Skip to Main Content

After years of declines, the pharmaceutical industry is experiencing a greater rate of success with its clinical trials in recent years, according to a new analysis.

Between 2012 and 2014, more than 11 percent of clinical trials succeeded, which meant compounds being tested survived the arduous journey from the laboratory to the pharmacy counter. This reversed a downward trend seen over the past 20 years, according to executives at McKinsey & Co., the consulting firm that conducted the analysis and does consulting work for drug makers.

Between 1996 and 1999, the cumulative success rate was 16.4 percent, but gradually began declining in subsequent years. From 2000 to 2003, 10.8 percent of trials succeeded before falling to 10 percent between 2004 and 2007, and then bottoming out at just 7.5 percent between 2008 and 2011.


“In some ways, this is rather remarkable,” said Martin Moller, one of the coauthors, who heads McKinsey pharma R&D consulting in Europe, the Mideast, and Asia. “We’ve been used to seeing a decline. This is a turning of the trend. Whether that is sustainable is an open question.”

The analysis, which was published last month in Nature Reviews Drug Discovery, examined more than 9,200 novel compounds that were developed from 1996 through 2014.


The declines in past years likely reflected industry cutbacks, according to the McKinsey consultants.

From 2007 through 2010, many drug makers were cutting back on R&D amid large mergers and reorganizations that resulted in what was described as “pipeline pruning.” Basically, most of the largest companies turned their backs on researching medicines for combating certain diseases and became much more selective about how they invested their research dollars.

For this reason, the McKinsey team suggested the more judicious use of resources resulted in higher quality product pipelines which, in turn, could account for the improved showings in clinical trials. The larger proportion of failing compounds in Phase 1, or early-stage trials, may indicate that companies are running more thorough evaluations and doing so sooner.

“Basically, the industry is learning how to fail earlier,” said Moller. “And that’s a good thing.”

This could be seen by looking at data for both Phase 2 and Phase 3, or mid- and late-stage, studies. From 2012 through 2014, the likelihood that a Phase 3 study would advance to product registration was 64 percent, up from 60 percent during the previous three-year period. Similarly, the odds of a Phase 2 study moving to the next round was nearly 40 percent, up from about 30 percent during 2007 and 2010.

There was a caveat, though. The increasing success rates for midstage trials may be slightly inflated not only because there are more life-saving medicines being shepherded to later-stage trials thanks to new trial designs, but also because of breakthrough regulatory mechanisms. “Such an approach carries a risk of a later-stage failure,” the consultants wrote.

For instance, drugs that were developed for treating rare diseases, which can benefit from speedier development paths, had a higher overall success rate from Phase 2 to regulatory approval. During the past three years, this amounted to 29 percent compared with 10 percent for drugs used to treat other maladies. In Phase 3 trials, the success rates were 73 percent and 64 percent, respectively.

In any event, what remains unclear is whether the ongoing rebound in clinical trial successes can continue, given the variables and uncertainty surrounding drug development.

“It’s going to take a little time before we come back to [the] level of confidence around [what we saw in] this area in the late 1990s and early 2000s,” said Moller. “Everyone still sees pharma R&D as a precarious game … Even though the aggregate success rates now have turned, it’s still true that there will be enormous variations among individual companies.”

  • Thanks for this “top line” awareness. For those of us in R&D/Clinical Operations, more details would be interesting. Such as:
    Who is doing better: CRO or Sponsor run trials?
    Which therapeutic areas are seeing the success and which are not?
    What parts of the world are doing better or is everyone in all countries showing the same improvement?
    How does the success relate to the amount Sponsors are paying Investigators as the per-patient-grant?
    Are the studies succeeding because the compounds are better or because the studies have been able to meet the enrollment goals?

    • Excellent points. Another way to increase the odds of success is to stay away from therapeutic classes that have a high failure rate in Phase 3 such as psychotropic drugs. Not one was approved in 2015, and this was at one time the leading class of prescription drugs.

  • I’d be curious to see how the use of cloud and other modern technologies in clinical trials might have impacted (and continue to impact) the increased success in clinical trials. In 2012, when success rates started rising, cloud-based electronic systems were being adopted to help improve efficiency, control, and compliance throughout studies.

    • Lisa, all of the new technologies can improve clinical trial execution and maybe expose a poor candidate earlier in development. About 30 years ago Dr. Leigh Thompson of Eli Lilly applied the term “fail fast” to clinical development. In the end it comes down to protocol development. It is a truism that an excellent protocol can make an average drug succeed, and a poorly designed protocol can make an excellent drug fail. Maybe some lessons are being learned.

  • Sorry folks but Merck is still the gold standard. When Roy Vagelos ran the company and Ed Scolnick ran R&D the success rate for phase 3 trials was 95% because extended Phase 2 trials were such that no compound advanced to Phase 3 unless there was a 95% probability of success.

Comments are closed.