Document 82XkE9KgKxgvNeaQp0jOywYOB

BIO-MEDICAL. RESEARCH DOCUMENT DESCRIPTION FORM 63 68 69 76 Duplicate.in all cards: --> year as-1961- File number [Right justify [Numeric only] Author(s), as Last Name FS (No Punctuation) and coden for journal as JAMA preceeded by one blank space 20 21 F/H. /c // 77fSSV/FZ. ^/3 .JS'M/f/r. /rt-- * 40 41 PtfPPtnZ , /V 7778 I1 Sub-Index Code 60 61 62 11 12 13 Title of Report: end with space-hyphen-hyphen-space. Follow with Index Terms. separated from each other with comma-space. Avoid other punctuation; do not abbreviate. 12 61 62 C 21 ' /fart?#f 22 23 /> #jr?r, fof 24 R&S 113975 'i S urce (Journal, Vol., NumberPages, Date ) 12 L/fauC&/ /l/t /"f7? 1 62 31 32 Brief Summary 12 10 SUMMARY: 61 62 61 62 63 64 i i R&S 113976 1"/ 7'; ' v'i \ Wir & THE LANCET. NOVEMBER24, 1979 Epidemiology of Cancer hepatic angiosarcoma associated WITH ANDROGENIC-ANABOLIC STEROIDS Henry Falk Hans Popper Louis B. Thomas Kamal G. Ishak Centerfor Disease Control, Atlanta, Georgia;National Cancer Institute, NationalInstitutes ofHealth, Bethesda, Maryland; Mount SinaiSchool ofMedicine ofthe City Unioertity ofNew York, Neto York; andArated Forces Institute ofPathology, Walter ReedArmy Medical Center, Washington, D.C., USA. Summary A retrospective epidemiological study of deaths from hepatic angiosarcoma (HAS) in the U.S. showed that during 1964-74 there were 168 such cases, of which 37 (22%) were associated with previously known causes (vinyl chloride, `Thoro trast', and inorganic arsenic) and 4 (3*1%) of the remaining 131 cases with the use of androgenic-anabolic steroids. It is suggested that the long-term use of andro genic-anabolic steroids is the fourth cause of HAS, the majority of cases still being of unknown etiology. More over, the presented cases serve as a link in a spectrum of hepatic disorders recently recognised to be caused by environmental agents such as vinyl chloride, arsenic, and thorotrast, and by contraceptive and anabolic steroids. Similar precursor stages, usually not recognised by clinical laboratory tests and consisting of areasofhyper plasia of hepatocytcs and sinusoidal cells and sinusoidal dilatation, lead potentially to hepatic adenoma, car cinoma, pcliosis, and angiosarcoma. INTRODUCTION The discovery that vinyl chloride monomer induced hepatic angiosarcoma (HAS) in polymerisation workers increased interest in this rare tumour.1'1 Recent studies of chemically induced and idiopathic cases have shown that, independent of aetiology, the tumour-precursor stage consists of areas of hyperplasia of hepatocytcs and sinusoidal cells associated with excess of reticulin and sinusoidal dilatation.4 With subsequent loosening of the lobular plate arrangement, HAS develops by increas ingly severe atypia of sinusoidal lining cells. This is often associated with bloody cysts, mainly in sarcoma tous areas. So far, androgenic-anabolic steroids have been impli cated in: cholestasis;5 hepatocellular tumours (benign, malignant, and equivocal),*4 which are almost always accompanied by conspicuous sinusoidal dilatation; and pcliosis--enlarged bloody cysts which develop in appar ently normal liver.*-1* Women taking oral contraceptive steroids have a small but increased risk for a range of hepatic lesions which includes benign adenoma,11-11 focal nodular hyperplasia,11 and hepatocellular cardnoma;14 these conditions are often associated with sinus oidal dilatation and vascular abnormalities. We describe here 4 cases of HAS in patients taking androgenic-anabolic steroids who were identified in a retrospective epidemiological study of HAS during 1964-74. METHODS The procedures have been presented in detail elsewhere.15 Briefly, the Center for Disease Control obtained information about cases of HAS diagnosed in the U.S. during 1964-74 in a variety of ways-which included a postal questionnaire to all pathologists in the country, a review of records at the Armed Forces Institute of Pathology (A.F.I.P.), and a review of death certificates with I.GD.