Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
Review
The role of thiopurine metabolite monitoring in inflammatory bowel disease Expert Rev. Gastroenterol. Hepatol. 8(4), 383–392 (2014)
Lauren Beswick, Antony B Friedman and Miles P Sparrow* Department of Gastroenterology, Alfred Health and Monash University, Commercial Road, Melbourne, 3004 Victoria, Australia *Author for correspondence: Tel.: +61 390 762 223; +61 395 910 466 Fax: +61 390 762 194; +61 395 910 966 [email protected]
Thiopurines are the mainstay of medical management in inflammatory bowel disease (IBD), especially in the maintenance of disease remission. Given the limited IBD armamentarium it is important to optimize each therapy before switching to an alternative drug. Conventional weight based dosing of thiopurines in IBD leads to intolerance or inefficacy in many patients. More recently increased knowledge of their metabolism has allowed for dose optimization using thiopurine metabolite levels, namely 6-thioguanine nucleotides and 6-methylmercaptopurine, with the potential for improved outcomes in patients with IBD. This review will outline the current understanding of thiopurine metabolism and pharmacogenomics and will describe the clinical application of this knowledge in the optimization of thiopurines in individual patients. KEYWORDS: 6-mercaptopurine • allopurinol • azathioprine • inflammatory bowel disease • metabolism • thiopurine methyl transferase • thiopurines
The thiopurines, 6-mercaptopurine (6MP) and its pro-drug azathioprine (AZA), are established therapies in the management of inflammatory bowel disease (IBD). They are utilized in both Crohn’s disease (CD) and ulcerative colitis (UC) as induction, maintenance and steroid-sparing agents. However, up to 60% of patients do not respond to conventional thiopurine dosing and up to 30% suffer adverse effects to these agents [1]. Historically, thiopurines have been dosed using a weight-based regimen, but recent developments in dose optimization have enabled the measurement of intracellular metabolites that are associated with either efficacy or intolerance to these agents. This allows a more individualized approach to thiopurine dosing, which will hopefully translate to an increase in the numbers of patients achieving long-term steroidfree clinical remission. Despite these advances, thiopurine metabolite testing is still underutilized in everyday clinical practice, with availability and reimbursement of the test impacting greatly on its regular use [2,3]. The role of thiopurine monotherapy in CD has recently been questioned, in contrast to their accepted role in combination therapy with anti-tissue necrosis factor agents in both
informahealthcare.com
10.1586/17474124.2014.894878
CD and UC, and in the prevention of postoperative recurrence in CD [3–6]. However, two recent studies evaluating the early administration of AZA in CD have shown that AZA is no more effective than placebo or conventional therapy when given as monotherapy soon after diagnosis [7,8]. Both of these studies used conventional weight-based dosing of thiopurines without the use of thiopurine metabolite testing. Whether the early use of thiopurines dose optimized with metabolite testing in this setting would have led to different outcomes remains to be determined. Despite these discordant recent results on the role of thiopurines as induction agents in CD, in the real world, thiopurines remain the primary agents used for the maintenance of steroid-free remission in both CD and UC. Thiopurine pharmacology
When absorbed, the pro-drug AZA is rapidly converted, predominantly by glutathione-stransferase, to 6MP. Subsequently, 6MP is metabolized via three competing enzymatic pathways: thiopurine methyltransferase (TPMT), xanthine oxidase (XO) and hypoxanthine phosphoribosyltransferase (HPRT). Thiopurine metabolism via the XO pathway leads to the
2014 Informa UK Ltd
ISSN 1747-4124
383
Review
Beswick, Friedman & Sparrow
