Rapid detection of the irinotecan-related UGT1A1*28 polymorphism by asymmetric PCR melting curve analysis using one fluorescent probe.

BACKGROUND: Determination of UGT1A1 (TA)n polymorphism prior to irinotecan therapy is necessary to avoid severe adverse drug effects. Thus, accurate and reliable genotyping methods for (TA)n polymorphism are highly desired. Here, we present a new method for polymerase chain reaction (PCR) melting curve analysis using one fluorescent probe to discriminate the UGT1A1*1 [(TA)6 ] and *28 [(TA)7 ] genotypes.
METHODS: After protocol optimization, this technique was applied for genotyping of 64 patients (including 23 with UGT1A1*1/*1, 22 with *1/*28, and 19 with *28/*28) recruited between 2016 and 2021 in China-Japan Friendship Hospital. The accuracy of the method was evaluated by comparing the results with those of direct sequencing and fragment analysis. The intra- and inter-run precision of the melting temperatures (Tm s) were calculated to assess the reliability, and the limit of detection was examined to assess the sensitivity.
RESULTS: All genotypes were correctly identified with the new method, and its accuracy was higher than that of fragment analysis. The intra- and inter-run coefficients of variation for the Tm s were both ≤0.27%, with standard deviations ≤0.14°C. The limit of detection was 0.2 ng of input genomic DNA.
CONCLUSION: The developed PCR melting curve analysis using one fluorescent probe can provide accurate, reliable, rapid, simple, and low-cost detection of UGT1A1 (TA)n polymorphism, and its use can be easily generalized in clinical laboratories with a fluorescent PCR platform.

INTRODUCTION

Glucuronidation, which is catalyzed by the UDP‐glucuronosyltransferases (UGTs), has an essential role in the metabolism of xenobiotic and lipophilic endobiotic substrates, such as bilirubin and the chemotherapeutic drug irinotecan. UGT1A1, belonging to the UGT1A subfamily, is responsible for hepatic glucuronidation of bilirubin. The UGT1A1 gene is located on chromosome 2q37. Multiple variations in the UGT1A1 gene can alter the enzyme activity of UGT1A1 protein. , , Among them, the promotor (TA)n polymorphism (rs3064744) is one of the most frequently investigated. Its wild‐type sequence contains six repeats of TA [(TA)6, UGT1A1*1], whereas the other alleles contain either five TA repeats [(TA)5, UGT1A1*36] with normal or increased UGT1A1 expression or seven [(TA)7, UGT1A1*28] and eight TA repeats [(TA)8, UGT1A1*37] with decreased UGT1A1 expression and thus a reduction of its enzyme activity. , , UGT1A1*28 is the most common (TA)n variant, with reported allelic frequencies of 29%–45% in Caucasians, 42%–51% in Africans, and 16% in Asians. The UGT1A1*36 and *37 variants occur almost exclusively in Africans. ,

Decreased UGT1A1 enzyme activity leads to disordered glucuronidation, resulting in defects in bilirubin metabolism. Individuals who are homozygous or compound heterozygous for the UGT1A1*28 and *37 alleles develop the inherited Gilbert's syndrome (OMIM 143500) characterized by mild unconjugated hyperbilirubinemia, which is commonly a benign condition, and a more aggressive childhood subtype, Crigler‐Najjar syndrome (OMIM 218800, 606785). ,

