Results from patient-derived xenograft models support co-administration of allopurinol and 6-mercaptopurine to reduce hepatotoxicity and improve event-free survival in pediatric acute lymphoblastic leukemia

The prognosis of pediatric acute lymphoblastic leukemia (ALL) has significantly improved over the past decades. Maintenance therapy, with 6-mercaptopurine (6-MP) and methotrexate (MTX) as its cornerstone drugs, is crucial for achieving long-term remission. However, the clinical use of 6-MP is often compromised by hepatotoxicity-related treatment interruptions or dose reductions, which potentially increase the risk of relapse. These challenges are closely linked to the complex and variable metabolism of purines among individuals.¹ To address this challenge, allopurinol, a xanthine oxidase inhibitor, has been proposed as an adjunct therapy to optimize 6-MP metabolism and mitigate hepatotoxicity. Previous studies have shown that allopurinol effectively reduces liver toxicity and modulates thiopurine metabolism,^2-5 but its impact on long-term outcomes remains insufficiently explored. Understanding whether allopurinol can improve tolerability while maintaining the therapeutic efficacy of 6-MP is critical for refining ALL maintenance therapy strategies. To address this gap, we utilized patient-derived xenograft (PDX) models to evaluate the impact of adding allopurinol on survival in mice, and conducted a large retrospective study to assess whether addition of allopurinol in maintenance therapy influences the prognosis of children with ALL and to evaluate the safety of allopurinol in this setting.

Using preclinical PDX models established from four different ALL patient samples, we compared the survival outcomes between mice receiving combination therapy (halved 6-MP with allopurinol) and those receiving full-dose 6-MP monotherapy. Our results demonstrated the life span in the combination group were comparable to those in the monotherapy group (Figure 1A). Importantly, we observed a significant shift in thiopurine metabolism, with higher DNA-incorporated thioguanine (DNA-TG) levels (associated with antileukemic effects)⁶ and reduced 6-methylmercaptopurine (6-MMP) levels (associated with hepatotoxicity)⁷ in the combination group (Figure 1B). In the AL5611 PDX, the spleen size and weight in the combination group were smaller than in the monotherapy group, and the leukemia burden in the spleen and bone marrow was also lower (Figure 1C). These results highlight the potential of allopurinol to modulate 6-MP metabolism effectively, supporting its clinical evaluation in ALL patients to optimize therapy without compromising survival outcomes.

We conducted a retrospective cohort study at the Shanghai Children’s Medical Center (SCMC) affiliated to Shanghai Jiao Tong University School of Medicine between December 2014 and June 2023. All patients were in the cohort of Chinese Children’s Cancer Group (CCCG)-ALL-2015 trial (clinical trial number: ChiCTR-IPR-14005706).⁸ Inclusion criteria encompassed pediatric ALL patients who underwent continuation therapy as defined in the protocol. High-risk patients and those who did not proceed to continuation therapy were excluded. The study was approved by the Institutional Review Board of SCMC, (no. SCMCIRB-K2024169-1).

A total of 752 pediatric ALL patients were included in the analysis, among whom 459 experienced hepatotoxicity, defined as alanine transferase (ALT) >2× upper limit of normal (ULN) or elevated bilirubin levels. Allopurinol co-treatment was initiated in 149 patients according to their liver function and hematology parameter: i) repeated ALT levels exceeding ten-times ULN after resuming 6-MP therapy following temporary discontinuation, or ii) persistent ALT elevation between two-times and ten-times ULN for more than 8 weeks despite 6-MP dose adjustments, accompanied by white blood cell (WBC) counts remaining above the target range. The median time to allopurinol initiation from the initial detection of liver function abnormalities was 15.6 weeks (approximately 4 months). The demographics and clinical characteristics are summarized in Online Supplementary Table S1.

Patients in the allopurinol group began treatment with a median allopurinol dose of 35.7 (interquartile range [IQR]; 32.8-39.9) mg/m²/day and a median duration of use of 47 (IQR, 28-63) weeks. To minimize the risk of excessive myelosuppression, the 6-MP dose was reduced upon initiation of allopurinol, from a median dose of 49.3 (IQR, 40.3-58.2) mg/m²/day to 25.0 (IQR, 18.7-32.5) mg/m²/day, approximately half the dose before allopurinol. MTX dosing was also reduced concurrently by one-third. Subsequent dose adjustments of 6-MP were made based on regular blood count monitoring.

