According to recent press reports and a regulatory filing by Theravance, the commercialization and manufacturing agreement between Astellas and Theravance has been terminated. Theravance developed Vibativ or telavancin through phase III for two indications – complicated skin infections and hospital-acquired pneumonia. Telavancin is an intravenous drug for the treatment of severe infections caused by Gram-positive bacteria like staph. Both sets of trials indicated that telavancin was non-inferior to vancomycin in these indications. But telavancin was associated with increased toxicity risk and, for the treatment of skin infections, a risk management strategy was required in the US and in Europe because of the risk to pregnant women and to those with pre-existing renal insufficiency.
In the middle of the development of telavancin, the FDA changed its mind about what an appropriate endpoint should be for hospital-acquired pneumonia (HAP). Until that point it had been clinical outcome – cure or failure. But suddenly it was to be mortality. Theravance’s trials were well under way if not completely enrolled at that point and were not designed (no trial can be designed like that anyway) to assess mortality. The FDA rejected telavancin for HAP on this basis. Europe, where clinical outcome is still considered a valid endpoint, accepted the telavancin marketing application and approved the drug for HAP, but only for those cases where another alternative was not suitable.
Telavancin’s sales have been dismal at about $4.5 million for the 3rd quarter 2011. One could argue that this is not surprising since it is really, just another IV only Gram-positive only antibiotic that failed to study or prove superiority to any of the other drugs out there. The EU approval for HAP also just arrived in mid-2011. Further, the trials compared telavancin to vancomycin whereas many feel that, certainly for HAP, linezolid is a superior drug. Given the risks of telavancin, it is obviously hard to make the argument that physicians should adopt this new therapy. One wonders what Astellas was thinking when they made the deal to begin with. Although, Astellas clearly left themselves an out. They apparently were not responsible for any of the development costs, and only paid $65 million upfront in 2005. Additional milestone payments were also part of the agreement going up to an additional $156 million. Astellas was responsible for ex-US regulatory filings and for all commercialization with Theravance getting a 20% royalty.
Apparently the relationship between the two companies was not always smooth. Recent manufacturing problems have led to drug supply shortages and this has also slowed sales. If they will have to change manufacturers, this could lead to a very significant delay of 1-2 years in drug supply and sales.
Theravance says it will now pursue other options including identifying another partner or partners. But it is difficult to see another partner signing on under these conditions.
At the same time, Theravance has a much more interesting product in its pipeline (if one is interested in IV only, Gram positive only compounds), TD-1792. Like telavancin, it is related to vancomycin. But it is much more potent than either telavancin or vancomycin. It now seems unlikely that this product will ever be developed either.
One might almost think that Theravance’s effort in antibiotics was cursed. First, there was the increased toxicity risk of its lead candidate where superior efficacy was not and could not be proven. Then, the FDA changes course in mid-stream, putting a key indication out of reach for Theravance. Then a severe manufacturing problem arises. Finally, a key commercialization partner caves under all the pressure.
Are there implications from the Theravance debacle for other antibiotic developers? I think there are several. IV only Gram-postive only drug candidates should be approached with caution since the market is already crowded and superiority will be impossible (probably) to show. If one embarks on such a project, comparison to a drug that has shown superiority to vancomycin, like linezolid, is advisable. Manufacturing these large, natural-product-based molecules is not trivial – although in the case of Theravance, this appears to be a problem of a particular manufacturer and not necessarily an indictment of their manufacturing process. Finally, cautious partners remain cautious.
Mortality? Why? It seems like a step backwards for the FDA.
ReplyDeleteBecause that is all they could identify in the historical database to justify an NI margin. Of course, there are other methods like pharmacometrics they could use to identify a treatment effect for clinical outcome - but they have not gotten there yet . . .
ReplyDelete