1050 9. Komfield H, Riedel N, Viglianti GA, et al. Cloning of HTLV-4 and its relation to simian and human immunodeficiency viruses. Nature 1987; 326: 610-13. 10. Wong Staal F, Gallo RC. Human T-cell lymphotropic viruses. Nature 1985; 317: 395-403. 11. Lasky LA, Naknamura G, Smith DH, et al. Delineation of a region of the HIV-1 gp120 glycoprotein critical for interaction with the CD4 receptor. Cell I 1987; 50: 975-85. 12. Patterson S, Oxford JB. Early interactions between animal viruses and the host cell. Vaccine 1986; 4: 79-89. 13. Oudin J, Michael M. Une nouvelle forme d’allotypie des globulines der serum de lapin, apparement leie a la function et a la specificite anticorps. CR Seances Acad Sci 1963; 257: 805-08. 14. Kunkel HC, Mannick M, Williams RC Individual antigenic specificity of isolated antibodies. Science 1963; 257: 1218-19. 15. Jeme NK. Towards a network theory of the immune system. Ann Immunol (Pans) 1974, 125c: 373-89. 16. NisonoffA, Lamyoi E. Implications of the presence of an internal image of the antigen in anti-idiotypic antibodies: Possible application in vaccine production. Clin Immunol Immunopathol 1981, 21: 397-406. 17. Roitt IM, Cooke AM, Male DK, et al. Idiotypic networks and their possible exploitation for manipulation of the immune response. Lancet 1981; i: 1041-44 18. Gaulton GN, Greene MI. Idiotypic mimicry of biological receptors. Ann Immunol 1986; 4: 253-80. 19. Chanh TC, Dreesman G, Kennedy RC. Monoclonal anti-idiotypic antibody mimics the CD4 receptor and binds HIV Proc Natl Acad Sci USA 1987; 84: 3891-95 20 McDougal JS, Nicholson JKA, Cross G, Cort S, Kennedy SP, Mawle AC. Binding of HIV to the CD4 molecule J Immunol 1986; 137: 2937-44. 21. Schick MR, Dreesman GR, Kennedy RC. Induction of an anti-hepatitis B surface antigen response in mice by non internal image (Ab-2&agr;) anti-idiotypic antibodies. J Immunol 1987; 138: 3419-25. 22. Kennedy RC, Dreesman GR, Chanh TC, et al. Use of resin bound peptide for identifying a neutralising antigenic determinant associated with the HIV envelope. J Biol Chem 1987; 262: 5769-74. 23. Dreesman GR, Kennedy RC. Anti-idiotypic antibodies Implications of internal image based vaccines for infectious diseases. J Infect Dis 1985; 151: 761-65. 24. Uytdehaag FGCM, Bunschoten H, Weijer K, Osterhause ADME. From Jenner to Jeme: Towards idiotypic vaccines. Immunol Rev 1986; 90: 93-113. 25. Kennedy RC, Eichberg JW, Lanford RE, Dreesman GR. Anti-idiotypic antibody vaccine for type B viral hepatitis in chimpanzees. Science 1986; 232: 220-23 26. Benjamin RJ, Waldmann H. Induction of tolerance by monoclonal antibody therapy Nature 1986; 320: 449-51. 27. Bona CA, Hohler H. Anti-idiotypic antibodies and internal images in monoclonal and anti-idiotpyic antibodies. In: Venter JC, Fraser CM, Lindstrom J, eds. Probes for receptor structure and function. New York: Alan Liss, 1984: 141-49. 28. Jeme NK, Roland J, Cazeneve PA. Recurrent idiotopes and internal images EMBO J 1982; i: 243-50 29. Dalgleish AG. Antiviral strategies and vaccines against HIV J Coll Physicians 1986; 20: 258-67. OESOPHAGEAL VARICES ASSOCIATED WITH BUSULPHAN-THIOGUANINE COMBINATION THERAPY FOR CHRONIC MYELOID LEUKAEMIA NIGEL S. KEY PAULINE M. EMERSON NORMAN C. ALLAN PETER M. A. KELLY ROGER W. G. CHAPMAN JAMES O’D. MCGEE Departments of Haematology and Gastroenterology, and Nuffield Department of Pathology, John Radcliffe Hospital, Oxford; and MRC Clinical and Population Cytogenetics Unit, Western General Hospital, Edinburgh Summary 5 patients receiving continuous busulphan and 6-thioguanine for chronic myeloid leukaemia (CML) were found to have oesophageal varices associated with abnormal liver function tests. 