More importantly, extensive studies indicate that UGT1A1*28 carriers (homozygotes and heterozygotes) have a significantly higher risk for life‐threatening, adverse effects from irinotecan treatment in multiple races and populations. , , , , , Irinotecan, a camptothecin analog, is a chemotherapeutic drug widely used in the treatment of metastatic colorectal cancer (first‐line treatment) and occasionally used for treating other solid tumors. A significant proportion of patients taking irinotecan experience life‐threatening adverse effects including leukopenia, neutropenia, and/or diarrhea, due to the reduced elimination of SN‐38 (7‐ethyl‐10‐hydroxycamptothecin), the active metabolite of irinotecan, which is primarily glucuronidated by hepatic UGT1A1 protein. , , A meta‐analysis based on 58 studies including 6087 cancer patients showed that individuals carrying UGT1A1*1/*28 and *28/*28 have a greater prevalence of diarrhea and neutropenia than those carrying UGT1A1*1/*1 with odds ratios of 2.18 and 2.15, respectively, specifically for patients with metastatic colorectal cancer. In 2005, the Food and Drug Administration (FDA) added the genetic status of (TA)n polymorphism to the drug label for irinotecan, recommending that patients with a UGT1A1*28/*28 genotype should receive a lower starting dose of irinotecan. Moreover, the results from an updated meta‐analysis conducted in 2010 involving 1998 cancer patients indicate that the UGT1A1*28/*28 genotype is associated with a 2‐fold increased risk of neutropenia not only at medium doses but also at low doses. Thus, there is an urgent need to develop rapid and accurate genotyping methods for (TA)n polymorphism. If the (TA)n genotype is determined prior to the therapy initiation, the occurrence of the adverse reactions to irinotecan could be prevented.

To date, multiple assays have been applied in attempts to facilitate precision irinotecan therapy, including denaturing gradient gel electrophoresis, dual hybridization probe melting analysis, , , , SYBR Green I melting analysis, a single‐strand conformation polymorphism (SSCP) method, pyrosequencing, , , the FDA‐approved Invader® assay, fragment analysis, , , hydrolysis probes, , high‐resolution melting (HRM) curve analysis, , a denaturing high‐performance liquid chromatography (DHPLC) method, a microarray with LNA‐probes, a restriction fragment length polymorphism (RFLP) method, and a three‐dimensional polyacrylamide gel‐based DNA microarray. Two studies have compared these established genotyping methods. Baudhuin et al. compared the direct sequencing, fragment analysis, and Invader® assay methods. They found that although all samples had concordant genotypes, the interpretation of sequencing data was challenging, and the Invader® assay required more concentrated DNA and was more expensive. In 2020, Sissung et al. compared eight (TA)n polymorphism genotyping technologies (i.e., direct sequencing, pyrosequencing, gel sizing, DMET Plus arrays, Pharmacoscan arrays, Illumina MiSeq, fragment analysis, and fluorescent PCR), and from their results, they recommended that all genotyping be conducted with fluorescent PCR, as the results from the other platforms were often ambiguous or incorrect. They also suggested that a novel methodology based on fluorescent PCR will be a promising direction for the development of an accurate and reliable genotyping platform for (TA)n polymorphism that can easily be generalized.

In the present study, we developed a novel method for UGT1A1*1/*1, *1/*28, and *28/*28 genotyping based on asymmetric PCR and melting curve analysis with one fluorescent probe and demonstrated that this method is accurate, stable, rapid, and simple with cost‐efficient performance.

MATERIALS AND METHODS

Patients

Sixty‐four participants, who previously underwent UGT1A1 (TA)n promoter polymorphism genotyping by fragment analysis, were recruited between 2016 and 2021 in China‐Japan Friendship Hospital. Written informed consent was obtained for all participants. The study protocol was developed in accordance with the Declaration of Helsinki II and approved by the ethics committee of the hospital. Two‐milliliters of whole blood samples from all participants were collected into EDTA‐anticoagulated tubes and stored at 4°C before DNA extraction. The clinical characteristics of the participants are presented in Table S1.

DNA sample preparation

Genomic DNA was extracted using a DNA extraction kit (Tianlong Science and Technology Co. Ltd) according to the manufacturer's instruction. The DNA was dissolved in TE buffer (10 mmol/L Tris and 0.1 mmol/L EDTA, pH 8.0) to 30–60 ng/μl as measured at 260 nm by a NanoDrop 1000 (Thermo Fisher Scientific) and then stored at −20°C. For sensitive analyses, DNA was diluted to 0.1–0.2 ng/μl using nuclease‐free water (Ambion, Life Technologies Corp).

