Childhood brain tumors - InSight in Sight

Abstract

Brain tumors are the most common solid tumors in children with an estimated age-adjusted incidence of 6.21 per 100,000. Recent advances in the diagnosis, treatment and surveillance of childhood brain tumors have considerably improved survival, with a current five-year survival rate reaching 75% in developed countries. This improved survival rate stresses the importance of awareness of the adverse effects coinciding with the brain tumor or its treatment. One of these adverse effects is an impaired visual function, which poses a substantial burden on the health, quality of life, and participation in daily life of children with a brain tumor.

This thesis provides insight into the ophthalmological consequences in children with a newly diagnosed brain tumor and in the potential role of retinal optical coherence tomography (OCT) as an objective, non-invasive testing method of the visual function. To investigate these aims, we designed and conducted the CCISS study (‘Child Central nervous tumors InSight in Sight’), a prospective nationwide longitudinal cohort study investigating visual impairment in 170 children newly diagnosed with a brain tumor between May 2019 and August 2021 in the Netherlands.

Our analyses have shown that a major part of the children (79%) have ophthalmological abnormalities at brain tumor diagnosis and that ophthalmological abnormalities during examination were present in the majority of children (65%) who initially presented without visual symptoms. These findings stress the importance of standardized ophthalmological examination and the awareness of clinicians for latent ophthalmological abnormalities in children with a newly diagnosed brain tumor.

In addition, we investigated the value of retinal OCT thickness measurements to discriminate an abnormal visual function in children with a newly diagnosed brain tumor. We found relatively high negative predictive values of average circumpapillary RNFL thickness (83%) and average macular GCL-IPL thickness (82%), but the positive predictive values (respectively 33% and 57%) were low, demonstrating a moderate diagnostic accuracy. This is also is in line with the low to moderate sensitivity and specificity. Therefore, at this moment, RNFL thickness and GCL-IPL thickness measurements cannot be used for clinical care decision making in children with a newly diagnosed brain tumor, nor should these measurements replace the thorough standard ophthalmological examination.