Clinical Use and Practical Application of TPMT Enzyme and 6-Mercaptopurine Metabolite Monitoring in IBD Ernest G. Seidman, MD

Transcription

1 GENETIC TESTING FOR IBD Clinical Use and Practical Application of TPMT Enzyme and 6-Mercaptopurine Metabolite Monitoring in IBD Ernest G. Seidman, MD Division of Gastroenterology, Hepatology, and Nutrition, Faculty of Medicine, University of Montreal, Montreal, Canada 6-mercaptopurine (6-MP) and its parent drug azathioprine (AZA) have been proven to be effective for both steroid-dependent and chronically active, or steroid-resistant inflammatory bowel disease, as well as for the prevention of relapse. Concerns about toxicity, delayed onset of action, and therapeutic failure (1 out of 3 patients) have restricted their use. Recent pharmacogenetic advances have led to the development of novel strategies to optimize and individualize therapy with AZA and 6-MP, maximizing efficacy while minimizing toxicity. We have defined a range of optimal therapeutic 6-MP metabolite levels, as well as an association of metabolite levels with medication-induced toxicity and the genotype of the main catabolic enzyme, thiopurine methyltransferase (TPMT). Measurement of 6-MP metabolite levels and TPMT molecular analysis provide clinicians with useful tools for optimizing therapeutic response to 6-MP/AZA, as well as for identifying individuals at increased risk for druginduced toxicity. [Rev Gastroenterol Disord. 2003;3(suppl 1):S30 S38] 2003 MedReviews, LLC Key words: Crohn s disease Ulcerative colitis Therapeutic outcomes Azathioprine 6-mercaptopurine Pharmacogenetics Adherence It is well established that the immunomodulatory agents 6-mercaptopurine (6-MP) and its pro-drug azathioprine (AZA) have significant beneficial effects on the long-term clinical course of inflammatory bowel disease (IBD). Their efficacy has been demonstrated in both pediatric and adult patients for the treatment of key, specific indications (summarized in Table 1). Placebo-controlled trials have demonstrated that AZA and 6-MP improve long-term remission rates for both ulcerative colitis and Crohn s disease (CD). 1 Significant steroid-sparing effects S30

2 TPMT Enzyme and 6-Mercaptopurine Table 1 Indications for the Use of Azathioprine/ 6-Mercaptopurine in IBD Steroid-dependent disease (steroid-sparing effect)* Steroid-resistant disease (induction of remission)* Relapse prevention (maintenance of remission)* Perianal disease Fistulizing disease Prevention of post-operative recurrence IBD, inflammatory bowel disease. * Evidence based on randomized, controlled trials *P <.01 * Steroid dependence * Steroid resistance * Sustained remission 6-MP Placebo Figure 1. 6-mercaptopurine (6-MP) prevents relapse in Crohn s disease. The results are summarized for this multicenter trial, in which pediatric patients with new onset of Crohn s disease were randomized to receive 6-MP (1.5 mg/kg/day) or placebo after induction of remission with prednisone. After 18 months, the 6-MP group had a significantly greater percentage of patients with a sustained remission, off corticosteroids. Furthermore, steroid dependence and resistance were also avoided. Adapted from Markowitz et al. 2 have also been observed. In one key, randomized, placebo-controlled trial in pediatric patients with CD, 2 6-MP (1.5 mg/kg/day) significantly reduced relapse rates and steroid dependence and resistance over an 18-month period (Figure 1). This study, along with others, suggests that although steroids often induce remission, they do not improve overall, long-term outcomes of the course of the disease. It can be argued that immunosuppressive therapy using AZA or 6-MP should be considered for the long-term maintenance of remission, to prevent steroid dependency or resistance. 3 The induction of remission of steroid-refractory, active CD may also be elicited using AZA or 6-MP. 1 The pooled odds ratio for response of fistulae (4.6) is also significantly higher than with placebo. Furthermore, it has been shown that postoperative recurrence is more often prevented using 6-MP, compared with 5-aminosalicylates (5-ASA) or placebo. Despite the increasingly widespread use of these drugs and the generally acceptable safety profiles, some patients develop adverse reactions to these agents. 