Document Nebz6Nz8r9w987NOYDz3EdYdw

ye; Anti-p53 Antibodies in Sera of Workers Occupationally Exposed to Vinyl Chloride 9 s'- 0(-f Glenwood E. Trivers, Helen L. Cawley, Virna M. G. DeBenedetti, Monica Hollstein, Marie J. Marion, William P. Bennett, Marie L. Hoover, Carol C. Prives, Carlo C. Tamburro, Curtis C. Harris* Background: The p53 tumor suppres sor gene (also known as TP53) is often mutated in a wide variety of cancers, including angiosarcoma of the liver (ASL). Anti-p53 antibodies have been detected in the sera of patients with leukemia, childhood lymphoma, or cancers such as those of the breast, lung, colon, esophagus, and liver (hepatocellular carcinoma). Purpose: The objective of this study was to determine the prevalence and time of appearance of serum anti-p53 an tibodies during the pathogenesis of ASL associated with occupational ex posure to vinyl chloride. Methods: En zyme-linked immunoassay (EIA) was used to detect anti-p53 antibodies in 148 serum samples from 92 individuals occupationally exposed (in France or in Kentucky) to vinyl chloride; IS of these individuals (six from France and nine from Kentucky) had ASL. A subset of coded EIA-positive and EIA-negative sera was further analyzed for anti-p53 antibodies by immunoblotting and immunoprecipitation. Nucleotide se quence analysis of exons 5-8 of the p53 gene was conducted on ASL DNA from six patients. We tested sera from 31 men who had no occupational exposure to vinyl chloride; they made up the control group. Statistical analyses were done using the Kruskal-Wallis chisquared approximation and the Wilcoxon two-sample test for normal approximation. All P values result from two-sided tests. Results: Fourteen serum samples (from nine individuals) were positive in the EIA. Five of the 15 individuals with ASL were positive for anti-p53 antibodies by EIA, immuno blotting, and immunoprecipitation: one individual at 11.3 and 10.8 years before diagnosis, another at 4 months before and shortly after diagnosis, and three when diagnosed or shortly thereafter. Four of the 77 vinyl chloride-exposed workers without diagnosed ASL were positive for anti-p53 antibodies; two of the four had symptoms related to vinyl chloride toxicity. Tumors from three of the six vinyl chloride-exposed workers from which sufficient DNA for analysis was obtained had A:T to T:A missense mutations of the p53 gene. Anti-p53 antibodies were detected in two of these individuals. Among the control group, two of 15 serum samples from 15 lung cancer patients and zero of 15 serum samples from control subjects without cancer had anti-p53 antibodies at substantially lower levels than the nine (10%) of 92 vinyl chloride-exposed workers who were positive for anti-p53 antibodies. Conclusions and Implications: Serum anti-p53 antibod ies can predate clinical diagnosis of certain tumors, such as ASL, and may be useful in identifying individuals at high cancer risk, such as workers with occupational exposure to vinyl chlor ide. [J Natl Cancer Inst 1995;87:1400- 7]. Angiosarcoma of the liver (ASL) is the most common of the malignant mesen chymal tumors affecting human liver (/). Moreover, it is an extremely rare cancer in humans, except in instances of cata strophic exposure to carcinogens, the usual causal source (2). ASL has been as sociated with exposure to thorotrast (thor ium dioxide), arsenic solutions, and more recently with occupational exposure to vinyl chloride [reviewed in (2)]. In 1974, three ASL deaths in Kentucky were traced to the workers' common exposure to vinyl chloride during their years of employment in the manufacture of poly vinyl chlonde (3), Vinyl chloride-induced ASLs in rats were also reported in that year (4), and observations in humans were later confirmed in workers from European factories (5). Vinyl chloride-in duced ASL evolves insidiously, frequent ly producing surface damage, i.e.. Raynaud's syndrome (blood vessel ob- struction in fingers, toes, or ears causing tissue damage at those sites), but few major symptoms until late in the disease (3,6), Protection against exposure to vinyl chloride in industrial environments was initiated m the 1970s, as were studies to monitor cancer incidence in previously exposed workers (7). Recent molecular epidemiology studies of vinyl chlondeexposed workers have found K-ras (8) and p53 (9) mutations in liver tissue of ASL patients and mutated p21 Ras pro tein (10) in the sera of workers with and without ASL. The p53 gene (also known as TP53) is mutated in a wide variety of human can cers, including cancer of the liver [re viewed in (11-15)]. The predominant genetic alterations are missense mutations in four common regions of the most con served domains of the protein, exons 5-8, resulting in altered protein conformation (16,17), loss of suppressor function (18), and immunochemically detectable ac cumulation of p53 in the nucleus of tumor cells [reviewed in (11,19-21)]. Reduced p53 tumor suppressor function can also be achieved by, for example, a missense mutation that does not alter overall protein conformation (2223). The p53 mutational spectra may reflect mutagenic mechanisms leading to cancer and provide clues to their etiologic origin (11,14). Antip53 antibodies, initially detected in sera from breast cancer patients (24), have been detected in sera from a variety of cancer patients, including, but not limited to, those with childhood lymphoma (25), breast cancer (26-28), lung cancer (2932), leukemia (33), colon cancer (30), hepatocellular carcinoma (HCC) (34), and esophageal cancer (Cawley HL, * Affiliations of authors: G. E. Tnvers, H, L. Cawley, V, M G. DeBenedetti, W P Bennett, C. C Harris, Laboratory of Human Carcinogenesis, Division of Cancer Etiology, National Cancer in stitute, Bethesda, MD; M. Hollstein. German Cancer Research Center, Heidelburg. Federal Republic of Germany: M. J, Marion, Unite 271. Instuui National de la Sami et de la Recherche Medicale. Lyon. France; M. L. Hoover, Coriell Institute. Camden, NJ; C. Prives, Department of Biological Sciences. Columbia University, New York, NY, C. C. Tamburro. Division of Occupational Toxicology 1198. University of