Thiopurine S-methyltransferase Phenotype Research Articles

Expressing thiopurineS-methyltransferase activity as units per litre of whole-blood overcomes misleading high results in patients with anaemia

Thiopurine S-methyltransferase (TPMT) phenotype analysis, expressed as TPMT activity, is established as a routine pharmacogenomic test to screen patients prior to initiating thiopurine drug therapy. Conventionally measured TPMT activity is corrected for red blood cell (RBC) parameters. Here we present evidence that supports the simplification of the TPMT assay: by expressing TPMT activity in mU/L whole blood, without undertaking any haemoglobin (Hb) correction. Hb concentrations were compared in consecutive samples that had been received for TPMT phenotype analysis and which were stratified into samples with high (n = 111) and samples with normal (n = 50) Hb-corrected enzyme activity. TPMT activity was also measured in samples received for full blood count determination, stratified into those with low (n = 50) and normal (n = 50) Hb. A reference interval for TPMT activity in mU/L was derived from a correlation between activity expressed in conventional units and that expressed in mU/L (n = 1563), supported by comparison with associated genotype (n = 201). In the high TPMT activity group, 83% of specimens had a low Hb concentration compared with 14% of specimens in the normal TPMT group. Samples with a low Hb concentration were found to have significantly higher Hb-corrected TPMT activity than samples with a normal Hb concentration: 83 versus 44 nmol 6-methyl thioguanine /g Hb/h, P < 0.0001. These results strongly suggest that misleading high Hb-corrected TPMT activity is found in anaemic patients. Based on the reference interval for enzyme activity of 70-150 mU/L, phenotype-genotype concordance compared well with the conventional approach (88% versus 89%). Furthermore, distribution of TPMT phenotypes with activity expressed in mU/L was identical: 0.5% deficient, 11% low, 86% normal and 2.5% high, to when it was expressed in conventional units. Expressing TPMT activity in mU/L can overcome misleading high Hb-corrected TPMT results occurring in patients with anaemia, which could lead to inappropriate treatment. Removing the need to measure RBC indices further simplifies TPMT phenotyping, leading to a more robust assay, with reduced turn-around time and cost.

Thiopurine S -Methyltransferase Pharmacogenetics in a Large-Scale Healthy Italian–Caucasian Population: Differences in Enzyme Activity

To investigate the influence of genotype, age and gender on the thiopurine S-methyltransferase (TPMT) phenotype in healthy Italian-Caucasian subjects. The study investigated the TPMT genotype and the TPMT phenotype of 943 healthy Italian-Caucasian subjects of different age and gender (age range: 0.08-68 years; 623 males 320 females). TPMT red blood cell activity was measured in all samples and genotype was determined for the TPMT alleles *2, *3A, *3B and *3C. TPMT activity levels in our whole population ranged from 1.6 up to 75.2 U/gHb. Significant TPMT activity differences between wild-type and heterozygous subjects were observed. We divided our TPMT activity into four categories according to our frequency distribution: low (0.1%), intermediate (32.9%), normal (60%) and high (7%), with arbitrary cut-off values of 8.0, 19.4 and 37.0 U/gHb, respectively. The whole population had a total of 94.5% of homozygous wild-type subjects, 5.4% heterozygous variants and one (0.1%) compound heterozygous variant TPMT*3B/*3C. The overall concordance rate between TPMT genotypes and phenotypes was 71.6%. The TPMT activity was significantly higher in wild-type children (0.08-17 years) than in wild-type adults (aged 18-68 years). Moreover, it was noted that wild-type infants from 0.08 to 5 years had a 9% higher average TPMT activity than the other wild-type groups, and only in children from 0.08 to 2 years was the TPMT activity higher in males than in females. The data obtained in this study show that genetic factors seem to be the major aspect in TPMT phenotype variability in adults, whilst, in children, other physiological factors should be taken into consideration when assessing the TPMT phenotype, such as age and gender.

