gms | German Medical Science

Objective: Despite therapeutic improvements, the median overall survival after glioblastoma (GBM) diagnosis remains 14.6 months. Overall survival can be improved by early recognition of tumor progression. Yet, one of the major challenges is to differentiate between true and pseudoprogression, which is a reversible reaction that can occur following radiochemotherapy with radiologic changes mimicking tumor recurrence, with or without neurological deterioration. The aim of this study was to define predictive clinical and molecular markers to identify pseudoprogression.

Methods: Diagnosis of tumorprogression vs. pseudoprogression was confirmed either by histopathological findings or by re-approved assessment according to the RANO criteria on follow-up MRIs. Age, molecular markers (IDH (Isocitrate dehydrogenase) mutation, MGMT (O-6-methylguanine-DNA methyltransferase) promotor methylation), Karnofsky performance status scale (KPS) extent of resection, timing following radiation and number of operations were correlated with occurrence of pseudoprogression. For statistical analysis, chi-square values, independent t-test, Kaplan-Meier survival analysis and boxplots were used. Results with p<.05 were considered to be significant.

Results: From 2002 to 2015 in 151 glioblastoma patients (135 primary GBM, 16 secondary GBM, f=49, m=102, mean age=55 years, IDH mutation positive n=13, MGMT positive n=42) initial MRI findings were suspicious for tumor progression. Complete resection was achieved in 116, while 35 were incomplete resected. In 85 patients (56.3%) a pseudoprogression was diagnosed histopathologically (n=30) or via follow-up MRI (n=55) in a median timeframe of 4 months (range 1-23 months) following radiochemotherapy. There was a significant correlation between the diagnosis of pseudoprogression and patients < 50 years (p=0.024) and positive MGMT (p = 0.001). IDH mutation (p=0.17), KPS (p=0.394), number of previous resections (p=0.09) and extent of resection (p=0.338) did not reveal a significant impact. Overall survival (OAS) as well as progression-free survival (PFS) were significantly increased in patients with diagnosed pseudoprogression (14 vs 35 months OAS (p=0.002), 6 vs 23 PFS (p=0.007), respectively).

Conclusion: Pseudoprogression is associated with MGMT status and older age. There is no correlation with IDH mutation, KPS, number of resections and extent of resection. OAS and PFS were increased in patients with a diagnosed pseudoprogression. Following this first analysis, identifying important markers, we intend to develop a score system to predict pseudoprogression and to avoid unnecessary surgeries.