LILLE, France (TheStreet) -- Torture a prisoner long enough and he'll tell you what you want to hear. Clinical trials are similarly vulnerable. Just ask Genfit (GNFTF), which put its fatty liver drug candidate GFT505 through the statistical equivalent of waterboarding and sleep deprivation to extract the desired result from a mid-stage study: The drug works!

But GFT505 achieved the primary endpoint of the study -- clearance of nonalcoholic steatohepatitis (NASH) -- only after excluding placebo patients who responded really well. Even disappearing patients from the study wasn't enough. Genfit also had to change the statistical analysis plan after the study was completed. Genfit acknowledges all this in its press release.

The GFT505 study in NASH was a failure, analysts said. That's bad for Genfit and good for Intercept Pharmaceuticals (ICPT - Get Report), which is developing a rival drug known as OCA to treat NASH patients.

NASH is commonly referred to as fatty liver disease because patients literally have livers which are inflamed and enlarged due to the accumulation of fat. In more advanced NASH patients, the liver becomes fibrotic, which can lead to cirrhosis.

Demonstrating a benefit for NASH patients is a big challenge. Intercept's drug OCA was able to achieve this goal convincingly. The mid-stage study of OCA in NASH was stopped early because the drug's ability to resolve NASH symptoms and improve liver fibrosis was overwhelmingly positive and highly statistically significant. Subsequently, the FDA awarded OCA Breakthrough Therapy Designation status. Intercept is now preparing a large phase III study of OCA in NASH.

Genfit's similar mid-stage study of GFT505 is a mess. During a Thursday afternoon conference call, company executives refused to divulge the unadulterated results from the study. Instead, investors and analysts were offered a list of lame excuses and heavily edited data, which, even if you believed it, showed GFT505 to have a modest effect on NASH clearance and no positive benefit on liver fibrosis. Despite a failed phase II study, Genfit intends to advance GFT505 into a phase III study.

Torture rarely pays off in the end.

Adam Feuerstein writes regularly for TheStreet. In keeping with company editorial policy, he doesn't own or short individual stocks, although he owns stock in TheStreet. He also doesn't invest in hedge funds or other private investment partnerships. Feuerstein appreciates your feedback; click here to send him an email.