1.0–1.5 mg/kg/day 6MP). In the original placebo-controlled studies, response rates varied between 42 and 80% [12–14]. Given Thioxanthine TPMT TPMT Thioxanthine variable results using conventional dosing, thiopurine metabolite measurements are Glutathione HPRT 6-MP AZA 6-TIMP increasingly being used to optimize thiopurine therapy in IBD and improve cliniAllopurinol IMPDH XO cal outcomes. More specifically, the measurement of thiopurine metabolites 6-TXMP 6-TU during initiation of treatment has been evaluated in recent studies, with variable GMPS results. A small German-based study in 2007 compared conventional weight6-TGMP based dosing of thiopurines to 6TGN concentration-adapted thiopurine therapy. Kinase There was no difference in both groups in Incorporation terms of 6TGN levels or clinical remission into DNA and 6-TGN 6-TGDP rates at 24 weeks of follow-up. However, inhibition of replication this study was limited by the small numKinase bers of participants and a moderate dropout rate. In addition, both study groups Rac-1 Apoptosis 6-TGTP had sub-therapeutic 6TGN levels at the inhibition end of follow-up [15]. A more recent multicenter US study compared weight-based Figure 1. Thiopurine metabolism. 6-MMP: 6-methylmercaptopurine; 6-MP: 6-mercaptopurine; 6-TGDP: 6-thioguanine dosing to individualized dosing based on diphosphate; 6-TGMP: 6-thioguanine monophosphate; 6-TGN: 6-thioguanine nucleotides; thiopurine metabolites. Again, there was 6-TGTP: 6-thioguanine triphosphate; 6-TIMP: 6-thioinosine monophosphate; only a trend toward clinical remission 6-TU: 6-thiouric acid; 6-TXMP: 6-thioxanthine monophosphate; AZA: Azathioprine; (p = 0.11) and increased median 6TGN GMPS: Guanosine monophosphate synthase; HPRT: Hypoxanthine levels (p = 0.07) favoring the individualphosphoribosyltransferase; TPMT: Thiopurine methyltransferase; XO: Xanthine oxidase. ized dosing arm; however, this study again production of 6-thiouric acid, an inactive metabolite excreted in lacked statistical power due to the small number of patients urine. TPMT methylates 6MP into 6-methylmercaptopurine recruited [16]. Given these recent inconclusive results from small, (6MMP). Finally, metabolism via HRPT followed by inosine underpowered studies, the comparison of conventional weightmonophosphate dehydrogenase and guanosine monophosphate based dosing to individualized therapy based on thiopurine synthase leads to the production of the 6-thioguanine nucleoti- metabolites needs further larger prospective studies to help deterdes (6TGN) which are the active metabolites responsible for mine the best dosing strategy. thiopurine efficacy. Supra-therapeutic levels of 6TGN are potentially myelotoxic, while high levels of 6MMP are associated with Thiopurine metabolites: measurement & utility hepatotoxicity, albeit with low sensitivity and specificity [9]. 6TGN and 6MMP levels are measured with high-performance 6TGN encompasses 6-TG monophosphate (6TGMP), liquid chromatography in human blood, with values expressed diphosphate and triphosphate (6TGTP). 6TGN have a number in pmol/8 108 red blood cells (RBCs) [17]. The therapeutic of actions, which ultimately lead to apoptosis and inactivation range for use in clinical practice for 6TGN is 235–400 pmol/ of T-lymphocytes. First, 6TGN are incorporated into DNA in 8 108 RBCs. For 6MMP, a level of less than 5700 pmol/ place of guanine and adenosine, leading to strand breakage and 8 108 RBCs mitigates the risk of hepatotoxicity. Data sugcell cycle arrest. Second, the 6TGN that are incorporated into gest that 6TGN concentrations in excess of 235 pmol/8 108 DNA show reduced stability, leading to changes in DNA struc- RBCs are associated with clinical remission in a significant proture and activation of the mismatch repair system. Third, portion of patients [18]. The 6TGN upper limit is based on 6TGTP competitively inhibits Rac1, which in turn blocks the studies showing that the proportion of patients in remission activation of Vav and the Rac1 target genes such as NF-kB does not increase significantly with 6TGN concentrations and STAT 3 [10,11]. The intracellular metabolism of thiopurines greater than 450 pmol/8 108 RBCs, whereas there is an is outlined in FIGURE 1. increased risk of myelotoxicity above this level [19].
Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
6-MMP
6-MMP
Dosing of thiopurines
6TGN levels & efficacy
Conventionally, thiopurines have been prescribed in IBD using a weight-based dosing regimen (2.0–2.5 mg/kg/day AZA and
Although the relationship between 6TGN levels and efficacy was first described in IBD patients in 1996, it was the follow-
384
Expert Rev. Gastroenterol. Hepatol. 8(4), (2014)
Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
The role of thiopurine metabolite monitoring in IBD
up landmark study in 2000 that confirmed this correlation [18]. Clinical response in 92 pediatric patients (79 CD, 8 UC and 5 indeterminate colitis patients) was correlated with thiopurine metabolite levels. Higher 6TGN levels were observed in responders compared to non-responders (median 6TGN 312 vs 199 pmol/8 108 RBCs; p < 0.0001). If 6TGN levels were greater than 235 pmol/8 108 RBCs, patients had an odds ratio of 5.0 (p < 0.0001) of being a thiopurine responder. There was no correlation with response and other factors including 6MMP levels, drug dose and concurrent medications. The dose of thiopurine correlated poorly with 6TGN levels (r = 0.0009) [18]. A meta-analysis in 2006 of 12 studies with 941 patients showed that patients with 6TGN levels >230– 260 pmol/8 108 RBCs were more likely to be in remission than those below the threshold level (62 vs 36%) with a pooled odd ratio of 3.27 (p < 0.0001) [20]. A more recent pooled analysis in 2013 included 20 studies of 2234 patients and showed a pooled odds ratio of 2.09 (p < 0.00001) for clinical remission using a threshold of 230 pmol/8 108 RBCs [21]. In contrast, a prospective multicenter Spanish study of 133 IBD patients did not find a correlation between clinical response and 6TGN levels, although the positive predictive value for clinical response if 6TGN levels were above 260 pmol/8 108 RBCs was 87% [22]. However, in this study, thiopurine doses were not adjusted depending on metabolite results, making this a purely observational study. Dose optimization studies using 6TGN levels have also been reported. Two retrospective Australian studies have shown that optimization of thiopurines in patients with sub-therapeutic 6TGN levels can lead to improvement in clinical outcomes in 88 and 78% of patients after dose escalation of thiopurines, respectively [23,24]. A French study of 55 patients with steroiddependent or active IBD for at least 6 months while on stable doses of AZA evaluated dose escalation based on 6TGN levels and clinical outcomes. In patients with a baseline 6TGN of 100–200 pmol/8 108 RBCs in whom dose escalation was carried out, clinical remission was attained in 77%. In those with a baseline 6TGN of 300–400 pmol/8 108 RBCs, only 24% of patients achieved clinical remission, while no patients with a baseline 6TGN above 400 pmol/8 108 RBCs improved after dose escalation [25]. 6TGN levels & myelosuppression
Thiopurines are associated with myelosuppression, most frequently leucopenia, regardless of TPMT activity. Studies have shown that myelosuppression can occur from within days to many months after the initiation of thiopurine therapy, even in patients with a normal TPMT phenotype [18,26]. The same landmark study that described a correlation between 6TGN levels and clinical efficacy also demonstrated a correlation between 6TGN levels and leucopenia. Leucopenia was observed in 14% of patients and was associated with higher 6TGN levels (p < 0.03) [18]. This relationship has also been described in cardiac and renal transplantation settings as well as in pediatric patients with acute lymphoblastic leukemia [27–29]. A Swedish informahealthcare.com
Review
retrospective study has shown that high 6TGN levels have been associated with an increased risk of adverse events with thiopurines, in particular myelotoxicity. In a cohort of 266 IBD patients in whom metabolites were measured, 29 patients had a 6TGN level greater than 400 pmol/8 108 RBCs. About 41% of these patients experienced an adverse event including gastrointestinal side effects and myelosuppression, which was significantly higher than those with levels below 400 pmol/8 108 RBCs (p = 0.001) [30]. 6MMP levels & hepatotoxicity
