Document 0Xx8EbDk1D0ZZ6QvKMao0Jwb

REGULATORY TOXICOLOGY AND PHARMACOLOGY 21, 307-321 (1995) Cell- Type-Specific Leukemia Analyses in a Combined Cohort of More Than 208, 000 Petroleum Workers in the United States and the United Kingdom , 1937- 1989 OTTO WONG* AND GERHARD K. RAABEt Applied Health ~cience8, Inc.. 181 Second Avenue, Suite 628 O. Bo% 2078. San Mateo. California 94401; and Medical Department. Mobil Oil Corporation. P. O. Bo% 1038. Princeton. New Jersey 08543 Received September 10. 1994 A large number of epidemiologic studies of workers in the petroleum industry have been conducted to investigate the carcinogenic and other health effects of exposure to petroleum products during manufacture and distribution. Of particular interest is the relation- ship between exposure to benzene or benzene-contain- ing liquids in the petroleum industry and leukemia risk. However, few studies have investigated cell- type-specHic leukemia risk. In the present investigation, all co- hort studies of petroleum workers in the United States and the United Kingdom were combined into a single database for cell-type-specific leukemia analysis. The majority of these workers' were petroleum refinery em- ployees. but production, pipeline, and distribution workers in the petroleum industry were also included. The combined cohort consisted of 208,741 petroleum workers. Between 1937 and 1989,. these workers contributed a total of 4,665,361 person-years of observation. More than 56 thousand deaths were reported among these workers during the 53 years of observation. Cell-type-specific leukemia risks were calculated using a meta-analysis procedure appropriate for combining occupational cohort studies. These risks were expressed in terms of cell-type-specific leukemia stan- dardized mortality ratios (meta-SMRs). The meta-SMR for acute myelogenous leukemia was 0.98. The lack of an increase of acute myelogenous leukemia was attrib- uted to the low levels of benzene exposure in the petro- leum industry. particularly in comparison to benzene exposure levels in some previous studies of workers in other industries, who had been found to experience increased risk of acute myelogenous leukemia. Similarly, no increase in chronic myelogenous, acute lymphocytic, or chronic lymphocytic leukemias was found in petroleum workers (meta- SMRs of 0.89, 1.16, and 0. 84, respectively). Meta-analyses restricted to refinery studies or to studies with at least 15 years of follow-up yielded similar results. The findings of the present investigation are consistent with those from several re- cent case-control studies. ~ 19911 Academic Pr_. lac. INTRODUCfION Workers in the petroleum industry are potentially ex- posed to hydrocarbons through inhalation of vapors from as well as dermal contact of crude oil, feed stocks, intermediate products during refining, and end prod- ucts, such as gasoline (Runion, 1988; Raabe, 1993). Of special concern is benzene, which has been linked to an increased risk of leukemia, particularly acute myelogenous leukemia (AML). The first case report suggesting an association between exposure to benzene and leuke- mia appeared in 1928 (Delore and Borgomano, 1928). In the following years, additional case reports were pub- lished (Hunter, 1939; Vigliani and Saita, 1964; Brow- ning, 1965; Gouguel et aL 1967; Girard and Revol, 1970; Aksoy et aL 1972). Most case reports were from En- gland, Italy, France, and Turkey. Documentation, if any at all, of exposures to benzene or benzene-containing mixtures in these case reports was extremely poor. Fur- thermore, these reports lacked information concerning the size of population at risk, and, therefore, no risk estimates could be derived. Thus, case reports by themselves do not provide definitive evidence for causation but rather suggest potential areas for future epidemio- logic research. One obser:vation, however, emerged from these case reports: The predominant cell type of the re- ported leukemia cases was AML. It was not until the 1970s that epidemiologic studies of individuals exposed to benzene were carried out. J udging by present day standards, some of the early studies were rather crude. For example, in the Thorpe (1974) study of leukemia incidence among employees at eight Esso petroleum affiliates in Europe, only active employ- ees and annuitants, but" not separated employees, wer~. included. Furthermore , person-years of observation were estimated and not actually calculated. Similarly, in the Infante et aL (1977) study of rubber hydrochloride workers exposed to benzene at two Plio film manufactur- ing plants in Ohio, the cohort was incomplete (an unknown number of employment records were missing and no maintenance workers were included). Furthermore 307 0273- 2:100/9.'; $6. Copyright i9 1995 by Academic Press. Inc. All ri~hl' of reproduction in any form reserved. GARABRANT/BERRYMAN 00379 : 3 ()g' WONG A:-iD RAABE mortality ascertainment in the Infante et al. (1977) Both in t'ivo and in t'itro studies have indicated that the study was only 75CO complete. Likewise, employment committed stem (precursor or progenitor) cells are capa- histories were also incomplete. It appeared that some in- ble of self-renewal and amplification. Previously a num- dividuals included in the study had worked at the facili- ber of hypotheses have been proposed regarding the pos- ties for only a few days, and some might not have worked sible role of stem cells in benzene leukemogenesis. How- there at all (Wong et al., 1983), Subsequent updates of ever, in most of these hypotheses little or no distinction the Infante et al. (19ii) study by others (Rinsky et ai., was made between the pluripotential and the committed 1981. 