Plasmid construction

Plasmids with UGT1A1 (TA)n promoter genotypes UGT1A1*1 and *28 were obtained from Sangon Biotech Co., Ltd. In brief, a fragment containing the UGT1A1*1 or *28 polymorphism was inserted into the plasmid pUC57. The insert sequences of plasmids are presented in Table S2. The heterozygote of UGT1A1*1 and *28 alleles was constructed by mixing the plasmids equally.

Fragment analysis using capillary electrophoresis

The initial genotyping of UGT1A1 (TA)n promoter polymorphisms was achieved using PCR amplification and fragment analysis. The sequences of PCR primer pairs were designed as follows: 5’–FAM‐CTCCCTGCTACCTTTGTGGACTGA–3′ (forward primer; FAM is a fluorescent dye with emission at 518 nm and excitation at 492 nm; TaKaRa Bio, Inc.), 5’–ACAACGAGGCGTCAGGTGCTA–3′ (reverse primer; TaKaRa Bio Inc.). PCR was carried out in a 10‐μl reaction volume containing 30–60 ng of genomic DNA and GoTaq@ DNA Polymerase supplied in 2 × GoTaq@ Green Reaction Buffer (pH 8.5), dNTP (400 μM each) and 3 mM MgCl2 (M7122: Promega), with 0.4 μM of each primer. The PCR cycling program consisted of initial denaturation at 94°C for 5 min followed by 40 cycles of 94°C for 15 s, 55°C for 25 s, and 72°C for 50 s, and a final extension at 72°C for 10 min was performed (C1000: Touch™ Thermal Cycler, Bio‐Rad Laboratories, Inc.). The amplified product for UGT1A1*1 was 339 bp, and that for UGT1A1*28 was 341 bp.

The amplicons were diluted 1:20 with nuclease‐free water (Ambion, Life Technologies Corp.). Then, 1 μl of the diluted DNA was mixed with 9 μl Hi‐Di™ formamide (Applied Biosystems) and 1 μl genescan™ 500 LIZ® Size Standard (Applied Biosystems). The mixtures were denatured at 95°C for 5 min using a C1000 Touch™ Thermal Cycler (Bio‐Rad Laboratories, Inc.) and then quenched on ice. Samples were separated by capillary electrophoresis on the ABI 3500 Genetic Analyzer (Applied Biosystems) using POP‐7™ Performance Optimized Polymer (Applied Biosystems) and a 50‐cm capillary array following the parameters below: 60°C run temperature, 15 kV pre‐run voltage for 180 s, 1.6 kV injection voltage for 8 s, and a 19.5 kV run voltage for 1330 s with a 1‐s data delay. Sample migration distances were analyzed using genemapper® Software (Applied Biosystems) to determine the genotype.

Asymmetric PCR and melting curve analysis with one fluorescent probe

The PCR primers were designed as follows: 5’–TGAACTCCCTGCTACCTTTG–3′ (forward primer; Sangon Biotech Co., Ltd.), and 5’–CAACAGTATCTTCCCAGCAT–3′ (reverse primer; Sangon Biotech Co., Ltd.). The fluorescent probe used to detect (TA)n polymorphism was designed as 5’–BHQ2‐GCCATATATATATATATAAGTAGG‐Cy5–3′ (BHQ2 is a quencher dye; Cy5 is a fluorescent dye with emission at 670 nm and excitation at 649 nm; Sangon Biotech Co., Ltd.).

