Tobacco Documents Online

Human Nasopharyngdal Carcinomas Positive for Epstein-Barr Virus DNA in North America,. Ronald Glaser, 3.4 Melhan N.onoyama, 5 Romuald T. Szymanowskl, 6 and William Graham 7 ABSTRACT~Ne~opharyngeII carcinomas (NPC) from 2 black patients and 1 Caucasian pali-.n! were positive for Epstein-Bert virus (EBV) DNA. Of Ihe tumom, 2 were lymphoepllhellome| (undiffe¢enllaled NPC) end 1 was a moderately differentiated NPC. All 3 patients had high IgG tlters against EBV eady antigen and high IgG ind IgA tller~ agalnlt virus capsld antigen (VCA). In one patient, the levels of antI-VCA IgA were dllferent than those o! antI-VCA IgG over the course of the disease, JNCI 84: 1317-1319, 1980. EBV is a human herpesvirus with oncogenic prop- arises. In addition to being associated with the self- limiting lymphoproliferative disease infectious mono- nucleosis (1), It has been associated with NPC (2.-4). Cantonese Chinese are at high risk to NPC (5, 6), and a very consistent association between EBV and Chines~ pi~tients with NPC has been established as demon- strated by high antibody titers against EBV EA and VCA (3, 7). NPC patients from various areas of the world, including Africa and Europe, have been shown IO have EBV DNA associated with the tumor cells (8, 9). An Eskimo population in Alaska also appears to be at risk for NPC, and studies are presently being i~erformed to examine the possible association of EBV in those cases (I0). High geometric mean titers against" EBV antigens have also beenobserved in North Ameri- can NPC patients of varied racial backgrounds (11-13). Evidence suggests that antibody against EA increases with the stage of disease (7, 14) and that it may be possible to use such antibody data to predict the course of the illness (7, 13). In addition, the presence of IgA antibodies against EBV antigens may also be NPC related (15). These antibody titers may be useful in monitoring disease st;~tus because EBV antigen-specific IgA titers are high in untreated NPC patients and dr'crease when the patient .is in good-health (15). Ahbough data suggest that North American NPC I,:ttlents have high anti-EBV titers (11-13), little is known of the association of EBV with NPC in North American patients in regard to EBV DNA-posltive htmors. In this study, we present data on 3 NPC patients of non-Oriental background with EBV DNA- imsitive turfiors. MATERIALS AND METHODS If assays.--Patients" sere were obtained from Henry l"-rd Hospital, Detroit, Michigan, and Milton S. Her- ~hey Medical Center, Hershey, Pennsylvania, and as- s:tx,'ed for IgG and IgA against EBV-specific VCA by the indirect IF test (16). Smears of HR-1 cells fixed ~itlt acet6ne were adsorbed with twofold dilutions of st'rum samples and allowed to incubate for 30 minutes at 87° C. The cells were washe~t with phosphate- buffered saline and readsorhed with fluorescein isothio- cyanate conjugated to either rabbit antihuman IgG or goat ant/human IgA. To determine the anti-EA titer, D98/HR-I cells treated with iododeoxyuridine for $ days were used. Under these conditions, only EA is synthesized (16). The cells were fixed with acetone, and the indirect IF test was performed for anti-EA IgG. The antibody titer against EBNA was determined by the anticomplementary IF test with Raji cells fixed in acetone-methanol according to the procedure of Reed- man and Klein (17). DNA-DNA reassociation kinetics.