BOSTON (TheStreet) -- French investors holding shares of Genfit (GNFTF) are having trouble coping with the failure of the company's fatty liver disease drug GFT505. I know this because many of them (it feels like all of them!) have reached out to me via Twitter (TWTR) - Get Report in the past few days to express their outrage over my bearish analysis of last Friday's GFT505 study results in NASH.
This is the first time I've been Twitter trolled in French. I like it. Much classier.
French pride appears at stake here, as well.
My responses follow.
1. The primary endpoint of Genfit's study was NASH clearance without worsening of liver fibrosis. To be deemed cleared of NASH in the Genfit study, a patient needed to score a zero, meaning elimination, in at least one of the three components of the NAFLD Activity Score (NAS) -- steatosis (fat accumulation), inflammation or ballooning. If a patient showed no improvement in two NAS components but scored a zero in the third, that was good enough to be counted as a responder. Genfit defined no worsening of liver fibrosis as no progression to advanced fibrosis.
Genfit's definition of NASH clearance should have been relatively easy to achieve, yet GFT505 failed to do so. Only after Genfit threw out the placebo patients with baseline NAS scores of 3 and made other statistical adjustments did GFT505 "work" -- NASH clearance reported in 22.4% of GFT505 patients (NAS of 4 or greater), compared to 12.7% for placebo patients.
Intercept included NASH resolution as a secondary endpoint in the "FLINT" study of OCA. NASH resolution was defined as no NASH, meaning NAS scores of zero across all three components. This is a much stricter, harder-to-achieve definition compared to what Genfit used. As reported by Intercept, 22% of OCA patients achieved NASH clearance compared to 13% for placebo. The result trended in OCA's favor but was not statistically significant.
Yes, Intercept's NASH resolution data for OCA are better than Genfit's NASH clearance data for GFT505. No question about it, particularly because OCA achieved the primary endpoint of the FLINT study -- a two point or more reduction in NAS score with no worsening of liver fibrosis -- with high statistical significance. The study was stopped early because of OCA's efficacy. The detailed presentation of the FLINT study data last year showed significant improvements attributed to OCA across all three of the histological components of the NAS score. Genfit has yet to disclose comparable data for GFT505.
2. Genfit can't remove patients from a study analysis and claim to achieve the primary endpoint. Genfit included patients with NAS 3 baseline scores because the company believed this would make it easier for GFT505 to succeed. The strategy backfired, so it's disingenuous at best for Genfit to now blame NAS 3 patients for the study's failure.
Intercept's FLINT study also enrolled some patients with baseline NAS scores of 3. The company has also disclosed that about 20% of patients did not have a definitive diagnosis of NASH at baseline. Still, this didn't prevent OCA from achieving the primary endpoint of the study. Only Genfit had that problem.
3. Thanks for reminding us about FDA granting Breakthrough Therapy Designation to Intercept's OCA for NASH based on the stellar data from the FLINT study. Genfit has earned no comparable designation for GFT505.
The target population for any NASH drug will be patients with more advanced disease (NAS scores of 4 and higher) and liver fibrosis. NASH gets serious when fibrosis develops because it's one step closer to cirrhosis and liver transplant. Intercept, smartly, is targeting more advanced NASH patients because this is where OCA will have the greatest benefit.
4. OCA demonstrated a statistically significant improvement in fibrosis -- 35% vs. 19% placebo. I don't know if Intercept presented liver fibrosis data specifically excluding NAS 3 baseline patients, but let's assume these patients have no or relatively little fibrosis. Intercept does have fibrosis data separated by baseline stage. For F1 (early fibrosis) patients, there was an improvement trend favoring OCA over placebo (35% vs. 14%; not statistically significant.) For F2 and F3 baseline patients, the improvement in liver fibrosis was 42% for OCA vs. 25% for placebo -- statistically significant.
Let's stop here for a moment to remember Genfit's Friday announcement which stated that GFT505 demonstrated no improvement in liver fibrosis compared to placebo. None. It's true that Genfit ran a one-year study of GFT505 which may not be long enough to show significant improvements in liver fibrosis, but 12 months is certainly long enough to show some positive trend. Genfit did not, which is enough evidence to conclude that GFT505 is an inferior NASH drug compared to Intercept's OCA.
More tweets from French fans of Genfit:
Translation (using Twitter's native translation tool): @adamfeuerstein just do Ur job! Do your job properly and you can shave you the morning proud of you. There are job anyway!
Translation: @adamfeuerstein , Genfit, icpt j ' eagerly await the sequel...that will make the fda ac icpt following safety problems!
Then you missed this story I wrote last May: "Intercept Pharma, Government Scientists Spar Over Negative Safety of Liver Drug, Emails Show."
No one is saying OCA is the perfect NASH drug. The drug's negative impact on lipids in the FLINT study has been well documented. But GFT505's safety profile, given that it's a PPAR agonist, is not entirely clean, either.
Translation: @adamfeuerstein @Sport234a just be honest and professional. Insane UR, U R a lier! that's Ur Karma! Who sows the wind reaps the storm.
Translation: @Ji_Bus @adamfeuerstein Adam you're still in the forest!
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.