-A. code 197-8 (liver cancer, not speci fied whether primary or secondary). Pathology reports and specimens were reviewed for each identified case (all nonA-F.l.P. material was seen by L.B.T. and FLP.), and the final study group consisted ofthe 168 histologically confirmed cases who had died during 1964-74. Permission to review the pa tient's medical records was tough: from thenert of kin who were asked by telephone details of the patient's occupation, residence, and possible exposure to chemicals. Sometimes phys icians, friends, or employers provided additional information. RESULTS Of the 168 cases of HAS in this study, 37 (22%) had. been exposed to previously reported etiological agents (vinyl chloride, Thorotrast', inorganic arsenic).1-15 Of the remaining 131-cases, 4 (3-1%) had documented his tories of treatment with androgenic-anabolic steroids. Because medical records were not always available and next of kin were often unaware of the patient's full drug HEPATIC ANGIOSARCOMA ASSOCIATED WITH ANDROGENIC-ANABOLIC STEROID USE Care no. 1 2 3 4 Age at death 58 56 52 52 Year of death ' Underlying. condition 1974 Irregular menses . 1973 Multiple myeloma 1971 Hypogonadism 1967 Osteoporosis Androgenic-anabolic ' steroids used Testosterone enaothate (`DdatestryT) Fluoxymcstcrone OHaloiestin*) - Testosterone Mcthyltestacterooe Other unidentified androgens Methandrostenolone Stanozolol ('Winttrol" Methyltestosterooe CGynetoneT Duration 2jr lOyrt 23 ri 15 exxf Latent period* 2 yr 13 >T 35 yr 2 vr Time dipxd from lira taking androgenic-anabolic steroids todiagnotii ofHAS. fNot used continuously. (Discontinued 3 year* before diagnosis ofHAS. - Reprinted by the U.S. DEPARTMENT OF HEALTH. EDUCATION, AND WELFARE PUBLIC HEALTH SERVICE from THE LANCET, November 24, 1979 V THE LANCET, NOVEMBER 24,1979 1121 history, the rates of androgenic-anabolic steroid use may be erroneously low. 4 Case-reports (Table) , Case /.--A Black, female English teacher had irregular menses for approximately 10 years, and in the early 1960s she was given norethynodrel 5 mg plus mestranol 0-075 mg (`Enovid-5') daily for 2-3 years. The treatment was then changed . to chlordiazeporide 10 mg plus water-soluble esterified oestrogen! 0-4 mg ('Menrium 10-4') daily, and in 1968 it was changed to 20 mg estradiol valerate ('Delestrogen') intramus cularly every month. From early 1971 she received instead 100 mg testosterone enanlhate ('Delatestryl') intramuscularly every 6 weeks. Menses ceased in July 1972.2 months later she had a raised alkaline phosphatase level of 236IU/1 but she was symptomless and physical examination was negative. The tes tosterone injections were discontinued in January, 1973. In March, 1973, the liver was enlarged and activities of both alkaline phosphatase <450 IU/1) and lactic dehydrogenase (LDH) <305 lU/ml) were high. A liver scan showed multiple small filling defects, and a needle biopsy specimen of the liver revealed angiosarcoma. She received chemotherapy for 5 months with only transient response and died in January, 1974, with gross ascites and hepatic failure. Case 2.--A White, male military officer had active pulmon ary tuberculosis in 1957 which was successfully treated with isoniazid, para-aminosalicylic add, and pyridoxine for 2 years. In 1958 he had multiple vertebral compression fractures. Pro tein electrophoresis and bone-marrow biopsy findings were consistent with multiple myeloma. Urethane therapy was started, but was discontinued by the patient after 4 months. In 1960 severe lumbar pain with radiation to the leg led to ther apy with cyclophosphamide (`Cytoxan') 50 mg/day until July, 1973, and fluoxymesterone (`Halotestin') 30-40 mg/day until July, 1970. In July, 1973, he had abdominal swelling, weak ness, a grossly enlarged liver, jaundice, asates, and cachexia. Bilirubin was 7-6 mg/di (direct, 5-2), serum glutamic oxaloUansaminase (SGOT) was 244 IU/ml, and LDH was 574 lU/ml. A liver scan showed multiple filling defects. He died 10 days after admission. HAS was first diagnosed at necropsy. There were periesophageal metastases. The bone-marrow also revealed multiple myeloma. Case 3.