Louisville, KY. Correspondence to: Glenwood . Trivers, Ph.D.. National Institutes of Health, Bldg. 37. Rm. 2C04. Bethesda, MD 20892. See "Notes'- section following "References." .ItM iirirjh o iis u fu s reproduced lluu m aterm l with I il.1 IO o p y I I ip il o w n e i F u r 1h e r r e p t o d u c lio ii iu 1400 REPORTS Journal of the National Cancer Institute, Vol. 87, No, 18 September 20, 1995 | [ ASI 000013936 Meltzer S, DeBenedetti VM, Trivers GE, Bennett WP, Harris CC: manuscript m preparation). Anti-p53 antibodies have also been detected in patients with chronic obstructive pulmonary disease who later developed lung cancer (31) or other malignancies (Trivers GE, Cawley HL, DeBenedetti VM, Caron G, Miller RD, Bennett WP, et al.: manuscript in preparation). These antibodies are as sociated with cellular p53 accumulation and p53 mutations in individual tumors (26,29J2). As a putative tumor-specific antigen, mutated p53 protein has been as sociated with a poor prognosis in patients with cancer, including those with breast, stomach, colon, and bladder cancers [reviewed in ((3)]. Likewise. anti-p53 an tibodies have also been suggested to indi cate a poor prognosis in patients with lung (35) or breast (27J6J7) cancer. Most studies of serum anti-p53 antibodies, however, have used patients with rela tively late stage cancer. Here, we describe the detection of antip53 antibodies in sera from individuals occupationally exposed to vinyl chloride. Our objective was to determine the preva lence and time of appearance of these an tibodies during the pathogenesis of ASL. Subjects and Methods Vinyl Chloride-Exposed Workers and Unexposed Control Subjects Exposed workers. One hundred forty-eight ,erum samples were obtained from 92 workers recruited (via informed consent approval by the in ternal review boards of the University of Louisville, Kentucky, and the National Institutes of Health and Medical Research, Insmut National de la Santd de la Recherche Medicate [INSERM], in Lyon, France) for two industry-initiated or industry-assisted projects in Kentucky (one factory) and France (INSERM, five factories) to study vinyl chloride-in duced ASL and workers at risk for development of ASL due to prolonged, high exposure to vinyl chloride. Sera were obtained from 87 workers m France and from nine workers in Kentucky and were stored at -70 'C from 6 months to 18 years before being used for this study Aliquots of the sera from the French workers were used previously to study p2l Ras protein UO). The workers in this study held a variety of positions providing high exposure to vinyl chloride for 12-34 years, including an ex tended period before 1974 (0-34 years) without protection from exposure, and a shorter period after 1974 (0-18 years) with protection (7). Medical and occupational histones of the workers were provided, including job classifications, smoking histones, in formation on alcohol consumption, clinical records, and years of exposure to vinyl chlonde (high [before 1974]; low [after 1974]), along with the results of penodic. extensive physical examinations. Tumors, benign and malignant, were venfied by his topathologic examination. There were 92 factory donors (nine Kentuckians [aged 38-72 years; mean 1 SD = 51.2 11 years); S3 French (aged 34-71 years; mean + SD = 58 6 years]). All nine Kentuckians had diagnosed ASL. Of the 83 French, 41 were asymptomatic, 26 were symptomatic for toxic effects of vinyl chloride ex posure, eight had benign tumors, and eight had malignancies (six ASLs and two HCCs). Twentyfour workers (nine in Kentucky and 15 in France) donated multiple (two to eight) blood samples, some before and after diagnosis of ASL and/or surgery. Asymptomatic patients, those with symptoms, and those with angiomas or HCCs had similar years of vinyl chlonde exposure in three exposure categories: high (before 1974, mean SD = 13 3 0.9 years), low (after 1974, mean SD = 7.3 0.5 years), and the total accumulated time (mean SD = 20.7 1 2 years). ASL patients had approximately 50% more time (20 5.7 years) in the higher exposure period, 50% less time (3.7 i 4.2 years) m the lower ex posure periods, and only a slightly greater (+15%) total time of exposure. Unexposed control subjects. Serum samples as well as medical and smoking histones were obtained from 30 men (aged 41-81 years, mean SD = 64 12 years; from an ongoing study of lung cancer, smoking, and anti-p53 antibodies), 15 of whom had lung cancer, 15 who did not have a history of any cancer, and none of whom had industnal exposure to vinyl chlonde. One HCC patient from Kentucky without vinyl chloride exposure was included in the study as an additional control subject. Examination of a larger number of unexposed control subjects was considered unnecessary on the basis of published reports (24-28) that indicated that antibody-positive serum donors are found predominantly among can cer patients. In a recent study (37) that included 200 normal donors, only one antibody-positive serum (0 5%) was found in a patient who had no symptoms of cancer. Moreover, a survey of nine of the refer enced reports (24-26.28-31J437) revealed a total of 167 (14%) positive donors among 1197 cancer patients, compared with three (0.4%) positive donors among 736 noncancer control donors, in cluding 40 and 67 patients with organ-specific, non cancer clinical symptoms affecting the lung and liver, respectively. Enzyme Immunoassay to Detect Anti-p53 Antibodies in Human Serum We used modifications of enzyme immunoassay (EIA) methods previously described (38). Briefly, purified human p53 (39.40). unrelated protein as control (bovine serum albumin (BSA) fraction 5; Sigma Chemical Co.. St. Louis, MO; selected in ex perimental companion with mock protein and horse and human serum albumin), and ail other reagents were added to wells of micronter plates in 50-uL volumes. The plates were washed five times follow ing each reaction (37 'C for I hour). We dried 2.0 ng/well of p53 and 2.0 ng/well of BSA in tnplicate wells and stored them at -20 "C. Plates were washed, then blocked with 4% goat serum (GTS) (Life Technologies, Inc. [GIBCO BRL], Gaithers burg. MD) in lx phosphate-buffered saline (PBS) containing 0.05% Tween-20 (GTS-T,0j. The serum samples, stored at -70 "C. were thawed, diluted 1 100 in 1% GTS-Tsq. and applied simultaneously to p53 and BSA in single, tnplicate-wetl columns. 