3 of these cases presented with gastrointestinal haemorrhage and 1 patient died. The 2 other cases had varices discovered at endoscopy. Nodular regenerative hyperplasia (NRH) of the liver was identified as the cause of portal hypertension in the 4 patients on whom liver biopsies were done. The administration of busulphan and thioguanine in combination is likely to be associated with the development of NRH, with portal hypertension and oesophageal varices occurring in a substantial proportion of cases. Introduction NODULAR regenerative hyperplasia (NRH) of the liver is rare and is characterised by diffuse nodules of regenerative hepatocytes. The nodules are not separated by fibrous septa, which distinguishes NRH from cirrhosis, and there is no progression to cirrhosis. Definitive histological diagnosis from a liver biopsy specimen may be difficult and a wedge biopsy may sometimes be required." NRH may be clinically silent or present with manifestations of portal hypertension (oesophageal varices, ascites, and spleno- megaly). Liver function tests may be normal or show slightly increased alkaline phosphatase (ALP) and y-glutamyl- transferase (y-GT) values.5,6. Recently 3 patients (cases A-C) with chronic myeloid leukaemia (CML) in chronic phase presented with bleeding oesophageal varices. Their liver biopsies revealed NRH. All 3 had been receiving long-term continuous busulphan and 6-thioguanine in the doses recommended in the Medical Research Council’s (MRC’s) CML trial protocol. These cases prompted a search for occult varices in the other 2 patients (cases D and E) with CML in this hospital who were on the same regimen. Both these patients were found to have lower oesophageal varices and in 1 case liver biopsy demonstrated NRH. The object of the MRC-CML trial is to compare busulphan alone with the combination regimen of busulphan (in a lower dose) and thioguanine (BU-TG). Study of busulphan in combination with 6-mercaptopurine, an anti-metabolite closely related to thioguanine, suggested that remission was achieved more rapidly with less risk of toxicity than with busulphan alone.7 In addition, the trial aims to identify whether the BU-TG combination has a beneficial effect on survival. Patients randomised to this arm receive 2 mg busulphan plus 80 mg thioguanine by mouth daily for 5 days per week until the total leucocyte count falls to 10-20 x 109/1. Alterations are then made in the number of days per week that the drugs are given with the aim of maintaining the white-blood-cell count (WBC) in the range 5-10 X 109/1. Case-reports Case A.-A 42-year-old man presented with a 4-month history of malaise and weight loss. Examination revealed splenomegaly of 12 cm. Haemoglobin (Hb) was 8-2 g/dl, WBC 397 x 109/1, and platelets 991 x 109/1. Peripheral blood and bone marrow films were typical of CML in chronic phase, and cytogenetic analysis showed a Philadelphia (Ph1) chromosome. Liver function tests were normal and his alcohol intake was restricted to 4 pints of beer per week. He was on no regular medication. He was enrolled into the BU-TG arm of the MRC trial. He entered remission after I weeks with no palpable splenomegaly and a WBC of 9-8 x 109/1. About 6 months after diagnosis, liver function was abnormal for the first time with an ALP of 389 IU/1 (normal 100-300). 17 months after diagnosis he was admitted after an episode of haematemesis and malaena. The spleen was palpable (4 cm) for the first time since entering remission. Liver function tests were: aspartate aminotransferase (AST) 55 IU/1 (normal 10-35), y-GT 79 IU/1 (15-40) and ALP 582 IU/1. Hepatitis B surface antigen (HBsAg) was not detected and clotting studies were normal. Upper gastro-intestinal endoscopy showed extensive, bleeding, lower oesophageal varices. Abdominal ultrasound failed to detect any thrombosis in the hepatic, portal, or splenic veins but a large solid mass arising from the right kidney was identified. Intravenous urography and renal angiography suggested a hypemephroma. At laparotomy, the liver was of normal size and "a little irregular but without frank cirrhotic change". The right kidney (845 g) and spleen (1385 g) were removed. Histology confirmed a renal adenocarcinoma. An intra-operative biopsy of the