4 6 Clinicians express concerns about short- and long-term potential adverse effects, particularly with respect to myelotoxicity and hepatotoxicity. 7 This article reviews the recent evidence-based approach using pharmacogenetic methodology and metabolite monitoring to individualize therapy with these agents, thereby enhancing therapeutic efficacy while minimizing toxicity (Table 2). Pharmacogenetics: The Clinical Utility of TPMT Genotyping In search of potential explanations for therapeutic failures in approximately 1 out of 3 IBD patients treated with AZA or 6-MP, we sought to determine whether inter-individual differences in the metabolism of these drugs might be involved. 8 The immunosuppressive effects of AZA and 6-MP are mediated via their intracellular conversion along two competing pathways (Figure 2) to yield the nucleotide metabolites 6-thioguanine (6-TG) and 6-methylmercaptopurine (6-MMP). 6-MMP is a catabolic metabolite and is largely the product of the activity of the thiopurine methyltransferase (TPMT) enzyme. Genetically determined variations in the activity of TPMT can lead to differences in the likelihood of response to and toxicity from 6-MP and AZA. 9 Clinical studies have established an inverse correla- Table 2 Pharmacogenetics of Purine Analogue Therapy in IBD: An Evidence-Based Approach Knowledge of TPMT genotype: Reduces the risk of leukopenia Allows tailored" starting dose Reduces time to response Knowledge of TPMT genotype: Identifies the reason for most nonresponders Identifies nonadherence Identifies cause of nonidiosyncratic adverse events (hepatic + bone marrow) Increases response rate IBD, inflammatory bowel disease; TPMT, thiopurine methyltransferase. S31

3 TPMT Enzyme and 6-Mercaptopurine continued tion between red blood cell (RBC) TPMT activity and accumulation of the 6-TG metabolite in erythrocytes of patients treated with 6-MP and AZA. 10 Patients with low levels of TPMT activity thus accumulate higher levels of the anabolic, immunosuppressive 6-TG metabolite. Consequently, they are at a much higher risk of myelotoxicity after receiving standard doses of 6-MP and AZA than those individuals with high enzyme activity. Determining polymorphisms of TPMT by genotyping has been shown to be a clinically useful and costeffective strategy to help clinicians identify patients at risk for toxicity from 6-MP and AZA Polymorphisms of the TPMT gene can be detected using the polymerase chain reaction restriction length polymorphism method, performed with a sample of genomic DNA extracted from peripheral blood leukocytes. Among IBD patients, as in control or nondiseased populations, approximately 11% of individuals have a heterozygous genotype, most frequently involving the TPMT-3A allele. 16 Such patients, heterozygous for a mutation of TPMT, have approximately half of the capacity to methylate 6-MP, thus shunting much more of the drug toward the phosphoribosylation pathway and 6-TG (Figure 2). They require smaller drug doses to optimize therapeutic potential while avoiding myelotoxic levels of 6-TG (Table 3). Patients with homozygous mutations of TPMT (0.3% of the population) are at extreme risk of very severe leukopenia with potential septic complications and should not be exposed to these AZA XO TPMT 6-TU 6-MP 6-MMP Circulation drugs. The risk of AZA or 6- MP induced leukopenia can be avoided by determining TPMT genotype or by assaying for enzymatic phenotype, measuring the enzyme activity in vitro. 15 TPMT genotyping or assaying enzymatic activity represent practical, reliable, and cost-effective screening TPMT genotyping can identify patients with very low or absent TPMT enzyme activity, in whom thiopurine drugs should be avoided. methods of identifying patients at very high risk for developing severe, potentially fatal bone marrow toxicity from thiopurine drugs. It is strongly recommended that TPMT genotype or phenotype be determined before the initiation of AZA or 6-MP therapy in patients with IBD or rheumatic disorders These diagnostic tools 6-TG nucleotides HPRT TPMT 6-TImP 6-MMP ribonucleotides Intracellular DNA RNA Purine synthesis Figure 2. Metabolic pathways of 6-mercaptopurine (6-MP) metabolism. Azathioprine (AZA) is rapidly converted to 6-MP by a