10 samples per test plate. Rabbit anti-p53 antiserum 11 40 000 dilution. CM-1, from Dr David Lane. Department of Biochemistry, Medical Sciences In stitute. The University Dundee, Scotland. U,K,) in 1% GTS-T:q was added in a single column as posi tive control. Detection of bound antibody involved use of alkaline phosphatase-conjugated, goat anti rabbit immunoglobulin G (IgG) or alkaline phos phatase-conjugated, goat anti-human IgG antiserum (Jackson ImmunoResearch Laboratories, Inc.. West Grove. PA). We added 0.1% paranitrophenyl phos phate (Sigma Chemical Co.) in 10 tnM magnesium chlonde and 10 miVf diethanolamine (pH 9,6) to the wells, and the product paramtrophenol was measured at an absorbance of 405 nm after 15 minutes, up to 6 hours, and after storage overnight at 4 "C. Criteria for EIA-positive assays and samples. Each Seram sample was assayed three times. A posi tive assay required a mean p53-ro-BSA ratio of I 5 or greater (without the deletion of the lowest wells m individual triplicate assays) and a difference be tween the two means equal to two standard devia tions or greater. A positive sample required at least two positive assays. All samples were examined by EIA. Those samples with positive EIA results, along with a selected number of negative samples, were submitted for analysis by immunoblotting and precipitation. Statistical evaluations of EIA. EIA results were converted to specific activities and compared in SAS (PROC NPAR1 WAY; SAS Institute Inc., Cary, NC), using the Kruskai-Wallis chi-squared approximation and the Wilcoxon two-sample test for normal approximation. All P value results are from two-sided tests. Although the ratios were used for their readily observable magnitudes of differen ces desired for determining antibody positivity, we used specific activity (p53 minus BSA A405) for the statistical analyses of the results. We considered raw data to be more reliable for this purpose than a derived factor Immunoblotting Procedures used were modifications of methods previously described (41) p53 (10 ng/lane) and BSA fraction 5 (10 ng/lane) as control antigen were fractionated in 12% polyacrylamide gels containing 10% sodium dodecyl sulfate The proteins were transferred by electrophoresis to nitrocellulose membranes (Bio-Rad Laboratories. Richmond, CA) in buffer (lx PBS. 0.1% Tween 20, 0.02% NaN-,. and 3% BSA) for 1 hour, then incubated for 2 hours with 1 : 100 human sera and 1 ; 10 000 rabbit amip53 (CM-I) as positive control. Antibodies were de tected with I : 1000 anti-rabbit IgG (Jackson ImmunoResearch Laboratories, Inc.) and I 1000 anti-human IgG (Jackson ImmunoResearch Laboratories) as second antibody treatment (I hour), followed by a 1-minute incubation in ECL Western detection reagents (Amersham Life Science Inc , Arlington Heights. IL). Molecular weight markers (49 500-205 000 daltons, prestained sodium dodecyl sulfate-polyacrylamide gel electrophoresis stan dards; Bio-Rad Laboratories) were run on each gel to identify the p53 bands. The nitrocellulose msra- Journal of the National Cancer Institute, Vol. 87, No. 18, September 20, 1995 AS! 000013937 REPORTS 1401 branes were exposed to Hypertiim (Amersham Lift Science Inc.). All sera that were positive at a 1 100 dilution were reassayed at I 100 or higher dilutions; sera negative at I 1 100 were reassayed at 1 : 100 or lower dilutions (1:50: I 25), where interpretation could become difficult because of increasing background staining, but frequently al lowed the reproducible confirmation of an EfA result. Purified p53 was prepared by modifications of procedures previously described (39). Recombinant baculoviruses (from Drs, D. O. Reilly and L. Miller, University of Georgia. Athens) transfected with the wild-type human p53 gene were used to infect SF9 insect cells (Invitrogen. San Diego. CA) grown in culture. The harvested cells were lysed, and the p53 protein was immunoaffinity purified on p53-specific monoclonal antibody (PA842I; Oncogene Science, Inc.. L'niondale. NY) columns (A-Sepharose. On cogene Science), using monoclonal antibody (On cogene Science `(SO!) or rabbit anti-p53 antibody (CM-l) to define the elution profile (data not shown). The concentration of protein in column fractions was determined by protein assay (Bio-Rad Laboratories) and by silver siajmng (Bio-Rad Laboratories) after electrophoresis in poly acrylamide gels. The purified pi3 protein produced one to three bands further characterization was also done by immunoblotting using additional p53specific monoclonal antibodies. The choice of a control antigen was determined by experimentation. Using antt-p53 antibody-positive and antibody negative human sera and rabbit anti-human p53 (CM-l) antiserum, we tested mock protetn (immunoaffimty purified from SF9 cells infected with nomransfected baculoviruses), several different preparations of human sentm albumin, horse serum albumin, and BSA (Sigma Chemical Co.), and proteins from p53-expre$sing (p53**> and p53-nonexpressing (p53 ) Caiu 6 tumor cells (American Type Culture Collection. Rockville, MD- No. ATCC HTB 56, from a lung adenocarcinoma that does not express p53; transfected with p53 mutl43ala in this laboratory by Dr. E. FeileyBosco) (42) No control antigen was better than BSA fraction 5. Therefore, BSA fraction 5 was used as the negative control m the EJA. Immunoprecipitation In vitro translation of '*S-labeled pS3. Proce dures were carried out according to directions pros ided with the Promega Kit (TnT SP6 Coupled Rabbit Reticulocyte Lysaie System; Promega Corp.. Madison. WD. DNA (plasmid: p-Select p53. 