Impact of the heterozygous TPMT*1/*3C genotype on azathioprine-induced myelosuppression in kidney transplant recipients in Thailand

Background: Thiopurine S-methyltransferase (TPMT) is a polymorphic enzyme associated with detoxification of azathioprine, an immunosuppressant used after renal transplantation in several Asian countries. Patients with variations of the TPMT gene may be at risk for myelosuppression after they receive a standard dosage of the drug. The frequency of TPMT*3C has been reported to be higher in the Thai population than in other Asian populations, possibly putting the Thais at higher risk for myelosuppression. Objective: The aim of this study was to assess the impact of the heterozygous TPMT*1/*3C genotype on azathioprine-induced myelosuppression in kidney transplant recipients in Thailand. Methods: This study was conducted at Srinagarind Hospital, Khon Kaen University, Khon Kaen, Thailand, and Chulalongkorn Hospital, Chulalongkorn University, Bangkok, Thailand. Eligible patients underwent kidney transplantation from deceased or living-related donors from 1984 to 2007. Electronic medical records were assessed retrospectively for the 6-month period after initiation of azathioprine treatment. TPMT genotyping and phenotyping were studied prospectively using real-time polymerase chain reaction and biochemical assay, respectively. The odds ratios (ORs), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were determined. Results: A total of 139 patients were enrolled (89 men, 50 women; median age, 42 years [range, 17–70 years]; mean weight, 58 kg [range, 37–87 kg]). The heterozygous TPMT*1/*3C genotype was found in 9 of the 139 patients (6.47%) (95% CI, 3.00–11.94). The TPMT activity of those patients was significantly lower than that of patients with the homozygous wild-type genotype (median, 21.37 vs 37.12 nmol 6-methylthioguanine/g · Hb/h, respectively; P < 0.001). The risk for azathioprine-induced myelosuppression in the patients with the heterozygous TPMT*1/*3C genotype was significantly higher than that in patients with the wild-type genotype (adjusted OR, 14.18 [95% CI, 3.07–65.40]; P < 0.005). The sensitivity and specificity of TPMT*3C genotyping for the prediction of azathioprine-induced myelosuppression in these kidney transplant recipients were 27% and 97%, respectively. Assuming a prevalence of azathioprine-induced myelotoxicity of 7% according to previously published data, the PPV and NPV were estimated to be 50% and 95%, respectively. Conclusion: In these kidney transplant recipients, patients who carried the TPMT*3C allele were at a higher risk for azathioprine-induced myelosuppres-sion than noncarriers.

Improving pharmacovigilance in Europe: TPMT genotyping and phenotyping in the UK and Spain

Thiopurine S-methyltransferase (TPMT) is the rate-limiting step in the conversion of thiopurine drugs including azathioprine (AZA) to inactive metabolites. Heritable deficiency of TPMT activity increases risk for adverse events, most notably, myelosuppression leading to leukopenia and neutropenic sepsis. The reported European Commission study was undertaken to identify current evidence for the clinical utility of testing for TPMT status and extent of uptake, by either genotyping or phenotyping, in the clinical setting. Data presented here for the UK and Spain indicate that there has been a considerable increase in the uptake of TPMT testing in recent years. There are some data that support routine TPMT testing before AZA prescribing for reducing AZA-related adverse events. Key data include evidence in favor of TPMT testing in addition to the current practice of routine monitoring for reducing the number of AZA-related episodes of myelosuppression, averting deaths from neutropenic sepsis and improving health-related quality of life. Further data are needed for determining the cost-effectiveness of routine TPMT testing.

