Document N2ZpgE84OdpEnqpoZ4gX8gVww

302T736030 A R 2 2 6 - 1 308-1 F-261 T-340 P-001 FEB U '9 $ 09:>3 Po&Wt* brand tax transmittal memo 7871 j/ T*~ 7o *u / p h u j+ r y J .O J a J U *jt+ \ LO&U,. UJorfcj u M. * "*** f7 y -> /3 S X y y j-b o s o AN INVESTIGATION INTO THE OOCUKREHCB 07 LEUKEMIA AT WASHIMCTOH KORAS JUDY WALRATH, PH.D. EPIDEMIOLOGIST CHERRY BURKE, PH.D. EPIDEMIOLOGIST EPIDEMIOLOGY 8ECTX0N MEDICAL DIVISION EMPLOYEE RSLATIOKC DEPARTMENT E.I. DU PONT DE NEMOURS AMD COMPANY REVISED: APRIL 1989 ) EID584220 3 -*l?77Wl F-?fii T-^jpi p-nri? FFR 11 ;P RP.--I -2- MPBHWW In April 1989, an additional casa of acuta myelogenous leukemia vas diagnosed in a 51-year old male salary employee at Washington Works. We can assune that the expected nunber of leukemia cases did not increase measurably in this short time period (January 1989-April 1989). Observed and estimated expected numbers of leukemia cases with standardized incidence ratios for 1956-April 1989 are as follows: Leukemia Incidence at Washington Works < ____Malta. 19g3~APril 198?______ Observed Ratio Wage Salary 6 -1 2.7 2.2 1*5 Total 9 4.2 2.1* Statistically significant at the 10% level (two-tailed) The standardized incidence ratio of 2.1 for the total joup la aeaeiaeioaioy eignitioene me no 10 level (tve-ealled test). with the diagnosis of this additional case, a case control study is being initiated. Although the statistical power of this study will be limited, due ts the small number of leukemia cases, a mors thorough investigation into possible associations with work area and chemical exposures is warranted. April 1989 EID584221 -3- SVWMftBX An investigation into the occurrence of leukemia at the Washington Works site in Parkersburg, WV was-begun after employee exproccod oonoorn ever recent aiayijuwa of this uv/iiUltluu among uxeir coworxers. Routine surveillance by the Epidemiology Section of the Corporate Medical Division had not revealed any statistically significant excess incidence of leukemia among Washington Works active employees from 1956-83. There vas also no statistically significant excess mortality fron the group or lymphatic.and hematopoietic tissue malignancies, which includes leukemi&j asiivmy aytivt* employees and pensioners from 1957-84. Analysis of cases or deaths occurring after 1983 or 1984, respectively; had not yet been accomplished at the time employee concern Was raised. This study estimates the number of cases of leukemia expected at the site through 1988 end compares this to the actual number diagnosed. In additionf LL.wi.yu ..up. Sight cases of leukemia occurred among male Washington Works employees from 1956-68, compared with an estimated 4.2 cases expected. This is not a statistically significant excess. However, four cases were diagnosed since the last surveillance period (i.e. during 1984-88), which represents an unusually high Incidence. Two of the 4 cases were of acute non-lymphocytic leukemia, 1 of chronic myelogenous leukemia, and 1 of acute, unspecified, leukemia. Seven of the 8 leukemia cases were deceased by 1988; no additional cases due to this cause were identified thrnuoh tne Epidemiology Section's Mortality Registry. The number of deaths due to the group of lymphatic and hematopoietic cancers was not remarkably different from expected. Preliminary reviaw of work histories did not reveal any commonality of exposure. Zt is recommended that a thorough review of the work areas and potential chemical exposures, particularly to benzene and hydrocarbons, be initiated, and that the site be monitored closely for the occurrence of additional cases. ) EID584222 % 3ju7T3623 F - i l T-3-iO P-OCW FEB 11 ' 9 i 09:31 AN INVESTIGATION INTO THE OCCURRENCE OF LEUKEMIA AT WASHINGTON WORKS BACKGROUND An investigation into