s. although thioguanine is one of several agents with activity in the treatment of the chronic phase of chronic myelogenous leukemia, more objective responses are observed with myleran ® (busulfan), and therefore busulfan is usually regarded as the preferred drug.

megaly, thrombocytopenia, and oesophageal varices). histopathological features associated with this toxicity include hepatoportal sclerosis, nodular regenerative hyperplasia, peliosis hepatitis, and periportal fibrosis. thioguanine therapy should be discontinued in patients with evidence of liver toxicity as reversal of signs and symptoms of liver toxicity have been reported upon withdrawal. patients must be carefully monitored (see precautions, laboratory tests). early indications of liver toxicity are signs associated with portal hypertension such as thrombocytopenia out of proportion with neutropenia and splenomegaly. elevations of liver enzymes have also been reported in association with liver toxicity but do not always occur. the most consistent, dose-related toxicity is bone marrow suppression. this may be manifested by anemia, leukopenia, thrombocytopenia, or any combination of these. any one of these findings may also reflect progression of the underlying disease. since thioguanine may have a delayed effect, it is important to withdraw the medication temporarily at the first sign of an abnormally large fall in any of the formed elements of the blood. evaluate patients with repeated severe myelosuppression for thiopurine s-methyltransferase (tpmt) or nucleotide diphosphatase (nudt15) deficiency. tpmt genotyping or phenotyping (red blood cell tpmt activity) and nudt15 genotyping can identify patients who have reduced activity of these enzymes. patients with homozygous tpmt or nudt15 deficiency require substantial dosage reductions. bone marrow suppression could be exacerbated by coadministration with drugs that inhibit tpmt, such as olsalazine, mesalazine, or sulphasalazine. it is recommended that evaluation of the hemoglobin concentration or hematocrit, total white blood cell count and differential count, and quantitative platelet count be obtained frequently while the patient is on thioguanine therapy. in cases where the cause of fluctuations in the formed elements in the peripheral blood is obscure, bone marrow examination may be useful for the evaluation of marrow status. the decision to increase, decrease, continue, or discontinue a given dosage of thioguanine must be based not only on the absolute hematologic values, but also upon the rapidity with which changes are occurring. in many instances, particularly during the induction phase of acute leukemia, complete blood counts will need to be done more frequently in order to evaluate the effect of the therapy. the dosage of thioguanine may need to be reduced when this agent is combined with other drugs whose primary toxicity is myelosuppression. myelosuppression is often unavoidable during the induction phase of adult acute nonlymphocytic leukemias if remission induction is to be successful. whether or not this demands modification or cessation of dosage depends both upon the response of the underlying disease and a careful consideration of supportive facilities (granulocyte and platelet transfusions) which may be available. life-threatening infections and bleeding have been observed as consequences of thioguanine-induced granulocytopenia and thrombocytopenia. the effect of thioguanine on the immunocompetence of patients is unknown. pregnancy drugs such as thioguanine are potential mutagens and teratogens. thioguanine may cause fetal harm when administered to a pregnant woman. thioguanine has been shown to be teratogenic in rats when given in doses 5 times the human dose. when given to the rat on the 4th and 5th days of gestation, 13% of surviving placentas did not contain fetuses, and 19% of offspring were malformed or stunted. the malformations noted included generalized edema, cranial defects, and general skeletal hypoplasia, hydrocephalus, ventral hernia, situs inversus, and incomplete development of the limbs. there are no adequate and well-controlled studies in pregnant women. if this drug is used during pregnancy, or if the patient becomes pregnant while taking the drug, the patient should be apprised of the potential hazard to the fetus. women of childbearing potential should be advised to avoid becoming pregnant.

itiated to establish the etiology of the hepatic dysfunction. deterioration in liver function studies during thioguanine therapy should prompt discontinuation of treatment and a search for an explanation of the hepatotoxicity. administration of live vaccines to immunocompromised patients should be avoided.

