Effect of a Daily Text Messaging and Directly Supervised Therapy Intervention on Oral Mercaptopurine Adherence in Children With Acute Lymphoblastic Leukemia: A Randomized Clinical Trial

Abstract

Importance: Suboptimal adherence to oral mercaptopurine treatment in children with acute lymphoblastic leukemia (ALL) increases the risk of relapse. A frequently expressed barrier to adherence is forgetfulness, which is often overcome by parental vigilance.

Objective: To determine whether a multicomponent intervention, compared with education alone, will result in a higher proportion of patients with ALL who have mercaptopurine adherence rates 95% or higher, for all study participants and among patients younger than 12 years and vs those aged 12 years and older.

Design, setting, and participants: The adherence intervention trial was an investigator-initiated, multi-institutional, parallel-group, unblinded, randomized clinical trial conducted between July 16, 2012, and August 8, 2018, at 59 Children's Oncology Group institutions in the US, enrolling patients with ALL diagnosed through age 21 years and receiving mercaptopurine for maintenance. The date of final follow-up was January 2, 2019. Data analysis was performed from February to October 2019.

Interventions: Patients were randomized 1:1 to education alone or the intervention package, which consisted of education and personalized text message reminders daily to prompt directly supervised therapy. Four weeks of baseline adherence monitoring were followed with a 16-week intervention.

Main outcomes and measures: The primary end point was the proportion of patients with adherence rates 95% or higher over the duration of the intervention for all study participants, and for those younger than 12 years vs those aged 12 years and older.

Results: There were 444 evaluable patients (median age, 8.1 years; interquartile range, 5.3-14.3 years), including 230 in the intervention group and 214 in the education group. Three hundred two patients (68.0%) were boys, 180 (40.5%) were non-Hispanic White, 170 (38.3%) were Hispanic, 43 (9.7%) were African American, and 51 (11.5%) were Asian or of mixed race/ethnicity. The proportion of patients with adherence rates 95% or higher did not differ between the intervention vs education groups (65% vs 59%; odds ratio, 1.33; 95% CI, 1.0-2.0; P = .08). Exploratory analyses showed that among patients aged 12 years and older, those in the intervention group had higher mean (SE) adherence rates than those in the education group (93.1% [1.1%] vs 90.0% [1.3%]; difference, 3.1%; 95% CI, 0.1%-6.0%; P = .04). In particular, among patients aged 12 years and older with baseline adherence less than 90%, those in the intervention group had higher mean (SE) adherence rates than those in the education group (83.4% [2.5%] vs 74.6% [3.4%]; difference, 8.8%; 95% CI, 2.2%-15.4%; P = .008). No safety concerns were identified.

Conclusions and relevance: Although this multicomponent intervention did not result in an increase in the proportion of patients with ALL who had mercaptopurine adherence rates 95% or higher, it did identify a high-risk subpopulation to target for future adherence intervention strategies: adolescents with low baseline adherence.

Trial registration: ClinicalTrials.gov Identifier: NCT01503632.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Mascarenhas reported receiving grants and clinical trial support to his institution from the National Cancer Institute through the Children’s Oncology Group during the conduct of the study; personal fees from Bayer; clinical trials support to his institution from Bayer, Eli Lilly, Loxo-Oncology, AstraZeneca, Jazz, Salarius, Thermo Fisher Scientific, ER SQUIBB, Merck, Novartis, Amgen, and Incyte outside the submitted work; and travel and accommodation related to consulting from Bayer, AstraZeneca, Salarius, and Thermo Fisher Scientific with no personal remuneration outside the submitted work; and reporting serving on the board of directors for the Children’s Oncology Group Foundation (present), American Society of Pediatric Hematology and Oncology (past), and the Pablove Foundation (past). Dr Lew reported receiving travel reimbursement for study committee activity from BMS outside the submitted work. Dr Raetz reported receiving grants from Pfizer and serving on the Data Safety Monitory Board for Celgene outside the submitted work. Dr Relling reported receiving grants from Servoer and the National Institutes of Health outside the submitted work. Dr Landier reported receiving grants from National Cancer Institute during the conduct of the study. No other disclosures were reported.

Figures

Figure 1.. Enrollment and Retention of Patients by Study Group

Day 29 and day 57 represent days during the trial when the patient returned for a scheduled clinic visit. aAmong the 163 patients who did not meet the inclusion criteria, 111 were using a pillbox, 20 did not want to receive text message reminders, 12 did not want to use the Medication Event Monitoring System (MEMS) device, 5 did not have a designated caregiver, and 15 had other reasons. Among the 295 patients who declined participation, 114 were not interested, 62 had another reminder system, 24 said they already remembered, 23 said they were too busy or the timing was bad, 3 had physician refusal, and the other 69 had other or no reasons. Lack of access to a text-capable device was not a confounding feature, because for patients who did not have access to a cellular telephone with texting capabilities, we issued a study telephone with unlimited data for the entire study period.

Figure 2.. Mean Fitted Adherence Rate by Treatment Group, Intervention Package (IP) and Education Alone

Graphs show adherence rates for entire cohort (A), for children younger than 12 years at study participation (B), for children aged 12 years and older at study participation (C), for children aged 12 years and older with baseline adherence less than 90% (D), and for children aged 12 years and older with baseline adherence greater than or equal to 90% (E). The IP included education and daily personalized text message reminders prompting directly supervised therapy.

Figure 3.. Proportion of Patients With Adherence Rates 95% or Higher Before and After the Intervention Period

The intervention package (IP) included education and daily personalized text message reminders prompting directly supervised therapy. Proportions of patients were adjusted for baseline adherence, paternal education, and time in the adherence study.