nonenzymatic process. Initial 6-MP transformations occur along competing catabolic (XO, xanthine oxidase; TPMT, thiopurine methyltransferase) and anabolic (HPRT, hypoxanthine phosphoribosyltransferase) enzymatic pathways. The latter intracellular enzyme (dashed line) transforms the drug into 6-thioguanine nucleotides (6-TG), which have been shown to be the most important factor correlated with treatment efficacy. TPMT converts the drug into 6-methylmercaptopurine ribonucleotides (6-MMP). Patients heterozygous for a mutant allele of TPMT will convert a higher proportion of the drug into 6-TG. This translates into a higher success rate, but with an increased risk of myelosuppression. Patients with excessive TPMT activity, on the other hand, generate too little 6-TG with overabundant 6-MMP levels. They are at increased risk of hepatotoxicity, and much less likely to respond to thiopurine therapy. also identify patients who are heterozygotes and who thus require substantial reduction in doses with careful monitoring of drug metabolite levels to avoid bone marrow toxicity. Clinicians should be aware, however, that allelic variants of TPMT do not explain all leukopenic events, and that TPMT measurements cannot replace the need for continued monitoring of leukocyte counts in patients treated with AZA or 6-MP. Clinical Utility of Monitoring Metabolite Levels The studies reviewed above established that TPMT genotyping can identify patients with very low or absent TPMT enzyme activity, in whom thiopurine drugs should be avoided, as well as heterozygous patients with intermediate levels of enzyme activity, in whom substantially lower doses of AZA or 6-MP S32

4 TPMT Enzyme and 6-Mercaptopurine Table 3 Genetic Determinants of Thiopurine Methyltransferase Activity Gene Frequency Enzyme Activity Allele 89% Normal to high TPMT H /TPMT H 11% Intermediate TPMT H /TPMT L 0.33% Low to Absent TPMT L /TPMT L If heterozygote: approximately 50% activity, yielding less 6-MMP and more 6-TG. Very likely to respond to therapy; however, much higher risk of myelotoxicity; requires reduction in dose (33% of usual). If homozygous mutation: minimal activity, yielding negligible 6-MMP and high 6-TG. Very high risk for severe myelotoxicity; drug contraindicated. If wild type: normal to high activity, yielding normal to excessive 6-MMP levels; variable outcome. TPMT, thiopurine methyltransferase; 6-MMP, 6-methylmercaptopurine; 6-TG, 6-thioguanine. should be administered (Table 3). Our next research goal was to determine which metabolite best predicts a favorable outcome in IBD patients treated with AZA or 6-MP, and whether a reliable therapeutic window" could be established for a particular metabolite. In a study among pediatric patients, 17 we found that among the 9% of patients who were heterozygous for the TPMT allele, all achieved a clinical response. On the other hand, only 46% of patients homozygous wild type" for TPMT achieved a complete clinical response. Of the remaining 45% of patients, two thirds did not respond and one third were partial responders. We then analyzed patients blood samples for erythrocyte levels of 6-MMP and 6-TG after at least 4 months of therapy, on a dose that was stable for at least 1 month. Overall, the only factor that correlated with successful outcome and complete clinical response was the intracellular 6-TG level. 17 Erythrocyte 6-TG levels were significantly higher among clinical responders than among nonresponders (Table 4). Quartile analysis showed that the best probability of treatment response was not significantly increased until RBC 6-TG levels were greater than 235 pmol/ RBC. Moreover, the odds ratio of a therapeutic response for a 6-TG level higher than the cutoff of 235 was an impressive 5.0 (95% confidence interval, ; P <.001). Logistical regression analysis confirmed that intracellular 6-TG levels were significantly associated with a complete therapeutic response, independent of any other confounding factors, including drug dose (P =.002). Thus, the patient s individual metabolism of the drug was the key factor, rather than the dose per se. Although some patients will respond at 6-TG levels below the cutoff (235 pmol/ RBC), the likelihood of response increases at higher 6-TG levels (Figure 3). Indeed, inadequate dosing is the most common cause of treatment failure (Table 5). Measuring drug metabolite levels and adjusting Measuring drug metabolite levels and adjusting the dose to achieve 6-TG levels above the therapeutic threshold can convert a treatment failure into a responder. the dose to achieve 6-TG levels above the therapeutic threshold can convert a treatment failure into a responder. 