Children and adolescents with sickle cell disease demonstrate improved health-related quality of life and low decisional regret after hematopoietic cell transplantation: A sickle cell transplant advocacy and research alliance (STAR) study

Abstract Hematopoietic cell transplantation (HCT) for pediatric sickle cell disease (SCD) has steadily increased, however data on long-term effects is limited. Previous registry studies focused on survival, rejection and GVHD. Data on long-term health-related quality of life (HRQoL) post-HCT in SCD is limited by small sample sizes and short follow-up. The objective of Project Sickle Cure (PSC) is to measure HRQoL and decisional regret (DR) in children with SCD, hypothesizing that HCT will improve HRQoL longitudinally with low rates of DR. PSC is a prospective observational multicenter study of SCD participants undergoing allogeneic HCT. Since 2021, 15 STAR sites in North America have enrolled 89 participants, with 65 having >6 mo post-HCT follow-up and included in this analysis. Baseline demographics, disease, and HCT characteristics were collected and summarized. Participants and their primary caregivers complete HRQoL measures (PedsQL Generic Core [GC], PedsQL Family Impact Module [FIM], and Patient Reported Outcomes Measurement Information System [PROMIS]) measures alongside a new DR Scale (DRS; higher scores indicate greater regret). Data are collected pre-HCT and post-HCT at 30, 90, and 180 days (d) and 1, 2, and 3 years (yrs). General linear mixed-effects models were used to evaluate preliminary changes over time. Results are presented as least-square means (LSM) and differences in least-square means (ΔLSM) from baseline. In this interim exploratory analysis, no statistical adjustments to p-values (e.g., alpha spending or corrections for multiple testing) were applied; results may differ after full enrollment and completion of planned analyses. Mean duration of follow up was 2.3 yrs (SD=1.0), with 85% >1 yr post-HCT. Participants were 46% male with mean age at HCT of 10.1 yrs (SD=5.3). Most had HbSS genotype (94%), and disease severity was severe, less severe, and mild in 35%, 35%, and 30%, respectively. Donor was matched related in 75%, bone marrow graft source in 68%, and conditioning was myeloablative in 53%, reduced intensity in 38%, and nonmyeloablative in 9%. Acute GVHD of any severity was reported in 34%, and any chronic GVHD in 5%. Many participantshad significant improvement from baseline in multiple domains of HRQoL, some with sustained improvement and from the earliest timepoint. For PedsQL GC, patient (pt) and parent-proxy HRQoL significantly increased from LSM=71.7 and 73.7 at baseline starting d 180 and 1 yr to 78.3 and 81.2 by yr 2, respectively (pt ΔLSM (95% CI)=6.6 (0.6,12.6), p=.032; parent-proxy ΔLSM (95% CI)=7.5 (0.1,14.9), p=.046). For PedsQL FIM, parent-reported total family functioning significantly increased from baseline of LSM=76.3 to 88.3 starting d 180 to yr 3 (ΔLSM (95% CI)=12.1 (2.9, 21.2), p=.011). Patient PROMIS scores significantly improved from baseline for anxiety (starting d 30) and fatigue up to yr 1 (anxiety ΔLSM (95% CI) =-6.0 (-10.5,-1.5), p=.009; fatigue ΔLSM (95% CI)=-6.8 (-12.4,-1.2), p=.019), as well as mobility from yr 2 to yr 3 (ΔLSM (95% CI)=7.2 (0.3,14.1), p=.040) and pain interference from yr 1 up to yr 2 (ΔLSM (95% CI)=-7.9 (-14.8, -1.0), p=.024). Parent-proxy scores significantly improved from baseline up to yr 3 for anxiety (from d 180; ΔLSM (95% CI)=-8.6 (-15.4,-1.8), p=.013), fatigue (from 1 yr; ΔLSM (95% CI)=-9.2 (-16.1,-2.3), p=.009), mobility (from yr 2; ΔLSM (95% CI)=6.6 (1.3,12.0), p=.016), and pain interference (from d 90; ΔLSM (95% CI)=-11.0 (-17.8,-4.2), p=.002). Pt and parent-proxy depression did not significantly differ from baseline. Though pt peer relationship scores significantly decreased from baseline to d 180 (ΔLSM (95% CI)=-4.8 (-9.1,-0.4),p=.031), all scores remained above cut-off for impairment. For DRS, pt and parent-proxy scores indicated low regret up to yr 3 (LSM=17.5 and 11.2, respectively), which did not significantly differ from baseline at d 90. Most PROMIS HRQoL scores were normal pre-HCT in our cohort. Previous studies with smaller sample sizes have demonstrated both low and high baseline scores. However, it is the favourable change from baseline which is of greater relevance to patients and providers. In conclusion, our interim analysis demonstrated significant improvements in multiple domains of HRQoL at longitudinal timepoints, as early as 30 d and lasting up to 3 yrs post-HCT with low DR. Our findings are strengthened by our large cohort, duration of follow-up and the use of diverse measures, including a novel DRS.