and agreements with epidemiologically determined values may be fortuitous. Considering these possibilities, the most likely lifetime risks of cancer death at the average exposure levels in Sweden were estimated for certain pollution fractions or indicator compounds in urban air. The risks amount to approximately 50 deaths per 100,000 for inhaled particulate organic material (POM), with a contribution from ingested POM about three times larger, and alkenes, and butadiene cause 20 deaths, respectively, per 100,000 individuals. Also, benzene and formaldehyde are expected to be associated with considerable risk increments. Comparative potency methods were applied for POM and alkenes. Due to incompleteness of the list of compounds considered and the uncertainties of the above estimates, the total risk calculation from urban air has not been attempted here.

Benzene toxicity and risk assessment, 1972-1992: implications for future regulation

Acute and chronic exposure to benzene vapors poses a number of health hazards to humans. To evaluate the probability that a specific degree of exposure will produce an adverse effect, risk assessment methods must be used. This paper reviews much of the published information and evaluates the various risk assessments for benzene that have been conducted over the past 20 years. There is sufficient evidence that chronic exposure to relatively high concentrations of benzene can produce an increased incidence of acute myelogenous leukemia (AML). Some studies have indicated that benzene may cause other leukemias, but due to the inconsistency of results, the evidence is not conclusive. To predict the leukemogenic risk for humans exposed to much lower doses of benzene than those observed in most epidemiology studies, a model must be used. Although several models could yield plausible results, to date most risk assessments have used the linear-quadratic or conditional logistic models. These appear to be the most appropriate ones for providing the cancer risk for airborne concentrations of 1 ppb to 10 ppm, the range most often observed in the community and workplace. Of the seven major epidemiology studies that have been conducted, there is a consensus that the Pliofilm cohort (rubber workers) is the best one for estimating the cancer potency because it is the only one with good exposure and incidence of disease data. The current EPA, OSHA, and ACGIH cancer potency estimates for benzene are based largely on this cohort. A retrospective exposure assessment and an analysis of the incidence of disease in these workers were completed in 1991. All of these issues are discussed and the implications evaluated in this paper. The range of benzene exposures to which Americans are commonly exposed and the current regulatory criteria are also presented.

Determinants of lung cancer risk among cadmium-exposed workers

Workers at a cadmium recovery plant in Globe, Colorado, showed an increased risk of lung cancer, which some investigators have attributed to cadmium exposure. We conducted a cohort mortality analysis of this work force and a case-control analysis of the lung cancer cases within this work force in order to assess the probable causes of the lung cancer excess. The Globe plant began as a lead smelter about 1886, switched to arsenic production in 1920, and became a cadmium metal production facility in 1926. Cadmium, arsenic, and cigarette smoking are three potential lung carcinogens found in this workplace. Industrial hygiene data collected from 1943 onward served as the basis for the National Institute for Occupational Safety and Health (NIOSH)-derived exposure algorithm that assigned cadmium exposure estimates to employees based on their work area in the plant and calendar time. Few exposure data existed for substances other than cadmium. Feedstock ore concentrations were used as a surrogate measure of air arsenic levels. The arsenic content of the fines used as feedstock prior to 1940 was considerably higher than that of the fines used after 1940. Smoking histories had been obtained previously for 45% of the workers. A case-control analysis of the 25 cases of lung cancer known to have occurred among these workers through 1982 was conducted using three controls per case, matched by closest data of hire and age at hire. Potential causal agents for lung cancer included cadmium exposure, cigarette smoking, and arsenic exposure. Exposure variables for each case and control included estimated cumulative cadmium exposure in milligram-years per cubic meter, cigarette smoking history, and plant arsenic exposure status at the time of hire. Estimated cumulative cadmium exposures of cases and controls did not differ overall or within the date-of-hire strata. Cases were more than eight times more likely to have been cigarette smokers than were controls. Lung cancer risk in this workplace was more closely related to the period of hire, not to the cumulative cadmium exposure. The period of hire appears to be a surrogate for arsenic exposure as related to feedstock. The measures used here seem to indicate that exposure to arsenic and cigarette particulates, rather than to cadmium particulates, may have caused the increased rate of lung cancer of these workers.

Public health versus public policy? An appraisal of British urban transport policy

The most visible aspect of the relationship between transport and health is in the realm of road traffic accidents (RTAs). But the effects of transport policy upon the public health are much wider-ranging. They include exacerbation of social and health inequalities, separation from directly health-protective amenities and social isolation, as well as health damage by physico-chemical pollution. It is arguable that these factors may account for more years of life lost annually than do RTAs. The current focus on accident reduction as the sole means of avoiding adverse health effects of transport distracts attention from the wide health erosive effects of road traffic. Universal mobility and accessibility, independent of the car, is required for public health protection and advancement, since private transport carries many external, and poorly recognised health costs.