Personalized medicine to implementation science: Thiopurines set for the leap

See article in JGH Open, this issue. DOI: https://doi.org/10.1002/jgh3.12798. The concept of the four Rs: right drug at the right dose at the right time for the right person is so supremely divine that it borders on being Don Quixotic. Yet, constant inroads by "omics" is bringing into easy grasp a hitherto remote dream. While fully cognizant of the fact that inflammatory bowel disease (IBD) is a heterogeneous disease, the treatment adage "one size fits all" still persists largely because of compulsion of lack of options. Fully knowing that the bewildering heterogeneity of disease hinders kind results for everyone, universal algorithms are still adhered to. Even the most discerning physician with in-depth knowledge of the nuances of therapeutic tools knows that responses may not go along with expectations. This fjord has been planked by our gradual advances and the promise of precision medicine. Personalized medicine is an emerging practice that uses an individual's genetic profile to guide decisions regarding the prevention, diagnosis, and treatment.1 Thiopurines have been gradually falling from grace possibly due to sequential introduction of various targeted monoclonal antibodies. The unbridled enthusiasm to achieve faster and higher treatment goals has led to impatience with the slow-acting thiopurines in the Western world, where easy access to biologics is not an issue. Recent AGA guidelines appear to bypass thiopurines in their quest for more efficient therapy.2 This is despite the constant fact that thiopurines remain a trusted ally of most IBD patients with limited resources. Thiopurines have faithfully served a wide proportion of IBD patients with moderate to excellent results. Certainly, use of thiopurines needs reinvention, and so it is fitting that this drug class is at the forefront of personalized medicine, ahead of other targeted therapies. Three different thiopurine formulations are clinically useful: azathioprine, 6-mercaptopurine (6-MP), and thioguanine (TG). Azathioprine, a pro-drug, is converted to 6-MP, enzymatically (glutathione S transferase) as well as non-enzymatically. As demonstrated in Figure 1, 6-MP can be channelized through three different pathways, the end products of which impact the efficacy as well as the toxicity of thiopurines.3, 4 The thiopurine S-methyl transferase (TPMT) enzyme converts 6-MP to 6-methyl mercaptopurine (6-MMP) and xanthine oxidase converts 6-MP to thiouric acid, both inactive metabolites. Hypoxanthine-guanine phosphoribosyl-transferase (HPRT) enzyme, through a series of enzymatic reactions, converts 6-MP into active 6-thioguanine nucleotides (6-TGN) metabolites (6-thioguanine triphopshate: 6-TGTP and deoxy-6-thioguanine triphopshate: d′6-TGTP), which are responsible for efficacy and myelosuppression. 6-thioguanine, the third therapeutic formulation, bypasses the pathways mentioned above and can restore therapeutic efficacy in patients with ineffective metabolism.5 Changes in the levels of enzymes responsible for gatekeeping the three major metabolic pathways impact the clinical efficacy and adverse effects of thiopurines (Fig. 1). The deficiency or reduction in TPMT levels due to various genetic polymorphisms is well recognized as a cause of thiopurine-related leukopenia. The pretreatment testing for susceptibility to TPMT-related leukopenia can be performed either at the phenotypic (testing for TPMT enzyme assay) or the genotypic level (polymorphisms).6, 7 There are some pros and cons of either test: genotype is not affected by physiological variables, but the testing may miss novel or unrecognized mutations. The phenotypic tests are best performed before starting the therapy because enzyme levels may vary with certain factors (transfusions and age) and decline once thiopurines are initiated. On the other hand, enzyme levels may help detect deficiency even in those where novel mutations are present.7 TPMT enzyme assay may misclassify TPMT deficient individuals as TPMT intermediate in a subset. Although thiopurine-related leukopenia is more frequent in Asian populations, TPMT mutations are not common. In 2014, a polymorphism in Nudix Hydrolase 15 (NUDT15) was first identified as a cause of thiopurine-related leukopenia in South Korea.8 Subsequently, this polymorphism has been identified as an important cause of thiopurine-related leukopenia in other Asian countries, including South Asia.9-11 NUDT15 has a role in hydrolysis of thio-guanosine-triphosphate into the monophosphate form. In the absence of NUD15 activity, this conversion is compromised, predisposing such individuals to increased leukopenia with thiopurine use. Multiple other polymorphisms have been suggested as a possible factor in thiopurine-related cytopenia, but the evidence is either conflicting or emerging (Table 1).12-17 Thiopurine methyl-transferase TPMT Controls the alternate metabolic pathway to form 6-methyl mercaptopurine. Deficiency may direct the thiopurine to increased production of thioguanine nucleotides amplifying