~Tumor biopsy specimens were kept Irozen at -76° C until assayed for EBV DNA. The DNA was extracted from the tissue by treatment with pronase and sodium dodecyl sulfate followed by phenol extraction. The DNA was purified with ethanol, resuspended in Tris-EDTA, and treated with gNase (18). The DNA was sonicated to obtain an average mass of 2×10s daltons, and DNA-DNA reas- sociation kinetics were performed as previously "de- scribed (18). Sonlcated tumor cell DNA (1 rag) was mixed with a tritium-labeled EBV DNA probe (2×10* cpm), heat denatured in 0.0095 M EDTA, and rapidly cooled. The salt concentration was adjusted to 1.5 M, and the DNA was incubated at 66° C. Samples were taken at various times, and the degree o[ reassociation was determined with S~ nuclease (18). RESULTS Of the NPC patients, 2 were from the Detroit, Michigan, area. The first patient (NPC-I) was a ABBREVIATIONS USED: EA==early antigen; EBNA=Epsteln.Barr virus- associated nu'c|earantigefi( EBV=Epst~in:Barr virus: 1F=immuno- fluorescence; NPC= nasoph'a.ryngeal carcinoma(s); VCA=virus cap- sid antigen. s Received August 2, 1979; accepted October 24, 1979. ~ Supported in part by Public Health Sen.ice grants CA16058. CA21665, and CA~3807 from the National Cancer Institute and contracts NOI-CPSI0"2I and NOI-CPS$SI6 from the Virus Cancer Program, Division of Cancer Cause and Prevention, National Cancer Institute. ~ Department of Medical Microbiology, The Ohio State University College of Medicine and Comprehensive Cancer Center, 333 West 10th Ave., Columbus, Ohio 43210. • Recipient of a Leukemia Society of America Scholar Award. s Life Sciences Biomedical Research Institute, St. Petersburg. Fla. ~3710. • Henry Ford Hospital, Detroit. Mich. 48202. • Milton S."Hershey Medical Center, l|crshey, Pa. 1317 J.~c~, vo~_ ~, No. 6. JUNE 1980 T!04132463

1318 Antibody titer Patient Anti- Anti- No. of EBV DNA No. Anti- VCA VCA Anti- copies/cell EA IgG IgA EBNA NPC-1 1:512 1:1,024 1:160 1:16 4 NPC--2 1:512 1:2,048 1:160 1:4,096 1 NPC-3" 1:256 1:512 1:2,560 1:256 13 • At time of initial surgery. year-old black female whose antibody titers at the time the biopsy was taken were as listed in table 1. The tumor was classified as a moderately differentiated epidermoid carcinoma of the nasopharynx and con- tained 4 EBV genome equivalents/cell as determined by DNA-DNA reassociation kinetics (table I). The second patient (NPC-2) was a 17-year-old black female whose antibody titers were as listed in table 1. The tumor was classified as a lymphoepithelioma and contained I EBV genome equivalent/cell. Follow-up data of the antibody titers of NPG-1 and NPC-2 could not be obtained. The third patient (NPC-3) was a 45-year-old Cau- casian female from central Pennsylvania. The tumor cells contained 13 EBV genome equivalents/cell (text- fig. 1). At the time the biopsy was received, the patient's antibody titers were as listed in table 1. This tumor was also classified as a lymphoepithelioma. We obtained several serum samples from patient NPC-$ over the course of almost 2 years. Her anti-EA and anti-VCA IgG titers rose to 1:8,192 2't0 days after hr$ TEXT-FIGURE I.---DNA-DNA reassociation kinetics. O=Tumor from NPC-S; ~=a mixture of EBV genome-~sitive Raji cells + EBV genome-negative BJA.B ceils to make 10 genomes/cell; X=control BJA-B cells. Experiments were conducted by mixture of ! mg soni~ted tumor cell DNA and 0.5 genomes EBV DNA labeled with [~tl]th~'midine in vivo by superinfection of Raji cells with itR-I vlru~ (18). Degree of hybridization was m~sur~ by cleasc ~cnsiti~ity. ~=DNA concentmtlon (tool nucleotides/liter); C~tontentzation of DNA [~gmcnts whhout duplex regions. jXCl. VOl_ 6,1, NO. 6. JUNE 1980 10,240 2,560 ~60 40" ~ EA ~'--,'O VCA-IqA o-,-,,,o VCA- IgG ~ EBNA zbo ~6o ~ DAYS AFTER ~MISSION TEXT-FIGURE 2.