The impact of allopurinol co-treatment on event-free survival (EFS) was evaluated using time-dependent multivariate Cox regression analyses, treating allopurinol initiation as a time-dependent variable. The analysis revealed a significant improvement of EFS among allopurinol co-treated patients compared to those didn’t receive allopurinol (adjusted hazard ratio [aHR] =0.29, 95% confidence interval [CI]: 0.10-0.80; P=0.017; Table 1; Figure 2B). To address potential confounding factors, propensity score matching (PSM) was also conducted to minimize selection bias and ensure comparability between the allopurinol and non-allopurinol groups. After PSM, allopurinol co-treatment remained significantly associated with improved EFS (aHR=0.25, 95% CI: 0.09-0.69; P=0.008; Online Supplementary Table S2; Online Supplementary Figure S1). These findings, consistent with multivariable Cox regression analyses, confirm that allopurinol co-treatment is an independent prognostic factor for EFS.

Figure 1.Preclinical evaluation of the effects of allopurinol on 6-mercaptopurine efficacy and metabolism in patient-derived xenograft models of acute lymphoblastic leukemia. (A) Survival analysis in 4 allopurinol (ALLO) patient-derived xenograft (PDX) models comparing 6-mercaptopurine (6-MP) and ALLO combination therapy with 6-MP monotherapy. (B) 6-MP metabolite and DNA-incorporated thioguanine (DNA-TG) levels in PDX following ALLO co-treatment. The levels of 6-TU and 6-MMP in red blood cells (RBC) were normalized by hemoglobin (Hb) level, and the DNA-TG levels in bone marrow (BM) and spleen were normalized by DNA content. (C) Proportion of leukemia cells (huCD45⁺/huCD7⁺) in peripheral blood before and after treatment in PDX-5611. (D, E) Images and weights of the spleens after treatment in PDX-5611. (F, G) Proportion of leukemia cells in the spleen and bone marrow after treatment. *P<0.05; **P<0.01; ***P<0.001; Student’s t test.

To explore the metabolic effects of allopurinol co-treatment, DNA-TG levels in bone marrow were analyzed using available stored leftover frozen bone marrow samples. Allopurinol co-treatment led to significant changes in thiopurine metabolism. DNA-TG levels in bone marrow increased by 3.4-fold (from 131 to 444 fmol/µg DNA; P<0.001; Figure 2E). Liver function also improved markedly after allopurinol initiation. Prior to treatment, alanine transaminase (ALT) levels in the allopurinol group averaged ~500 U/L. Following allopurinol co-treatment, ALT levels decreased significantly in the first month and stabilized at a lower level after the second month and remained consistent thereafter, indicating sustained liver function improvement (Figure 2F, G). In our study, allopurinol co-treatment significantly improved WBC control, with the proportion of patients maintaining WBC within the target range increasing from 34.2% to 68.5% (P<0.001; Figure 2C). Although leukopenia (WBC <2×10⁹/L) increased from 37.6% to 53.7% (P= 0.005), this did not result in a higher incidence of infection. This observation underscores the safety of allopurinol co-treatment when combined with close monitoring and dose adjustments. Our study is the first to evaluate the impact of allopurinol co-treatment on EFS in pediatric ALL patients, integrating clinical and preclinical data to highlight its dual benefits in mitigating hepatotoxicity and improving prognosis. While the addition of allopurinol has been studied to improve patient tolerance to 6-MP in both inflammatory bowel disease^9,10 and pediatric ALL^2-5 by balancing thiopurine metabolism, previous studies primarily focused on biochemical markers, such as ALT and 6-MP metabolites. Most of these findings have been based on case reports or small retrospective studies, leaving its impact on long-term outcomes, including EFS, largely unexplored. Although Källström’s phase II study demonstrated a significant increase in erythrocyte 6-TGN levels after allopurinol addition,¹¹ direct evidence linking allopurinol co-treatment to prognosis in ALL patients has been lacking. To address this gap, our study employed a large, retrospective cohort and PDX models to provide evidence supporting the use of low-dose allopurinol in pediatric ALL patients. Our findings confirm that, as previously reported, allopurinol co-treatment effectively improves liver function in patients with hepatotoxicity, enhancing tolerance to 6-MP. More importantly, we found that patients in the allopurinol group had the best prognosis, outperforming both hepatotoxicity without allopurinol group and no hepatotoxicity group.

