Thread by @VincentRK

So, you want to lead a randomized trial?

I ran a poll: How many people have near veto power over the approval & design of an investigator initiated oncology randomized trial? Only 23% guessed right.
Answer: >50! @RielyMD @mtmdphd

Here are the steps to do an RCT? Thread 1/

2/ Almost all investigator initiated randomized phase III trials in oncology in the US have to be done in the cooperative group setting. Once you have a brilliant idea you must first convince the disease committee of your concept. At least 1-2 ppl have veto power at this stage.

3/ Next it goes to the Executive Committee of the group eg., @eaonc @SWOG @ALLIANCE_org

Here they have different priorities including whether there are resources relative to other cancers that are also competing for trials etc. At least 5 ppl effectively have veto power here

4/ Next it goes to the disease specific Steering Committee of @theNCI @NCICTEP_ClinRes

~30 people (specialists in the disease from around the country, statisticians, administrators etc) have effective veto. All it takes is 1-2 strong voices & your whole idea can be shredded

5/ Next (if you are extremely lucky so far) CTEP needs to approve. Then you will be tasked with writing a revised concept that goes through the whole process again till it is fully approved. Very very few concepts are approved as is. It has never happened to me in 20 years.

6/ Next if (and it’s a big IF) your concept is approved, you will have to write the full protocol document and that needs approval by @theNCI @NCICTEP_ClinRes. Usually needs revisions. This can go back and forth for a while. But you are not done yet. Many more need to approve!

7/ Unless you are using commercial drug paid by insurance, and need no funding, and have no lab correlatives: many additional people have to approve. This is pretty much all trials. The main source for drug & funding is Industry. At least 3-5 ppl have veto power for this step.

8/ The trial now has to be approved by the FDA unless it is IND exempt. At least 1-2 ppl have veto power here to make changes to design or treatment.

9/ Next the Central IRB must approve. At least a few people have veto power here to change the design or require changes.

After all this you hope that the trial is embraced by the community and accrues. But I have seen trials close down despite all this work due to poor accrual

Many times the trial that opens is quite different than what the principal investigator who had the first idea wanted to do. But faced with disapproval versus compromise, many including me choose compromise because not doing a trial is a worse option for informing practice

Many times the recommendations of one group with veto power is in direct contradiction to another group with veto power. Which means the trial doesn’t is in stalemate or doesn’t happen after 1-2 years of hard work. This is not theoretical. It’s happenning right now!

All of this sounds pretty depressing and can put off people from trying to do randomized trials. That’s not my goal here. It’s for people to understand and appreciate how difficult the process is. Unless you try, you have no idea how the process works.

Second, it’s worth it and rewarding. Sometimes the changes suggested are very valuable and do lead to a better trial. Third, we need changes to the process.

We need a reformed system where a few people who are genuine experts, who have skin in the game have veto power. People who know not only what the ideal trial is, but also what’s feasibile, and understand the urgency. Cancer patients need this. The rest can suggest but not veto

One more important thing to add: In these processes, Pharma is usually the easiest to work with. We can get a quick informal verbal yes/no answer — sometimes even before we right the formal concept. Then we work the details with 2-3 key people who are usually willing to negotiate