3 ms·

Avexis, the company that Novartis acquired to get this drug, spent ~$700M developing it. The drug will treat a few hundred patients a year. Without a high price

Avexis, the company that Novartis acquired to get this drug, spent ~$700M developing it. The drug will treat a few hundred patients a year. Without a high price, it isn't profitable. for these drugs to get covered at such a high price, they have to be very effective, which this drug seems to be However, did Novartis have to charge $2M? Avexis was trading at a $3B market cap not long before Novartis bought it for $8.7B. Analysts were projecting a $1-1.25M price tag, which is closer to the cost effective range Novartis probably paid up for avexis bc it was a competitive deal process and they thought they could get away with a higher drug price than the market expected. At the time of the acquisition people thought Novartis was paying a premium for the platform, by charging $2M for the drug it seems like more of a pricing play than platform deal I wrote a case study on avexis with more detail here: https://www.baybridgebio.com/blog/most_expensive_drug.html https://www.baybridgebio.com/blog/most_expensive_drug.html

JamesBarney 7y ago
I've always been curious how that 700M breaks up? How much of that is FDA trials and how much is paying 1000 phds to work on designing drugs for 7 years. Of course the follow up question is there any way to get that number lower.
- aaavl2821 7y ago
  It seems very high to me as well, especially since they were allowed to submit for approval with only one 15-patient study completed and interim data from a few other studies. I think they have tested the drug in under 100 pts to date Per the financial statements, ~$500M of that $700M was r&d (the rest was general and administrative expense). A lot of the r&d spend was payments to third party research and manufacturing firms. Given the small clinical trial size I'd imagine more money went to contract manufacturing than research. There are very few companies and people that can make gene therapies at scale at acceptable quality levels. It seems like this expertise is quite expensive. There was probably a lot of process development here since making these drugs at scale is so new (here "at scale" means "enough for ~100 patients") They also spent a lot building out their own manufacturing site. They also had a $135M license payment, I think to the company they licensed their viral vector from
- continuations 7y ago
  Maybe that $700M includes money they spent on ideas that turned out to be failures. So something like they spent $100M on this one winning idea but also $600M on 50 other ideas that went nowhere.

JamesBarney 7y ago

I've always been curious how that 700M breaks up? How much of that is FDA trials and how much is paying 1000 phds to work on designing drugs for 7 years. Of course the follow up question is there any way to get that number lower.

aaavl2821 7y ago
It seems very high to me as well, especially since they were allowed to submit for approval with only one 15-patient study completed and interim data from a few other studies. I think they have tested the drug in under 100 pts to date Per the financial statements, ~$500M of that $700M was r&d (the rest was general and administrative expense). A lot of the r&d spend was payments to third party research and manufacturing firms. Given the small clinical trial size I'd imagine more money went to contract manufacturing than research. There are very few companies and people that can make gene therapies at scale at acceptable quality levels. It seems like this expertise is quite expensive. There was probably a lot of process development here since making these drugs at scale is so new (here "at scale" means "enough for ~100 patients") They also spent a lot building out their own manufacturing site. They also had a $135M license payment, I think to the company they licensed their viral vector from
continuations 7y ago
Maybe that $700M includes money they spent on ideas that turned out to be failures. So something like they spent $100M on this one winning idea but also $600M on 50 other ideas that went nowhere.