18 The vast majority of clinical responses (92%) occurred in the absence of leukopenia, demonstrating that leukopenia is not a reliable Table 4 6-TG Levels Reflect Therapeutic Efficacy Laboratory 6-TG Concentration 6-TG Concentration (IBD population) in Responders in Nonresponders P Reference Seidman, University of Montreal (pediatric) <.001 Dubinsky et al, Cuffari, Johns Hopkins (adults) <.001 Cuffari et al, TG, 6-thioguanine. S33

5 TPMT Enzyme and 6-Mercaptopurine continued Table 5 Reasons for Failure of Azathioprine/6-Mercaptopurine Therapy in IBD Explanation % of Failures 6-TG Level* 6-MMP Level* 6-MMP/6-TG Ratio Inadequate dose 74 Low (<230) Low (<5700) Normal (5 25/1) Predominant TPMT 15 Low (<230) High (>5700) High (30 100/1) Lack of adherence 6 Low (<<230) Low (<<5700) Normal (5 25/1) Drug failure (ineffective) 5 Normal ( ) Normal (<5700) Normal (5 25/1) IBD, inflammatory bowel disease; 6-MMP, 6-methylmercaptopurine; 6-TG, 6-thioguanine; TPMT, thiopurine methyltransferase. *pmol/ RBC. Ratio of less than 3/1 suggests TPMT deficiency. marker of therapeutic response and is definitely not required to control IBD. A comparable study in adult IBD patients 18 revealed strikingly similar 6-TG levels among responders as observed in our pediatric studies (Table 4). In addition to under-dosing, other reasons for treatment failure with inadequate 6-TG levels can be readily identified by serial metabolite measurement (Table 5). These include nonadherence to therapy (either absent or inconsistent adherence) and aberrant metabolism with excessive TPMT activity. In the latter situation, excessive TPMT activity generates high 6-MMP levels, with inadequate production of 6-TG. Dose escalation in that situation usually fails to substantially increase 6-TG levels, whereas 6-MMP levels rise, often to the point of inducing hepatotoxicity, 19 as discussed below. As shown in Table 5, the probability of successful treatment can be estimated by simply determining the metabolite ratio 6- MMP/6-TG. Patients with normal TPMT genotype and enzyme activity typically have 6-MMP/6-TG ratios of 10 15/1. The subgroup of individuals with normal genotype but excessive TPMT enzyme activity usually have 6-MMP/6-TG ratios of >30/1 and often as high as 100/1 (Table 5). Patients heterozygous for TPMT usually have low ratios, producing 6-TG Levels (pmol/ RBC) Risk of myelotoxicity (>450) almost as much 6-TG as 6-MMP. In the latter scenario, response to therapy is usually excellent, although myelotoxic levels of 6-TG may occur unless the drug dose is adjusted downward. Finally, we sought to determine whether toxicity to AZA or 6-MP correlated with metabolite levels. In the subgroup of patients with hepatotoxicity (elevated transaminases), intracellular 6-MMP levels were significantly higher (Figure 4). Levels greater than 5700 pmol/ RBC increased the likelihood of hepatotoxicity (elevated transaminases) from 3% to 18%. Myelotoxicity, on the other hand, correlates with high levels of 6-TG, usually exceeding 450 pmol/ RBC (Figure 3). Figure 3. 6-Thioguanine (6-TG) ribonucleotide levels correlate with outcome of 6-MP or azathioprine therapy in IBD. 6-TG levels are measured by high-performance liquid chromatography in erythrocytes (red blood cells, RBC) in lieu of leukocytes, for ease of assay. Results are expressed as pmol/ RBC. Significantly increased efficacy is achieved when RBC 6-TG levels are greater than 235 pmol/ RBC. Most treatment failures have 6-TG levels lower than this threshold. However, 6-TG levels > 450 pmol/ RBC are associated with an increased risk of myelotoxicity. Optimal therapeutic response range ( ) Suboptimal response S34

