Testing for thiopurine S-methyltransferase and nudix hydrolase 15 gene polymorphism during induction therapy in acute lymphoblastic leukemia

INTRODUCTION

6-Mercaptopurine (6-MP) is an anti-metabolite and myelosuppressive agent belonging to a broader class of drugs called thiopurines, which also includes thioguanine and azathioprine. These purine analogs act primarily by interfering with nucleic acid synthesis and have found their use not only in the management of myeloid and lymphoid malignancies but also in various autoimmune conditions.^[1] 6-MP is the most commonly prescribed drug in the treatment of acute lymphoblastic leukemia (ALL). It plays an important role in both induction and maintenance therapy of most approved protocols for over 40 years.

The metabolism of this drug mainly involves the enzymes thiopurine S-methyltransferase (TPMT) and nudix hydrolase 15 (NUDT15).^[2] Genetic polymorphisms (GP) encoding these enzymes are responsible for a variable toxicity profile of 6-MP in individuals, and their prevalence varies among different ethnic groups.

Although the prevalence of TPMT mutation is relatively rare in the Asian sub-groups, they were noticed to experience severe 6-MP-related toxicity. Later, a genome-wide association study conducted revealed a higher prevalence of NUDT15 GP, which explained the high toxicity of 6-MP in the population of Asian descent.^[3]

The toxicity profile of 6-MP includes myelosuppression, hepatotoxicity, gastrointestinal disturbances, and alopecia. GP in 6-MP metabolizing enzymes further exaggerates this toxicity, particularly myelotoxicity, leading to recurrent episodes of febrile neutropenia and treatment interruptions, and has a direct impact on the outcome of the patient.^[1]

Clinical pharmacogenetic implementation consortium (CPIC) guidelines describe the recommended dosage adjustments for those with GP.^[4] Hence, it may be useful to test all patients pre-emptively before introducing this drug in ALL induction, to prevent multiple treatment interruptions and also improve the induction outcome.

MATERIAL AND METHODS

This study is a retrospective analysis, conducted in the Department of Hematology at a tertiary referral center in central Uttar Pradesh, where data between September 2021 and March 2023 were collected and analyzed. Since patient affordability was a matter of concern, all patients could not be tested pre-emptively, and certain pre-defined criteria were used, as per institution protocol, to choose the patients in whom the test was absolutely mandated.

The primary aim of this study was to determine the prevalence of TPMT and NUDT15 GP among those patients presenting with 6-MP-related toxicity in the induction of the ALL treatment protocol. The secondary aim was to compare the drug’s toxicity profile among those with and without the enzyme gene mutation.

Patients with ALL with an age <55 years were enrolled in our study. Patients <35 years of age were started on the Berlin-Frankfurt-Munster 1995 protocol (BFM) 95 protocol, and those ≥35 years were started on the Medical Research Council-United Kingdom (MRC) UK protocol. Each of these protocols had 4 phases: Induction, extra-compartmental therapy, re-induction, and consolidation. Induction was further divided into induction-A and induction-B. 6-MP was introduced and formed the backbone of induction-B. It was given at a 60 mg/m² single daily dose for 28 days during this phase. All patients underwent a bone marrow examination at the end of induction A and were deemed to have a morphological remission and a negative minimal residual disease before entering induction B.

All patients entering the induction-B and meeting any one or more of the following criteria after starting 6-MP are tested for TPMT and NUDT 15 gene polymorphism after taking an informed and written consent.

1. Persistent absolute neutrophil count (ANC) <500/mm³ and/or platelet count of <25,000 for ≥7 days, even with granulocyte colony-stimulating factor support encountered ≥2 times during induction of the ALL protocol

   OR

2. Two or more episodes of febrile neutropenia with a temperature >100.4 F and ANC <500/mm³ or a falling trend of ANC to reach <500/mm³

   OR

3. Serum glutamic-pyruvic transaminase (SGPT) >5 times and/or total serum bilirubin (TSB) >3 times the upper limit of normal range encountered even once

   OR

4. A single or a cumulative treatment interruption of ≥14 days during or at the end of induction due to any of the above- mentioned reasons leading to consumption of <50% of the cumulative 6-MP dose.