the toxicity Heterozygous: 30–70% of dose Homozygous: Avoid or 10% Nudix hydrolase or nucleoside diphosphate-linked moiety X-type motif 15 NUDT15 Preferentially hydrolyzes 6-thio-guanosine triphosphate to 6-thio (deoxy)-guanosine monophosphate Reduced incorporation into DNA and lower cytotoxic and immunosuppressant effects Deficiency increased the triphosphate levels and may increase cytopenias Heterozygous: 30–70% of dose Homozygous: Avoid or 10% Inosine triphosphate pyrophosphohydrolase ITPase Hydrolyzes ITPA so as to prevent excess 6-TITP Deficiency will lead to accumulation of 6-TITP and increased toxicity rs1127354 or 94 C > A rs7270101 or (IVS2 + 21 A > C) May cause neutropenia and other adverse effects like rash, pancreatitis Conflicting evidence ATP-dependent efflux of thiopurine metabolites limiting the toxicity Reduced function results in intracellular accumulation of 6-TGN The clinical impact of pretherapy genotype testing has been studied in multiple randomized trials. In the TROPIC trial, the pretreatment testing groups (TPMT mutations) had a similar hematological adverse event rate as compared with the standard therapy. Still, the testing was beneficial if only patients with TPMT polymorphisms were considered.18 In a multicenter Korean randomized trial, myelosuppression was less frequently noted in patients with polymorphism testing as compared with those who did not (16.7% vs 35.9%).19 In a Chinese study, where NUDT15 testing was compared with the control group, leukopenia occurred less frequently in the testing arm (23.7% vs 32.4%).20 Therefore, the evidence through RCTs suggests that the pretherapy genotype assessment reduces the myelotoxicity associated with thiopurine use (Table 2).18-20 IBD Multiple centers in Korea Testing for NUDT15, FTO and three common TPMT variants Homozygous: alternative Heterzygous: 50 of azathioprine equivalents Crohn's disease Two hospitals in China NUDT15 C415T CC: standard dose CT: 50% of standard dose, t TT: alternative drugs Because of clear evidence, we usually test for common polymorphisms of TPMT and NUDT15 before initiating thiopurines and modify the dose based on genotype testing.21 The presence of a particular genotype should be utilized to classify the phenotype as a normal, intermediate, or poor metabolizer. Poor metabolizers should ideally avoid thiopurines; intermediate metabolizers should be initiated on a reduced dose (30–80%), while normal metabolizers can be started on a normal dose. The pretesting for polymorphisms does not completely obviate the risk of leukopenia completely, and the follow-up should continue to assess for changes in hemogram and liver function tests. Once the thiopurines have been initiated, the assessment of metabolites may provide an opportunity to modify the treatment to improve effectiveness and reduce adverse effects. Measurement of two metabolites, as reported by Yeo et al. in this issue, could provide an opportunity to address nonresponse and adverse effects of thiopurine therapy.22 The measurement of 6-TGN, which is the active metabolite, assesses the therapeutic benefit of thiopurines. On the contrary, the levels of inactive 6-MMP estimate the shunting of thiopurines by TPMT. Shunting can cause nonresponse and may need an increase in the dose of thiopurine or the addition of drugs like allopurinol which provide more drug for the HPRT pathway and increase 6-TGN production. On the contrary, an elevated 6-TGN in the wake of nonresponse to the thiopurine therapy suggests resistance and warrants a change of therapy. A low level of both 6-TGN and MMP suggests issues with drug compliance or low dose (Table 3). Although simplistic, the clinical utility of the metabolite measurements has not gained enough currency. The present study adds to the growing evidence and calls for more frequent use of metabolite measurements, especially in difficult cases. Appropriate dose not being administered Check adherence Increase dose 6-MMP/TGN > 11 Shunting via the TPMT pathway Add allopurinol Therapeutic dose If there is no clinical effect—consider resistance and change therapy Reduce dose If no clinical effect—change to another therapy However, there are practical difficulties in using this strategy—the metabolite measurements may not be routinely available, additional costs to a cheap IBD therapy and the levels representing low or high values are not standardized and vary across studies. Further, the data regarding the predictive value of these levels are not homogeneous. In a study comparing two different target levels (600–1200 pmol/8 × 108 RBC vs 320–630 pmol/8 × 108 RBC), the higher target levels were associated with numerically more (but not statistically more) discontinuations related to adverse events.23 In an RCT, an adjustive strategy to target a TGN level of 250–400 pmol/8 × 108 erythrocytes in Crohn's disease did not achieve a better clinical efficacy.24 Observational studies, however, d

Abstract licence: CC BY-NC

39 more references, whose wording the paper’s licence keeps private.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.