1.27 p.g/y.L, prepared by Dr. Felley-Bosco {42)], 20 jaCi or '_S-labeled cysteine iDu Pont NEN, Boston. MA). and the kit reagents were mixed in a total volume of 50 pL. then incubated at room tempera ture for 90 minutes Plasmid DNA (2 pg per reac tion) produced 200-300 ng p53 (by Coomassie blue .taming) in 50 pL of the TnT system, Immunoprecipitation procedures, A 10-p.L volume of serum was diluted in 90 pL of 1% goat serum m GTS-T:q. We added 20 pL of this mixture io 80 pL (final dilution = 1-50) of the reaction solution, containing 4-16 pg (estimated by im munoblotting) of 3 S (Du Pont NENWabeled p53 protein. The positive control was l : 10 monoclonal antibody DO-1 (Oncogene Science), Protein A-con- jugated agarose beads (Oncogene Science) were added, and the precipitated mixture was fractionated in a polyacrylamide gel along with Rainbow 14Cmethylated, molecular weight markers (14 300200 000 daltons; Amersham Life Science) and exposed to film The negative controls consisted of sera from cancer-free donors without vinyl chloride exposure. All assays for matched sera from the same individual were done in a blinded manner and not necessarily on the same day with the same UClabeled p53 translation product. DNA Sequence Analysis Tumor and nontumor DNA from four ASLs (cases 26, 14. 8. and 3) and one HCC from the French workers had been previously examined for mutations in the p53 gene (9). DNA from an addi tional ASL (case 17) was obtained from a cell line generated with percutaneous biopsy tissue (Marion MJ, Boivm S, Manel-Planche G, Hollstein M. Montesano R: manuscript m preparation). Along with this case. DNA from the tumor of one ASL patient and DNA from the tumor of one HCC patient (both from Kentucky) were analyzed for p53 mutations during this study, making a total of six ASLs and two HCCs assayed for p53 mutations and an tibodies. Exons 5-8 were amplified by polymerase chain reaction (PCR) using mtronic primers and se quenced as previously described (43) Positive samples were confirmed by sequencing from a second independent PCR from tumor DNA. Results Of the 148 serum samples assayed by EIA at a 1 : 100 dilution, a total of 14 samples from nine workers had anti-p53 antibodies. Table l shows the antibody detection results for the nine positive workers and one negative worker as p53/BSA ratios. Mean ratios for EIApositive samples ranged from 1.53 to 24.00, with those at higher levels giving titers (i.e.. dilution factor to detect an tibodies) of 400-1600 (worker 24. samples a and b, Ftg, C, worker 23, Fig. 2). Six of nine workers with anti-p53 an tibodies provided multiple serum samples. Worker 24, who died of ASL, provided three serum samples at successive 6month intervals (11.3 years to 10.3 years prior to diagnosis). The first two samples contained detectable anti-p53 antibodies (Table 1, Fig. 1); the third did not. Con trol sera included two positive for antip53 antibodies among 15 serum samples from lung cancer patients (data not shown) and none positive in 15 serum samples from control subjects without cancer diagnosis (Fig. 3); these preva lence rates were significantly lower than the five positive serum samples among 15 workers with diagnosed ASL (P<.002). All EIA-posmve serum samples and an approximately equal number of randomly selected negative samples, combined with similarly selected samples from other studies, were coded and tested together in the immunoblotting and immunoprecipi tation assays. Forty-two serum samples (27 from workers in France and 15 from Kentucky workers) were examined. Nine serum samples (from six workers), all EIA positive, were positive by immuno blotting. Ten serum samples (from seven workers) were positive by immunoprecip itation, including the six that were EIA positive. Seven workers were antibody positive in at least two of the three assays. Table 1 shows the close relationship be tween the increasingly higher EIA ratio and the detection of antibodies by im munoblotting and immunoprecipitation. Five of the 15 workers with ASL had anti-p53 antibodies that were detected by the three anti-p53 antibody detection as says. Two anti-p53-positive ASL workers had detectable antibodies prior to the diagnosis of ASL: worker 23, 4 months before and on the day of diagnosis (Fig. 2, A and B); worker 24, 11.3 and 10.8 years before the diagnosis (Fig. 1). Workers 17 and 26 were positive at the time of diagnosis, and worker 25 was positive 7.5 years after partial hepatectomy (Fig. 2, A and B). Anii-p53 anti bodies were also detected by all three assays in sera from two of 43 asymptom atic workers (workers 21 and 22) (Table l). However, two of 26 workers with nontumor clinical symptoms (workers 19 and 20) had anti-p53 antibodies that were detectable only by the EIA. As previously reported (9), the two ASLs among the five tumors (four ASLs and one HCC) available from France for sequencing had A : T to T : A missense mutations; the ASLs from worker 3 (exon 7, codon 255: ATC to TTC) and worker 26, with the highest anti-p53 antibody titer (exon 7, codon 249: AGG to TGG). DNA from an ASL cell line (Marion MJ, Boivin S, Manel-Planche G, Hollstein M, Montesano R: manuscript in preparation: see "Subjects and Methods" section) from worker 17 (Table 1; Fig. 2, A and B) was examined and was also found to have an A : T to T : A mutation (exon 5. codon 179: CAT to CTT; the data have not been previously published). The only workers m Kentucky with sufficient DNA avail- 1402 REPORTS Journal of the National Cancer Institute, Vol. 87, No. 18, September 20, 1995 AS! 000013938 Table 1. p53 antibodies in vinyl chloride workers with and without angiosarcoma of the liver (ASL) by descending enzyme-linked immunoassay iHlA)' Mean EIA Worker No. 26 25 Smoking Age.y status (Years) exposedt Occupation 59 Smoker [26] 16 Autoclave worker 53 Never smoker [28] 19 Autoclave cleaner 24 Sample a Sample b Sample c 23 Sample a Sample b h Sample a Sample b 2! Sample a Sample b 20 19 Sample a Sample b Sample c Sample d 17 Sample a Sample b 3 Sample a Sample b Sample c Sample d 72 Unknown [34]32 Chemical worker 53 Unknown [28)27 Vat cleaner 68 Smoker 74 Smoker [28]23 Polymerization worker 55 Never smoker (36)23 Polymerization 57 Never smoker worker 56 Smoker [28)14 Synthesis and polymerization worker 60 Smoker 61 Smoker 62 Smoker 64 Smoker [25)14 Mechanic/maintenance worker 57 Smoker 61 Smoker [23)5 Autoclave operator 61 Smoker 61 Smoker (28)28 Autoclave cleaner Time| between assay and diagnosis At diagnosis 7 y after resectiontt; 3 y, 4 