Our team is routinely building analytics dashboards for clients, which means that sometimes we find numbers that defy expectations. When this happens, we like to dig in and write about it. In this case, the number that has caught our eye is the decrease in newly registered clinical trials on ClinicalTrials.gov from 2014 to 2015.

The number of new studies registered at ClinicalTrials.gov (CT.g) decreased by 2.4% year over year, from 22,916 in 2014 to 22,350.  However, this is still an increase of 11.9% in studies registered in 2013 with a total of 19,969:

clinical studies by submission date

But actually, when we dug further we found that there was a 2.8% increase in academically sponsored trials in 2015 (16,140 in 2014 to 16,593 in 2015). The decrease in new clinical trials is taking place not in academia, but in industry and NIH sponsored studies. 

clinical studies by sponsor

Why is this decrease a signal of the health of the biopharmaceutical industry?  From CT.g we can further drill into the types of studies being started, and the biggest decrease is seen in Phase 1 studies which have decreased by 668 (24%) from 2,773 studies in 2014 to 2,105 in 2015.  Phase 3 studies have shown a mild decrease of 10 (0.5%) studies from 1,822 to 1,812 from 2014 to 2015 respectively.  There are a few studies designated Phase as Phase 0 or I/2 etc which we have not considered, but are shown below:

clinical studies by phase

While we should be concerned with the number of Phase 1 studies being registered on CT.g, this is not a good snap shot of the industry as there are so many other factors in play:

  1. Phase I studies do not have to be registered in CT.g
  2. Companies register Phase I studies so they can be published, since without the NCT number, journals will not accept the paper for publication
  3. Phase I studies are becoming much larger with multiple arms.  Historically Phase I studies were 12 to 20 subjects and each company would run several Phase 1 trials.  The trend is going to larger studies with many adaptive dose increases.  An example of this is the recent concerning study in the news by Biot, with 128 subjects due to be enrolled.  (Only 90 were enrolled before there was an SAE and at least one subject [volunteer] died of brain damage).
  4. There is some cyclic activity as 2013 had a very similar number of Phase 1 study starts (1,993) as in 2015 (2,105)

Finally, we have not considered all the other industry sponsored studies which are not given a phase and are conducted by device and software companies.  Here the numbers have remained consistent from 2014 (12249) to 2015 (12263) and an increase (14.9%) since 2013 of 10,683 studies.

We will continue to monitor these clinical trial starts, and see if this trend continues or is just a cyclical event we have observed.