--A White man had a history of hypogonadism lor approximately 50 years. He was cryptotchid as a child aiid cor rective surgery at age 12 was followed by gonadal failure:' By 1944, at age 25, he had received, because of bilateral atrophic testes and poor development of secondary sexual character istics, hormone injections (not specified) for approximately 5 to 6 years, and methyltestosterone for 2 years, with limited ben efit. He continued to be given testosterone inegularly for several years. From the late 1950s androgenic hormone prep arations (mostly oral, but on occasion injectablcs) were given regularly. Specific information about the dosage and type of androgenic steroid given is not available. In June, 1971, he had nausea, vomiting, anorexia, and questionable hepatome galy. The total bilirubin was 4-1 mg/dl, alkaline phosphatase was 210 IU/1, LDH was 225 IU/ml, and SGOT was 130 IU/ml. Multiple filling defects were seen on hepatic scan, and a biopsy specimen taken at laparotomy showed HAS. In the absence of demonstrable metastases an orthotopic liver trans plantation was performed and immunosuppressive therapy was started. 2 months later liver function became abnormal, and during the final month of life septicaemia and uncorrectable acidosis developed. Necropsy showed widespread angiosarcoma that involved not- only the transplanted liver,- but also the lungs, adrenal glands, kidney, pancreas, spleen, bone-marrow, peritoneum, pleura, pericardium, lymph-nodes, and the ab dominal incision scar. The patient was a designer-draftsman for radio and telephone companies for 22 years. During 7 of hose years he-was`exposed to carbon tetrachloride (used for 'leaning instruments) for approximately 15 min/week in a well-ventilated room. In the early 1950s he had a small farm and was briefly exposed to arsenical pesticides about once a year. Case 4.--A White, male cook had a 19-year history of progressively severe back problems dating from an injury in 1948. Three lumbosacral fusion operations were not successful and compression fractures recurred. By early 1965, possibly even earlier, he had received anabolic steroids (methandrostenolone); information about the exact dates and dosage of this therapy was not available from the medical records. In October, 1965, because of multiple compression fractures and osteoporosis, he was given stanozolol (`Winstrol'), glutamic add, caldum, and vitamin D foj at least 3 months. In late 1966 the patient was placed on methyltestosterone plus ethinyloestradiol (`Gynetone'). In April, 1967, he had jaundice with a bilirubin of 8-5 mg/dl and an SGOT of 244 IU/ml. His liver was not enlarged. The methyltestosterone therapy was stopped. In May, 1967, progressive hepatic failure set in and he died 4 weeks later. At necropsy the liver was enlarged by multiple nodules ofHAS; metastatic angiosarcoma was present in lymph-nodes. He had consumed alcohol excessively for several years in early adulthood, but drank little alcohol in subsequent years. Except for the exposures noted in cases 3 and 4, the 4 patients did not abuse alcohol, and were not exposed to vinyl chloride, thorotrast, or arsenic. PathologicalFindings The lesions in the liven of these 4 patients were nearly identical and conformed with the previously reported morphological description of HAS.4 Briefly, the hepatic changes included complex precursor lesions and areas of roulticenuic angiosarcomatous transformation leading to many large nodules of angiosarcoma. The precursor lesions, best seen in portions of the liver not involved by angiosarcoma, consisted of areas with mixed nodular hyperplasia of hepatocytes and sinusoidal lining cells, with or without sinusoidal dilatation. In some parts of the specimens, angiosarcomatous transform ation of the sinusoidal lining cells was associated with conspicuous sinusoidal dilatation and disruption of Fig. 1--Trabecular angiosarcoma In case 3. Note trabecule of hyperplastic hepatocytes or fibrous tissue in widened Mood spaces. The trabecule* are covered by angiosarcoma cells which are also found in portal connective tissue. Hematoxylin andeosm,60x. Insert--Close-up of hyperplastic hepatocytes and angiosarcoma cells. 250*. Kui'-r- 3J CD 00 to -4 >1 1122 THE LANCET, NOVEMBER 24, 1979 hepatic plates to that large blood spaces formed (fig. 1). Hyperplastic hepatocytes were often found in plates two or more cells thick which were covered by angiosarcoma cells (fig. 1, insert). Both intralobular and trabecular patterns of angiosarcoma were contiguous with solid nodules of angiosarcoma. In case 3 a preoperative biopsy, as well as the resected liver, showed the .same changes.. described in detail for HAS;4 it has also recently been observed after long-term exposure to contraceptive steroids.'