6 TPMT Enzyme and 6-Mercaptopurine 6-MMP (pmol/ RBC) Nongenetic Determinants of TPMT Activity In addition to the common polymorphisms of the TPMT gene, other factors are known to influence its enzyme activity. We have observed that its activity is lower in postpubertal males than in females, suggesting possible hormonal influences. TPMT activity has been reported to be higher in uremic patients. Drugs that influence substrate availability include allopurinol and methotrexate. The former is contraindicated in patients receiving AZA or 6-MP, as allopurinol blocks xanthine oxidase, the enzyme Hepatotoxicity risk zone Mo 1 Mo 2 Mo 3 Increased ALT & AST Normal ALT & AST Qtr Qtr Qtr Qtr Dose: /75 50/75 Figure 4. High 6-methylmercaptopurine (6-MMP) metabolite levels are associated with hepatotoxicity (elevated transaminases) in inflammatory bowel disease. 6-MMP levels above 5700 pmol/ RBC are associated with an increased risk of hepatotoxicity (18% vs 3%). Decreasing the dose of the drug (in the example shown from 75 mg daily to 50 mg alternating with 75 mg daily) will normally result in decreased 6-MMP levels and normalization of serum transaminases within a few weeks. However, the 6-thioguanine level may fall below the threshold of the target range. ALT, alanine aminotransferase; AST, aspartate aminotransferase. responsible for the low bioavailability of these drugs due to its first-pass metabolism to the biologically inactive 6-thiouracil metabolite (Figure 2). Drugs that inhibit TPMT include furosemide and ASA, as well as sulphasalazine and 5-ASA. The addition One of the advantages of determining TPMT status prior to initiating therapy is the ability to administer full throttle" doses of AZA or 6-MP from the very beginning. of sulphasalazine or 5-ASA to patients on a stable dose of azathioprine has been reported to cause leukopenia (usually mild). 20 This occurs as a result of TPMT inhibition, shunting more drug to 6-TG while reducing 6-MMP production. Theoretically, this drug combination may be strategically employed in patients failing therapy with AZA or 6-MP due to overproduction of 6-MMP. 19 Finally, the administration of AZA or 6-MP, the substrates for TPMT, induce its gene expression, resulting in increased enzyme activity (10% 33% after 3 months of therapy). Practical Issues and Frequently Asked Questions What Should One Do before Initiating Therapy? Determine that the patient is not leukopenic, and measure liver and pancreatic enzyme levels. Assure that the patient is not taking drugs that may interfere with AZA/6-MP metabolism (as detailed above), particularly allopurinol. Obtain results of TPMT genotype or enzyme activity, to reduce the risk of adverse events, such as myelotoxicity. 12,13 What Starting Dose Should Be Initiated? One of the advantages of determining TPMT status prior to initiating therapy is the ability to administer full doses of AZA (2.5 mg/kg) or 6-MP (1.25 mg/kg) from the very beginning. This has the potential to reduce lag time to reach therapeutic 6-TG levels, hastening response time and diminishing steroid exposure. The identification of TPMT heterozygous state, on the other hand, indicates that the starting dose should be about 33% of the usual dose. The drugs are contraindicated in cases of homozygous TPMT deficiency. The long half-life of these drugs allows for fine tuning the dose by alternating the daily dose. For example, if 6-TG levels in a nonresponder are slightly below threshold while taking 100 mg of AZA daily, one can administer 100 and 125 mg on alternating days to yield an increase of 12.5%, as opposed to 125 mg per day, which is a 25% increase in dose. S35

7 TPMT Enzyme and 6-Mercaptopurine continued What Tests Should Be Performed for Monitoring While on Therapy and at What Frequency? Complete blood count (CBC) monitoring should be performed at 2, 4, and 8 weeks after initiating therapy, irrespective of TPMT status. Liver and pancreatic enzymes are tested concurrently. In our experience, symptomatic idiosyncratic pancreatitis may be averted if pancreatic enzyme elevations are observed at 2 weeks, prior to onset of clinical symptoms (usually after 3 4 weeks). The CBC should be repeated every 3 months thereafter, or 15 days after a dose adjustment. 