Highly Multiplexed Genotyping of Thiopurine S-Methyltransferase Variants Using MALDI-TOF Mass Spectrometry: Reliable Genotyping in Different Ethnic Groups

To avoid severe hematotoxicity in patients, determination of the TPMT (thiopurine S-methyltransferase) genotype before commencing thiopurine therapy has become accepted. We used MALDI-TOF mass spectrometry (MS) based on Sequenom iPLEX technology to develop novel multiplex assays for comprehensive testing of TPMT. Two assays, a 15-plex and a 7-plex assay, consisting of multiplex PCR, shrimp alkaline phosphatase treatment, primer extension, and MALDI-TOF MS analysis, allow detection of all currently known functionally relevant 24 TPMT alleles (TPMT*2 to *18, *20 to *23). Previously identified variant DNA samples and newly constructed synthetic templates were used as quality controls. Assay evaluation performed on a panel of 586 genomic DNA samples previously genotyped by other methods (denaturing HPLC, sequencing) resulted in 100% agreement. Analyses of the distribution of TPMT alleles in 116 samples from a Ghanaian population revealed a TPMT*8 allele frequency of 3.4%. In a Korean population of 118 unrelated individuals, we found a TPMT*6 allele frequency of 1.3%. The newly developed multiplex MALDI-TOF MS assay allows efficient genotyping for all currently known functional TPMT variants. To achieve the most accurate prediction of TPMT phenotype, molecular diagnosis of TPMT should include all these variants.

Thiopurine methyltransferase Genotype Is Not a Risk Factor for Secondary Malignant Neoplasias after Treatment for Childhood Acute Lymphoblastic Leukemia on Berlin-Frankfurt-Muenster Protocols.

The thiopurines mercaptopurine and thioguanine are important components of contemporary polychemotherapeutic treatment protocols for acute leukemias. Thiopurines are prodrugs that undergo intestinal and hepatic metabolism. Activation occurs via a multistep pathway to form thioguanine nucleotides, which are thought to be the major cytotoxic compound through triggering cell cycle arrest and apoptosis. This process is in competition with direct inactivation of thiopurines or their metabolites by thiopurine S-methyltransferase (TPMT). TPMT is a cytosolic enzyme that is ubiquitously expressed in the human body and catalyzes the S-methylation of thiopurines. The TPMT locus is subject to genetic polymorphism, with heterozygous individuals (5 to 11% of Caucasians) having intermediate TPMT activity, and homozygous individuals (0.3 to 0.5% of Caucasians) having low TPMT activity. At least 20 variant TPMT alleles (*2 to *18) have been described so far that confer decreased enzyme activities compared to the TPMT*1 wild-type allele. TPMT genotype is highly concordant with TPMT phenotype. With regard to long-term adverse effects, patients who have diminished TPMT activity were shown to be at increased risk of developing chemotherapy-induced acute myeloid leukemia and radiation-induced second brain tumors tumors after exposure towards mercaptopurine during therapy for childhood acute lymphoblastic leukemia (ALL). To investigate if such an association is generalizable to other entities of secondary malignant neoplasms (SMN) and different treatment approaches for ALL, we collected specimens of 72 patients who developed a SMN after ALL treatment on Berlin-Frankfurt-Muenster (BFM) protocols, analyzed their TPMT genotype and compared genotype frequencies to these in the general ALL patient population. The 72 patient cohort consisted of 49 hematological SMN (half of them received cranial irradiation), 12 brain tumors, and 11 other solid SMN. Neither in the entire patient group nor in subgroup analyses, differences in allele frequencies of TPMT variants conferring diminished enzyme activity were detectable when comparing SMN patients to the overall ALL patient population. Thus, low TPMT activity does not seem to play a major role in the etiology of SMN after treatment for childhood ALL according to BFM treatment strategies. Factors potentially helpful for the explanation of the previously described relationsship of TPMT activity with SMN will be presented.

