Results
Of 626 eligible pediatric HCT survivors surveyed, 199 participated (50.3% female) at a median age of 37.8 years (range 18-61) at survey and median 27.6 years (range 1-46) from transplant (Table 1). Of note, 59 (29.6%) of participants underwent transplant at age <10 years and 21 (10.6%) were transplanted at age <3 years. The majority underwent transplantation for malignant conditions, including 106 patients with acute leukemia (53.3%) and 20 with chronic leukemia (10.1%). Among non-malignant conditions, 44 patients (22.1%) received transplants for hematologic conditions, most commonly aplastic anemia. Nine patients (4.5%) received transplants for immunodeficiencies, histiocytic disorders, or other indications. Most (175/199 patients) initially received allogeneic transplants (87.9%) and a subset of 20 patients (10.1%) received multiple transplants. Across all transplants, 132 patients (66.3%) underwent conditioning with total body irradiation (TBI), with 123 (61.8%) receiving ≥1000cGy. Eighteen (9.0%) also received additional cranial irradiation prior to TBI (median 1200cGy, range 480-5400); 1 additional patient received cranial irradiation alone without TBI. The prevalence of moderate/severe hearing issues and history of stroke/seizures was 40/198 (20.2%) and 21/197 (10.7%), respectively. Compared with non-participants, participants were more likely to be female, currently older, and transplanted at an earlier age, but did not differ across other demographic or clinical characteristics (Supplemental Table S1).
On the CCSS-NCQ, 32.5% of survivors reported problems with task efficiency, 25.4% with memory, 21.8% with emotional regulation, and 18.9% with organization compared with an expected 10% in the general population (Table 2; all p<0.01 ). Results were similar after application of inverse probability weights, with 33.0% reporting impairment with task efficiency, 27.6% with memory, 22.6% with emotional regulation, and 18.4% with organization (Supplemental Table S2). In unweighted analysis, female survivors reported more issues with emotional regulation and memory relative to males. Race/ethnicity, current age, underlying diagnosis, year of transplant, number of transplants, or history of cGVHD were not associated with a significant difference in CCSS-NCQ scores. Characteristics associated with impairments in most or all CCSS-NCQ domains included hearing issues, history of stroke or seizure, and self-reported sleep disturbances. Survivors >25 years of age with less educational achievement also reported significantly worse scores in all domains (allp<0.05 ). Survivors who received cranial irradiation were more likely to report impaired scores in task efficiency and organization compared with survivors who did not receive any radiation or those treated with TBI alone (p<0.05 ). In multivariable regression analysis adjusted for sex and current age, hearing issues were independently associated with impaired task efficiency (OR 2.71, 95% CI 1.19-6.17) and organization (OR 3.39, 95% CI 1.38-8.32), while history of stroke or seizure was independently associated with impaired emotional regulation (OR 4.78, 95% CI 1.81-12.6) and memory (OR 3.98, 95% CI 1.46-10.8; Figure 1 and Supplemental Table S3). Sleep disturbances were associated with impairments across all CCSS-NCQ domains. Younger age at transplant was generally associated with increased risk of impairments in CCSS-NCQ domains, although estimates were not statistically significant in adjusted models. Additionally, results were essentially unchanged if models were further adjusted by radiation exposure.
Respondents reported average Neuro-QoL (49.6±0.7) compared with an expected mean score of 50 in the general population (p=0.52 ; Table 3). The proportion with impaired cognitive quality of life with T-score <40 was 19.1% compared with 16% expected in the general population sample (p=0.23 ).21 These values were similar after accounting for inverse probability weights, with mean Neuro-QoL score of 49.2±0.7 and 20.5% impairment (Supplemental Table S2). In unweighted analysis, characteristics associated with lower Neuro-QoL scores included younger age (<10 years) at HCT, prior cranial irradiation, hearing issues, history of stroke or seizure, and sleep disturbances (allp<0.05 ). Sex, race/ethnicity, current age, underlying diagnosis, year of transplant, number of transplants, or history of cGVHD were not associated with a significant difference in Neuro-QoL scores. In survivors >25 years old, lower Neuro-QoL scores were associated with less educational achievement, defined as less than college completion (p<0.01 ). In multivariable regression analysis adjusted for sex and current age, impaired Neuro-QoL was independently associated with hearing issues (OR 4.79, 95% CI 1.91-12.0), history of stroke or seizure (OR 5.22, 95% CI 1.73-15.7), and sleep disturbances (OR 6.90, 95% CI 2.53-18.9; Figure 1 and Supplemental Table S3). Younger age (<10 years) at time of transplant was not significantly associated with impaired Neuro-QoL in multivariable analysis compared with age ≥ 10 years (OR 1.63, 95% CI 0.64-4.16).