--Antibody liters from serum samples of patient NPC-3 from time of admission to recurrence of tumor (indicated by arrow). surgery, when the first serum sample had been ob- tained (text-fig. 2). A decrease in the anti-EA and anti- VCA IgG titers to 1:512 and 1:'t,096, respectively, oc- curred between 240 and 360 days after admission. Though the changes in the anti-VCA IgG titers in this phase of disease were only twofold, which generally is not considered significant, the fact that multiple serum sample~ gave the same results over time suggests that the change in antibody titers may not be Srtifactual. Approximately '180 days after admission and just prior to recurrence of the tumor, the anti-EA and an.ti-VCA IgG titers increased to 1:4,096 and 1:8,192, respectively. We then examined the anti-VCA IgA titers in patient NPC-$. The levels of anti-VCA IgA were different from the levels of anti-VCA IgG over the time period studied. The anti-VCA IgA titer dropped from the initial very high titer of 1:2,560 to 1:160 by 300 days after surgery and then rose again to 1:640 at approxi- mately the same time the anti-EA and anti-VCA IgG titers increased, i.e., just prior to the recurrence of the tumor. The drop in the anti-VCA I.gA titer after surgical removal of the tumor and initiation of chemo- therapy corresponded with a period of good health in this patient. DISCUSSION Because NPC is not a common disease in North America, little is known regarding the incidence of EBV DNA-positive tumors, though serologic studies have been performed (11-13). We have examined 3 Ti04132464

lit NI'C for the presence of EBV DNA and have also .,ss;tyed anti-EBV tilers in 1 Caucasian and £ black North American NPC patients. In the 3 cases of NPC in persons of non-Oriental b:wkground examined in this study, all tumors con- i:dned varying amounts of EBV DNA. The number of - , enome equivalents per cell ranged from 1 to 13 EBX g . b ;ts assayed y DNA-DNA reassociation kinetics. Of the 3 tulnors, 2 were histopathologically classified as lym- phoepitheliomas (undifferentiated NPC); 1 was clas- sified as a moderately differentiated carcinoma. Recent data suggest that EBV DNA-posidve tumors from European Caucasian patients are primarily of the tmdifferentiated histologic type (19). Similar results have been found in a collaborative study with Dr. U. l,r:,sad, University of Malaya, Kuala Lumpur, Malaysia, in wbich preliminary data suggest that a .high per- cenlage of EBV DNA-positive NPC from that geo- graphic area are either lymphoepitheliomas or un- differentiated carcinomas (Glaser R, Prasad U, Non- o.vama M: Unpublished data). We found that the 3 NPC patients studied also had very high IgG tilers against the EA and VCA compo- ,tents of EBV as well as high anti-VCA IRA titers. An increase in IgG antibodies against EA and VCA and in anti-VCA IgA tilers occurred in patient NPC-3 just prior to the reappearance of a tumor at the original .,i~e. The correlation of the changes in anti-EA and :mti-VCA tilers with the recurrence of disease has been shown in NPC patients from other areas of the world ~4, 7, 14). All 3 patients, NPC-3 in particular, had high IgA tilers against VCA. Also, in patient NPC-3 we found flint the tiler dropped over the course of approximately 300 days from the initial tiler of 1:2,560 to 1:160, k.veled off, and then increased to 1:640 concomitant with the reappearance of the tumor. The anti-VCA IgA tilers were different when compared to the anti-VCA IgG tilers. This difference has been observed in Chi- nese NPC patients as well (20), and further studies will be necessary to clarify this observation. The patient was clinically in good health over the course of the study until the reappearance of the tumor. This period of good health might be reflected in the drop of anti- VCA IRA observed (20). • Or/ml~ts, and that at least in l of the 3 "patients studied in some detail, the antibody pattern is similar to that described in cases outside of North America. These patterns include high anti-EA and :mti-VCA (IgG) levels, IRA against VCA, and an increase in anti-EA and anti-VCA (IgG and IgA) associated with the course of the disease (7, 11-15, 19). Most im p~ l,~.Lagictt~__. A',tatacafi~mad~/n~theat um~m - EBV DNA in North American NPC 1; REFERENCES (1) HENL£ G. HENLE W, DIEHL V. Relationship of BurkiWs tumor- associated herpes virus to infectious mononucleosis. Proc Nail Acad Sci