beauregard · @beauregard2
4th Nov 2019 from TwitLonger
Okay, here's a quick reality check on that very reality-challenged short report on $FGEN. I'll try to keep this as simple as possible, but to do so I need to first provide a little background.
Prior to the massive (10k patients) phase-3 studies that have recently (2019) been pooled for safety analysis, FGEN or its partners ($AZN and Astellas) conducted smaller phase-2 and phase-3s in China and Japan.
All of the Chinese and Japanese studies were open-label (including one that became open-label after just 8 weeks). It has been reported that doctors pushed some patients to switch to standard of care (ESAs), producing the mild discontinuation imbalance, which the short report so thoroughly exploited for to sow FUD (fear, uncertainty, and doubt).
Key point #1: There is no evidence whatsoever that Roxadustat patients discontinued at a higher rate in the massive, pivotal closed-label, Phase 3s. In fact, the companies specifically and repeatedly emphasized that there was a MUCH HIGHER DISCONTINUATION RATE ON PLACEBO. In fact, the discontinuation rate was so much higher on placebo that the companies had to hold their tongues on making claims of non-inferiority until the FDA had suggested a statistical approach to accounting for such higher placebo dropouts. This is the OPPOSITE of the argument made in the short report.
Key point #2: The short report repeatedly refers to higher Roxa discontinuations without mentioning that these are from the OPEN-LABEL Chinese and Japanese trials only (did not apply to the small US phase 2s). It repeatedly attempts to create the impression that the discontinuations were found in the big pivotal P3s. In many sentences, the obfuscation is so deliberate as to be legally actionable.
Key point #4: A key advantage that Roxa has over Standard of Care is that Roxa does not require iron supplementation. This is game changing and the advantages are obvious to doctors and insurance companies alike. However, the short report desperately spins this advantage, by saying things like the companies "manipulated" the results by "banning" iron. In reality, the makers of ESAs would give anything to find that their drugs also worked without iron.
Key point #5: The short report ignores all recent statements made by the companies and instead cites statements made when the companies were under a self-imposed gag order not to discuss statistics until an agreement with the FDA had been reached on endpoint analysis. In July, the companies met with the FDA and reached those agreements.
Key point #6: Prior to the July meeting with the FDA, the companies withheld conclusive statements on MACE (the preferred FDA endpoint) but not MACE+ (the preferred EMA endpoint, where a statistical plan was already in place). This was a temporary act of caution. It was not an attempt to substitute MACE+ for MACE as has been falsely insinuated by the short report.
Key point #7: Now that gag order has been lifted both AZN and FGEN have confirmed that they have a statistical plan for MACE (not MACE+) with the FDA. The CEOs of BOTH companies have affirmed that they have "very high confidence" that the phase3 data "CONFIRMS CARDIOVASCULAR SAFETY IN BOTH DIALYSIS DEPENDENT AND NON-DIALYSIS DEPENDENT." They made these statements in full possession of the data. They will release that data this week at a major conference.
Key point #8: They haven't merely stated that the phase 3 data confirms cardiovascular safety. Both companies have confirmed the statement with their wallets: They have already initiated trials to prove the drug in chemotherapy-induced anemia, an investment that would be worthless if the data proved the drug was unsafe. In addition, they have announced several extra sessions (dinners for KOLs and analysts) to trumpet the results after their official release at the medical conference--not typical behavior of companies in possession of bad data.
Key point #10: The Short report claims that Roxadustat suffered from higher rescue rates. In fact, that statement came from one small phase-2 and totally ignores the fact that in large phase 3 trials ROXADUSTAT MASSIVELY REDUCED RESCUE RATES. In fact, in the massive ANDES trial, Roxadustat reduced rescue therapy by an astonishing 81%. Of course, the short analyst doesn't want that known.
Key point #11: Roxadustat has additional beneficial effects besides treating anemia. To wit, it actually helps MAINTAINS KIDNEY FUNCTION. ESAs don't do that. It also REDUCES LDL CHOLESTEROL. ESAs don't do that either. The short report ignores this, while declaring vaguely that HIF-inhibitors (such as Roxadustat) have (mysterious, unnamed) negative outcomes.
In sum, the short report cherry picks from small open-label foreign trials while completely ignoring contrary evidence from larger well-controlled randomized phase 3 trials.