1987) have improved mortality ascertainment, but stem cells. Recent research has indicated that the com- the cohort incompleteness remained a major limitation mitted stem cells or progenitor cells play an important of the study. part not only in the maintenance of normal, steady-state In a 1983 report. Wong discussed some of the unre- hemopoiesis but also in the etiology of cell-type-specific solved issues regarding the relationship between ben- leukemias. zene and leukemia. The sarpe issues were iterated by a Based on an investigation of immunoglobulin gene or- group of scientists convened by the International ganization and expression in hemopoietic stem cell leu- Agency for Research on Cancer (McMichael, 1988). kemia, Ford et al. (1983) have demonstrated that aclonal One of the major issues raised by both Wong (1983) event leading to blast crisis can occur in a committed B and :'vIcMichael' (1988) is the relationship, if any, cell precursor rather than in the pluripotential stem cell between benzene exposure and leukemia cell types other precursor. Similarly, based on human DNA analyses, than AML. Maher et aL (1993) concluded that CML and CLL do not In the past, leukemia was considered a single statisti- arise from a common progenitor and that these malig- cal category in most occupational epidemiologic studies, nancies arise from separate stem cells. Thus, recent ad- partly because of the historical nomenclature, unavail- vances in these areas confirm the epidemiologic obser- ability of cell-type-specific rates for comparison and, vation that "leukemia" is a group of distinct malignan- most importantly, the paucity of cases by cell-type in in- cies, which should be analyzed separately- dividual studies (Wong, 1987a,b). Recently epidemio- The need to analyze leukemia data by cell type in re- logic studies have demonstrated the importance of cell- lation to occupational exposures such as benzene has type-specific analysis in studying leukemia. It has now created some interesting challenges to epidemiologists. been recognized that the diseases collectively known as Workers in the petroleum industry represent one of the leukemia are several distinct malignancies with different largest populations exposed to benzene. Numerous stud- etiologic factors. Linet (1985) began her book on leuke- ies based on petroleum workers have been conducted in mia with the following: " Earlier epidemiologic reviews the United States and in the United Kingdom. Even considered leukemia as a single disease with some minor though most of these studies consist of several thousand differences in risk factors for each of the histopathologic workers who have been observed over several decades, types. In the last two decades, epidemiologic studies few individual studies offer adequate data for cell-type- have provided evidence that diseases collectively identi- specific analysis. The objective of the present investiga- fied as leukemia are several distinct malignancies char- tion is to combine all the studies of petroleum workers acterized by varying age, sex, race, and ethnic group pat- in the United States and in the United Kingdom into terns, dissimilar secular trends, and different etiologic a single large database, which can then be analyzed to factors. " Similarly, the diversity of different types of leu- evaluate cell- type-specific leukemia risks in these petro- kemia has long been recognized by hematologists, as in- leum workers. dicated, for example, in the widely used textbook of he- matology by Wintrobe et aL (1981): "Few authorities MA TERlALS AND METHODS would disagree concerning the recognition of chronic myeloid leukemia (CML), acute myeloid leukemia Based on a consideration of both study design and (AML), polycythemia vera (PV), idiopathic myelofi- data quality, only cohort studies were included in the brosis (lMF). chronic lymphocytic leukemia (CLL), analysis. Studies based on proportional mortality ratios acute lymphoblastic leukemia (ALL), Hodgkin s disease (PMRs) were excluded from analysis because of the fol- (HD), the non- Hodgkin' s lymphomas (NHL), multiple lowing limitations. In addition to the well-known meth- myeloma (MM), and macroglobulin(!mia as distinct en- odological deficiencies of PMR studies (Wong and De- tities- " coufie, 1982; WongetaL, 1985), some of the PMRstudies The question of whether all leukemia cell types can of petroleum workers also suffered