USA 19fi8;59:94ol01. (2) OLo LJ, BoYsr EA, OE'I-rcrtq HF, et al. Precipitating an,ihody in human serum to an antigen present in cultured Burkitt's lymphoma cells. Proc Natl Acad Sci USA 1966;56:1699-1704. (3) HENLE W, HENLE G, HO H-C. et al. Antibodies to Eps,ein-Barr virus in nasopharyngeal carcinoma, other head and neck neo- plasms, and control groups. J Natl Cancer Inst 1970:44:225- 231. (4) De-T~ G, Ho JH, AsLaSm DV. et al. Nasopharyngeal carci- noma. IX. Antibodies to EBNA and correlation with response to other EBV antigens in Chinese patients, lnt J Cancer 197b; 16:713-721. (5) Dtc.~'¢ KH. THOMAS GH, Hs,U ST. No,es on carcinoma of the nasopharynx. Caduceus 1930;9:'15-68. (6) CLtrVORn P. On ,he epidemiology of nasopharyngeal carci- noma. Int J Camcer 1970;5:287-309. (7) H~'NLr W, HO H-C, H~LE G. KW^S HC. Antibodies to Ep- s,ein-Barr virus-related antigens in nasopharyngeal carci- noma. Comparison of active cases and long-,ecru survivors. J Natl Cancer Inst 1973;51:361-369. (8) Zt|R H~US~N H, $1IL'LTE-HOLTHAUSEN H, KLEIN G, et al. EB virus DNA in biopsies of Burkitt tumors and anaplastic caret- . nomas of the nasopharynx. Nature 1970;228:1056-1058. (9) No~o'g^Mx M, Hua:qo CH, P^oa~o JS, et al. DNA of Eps,cin- Ban- virus detected in tissue of Burkiu's lymphoma and naso- pharyngeal carcinoma. Proc Natl Acad Set LISA 1973;70:3265- 3268. (10) LAN,Eg AP, BENDER TR, BLOT '~VJ, FR^UI*.IENI JF JR, HURLBURT WB. Cancer incidence in Alaska natives. Int J Cancer 1976;18: 409-412. (I1) GOt.D,',tA~ JM, GOO~,^N ML, M, LLER D. Antibody to Epstein- Barr virus in American patients with carcinoma of the naso- pharynx. JAMA 1971;216:1618-1622. (12) HENDERSON BE, LOUIE E. BOODANOFF E, HENLE W, ALENA B. HENLE G. Antibodies to herpes group viruses in patients with nasopharyngeal and mher head and neck cancers. Cancer Res 1974;34:1207-1210. (13) nE SCHR'¢VER A. KLEIN G, HEYI.E W. HV,'qLE G. EB virus-asso- ciated antibodies in Caucasian p:uiems wi~h carcinoma of the nasopharynx and in long-term survivors after treatment, lnt ' J Cancer 1974;13:319-325. (I-/) HELLE W, Ho JH. Hr.~LE G, C..~^u JC. Kw^~ HC. Nasopha- ryngeal carcinoma: Significance of changes in Eps,ein-Barr virus-related antibody pat,erns following therapy. Int J Can- cer 1977;20:~6~-672. (15) |"IENLE G, HENLE W. Eps,ein.Barr virus specific lgA sermn anti- bodies as an outstanding feature of nasopharyogeal carci- noma. lm J Cancer 1976;17:1-7. (16) GL~.SER R, NONOYAMA .~1. DECKER B, RAPP F. Synthesis of Ep- stein-Barr virus antigens and DNA in activated Burkitt so- matic cell hybrids. Virology 1973;55:6-°-69. (17) REEDMAN B, KLEIN G. Cxlhtlar localization of an Epstein-Barr virus (EBV) associaled complement-fixlng antigen in pro- ducer aud hint-producer lymphoblasmid cell lines. Int J Can- cer 1973;1 h499-520. (18) TANAKA A. MrYA(;, Y, ~:~JIM^ Y, NONOYAMA M. hnproved pro- dncdon of Eps,ein-Ban- virus DNA [or nncleic acid hybridiza- ,ion studies. Virology 1976;74:81-85. (19) ANDERSSON-ANVRET M, FORSBY N, KLEIN G, ]{ENLE 't.V, BIORK- LU~;D A. Relationship be,ween the Epstein-Barr virus genomc and nasopharyngcal carcinoma in Caucasian patients. Int J Cancer ! 979;23:762-767. (20) Ho HC, Kw^~ HC, No MH, DE-THI~ G. Sernm IgA antibodies to Epsteln-Ban. capsid antigen preceding symptoms of naso- pbaD.ngeal carcinoma. Lancet 1978;1:436. jNcl. rot__ ~. NO. S. JUNE 1980 T!04132465