Whole-blood thiopurine S-methyltransferase activity with genotype concordance: a new, simplified phenotyping assay

We have developed a new thiopurine S-methyltransferase (TPMT) phenotyping method that measures TPMT activity in whole blood. To evaluate this assay, we compared it with conventional TPMT phenotyping, which uses a red blood cell (RBC) lysate and genotyping for analysis of common TPMT mutations. Whole-blood and RBC lysates were prepared from 402 patients' samples received for routine analysis. The TPMT activity of lysates was determined using 6-thioguanine as substrate with high-performance liquid chromatographic (HPLC) analysis and fluorimetric detection. DNA was extracted from buffy coats using phenol-chloroform extraction. A multiplex amplification refractory mutation system (ARMS) strategy was used to screen for the common TPMT mutations TPMT*2 and TPMT*3 (TPMT*3A, TPMT*3C and TPMT*3D). TPMT activities were higher in the whole-blood (mean TPMT activity 51 nmol 6-MTG/gHb/h) compared with the RBC lysate (37 nmol 6-MTG/gHb/h). Overall, concordance with TPMT genotypic analysis was 97% for both the new whole-blood and standard RBC lysate methods. Between low TPMT activity and heterozygotes, both phenotypic methods gave a concordance of 79%. Using multiplex ARMS testing for TPMT*2 and 3*C mutations to define the cut-off between low and normal TPMT activity, we have demonstrated that the new whole-blood TPMT phenotyping method performs as well as the conventional RBC lysate assay. This new method overcomes the need to prepare a RBC lysate, a process which is time consuming and increases analytical variation. The resulting assay is better suited to a regional or national TPMT phenotyping service.

Identification and functional analysis of two rare allelic variants of the thiopurine S-methyltransferase gene, TPMT*16 and TPMT*19

Human thiopurine S-methyltransferase (TPMT) catalyses the S-methylation of thiopurine drugs. TPMT is genetically polymorphic and is associated with large interindividual variations in thiopurine drug toxicity and therapeutic efficacy. During routine genotyping of patients with Crohn's disease, one novel missense mutation, 365A > C ( TPMT*19, Lys 122Thr), and a recently described missense mutation, 488G > A ( TPMT*16, Arg 163His), were identified in a Caucasian and a Moroccan patient, respectively. Using a heterologous yeast expression system, kinetic parameters ( K m and V max) of the two variants with respect to 6-thioguanine S-methylation were determined and compared with those obtained with the wild-type enzyme. The Lys 122Thr exchange did not significantly decrease the intrinsic clearance value ( V max/ K m) of the variant enzyme. In contrast, the Arg 163His substitution significantly decreased the intrinsic clearance value by three-fold. The Arg 163 is located in a highly conserved region of the human TPMT protein and, as such, the Arg 163His substitution is expected to result in a marked reduction of enzyme activity, as confirmed by the in vitro data. Phenotyping by measurement of red blood cell TPMT activity indicated that the patient heterozygous for the Lys 122Thr mutation had normal TPMT activity, whereas the patient heterozygous for the Arg 163His mutation was an intermediate methylator, which demonstrated a positive correlation between TPMT phenotyping and the in vitro data. The identification of a novel non-functional allele of the TPMT gene improves our knowledge of the genetic basis of interindividual variability in TPMT activity. These data further enhance the efficiency of genotyping methods to predict patients at risk of an inadequate response to thiopurine therapy.

Comprehensive analysis of thiopurine S-methyltransferase phenotype-genotype correlation in a large population of German-Caucasians and identification of novel TPMT variants.

