(Bloomberg) -- When you roll the dice with the FDA, it's generally a good idea to stack the odds in your favor.
Massachusetts-based Chiasma's shares fell as much as 62 percent on Monday, the most since its IPO last year, after its lead drug for acromegaly Mycapssa failed to get FDA approval Friday afternoon. The agency said in a letter it didn't think the company's application provided "substantial evidence of efficacy to warrant approval," according to Chiasma. The agency had concerns about the company's trial design and wants a new one. It "strongly recommended" the new study be a randomized, controlled, double-blind trial, which is more statistically robust -- and likely more expensive.
The news, and the market's violent recoil, highlights a fine line small biotechs sometimes walk. Small firms working to get a drug approved hope to do so in the leanest and quickest way possible. That's understandable -- these firms often have no revenue and rely on capital markets for funding. But this approach can result in trials that don't always measure up to the FDA's statistical ideal.
Now Chiasma won't see revenue any time soon from a drug it hoped might get approved Friday. Investors will now scrutinize its limited cash supply at a time when raising more money is problematic. And it faces the prospect of a pricey new trial to get back on the approval docket.
The data the FDA found lacking came from what's known as an "open-label, single-arm" trial. That's when all of the patients in the study get an experimental treatment, with no comparison against another treatment or placebo. The drug's effects are measured by some other benchmark -- in this case, a hormone level -- and the results can be harder to interpret. In a double blind, randomized study -- the scientific and statistical gold standard -- the treatment is compared to a control arm, and no one knows who is taking what through the trial.
Single-arm studies are less common in the Phase 3 trials companies use to seek FDA approval. Drugs do get approved that way, particularly rare-disease drugs and some cancer medications. Roche and AbbVie's blood-cancer drug Venclexta got accelerated approval with such a study on April 11.
But the FDA wants a good reason for such a trial, and strong evidence of efficacy to overcome its ingrained statistical caution. Telesta Therapeutics also tried and failed to get approval for its bladder-cancer drug with a single-arm trial in February, with the agency requesting an additional phase 3 trial.
The FDA had warned Chiasma in December 2014 that its study design could be a problem, the company noted in its 10-K filing. But Chiasma may have underestimated the seriousness of the agency's concerns.
As for Chiasma's path forward, it's all about cash. The company announced on a conference call Monday that it had approximately $134 million in cash at the end of the first quarter. That's above average for firms on the Nasdaq Biotech Index with less than $2 billion in market cap. But Chiasma may have to make that cash stretch far longer than it expected. Before the FDA's rejection, Chiasma had expected the cash to last through at least mid-2017. CEO Mark Leuchtenberger said on the call the company is now "revisiting all priorities with an eye to extending this cash runway."
Raising more money isn't easy in the current biotech market; the NBI is down 13.4 percent so far this year. Selling shares is a mite tougher after you get a rejection from the FDA and your share price is cut in half. The company recently started a controlled, randomized trial seeking European approval. It will likely need yet another one in the U.S. in order to satisfy the FDA. That will cost in money and years.
It's not just a major setback for Chiasma, but a warning sign the FDA is going to whip out its extra-large magnifying glass when companies don't have control data.
This column does not necessarily reflect the opinion of Bloomberg LP and its owners.
To contact the author of this story: Max Nisen in New York at email@example.com.
To contact the editor responsible for this story: Mark Gongloff at firstname.lastname@example.org.
©2016 Bloomberg L.P.