Table 1.Multivariate Cox regression analysis for event-free survival.

Figure 2.Clinical evaluation of allopurinol co-treatment during maintenance therapy in pediatric acute lymphoblastic leukemia patients, including its impact on event-free survival and laboratory parameters. (A) Patient enrollment and group classification. (B) Kaplan-Meier curves for event-free survival (EFS) in the hepatotoxicity (HTX) with allopurinol (ALLO) group, no HTX group, and HTX without ALLO group in the entire population, low-risk patients, and intermediate-risk patients. Black dots represent the time points when allopurinol treatment started. Adjusted hazard ratio (aHR) with 95% confidence intervals (CI) and P values were estimated using the time-dependent Cox regression analysis, adjusted for age, sex, final risk, immunophenotype, day 19 minimal residual disease (MRD) level, and liver function abnormality. Changes in laboratory parameters following ALLO co-treatment, including white blood cell (WBC) (C), absolute neutrophil count (ANC) in peripheral blood (D), DNA-incorporated thioguanine (DNA-TG) in bone marrow (BM) (E), alanine aminotransferase (ALT) (F), and mean change in ALT from baseline (adjusted MMRM analysis) (G). ***P<0.001; Wilcoxon signed-rank test.

Previous studies have commonly used allopurinol doses of 50 mg/m²,^3,4,11 finding this dose effective for modifying thiopurine metabolism and well tolerated in ALL patients, with some studies even exploring higher or fixed doses.^12-14 However, in our results from PDX, we found that a small dose of 3 mg/kg allopurinol was sufficient to achieve desired outcomes in mice. Additionally, given the increased sensitivity to allopurinol observed in the Chinese population,¹⁵ we are inclined to use lower doses in clinical practice. In our study, most patients received a 30-40 mg/m² dose of allopurinol and achieved favorable outcomes.

Based on the positive results, we recommend that for patients experiencing hepatotoxicity during maintenance therapy, a low dose of allopurinol (25-50 mg/m²) may be considered. Simultaneously, the 6-MP dose should be halved from the pre-allopurinol dose. During the initial adjustment phase, a complete blood count should be monitored twice weekly, and the doses of 6-MP and MTX should be adjusted according to the blood counts, following the same principles. Generally, there is no need to adjust the allopurinol dosage.

Despite its retrospective nature, our study has several strengths, including a large sample size, and the incorporation of PDX models. Limitations include potential selection biases, clinician discretion in initiating allopurinol, and variability in 6-MP and MTX dose adjustments. Furthermore, the observed increase in EFS with the allopurinol and 6-MP combination may be protocol-specific, which should be considered when applying the findings to other protocols In conclusion, allopurinol co-treatment effectively reduces 6-MP-induced hepatotoxicity, improves WBC control, and enhances EFS in pediatric ALL patients. Our findings provide important evidence supporting the co-administration of allopurinol during maintenance therapy for pediatric ALL patients with hepatotoxicity. A larger randomized trial is needed to evaluate allopurinol’s impact on outcomes and toxicity in pediatric ALL.

Footnotes

• Received October 29, 2024
• Accepted January 7, 2025

Correspondence

*YM, QZ, MW and JC contributed equally as first authors

S. SHEN - shenshuhong@scmc.com.cn

X. CHEN - cxxsunny9@aliyun.com

Disclosures

No conflicts of interest to disclose.

Funding

This work was supported by National Natural Science Foundation of China (no. 82170155), the Three-year Action Plan for Promoting Clinical Skills and Innovation Ability of Municipal Hospitals (SHDC2022CRT008), Key Project of National Clinical Research Center for Child Health and Disorders (no. NCRCCHD-2021-KP-04), Innovation Program of Shanghai Municipal Education Commission (2021-01-07-00-02-E00091), Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province (2022ZQNZD011), Shanghai Health Commission Clinical Research Project (20214Y0156).