The thiopurine S-methyltransferase (TPMT) genetic polymorphism has a significant clinical impact on the toxicity of thiopurine drugs. It has been proposed that the identification of patients who are at high risk for developing toxicity on the basis of genotyping could be used to individualize drug treatment. In the present study, phenotype-genotype correlation of 1214 healthy blood donors was investigated to determine the accuracy of genotyping for correct prediction of different TPMT phenotypes. In addition, the influence of gender, age, nicotine and caffeine intake was examined. TPMT red blood cell activity was measured in all samples and genotype was determined for the TPMT alleles *2 and *3. Discordant cases between phenotype and genotype were systematically sequenced. A clearly defined trimodal frequency distribution of TPMT activity was found with 0.6% deficient, 9.9% intermediate and 89.5% normal to high methylators. The frequencies of the mutant alleles were 4.4% (*3A), 0.4% (*3C) and 0.2% (*2). All seven TPMT deficient subjects were homozygous or compound heterozygous carriers for these alleles. In 17 individuals with intermediate TPMT activity discordant to TPMT genotype, four novel variants were identified leading to amino acid changes (K119T, Q42E, R163H, G71R). Taking these new variants into consideration, the overall concordance rate between TPMT genetics and phenotypes was 98.4%. Specificity, sensitivity and the positive and negative predictive power of the genotyping test were estimated to be higher than 90%. Thus, the results of this study provide a solid basis to predict TPMT phenotype in a Northern European Caucasian population by molecular diagnostics.

Pyrosequencing of TPMT alleles in a general Swedish population and in patients with inflammatory bowel disease.

Interindividual differences in therapeutic efficacy in patients treated with thiopurines might be explained by the presence of thiopurine S-methyltransferase (TPMT) alleles that encode for reduced TPMT enzymatic activity. It is therefore of value to know an individual's inherent capacity to express TPMT. We developed a pyrosequencing method to detect 10 single-nucleotide polymorphisms (SNPs) in TPMT. A Swedish population (n = 800) was examined for TPMT*3A, TPMT*3B, TPMT*3C, and TPMT*2. Patients with inflammatory bowel disease (n = 24) and healthy volunteers (n = 6), selected on the basis of TPMT enzymatic activity, were investigated for all 10 SNPs to determine the relationship between TPMT genotype and phenotype. In the general population we identified the following genotypes with nonfunctional alleles: TPMT*1/*3A (*3A allelic frequency, 3.75%), TPMT*1/*3C (*3C allelic frequency, 0.44%), TPMT*1/*3B (*3B allelic frequency, 0.13%), and TPMT*1/*2 (*2 allelic frequency, 0.06%). All nine individuals with normal enzymatic activity were wild-type TPMT*1/*1. Thirteen individuals with intermediate activity were either TPMT*1/*3A (n = 12) or TPMT*1/*2 (n = 1). Eight individuals with low enzymatic activity were TPMT*3A/*3A (n = 4), TPMT*3A/*3C (n = 2), or TPMT*1/*3A (n = 2). Next to wild type, the most frequent alleles in Sweden are TPMT*3A and TPMT*3C. A previously established phenotypic cutoff for distinguishing normal from intermediate metabolizers was confirmed. To identify the majority of cases (90%) with low or intermediate TPMT activity, it was sufficient to analyze individuals for only 3 of the 10 SNPs investigated. Nevertheless, this investigation indicates that other mutations might be of relevance for decreased enzymatic activity.

Determination of thiopurine S-methyltransferase phenotype using thin-layer chromatography and quantitative scanning.

To develop a non-high-performance liquid chromatography method for the determination of thiopurine- S-methyltransferase (TPMT) phenotype using thin-layer chromatography and quantitative scanning. TPMT reaction was performed using a radiochemical assay. The reaction product [(14)C]-6-methylmercaptopurine was separated using thin-layer chromatography and quantified by means of radioactive scanning. Day-to-day variance was determined to validate results. Determination of TPMT phenotype using thin-layer chromatography and quantitative scanning is reliable (day-to-day variance 8.5+/-1.7%, mean+/-SEM). Mean TPMT activity in 314 randomly selected patient samples was 11.8+/-3.3 units/ml red blood cells (mean+/-SD, range 3.5-25 units/ml). We developed a new assay variant for the determination of TPMT phenotype that is easy to perform, reliable and reduces production of radioactive waste. This may lead to more frequent pretreatment determination of TPMT phenotype and increase drug safety and efficacy by individualising thiopurine doses.

Preponderance of thiopurine S-methyltransferase deficiency and heterozygosity among patients intolerant to mercaptopurine or azathioprine.

