The significance of Abscopal Responses

The ultimate aim of any biotech company is to gain FDA marketing approval for their drug.  This is bloody hard, as it should be.  In essence, a new drug has to show that it is statistically superior to the best drug currently being used to treat a given condition.

“Statistically significant" superiority (P < .05) is the yardstick used by the FDA.  Basically, one must show that the difference in responses between two treatment arms was not due to chance.  If a trial was to compare overall survival in breast cancer, for instance, and the expected results were 62 months vs 60 months, a trial would require thousands of patients to be able to show statistical superiority; and might take ten years to complete.  Less patients would be required if the expected results were 80 months vs 60 months, but the trial would still take a long time before it could be stopped for success.

The FDA recognises this dilemma and offers drug companies the opportunity to conduct shorter trials by using “surrogate markers” to predict overall survival.  One common marker is Progression Free Survival (PFS), a measure of the time it takes for tumours to grow by 20%, or for a new lesion to appear, whichever comes first.  This shortens the trial duration, but you still need enough patients to prove superiority.

If the expected difference between treatment arms is large, fewer patients would be required.  And this is where Abscopal Responses could be important.

Graham Kelly: The patients we are using in DARRT-1 share one certainty – their cancers have stopped responding to all approved (and most non-approved) treatments, the cancers will grow inexorably, the patients will be treated palliatively with high-dose opioids, and they will die.

If the benefit of NOX66 was limited to its effect on the irradiated lesions in these patients, it probably would mean short-term improvements in quality of life issues (and that might be enough to get you a marketing approval), but it isn’t likely to have any major impact on PFS or overall survival because the rest of the cancers will continue on their merry way. That sort of limited response puts you on a knife-edge when it comes to a Phase 3 trial outcome, plus it means needing to use many hundreds of patients.

We are focusing on an abscopal response because that gives us a red hot shot at improving PFS and OS. We only need to show an abscopal response in about 15% of men, and it only needs to be partial (say, lasting 6 months) for that to give us a statistically meaningful increase in PFS with a trial length of 12 months. This is a far more certain outcome and one that could mean needing to use a small number of patients. If we continue to see even a partial (with the odd complete response thrown in) response in 20% of men, then a Phase 3 study would only need 100-150 men in each of a test and control arm. We could knock that over in 18 months.

A Phase III trial will probably compare palliative radiation to palliative radiation plus NOX66.  The achievement of abscopal responses under trial conditions in DARRT-1 increases the likelihood that NOX66 will achieve statistical superiority in a registration trial.