Multiple studies have demonstrated the importance of oral chemotherapy dosing to outcome in the treatment of pediatric acute lymphoblastic leukemia. Dosing of oral mercaptopurine is adjusted by clinical tolerance using dose escalations or reductions depending on the patient’s neutrophil and platelet counts. Current regimens use guidelines to attempt to maximize anti-leukemic effect but minimize risk of infection or bleeding from extreme cytopenias. A recent approach to help predict patient tolerance involves testing of drug metabolites or TPMT activity phenotype or genotype, but these tests remain a send-out test for most centers and have not been universally adopted as standard of care. As a potential additional clinical tool, we sought to determine whether tolerance of mercaptopurine during an early consolidation course could help predict tolerance of mercaptopurine during maintenance.

We performed a retrospective chart review of all standard risk ALL patients treated at our center in the last 10 years. Patients qualified for the study if they received standard consolidation using treatment regimens based on Children’s Oncology Group (COG) studies AALL0331 or AALL0932 which include mercaptopurine for 28 days without other myelosuppressive chemotherapy (only other chemotherapy includes a single vincristine dose on day 1 and intrathecal treatments on days 1, 8, and 15). Patients receiving consolidation including other myelosuppressive chemotherapy (such as arm LRAsp-IV on AALL0331 which included PEG-asparaginase in consolidation or arm IS-IV on AALL0331 with intensified consolidation for slow early responder patients) were excluded from the study. We hypothesized that patients who had delay in the start of their interim maintenance course following standard consolidation due to cytopenias demonstrate a clinical sensitivity to mercaptopurine that could result in increased risk of interruptions in oral chemotherapy early in maintenance therapy. Thus we correlated delays in start of interim maintenance with interruptions in dosing of mercaptopurine due to cytopenias during the first 2 cycles of maintenance. Data was analyzed using JMP® Statistical Discovery software.

A total of 55 patients met our eligibility requirements (standard risk ALL, treated with standard consolidation per the regimens noted above, and completed at least 2 cycles of maintenance chemotherapy). Of these 55 patients, 11% (N=6) had a delay in the start of their interim maintenance course due to cytopenias following standard consolidation. The average duration of this delay was 12.5 days with a range of 7-21 days. Among the 6 patients who had delayed interim maintenance, 5 (83%) had mercaptopurine held during the first 2 cycles because of cytopenias. This was higher than the 61% (30 out of 49) of patients requiring dose interruptions for cytopenias in maintenance among those patients who started interim maintenance without delay. (P=0.28) The interval from start of maintenance until time of first dose interruption and number of interruptions in the first two cycles of maintenance therapy were also examined. Patients who had delays in start of interim maintenance had a mean time to first dose interruption of 51 days which was not different than the mean time of 55 days to first interruption in all other patients. (P=0.76) Both groups had a median of 43 days to time of first interruption. The number of dose interruptions ranged from 0-3. There was not a difference in the number of dose interruptions among patients who did and did not have a delay in interim maintenance. (P=0.28)

Our analysis demonstrated that 65% of all patients in our cohort required dose interruptions of mercaptopurine in the first 2 cycles of maintenance therapy, suggesting this is a very common complication. A higher proportion of patients with delays in starting interim maintenance experienced such interruptions compared to those without delays. Although this was not statistically significant it should be evaluated in larger cooperative group studies. Our data does suggest that the time period between day 29 and day 57 of maintenance therapy is a high risk period for cytopenias. This data supports close monitoring of counts early in maintenance (every 2 weeks rather than every 4 weeks with vincristine visits) to minimize risks of infection or bleeding associated with prolonged cytopenias.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.