olerability. ninety-six (59%) of 163 pediatric patients with previously untreated acute nonlymphocytic leukemia obtained complete remission with a multiple-drug protocol including thioguanine, prednisone, cytarabine, cyclophosphamide, and vincristine. remission was maintained with daily thioguanine, 4-day pulses of cytarabine and cyclophosphamide, and a single dose of vincristine every 28 days. the median duration of remission was 11.5 months. fifty-three percent of previously untreated adults with acute nonlymphocytic leukemias attained remission following use of the combination of thioguanine and cytarabine according to a protocol developed at the memorial sloan-kettering cancer center. a median duration of remission of 8.8 months was achieved with the multiple-drug maintenance regimen which included thioguanine. on those occasions when single-agent chemotherapy with thioguanine may be appropriate, the usual initial dosage for pediatric patients and adults is approximately 2 mg/kg of body weight per day. if, after 4 weeks on this dosage, there is no clinical improvement and no leukocyte or platelet depression, the dosage may be cautiously increased to 3 mg/kg/day. the total daily dose may be given at one time. the dosage of thioguanine used does not depend on whether or not the patient is receiving zyloprim (allopurinol); this is in contradistinction to the dosage reduction which is mandatory when purinethol ( mercaptopurine ) or imuran (azathioprine) is given simultaneously with allopurinol . procedures for proper handling and disposal of anticancer drugs should be considered. several guidelines on this subject have been published. 1-8 there is no general agreement that all of the procedures recommended in the guidelines are necessary or appropriate.

oprim ® (allopurinol). unlike purinethol (mercaptopurine) and imuran ® (azathioprine), thioguanine may be continued in the usual dosage when allopurinol is used conjointly to inhibit uric acid formation. less frequent adverse reactions include nausea, vomiting, anorexia, and stomatitis. intestinal necrosis and perforation have been reported in patients who received multiple-drug chemotherapy including thioguanine. hepatic effects liver toxicity associated with vascular endothelial damage has been reported when thioguanine is used in maintenance or similar long-term continuous therapy which is not recommended (see warnings and dosage and administration). this usually presents as the clinical syndrome of hepatic veno-occlusive disease (hyperbilirubinemia, tender hepatomegaly, weight gain due to fluid retention, and ascites) or signs and symptoms of portal hypertension (splenomegaly, thrombocytopenia, and esophageal varices). elevation of liver transaminases, alkaline phosphatase, and gamma glutamyl transferase and jaundice may also occur. histopathological features associated with this toxicity include hepatoportal sclerosis, nodular regenerative hyperplasia, peliosis hepatitis, and periportal fibrosis. liver toxicity during short-term cyclical therapy presents as veno-occlusive disease. reversal of signs and symptoms of this liver toxicity has been reported upon withdrawal of short-term or long-term continuous therapy. centrilobular hepatic necrosis has been reported in a few cases; however, the reports are confounded by the use of high doses of thioguanine, other chemotherapeutic agents, and oral contraceptives and chronic alcohol abuse.