AH sites combined SNCS 20,625 TNCS 125,542" B~ml cavity and SNCS ~ 1,005 TNCS 4,508 Lip SNCS TNCS Tong~, SNCS TN~S Salivary gland SNCS Gum and mouth SNCS TNCS 285 537 161 940 179 449 197 1,263 183 TNCS 1,319 "l~ige~tive system SNCS 6,031 TNCS 3~,787 • ..~ophagua :. ": SNCS 275 T~C8 1,5§9 Stomach SNCS 1,721 TNCS 4,247 Intestines and r~ctum combined SNCS 2,879 TNCS 18,157 Intestinei SNCS 1,717 TNCS 12,552 Small in u~dne SNCS 62 TNCS • 311 C~lon excluding rectum SNCS 1,655 TNCS 12,241 Rectum and rectosigmoid junction SNCS 1,162 TNCS 5,605 Liver, ga!!blad~ero and biliary tract SNCS 515 TNCS 2,163 Liver SNCS 321 TNCS 1,009 G.'dlhhdder and other billao" tract SNCS 194" TNCS 1,154 " Pa ncrea.~ SNCS 493 TNC~ 4,016 Other digestive system SNCS 148 T.NCS 635 288.9 8,548 277.7 53,970 ~.1I~.q. 691 • "b ! 2,806 4.0 1.2 " 457 2.3 " 117 2.1. 588 2.4 69 1.0 19it 2.8 134 2.8 714 2.6 12~ 87.2 3,0~9 66.6 13,923 4.0 188 3.5 776 25.2 956 9.1 2,131 1,377 8,005 755 5,208 40 133 715 5,075 622 2,797 • 214 156 512 392 ~ 265 1,879 69 41.2 39.3 24.6 27.1 0.8 0.7 23.8 26.4 16.6 12.2 7.4 4.6 4.6 2.2 2.8 2.4 7.1 8.7 2.1 1.4 283.7 10,170 305.0 820 225.6 1,087 309.0- 54,804 9.56.8"~ 8,844 330.2 7,205 ~.o,~.~-z.1:~6 ~.3 3,~ a.~ .i~'-P.~,6 1~ 7.s 39 1.2 0.7 -- 2.s 48 o.~ s 0.3 4 3.8 36 1.1 6 1.5 2 3.3 228 L1 8~ 3.1 35 • 4.0 397 1.8 93 3.4 12.1 1,376 5.8 " 491 18.6 242 45.8 8,084 35.0 986 37.2 97~ 126 ..o.6 1.3 15. 23.8 851 2~.0 49 13,7 40 29.0 5,77(] " 24.8 604 22.9 725 20.7 454 13.9 42 11.4 44 16.0 2,188 9.6 346 13.0 233 " 7.0 256 7.8 23 5.9 22 5.1 925 3.9 216 8.1 115 5.1 134 4.1 19 5,1 12 2.9 279 1.2 159 6.3 49 1.9 122 3.7 4 0.8 I0 2.2 646 2,7 47 L8 66 3.9 , ',82 s,8. ~3 s.9 . 13 10.7q/1,5o9 6.5"Y. :~;5 14.i~' 25o 2.3 ~6 2.0 5 z.z 7 1.3 322 1.5 37 1.3 44 ~73.2 2:]1.5 ~.9 0.1 0.~ 1.1 0.5 1.4 57.0 ~.4 1.5 3.4.. 18.9 7.9 24.4 3L7 12.1 24.1 0.5 11.9 ~.6 ~.3 7.6 S.9 3.8 3.0 1.~ 2.~ 2.2 2.1, 1.4 • Average anne.el number of new cases per 100,000 lx~pulation, age-adjusted to the 1950 United States population standard. P, ate:z beryl on fewer tb.an 100 cases ha~e a standard error of >10~, and those based on fewer than ~ cases have a standard error cf ~.20%. Due to rounding, the rates for all ]eukemias and all lymphomas do not equal the sum of the subtypes. # TNCS c~unts for all races and sexes combined include 719 cases among patients with race not specified. J NATL CANCEI~ INST VOL. 60, NO. 3, .MARCH 1978 y TI041324

o-~ - . . .¢ "~ Cancer Incidence and Mortality Trends in the United States: 1935-741 Susan S, Devesa 2 and Debra T. Silverman 2, 3 ABSTRA~'r--usIng Incidence data from three national cancer • su..',~ya sad mortality data for the entire United States, one can mal~ several observations concerning the trends In cancer occurrence. Rates among males have been Increasing, whereas those amos2 females have been decreasing. In the past, cancer of all sites combined occurred more frequently among females; now melee have rates higher than do females of the same race. White predominance has been replaced by a nonwhite excess In Incidence and mortality rates among males and a nonwhite exceel In female mortality; racial differences In Incidence rates among females have been diminishing. Increases have occurred in cancers of the lung, prostate gland, breast among nonwhite females, pancreas, kidney, bladder among males, and esopha- ggl among nonwhites; melanomas have Increased among whNes, and lymphomas have Increased In each race-sex group.. The reported Increases may be due partly to Improvements in <l]agnosls, which probably ended ~or the different pdmary sites .... and probably affected nonwhites more than whites. Meanwhile, decreases have occur~ed In cancers of the uterus (particularly :=.;'cant'ix), stomach, and liver. Intestinal cancer has Increased, "" " whereas rectal cancer has decreased. Possible problems In "xpecHyIng the site of origin of tumors arising in the area of the " roc~osigmo~d Junction make it difficult to determine how. accuo rately the observed trends reflect the true situation. ~Jfien Intes- tinal and rectal cancers are considered together, the rates for nonwhites hays been Increasing toward the more stable level o! the whiles. The incidence of thyroid cancer has Increased, wheraae the mortality has decreased. Without lung cancer, the. Incid~nce of all cancer types con~bined among white males ~ould'b~ decreasing In recent yeats rather than Increasing° The