6-MP metabolite levels can be determined after 2 3 weeks on therapy or after a dose change, when a steady state is reached. Metabolite levels are monitored to the point of clinical response, serving as a baseline measure for future comparison. This allows the clinician to determine whether a late relapse is due to patient nonadherence to therapy. As an example, a 29-yearold male was taking 100 mg of 6-MP for 3 years, with a baseline 6-TG level of 244 pmol/ RBC, slightly above the desired threshold. The 6- MMP level was 2400 pmol/ RBC at that time. When a symptomatic relapse occurred 6 months later, his 6-TG and 6-MMP levels were 160 and 1903 pmol/ RBC, respectively. This downward trend for both metabolites on a fixed dose allowed the clinician to correctly identify lack of adherence to therapy as the cause of the relapse. When confronted with the metabolite results, the patient admitted to having become less adherent, forgetting to take the pills 2 days a week, on average. Had baseline levels not been performed, one may have incorrectly assumed that the dose prescribed was inadequate. Baseline values are also helpful to identify potentially toxic metabolite levels. For example, an 18-year-old Table 6 Indications for Thiopurine Metabolite Testing in IBD Patients Treated with Azathioprine or 6-Mercaptopurine 1. After initiating therapy 2. Following dose adjustments (including until target 6-TG reached) 3. At the time of a clinical relapse 4. Serially (twice yearly) to monitor adherence and/or TPMT induction 5. At the time of an adverse event (hepatotoxicity or myelotoxicity) 6. To monitor effect of change in weight or concomitant drug use (5-ASA, sulfasalazine, furosemide, etc) IBD, inflammatory bowel disease; 6-TG, 6-thioguanine; TPMT, thiopurine methyltransferase; 5-ASA, 5-aminosalicylates. female patient who had been well on 150 mg of AZA for 4 years presented with febrile leukopenia (white blood cell count 2300). Previous white blood cell counts had been normal. 6-MP levels, drawn for the first time, revealed 6-TG and 6-MMP levels of 602 and 660 pmol/ RBC, respectively. The low ratio of 6-MMP/6-TG (1.1) is very suggestive of the diagnosis of TPMT deficiency. The myelotoxic 6-TG level (>450 pmol/ RBC), along with an acute viral infection, led to the delayed symptomatic leukopenia. The indications for metabolite monitoring are summarized in Table 6. Why the Hepatotoxicity and What to Do about It? Hepatotoxicity from AZA or 6-MP usually presents as an increase in transaminase levels (typically one to four times the upper limits of normal). Indirect hyperbilirubinemia may also be observed. The presence of cholestasis should alert the clinician to look for other causes of liver disease, such as sclerosing cholangitis. Increased 6-MMP levels are causal, and the risk of hepatotoxicity increases (3% to 18%) when levels exceed 5700 pmol/ RBC. 17 Lowering the drug dose (10% 25%) is usually associated with normalization of transaminases within a few weeks. The identification of hepatotoxic 6-MMP levels is clinically useful, as it identifies a specific cause of elevated transaminases, rather than other potential causes (alcohol or nonalcoholic steatohepatitis, autoimmune or viral hepatitis, obesity, other drugs, etc). Hepatotoxicity due to excessive 6-MMP levels is not a reason to discontinue therapy; however, lowering the drug dose may also decrease 6-TG levels to below the threshold of the therapeutic window. Other strategies include fractionating the dose (75 mg twice daily rather than 150 at bedtime), as well as adding 5-ASA or sulphasalazine to inhibit the TPMT enzyme. The identification of TPMT deficiency and 6-MMP under-production translates to a reduced risk of hepatotoxicity, with less need to monitor transaminases. Why Not Follow White Blood Cell Counts and Maintain Leukopenia? In the era before metabolite measurement and the establishment of a therapeutic window for 6-TG levels, this was practiced by some clinicians to increase the likelihood of a clinical response. Whereas leukopenia means that the 6-TG level is high (usually S36

