Third time is the charm for InterMune (ITMN.DL). Of course, so was the first.

It was the middle trial that ended up delaying a U.S. approval for its lung drug Esbriet by five years.

Haven't been following biotech for that long? Let's get you caught up.

  • In 2009, the biotech said Esbriet helped patients in both trials, but only one of the two phase 3 trials in a lung disease called idiopathic pulmonary fibrosis demonstrated a statistically significant difference between Esbriet and placebo.
  • An FDA advisory committee voted 9-3 recommending approval of the drug.
  • The FDA ignored its outside advisors and rejected the drug.
  • EU regulators were more forgiving of the mixed data and approved Esbriet.

Source: InterMune.

InterMune went back to the drawing board, designing a new clinical trial, dubbed ASCEND, with more patients than the first two trials to improve the chances of showing statistical significance. The company also measured efficacy at 52 weeks rather than 72 weeks in the previous trials, because the trials showed a treatment effect at 48 weeks, but then the placebo group didn't decline as fast as the Esbriet group so, by 72 weeks, the difference wasn't statistically significant. InterMune also tweaked the primary endpoint; instead of comparing the relative decline in lung capacity -- called percent predicted forced vital capacity, or FVC -- the new trial measured how many patients saw a 10% decline in FVC, which is clinically meaningful.

The changes worked. Big time.

After a year on the drug, 16.5% of patients experienced an FVC decline of 10% or more (patients that died were also included) compared to 31.8% of patients that took placebo. The difference had a p-value of less than 0.000001. Or in other words, there's a 0.0001% chance that the difference happened by chance alone.

Yeah, I'd call that statistically significant.

Ironically, the relative change in FVC was statistically significant as well, so InterMune didn't really have to change the endpoint. Of course, maybe that shouldn't be too surprising since Esbriet did pass one of the two trials with that endpoint.

InterMune plans to submit its new application to the FDA in the third quarter. As far as I can tell, the data looks good enough to get approved by the FDA this time around.

Was the FDA wrong?
Clearly.

In hindsight, at least.

The main argument at the advisory committee meeting seemed to be that there weren't any approved treatments for IPF, so why not just approve it despite the mixed data? As much as investors and companies would like, that's generally not how the FDA works.

While I hate defending the FDA's overly cautious stance, I think the agency was right to be careful with Esbriet considering how hard it's been to treat IPF. In the intervening five years since Esbriet's last trials, a trial for Gilead Sciences' (GILD 0.09%) ambrisentan was stopped because of lack of efficacy; Actelion Pharmaceuticals' Tracleer failed to show an effect in IPF, and so did Novartis' (NVS 2.25%) Gleevec. 

There was a report from an unnamed source that Boehringer Ingelheim's nintedanib has passed a phase 3 trial, but the company hasn't confirmed that, as far as I know. Both companies should report data at the American Thoracic Society in May, so we'll know soon enough who might win in the marketplace, assuming they're both approved.