7140 Assessing Bone Mineral Density and Cortical Geometry Using High-resolution Peripheral Quantitative Computed Tomography and Dual Energy X-ray Absorptiometry in Pediatric Survivors of High-risk Neuroblastoma with Growth Failure

Abstract Disclosure: S. Gera: None. M. Guo: None. N. Pollock: None. I. Hampton: None. M. Song: None. D. Weber: None. M. Denburg: None. S. Mostoufi-Moab: None. Introduction: Survival of children with high-risk neuroblastoma (HR-NBL) has increased with multimodal therapy. Cis-retinoic acid (cis-RA), cornerstone of HR-NBL therapy, can cause osteoporosis and premature physeal closure, posing a threat to skeletal structure. HR-NBL survivors demonstrate profound growth failure. DXA measures areal bone mineral density (aBMD). DXA is confounded by short stature and fails to distinguish between trabecular and cortical bone. We previously reported no significant difference in height-per-age adjusted aBMD-Z in HR-NBL survivors compared to healthy controls[1]. However, aBMD-Z in severe short stature may attenuate bone density deficits. High-resolution peripheral quantitative computed tomography (HRpQCT) yields 3D measures of volumetric BMD and bone architecture. Objectives: Assess volumetric BMD (vBMD) and cortical structure using distal tibia HRpQCT and leg lean mass by DXA in HR-NBL survivors compared to healthy controls in a cross-sectional study. Methods: We prospectively enrolled 20 HR-NBL survivors (median age 12.2 yrs, range 9.5 to 15.8, 7 female) treated with cis-RA (median age 2.6 yrs at start of therapy, range 0.3-9.1), and 20 age-, sex-, and race-matched healthy controls. We assessed leg lean mass adjusted for leg length by DXA. For HRpQCT, we compared tibia length (cm), cortical and trabecular vBMD (mg HA/cm3), geometry parameters (cortical and trabecular area [mm2]), and structural parameters (cortical thickness [μm] and porosity, trabecular number [mm-[1]] and separation [μm]). Outcomes were compared using Student’s t-test or Mann-Whitney U test. Linear regression models assessed difference in outcomes of bone microarchitecture and leg lean mass between both groups. Results: Compared to controls, tibia length was significantly shorter in HR-NBL survivors (32.3 ± 0.8 vs. 35.7 ± 0.6, P<0.005). HR-NBL survivors demonstrated lower cortical area (187.6 ± 43.8 vs. 218.7 ± 42.3, P<0.05), cortical perimeter (62.4 ± 7.71 vs. 69.5 ± 8.1, P<0.01), and leg lean mass Z-score (-1.47 ± 1.34 vs. -0.20 ± 0.84, P<0.001) compared to controls. Cortical geometry deficits were explained by muscle deficits (P<0.001). Trabecular area, BMC, areal and volumetric BMD, and measures of bone structure were not significantly different in HR-NBL survivors compared to controls. IGF-1 and 25-OH D levels had no significant effects on DXA or HRpQCT parameters. Conclusion: Bone density is not severely impacted in HR-NBL survivors. Muscle deficits persist years after treatment and explain cortical geometry deficits. HRpQCT substantiates prior DXA evaluation in HR-NBL survivors with significant short stature, demonstrating preserved bone density and structure. 1.Guo et al. Sarcopenia and preserved bone mineral density in pediatric survivors of high-risk neuroblastoma with growth failure. J Cachexia Sarcopenia Muscle. 2021 Aug;12(4):1024-1033. Presentation: 6/3/2024