Discussion
Our study demonstrated that pediatric HCT survivors reported higher impairments in specific CCSS-NCQ neurocognitive domains of emotional regulation, task efficiency, memory, and organization but average cognitive quality of life compared with general population norms at a median of 27 years post-transplant. Despite reported problems with specific cognitive abilities, reassuringly these problems were not perceived to have significant consequences on daily life as reflected on the Neuro-QoL. This could represent adaption to cognitive difficulties over time or an adjustment in expectations. These results reflect the long-term follow-up period of our patient population, in contrast with a shorter-term study in pediatric HCT survivors that described largely stable intelligence and academic achievement through 5 years post-HCT.12 Given the limited studies of late neurocognitive effects among pediatric HCT survivors, some comparisons can be drawn from CCSS studies of childhood cancer survivors at a similar follow-up interval.4,9,25 In a report from the CCSS featuring over 6,000 survivors (mean age 32 years), a lower percentage (12.5-22.4%) reported impairments in the CCSS-NCQ domains.9 Although not directly comparable, this suggests that pediatric HCT survivors may be more likely to report neurocognitive impairments compared with general pediatric cancer survivors.
Various treatment characteristics and exposures have been explored in terms of their association with neurocognitive dysfunction. Previous literature has suggested that younger age at time of cancer diagnosis or HCT is a risk factor for worse neurocognitive function.4,14,15 While a cohort study of pediatric HCT survivors initially demonstrated overall stability in cognitive functioning at 5 years post-HCT,12 a follow-up study with an expanded cohort showed that patients <3 years at time of transplant may be particularly susceptible to the cognitive impact of TBI, with lower IQ (16 points on average) at 5 years post-HCT compared with those who did not receive TBI.26 In our study, younger age <10 years at time of HCT was suggestive of worse outcomes after multivariable analysis, although the study may have been underpowered to detect a statistically significant association. Additionally, the sample size was too small to draw any conclusions in the cohort <3 years at time of HCT. As above, radiation exposures such as TBI and cranial irradiation have been associated with neurocognitive dysfunction.3,11 While we did not find a difference between non-irradiated survivors versus those who underwent TBI only, those who received pre-HCT cranial irradiation were more likely to report worse neurocognitive function. In our cohort, survivors treated with cranial irradiation reported impairments in the CCSS-NCQ domains comparable to childhood CNS tumor survivors,8,27,28 as well as pediatric ALL survivors treated with cranial irradiation, with significant late cognitive deficits detected at a median follow-up of 28.5 years after treatment.29
Subsets of pediatric HCT survivors were more likely to have impaired neurocognitive function, including those with hearing conditions (moderate to severe hearing loss or more bothersome tinnitus), history of stroke or seizures, or sleep disturbances. The association of poor long-term neurocognitive outcomes in survivors with a higher burden of chronic health conditions is unsurprising, given previous studies.25 Specifically for long-term pediatric cancer survivors, hearing and visual deficits have been associated with impaired emotional regulation and organization,4 while history of stroke is associated with worse health-related quality of life and neurocognitive function, particularly task efficiency and memory.30 Lastly, we found that self-reported sleep disturbances were significantly associated with greater impairments in all CCSS-NCQ domains and worse cognitive quality of life. Given the cross-sectional nature of our study, we were unable to determine if sleep issues had a causal effect on neurocognitive function. However, pediatric cancer survivors who reported greater fatigue and poorer sleep quality have also exhibited greater impairments in all CCSS-NCQ domains.31 Studies of pediatric cancer survivors have reported a greater burden of sleep disturbances and fatigue compared with non-cancer controls.6,32 Additional research to better characterize the sleep disorders experienced by our patients may help determine specific targets for intervention.
For pediatric HCT survivors over the age of 25 years at the time of survey completion, greater impairments in all CCSS-NCQ domains and worse Neuro-QoL scores were associated with less educational achievement (defined as less than college education). Our findings are similar to other studies showing that childhood cancer survivors with neurocognitive deficits are less likely to attain educational milestones, obtain employment, or live independently.3,4 These results may be time-dependent, as differences in educational achievement may not be fully appreciated until an extended time interval after HCT. This highlights the importance of early referrals for resources or interventions to potentially mitigate HCT treatment effects. The Children’s Oncology Group’s Long-Term Follow-Up guidelines recommends routine formal neuropsychological evaluation for patients at risk for late neurocognitive deficits, including those who receive any cranial irradiation (including TBI) or specific chemotherapy agents commonly used during HCT, including intrathecal methotrexate, high-dose intravenous methotrexate and/or cytarabine.33 While previous pediatric HCT reviews3,15 have suggested the implementation of similar routine screening, further studies will be needed to study if earlier testing and intervention improves long-term outcomes.
Our study benefited from a very long average follow-up duration, greatly exceeding that of other studies of neurocognitive outcomes among pediatric HCT survivors. Additionally, our study also featured one of the largest samples examining neurocognitive outcomes in this patient population. However, there were several limitations to this study. Our study was conducted at a single center and limited by lower response rate; thus, our results may not be fully representative of the overall population of pediatric HCT survivors. Non-responders were more likely to be younger and male; however, when inverse probability weighting was applied to represent the entire cohort, we did not find a meaningful difference in our results. It is unclear how potential response bias would have influenced these results, as those with lower function or quality of life may have been both less likely to respond or possibly more likely to participate given the nature of the survey. Lastly, while our study was based on self-reported measures with no objective neurocognitive testing, many components of cognition can only be captured by patient self-report and we relied on well-validated surveys which enabled comparisons with the general population.