Abstract 3290
Background
Meningioma growth rates are highly variable, even within benign subgroups, causing some cases to remain stable while others grow rapidly despite radiotherapy. Biomarkers that differentiate meningiomas by aggression and enable prediction of their biological behavior would therefore be clinically beneficial.
Methods
Microarrays were used to identify microRNA (miRNA) expression in primary recurrent, non-recurrent and secondary meningiomas of all grades. miRNAs found to be deregulated in the microarray experiments were validated by quantitative real-time PCR using samples from a cohort of 191 patients (median age 56). Statistical analysis of the resulting dataset revealed miRNA predictors of meningioma recurrence.
Results
miRNAs exhibiting differential expression (independently of histological grade) in primary recurrent, non-recurrent and secondary meningiomas were identified. The most effective predictive model included miR-331-3p, extent of tumor resection and its localization as predictive markers. The model with a recurrence probability cut-off of 28% and small number of the input data (7) had a high area under the curve (AUC) (0.829), sensitivity (75%), specificity (75%), and acceptable leave-one-out cross-validation (LOOCV) test error (23.2%). miR-18a-5p, miR-130b-3p, miR-146a-5p, miR-1271-5p, age at diagnosis, gender and histological grade showed to be supportive but not predictive factors in the tested models.
Conclusions
This model is a novel predictor of meningioma recurrence that could facilitate optimal postoperative management. Moreover, combining this model with information on the molecular processes underpinning recurrence could enable the identification of distinct meningioma subtypes and targeted therapies.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Ministry of Health of the Czech Republic (15-29021A); Palacky University Olomouc (LF 2019_003); Ministry of Education, Youth and Sports of the Czech Republic (LO1304, LM2015091); European Regional Development Fund (ENOCH CZ.02.1.01/0.0/0.0/16_019/0000868).
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
5801 - Implementing a Population-Based Cancer Registries Network in Mexico
Presenter: Alejandra Palafox-Parrilla
Session: Poster Display session 3
Resources:
Abstract
4385 - Awareness of cancer risk factors among high school students
Presenter: Monika Rucinska
Session: Poster Display session 3
Resources:
Abstract
3355 - COMPRAYA: Dutch nationwide infrastructure for COMPRehensive health outcome and intervention research among patients with Adolescent and Young Adult cancer
Presenter: Olga Husson
Session: Poster Display session 3
Resources:
Abstract