the occurrence of leukemia at the Washington Works site in Parkersburg, WV vas begun after employees expressed concern over recent diagnoses of this condition among their coworkers * Leukemia Classifictiong_and Etiologies The disease entity of leukemia comprises a diverse grouping of malignancies of~ the blood-forming organs. These different types of leukemia vary according to cell type and stage of disease; ' The types of leukemia generally fall into four histopathological categories (Linet, 1985): (1) Acute Lymphocytic Leukemia - This type of leukemia is more common in the first two decades of life, uncommon in ages 25-60, and increase in inoidonc*. with age, particularly after age 60. '' (2) Chronic Lymphocytic Leukemia - This type of leukemia is extremely rare before age 40, and increases in incidence with age, especially after age 50. (3) Acute Non-lymphocyt 1r Leukemia - Thia category Include the acute monocytic, scuta m y d v m y U Lc, *nd acuta promveloevtic lauremiac. as well as some rare forms. Fifty percent of adult leukemias are of this type. Incidence is low before age 20, but increases progressively with age thereafter. (4) chronic Granulocytic (Myelocytic) Leukemia - This type of levU&tmia is highest in incidence among adults agvd 30-C0. This type may be referred to as chronic non-lymphocytic leukemia in this report. These distinct types of leukemias are characterised by differing demographic (age, sex, race and ethnicity) and secular patterns, as vail as varying hypothesized etiologies (Linet, 1985). Exposure to radiation has been known to be associated with leukemia, except for the chronic lymphocytic type. Various chemicals have been shown to be toxic to bone marrow and some of these chemicals may also be leukemogenic. Benzene has shown strong evidence of causing leukemia, mostly of t.hf* rut* non-lymphocytic type, in 4eval occupational studies. Organic hydrocarbons have shown some evidence of causing leukemia. Various therapeutic agents have torn iwpocted to nave leukemogenic potential, including alkylating agents used in cancer chemotherapy. Infectious agents also appear to play a role in the etiology of leukemia. This association has been demonstrated in experimental animal.... EID584223 C) -5- studias, and has been supported by studies of in utero exposure and through documentation of clusters in humans. P - r w in n r.aiiVQnn' a T n n Srfo n go % Wut y a Routine surveillance by the Epidemiology .Section of the Corporate Medical Division did not reveal any remarkable excess incidence of leukemia among Washington Works active employees from 1956*83, nor any excess mortality from the group of lymphatic and hematopoietic tissue malignancies, which include leukemia, among active employees and pensioners from 1957*84, as compared to -the rest of the Corporation. Incidence and mortality figures are shown below: Leukemia Incidence at Washington Works Kftles. 1956*83_____ .____ _ ... fifcsfirytd Wage Salary Total 2 -2 4 2.1 0.9 1*1 1*2 3.3 1.2 Lymphatic/Hematopoietio Mortality at Washingtdn Works WilWi 19ff7***84 gfefggYfld Expscted Wage salary 4 4.8 0.8 1*1 1*1 Total S 7.9 1.0 Sfl/W . B M I f l S m flWP BB8ISI Analysis of eases or deaths occurring after 1983 or 1984, respectively, had not yet been accomplished at the time employee concern was raised. This study estimates the number of rases of leukemia expected at. tho cite through 1988 a,l compares this to the actual number diagnosed. Cases of leukemia diagnosed after 1983 were identified from the Epidemiology Section's Cancer Registry and added to the cases from 1956*83. This case series was categorized by histopathological type, year of diagnosis and age at diagnosis. The expected number of cases, based on the Du Pont Corporate-wid was estimated tor Washington Works for the period 1984*88 and added to the expected number for 1956-83. 