Doctor: Karl Champeaux
Title: Meningioma – a natiowide study on survival
Supervisor: Matthieu Resche Rigon
Doctoral school: ED 393 Epidemiology and Biomedical Information Sciences, Université Paris Cité
Date of thesis defense: 07/2020
Jury: Matthieu Resche-Rigon, Sébastien Froelich, Michel Kalamarides, Philippe Tuppin, Isabelle Baldi, Évelyne Emery, Luc Bauchet, Franck-Emmanuel Roux
Meningioma is the most common primary intracranial tumour. Despite a consistent literature, there is less than a handful of studies reporting on outcome after meningioma surgery using nationwide administrative medical databases. The recent existence of the French nationwide healthcare database (SNDS) offered us the opportunity to study globally what we have also done locally. The reported incidence of meningioma has varied substantially over time and by the method of identification, from 1 to 8.4 per 100 000 with 5/100 000 in our study. We conclude that the SNDS is a reliable and effective source for studying the epidemiology of surgically treated meningiomas, including the precise location of the tumour asserting the usefulness of such a data base to investigate the patients’ outcome after meningioma surgery. Based on a single centre retrospective study of 501 patients operated on for a WHO grade I meningioma at Queen Elizabeth University Hospital, Glasgow, Scotland, we found that treatment failure correlated with venous sinus invasion, incomplete resection, and progressing tumour. Shorter survival correlated with increased age and redo surgery for recurrence. We recommend the cumulative incidence competing risk approach in meningioma studies where unrelated mortality may be substantial, as this approach results in more accurate estimates of disease risk and associated predictors. Then, in a multicentre cohort study involving 6 centres in France and in the United Kingdom, we confirmed the poor prognosis associated with malignant meningioma, the treatment of which remains challenging. Patients under 65-yr-old with primary malignant meningioma may live longer after complete resection and postoperative radiotherapy but even with aggressive treatments, local control remains difficult to achieve. Finally, we investigate the nationwide survival after meningioma surgery and search for associated factors using the SNDS. This nationwide study found 28 891 patients of which 75.4% were female. Median age at surgery was 59 years, IQR[49 – 68]. Cranial convexity (24.4%) and middle skull base (21.7%) were the most common locations. 91.2% of the tumours were benign and 2.7% malignant. 7.6% of the patients underwent redo surgery, 9.2% radiotherapy (RT) and 3.2% stereotactic radiosurgery (SRS) for recurrence. Median follow-up was 4.9 years, IQR[2.5-7.6]. At data collection, 8.3% were dead. 191 patients died within a month of surgery and 623 within a year. Overall survival rates at 5 years was: 92.6%, 95% CI[92.2-92.9]. In the multivariable analysis, gender (HR=0.6, 95% CI[0.55-0.65], p<0.001), age at surgery (HR=1.07, 95% CI[1.07-1.07], p <0.001), type 2 neurofibromatosis (HR=4.18, 95% CI[2.88-6.07], p<0.001), parasagittal (HR=1.18, 95% CI[1.03, 1.35], p=0.017) or falx cerebri location (HR=1.18, 95% CI[1.02-1.37], p=0.031), atypical (HR=1.32, 95% CI[1.14-1.54], p<0.001) or malignant (HR=2.97, 95% CI[2.56, 3.45], p<0.001), redo surgery (HR=2, 95% CI[1.79, 2.24], p<0.001) or RT for recurrence (HR=2.14, 95% CI[1.93, 2.37], p<0.001) were established as independent prognostic factors of survival. After removal of meningioma, survival of patients presenting no co-morbidities is long and outcome is better in women, younger adults, and patients with convexity and benign tumour. Once the tumour had relapsed, cure is unlikely and none of the treatment studied succeed to improve the survival. This study on meningioma using the SNDS is the first of its kind. Despite some limitations, the SNDS is an invaluable tool to assess meningioma outcome. It offers incomparable means to explore associations with other pathology, medication or other surgical treatment which has and could not be assessed before.