mo, before death Clinical symptoms: survival; mutation p53/ BSA Immuno- ratio Result blottingll precipitanonl ASL#: p53MUT 249: AGG>TGG** ASL# 24.00 19.50 +4* 4-+ 4- 4- 11 y, 4 mo, before diagnosis 10 y. 10 mo. before diagnosis 10 y. 4 mo, before diagnosis 4 mo before diagnosis At diagnosis 4 y before assay 3 y before assay 6 y before assay 4 y before assay Not applicable ASL# ASL# None None None None RS 4 06 2.69 0.70t + 4- 3.90 2.60 + 3 00 2.34 4+ 1.53 2.02 1.95 4- 4- 4- 44- 44- + 4- + - + + + 44- 4 _ 4 y before assay 3 y before assay 2 y before assay 1 mo before assay RS RS; hepatomegaly (recent liver nodule)ll II 1.88 1,72 1.62111 1.46H 4- + 3 y, 8 mo, before diagnosis ASL# p53MUT** 1.04 At death 179:CAT>CTT 1.55 4- At surgery 21 d after surgery 36 d after surgery 63 d after surgery ASL: PRT.HEPTY#,#*( 1.03 0.87 255:ATC>TTC** 1.11 1.00 _ - - _ - - 4- ND ND ND ND _ - - + ND ND ND ND Table contains data from 10 vinyl chloride workers (six of 15 with ASL and four of 77 without an ASL diagnosis) examined for serum antibodies to p53 (see "Subjects and Methods" section). Values - [total years] and years of unprotected exposure to highest levels of vinyl chloride (before 1974, when protective measures were begun). Time interval in years, months, and days between collection of serum sample and diagnosis of cancer. Mean of three triplicate-well EIAs. - = negative; + = positive; ++ = highly positive; = questionably negative. ((Results of three immunoblottings. - = negative; + = positive: ++ = highly positive; = questionably negative; ND = not done. ^Results of three immunoprecipitations of in vitro translated 55S-labeled p53. Volumes of sera received from these patients were frequently insufficient to support required attempts to obtain conclusive results in assays with higher concentrations. ND = not done. #Dead. **Missense p53 mutation showing codon and amino acid changes in exons 7. 5. and 7. respectively, from patients 26, 17, and 3. (Recurrence of ASL seen during aborted liver explant. tiDetection indicates a highprobability of the presence of p53 mutations; the loss of detectable anti-p53 antibody may be due to either antigen depletion or excess, causing antigen levels to recede below assay detection levels. RS = Raynaud's syndrome. II IILiver lesions were found subsequent to the positive assays for p53 antibodies. 1*1 Only one of three triplicate-well assays was positive. ##PRT.HEPTY = partial hepatectomy. performed before serum samples were taken, and no serum samples were available just before or just after surgery, eight serum samples collected in 7 months were negative, the first three of which (21-63 days) are shown in the table. able for analysis (one antibody-negative worker with ASL and the antibody-nega tive HCC control subject) were found to iave wild-type p53 in the exons ex amined. Two of the three workers with mutations (workers 17 and 26) had antip53 antibodies. Worker 3 was negative for anti-p53 antibodies. In addition to the 15 workers with ASL, 26 workers had clinical symptoms related to vinyl chloride toxicity. These symptoms included Raynaud's syndrome (15), hepatomegaly (9), and splenomega ly (2), occurring alone or in various com binations. Seven workers with benign angiomas and two with HCC were nega tive for detectable anti-p53 antibodies. None of the hepatic angiomas were avail able for sequence analysis. Worker 17 was diagnosed with an hepatic angioma 3.4 years before presenting with ASL. He provided two serum samples; one at 3.7 years before diagnosis and one at the time of the diagnosis, after admission to the Journal of the National Cancer Institute, Vol. 87, No. 18, September 20, 1995 ASI 000013939 REPORTS 1403 Fig. 1. Detection of serum anu-p53 antibody in vinyl chloride-exposed worker 24 by the enzyme immunoassay (EIA), immunoprecipitation, and immunoblotting (IMMUNOBLOT) assays (see "Subjects and Methods'' section, sample analyses are vertically aligned for cross-reference). A) Detection in serum sample prepared 11 years, 4 months, before diagnosis of angiosarcoma of the liver, B) Detection in serum sample prepared 10 years, 10 months, before diagnosis of angiosarcoma of the liver. EIA and immunoprecipitation produced titers of 1600 and 800, respectively, from both serum samples. Immunoblottmg achieved detection at l 100 dilu tion only. A third serum sample prepared at 10 years, 4 months, before diagnosis (6 months later) was negative. Marker = molecular weight indicator (see "Subjects and Methods" section); negative = normal serum as negative antibody control; positive = rabbit anti-p53'Specific antibody; kD = kilodaltons of molecular mass. IMMUNOPRECIPITATION | g | 65kD -- 46kO -- 1 I 50115701100 WORKER NO. rMOO221 501 1 231 50, 1r--SO--i r1 5^0i DSILEURTUIOMN II 1111- Vv* \ay\"1y \%\ \VV\X%v0, \^-A \Y*O, % % %% !i..17..-i WORKER NO, 22 23 25 26 1 " '! \\ \<\5. O, V % \ A\^A W % \\\A \ % \ V\ \\ % Fig. 2. Detection of serum ami-p53 antibody (see "Subjects and Methods" section) in vinyl chlonde-exposed workers 17. 22. 23. 25, and 26 by the enzyme immunoassay, im munoblottmg (IMMUNOBLOT), and im munoprecipitation assays (see "Subjects and Methods" section). Marker = molecular weight indicator (see "Subjects and Methods" section); CM-1 = rabbit anti-p53spectfic antibody as positive control; normal = normal human serum as negative control; Y = years; DX s diagnosis, DO l = p5 3specific antibody-positive control; and M = months. A) Immunoblotting results show that worker 17 was negative before but posi tive at the time of diagnosis; worker 22. an asymptomatic worker who is still under ob servation, has had two positive serum samples 4 and 3 years before these assays were performed, requiring, respectively, a maximum dilution of 1 : 100 and I 50 for detection; worker 23, an ASL patient now dead, had positive serum samples 4 months before and at the time of cancer diagnosis (this worker had an anti-p53 antibody titer of 400 by EIA, data not shown). B) Im munoprecipitation shows the same positive results for these workers and a more distinct ly decreased titer in the second serum sample from asymptomatic worker 22. 