Many drugs are inactivated by enzymes. Thus, patients who are intermediate, and, in particular, poor drug metabolizers can have higher concentrations of active drug metabolites, increasing the risk for concentration-dependent adverse events. In this issue of Clinical Pharmacology and Therapeutics, two manuscripts address this fundamental area in pharmacokinetics, focusing on enzymes that metabolize drugs with narrow therapeutic indices (Figure 1).

The first paper of interest is a comprehensive review in which White et al.¹ describe the role of dihydropyrimidine dehydrogenase (DPD) deficiency and recommend implementation of DPYD genotyping as a way to identify patients at a high risk for severe adverse events and to personalize the use of fluoropyrimidines. The DPYD gene encodes DPD, an enzyme involved in the metabolism of fluoropyrimidines (5-fluorouracil, capecitabine, and tegafur), chemotherapeutic agents frequently prescribed for several common cancers. However, the use of fluoropyrimidines is limited by a wide range of frequent side effects, including diarrhea and bone marrow suppression. As such, DYPD genotyping helps identify patients who are carriers of DPYD functional variants and therefore are phenotypically DPD deficient. This testing presents an opportunity for genotype-guided dose adjustment of fluoropyrimidines in routine clinical practice, as widely supported by pharmacogenomic experts around the world. Current recommendations by Clinical Pharmacogenetics Implementation Consortium (CPIC) include dose adjustment in heterozygous carriers,² and the Dutch Pharmacogenomics Working Group proposes a score based on four DPYD variants that can be used to personalize treatment.³

Whether to genotype or not in routine clinical care does raise some potential concerns in this instance. First, there is a risk for decreased effectiveness in cancer treatment, as pre-emptive genotyping could lead to dose reductions. Other concerns include costs and the potential for treatment delays. Nevertheless, data indicate that the effectiveness of fluoropyrimidine-based treatment does not decrease the effectiveness of cancer treatment,⁴ and, as Deenen et al.⁵ referenced, fluoropyrimidine dose reductions based on genotype information significantly decreased the rates of serious toxicities. Despite this evidence and genotype endorsement by multiple regulatory agencies, Australian authorities have not issued a position statement. Therefore, after review of the evidence, the authors favored the recommendation to implement genotyping of patients who undergo therapy with fluoropyrimidines in routine clinical practice in Australia.

The paper by Pristup et al.⁶ describes an endogenous substrate for thiopurine S-methyltransferase (TPMT), a critical enzyme in the metabolic pathway of thiopurines, including azathioprine and 6-mercaptopurine. Thiopurines are used in the treatment of patients with cancer, autoimmune and other inflammatory conditions, and patients undergoing solid organ transplants.^7-9 Despite their multiple indications, thiopurine use is limited by frequent adverse events, which can be severe and even fatal. One of these adverse events—bone marrow suppression—is dose dependent. Approximately, 1 in 300 patients of European ancestry are homozygous for variants that lead to the poor TPMT metabolizer phenotype and 11 in 300 patients are heterozygous.¹⁰ For example, current CPIC guidelines recommend either adjustment of the dose or use of an alternative drug for heterozygous and homozygous patients with inflammatory conditions who receive azathioprine.¹¹ However, prior to the study by Pristup et al., there was no known TPMT endogenous substrate.

Molybdenum is a trace element, essential for almost all organisms.¹² Physiologically, there are several molybdenum dependent enzymes that catalase redox reactions; these reactions are controlled by molybdenum cofactors. In this issue, Pristup et al. report the novel finding that TPMT is a critical enzyme in the synthesis of urothione, the urinary excreted metabolite of molybdenum cofactor. Pristup and colleagues also suggest that the phenotypic consequences of TPMT variants could be different between the endogenous substrate (as urothione excretion was similar between TPMT wild-type and carriers of one nonfunctional TPMT allele) and the drug⁶ (as carriers of one nonfunctional TPMT allele are likely intermediate metabolizers).

In summary, these two papers highlight the importance of better understanding the role of the enzymes involved in high-risk pharmacokinetics, in particular as they pertain to drug classes with high rates of concentration-dependent side effects.