For PCR, different concentrations of Mg2+ (1, 2, 3, 4, and 5 nM) as well as different proportions of forward and reverse primers (forward primer: reverse primer = 2:1, 1:1, 1:2, 1:4, 1:8, 1:16) were tested to optimize the protocol. Finally, PCR was carried out in a 25‐μl optimized reaction mixture containing 150–300 ng of genomic DNA and TaKaRa Ex Taq® Polymerase (0.125 U) supplied in 10 × Ex Taq Buffer (pH 8.5), dNTP (200 μM each), and 2 mM MgCl2 (RR01AM, TaKaRa Bio, Inc.), with 0.1 μM forward primer, 0.8 μM reverse primer, and 0.4 μM probe.

The PCR amplification and melting curve analysis were performed on a slan®‐96P fluorescent quantitative PCR system (Hongshitech). The PCR cycling program consisted of initial denaturation at 95°C for 5 min followed by 50 cycles of 95°C for 20 s and 60°C for 1 min for amplification. The amplified products for UGT1A1*1 and *28 were 240 bp and 242 bp, respectively. The melting curve program included three steps: denaturation at 95°C for 2 min, renaturation at 45°C for 2 min, and subsequent melting with continuous acquisition of fluorescence from 45 to 75°C at a ramp rate of 0.08°C/s.

The analytical sensitivity of the present method was evaluated by examining its performance with varying amounts of input genomic DNA used for PCR. We selected two samples of each genotype with an initial concentration ranging from 31.9 to 40.5 ng/μl, and prepared doubling dilutions 11 times to a lowest concentration ranging from 0.016 to 0.020 ng/μl (Table S3). Thus, an input DNA amount as low as 0.1 ng was used to test the limit of detection.

Sanger sequencing

The genotype of each patient was confirmed by Sanger sequencing. PCR was performed as described for fragment analysis except that the forward primer was not fluorescently labeled, and the reaction volume was 50 μl, containing 150–300 ng of genomic DNA. The amplicons were then sent to Tsingke Biotechnology Co., Ltd. for unidirectional sequencing using an ABI 3730xl DNA Analyzer (Applied Biosystems).

Statistics

The data are presented as mean ± standard deviation (SD), and the coefficients of variation (CVs) were calculated using sas (version 9.3; SAS Institute).

RESULTS

Protocol optimization

To optimize the protocol for the new melting curve method, different concentrations of Mg2+ (1, 2, 3, 4, and 5 nM) as well as different proportions of forward and reverse primers (F: R = 2:1, 1:1, 1:2, 1:4, 1:8, 1:16) in the PCR reaction mixture, were tested. As shown in Figures S1 and S2, 2 nM of Mg2+ and an F:R ratio of 1:8 were finally chosen as the optimized conditions for subsequent analyses.

Accuracy

We used three methods for genotyping the (TA)n polymorphism in 64 patients (Figure 1). Using Sanger sequencing as the gold standard for genotyping, DNA samples were first genotyped via a reference method, fragment analysis using capillary electrophoresis. The genotypes of the 64 patients included UGT1A1*1/*1 in 23 patients, *1/*28 in 22 patients, and *28/*28 in 19 patients. The melting curve approach was then performed blindly with fragment analysis. The accuracy of this melting curve analysis was validated and compared with fragment analysis. Among the samples, the results of melting curve analysis reached 100% concordance with Sanger sequencing in all genotypes. The fragment analysis showed 100% concordance with Sanger sequencing for UGT1A1*1/*1 and *1/*28 samples. However, for UGT1A1*28/*28, when applying fragment analysis, two of the 19 *28/*28 individuals were misclassified as *1/*28, resulting in 89.47% concordance with Sanger sequencing (Table 1). These results indicated that the melting curve analysis offered a higher accuracy than fragment analysis.

