A randomized trial compares patients randomized to treatment to those randomized to the control arm. And that comparison failed to show any improvement in either PFS or OS.
Any attempt to subsequently compare to ECAs is not going to fly. In the words of an RA, the ECA comp will be hypothesis forming.
Not that complicated.
One of your many many BS claims and lies here
You are misstating the DCVax-L trial design and what regulators accept. 1. “Randomized comparison failed” is a strawman. The DCVax-L Phase 3 had mandated crossover: control patients received DCVax-L at progression. That design precludes a clean OS comparison between randomized arms (the control arm is “contaminated”). The trial was not powered to show OS between original randomization groups for exactly that reason. So calling the OS/PFS result “a failed randomized comparison” ignores the protocol.
2. PFS isn’t decisive in GBM with immunotherapy. Dendritic-cell vaccines can cause pseudo-progression and imaging artefacts; GBM PFS under RANO is notoriously noisy for immune therapies. That’s why OS (with appropriate controls) is the meaningful endpoint.
3. External control arms (ECAs) are explicitly acceptable in this setting. Regulators (including MHRA) have published guidance on using RWE/ECAs when RCTs are infeasible or unethical in small, lethal indications. ECAs are not “hypothesis-forming only” by default; they can support marketing decisions when pre-specified, well-matched, and methodologically robust (eligibility alignment, contemporaneous datasets, prespecified SAP, sensitivity analyses). Many cell/gene/oncology approvals have relied on single-arm data contextualized with external controls.
4. Peer-reviewed ECA results for DCVax-L exist. The Phase 3 program reported statistically significant OS advantages vs large, contemporaneous external control cohorts in both newly diagnosed and recurrent GBM under a prespecified analysis plan. You can dispute interpretation, but calling all ECA comparisons “hypothesis-forming” is simply wrong in today’s regulatory science. But then again, Wrong is probably your first name and Exwannebuzz your last
The trial’s crossover design makes a head-to-head OS comparison non-informative by design; that’s why robust ECAs were used. And contrary to your claim, ECAs are an accepted basis for decisions in small, high-mortality settings when done correctly.