Prior to the utilization of thiopurine metabolites to guide and optimize thiopurine therapy, if an IBD patient developed hepatotoxicity in the setting of these medications, they would be labeled intolerant to the offending drug. In cases of hepatotoxicity with AZA, 6MP may subsequently be trialed with a tolerability of approximately 50%, whereas initial or subsequent hepatotoxicity to 6MP would mean that thiopurines would be subsequently contraindicated. This narrows the potential therapeutic options in these patients. However, the initial pivotal Canadian metabolite study also demonstrated that high levels of 6MMP were associated with hepatotoxicity with elevated levels of transaminases. The incidence of hepatotoxicity in this study was 17% with median 6MMP levels of 5463 pmol/8 108 RBCs in those patients with abnormal liver function tests compared to 2213 pmol/8 108 RBCs in those with normal liver function tests (p < 0.05). The risk of hepatotoxicity increased threefold (18 vs 6%, p < 0.05) when 6MMP exceeded 5700 pmol/8 108 RBCs. There was no correlation of 6MMP levels with clinical efficacy or thiopurine dose. In addition, there was no relationship between hepatotoxicity and 6TGN levels [18]. These studies identified a subgroup of thiopurine ‘shunters’ who preferentially produce 6MMP over 6TGN and are defined by having a 6MMP:6TGN ratio of greater than 20. This group of patients, which is probably 10–15% of all patients treated with thiopurines, is at increased risk of hepatotoxicity and is likely to be refractory to standard therapy with thiopurines due to this metabolic profile. Use of thiopurine metabolites in clinical practice
Despite an increased understanding of the pharmacology of thiopurine metabolism, the role of metabolite testing in clinical practice continues to be debated. Two recent retrospective studies have described the use of thiopurine metabolites to improve clinical outcomes in IBD patients treated with thiopurines. Both studies used a metabolite-directed algorithm, which categorized patients into five subgroups – non-adherent to thiopurine therapy, underdosed/rapid metabolizers, thiopurine shunters, refractory and overdosed patients (TABLE 1). In the first study, thiopurine metabolites were measured in 63 patients with active IBD despite being on a stable thiopurine dose for at least 3 months. Based on metabolite results, 11% patients were non-adherent, 29% were underdosed, 40% were truly refractory to treatment, 13% were 385
Review
Beswick, Friedman & Sparrow
Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
Table 1. Thiopurine metabolite results, interpretation, recommended action and prevalence. Metabolite result†
Interpretation
Action recommended
Approximate prevalence (%)
Group 1
No/very low 6TGN (
Review
The role of thiopurine metabolite monitoring in inflammatory bowel disease Expert Rev. Gastroenterol. Hepatol. 8(4), 383–392 (2014)
Lauren Beswick, Antony B Friedman and Miles P Sparrow* Department of Gastroenterology, Alfred Health and Monash University, Commercial Road, Melbourne, 3004 Victoria, Australia *Author for correspondence: Tel.: +61 390 762 223; +61 395 910 466 Fax: +61 390 762 194; +61 395 910 966 [email protected]
Thiopurines are the mainstay of medical management in inflammatory bowel disease (IBD), especially in the maintenance of disease remission. Given the limited IBD armamentarium it is important to optimize each therapy before switching to an alternative drug. Conventional weight based dosing of thiopurines in IBD leads to intolerance or inefficacy in many patients. More recently increased knowledge of their metabolism has allowed for dose optimization using thiopurine metabolite levels, namely 6-thioguanine nucleotides and 6-methylmercaptopurine, with the potential for improved outcomes in patients with IBD. This review will outline the current understanding of thiopurine metabolism and pharmacogenomics and will describe the clinical application of this knowledge in the optimization of thiopurines in individual patients. KEYWORDS: 6-mercaptopurine • allopurinol • azathioprine • inflammatory bowel disease • metabolism • thiopurine methyl transferase • thiopurines
