A randomized trial of a mercaptopurine (6MP) adherence-enhancing intervention in children with acute lymphoblastic leukemia (ALL): A COG ACCL1033 study

10007 Background: We previously reported that > 40% of children with ALL are non-adherent to 6MP, and > 52% of ALL relapses are attributable to 6MP non-adherence. The most common barrier is forgetting to take 6MP; the most common facilitator is parental vigilance. These observations informed a randomized trial to enhance 6MP adherence (COG-ACCL1033, #NCT01503632; 89 COG sites). Results are described here. Methods: The Intervention Package (IP) consisted of: i) Education; ii) 6MP schedules; iii) daily personalized text message reminders from physician to patient and caregiver, to prompt iv) directly supervised therapy (DST), with text back response by patient/caregiver. Baseline adherence was measured for 4 wks, followed by intervention for 16 wks to examine the impact of IP vs. Edu (education) on 6MP adherence (measured electronically by MEMs Cap) in all patients, ≥12yo, < 12yo. Longitudinal binomial logistic regression using generalized estimating equations was used. Missing data were handled by multiple imputation. Results: 444 patients were randomly assigned to IP (n = 230) or Edu (n = 214). Baseline characteristics (age at study: 8.6y vs 7.5y; males: 67% vs 69%; non-Hispanic whites: 40% vs 42%) and adherence rates (92% vs 94%) were comparable (except paternal education: 49% vs 38%, p = 0.04). No study arm*time interaction was found; thus, the 16-week overall mean fitted adherence rates were compared between IP and Edu, adjusting for baseline adherence, time on study, parental education. All patients: Adherence rates were marginally higher on IP (94% vs 92.5%, p = 0.09). On IP, for times with a record of text response, adherence rates were higher (94%) when compared with times with no response (89%), p = 0.002. < 12yo: Adherence rates were comparable (IP: 94.4% vs Edu: 93.7%, p = 0.5). ≥12yo: Adherence rates were significantly higher on IP (93.1% vs 90.0%, p = 0.037). ≥12yo with baseline adherence < 90%: IP had the highest impact for this subgroup (83.4% vs 74.6%, p = 0.008). Conclusions: A 16-week comprehensive intervention resulted in higher 6MP adherence rates in children with ALL who were 12y or older at study. IP was most impactful in adolescents with baseline non-adherence. Clinical trial information: #NCT01503632.