* The potential for progression of the precursor stage is now becoming apparent, as exemplified by the present report, although the magnitude of the risk and the distribution of lesions seen will undoubtedly vary for different etiological factors (fig. 2). DISCUSSION Further investigation of 168 deaths from HAS during 1964-74, of which 22% were associated with the 3 pre viously identified causative agents for this tumour, showed that 4 of the other 131 (3-1%) cases previously thought to be idiopathic had been exposed to large doses of androgenic-anabolic steroids. Although estimates of the extent of androgenic-anabolic steroid use are very crude because of scarcity of data, it is thought that con siderably fewer than 1% of the general U.S. population are likely to have had androgenic-anabolic steroids for long periods (personal communication. Drug Use Analy sis Branch, Division of Drug Experience, FDA). Thus, we suggest that these steroids arc a fourth potential cause of HAS although, given the small number of cases and the difficulty in precisely quantifying the expected number, this association remains to be confirmed. At this time the risk of HAS developing after the use of these steroids would appear to be less than after expo sure to vinyl chloride or thorotrast. The suspicion that prolonged therapy with andro genic-anabolic steroids may cause HAS arose because of observations that: 1) sinusoidal dilatation, often with ac tivation of sinusoidal lining cells, is present in the early stages of both steroid-induced and HAS precursor lesions;4*1* (2) hepatocellular proliferation, which is an essential part of the precursor lesion of HAS, is also seen after exposure to androgenic-anabolic steroids and can result in tumour formation; and (3) peliosis hepatis has been increasingly reported in those receiving these steroids: ' ' " Women who are taking oral contraceptive steroids have an analogous, although very small, risk for the de velopment of primarily benign hepatocellular tumours which are associated with sinusoids dilatation or vascu lar anomalies.'1*17 Recently, a case was reported of HAS developing in an elderly male treated for 12 years with diethylstilbostrol because of prostatic carcinoma,1* although it should be emphasised that so far an associ ation between oral contraceptive or oestrogenic hor mones and HAS has not been found in females. Although the suggested association between andro genic-anabolic steroids and HAS remains to be con firmed, it may be a link in a spectrum of liver diseases which have attracted much attention in the past decade ---these conditions, which appear related to the treat ments and environmental agents discussed above, arc hepatic adenoma, focal nodular hyperplasia (both often associated with intra-abdominal hemorrhage), hepato cellular carcinoma in the absence of cirrhosis and with potential for regression, peliosis hepatis, and HAS. The common feature in this spectrum of diseases seems to be the precursor lesion consisting of varying degrees of mixed hyperplasia of hepatocytes and sinus oidal cells and sinusoidal dilatation and which has been Fig. 2--Interrelated hepatic disorder! caused by new treatments and enviroommfal agents. Aetiologies! agents: AS--androgenic-anabolic steroids; CS--con traceptive steroids; VC--vinyl chloride; . 0--other unidentified; A--arsenic; T--thorotrast; F---all ofabove. The common morphological precursor stage, which may present as portal hypertension of the idiopathic or Banti syndrome type as'obscrvcd after exposure to vinyl chloride or arsenic, is often symptomless and is not relia bly detected by conventional hepatic laboratory tests. Thus its detection depends greatly on investigations such as liver biopsy. The dilatation of the sinusoids and other vascular lesions arc key features in the clinical manifestations ofthe entire group. They account for the hemorrhage into the hepatocellular tumours and subse quently into the peritoneal cavity seen with oral contra ceptives, the grossly visible bloody cysts in the usually multicenuic angiosarcoma, and the peliosis hepatis seen primarilywith androgenic-anabolic steroids. The androgenic-anabolic steroids, which appear to in duce hepatic adenoma and hepatocellular carcinoma (a case of cholangiocardnoma was also reported recently10) as well as peliosis and angiosarcoma, thus link the