Malignant Solitary Fibrous Tumours of the Pleura Are Not All the Same: Analysis of Long-Term Outcomes and Evaluation of Risk Stratification Models in a Large Single-Centre Series

Introduction: Malignant solitary fibrous tumours of the pleura (mSFTP) are extremely rare diseases (Methods: Observational retrospective cohort study on all proven cases of mSFTP surgically resected with radical intent between 2000 and 2019 in a single centre. Demographic, surgical and pathological data were examined. All patients were risk-stratified by using three prediction models: modified Demicco, De Perrot and Tapias. Overall survival (OS) and disease-free survival (DFS) were analysed. Results: There were 21 men and 13 women (median age, 67 years, range, 23–83 years). Twenty-one patients (62%) were symptomatic. The median follow-up was 111 months (range, 6–258 months). The 5-year OS and DFS were 81.2% and 77.4%, respectively. Nine patients (26.5%) experimented recurrences. At univariate analysis, the presence of necrosis (p = 0.019), nuclear atypia (p = 0.006), dimension greater than 11.5 cm (median value of our cohort) (p = 0.037) and relapse/disease progression (p = 0.001) were independent prognostic factor of worse OS. The administration of adjuvant treatment was a protective independent factor for survival (p = 0.001). Radicality of resection (p = 0.005); tumour dimension (p = 0.013), presence of necrosis (p = 0.041) and nuclear atypia (p = 0.007) and pleural pattern (p = 0.011) were independent prognostic factors of worse DFS. Analysing the three risk stratification models, the Tapias score was revealed as the best index to predict both OS (p = 0.002) and DFS (p = 0.047) in patients with mSFTP. Conclusions: Using the risk stratification model proposed by Tapias, patients with the highest risk of recurrence could be identified at the time of surgery to establish a more frequent imaging surveillance and longer follow-up. The role of adjuvant treatment in mSFTP therapy has not been established yet, but further analysis on patients with a high risk of recurrence, stratified according to risk models, along with biomolecular panels may tailor future post-surgical therapies.