1404 REPORTS Journal of the National Cancer Institute, Vol. 87, No. 18, September 20, 1905 AS I 000013940 i.4r *26 1.2 1.0 9 < 0.8 > c 0.6 Q. w 0.4 A *25 A#23a A#2to A#23b t2U #17 0.2 A Ab+: All Assays A Ab+: EIA Only #: Patient ID a. b, c: Serum ID O: Negative -------: Median A*22a A #22b A#2lb A#19a A#20a A#19b A#21a 0.0 rmififlmm < nra-rim Workers 15 Ab+ (5) Angiosarcomas 11 (0) Angiomas &HCC 26 (2) Clinical Symptoms 41 (2) No Symptoms 15 (0) Unexposed Fig. 3. Enzyme immunoassay (EIA) results shown as spectfic activities (mean p53 Aaos minus mean bovine serum albumin A-ios) of individual serum samples in groups of workers screened for anti-p53 antibody (see "Subjects and Methods" section) after 12-34 years of high industrial exposure to vinyl chloride: 31 serum samples from 15 workers with angiosarcoma of the liver (ASL), two of whom (patients 23 and 24) had detectable antibody before the diagnosis; 14 serum samples from 11 workers with other tumors (eight an giomas and three hepatocellular carcinomas [HCC]); 35 serum samples from 26 workers with clinical symptoms of vinyl chloride toxicity: 44 serum samples from 41 workers without symptoms; and 15 serum samples from 15 age-matched male control subjects without cancer and without industrial exposure to vinyl chloride. Results in workers with ASL were statistically significant when compared with those in workers with and without symptoms (P<.001 and PS.0005, respectively) and volunteers without vinyl chloride ex posure (PS 002) Results in workers with and without symptoms were not significant when compared with those m unexposed volunteers (P5.08 and P>. 1, respectively). However, the frequencies of positive vinyl chloride-exposed workers were statistically significant when compared with the results in antt-p53 antibody positive normal and symptomatic cancer-free control subjects (three positive cases in 736) compiled from the published studies cited (see "Subjects and Methods" section) (i.e., versus five positive cases in 15 vinyl chloride-induced ASL: PS.OOOl; versus four positive cases in 77 non-ASL workers. PS.OOl). hospital. Only the latter sample was posi tive, and the worker died shortly after ad mission to the hospital. Serum was unavailable immediately before and im mediately after surgery (partial hepatectomy for ASL) for antibody-negative worker 3, and eight samples taken from 1.5 to 6 months after surgery were nega tive. (Results for the first three serum samples are shown tn Table l.) However, antt-p53 antibodies were detected by EIA in two workers (19 and 20) without clini cally evident cancer (Table l). These two workers had Raynaud's disease. Worker 19 also had hepatomegaly and later developed a liver nodule that was found subsequent to the report of finding the anti-p53 antibodies. Discussion The objectives of this study were to begin an elucidation of the molecular pathogenesis of ASL and to identify biomarkers for use in its early diagnosis. Workers with aggressive ASL survived about 6 months after diagnosis, and about 50% had metastases at the time of diag nosis. Long-term survival after partial hepatectomy has been observed for some ASL patients; however, because of the metastatic disease, few ASL patients sur vive more than 1-3 years after complete resection (5,44). Workers in this study often presented late, and some died the day of the ASL diagnosis. tn this study of 92 workers occupation ally exposed to vinyl chloride, we have found serum anti-p53 antibodies in five (33%) of 15 workers with ASL and in four (5%) of 77 workers without clinical ly evident cancer. Two of the workers with clinical symptoms of vinyl chloride toxicity had antibodies detectable by one assay (EIA) only, a result similar to that reported in studies of breast cancer (28) and lung cancer (31) patients. These find ings could reflect variations in assay sensitivities in different laboratories. However, we did not detect anti-p53 anti bodies in three workers with HCC or in seven workers with liver angiomas. The incidence of anti-p53 antibody-positive ASL patients was somewhat higher than the 13%-25% incidence reported for breast cancer (24,27) and lung cancer (29,32) patients as well as for the in cidence of most of the other types of can cer reported to date, including 20 (25%) of 80 HCC patients (34). The actual in cidence could be lower because of the small number of ASL patients in this study. Two of three ASL patients with p53 mutations had p53 antibodies. The most notable result was that, in two ASL patients, anti-p53 antibodies were detected from 4 months and 11 years, respectively, prior to the cancer diagnosis, and one patient had anti-p53 antibodies before and after surgery. The detection of anti-p53 antibodies this far in advance of diagnosable ASL is not clear ly understood, but it could be the result of earlier antigenic presentation to the im mune system, due to the more cytolytic nature of vinyl chloride toxicity. Long term, high-level vinyl chloride exposure Journal of the National Cancer Institute. Vol, 87, No. 18, September 20, 1995 AS I 000013941 REPORTS 1405 leads to extensive tissue damage and ac tivation of multiple systemic disorders, including sclerotic syndrome, acro-osteolysis, Raynaud's-like symptoms, thrombo cytopenia, and liver damage (44). In addi tion, vinyl chloride is mutagenic and car cinogenic, producing DNA strand breaks and gaps, chromosomal fragments, di centrics, and rings (5). If this cellular damage occurs early in the carcinogenesis process, it provides a plausible descrip tion of conditions that might lead to early development of mutant cells (45) and pre sentation of an antigenic mutant p53 protein. The process by which mutated p53 protein activates an immune response in humans is not clearly known at present. Mutated protein may complex with Hsp70 protein before presentation to the immune system in some breast cancer patients (26); missense mutations may not be associated with the epitope of the al tered protein (27.31), and human anti-p53 antibodies may not be able to differen tiate between mutated and wild-type protein (29,30); however, it does appear likely that mutations and accumulation of large