FIGURE 1

Representative melting curve and electropherograms for UGT1A1*1/*1, *1/*28, and *28/*28 genotypes in patients. (A) Melting curve analysis with one fluorescent probe. (B) Fragment analysis. (C) Sanger sequencing

TABLE 1

Accuracy comparison of PCR melting curve analysis versus fragment analysis for UGT1A1*1/*1, *1/*28, and *28/*28 genotyping

Sanger sequencing genotype	N	Fragment analysis		PCR melting curve analysis using fluorescent probe
		N Correct/N Total	Concordance, %	N Correct/N Total	Concordance, %
UGT1A1*1/*1	23	23/23	100	23/23	100
UGT1A1*1/*28	22	22/22	100	22/22	100
UGT1A1*28/*28	19	17/19	89.47	19/19	100

The melting temperatures (Tms) of the UGT1A1*1 peak in *1/*1 and *1/*28 samples were 56.50 ± 0.09°C and 56.67 ± 0.08°C, respectively. For the UGT1A1*28 peak, the Tms in *28/*28 and *1/*28 samples was 52.28 ± 0.09°C and 52.07 ± 0.09°C, respectively.

The melting curve analysis was also validated by analysis of three replicates of each genotype using plasmids (Figure S3). The T ms of the UGT1A1*1 peak for the *1 plasmid and mixture of *1 and *28 plasmids were 56.29 ± 0.05°C and 56.66 ± 0.02°C, respectively. The T ms of the UGT1A1*28 peak in the *28 plasmid and mixture of *1 and *28 plasmids were 52.30 ± 0.03°C and 52.10 ± 0.03°C, respectively.

Precision

We evaluated the diagnostic reliability of the newly developed method. Five replicates of one sample of each genotype were tested on four independent days by different operators to determine the intra‐run and inter‐run precision. As shown in Table 2, we obtained an intra‐run T m CV ≤ 0.27% (ranging from 0.03% to 0.27%) with an SD ≤ 0.14°C (ranging from 0.02 to 0.14°C) and an inter‐run T m CV ≤ 0.27% (ranging from 0.17% to 0.27%) with an SD ≤ 0.14°C (ranging from 0.09 to 0.14°C). The melting curves also showed good reproducibility (Figure S4).

TABLE 2

Inter‐ and intra‐run mean ± SDs precision values of Tms for UGT1A1*1/*1, *1/*28 and *28/*28

Genotype	Day	UGT1A1*1 peak				UGT1A1*28 peak
		Intra‐run T m		Inter‐run T m		Intra‐run T m		Inter‐run T m
		Mean ± SD, °C	CV (%)	Mean ± SD, °C	CV (%)	Mean ± SD, °C	CV (%)	Mean ± SD, °C	CV (%)
UGT1A1*1/*1	1	56.47 ± 0.04	0.06	56.41 ± 0.09	0.17	/	/	/	/
	2	56.40 ± 0.03	0.05			/	/
	3	56.48 ± 0.08	0.14			/	/
	4	56.28 ± 0.06	0.10			/	/
UGT1A1*1/*28	1	56.62 ± 0.12	0.22	56.66 ± 0.13	0.23	52.05 ± 0.14	0.27	52.06 ± 0.14	0.27
	2	56.68 ± 0.02	0.03			52.11 ± 0.03	0.05
	3	56.82 ± 0.03	0.06			52.20 ± 0.03	0.06
	4	56.52 ± 0.08	0.14			51.89 ± 0.10	0.18
UGT1A1*28/*28	1	/	/	/	/	52.34 ± 0.13	0.25	52.31 ± 0.13	0.25
	2	/	/			52.34 ± 0.05	0.09
	3	/	/			52.40 ± 0.09	0.18
	4	/	/			52.15 ± 0.08	0.15

Limit of detection

To evaluate the limit of detection for the melting curve method, we diluted two samples of each genotype to an input DNA amount of 0.1 ng (Table S3). The present method could detect each genotype correctly with a sensitivity as low as 0.2 ng (Figure S5).