spec trum of hepatic disorders with potential for malignant transformation of both hepatocellular and sinusoidal cells. The most important purpose in calling attention to this range of conditions, however, is to encourage the search for the etiological factors (endogenous, medi cinal, and environmcntalrindustrial) which arc respon sible for the cases considered so far to be idiopathic. We thank Dr CUrlc W. Heath Jr,, Dr Glyn G. Caldwell, Dr John Herbert, Ms Joyce Cannon, Ms Alice Little, Mr Robert Albin, and Ms Renee Shirley of the Cotter for Disease Control; Dr Paul Lcavenon of the National Center for Health Statistics, and Dr Irving J. SclikofT of the Mount Sinai Hospital in New York City, for their help. We also thank the many physicians, pathologists, and tumour registrars who cooperated in this study. This study was partially supported by N1H-NIEHS Grant No. 2 P30KS0092806. Requests for reprints should be addressed to H.F., Chronic Diseases Division, Bureau of Epidemiology, Center for Disease Control, Public Health Service, U-S. Department of Health, Education, and Welfare, Atlanta, Georgia 30333, USA. References atfoot ofnext column THE LANCET, NOVEMBER24,1979 1123 I fi8 (/> 113979 DR FALK AND OTHERS'. REFERENCES 1. Creech JL, Johnson MN. Angiosarcoma of liver in the manufacture of |nl]h vinyl chloride.,7OecirpMrd 1974; It, 150-51. 2. Fill H, Creech JL, Heath CW, Jehmon MN, Key MM. Hepatic ditcare amoot worker, at a vinyl chloride polymerization plant. JAMA 1974; 210:59-6). 71. Spina, R, Kamineki R. Angiosarcoma of the Kver in vinyl chkxidcfrdyvinyl chloride worker,--1977 update of the NIOSH reenter. Occufi Med 1V7S; 10,427-29. 4. Popper H,Thomas LB. Telle, NC. Falk H, SdikoffIJ. Devetopment ofhepa tic anponremna in man iadoeed by vinyl chloride, thocount, and anenic--comparison with czki of unknown etiology. Am J Pathol 197*; 92,349-76. 5. Gordon BS, WolfJ, Krause T, Shai F. Peliocix hepatis atid cholestasis followin* administration of iWKtthandtolonc. Am 1 Che Patkal I960; IS, 156-6). 6. Johnson FL, Lemcr KG, Stepa] M, ei al. Association ofandrogenic-anabolic steroid therapy with development of hepatocellular carcinoma. Lancer 1972; i: 1273-76. 7. Psredinas FJ, Bull T8, Wcitaby D, Murray-Lyon IM. Hypoplasia and proItP* f hepatocyte, into hepatic vein, during long-term methyhestower- one therapy: possible relationships of them changes to the devetopment of pelioti, hepati, and liver tumours. Hutoparhology 1977; 1:225-46. *. Westaby D, Ogle SJ, Paradina, FJ, Randcll JB, Murray-Lyon IM. Liver damage from lonf-termmethylteitwcronc. Lancet 1977; ii: 261-63. 9.Hafhen SA, Boyer JL. I'clioti, hepati, associated with androgenic-anabelie steroid therapy---a severe form of hepatic injury. Ann interm Med 1974; *1,610-1*. 10. Nadcfl J, Koaek J. Pelioti, bepatis--iweive one, associated with oral androgen lherpy.Are* PatholLetMed 1977; 101,405-10. 11. Baam JK, Hohz F, Bookstein JJ, Klein EW. Possible association between benign hepatoma, and oral contraceptives. Lancet 1975; i:926-29. 12. Klatskin G. Hepatic tumours: possible relationship to use oforal contracep tives. Curmunhty 1977; 73:3*6-94. 13. Knowles DM, Casarelia WJ, Johnson PM, Volf M. The dinical, radiologic, and pathologic characterisation of benign hepatic neopiaa-ts--alleged ataociaiion with oral contraceptive,.Medicine 1978; 57,223-37. 14. Christophetsoo WM, Mays ET, Barrows G. HepaiooeUular carcinoma in youngwomen on oral contraceptives, Lancet 1978; u: 38-39. 15. Falk H, Tellcs NC, Ishak KG, Thomas LB, Popper H. Epidemiology ofTborotrast-itsdoetd hepatic angiosarcoma in the L'nited Stales- Environ Res 1979; M,65-73. 16. Popper H. Seliknff IJ, MaJtoni C, Squire RA, Thome, LB. Comparison of neoplastic hepatic lesions in man and ezperimental animals. In: Hiatt HH, Wauon JD, Cinne, JA, eds. Origin, of human canoer. Cold Spring Har bor Laboratory, 1977:13J9-82. 17. Sherlock S. Progress report: hepatic adenoma, ind oral contraceptives. Out 1975:16,735-36. I*. HodwUgeti C. Angiosarcoma of the liver associated with diethyktilbenrol. JAMA 1978;240:1510-11. 19. Thimg SN, Gerber MA. Precursor stage of hepatocellular neoplasm fallowmglongexposure tooral contraceptives. Hunt Pctkol(ia press',. 20. Sumncyer FW, Smith DH, Ishak KG. Anabolic steroid therapy and inirahepaticcholangiocarctnoma. Cancer 1979; 43:440--43.