Predicting neurocognitive function in pediatric brain tumor early survivorship: The neurological predictor scale and the incremental validity of tumor size. Issue 9 (16th June 2022)

• Record Type: Journal Article
• Authors: McCurdy, Mark D.; Raghubar, Kimberly P.; Christopher, Krystal; Okcu, M. Fatih; Wilde, Elisabeth; Desai, Nilesh; Chu, Zili D.; Gragert, Marsha; Stancel, Heather; Warren, Emily H.; Whitehead, William E.; Grosshans, David; Paulino, Arnold C.; Chintagumpala, Murali; Kahalley, Lisa S.
• Is Part Of: Pediatric blood & cancer. Volume 69:Issue 9(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-06-16
• Subjects: Cognitive function -- late effects -- outcomes -- proton radiation therapy; Tumors in children -- Periodicals; Blood -- Diseases -- Periodicals; Cancer in children -- Periodicals; 618.92
• DOI: 10.1002/pbc.29803
• Languages: English
• ISSNs: 1545-5009

Abstract

Abstract: Background: The Neurological Predictor Scale (NPS) quantifies cumulative exposure to conventional treatment‐related neurological risks but does not capture potential risks posed by tumors themselves. This study evaluated the predictive validity of the NPS, and the incremental value of tumor location and size, for neurocognitive outcomes in early survivorship following contemporary therapies for pediatric brain tumors. Procedure: Survivors ( N = 69) diagnosed from 2010 to 2016 were administered age‐appropriate versions of the Wechsler Intelligence Scales. Hierarchical multiple regressions examined the predictive and incremental validity of NPS score, tumor location, and tumor size. Results: Participants (51% female) aged 6‐20 years ( M = 13.22, SD = 4.09) completed neurocognitive evaluations 5.16 years ( SD = 1.29) postdiagnosis. The NPS significantly predicted Full‐Scale Intelligence Quotient (FSIQ; Δ R 2 = .079), Verbal Comprehension Index (VCI; Δ R 2 = 0.051), Perceptual Reasoning Index (PRI; Δ R 2 = 0.065), and Processing Speed Index (PSI; Δ R 2 = 0.049) performance after controlling for sex, age at diagnosis, and maternal education. Tumor size alone accounted for a significant amount of unique variance in FSIQ ( Δ R 2 = 0.065), PRI ( Δ R 2 = 0.076), and PSI ( Δ R 2 = 0.080), beyond that captured by the NPS and relevant covariates. Within the full model, the NPS remained a significant independent predictor of FSIQ (β = −0.249, P = 0.016), VCI (β = −0.223, P = 0.048), and PRI (β = −0.229, P = 0.037). Conclusions: Tumor size emerged as an independent predictor of neurocognitive functioning and added incrementally to the predictive utility of the NPS. Pretreatment disease burden may provide one of the earliest markers of neurocognitive risk following contemporary treatments. With perpetual treatment advances, measures quantifying treatment‐related risk may need to be updated and revalidated to maintain their clinical utility.