The thiopurines, 6-mercaptopurine (6MP) and its pro-drug azathioprine (AZA), are established therapies in the management of inflammatory bowel disease (IBD). They are utilized in both Crohn’s disease (CD) and ulcerative colitis (UC) as induction, maintenance and steroid-sparing agents. However, up to 60% of patients do not respond to conventional thiopurine dosing and up to 30% suffer adverse effects to these agents [1]. Historically, thiopurines have been dosed using a weight-based regimen, but recent developments in dose optimization have enabled the measurement of intracellular metabolites that are associated with either efficacy or intolerance to these agents. This allows a more individualized approach to thiopurine dosing, which will hopefully translate to an increase in the numbers of patients achieving long-term steroidfree clinical remission. Despite these advances, thiopurine metabolite testing is still underutilized in everyday clinical practice, with availability and reimbursement of the test impacting greatly on its regular use [2,3]. The role of thiopurine monotherapy in CD has recently been questioned, in contrast to their accepted role in combination therapy with anti-tissue necrosis factor agents in both
informahealthcare.com
10.1586/17474124.2014.894878
CD and UC, and in the prevention of postoperative recurrence in CD [3–6]. However, two recent studies evaluating the early administration of AZA in CD have shown that AZA is no more effective than placebo or conventional therapy when given as monotherapy soon after diagnosis [7,8]. Both of these studies used conventional weight-based dosing of thiopurines without the use of thiopurine metabolite testing. Whether the early use of thiopurines dose optimized with metabolite testing in this setting would have led to different outcomes remains to be determined. Despite these discordant recent results on the role of thiopurines as induction agents in CD, in the real world, thiopurines remain the primary agents used for the maintenance of steroid-free remission in both CD and UC. Thiopurine pharmacology
When absorbed, the pro-drug AZA is rapidly converted, predominantly by glutathione-stransferase, to 6MP. Subsequently, 6MP is metabolized via three competing enzymatic pathways: thiopurine methyltransferase (TPMT), xanthine oxidase (XO) and hypoxanthine phosphoribosyltransferase (HPRT). Thiopurine metabolism via the XO pathway leads to the
2014 Informa UK Ltd
ISSN 1747-4124
383
Review
Beswick, Friedman & Sparrow
1.0–1.5 mg/kg/day 6MP). In the original placebo-controlled studies, response rates varied between 42 and 80% [12–14]. Given Thioxanthine TPMT TPMT Thioxanthine variable results using conventional dosing, thiopurine metabolite measurements are Glutathione HPRT 6-MP AZA 6-TIMP increasingly being used to optimize thiopurine therapy in IBD and improve cliniAllopurinol IMPDH XO cal outcomes. More specifically, the measurement of thiopurine metabolites 6-TXMP 6-TU during initiation of treatment has been evaluated in recent studies, with variable GMPS results. A small German-based study in 2007 compared conventional weight6-TGMP based dosing of thiopurines to 6TGN concentration-adapted thiopurine therapy. Kinase There was no difference in both groups in Incorporation terms of 6TGN levels or clinical remission into DNA and 6-TGN 6-TGDP rates at 24 weeks of follow-up. However, inhibition of replication this study was limited by the small numKinase bers of participants and a moderate dropout rate. In addition, both study groups Rac-1 Apoptosis 6-TGTP had sub-therapeutic 6TGN levels at the inhibition end of follow-up [15]. A more recent multicenter US study compared weight-based Figure 1. Thiopurine metabolism. 6-MMP: 6-methylmercaptopurine; 6-MP: 6-mercaptopurine; 6-TGDP: 6-thioguanine dosing to individualized dosing based on diphosphate; 6-TGMP: 6-thioguanine monophosphate; 6-TGN: 6-thioguanine nucleotides; thiopurine metabolites. Again, there was 6-TGTP: 6-thioguanine triphosphate; 6-TIMP: 6-thioinosine monophosphate; only a trend toward clinical remission 6-TU: 6-thiouric acid; 6-TXMP: 6-thioxanthine monophosphate; AZA: Azathioprine; (p = 0.11) and increased median 6TGN GMPS: Guanosine monophosphate synthase; HPRT: Hypoxanthine levels (p = 0.07) favoring the individualphosphoribosyltransferase; TPMT: Thiopurine methyltransferase; XO: Xanthine oxidase. ized dosing arm; however, this study again production of 6-thiouric acid, an inactive metabolite excreted in lacked statistical power due to the small number of patients urine. TPMT methylates 6MP into 6-methylmercaptopurine recruited [16]. Given these recent inconclusive results from small, (6MMP). Finally, metabolism via HRPT followed by inosine underpowered studies, the comparison of conventional weightmonophosphate dehydrogenase and guanosine monophosphate based dosing to individualized therapy based on thiopurine synthase leads to the production of the 6-thioguanine nucleoti- metabolites needs further larger prospective studies to help deterdes (6TGN) which are the active metabolites responsible for mine the best dosing strategy. thiopurine efficacy. Supra-therapeutic levels of 6TGN are potentially myelotoxic, while high levels of 6MMP are associated with Thiopurine metabolites: measurement & utility hepatotoxicity, albeit with low sensitivity and specificity [9]. 