Increases In overall InCidence seen among nonwhite malee are due primarily to Increases In cancers of the lung, prostate gland, Intestine, and e~ophsgus. Among white females, the decreata In overall Incidence ts due to the decreases In cancers of the uterine cervix and the stomach; breast cancer ratee are steady • and continue to have a major Impact. Except for breast cancer, the exl~rience of nonwhite females is similar, with the declines in uterine cancer being greater end those In the other we attempt to achieve comparability among the three such surveys and to .compare the trends with the ported mortaliW ['or the entire countr},. METHODS Incidence data.--NCI has completed the TNCS in which infbrmation was gathered concerning the 18I ,027 cases of itlvasive cancer diagnosed among- residents of seven metropolitan areas and two entire States during 1969-71 (1). Data concerning carcinomas in situ were also gathered, bat these tumors are not included in this analysis. The population of the surveyed areas was in excess of 21 million, or more than 10% that of the entire country. The survey operation has been detailed in (2). NCI contracted with a medically oriented nonprofit" organization in each of the geographic areas to be surveyed. The local staff then identified cancer cases among hospitalized patients from pathology and au- topsy reports, surgeD• and radiotherapy records, dis- charge diagnoses, tumor registries, and medical record indexes. Information concerning identification of the individual, primary site and histology of the tumor, date and method of diagnosis, as well as demographic data for each patient .was then abstracted from the medical record. All death certificates mentioning were also abstradted; if the case had not been previously reported, an active search was initiated either in the hosph~l where the patient's death occurred or by in- quiry to the physician who had signed the death certifi- cate. All recoi'ds for an individual were consolidated, and extensive "editing" was performed, both manually belng lees than the declines observed for white females.---J" Neff Cancer lest ~0: 545-571, 1978, The stud)• of trends over time in the incidence of and mortality from cancer in a defined po~ulation may provide clues to the etiology of the disease as associa- tions arc identified, linking changes in environmental factors to corresponding changes in disease occurrence. The information may also be used not only to define the current magnitude of the cancer problem but also to estimate future needs for cancer control programs. Ahhough several State and regional cancer re,series have existed ['or many ?ears, no national reporting system was available to define the incidence of cancer in the entire United States. NCI has therefore made periodic survc)'s in selected areas of the country, i~ the following anal.vsis of trends in the incidence of cancer, ABBREVIATIONS U.S£D: NCI = National Cancer Institute: TNC~ - Third National ~nccr Survey; SMSA = standard mctropoli~n statistical arcs; SN~ = ~con~ National ~nccr Sue'c)'; SeA = seven common areas; FNCS = First National ~nccr Su~'ey; NOS = tint otherwise s~ffed; ICDA = lnlernationa[ Cl~sificarion of D~- eases. Adapted. Received August 10, 1977; accepted October 31, 1977. . BaD,natty Br, mch, National Cancer Institute. National Institutes Health, Pttblic Heah.h Sen'ice, U.S. Department of HeMth~ Educe- and Warfare. Bethesda. Md. ~1~. We thank S. Cutler for his encou~gement attd guidance: D. Cramer and E. Lisiccki for their assistance in the developmem o~ the SNCS file: V. VauHohen andJ. Gochenoucr for computer pro~am- mlng: J. Dav~, C. McDonald. and M. Krau~ for techn~l assistance; McKaT ~or mo~alhy data; P. White for ~pulation data; and ~ve~l mcm~rs of the Biomc~ B~anch for their ~dvJce and ¢om- mclll$. VOL. 60, .NO. 3, ,XtARCH 1978 545 j .~ATL CANCER INST T104132467