amounts of the protein are the in itiating source of its immunogenicity. The failure to detect anti-p53 antibodies in the third serum sample from previously posi tive worker 24 (Table 1, Fig. 3) is at tributable only to the poorly understood, multifactorial immune mechanisms and host responses functioning in p53-related carcinogenesis. The spectrum of p53 mutations can provide clues to the exogenous and en dogenous mutagenic mechanisms leading to cancer (11,14). The finding of one ad ditional missense mutation of the p53 gene raises the total to three of six vinyl chlonde-induced tumors with mutations and extends the results of this and a pre vious study of ASL (9). These mutations appear to be consistent with the premutagenic ethenoadenosine DNA adducts induced by vinyl chloride (46) and might be explained by the recurring or increas ing lesions caused by continuing high ex posure to vinyl chloride. To our knowledge, this is the first study of anti-p53 antibodies in patients with and without cancer who have had time-measured industrial exposure to a known chemical carcinogen, vinyl chlor ide. In other studies, however, anti-p53 antibodies were detected in one patient with chronic obstructive pulmonary dis ease and one with a benign tracheal tumor 4 months and 15 months, respectively, before a diagnosis of lung cancer (31,47) and in four of 36 women with a positive family history of breast cancer (28). In addition, we (unpublished results) have detected anti-p53 antibodies in four patients with chronic obstructive pul monary diseases before diagnoses of can cer of the lung (6 and 7 months), breast (5 months), and prostate (11 months). These results are consistent with the finding that p53 mutation and nuclear protein accum ulation can occur in preinvasive lesions during carcinogenesis in hun iaii esopha gus (48). Cancers in which alterations in p53 have been shown to occur as early events in carcinogenesis include cancers of the lung, skin, breast, liver, and esophagus [reviewed in (14)]. The current data on the relationship between p53 mutations and anti-p53 antibodies indi cate that studies to detect anti-p53 an tibodies in populations at high risk for cancers that typically exhibit early chan ges in the p53 gene might be informative (Fig. 3), particularly with respect to the usefulness of antibodies as surrogates for early p53 mutations and as potential aids in the early detection of some cancers. References (1) Lotze MT. Flickinger JC, Carr BI. Hepatobiliary neoplasms. In. DeVita VT Jr, Heilman S, Rosenberg SA, editors. Cancer: principles and practice of oncology. 4th ed. Philadelphia: Lippincott, 1993.883-914. (2) Fleming LE, Beckett WS. Occupational and environmental disease of the gastroentestinal system. In: Rom WN, editor. Environmental and occupational medicine. 2nd ed. Boston: Little, Brown & Co. 1992.633-47. (2) Creech JL Jr, Johnson MN. Angiosarcoma of liver in the manufacture of polyvinyl chloride. J Occup Med 1974; 16:150-1 (4) Maltoni C, Lefemine G. Carcinogenicity bioassays of vinyl chloride: 1 Research plan and early results. Environ Res 1974;7:387405. (5) Vinyl chlonde and polymers: vinyl chloride, polyvinyl chlonde and vinyl chloride-vinyl acetate copolymers. IARC Monogr Eval Carcinog Risk Chem Hum 1979; 19-377-438. (6) Creech JL Jr, Johnson MN. Block B. An giosarcoma of liver among polyvinyl chloride workers--Kentucky. Morbidity and Mortality Weekly Rep 1974:23.49-50. DHEW, CDC Publ No. 74-8017, (7) Tamburro CH, Greenberg RA. Identification of human toxicity and carcinogenicity by ethylene denvatives. In: Holmstead B, Lauwerys R, Mercier M, Roberfroid M, editors. Mechanisms of toxicity and hazard evaluation. New York: Elswonh/Nonh Hol land Biomedical Press 1980-319-34. (S) Marion MJ, Froment O, Trepo C. Activation of Ki-ras gene by point mutation in human liver angiosarcoma associated with vinyl chlonde exposure. Mol Carcinog 1991,4-450-A (9) Hollstem M, Manon MJ, Lehman T. Welsh J. Hams CC, Martel-Planche G, et al. p53 muta tions at A:T base pairs in angiosarcomas of vinyl chloride-exposed factory workers. Carcinogenesis 1994;15:1-3. (10) Devtvo I, Manon MJ, Smith SJ. Carney WP. Brandt-Rauf PW. Mutant c-Ki-ras p21 protein in chemical carcinogenesis in humans exposed to vinyl chlonde. Cancer Causes Control 1994:5.273-8. (11) Hollstem M, Sidransky D. Vogelstein B. Harns CC. p53 mutations in human cancers. Science 1991;253:49-53 (12) Levine AJ, Momand J, Finlay CA. The p53 tumour suppressor gene. Nature 1991351: 453-6. (13) Harris CC. Hollstem M. Clinical implications of the p53 tumor-suppressor gene [see com ment citations in Medline]. N Engl J Med 1993:329:1318-27. (14) Greenblatt MS. Bennett WP, Hollstem M. Harris CC. Mutations m the p53 tumor sup pressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res 1994;54: 4855-78. (15) Hollstein M, Rice K. Greenblatt MS, Soussi T, Fuchs R, Sorlte T, et al. Database of p53 gene somatic mutations in human tumors and cell lines. Nucleic Acids Res 1994;22:3551-5 1/6) Benchimol S, Pim D, Crawford L. Radioim munoassay of the cellular protein p53 in mouse and human cell lines. EMBO J 1982;1:1055-62. (17) Crawford LV, Pim DC, Lamb P. The cellular protein p53 in human tumours. Mol Biol Med 1984,2:261-72. (IS) Vogelstein B, Kinzler KW p53 function and dysfunction. Cell 1992;70:523-6. (19) Prives C, Manfredt JJ. The p53 tumor suppres sor protein; meeting review. Genes C v 1993:7.529-34. (20) Gerard P, Lavme AJ. A comparison of the biological activities of wild-type and mutant p53. FASEB J 1993:7-855-65 (21) Lane DP. On the expression of the p53 protein m human cancer. Mol Biol Rep 1994;19:23-9 (22) Ory K, Legros Y, Auguin C, Soussi T Analysis of the most representative tumourdenved p53 mutants reveals that changes in protein conformation are not correlated with toss of transactivation or inhibition of ceil proliferation. EMBO J 1994:13.3496-504, (23) Lane D. p53 and human cancers. Br Med Bull 1994;50:582-99. (24) Crawford LV, Pim DC, Bulbrook RD, Detec tion of antibodies against the cellular protein p53 in sera from patients with breast cancer Ini I Cancer 1982,30:403-8. (25) Caron de Fromentel C, May-Levin F. Mouriesse