Inefficiencies in the phase II to phase III transition as a modifiable factor that is impeding successful drug development for glioblastoma.

Abstract

2516 Background: Improving the outcomes of patients with glioblastoma (GBM) represents one of the most significant challenges in neuro-oncology. We have observed inefficiencies in the availability and use of phase 2 data when planning phase 3 studies, and have undertaken a detailed review of key design parameters of phase 2 and 3 trials in GBM to identify and quantify the impact of this phenomenon. Methods: Studies between 2005-2019 inclusive were identified though MEDLINE search using keywords and MeSH terms, and manual bibliography searches. P2Ts were restricted to those referenced by the corresponding P3Ts. Clinical, statistical and sponsor characteristics were extracted by two reviewers (AB&AG). For each P3T, corresponding Phase 2 trial (P2T) data was “optimally matched” (OM) where same drug was used in similar schedule and similar GBM population; “partially matched” (PM) where dis-similar schedule and/or treatment setting; and “lacking” in all other circumstances. The statistical data used in the P2/3 transition were compared by Pearson Correlation, Fisher’s Exact or Chi-square testing as appropriate. Results: Of 20 P3Ts identified, 6 (30%) lacked any phase 2 data. Of the remaining 14 P3T, 9 had 1 prior P2T, 4 had 2 P2T and 1 had 3 P2T, for a total of 20 P3T-P2T pairs (called dyads). Further, there were 13 OM dyads and 7 PM dyads. OM dyads showed strong concordance for mPFS (r2= 0.95, p < 0.01) and mOS (r2= 0.84, p < 0.01), whilst PM dyads did not (p > 0.05). We identified several inefficiencies in translation from P2T to P3T. Firstly, 3 P3T had statistical assumptions of primary endpoint that may have been too optimistic. 2 of these P3Ts aimed for an expected endpoint that was higher than the actual outcomes from a matched P2T. 1 P3T was unable to reach the desired sample size. We note that 4 P3Ts had actual primary endpoint HRs that were < 0.9 but with P > 0.05. Finally, we investigated whether there were absolute thresholds for efficacy in P2Ts to inform whether to proceed with P3Ts. For P2Ts in the newly diagnosed setting, all those with mPFS < 14 months and/or mOS < 22 months had subsequent negative P3Ts. For P2Ts in recurrent disease, all those with mPFS < 6 months and mOS < 12 months had negative P3Ts. Applying these thresholds to the studies in our review, 10 of the 12 negative P3Ts (83%) with matched P2Ts need not have been initiated, sparing 4739 patients’ from unnecessary trial participation. Conclusions: Our data strongly supports the vital role of properly designed P2Ts in informing P3Ts for drug development for primary CNS tumours.