6TGN and 6MMP levels are measured with high-performance 6TGN encompasses 6-TG monophosphate (6TGMP), liquid chromatography in human blood, with values expressed diphosphate and triphosphate (6TGTP). 6TGN have a number in pmol/8 108 red blood cells (RBCs) [17]. The therapeutic of actions, which ultimately lead to apoptosis and inactivation range for use in clinical practice for 6TGN is 235–400 pmol/ of T-lymphocytes. First, 6TGN are incorporated into DNA in 8 108 RBCs. For 6MMP, a level of less than 5700 pmol/ place of guanine and adenosine, leading to strand breakage and 8 108 RBCs mitigates the risk of hepatotoxicity. Data sugcell cycle arrest. Second, the 6TGN that are incorporated into gest that 6TGN concentrations in excess of 235 pmol/8 108 DNA show reduced stability, leading to changes in DNA struc- RBCs are associated with clinical remission in a significant proture and activation of the mismatch repair system. Third, portion of patients [18]. The 6TGN upper limit is based on 6TGTP competitively inhibits Rac1, which in turn blocks the studies showing that the proportion of patients in remission activation of Vav and the Rac1 target genes such as NF-kB does not increase significantly with 6TGN concentrations and STAT 3 [10,11]. The intracellular metabolism of thiopurines greater than 450 pmol/8 108 RBCs, whereas there is an is outlined in FIGURE 1. increased risk of myelotoxicity above this level [19].
Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
6-MMP
6-MMP
Dosing of thiopurines
6TGN levels & efficacy
Conventionally, thiopurines have been prescribed in IBD using a weight-based dosing regimen (2.0–2.5 mg/kg/day AZA and
Although the relationship between 6TGN levels and efficacy was first described in IBD patients in 1996, it was the follow-
384
Expert Rev. Gastroenterol. Hepatol. 8(4), (2014)
Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
The role of thiopurine metabolite monitoring in IBD
up landmark study in 2000 that confirmed this correlation [18]. Clinical response in 92 pediatric patients (79 CD, 8 UC and 5 indeterminate colitis patients) was correlated with thiopurine metabolite levels. Higher 6TGN levels were observed in responders compared to non-responders (median 6TGN 312 vs 199 pmol/8 108 RBCs; p < 0.0001). If 6TGN levels were greater than 235 pmol/8 108 RBCs, patients had an odds ratio of 5.0 (p < 0.0001) of being a thiopurine responder. There was no correlation with response and other factors including 6MMP levels, drug dose and concurrent medications. The dose of thiopurine correlated poorly with 6TGN levels (r = 0.0009) [18]. A meta-analysis in 2006 of 12 studies with 941 patients showed that patients with 6TGN levels >230– 260 pmol/8 108 RBCs were more likely to be in remission than those below the threshold level (62 vs 36%) with a pooled odd ratio of 3.27 (p < 0.0001) [20]. A more recent pooled analysis in 2013 included 20 studies of 2234 patients and showed a pooled odds ratio of 2.09 (p < 0.00001) for clinical remission using a threshold of 230 pmol/8 108 RBCs [21]. In contrast, a prospective multicenter Spanish study of 133 IBD patients did not find a correlation between clinical response and 6TGN levels, although the positive predictive value for clinical response if 6TGN levels were above 260 pmol/8 108 RBCs was 87% [22]. However, in this study, thiopurine doses were not adjusted depending on metabolite results, making this a purely observational study. Dose optimization studies using 6TGN levels have also been reported. Two retrospective Australian studies have shown that optimization of thiopurines in patients with sub-therapeutic 6TGN levels can lead to improvement in clinical outcomes in 88 and 78% of patients after dose escalation of thiopurines, respectively [23,24]. A French study of 55 patients with steroiddependent or active IBD for at least 6 months while on stable doses of AZA evaluated dose escalation based on 6TGN levels and clinical outcomes. In patients with a baseline 6TGN of 100–200 pmol/8 108 RBCs in whom dose escalation was carried out, clinical remission was attained in 77%. In those with a baseline 6TGN of 300–400 pmol/8 108 RBCs, only 24% of patients achieved clinical remission, while no patients with a baseline 6TGN above 400 pmol/8 108 RBCs improved after dose escalation [25]. 6TGN levels & myelosuppression