from incomplete as- be affected by a common etiologic agent has also been certainment of deaths (Reeve et aL, 1982; Thomas aL. examined at the cellular level, particularly with respect 1980, 1982; Wong and Tabershaw, 1980). Furthermore, to hemopoietic stem cells. There are two types of stem employment histories were not available in these PMR cells: the pluripotential stem cells, which are responsible studies. and analyses were limited. Finally, the petro- for the production of circulating elements, and their di- leum refineries included in these PMR studies have sub- rect descendants, the committed precursor cells which sequently been studied more thoroughly with the cohort are programmed to follow a single line of differentiation. study design. GARABRANT/BERRYMAN 00380 CELL- TYPE- SPECIFIC LEUKEMIA IN . PETROLEUM WORKERS TABLE 1 Description of Epidemiologic Study Cohorts in the Petroleum Industry in the United States and the United Kingdom 309 Organization Amoco Chevron Chevron (Gulf) Exxon Mobil Cohort location 10 refineries in US El Segundo (CA) refinery Richmond (CA) refinery Port Arthur (TX) refinery Baton Rouge (LA) refinery Baytown (TX) refinery Bayway (NJ) refinery Beaumont (TX) refinery Paulsboro (NJ) refinery Torrance (CA) refinery Number of workers 10, 763 773 523 844 894 722 860 119 855 991 Shell 2 California refineries Deer Park (TX) refinery Woodriver (IL) refinery 585 8.' 796 Texaco 13 refineries in US Production & pipeline in US American Petroleum Institute Institute of Petroleum Total Land-based terminals in US Marine vessels in US 8 refineries in UK Distribution centers in UK US and UK petroleum workers 19,077 098 026 109 . 34 569 23,306 208,741 Observation period 1970- 1986 1950- 1986 1950- 1986 1937- 1987 1970-1982 1970- 1982 1970- 1982 1945- 1987 1946- 1987 1959- 1987 1973- 1989 1948- 1983 1940- 1989 1947- 1977 1946- 1980 1946- 1989 1946-1989 1951- 1989 1951- 1989 1937- 1989 Person-years 125, 241 110, 594 205 397 526 386 785 80 ,684 59,257 166, 427 120 718 33, 331 ;;7 657 181, 782 300,991 358 318 220 414 239 125 227 143 931 640 626,471 665 361 Total deaths References 405 121 038 799 000 374 826 294 681 408 051 180 627 024 886 066 695 10, 193 743 411 Homstra, 1990; Nelson, 1985 DaggetaL, 1992; WongetaL 1986 Satin et aL 1994; Wen et aL 1983 Shallenberger et aL 1992; Hanis et cU. 1982 , 1985a Raabe et aL 1994; Milcarek et aL 1994; Collingwood cU. 1994; Morgan and Wong, 1984, 1985a; Enterline and Henderson 1985 Tsai et aL 1993; Marsh et aL 1991; Honda et aL 1995; Wongsrichanalai et al., 1989; McCraw et aL 1985; Joyner, 1983 Divine et aL 1985 Divine and Barron, 1986 1987 WongetaL I993;Wongand Trent, 1993 Rushton, 1993a c; Rushton and Alderson, 1981, 1983 In addition to cohort and PMR studies, there were also a number of community-based case-control studies of cell-type-specific leukemias, which included petroleum or petroleum-related occupations in the analyses. Most of these studies did not provide adequate infor- mation about the nature of exposure or details of employment. Methodological problems (such as control selection, recall of employment histories, adjustment of confounding factors) further complicated the interpretation. Therefore, case-control studies were not in- cluded in the quantitative analysis, but formed part of the basis of discussion. A couple of reports based on linkage between tumor registry and census data have also been published (Olsen and Jensen , 1987; Jakobsson et aL, 1993). The Olsen and Jensen (1987) survey reported a deficit of acute leukemia in men employed at gasoline stations in Denmark, whereas the Jakobsson et al. (1993) survey reported an increased risk of acute myeloid leukemia in petrol sta- tion attendants in Finland. Both surveys were judged to be inadequate. Using occupational information derived from census data as a surrogate for exposure to specific chemicals can introduce substantial misclassification. Moreover, estimating person-years at risk or expected leukemia cases in certain industries based on census data involves a potentially substantial amount of inac- curacy. The problem of multiple comparisons based on the large number of cancer sites and industries or occu- pations examined further weakens the statistical find- ings from these surveys. As such, these reports are inad- equate in assessing leukemia risk due to exposure to specific chemicals. In the present investigation, a meta-analysis procedure was used in combining data from individual cohort studies of petroleum workers. The methodological details of the meta-analysis have been described elsewhere (Wong and Raabe, 1989). Data needed to carry out the meta-analysis included observed leukemia deaths by cell type and the distribution of age-specific person-years from each individual study. These data were usually not presented in published reports. Instead, such data were requested from the original investigators. For some studies, multiple reports on various portions of the cohort and/or updates were available. For the present analysis, data for the entire cohort based on the latest available. update were used. A specific analysis was carried out for each of the four major leukemia cell types (AML, CML, ALL, and CLL). The 8th Revision of the International Classification of Diseases was used in classifying deaths. The codes for AML, CML, ALL, and CLL are 204.0, 204. , 205. , and 205. 