DISCUSSION
Among patients treated with a DFCI ALL Consortium regimen, 25% [95% CI, 14% – 37%] of caregivers report non-adherence to oral 6MP during therapy. This rate is strikingly similar to Medication Event Monitoring System (MEMS)-measured adherence rates from the COG.3,6 Given that self-reported non-adherence has a sensitivity as low as 53% but a positive predictive value nearing 95%,7 these data suggest that the frequency of non-adherence in this cohort is likely an under-estimate of the true prevalence, and even higher than the non-adherence rates observed in the COG.
Similar to other studies, non-adherence was associated with parent education level,3 and barriers to adherence included 6MP administration guidelines surrounding food and dairy intake. Only 49% of parents recognized “relapse” or “cancer returning” as risks associated with non-adherence. Studies have demonstrated that patients who understand the risks and benefits of following medication guidelines are more likely to adhere to prescribed regimens.8 We previously reported significant variability in approaches to patient education surrounding oral chemotherapy across DFCI ALL Consortium sites.9 Specifically, providers reported modifying chemotherapy teaching based on their perceptions of patient comprehension. Our finding that 25% of families in the DFCI cohort reported non-adherence to 6MP suggests that more frequent patient-provider contact, and individualized oral chemotherapy education do not mitigate barriers to adherence.
This study was limited by its small sample size, cross-sectional nature, and self-report adherence measure. Despite limitations, these exploratory data demonstrate that 1 in 4 families self-report oral 6MP non-adherence in the context of a chemotherapy regimen reliant on weekly interactions with the medical team. These findings support the generalizable nature of published non-adherence data across pediatric oncology consortia, reinforcing the urgent need for healthcare delivery intervention development to address non-adherence. Given that parental education is associated with non-adherence across consortia, future interventions targeting the efficacy of medical communication may be warranted. A prospective investigation of oral chemotherapy adherence using a MEMS-based approach is ongoing in the current DFCI ALL Consortium trial 16-001 (NCT03020030).
Acknowledgements:
We would like to thank Dr. Smita Bhatia for sharing her expertise in adherence research as well as her permission to adapt her adherence survey instrument for this study.
JMK is supported by a National Institutes of Health KL2 award from the Irving Institute for Clinical and Translational Research at Columbia University Irving Medical Center (NIH-KL2TR001874)
MB is supported by a T32 award from The National Cancer Institute Training Grant (T32CA094061).
KB is supported by a K07 award from the National Institutes of Health (K07CA211847).