AUTHORS
Ellen Van Der Plas, PhD, Simon Chung, MS2, Jason Farrar, MD3, Andrew Brown, PhD4, Timothy Koscik, PhD5
AFFILIATIONS
1 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
2 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
3 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
4 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
5 University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
ABSTRACT
Background/Objective:
Persistent neurocognitive impairment significantly impacts quality of life in pediatric survivors of acute lymphoblastic leukemia (ALL), yet therapeutic options for established neural damage remain limited. This ongoing prospective study uses a multimodal approach to identify early biomarkers of altered brain development during the first year of ALL therapy.
Methods:
Newly diagnosed ALL patients and healthy controls (ages 3–10) underwent serial evaluations across three time points, including computerized neurocognitive testing (NIH Toolbox; standard scores, mean=100, SD=15), non-sedated 3T brain magnetic resonance imaging (MRI), and blood sampling. MRI protocols evaluated neuroanatomy (T1w), microstructural integrity via biophysical modeling (soma and neurite density imaging [SANDI] diffusion-weighted imaging), and neurometabolite concentrations (magnetic resonance spectroscopy [MRS]). Plasma markers of axonal (neurofilament light, NfL) and astrocyte (glial fibrillary acidic protein, GFAP) injury were quantified using single molecule array. Treatment exposures were abstracted from the medical records with primary exposure modeling focused on intrathecal methotrexate (IT-MTX) dose. Linear mixed-effects models adjusted for age were used to evaluate longitudinal group differences and associations with IT-MTX exposure.
Results:
We evaluated 22 ALL patients and 30 controls (54.8% female; mean age = 6.8±2.0 years) across 124 total visits (patients: n=48; controls: n=76). Patients completed their initial visit at a mean of 3.5±2.4 months post-diagnosis. Compared to controls, ALL patients exhibited significant neurocognitive deficits in processing speed (mean difference in standard scores [MD]=-10.89; 95% Confidence Interval [CI]: -18.28, -3.50), working memory (MD=-7.41; 95%CI: -14.81, -0.01), inhibitory control (MD=-15.78; 95%CI: -21.64, -9.93), and mental flexibility (MD=-11.62; 95%CI: -17.35, -5.90). Biophysical modeling with SANDI revealed widespread, statistically significant loss of neurite fraction across the deep white matter in ALL patients relative to controls (cluster size=106,600mm3, group x time coefficient=0.13, p<0.00001). Conversely, patients demonstrated accelerated neurite fraction increases in the hippocampus (cluster size=418 mm3, group x time coefficient=-0.06, p<0.0001). These microstructural changes coincided with altered energy homeostasis on MRS, marked by increased lactate (MD parts per million [ppm]=0.044; 95%CI=0.02, 0.07) and decreased phosphocreatine (MD ppm=-0.06; 95%CI=-0.10, -0.01). Systemic injury markers revealed ~30-fold higher plasma NfL in ALL patients vs. controls at the initial visit, followed by partial recovery over time (Estimated change from visit 1 to visit 2=-59.47 pg/ml, 95%CI=-106.16, -12.78; Estimated change from visit 1 to visit 3=-117.36 pg/ml,95%CI=-168.47, -66.25). Higher average IT-MTX dose (mg/m2) positively correlated with elevated NfL levels (Estimate=1.16 pg/mL per 1 mg/m2 IT-MTX, 95%CI=0.65, 1.66). GFAP levels were somewhat elevated in ALL patients relative to controls (Estimate=295.56 pg/mL; 95%CI=1.37, 589.74, p=0.06), with no significant group by visit interaction.
Conclusion:
While pediatric ALL patients demonstrate early microstructural, metabolic, and cognitive deficits during the first year of treatment, the partial recovery of NfL alongside hippocampal neurite expansion potentially points to a neuroplastic response. These findings highlight a critical, dynamic window where early therapeutic interventions could mitigate or even prevent permanent cognitive impairment.
DISCLOSURES
Ellen Van Der Plas, PhD: No relevant conflicts of interest to declare.