Inter-interpreter reliability of neuropathological assessment of disease status after early repeat resection for suspected recurrent glioblastoma.

2050Background: Distinction of true progression from pseudo-progression in patients with suspected recurrent glioblastoma (rGBM) is an important and common diagnostic challenge in neuro-oncology. The “gold standard” for assessment of post-treatment disease status has remained surgery with histologic evaluation. However, after chemoradiation, tissue frequently contains a variable mixture of scar tissue and tumor cells. Neuropathologists are left to determine whether patients have active tumor or treatment effect. As validity and reliability are critical for any diagnostic test, our study assessed inter-interpreter reliability in the pathological assessment of disease status in this setting. Methods: Histopathology images from 13 patients with GBM who underwent surgery within one year of diagnosis for suspected rGBM were selected. 101 neuropathologists at 50 NCI-designated Cancer Centers received a case scenario of a patient with suspected rGBM, including histopathological information of the initial diagnos...