To assess thiopurine S-methyltransferase (TPMT) phenotype and genotype in patients who were intolerant to treatment with mercaptopurine (MP) or azathioprine (AZA), and to evaluate their clinical management. TPMT phenotype and thiopurine metabolism were assessed in all patients referred between 1994 and 1999 for evaluation of excessive toxicity while receiving MP or AZA. TPMT activity was measured by radiochemical analysis, TPMT genotype was determined by mutation-specific polymerase chain reaction restriction fragment length polymorphism analyses for the TPMT*2, *3A, *3B, and *3C alleles, and thiopurine metabolites were measured by high-performance liquid chromatography. Of 23 patients evaluated, six had TPMT deficiency (activity < 5 U/mL of packed RBCs [pRBCs]; homozygous mutant), nine had intermediate TPMT activity (5 to 13 U/mL of pRBCs; heterozygotes), and eight had high TPMT activity (> 13.5 U/mL of pRBCs; homozygous wildtype). The 65.2% frequency of TPMT-deficient and heterozygous individuals among these toxic patients is significantly greater than the expected 10% frequency in the general population (P <.001, chi(2)). TPMT phenotype and genotype were concordant in all TPMT-deficient and all homozygous-wildtype patients, whereas five patients with heterozygous phenotypes did not have a TPMT mutation detected. Before thiopurine dosage adjustments, TPMT-deficient patients experienced more frequent hospitalization, more platelet transfusions, and more missed doses of chemotherapy. Hematologic toxicity occurred in more than 90% of patients, whereas hepatotoxicity occurred in six patients (26%). Both patients who presented with only hepatic toxicity had a homozygous-wildtype TPMT phenotype. After adjustment of thiopurine dosages, the TPMT-deficient and heterozygous patients tolerated therapy without acute toxicity. There is a significant (> six-fold) overrepresentation of TPMT deficiency or heterozygosity among patients developing dose-limiting hematopoietic toxicity from therapy containing thiopurines. However, with appropriate dosage adjustments, TPMT-deficient and heterozygous patients can be treated with thiopurines, without acute dose-limiting toxicity.

Pharmacogenetics: a tool for individualizing antineoplastic therapy.

This article reviews the clinical relevance of pharmacogenetics in cancer chemotherapy, with emphasis on drugs for which genetic differences in enzyme metabolism have been demonstrated to affect patient outcome. About 10% of children with leukaemia are intolerant to mercaptopurine (6-mercaptopurine) because of genetic defects in mercaptopurine inactivation by thiopurine S-methyltransferase. However, mercaptopurine dose intensity, a critical factor for outcome in patients deficient in thiopurine S-methyltransferase, can be maintained by means of thiopurine S-methyltransferase phenotyping or genotyping. Patients with reduced fluorouracil (5-fluorouracil) catabolism are more likely to be exposed to severe toxicity. The measurement of dihydropyrimidine dehydrogenase activity in patients cannot be considered fully predictive, and the role of dihydropyrimidine dehydrogenase gene variants in this syndrome has yet to be clarified. With regard to irinotecan, patients with Gilbert's syndrome phenotype have reduced inactivation of the active topoisomerase I inhibitor 7-ethyl-10-hydroxycamptothecin (SN-38) caused by a mutation in the UDP-glucuronosyltransferase 1A1 gene promoter. This subset of patients is more likely to be exposed to irinotecan toxicity and could be identified by genotyping for gene promoter variants. Finally, the experience with amonafide represents a model for dose individualization approaches that use simple phenotypic probes.

Isolation of a human thiopurine S-methyltransferase (TPMT) complementary DNA with a single nucleotide transition A719G (TPMT*3C) and its association with loss of TPMT protein and catalytic activity in humans.

