Predicting Cognitive Dysfunction After Pediatric Posterior Fossa Tumor Surgery

Development and Internal Validation of a Machine Learning-Enhanced Nomogram to Stratify Risk of Chronic Cognitive Impairment Following Pediatric Posterior Fossa Surgery

The goal of this observational study is to develop and validate a clinical prediction model to identify risk factors for long-term cognitive dysfunction in children (ages 0-18 years) who have undergone surgical resection of a posterior fossa tumor. The main questions it aims to answer are:

Can a combination of preoperative and postoperative clinical, surgical, and neuroimaging factors accurately predict which children will develop long-term cognitive dysfunction after posterior fossa tumor surgery? Is white matter integrity-specifically fractional anisotropy (FA) of the superior cerebellar peduncle (SCP)-a key independent predictor of cognitive outcomes? Researchers will compare children who developed long-term cognitive dysfunction (cases) to those who did not (controls) to see if differences in imaging biomarkers (e.g., SCP FA, fMRI abnormalities), tumor characteristics (e.g., location, volume, histology), treatment factors (e.g., radiotherapy, surgical approach), and demographic variables (e.g., age) are associated with cognitive outcomes.

Participants were not asked to perform any tasks or receive any interventions as part of this study, because it is a retrospective analysis of existing medical records and imaging data. Data collected included:

Preoperative and postoperative brain MRI and DTI scans Tumor pathology and surgical reports Treatment details (e.g., radiation, chemotherapy) Neuropsychological assessment results at 1-year follow-up

Study Overview

• Status: Completed
• Conditions: Posterior Fossa Tumors; Pediatric Tumor of Brain
• Intervention / Treatment: Other: Not applicable- observational study

• Study Type: Observational
• Enrollment (Actual): 600

Contacts and Locations

Study Locations

• China
  • Sichuan
    • Chengdu, Sichuan, China
      • West China Hospital of Sichuan University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

This study includes 600 children aged 0-18 years who underwent surgical resection of a primary posterior fossa tumor (e.g., medulloblastoma, ependymoma, or pilocytic astrocytoma). All participants had preoperative brain MRI with diffusion tensor imaging (DTI), received standard-of-care treatment, and completed comprehensive neuropsychological assessments at approximately 1 year post-surgery. The cohort represents a real-world pediatric neuro-oncology population treated at multiple tertiary care centers.

Description

Inclusion Criteria:

• Age 0-18 years at time of surgery
• Histopathologically confirmed primary posterior fossa tumor (e.g., medulloblastoma, ependymoma, pilocytic astrocytoma)
• Underwent surgical resection of the tumor at the hospital
• Availability of preoperative brain MRI and diffusion tensor imaging (DTI)
• Completed standardized neuropsychological assessment at approximately 1 year post-surgery

Exclusion Criteria:

• Pre-existing neurological or neurodevelopmental disorders (e.g., autism, intellectual disability, cerebral palsy)
• History of prior cranial irradiation or chemotherapy before posterior fossa surgery
• Incomplete clinical, imaging, or follow-up data required for model variables
• Tumor recurrence or progression before 1-year cognitive assessment
• Non-posterior fossa primary brain tumors (e.g., supratentorial gliomas)

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
pediatric patients who underwent surgical resection of a posterior fossa tumor; Stratify this cohort into two subgroups based on outcome: Those with long-term cognitive dysfunction (n=172) Those without (n=428)	Other: Not applicable- observational study

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Incidence of Long-Term Cognitive Dysfunction	Presence or absence of long-term cognitive dysfunction, defined as a full-scale IQ (FSIQ) score < 85 or significant impairment in ≥2 cognitive domains (e.g., attention, memory, executive function, processing speed) on standardized neuropsychological assessment at 1-year follow-up.	1 year

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Model Discrimination Performance	Area Value under the receiver operating characteristic curve (AUC) of the final predictive model (nomogram) in the internal validation cohort.	1 year post-surgery
Model Calibration Accuracy	Agreement degree between predicted probability and observed frequency of cognitive dysfunction, assessed by Hosmer-Lemeshow test P Value.	1 year after surgery

Collaborators and Investigators

• Sponsor: West China Hospital

Study record dates

Study Major Dates

• Study Start (Actual): January 1, 2015
• Primary Completion (Actual): December 30, 2023
• Study Completion (Actual): January 1, 2024

Study Registration Dates

• First Submitted: January 5, 2026
• First Submitted That Met QC Criteria: January 13, 2026
• First Posted (Actual): January 20, 2026

Study Record Updates

• Last Update Posted (Actual): January 20, 2026
• Last Update Submitted That Met QC Criteria: January 13, 2026
• Last Verified: January 1, 2024

More Information

Terms related to this study

• Keywords: Predictive model; Machine learning; Cognitive dysfunction; Nomogram; Pediatric posterior fossa tumor; Cerebellar cognitive affective syndrome; Calibration curve
• Additional Relevant MeSH Terms: Brain Diseases; Central Nervous System Diseases; Nervous System Diseases; Mental Disorders; Neoplasms by Site; Neoplasms; Neurocognitive Disorders; Cognition Disorders; Nervous System Neoplasms; Central Nervous System Neoplasms; Brain Neoplasms; Cerebellar Diseases; Cognitive Dysfunction; Infratentorial Neoplasms; Cerebellar Cognitive Affective Syndrome
• Other Study ID Numbers: WestChinaH-HX-2025-012

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES
• IPD Sharing Supporting Information Type: STUDY_PROTOCOL

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No