From the GRACE Archives - Originally Published April 11, 2009 | By Dr Pennell 

This week in the Journal of Clinical Oncology, a group of clinical researchers and specialists in management science took it upon themselves to study and report, in depth, the processes that the National Cancer Institute (NCI) and its developmental clinical trials branch, the Cancer Therapy Evaluation Program (CTEP), use to design and approve new clinical trials for promising anti-cancer agents. There has been a general consensus for years that new clinical trials were taking way too long to move from concept to reality, and they wanted to look at the possible reasons behind it.

As you probably know, clinical trials are research studies that prospectively enroll patients with a particular disease state (for example, metastatic NSCLC) to an experimental treatment that is hoped to be better than the standard treatment. The most important trials are the phase III “randomized” trials, in which patients are randomly assigned to either the new treatment or the current standard treatment. This is done to balance out any random elements of the patients so that both groups are relatively evenly matched for things like age, sex, performance status, etc. This type of trial is what was used to bring drugs like Avastin (bevacizumab) into common use.

For example, say drug X shows great promise in an early phase trials, and I want to design a phase III trial to bring this great new drug to approval by the Food and Drug Administration (FDA). I write to CTEP with an idea to test this drug, in combination with, say, Tarceva (erlotinib) in a phase III trial compared to Tarceva alone in NSCLC patients. I send in the concept, and CTEP then initiates its review process to determine if this is something they want to fund.

Here is the general outline of the process. It begins with review of the concept, and if the concept is approved it moves to the protocol development group. Once the protocol is finished, it goes to an institutional review board (IRB; whose primary function is patient safety) as well as some other administrative reviews, before moving to activation.

Sounds straightforward, doesn’t it? Ha!

What the study group discovered was that for every step of this simple process, there were a dozen or more extra steps that needed to be passed, and if you failed any of them, you had to start all over again! Here is an illustration of the steps needed just to get the concept to a formal initial review (from the first blue box above to the second):

I know this is hard to read, but it illustrates the point: this is a ridiculously bloated and overly-cumbersome process. In fact, they found that there are 296 distinct processes required for new phase III trial activation: at least 239 working steps, 52 major decision points, 20 processing loops, and 11 stopping points.

The result? The average time from submitting a concept to enrolling the first patient was 602 days, or nearly 2 years! Some trials took 4-5 years to activate. This is actually better than a prior study of phase III trials activated through the Cancer and Leukemia Group B (CALGB; one of the large, NCI sponsored cooperative groups) in this decade, which required 370 distinct processes and took an average of 784 days!

While 160,000 lung cancer patients are dying each year in the United States alone, this is completely unacceptable. And it is not much better for small trials at individual institutions. It is not at all uncommon for a small phase II trial, designed by an unnamed intrepid clinical lung cancer researcher, to take a year or more from concept to opening of the trial. My own institution has 90+ processes involved in opening a trial, which I have to think is a bit excessive but seems relatively simple compared to what is shown above!

The study points out that there is no one particular step that is leading to the delay, making fixing this problem problematic. Part of the problem is the sheer number of people who have input into these trials. The IRB is a necessary protection for all new clinical trials, but why do there need to be so many regulatory steps? Certainly we could all take lessons from the pharmaceutical industry, which is able to get phase III trials up and running in a much shorter period of time when patent life and revenues are on the line!

In an accompanying editorial, David Steersma quotes John Godfrey Saxe: “Laws, like sausages, cease to inspire respect in proportion as we know how they are made”. I would hope that the recent spate of prominent papers exposing the ridiculously overcomplicated phase III trial initiation prcedures will somehow bring about reform and streamlining of the process. So far, it seems to be the problem everyone recognizes but no one is willing to tackle.

Comments are closed

• 

  Dr. West says:

  April 11, 2009 at 2:51 pm

  Dr. Pennell,

  I think this was a great topic to write about, as I think we really can’t underestimate how much our painfully inefficient regulatory process slows the process of conducting clinical trials. This research is the key to moving the field forward and getting new treatments available to people.

  This hyper-regulated process was developed out of concern for protecting patients, and there have been enough lapses of ideal management in clinical trials to lead to a need for good regulation. But a process that is paralyzed by the inefficiencies of sending us back to the drawing board over and over while patients are desperate for new treatments is just unacceptable.

  I believe this is a difficult problem but one that the medical community as well as the patient advocates and even political figures need to identify as a process that paralyzes our progress in cancer and desperately needs to be addressed.

  -Dr. West

•  

  Catharine says:

  April 14, 2009 at 12:25 pm

  Dr. Pennell -

  That was eye-opening! I wonder if there’s a legislator in Congress who would take this on as a personal cause?

  Thank you for sharing a summary of the study.

  -Catharine

• 

  Dr Pennell says:

  April 14, 2009 at 4:00 pm

  The problem is that the bureaucracy behind clinical trials has grown haphazardly, like accretions of barnacles on a pier (or the tax code). No one would have designed it this way if they had a choice, instead each additonal stakeholder just added something to make their part seem more important. And don’t worry, most of the steps have nothing to do with patient safety (like the very necessary IRB review).

  The way to fix this would be to have independent evaluators analyze the process to determine what was truly important and what was redundant or unnecessary. Then scrap everything and start from scratch, streamlining the process. Next, track the time it takes to get from step to step, and constantly make changes needed to keep things moving while keeping patient safety a priority.

•  

  phylsgirl says:

  April 14, 2009 at 4:46 pm

  Thank you Dr. Pennell for this post. This is a topic of interest since clinical trials may become part of a patient’s treatment plan. But I have one questions. The example given for the approval process was Phase III. Am I correct the aproval process is “simplier” for Phase I and Phase II? But I guess the bigger question is what can be done to simplify the process, without sidestepping the safety issues of coure so clinical trials and treatment can be made available quicker?

  Also I have to applaud your wit in titling this post, ‘duh’, indeed. Great topic, thanks.

  Namaste,
  Ree