Survival of patients with prior anti-angiogenic therapy

We read with interest the report by Cloughesy et al on the activity of cabozantinib in recurrent glioblastoma (GBM) patients with prior anti-angiogenic therapy.¹ In their subset analysis, they reported overall survival (OS) results using 2 different cabozantinib doses: 4.1 months (95% CI: 1.4–16.7) in the 140 mg group (n = 12); 4.6 (95% CI: 2.9–5.6) in the 100 mg group (n = 58); OS for combined groups was 4.6 months. Of the 70 patients in the analysis, 57 (81%) with prior anti-angiogenic therapy had bevacizumab (BEV). In their discussion, the authors raise the issue of “how best to define a meaningful response (eg, objective response rate, progression-free survival).”¹ We argue, because of the vagaries in interpreting MRIs in patients receiving BEV or other anti-angiogenic agents, the most meaningful endpoint is OS. Relative to this, we suggest literature can provide a valuable benchmark, as the OS outcome for these patients tends to be relatively uniform. Such a review of the literature² exemplifies this argument and compares well with the reported results in the cabozantinib study (ie, group 3 below):

A total of 2388 patients from 53 arms of 42 studies were analyzed in 3 groups²:

• Group 1: Thirty-two studies in which survival post-BEV was determined by subtracting PFS from OS (2045 patients); OS post-BEV = 3.36 months (95% CI: 3.12–3.66)

• Group 2: Two studies (94 patients) in which OS post-BEV is reported: OS = 3.26 (95% CI: 2.394.42)

• Group 3: Eight studies of salvage therapy after progression on BEV (249 patients): OS post-BEV = 4.46 months (95% CI: 3.68–5.54).

Funding

This work was supported by the National Institutes of Health National Cancer Institute CORE grant (no. 4P 30 CA014520).