Thiopurine S-methyltransferase (TPMT) is a cytosolic enzyme that catalyzes the S-methylation of mercaptopurine, azathioprine, thioguanine and most of their nucleotide metabolites. TPMT exhibits genetic polymorphism, with about 10% of individuals having intermediate TPMT activity because of heterozygosity at the TPMT locus and about 1 in 300 inheriting TPMT deficiency as an autosomal recessive trait. Although several mutant alleles have now been associated with inheritance of TPMT deficiency in humans, the expression of only TPMT*2 and TPMT*3A has been established by isolation and characterization of complementary DNA (cDNA) from individuals with low TPMT activity. Radiochemical assay, Western blot analysis, polymerase chain reaction (PCR) genotyping, and cDNA sequencing were used to analyze TPMT activity and protein levels in erythrocytes and to determine TPMT genotype. We established expression of another common mutant allele, TPMT*3C (containing only the A719G mutation), by sequence analysis of cDNA isolated from an individual with a heterozygous TPMT phenotype (7 units/ml packed erythrocytes). The TPMT*3C allele was also confirmed in an unrelated individual by sequencing TPMT coding exons after PCR amplification of genomic DNA. Moreover, Western blot analysis of erythrocytes obtained from five heterozygous individuals with the TPMT*3C allele (i.e., TPMT*1/TPMT*3C) exhibited about 50% less immunodetectable TPMT protein compared with homozygous wild-type individuals, and a TPMT-deficient individual with a TPMT*3A/TPMT*3C genotype had no immunodetectable TPMT protein. These data establish that the TPMT*3C allele is expressed in humans and is associated with lower immunodetectable TPMT protein and catalytic activity.

Molecular diagnosis of thiopurine S-methyltransferase deficiency: genetic basis for azathioprine and mercaptopurine intolerance.

Thiopurine S-methyltransferase (TPMT) catalyzes the S-methylation (that is, inactivation) of mercaptopurine, azathioprine, and thioguanine and exhibits genetic polymorphism. About 10% of patients have intermediate TPMT activity because of heterozygosity, and about 1 in 300 inherit TPMT deficiency as an autosomal recessive trait. If they receive standard doses of thiopurine medications (for example, 75 mg/m2 body surface area per day), TPMT-deficient patients accumulate excessive thioguanine nucleotides in hematopoietic tissues, which leads to severe and possibly fatal myelosuppression. To elucidate the genetic basis and develop molecular methods for the diagnosis of TPMT deficiency and heterozygosity. Diagnostic test evaluation. Research hospital. The TPMT phenotype was determined in 282 unrelated white persons, and TPMT genotype was determined in all persons who had intermediate TPMT activity (heterozygotes) and a randomly selected, equal number of persons who had high activity. In addition, genotype was determined in 6 TPMT-deficient patients. Polymerase chain reaction (PCR) assays were developed to detect the G238C transversion in TPMT*2 and the G460A and A719G transitions in TPMT*3 alleles. Radiochemical assay was used to measure TPMT activity. Mutations of TPMT were identified in genomic DNA, and the concordance of TPMT genotype and phenotype was determined. 21 patients who had a heterozygous phenotype were identified (7.4% of sample [95% CI, 4.7% to 11.2%]). TPMT*3A was the most prevalent mutant allele (18 of 21 mutant alleles in heterozygotes; 85%); TPMT*2 and TPMT*3C were more rare (about 5% each). All 6 patients who had TPMT deficiency had two mutant alleles, 20 of 21 patients (95% [CI, 76% to 99.9%]) who had intermediate TPMT activity had one mutant allele, and 21 of 21 patients (100% [CI, 83% to 100%]) who had high activity had no known TPMT mutation. Detection of TPMT mutations in genomic DNA by PCR coincided perfectly with genotypes detected by complementary DNA sequencing. The major inactivating mutations at the human TPMT locus have been identified and can be reliably detected by PCR-based methods, which show an excellent concordance between genotype and phenotype. The detection of TPMT mutations provides a molecular diagnostic method for prospectively identifying TPMT-deficient and heterozygous patients.