in small amounts. intravenous administration of 35 s-6-thioguanine disclosed a median plasma half-disappearance time of 80 minutes (range: 25 to 240 minutes) when the compound was given in single doses of 65 to 300 mg/m 2 . although initial plasma levels of thioguanine did correlate with the dose level, there was no correlation between the plasma half-disappearance time and the dose. thioguanine is incorporated into the dna and the rna of human bone marrow cells. studies with intravenous 35 s-6-thioguanine have shown that the amount of thioguanine incorporated into nucleic acids is more than 100 times higher after 5 daily doses than after a single dose. with the 5-dose schedule, from one-half to virtually all of the guanine in the residual dna was replaced by thioguanine. tissue distribution studies of 35 s-6-thioguanine in mice showed only traces of radioactivity in brain after oral administration. no measurements have been made of thioguanine concentrations in human cerebrospinal fluid (csf), but observations on tissue distribution in animals, together with the lack of cns penetration by the closely related compound, mercaptopurine, suggest that thioguanine does not reach therapeutic concentrations in the csf. monitoring of plasma levels of thioguanine during therapy is of questionable value. there is technical difficulty in determining plasma concentrations, which are seldom greater than 1 to 2 mcg/ml after a therapeutic oral dose. more significantly, thioguanine enters rapidly into the anabolic and catabolic pathways for purines, and the active intracellular metabolites have appreciably longer half-lives than the parent drug. the biochemical effects of a single dose of thioguanine are evident long after the parent drug has disappeared from plasma. because of this rapid metabolism of thioguanine to active intracellular derivatives, hemodialysis would not be expected to appreciably reduce toxicity of the drug. thioguanine competes with hypoxanthine and guanine for the enzyme hypoxanthine-guanine phosphoribosyltransferase (hgprtase) and is itself converted to 6-thioguanylic acid (tgmp). this nucleotide reaches high intracellular concentrations at therapeutic doses. tgmp interferes at several points with the synthesis of guanine nucleotides. it inhibits de novo purine biosynthesis by pseudo-feedback inhibition of glutamine-5-phosphoribosylpyrophosphate amidotransferase—the first enzyme unique to the de novo pathway for purine ribonucleotide synthesis. tgmp also inhibits the conversion of inosinic acid (imp) to xanthylic acid (xmp) by competition for the enzyme imp dehydrogenase. at one time tgmp was felt to be a significant inhibitor of atp:gmp phosphotransferase (guanylate kinase), but recent results have shown this not to be so. thioguanylic acid is further converted to the di- and tri-phosphates, thioguanosine diphosphate (tgdp) and thioguanosine triphosphate (tgtp) (as well as their 2′-deoxyribosyl analogues) by the same enzymes which metabolize guanine nucleotides. thioguanine nucleotides are incorporated into both the rna and the dna by phosphodiester linkages and it has been argued that incorporation of such fraudulent bases contributes to the cytotoxicity of thioguanine. thus, thioguanine has multiple metabolic effects and at present it is not possible to designate one major site of action. its tumor inhibitory properties may be due to one or more of its effects on (a) feedback inhibition of de novo purine synthesis; (b) inhibition of purine nucleotide interconversions; or (c) incorporation into the dna and the rna. the net consequence of its actions is a sequential blockade of the synthesis and utilization of the purine nucleotides. the catabolism of thioguanine and its metabolites is complex and shows significant differences between humans and the mouse. in both humans and mice, after oral administration of 35 s-6-thioguanine, urine contains virtually no detectable intact thioguanine. while deamination and subsequent oxidation to thiouric acid occurs only to a small extent in humans, it is the main pathway in mice. the product of deamination by guanase, 6-thioxanthine is inactive, having negligible antitumor activity. this pathway of thioguanine inactivation is not dependent on the action of xanthine oxidase, and an inhibitor of that enzyme (such as allopurinol) will not block the detoxification of thioguanine even though the inactive 6-thioxanthine is normally further oxidized by xanthine oxidase to thiouric acid before it is eliminated. in humans, methylation of thioguanine is much more extensive than in the mouse. the product of methylation, 2-amino-6-methylthiopurine, is also substantially less active and less toxic than thioguanine and its formation is likewise unaffected by the presence of allopurinol. appreciable amounts of inorganic sulfate are also found in both murine and human urine, presumably arising from further metabolism of the methylated derivatives. in some animal tumors, resistance to the effect of thioguanine correlates with the loss of hgprtase activity and the resulting inability to convert thioguanine to thioguanylic acid. however, other resistance mechanisms, such as increased catabolism of tgmp by a nonspecific phosphatase, may be operative. although not invariable, it is usual to find cross-resistance between thioguanine and its close analogue, purinethol (mercaptopurine). metabolism and genetic polymorphism several published studies indicate that patients with reduced tpmt or nudt15 activity receiving usual doses of mercaptopurine, accumulate excessive cellular concentrations of active 6-tgns, and are at higher risk for severe myelosuppression . in a study of 1028 children with all, the approximate tolerated mercaptopurine dosage range for patients with tpmt and/or nudt15 deficiency on mercaptopurine maintenance therapy (as a percentage of the planned dosage) was as follows: heterozygous for either tpmt or nudt15, 50-90%; heterozygous for both tpmt and nudt15, 30-50%; homozygous for either tpmt or nudt15, 5-10%. approximately 0.3% (1:300) of patients of european or african ancestry have two loss-of-function alleles of the tpmt gene and have little or no tpmt activity (homozygous deficient or poor metabolizers), and approximately 10% of patients have one loss-of-function tpmt allele leading to intermediate tpmt activity (heterozygous deficient or intermediate metabolizers). the tpmt*2, tpmt*3a, and tpmt*3c alleles account for about 95% of individuals with reduced levels of tpmt activity. nudt15 deficiency is detected in <1% of patients of european or african ancestry. among patients of east asian ancestry (i.e., chinese, japanese, vietnamese), 2% have two loss-of-function alleles of the nudt15 gene, and approximately 21% have one loss-of-function allele. the p.r139c variant of nudt15 (present on the *2 and *3 alleles) is the most commonly observed, but other less common loss-of-function nudt15 alleles have been observed. consider all clinical information when interpreting results from phenotypic testing used to determine the level of thiopurine nucleotides or tpmt activity in erythrocytes, since some coadministered drugs can influence measurement of tpmt activity in blood, and blood from recent transfusions will misrepresent a patient’s actual tpmt activity.