1, respectively. For comparison in the US studies, GARABRANT/BERRYMAN 00381 (%) 310 WONG AND RAABE TABLE 2 Comparison of Leukemia Cell-Type Distribution between Petroleum Workers and the General Population Petroleum workers; Cell type US and UK % General population Acute myelogenous leukemia Chronic myelogenous leukemia Acute lymphocytic leukemia Chronic lymphocytic leukemia Total 103 (46%) 43 (19%) 25 (11 %) 55 (24%) 226 (100%) 45 (45%) 1909%) 9(9%) 28 (28%) 101 000%) 148 (46%) 62 (19%) . 34 (10%) 83 (25%) 327 (100%) 100 Note. Goodness-of- fit test between the US-UK comhined cohort of petroleum workers and the general population , xL.- = 2.00 (P = 0.57). age-specific mortality rates for each of the four major leukemia cell types were derived from data provided by the National Center for Health Statistics (Selvin et al. 1983). Although in general the accuracy of diagnosis of leukemia cell types based on death certificates was not as desirable as that based on pathological reports (par- ticularly in the 1940s or 1950s), no potential bias was introduced in the present investigation, since diagnostic information in both petroleum workers and the compar- ison group (general population) was based on death cer- tificates. Furthermore, the majority of leukemia deaths in petroleum workers occurred in or after the 19605. For each US cohort, expected deaths by cell type were calculated by applying US specific rates to person-years. or the UK industrywide studies, cell-type-specific !lformation has been published, and such information was taken directly from the publications (Rushton, 1993a c). One Canadian cohort study (Schnatter et aI. 1992) and one Australian prospective survey (Christie et al. 1991; Peach et aL 1992) were not included in the analysis because cell-type-specific leukemia rates for Canada or Australia were not available. The statistical procedure for meta-analysis consists of summing up the observed and expected deaths for a spe- cific leukemia cell type from individual studies, and cal- culating the summary or meta-SMR. This simple proce- dure treats each individual cohort as a separate meta- stratum in data summarization, thus adjusting for individual studies while preserving the original adjust- ment using substrata specific to age, sex, race, and time period. In the present investigation, both cell-type-spe- cific leukemia meta-SMRs and their corresponding 95% confidence intervals (95% CI) were calculated. Although other statistical procedures for meta-analysis are avail- able, the meta-SMR procedure was used in this investi- gation because of its simplicity in computation and in- terpretation, its preservation of the adjustments in the individual studies, and its resemblance to the analyses in the original studies. Similar procedures have been used previously in analyzing data on asbestos and gastroin- estinal cancer (Morgan et aL 1985), artificial sweeten- s and bladder cancer (Morgan and Wong, 1985b), fOrmaldehyde and respiratory cancer (Nelson et aI. 1986), chemical dyes and bladder cancer (Matanoski and Elliott, 1981), and radiation and colon cancer (Day, 1985). In particular, this method was discussed in the interpretation of negative epidemiological evidence for carcinogenicity by the International Agency for Research on Cancer (Wald and Doll, 1985). An additional benefit of meta-analysis is that it takes into consideration the multiple-comparison problem ex- hibited in the individual cohort studies. As discussed be- low, a total of 19 individual cohorts have been included in the database, which means that there would be 76 (19 X 4) cell-type-specific SMRs. Approximately 4 statistically significant SMRs can be expected to occur simply by chance alone at the a = 5% significance level. By com- bining data from these 19 cohorts, excesses or deficits due to chance for a particular cell type are less likely to occur. For further discussion on the meta-analysis of cohort studies, the readers are referred to Wong and Raabe (1989, 1990). RESULTS Included in the analysis were studies conducted or sponsored by seven major petroleum companies in the United States. Most of these studies consisted of refinery workers (Hornstra, 1990; Nelson, 1985; Dagg et aI. CombIned (U.s :i-i" CamblneclIU.K.) COmbIned (Us.&UK.) : Jj _-"'___0 FIG. 1. Acute myelogenous leukemia among petroleum workers in the United States and the United Kingdom, 1937- 1989. A-Q, US studies; R-S, UK studies. GARABRANT/BERRYMAN 00382 GARABRANT/BERRYMAN 00383 GARABRANT/BERRYMAN 00384 GARABRANT/BERRYMAN 00385 GARABRANT/BERRYMAN 00386 GARABRANT/BERRYMAN 00387 GARABRANT/BERRYMAN 00388 GARABRANT/BERRYMAN 00389 GARABRANT/BERRYMAN 00390 GARABRANT/BERRYMAN 00391 GARABRANT/BERRYMAN 00392 GARABRANT/BERRYMAN 00393