8 TPMT Enzyme and 6-Mercaptopurine >450 pmol/ RBC), there is no evidence that such myelotoxic levels yield higher efficacy than 6-TG in the 300 pmol/ RBC range. Thus, maintained leukopenia is unnecessary and potentially dangerous. In addition to an increased risk of infections, data suggest an increased risk of malignancy in patients maintained leukopenic with AZA or 6-MP for maintenance of leukemia or IBD. Why Does AZA or 6-MP Fail? The ability to assess metabolite levels has provided explanations for most patients who fail to respond to these drugs (Table 5). Far and away, the major cause of treatment failure is inadequate dose, with 6-TG and 6- MMP levels both low. 21 In patients with excessive TPMT activity, 6-MMP levels are high, compared with inadequate 6-TG levels, with an abnormally increased 6-MMP/6-TG ratio (Table 5). Lack of adherence to therapy is a key finding, easily discernible by serial metabolite measurements, as discussed above. Only a minority of failures are due to drug failure, despite adequate 6-TG levels. Why Not Use 6-TG Instead? We reported that patients intolerant to AZA or 6-MP, or not responding due to excessive TPMT activity with preferential 6-MMP production, can do well when given 6-TG. 22 In particular, patients with idiosyncratic reactions, such as pancreatitis, fever, and/or rashes usually tolerate 6-TG. However, there is little long-term experience with this drug for indications other than childhood leukemia. Furthermore, hepatic veno-occlusive disease with portal hypertension has been observed in patients with leukemia treated with 6-TG, but not with 6-MP. Thus, 6-TG should be reserved for patients intolerant to other thiopurine drugs that have provided informed consent for its use, as part of an IRB approved clinical trial. Conclusions Overall, the available data strongly support the fact that 6-MP metabolite measurements and TPMT molecular analysis provide clinicians with useful tools to optimize and individualize AZA/6-MP therapy in IBD. In view of the fact that response to AZA/6-MP therapy was optimized at 6-TG levels greater than a cutoff of 235 pmol/ RBC, target levels can be used as a therapeutic guide for those patients not responding. Furthermore, metabolite monitoring allows the clinician to identify the reasons for which patients fail to respond to 6-MP therapy, including nonadherence and inadequate dose, as well as excessive TPMT activity. The ability to identify high-risk patients by TPMT genotyping and monitoring 6-TG levels at which myelosuppression occurs will allow clinicians to individualize 6-MP dosing to maximize and hasten therapeutic efficacy while minimizing myelotoxicity. Main Points The immunomodulatory agent 6-mercaptopurine (6-MP) and its pro-drug azathioprine (AZA) have significant beneficial effects on the long-term clinical course of inflammatory bowel disease (IBD), but clinicians have expressed concerns about short- and long-term potential adverse effects, particularly with respect to myelotoxicity and hepatotoxicity. The immunosuppressive effects of AZA and 6-MP are mediated via their intracellular conversion along two competing pathways to yield the nucleotide metabolites 6-thioguanine (6-TG) and 6-methylmercaptopurine (6-MMP); 6-MMP is the largely the product of the activity of the thiopurine methyltransferase (TPMT) enzyme. It is highly recommended that TPMT genotype or phenotype be determined before the initiation of AZA or 6-MP therapy in patients with IBD. Patients heterozygous for a mutation of TPMT require lower drug dosing to optimize the therapeutic potential while avoiding myelotoxic levels of 6-TG; patients with homozygous mutations of TPMT (0.3% of the population) are at extreme risk of very severe leukopenia with potential septic complications and should thus not be exposed to 6-MP/AZA. Intracellular levels of 6-TG correlate with successful outcome and complete clinical response to AZA/6-MP; the probability of treatment response is significantly increased when red blood cell (RBC) 6-TG levels are greater than 235 pmol/ RBC. Measuring drug metabolite levels and adjusting the dose of AZA/6-MP to achieve 6-TG levels above the therapeutic threshold can convert a treatment failure into a responder. Intracellular 6-MMP levels greater than 5700 pmol/ RBC increase the likelihood of hepatotoxicity, whereas high levels of 6-TG, usually exceeding 450 pmol/ RBC, correlate with myelotoxicity. Far and away, the major cause of AZA/6-MP treatment failure is inadequate dose, with 6-TG and 6-MMP levels both low. In patients with excessive TPMT activity, 6-MMP levels are high, compared with inadequate 6-TG levels. Lack of adherence to therapy is a key finding, easily discernible by serial metabolite measurements. Only a minority of failures are due to drug failure, despite adequate 6-TG levels. S37

