Molecular pattern characterization and clinical outcome of recurrent glioblastoma multiforme.

Glioblastoma multiforme (GBM) is the most frequent primary malignant brain tumor of adulthood and represents the subtype with the highest aggressiveness and incidence of recurrence among cerebral gliomas. Despite the technical-scientific developments in diagnostic neuroimaging, surgery and adjuvant therapies of the last decades, the prognosis remains poor due to the extensive invasion and diffusion of tumor cells within the brain tissue, which makes almost impossible complete surgical resection. Considering the importance that molecular pattern profiles and surgical resection can have for this type of tumors in the next future and starting from these premises, a retrospective study was carried out in order to highlight the modifications of molecular patternsof GBMs at recurrence and the impact that these genetic modifications can have on surgery and on OS in a cohort of patients with recurrent GBM who underwent repeat surgery. The aim of this project study was to assess the molecular pattern profiles at first and second surgery in patients with recurrent GBM, to identify possible evolution of these patterns in each GBMs and to evaluate if these changes affect the OS. In addition, other prognostic risk factors were evaluated in order to assess if they affect OS of patients with recurrent GBM. Between January 2006 and July 2020, 63 patients underwent second surgery for recurrent GBM. The mean age at first surgery was 59.2 (range 43-78 years). Median OS resulted 22 months (range 2-168 months). Median time between first and second surgery, resulted as PFS, was 10 months (range 1-96 months), whereas the median time between second surgery and death was 8 months (range 0.1-111 months). Within the study group as a whole, prognostic factors were independently associated with prolonged OS included sex, PFS, third surgery, MGMT methylation, gross-total resection and adjuvant chemotherapy at recurrence. Age, site of tumor location and size of tumor at second surgery did not statistically affect OS. Median PFS after initial resection was 10 months (range 2–96 months). PFS affects OS and it was statistically significant (p < 0.0001). In a multivariate analysis female sex (HR = 0.322, 95% CI 0.147–0.705; p = 0.005), PFS (HR = 0.959, 95% CI 0.934–0.986; p = 0.003), GTR at first and second surgery (HR = 0.195, 95% CI 0.091–0.419; p < 0.0001) and adjuvant chemotherapy at recurrence (HR = 29 0.407, 95% CI 0.206–0.809; p = 0.01) were associated with longer OS. GTR at first surgery and at recurrence was associated with a median OS of 29 months compared to patients underwent STR at recurrence (19 months) and compared to patients underwent GTR at recurrence but STR at first surgery (13 months). The results of this study confirmed that EOR at first and at recurrence is an important significant predictor of outcome in patients with recurrent GBM. According to our findings, repeat craniotomy should be offered to all patients in good performance status at the time of tumor recurrence with the aim of achieving a maximal resection when it is safe and feasible.