7 EID584224 3027736030 -6- F-261 T-3dQ P-006 FEB 11 `=>1 Expected numbers of leukemia eases vere estimated by calculating the increase per year in the expected number of cases for Washington Works between the surveillance, periods 1956-77 and 1956-83, and applying this change to th 5-year period from 1/1/84 to 13/31/88. Identification of Leukemia Cases From the records of the Du Pont Cancer Registry, four leukemia cases were reported among aotive employees at the Washington Works site from 1984-88. All vere wage employees. Site medical rooerda vere reviewed fur confirmation of diagnosis. From the Du Pont Mortality Registry, an additional , death from tSuites1* v iAo**viiiod open s*vi vf jMfdcax records, however, this case was determined to be a lymphoma (lymphocytic, lymphosarcoma) and is not included in these analyses. . 5! Thus 8 leukemia cases (6 wage, a salary) 1956- were available for analysis. Descriptive information on these eight cases Is in the Appendix. RESULTS Distribution bv Year .of. Diagnosis Year of diagnosis is distributed as follows by histopathological type: Lymphocytic Leukemia Non-Lymphocytic Leukemia ACUtl 1976 1969, 1978, 1986, 1988 1977, 1986 Unspecified 1988 Four cases (50% of the total case series) have been identified since the lest surveillance period (i.e. during 1984-88). Among an average yearly, population of 2150 employees, 4 cases of leukemia over 5 years represents an unusually high incidence. Distribution bv Histopathological category Distribution of the leukemia cases by histopathological category is as follows: EID584225 % -7- Lymphocytic Leukemia Non-Lymphocytic Leukemia Unspecified Total 1 (12.5%) 4 (50.0%) 1 (i2,sat a (vo.oa) 0 2 (25.0%) -0 2 (25.0%) Total 1 (12.5%) 6 'C?5.0%) 1 (12.5%1 8 Fifty percent of the cases are acute, non-lymphocytic leukemias, consistent with general population findings. Of the 4 recently diagnosed eases, 2 were acute non-lymphooytic leukemia, 1 was chronic myelogenous leukemia and X w as a o u te leukeauLa o f uuy^vllmd t y p e . Distribution bv Ace at Diagnosis Distribution of the leukemia cases by age at diagnosis is as follows: Lymphocytic Leukemia NonvLymphoaytio Leukemia Unspecified 11. 40 36, 42, 44, 57 (mean 44.8) 53 Ptirppl. 36, 44 (mean 40) The case of acute lymphocytic leukemia diagnosed at age 40 is uncharactsristic for this cell type, which is uncommon in ages 25-60. Since the incidence of acute and enronic non-iympnocytic leukemias increases with age, the distribution of these types at Washington Works say be slightly young but within normal range. Ages of the 4 recently diagnosed cases vere 42 and 44 for the acute non-lymphocytic cases, 44 for the chronic myelogenous case and 53 for the acute, unspecified ,case. Pedate of Observed and estimated expected numbers gf gases of leukemia diagnosed among employees and standardized incidence ratios for 1956-88 are shown below. EID58422 -8- Leukemia Xncidtnc* at Washington Works 1956-88____________ Observed Wage Salary Total 6 8 2.7 A,tiiiS 4.2 2.2 1*2 1.9' The standardized incidence ratios for leukemia of 2.2 for wage employees and 1.9 for tha total group approach, but do not reach, statistical significance at tha 10% level (tvowtailed test). :4>: * > : .-t -. . tjpdate of Lvmphatic/Hematopoietlc Tissue Mortality observed and estimated expected nunbars of deaths dua to lynphatic/henatoppietie malignando* among employooc and pensioners fron 1957-88 and standardized mortality ratios are shown below. Lymphatic/Hematepoietio Cancer Mortality at Washington Works Wales. 