Evaluation of four tumour growth models to describe the natural history of meningiomas
Résumé
Background: The incidence of newly diagnosed meningiomas, particularly those diagnosed incidentally, is continually increasing. The indication for treatment is empirical because, despite numerous studies, the natural history of these tumors remains difficult to describe and predict. Methods: This retrospective single-centre study included 294 consecutive patients with 333 meningiomas who underwent three or more brain imaging scans. Linear, exponential, power, and Gompertz models were constructed to derive volume-time curves, by using a mixed-effect approach. The most accurate model was used to analyse tumour growth and predictors of rapid growth. Interpretation: Meningiomas exhibit multiple growth phases, as described by the Gompertz model. The management of meningiomas should be discussed according to the growth phase, comorbidities, tumour location, size, and growth rate. Further research is needed to evaluate the associations between radiomics features and the growth phases of meningiomas.
Domaines
Mathématiques [math]Origine | Fichiers produits par l'(les) auteur(s) |
---|