Biomarker-based adaptive trials for patients with glioblastoma—lessons from I-SPY 2 One of the most important articles of the year. I have been saying this for years but this is the first time I saw it in print. They propose a new method of clinical trials where many different treatments can be tested against each other in a rational way. As they evaluate results, the trial adapts. If they see that certain biomarkers influence outcome, they can change the randomization based on that biomarker to increase the chances of good results. For example, if a drug only seems to help people with a specific target, they stop using it on patients who do not have that target - they can still get randomized to the other treatments.
If a treatment doesn't perform well, they drop it and rotate in the next one. If a treatment looks good, they drop it out of this trial and launch a traditional trial of it.
This way they can test a large number of treatments, identify biomarkers that may be used to personalize treatments, and significantly reduce the amount of time (and money) needed to find the cure. It also improves the chances that the trial will help the participants - as once it is clear a treatment doesn't work - or if it doesn't work with that patient's markers, it is dropped.
Hope in sight in war on cancer These researchers can now grow DIPG cells in the test tube, and they tested available drugs and may have found one which may have a chance of working. Human trials should start soon