1957--88 Wage salary Total Observed 8 -i 12 Esptgtafl 6.3 10.3 Ratio 1.3 1* 1.2 Seven of the 12 deaths in this category vara due to lenkomia (5 wage, 2 salary). All 7 vara eases im;ludd in the incidence analyses. No additional cases among active employees or pensioners vara identified Llu-uugh the Mortalizy Registry* Expected Nvnabnrc of Deaths Based on We*L Virginia Rates Assuming an average of 2150 male employees per year at Washington Works, rind using leukemia mgo-adjuoted death rates for west Virginia (National cancer Institute) for white males as follows, 1950-59t 7.7/100,000 1960--691 9.0/100,000 1970-79* 8.4/100,000 (Average - 8.4/100,000) 0.181 deaths from leukemia would be expected at tha site per year. For tha time period 1/1/1957 to 12/31/1988 (32 years), we would expect 5.8 leukemia deaths. Sevan deaths observed from leukemia from 1957-88 is not statistically significantly different from the 5.8 leukemia deaths expected. EXD584227 :OiT73to03Q -9- F-Il T-3-*0 P-009 FEB n _"52' ? : ~ PlggygStQH Over the entire study period (1956-88), an excess of leukemia incidence vas observed among active Washington Works males which vas not quite statistically significant. However, this type of comparison neglects the distribution of deaths by year of diagnosis. Had these cases been distributed more evenly over the study period, and no exposures to known leukemogenie agents such ae radiation or benzene identified, this finding would most likely be attributable to non-occupational factors or to chance. However, the distribution of cases by year of diagnosis is While 4 caco vere dlagno*d during the surveillance period from 1956-83, the remaining 4 ware identified from 1984-88. Dates of diagnosis for these 4 cases were from 1/86 to 7/88. Since analysis through 1983 did not indicate excess leukemia incidence, the apparent cluster of leukemia cases diagnosed in recent years warrants particular attention. A preliminary review of work histories did nnt reveal any commonality of exposure. It is recommended that a thorough review of the work histories and potanfcial ohemieel *xpvures, particularly to benzene and hydrocarbons, and with concentration on the recent cases, be initiated. She site should be monitored closely for the occurrence of additional cases. ) ** The various suspected etiologies of leukemia, differences between cell types, and the small number of cases in this study, limit the conclusions which can be drawn from this investigation. Genetic as well as environmental factors are associated with the development of leukemia. Ionizing radiation, benzene, chemotherapeutic agents, and infectious agents have all been Implicated in leukemogenssis. For exanple, one main frrm this case serioc had previously Lwi treated for thyroid cancer, and his leukemia may have been secondary to treatment for that neoplasm. However, this does not preclude a thorough investigation of his work history and exposures. ) EIDS8422B II 3027736030 -10- F-261 T-340 P-010 FEB 11 *92 09:34 BBftiftBWtnSS 1. Linet, H.S. (1985) TftO-LfVKealiHg?__ Epldesiolocrlc_Aspects. Kev York: Oxford University Press. 2. Riggan, W.B., Mason, T.J., at al. (1983) U^S. Cancer Mortality,Rates__and Trends. 1950-1979. Washington, DC: NCX/RPA. EID58422* \l 11 anoirrm h f ff tf""* tuy D A S or OSOS. *liJ li PABBU. CL Doser BIABOOOTl A O AI . DXAOMMIO BXKAS01OCIC _____ n u _______ HAToa.li/t* D m or sas ai PPM! Oi U 10/U 00/00 37 Aera Nsoccme URXBOA 13/C 307 00/33 00/33 MK/W 00/70 40 ABUS UMMOCRIS UHOMZA -- 12/40 04/00 ujyyy 00/77 30 CBaone BAMuerne imtri 10/7 > : 41 / 00/77 H30 01/3A 03/3* nue 00/70 U AOBS MSLoemq DBCIAJSP 00/70 344 03/41 eo/ot mes 01/00 42 10/41 01/00 ym 07/00 44 sens HliMIIA cnxic ns/BBon DS Ol/OO 40 s AUS > : H-- 02/3* 00/37 WB> 02/0* sa enone uoomu m e *?* 02/00 S3 12/31 U/7S HABI 07/00 <xWj 4 |1 30 5 ocra wwucmc .UuK 00/00 A r t Vri- > *: 30 3 > 3 EID584230 Ib