DISCUSSION

Genotyping of the UGT1A1 (TA)n promoter polymorphism prior to irinotecan therapy is of great importance to minimizing the risk of severe adverse drug effects linked to the UGT1A1*28 variant, which can cause reduced gene transcription and deficient UGT1A1 enzyme activity. Clinical trials have suggested that irinotecan therapy guided by UGT1A1 status can significantly increase the likelihood of complete tumor response and achieve a favorable clinical outcome without significantly increased toxicities. , , , The FDA has already recommended a reduced initial dose of irinotecan for patients carrying the UGT1A1*28 variant. Therefore, an accurate, rapid, simple, and reliable genotyping assay for the (TA)n polymorphism, especially for the UGT1A1*1 and *28 alleles, is required urgently.

The gold standard method for UGT1A1 (TA)n promoter genotyping is direct Sanger sequencing, but it is laborious, time‐consuming, and expensive. Fragment analysis by capillary electrophoresis is frequently used, , , but the operating procedure is complicated and also time‐consuming. Additionally, fragment analysis unavoidably returns a range of fragment sizes [stutter (n–1) repeats shorter than the allele] in addition to the targeted fragment due to Taq polymerase “slippage”, which might cause operator bias when discriminating genotypes, especially among inexperienced operators. , , In the present study, the recruited patients were initially genotyped using fragment analysis, and two UGT1A1*28/*28 samples were misclassified as *1/*28, due to misjudgment of stutter peaks by the operators. In addition, for UGT1A1*1/*28 heterozygotes, the amplification of the *1 allele was stronger than that of *28 allele, and thus, the judgment of the *28 peak might be subjectively influenced by the operator. Moreover, both methods require transfer of the PCR amplification products, which could induce laboratory contamination. A DNA sequencing platform is required for both techniques, which is expensive and may not be available in some clinical laboratories. These platforms also require routine, time‐consuming, and labor‐intensive maintenance. In fact, in China for example, although direct sequencing and fragment analysis are the only two methods approved by the National Medical Products Administration of China (also known as CFDA) for UGT1A1 (TA)n promoter polymorphism genotyping, a gene sequencer is not available in most clinical laboratories. Thus, the testing cannot be performed locally, and the results are therefore delayed for patients.

Melting curve analysis based on fluorescent PCR has been adapted for (TA)n polymorphism genotyping. Among the established methodologies, a melting curve with dual hybridization fluorescent probes has been most frequently reported. , , , These approaches are accurate and offer good reliability. However, they require at least two probes for the discrimination of UGT1A1*1/*1, *1/*28, and *28/*28 genotypes. Dye‐based melting curve analyses have also been applied, which only require intercalation dyes of double‐stranded DNA. A method based on SYBR Green I showed a T m difference of only 1.3°C between the 132‐bp products of UGT1A1*1 homozygotes and the 134‐bp products of *28 homozygotes, which resulted in limited resolution and difficultly identifying the difference. An HRM approach based on LC‐Green was reported to successfully distinguish UGT1A1*1/*1, *1/*28, and *28/*28 genotypes from a 70‐bp product with LC‐Green. Another technology based on PCR using a snapback primer and genotyping by HRM was able to discriminate the other two rare genotypes. However, generating reproducible melting curves is always a challenge for HRM analysis, which requires high‐resolution instrumentation and targeted software. Moreover, dyes cannot distinguish the target amplicons versus non‐specific PCR products, and thus, the by‐products of the PCR system adversely affect the test performance, greatly limiting its application.