The patients with the following criteria were excluded from the study:

1. Those ≥55 years were excluded from the study in view of associated co-morbidities and poor tolerability to chemotherapy. As per protocol, they were all counseled to undergo the testing pre-emptively

2. Those with underlying baseline liver disease

3. Those with underlying invasive fungal pneumonia needing treatment with azoles in view of their associated hepatotoxicity.

2 mL of peripheral blood was sent in ethylenediaminetetraacetic acid vacutainers for analysis. Genomic DNA was isolated using the QIAamp DNA Blood Mini Kit. Next-generation sequencing-based targeted sequencing was performed so that all exon-intron boundaries and untranslated regions of TPMT and NUDT15 are captured. The libraries were sequenced on an Illumina sequencing platform with 2 × 150 bp paired end-reads to obtain a mean depth of >80–100×. The Genome Analysis Toolkit best practices framework was followed for the identification of germline variants using Sentieon.^[5]

The star (*) alleles of TPMT and NUDT15 GP, as mentioned in CPIC guidelines, were analyzed, and individuals were classified into poor metabolizer (PM), intermediate metabolizer (IM), and normal metabolizer (NM) based on the diplotype.Indeterminate phenotypes were considered as NMs. On the basis of the results of GP, the dose of 6-MP was adjusted as per the CPIC guidelines.^[4] The doses of 6-MP in IM and PMs were re-adjusted to 20 mg/m²/day and 10 mg/m²/day, respectively. They were later titrated upward while closely monitoring the counts to identify the maximum tolerated dose.

The data were compiled in MS Excel 2019 in the form of a spreadsheet. The nominal/categorical variables were summarized as frequencies and percentages. They were analyzed using the Chi-square test. A P ≤ 0.05 was taken as statistically significant. All statistical analyses were done using Epi Info version 7.2.1.0 CDC, Atlanta, GA, USA, 2018 statistical software.

RESULTS

During the study period, a total of 96 newly diagnosed cases of ALL were treated in the Department of Hematology, SGPGI, Lucknow. Out of the above 96 cases, 8 were >55 years of age, 5 had probable/proven invasive fungal infection (IFI), and 4 had baseline liver involvement (clinical liver disease/deranged SGPT/TSB/radiologically proven liver disease) and were hence excluded from the study. Out of the remaining 79 patients, 41 satisfied the inclusion criteria. 4 of the 41 cases refused to give consent for the test; therefore, the testing for GP was carried out on the remaining 37 subjects. The patient selection outline has been depicted in Figure 1.

Close

Figure 1:

Flow chart depicting patient selection. 6MP: 6-Mercaptopurine, ALL: Acute lymphoblastic leukemia, IFI: Invasive fungal infection

Export to PPT

The baseline characteristics of the study population have been summarized in Table 1.

Out of the 37 patients tested, 13 (35.13%) showed the presence of a genetic polymorphism. 12 out of the 13 patients had NUDT15 enzyme mutation and only 1 patient had TPMT mutation. Out of the 12 patients with NUDT15 GP, 9 patients were IMs (heterozygous), and 3 were PMs (homozygous) of 6MP. The patient with the TPMT mutation was an intermediate drug metabolizer (heterozygous).

Figure 2 depicts the prevalence of enzyme gene polymorphism in our study population.

Close

Figure 2:

Prevalence of enzyme gene polymorphism in our study. GP: Genetic polymorphisms.

Export to PPT

In our study, 15 out of 37 (40.54%) patients presented with grade 4 thrombocytopenia (platelet count <25,000), 15 (40.54%) presented with/febrile neutropenia, 29 (78.37%) patients had grade 4 neutropenia (ANC <500), and 8 (21.62%) patients had deranged liver function test.

12 out of 15 (80%) patients with grade 4 thrombocytopenia had a positive genetic polymorphism, and this result was statistically significant with a P < 0.001.

Furthermore, 11 out of 15 (73.3%) patients presenting with febrile neutropenia and satisfying our inclusion criteria had a significantly higher prevalence (P < 0.001) of enzyme gene polymorphism.

In our study, 29 patients who satisfied the inclusion criteria presented with grade 4 neutropenia (ANC of <500). Out of them, 13 patients (44.8%) had enzyme gene polymorphism. Although this result is not statistically significant, it highlights an important finding that all 13 patients with gene polymorphism presented with grade 4 neutropenia. This finding emphasizes the importance of the incorporation of pharmacogenomics into our clinical practice.