Thiopurines are associated with myelosuppression, most frequently leucopenia, regardless of TPMT activity. Studies have shown that myelosuppression can occur from within days to many months after the initiation of thiopurine therapy, even in patients with a normal TPMT phenotype [18,26]. The same landmark study that described a correlation between 6TGN levels and clinical efficacy also demonstrated a correlation between 6TGN levels and leucopenia. Leucopenia was observed in 14% of patients and was associated with higher 6TGN levels (p < 0.03) [18]. This relationship has also been described in cardiac and renal transplantation settings as well as in pediatric patients with acute lymphoblastic leukemia [27–29]. A Swedish informahealthcare.com
Review
retrospective study has shown that high 6TGN levels have been associated with an increased risk of adverse events with thiopurines, in particular myelotoxicity. In a cohort of 266 IBD patients in whom metabolites were measured, 29 patients had a 6TGN level greater than 400 pmol/8 108 RBCs. About 41% of these patients experienced an adverse event including gastrointestinal side effects and myelosuppression, which was significantly higher than those with levels below 400 pmol/8 108 RBCs (p = 0.001) [30]. 6MMP levels & hepatotoxicity
Prior to the utilization of thiopurine metabolites to guide and optimize thiopurine therapy, if an IBD patient developed hepatotoxicity in the setting of these medications, they would be labeled intolerant to the offending drug. In cases of hepatotoxicity with AZA, 6MP may subsequently be trialed with a tolerability of approximately 50%, whereas initial or subsequent hepatotoxicity to 6MP would mean that thiopurines would be subsequently contraindicated. This narrows the potential therapeutic options in these patients. However, the initial pivotal Canadian metabolite study also demonstrated that high levels of 6MMP were associated with hepatotoxicity with elevated levels of transaminases. The incidence of hepatotoxicity in this study was 17% with median 6MMP levels of 5463 pmol/8 108 RBCs in those patients with abnormal liver function tests compared to 2213 pmol/8 108 RBCs in those with normal liver function tests (p < 0.05). The risk of hepatotoxicity increased threefold (18 vs 6%, p < 0.05) when 6MMP exceeded 5700 pmol/8 108 RBCs. There was no correlation of 6MMP levels with clinical efficacy or thiopurine dose. In addition, there was no relationship between hepatotoxicity and 6TGN levels [18]. These studies identified a subgroup of thiopurine ‘shunters’ who preferentially produce 6MMP over 6TGN and are defined by having a 6MMP:6TGN ratio of greater than 20. This group of patients, which is probably 10–15% of all patients treated with thiopurines, is at increased risk of hepatotoxicity and is likely to be refractory to standard therapy with thiopurines due to this metabolic profile. Use of thiopurine metabolites in clinical practice
Despite an increased understanding of the pharmacology of thiopurine metabolism, the role of metabolite testing in clinical practice continues to be debated. Two recent retrospective studies have described the use of thiopurine metabolites to improve clinical outcomes in IBD patients treated with thiopurines. Both studies used a metabolite-directed algorithm, which categorized patients into five subgroups – non-adherent to thiopurine therapy, underdosed/rapid metabolizers, thiopurine shunters, refractory and overdosed patients (TABLE 1). In the first study, thiopurine metabolites were measured in 63 patients with active IBD despite being on a stable thiopurine dose for at least 3 months. Based on metabolite results, 11% patients were non-adherent, 29% were underdosed, 40% were truly refractory to treatment, 13% were 385
Review
Beswick, Friedman & Sparrow
Expert Review of Gastroenterology & Hepatology Downloaded from informahealthcare.com by Nanyang Technological University on 04/24/15 For personal use only.
Table 1. Thiopurine metabolite results, interpretation, recommended action and prevalence. Metabolite result†
Interpretation
Action recommended
Approximate prevalence (%)
Group 1
No/very low 6TGN (