In the present study, we developed a cost‐saving method based on asymmetric PCR and a one fluorescent probe‐mediated melting curve method to distinguish UGT1A1*1/*1, *1/*28, and *28/*28, and our analysis showed that the method is accurate and sensitive with outstanding reliability. Our approach represents the first melting curve method combined with asymmetric PCR reported for (TA)n genotyping that requires only one fluorescent probe. The experimental procedure is simple: asymmetric PCR is used to obtain excess copies of single‐stranded amplicons, and the probe is hybridized to the targeted amplicons at low temperature and dissociated as the temperature increases during the melting analysis process. Compared with other methods, the operation requires less time and labor, while the judgment of genotypes is very simple and clear. The probe was designed to perfectly match the UGT1A1*1 sequence, resulting in a stable double‐stranded product and a high T m when binding to the *1 allele. Meanwhile, the two base pair mismatches caused a 2‐bp bulge between the probe and UGT1A1*28 allele, resulting in a less stable product and a lower T m. The difference in T m between the UGT1A1*1 and *28 peaks is > 4.2°C, which provides good discrimination of the three genotypes. The intra‐ and inter‐assay precision levels were satisfactory at lower than 0.27%, suggesting a better reliability than either the dual‐probes melting analysis method or TaqMan real‐time PCR method, and the CV for our method was similar to that of the SYBR Green I melting method. Thus, the presented method is able to distinguish the genotypes using the absolute T m values of the PCR products, and only one UGT1A1*1/*28 genotype sample is required as a positive control for each run. The method also offers good sensitivity and affords accurate genotyping with an input genomic DNA amount as low as 0.2 ng. Moreover, as a closed‐tube PCR assay, the risk of contamination is eliminated in the present method. Therefore, we believe that the method can be well‐generalized among different clinical laboratories with good reproducibility when applied using the conditions established in this study.

In addition to genetic testing of UGT1A1, a novel methodology to directly measure serum SN‐38G and SN‐38 will also be helpful to prevent severe adverse effects of irinotecan, as previous studies revealed that the area under the curve (AUC) ratio for SN‐38 glucuronide (SN‐38G)/SN‐38 could be a clinical indicator of irinotecan toxicity and applied for adjustment of the optimal dose. Atasilp et al. developed a technique based on high‐performance liquid chromatography (HPLC)/tandem mass spectrometry (MS/MS) that can simultaneously measure the serum concentrations of irinotecan, SN‐38, and SN‐38G, which may serve as an alternative method to predicting irinotecan toxicity. Studies previously reported that UGT1A1*28 carriers have high AUC values for irinotecan and SN‐38, but a low AUC ratio for SN‐38G/SN‐38. Thus, we speculated that the combination of detection of irinotecan and its metabolites with genetic testing will allow for more precise use of irinotecan.

In addition to UGT1A1*28, UGT1A1*6 (rs4148323), a single nucleotide substitution located in exon 1 that occurs at a relative high frequency in Asians (~20%) can cause an obvious reduction in UGT1A1 enzyme activity and lead to irinotecan‐induced diarrhea and neutropenia. , , , , Combined genotyping and interpretation of both UGT1A1*28 and *6 will more fully predict and avoid the adverse effects of irinotecan, especially for Asian populations. Thus, we will further develop a rapid and reliable genotyping method for both UGT1A1*28 and *6 in the future.

In conclusion, the present method has the following benefits for the discrimination of UGT1A1*1 and *28 alleles: (a) high accuracy with satisfactory reliability (intra‐ and inter‐ run CVs for Tms were < 0.27%); (b) high sensitivity (limit of detection was 0.2 ng genomic DNA); (c) simple and labor‐saving operation (basically PCR analysis, suitable for both experienced technicians and beginners); (d) avoidance of contamination of PCR products (the melting curve of the amplification product is directly analyzed in a closed‐tube); (e) low cost (requiring one fluorescent probe and one positive control for each run); (f) quick operation (the genotype can be identified within 2.5 h); and (g) high‐throughput potential in a 96‐well or 384‐well PCR analyzer. Therefore, we expect this method can be simply and easily generalized for any clinical laboratory with a fluorescent PCR platform.

CONFLICT OF INTEREST

The authors declare that no conflicts of interest exist.

Table S1

Click here for additional data file.

Table S2

Click here for additional data file.

Table S3

Click here for additional data file.

Figure S1

Click here for additional data file.

Figure S2.

Click here for additional data file.

Figure S3

Click here for additional data file.

Figure S4

Click here for additional data file.

Figure S5

Click here for additional data file.