H, Lemerle J. Chandrase Karan K. May P. Presence of circulating antibodies against cellular protein p53 in a notable proportion of children with B-cell lymphoma IntJ Cancer 1987;39:185-9. (26) Davidoff AM, Iglehart JD, Marks JR. Immune response to p53 is dependent upon p53/H$70 vomplexes in breast cancer. Proc Natl Acad Sty US A 1992:89:3439-42. (27) Schlichtholz B. Legros Y, Gillet D. The im mune response to p53 in breast cancer patients is directed against immunodominant epitopes 1406 REPORTS Journal of the National Cancer Institute, Vol. 87, No. 18, September 20, 1995 ASI 000013942 unrelated to the mutational hot spot. Cancer Res 1992;52:6380-4. (28) Green JA, Mudenda B, Jenkins J. Leinster SJ, Tarunina M, Green B. et al. Serum p53 autoantibodies: incidence in familial breast cancer. Eur J Cancer 1994;30A:580-4. (29) Winter SF, Minna JD. Johnson BE, Takahashi T, Gaidar AF, Carbone DP. Development of antibodies against p53 in lung cancer patients appears to be dependent on the type of p53 mutation. Cancer Res 1992,52:4168-74 (JO) Labrecque S. Naor N, Thomas D. Matlashewski G. Analysts of the anti-p53 antibody response in cancer patients. Cancer Res 1993:53:3468-71 i jl) Schlichtholz B. Tredaniel J. Lubin R. Zalcman G, Hirsch A, Soussi T. Analysis of p53 an tibodies in sera of patients with lung car cinoma define immunodominant regions in the p53 protein. Br J Cancer 1994;69:809-16. (J2) Gurnee DG Jr. Travis WD, Tnvers GE. De Benedetti VM. Cawley H, Welsh JA. et ai. Gender comparisons in human lung cancer analysts of p53 mutations, anti-p53 serum an tibodies and C-erbB-2 expression. Carcino genesis 1995:16:993-1002. (33) Preudhomme C. Lubin R, Lepelley P, Vanrumbeke M, Fenaux P. Detection of serum anti p53 antibodies and their correlation with p53 mutations in myelodysplasnc syndromes and acute myeloid leukemia. Leukemia 1994:91589-91 (34) Voikmann M. Muller M, Hoffmann WJ, Meyer M, Hageistein J, Rath U. et al. The humoral immune response to p53 in patients with hepatocellular carcinoma is specific for malignancy and independent of the alphafetoprotetn status. Hepatology 1993:18.55965. (35) Winter SF. Sekido Y. Minna JD. Mclntire D, Johnson BE. Gazdar AF. et al. Antibodies against autologous tuL.cr cell proteins in patients with small-cell lung cancer' associa tion with improved survival [see comment citation in Medline], J Natl Cancer Inst 1993:85:2012-8 (36) Mudenda B. Green JA. Green B. Jenkins JR, Robertson L. Tarunina M, et al. The relation ship between serum p53 autoantibodies and characteristics of human breast cancer Br J Cancer 1994;6:1115-9 (37) Peyrat JP. Bonneterre J. Lubin R, Vanlemmens L, Fournier J, Soussi T, Prognostic sig nificance of circulating P53 antibodies in patients undergoing surgery for locoregional breast cancer. Lancet 1995:345:621-2. 138) Tnvers GE, Hams CC. Rougeot C. Dray F. Development and use of ultrasensitive enzymeimmunoassays. In Conn PM, editor. Neuroendocrine peptide methodology San Diego1 Academic Press, 1989:629-54. (39) Wang EH. cnedman PN, Pnves C. Tne munne p53 protein blocks replication of SV40 DNA in vitro by inhibiting the initiation func tions of SV40 large T antigen. Cell 1989:57.379-92 (40) Fnedman PN. Kern SE, Vogelstein B, Pnves C. Wild-type, but not mutant, human p53 proteins inhibit the replication activities of simian virus 40 large tumor antigen. Proc Natl Acad Set U S A 1990:87 9275-9. (41) Garfin DE. [mmunoblotting. In: Rose NR. de Macario EC. Fahey JL, Fnedman H, Penn GM, editors. Manual of clinical laboratory im munology. 4th ed. Washington, DC: American Society for Microbiology, 1992:47-51. (42) Felley-Bosco E, Weston A. Cawley HM, Ben nett WP. Hams CC Functional studies of a (43) Lehman t A. Bennett W^Metcalf RA^yJjuh M,EckerJ,ModaliRV.e,aLp53^:`t ras mutations, and p53-heat shock 70 protein complexes in human lung carcinoma cell tin,. Cancer Res 199l;5L 4090-6 (44) Thomas L8, Popper H. Pathology of angiosar coma of the liver among vmvl chlondepolyvinyl chloride workers. Ann N Y Acad Set 1975;246:268-77. (45) Cheng KC. Loeb LA. Genomic instability and tumor progression: mechanistic considerations Adv Cancer Res 1993.60:121-56. (46) Barbm A, Bartch H. Leconte P. Radman M. Studies on the miscoding properties of GVechenoadenine and 324"*-ethenocytosme, DNA reaction products of vinyl chloride metabo lites, during in vitro DNA synthesis. Nucleic Acids Res 1981:9:365-87. (47) Lubin R, Zalcman G, Bouchet L, Tredaniel J. Legros Y, Cazals D, et al. Serum p53 an tibodies as early markers of lung cancer Na ture Med 1995:1:701-2. (48) Hameltn R. Flejou JF. Muzeau F. Potet F Laurent-Puig P, Fekete F, et al TP gene muta tions and p53 protein immunoreactivitv m malignant and premaJignant Barrett's esopha gus. Gastroenterology 1994:107 1012-8. Notes Supported in pan by Public Health Service grant P20ES06832-01 from the National Institute of En vironmental Health Sciences. National Institutes ot Health, Department of Health and Human Services. Manuscript received February 3, 1995: revised June 23, 1995; accepted July 5, 1995. ^Vnnouncing the first comprehensive resource that encompasses all aspects of the alcohol and other drug (AOO) field--treatment, prevention, education, medical research, and the psychosocial disciplines: The Alcohol and Other Drug Thesaurus A Guide to Concepts and Terminology in Substance Abuse and Addiction The 1993 AOD Thesaurus contains: A hierarchical list that presents terms in a tree-like format. An alphabetical index. A multitude of scope notes. An extensive network, of cross- references. An abundance of lead-in terms. Use the AOD Thesaurus to: Search ETOH, the National Institute on Alcohol Abuse and Alcoholism's Alcohol and Alcohol Problems Science Database, on BRS. Free-text search other online databases for alcohol and other drug information. To order the AOD Thesaurus, write to the National Clearinghouse far Alcohol and Drug Information, P.O. Box 2345, Rockville, MD 20847-2345, or call 301-468-2600 or 1-800-729-6686 and ask far Inventory Item CR33, The price is $25, including shipping. Journal of the National Cancer Institute, Vol. 87, No 18, September 20, 1995 ASI 000013943 REPORTS 1407