许多药物被酶灭活。因此，处于中间水平的患者，尤其是药物代谢不良的患者，活性药物代谢物的浓度可能更高，从而增加了发生浓度依赖性不良事件的风险。在本期《临床药理学和治疗学》中，有两篇手稿阐述了药代动力学的这一基本领域，重点关注代谢具有较窄治疗指数的药物的酶（图 1）。

第一篇感兴趣的论文是 White等人的综合评论。¹描述了二氢嘧啶脱氢酶 (DPD) 缺乏症的作用，并建议实施DPYD基因分型，以此作为识别严重不良事件高风险患者和个性化使用氟嘧啶的一种方法。DPYD基因编码 DPD，一种参与氟嘧啶（5-氟尿嘧啶、卡培他滨和替加氟）代谢的酶，氟嘧啶是几种常见癌症的常用化疗药物。然而，氟嘧啶类药物的使用受到各种常见副作用的限制，包括腹泻和骨髓抑制。因此，DYPD基因分型有助于识别携带DPYD功能变异并因此在表型上存在 DPD 缺陷的患者。该测试为在常规临床实践中进行基因型指导的氟嘧啶剂量调整提供了机会，得到了世界各地药物基因组专家的广泛支持。临床药物遗传学实施联盟 (CPIC) 目前的建议包括杂合子携带者的剂量调整，²荷兰药物基因组学工作组提出了基于四种DPYD变体的评分，可用于个性化治疗。^3个

在这种情况下，是否在常规临床护理中进行基因分型确实引起了一些潜在的担忧。首先，存在降低癌症治疗有效性的风险，因为先发制人的基因分型可能导致剂量减少。其他问题包括费用和治疗延误的可能性。然而，数据表明基于氟嘧啶的治疗的有效性不会降低癌症治疗的有效性，⁴并且，正如 Deenen等人。^5个参考，基于基因型信息的氟嘧啶剂量减少显着降低了严重毒性的发生率。尽管有多个监管机构的证据和基因型认可，但澳大利亚当局尚未发表立场声明。因此，在审查证据后，作者赞成对在澳大利亚常规临床实践中接受氟嘧啶治疗的患者实施基因分型的建议。

Pristup等人的论文。^图6描述了硫嘌呤 S-甲基转移酶 (TPMT) 的内源性底物，TPMT 是硫嘌呤代谢途径中的一种关键酶，包括硫唑嘌呤和 6-巯基嘌呤。硫嘌呤用于治疗患有癌症、自身免疫和其他炎症的患者，以及接受实体器官移植的患者。^7-9尽管有多种适应症，但硫嘌呤的使用受到频繁不良事件的限制，这些不良事件可能很严重甚至致命。这些不良事件之一——骨髓抑制——是剂量依赖性的。大约每 300 名欧洲血统患者中就有 1 名是导致 TPMT 代谢表型不良的变异的纯合子，每 300 名患者中有 11 名是杂合子。¹⁰例如，目前的 CPIC 指南建议对接受硫唑嘌呤的具有炎症的杂合子和纯合子患者调整剂量或使用替代药物。¹¹然而，在 Pristup等人的研究之前，没有已知的 TPMT 内源底物。

钼是一种微量元素，几乎对所有生物都是必需的。¹²在生理学上，有几种钼依赖酶可以催化氧化还原反应；这些反应由钼辅助因子控制。在这个问题上，Pristup等人。报告新发现，即 TPMT 是尿硫酮合成中的关键酶，尿硫酮是钼辅助因子的尿液排泄代谢物。Pristup 及其同事还提出，TPMT变异的表型结果在内源性底物之间可能不同（因为尿硫酮排泄在TPMT野生型和一个非功能性TPMT等位基因的携带者之间相似）和药物⁶（因为一种非功能性 TPMT 等位基因的携带者可能是中间代谢者）。

总之，这两篇论文强调了更好地理解酶在高风险药代动力学中的作用的重要性，特别是因为它们与具有高浓度依赖性副作用发生率的药物类别有关。