9 TPMT Enzyme and 6-Mercaptopurine continued References 1. Pearson DC, May GR, Fick GH, Sutherland LR. Azathioprine and 6-mercaptopurine in Crohn s disease: a meta-analysis. Ann Intern Med. 1995;123: Markowitz J, Grancher K, Kohn N, et al. A multicenter trial of 6-mercaptopurine and prednisone therapy in newly diagnosed children with Crohn s disease. Gastroenterology. 2000;119: Seidman EG. 6-Mercaptopurine to maintain remission in Crohn s disease: an old friend becomes a new hero. Gastroenterology. 2000;119: Present DH, Meltzer SJ, Krumholz MP, et al. 6- Mercaptopurine in the management of inflammatory bowel disease: short- and long-term toxicity. Ann Intern Med. 1989;111: Connell WR, Kamm MA, Ritchie JK, Lennard Jones JE. Bone marrow toxicity caused by azathioprine in inflammatory bowel disease: 27 years of experience. Gut. 1993;34: Kirschner BS. Safety of azathioprine and 6- mercaptopurine in pediatric patients with inflammatory bowel disease. Gastroenterology. 1998;115: Markowitz J, Grancher K, Mandel F, Daum F. Immunosuppressive therapy in pediatric inflammatory bowel disease: results of a survey of the North American Society for Pediatric Gastroenterology and Nutrition. Subcommittee on Immunosuppressive Use of the Pediatric IBD Collaborative Research Forum. Am J Gastroenterol. 1993;88: Cuffari C, Théorêt Y, Latour S, Seidman G. 6- Mercaptopurine metabolism in Crohn s disease: correlation with efficacy and toxicity. Gut. 1996;39: Yates CR, Krynetski EY, Loennechen T, et al. Molecular diagnosis of thiopurine S-methyltransferase deficiency: genetic basis for azathioprine and mercaptopurine intolerance. Ann Intern Med. 1997;126: Lennard L, Van Loon JA, Weinshilboun RM. Pharmacogenetics of acute azathioprine toxicity: relationship to thiopurine methyltransferase genetic polymorphism. Clin Pharmacol Ther. 1989;46: McLoed HL, Krynetski EY, Relling MV, Evans WE. Genetic polymorphism of thiopurine methyltransferase and its clinical relevance for childhood acute lymphoblastic leukemia. Leukemia. 2000;14: Stolk JN, Boerbooms AM, de Abreu RA, et al. Reduced thiopurine methyltransferase activity and development of side effects of azathioprine treatment in patients with rheumatoid arthritis. Arthritis Rheum. 1998;41: Colombel JF, Ferrari N, Debuysere H, et al. Genotypic analysis of thiopurine S-methyltransferase in patients with Crohn s disease and severe myelosuppression during azathioprine therapy. Gastroenterology. 2000;118: Regueiro M, Mardini H. Determination of thiopurine methyltransferase genotype or phenotype optimizes initial dosing of azathioprine for the treatment of Crohn s disease. J Clin Gastroenterol. 2002;35: Seidman EG, Furst DE. Pharmacogenetics for the individualization of treatment of rheumatic disorders using azathioprine. J Rheumatol. 2002;29: Corominas H, Domènech M, González D, et al. Allelic variants of the thiopurine S-methyltransferase deficiency in patients with ulcerative colitis and healthy controls. Am J Gastroenterol. 2000;95: Dubinsky MC, Lamothe S. Yang HY, et al. Pharmacogenomics and metabolite measurement for 6-mercaptopurine therapy in inflammatory bowel disease. Gastroenterology. 2000;118: Cuffari C, Hunt S, Bayless T. Utilisation of erythrocyte 6-thioguanine metabolite levels to optimise azathioprine therapy in patients with inflammatory bowel disease. Gut. 2001;48: Dubinsky M, Dubinsky MC, Yang H, et al. 6-MP metabolite profiles provide a biochemical explanation for 6-MP resistance in patients with inflammatory bowel disease. Gastroenterology. 2002;122: Szumlanski CL, Weinshilboum RM. Sulphasalazine inhibition of thiopurine methyltransferase: possible mechanism for interaction with 6-mercaptopurine and azathioprine. Br J Clin Pharmacol. 1995;39: Seidman EG. Recent advances in the diagnosis and treatment of pediatric inflammatory bowel disease. Gastroenterol Rep. 2000;1: Dubinsky MC, Hassard PV, Seidman EG, et al. An open-label pilot study using thioguanine as a therapeutic alternative in a subgroup of Crohn s disease patients resistant to 6-mercaptopurine therapy. Inflamm Bowel Dis. 2001;7: S38