Five patients among our study population had transaminitis, and 3 had deranged TSB. None of them had any enzyme gene polymorphism. The results of our study have been summarized in Table 2.

Figure 2 depicts the prevalence of GP in relation to various 6-MP-induced toxicities.

DISCUSSION

Pharmacogenomics is rapidly gaining importance and its implementation has become a part of many leukemia protocols in the West. Although still not routinely incorporated into Indian protocols, testing for gene polymorphism affecting 6-MP metabolism has been studied by many authors in our country. Despite the emergence of data on varied other gene polymorphisms, only TPMT enzyme mutation remains the most widely investigated to date.

Khera et al.^[2] conducted a prospective study on the prevalence of GP and their relation to 6-MP toxicity in Indian population. This study showed a prevalence rate of 28.5% compared to a higher value of 35.13% in our study population. This can be explained by the fact that only those patients presenting with grade 4 toxicity and febrile neutropenia were tested in our study, and hence, an obviously higher prevalence was noted.

Studies conducted by Khera et al.^[2] and Kodidela et al.^[6] tested patients in their maintenance phase of ALL therapy. This present study, to the best of our knowledge, is the first of its kind, in that the subjects chosen were the patients undergoing induction of the ALL protocol. An early detection was sought in an attempt to prevent drug toxicity and also to minimize interruption of chemotherapeutic medication, especially in a crucial phase like induction.

Moradveisi et al.^[7] pre-emptively tested all patients in their study population. Our institute is a state government-funded referral center that caters to low-income population groups for whom the affordability of the test is a matter of concern. Hence, in our study, only the patients presenting with 6-MP-associated toxicities were tested. The major limiting factor of doing this was the confounding effect of the intensive chemotherapeutic drugs other than 6-MP. We tried to overcome this by setting stringent cutoffs as inclusion criteria for patient testing.

In our study, out of the 37 patients tested, 12 had NUDT15 and only 1 patient had TPMT GP. This finding is consistent with the landmark paper by Yang et al.,^[3] which demonstrated the significantly higher incidence of NUDT15 GP in the population of Southeast Asian descent.

Many Asian studies^[2,8] also showed a high prevalence of inositol triphosphatase (ITPA) in relation to 6-MP-related toxicities, although patients with NUDT15 GP showed a significantly greater toxicity profile. Since a proven effect of ITPA mutation is yet to be well established, we have not tested our study population for the above GP.

Our study showed a significantly higher prevalence of GP in patients meeting the criteria for thrombocytopenia and febrile neutropenia (P < 0.001). The GP was also higher in those with grade 4 neutropenia, although significance could not be established (P = 0.053). None of the patients with deranged SGPT or TSB had any enzyme mutation. In studies conducted by Khera et al.,^[2] Shah et al.,^[9] and Evans et al.,^[8] patients with TPMT and/or NUDT15 enzyme mutations had higher incidence of hematological toxicity compared to hepatic derangement, similar to that of our study.

The two major drawbacks of our study were its small sample size and the fact that pre-emptive testing could not be performed in all patients. Furthermore, due to its retrospective nature, the results could be put forth as mere associations, and a causal relation could not be drawn. As a result of our stringent inclusion criteria, almost all patients had a cumulative 6MP dose of <50% and days of interruption of ≥14 days, and hence, these parameters could be put to use for statistical analysis.

CONCLUSION

To conclude, all previously conducted studies have provided pharmacogenomic data on 6-MP-related toxicity only in the maintenance phase. Our study, though small, is the first of its kind to throw light on the fact that testing may be considered even in the induction phase. This early detection may enable early dose adjustment, thereby improving clinical outcomes by reducing therapy interruption and the need for hospitalization. This may also prove to be cost-effective in the long-run. Our study also explores the possibility of using certain pre-defined inclusion criteria like recurrent grade 4 thrombocytopenia and febrile neutropenia, as soft indicators for GP, especially in low-income countries like India, where pre-emptive testing of all patients is still not feasible. However, multicenter prospective studies with larger sample sizes are needed to establish and validate uniform inclusion criteria and to recommend early induction testing for gene polymorphism.