(Notice: The following account is a largely hypothetical explanation of what is underlying Alnylam's current troubles)
The last time Alnylam expressed surprise and regret was in September 2010 when Novartis declined to exercise an option for wide, non-exclusive rights to Alnylam’s RNAi Therapeutics IP estate ('Adoption License'). Since then Alnylam has been hit with a string of other partnership setbacks and other not-so-positive news pieces. The official party line, of course, is that these are all isolated events and Big Pharma does what it does because it has to cut down on R&D that won’t generate significant revenues in the next 3-5 years. Meanwhile, the popular story line is that Big Pharma has given up on RNAi Therapeutics because of the delivery challenge.
I believe that neither of these views are correct, and instead like to propose that while it is true that delivery is a challenge, there are technologies that show great potential, and it is the fight for ownership over and access to these technologies that is at the root of the recent developments. As you’ve heard me preach over the last 3 years or so, Tekmira’s SNALP delivery technology is at the forefront of systemic delivery and has consequently become the major bone of contention as Alnylam, its partners, and Merck struggle to protect their investments in RNAi Therapeutics.
Novartis Poses Target Threat to Alnylam and Partners, but needs SNALP
In retrospect it is clear why Novartis has acted the way it did. Initially, I had thought that Novartis' target selection strategy might be uncoupled from delivery considerations and instead would focus on the industry’s hottest GPCRs and kinases. Instead, Novartis appears to have done the scientifically sensible (Big Pharma and scientifically sensible in the same sentence...must be a first in this blog) and held off with their decisions until there was more clarity which targets and indications could be addressed based on the most advanced delivery technologies. As such, I believe that Novartis has mainly picked a bunch of targets in the liver, solid cancers, and maybe a few ocular and CNS indications, too.
In the Novartis post-mortem conference call, I was surprised to learn that as part of selecting their targets, Novartis and Alnylam had conducted quite a few pre-clinical in vivo studies in anticipation of IND filings, meaning that SNALP delivery was most likely employed, and that along with the targets came rights to some of Alnylam's current, and future (!) delivery capabilities!!! This to me was a bombshell because I had believed that delivery was Alnylam’s biggest negotiating advantage in making sure that Novartis will pay the $100M exercise fee. Well, the reason for my belief was that since Alnylam liked to emphasize that the Adoption License excluded delivery, similar restrictions applied to Novartis target picks.
Note that Novartis, which only recently appears to have ramped up their RNAi-related staff to around 100 (not really what you consider a sign of their losing interest in RNAi Therapeutics), was named in the Complaint by Tekmira as a company that Alnylam might have improperly shared SNALP delivery with. It may also be of interest that the first author of a high-profile Nature Biotech paper in 2005, David Morrissey, describing the use of SNALP delivery in a mouse model of hepatitis B is now heading RNAi Therapeutics efforts at Novartis.
31 targets, exclusive, heavily focused on the liver and solid cancers…what does that leave for Alnylam and its other Big Pharma partners? Is ‘5x15’, Alnylam’s strategy launched in January, i.e. after Novartis, a reflection of Novartis having picked higher-profile targets such as hepatitis C?
So 2 months after Novartis, Roche announces that it does not view RNAi Therapeutics a sufficiently broad commercial opportunity to justify the significant operations they had going. Since Roche, like Alnylam and Novartis, were likely to have focused their efforts on liver and cancer targets as they were looking at the same delivery technologies, this comment can also be interpreted as ‘After Novartis has taken many of the targets that we were interested in, what we are left with isn't attractive enough to make our large-scale effort worthwhile. Moreover, we are contrite that Novartis got so much paying so little'. And when they licensed the liver, metabolic, and cancer fields from Alnylam for $50M each, they must have done so in view of SNALP delivery.
Soon after, or even a little bit before the 2007 alliance, Roche together with Alnylam already got underway with validating targets and moving them into clinical development. But then another snag hit: Despite of what Alnylam might have promised Roche, Tekmira’s SNALP turned out to be more difficult to replicate than they thought. This is very apparent from Roche’s Factor VII patent application which started out with lipidoids up to the mouse stage, but then adopted Tekmira’s SNALPs for the non-human primate studies. If you are from Big Pharma and reading this, it is a lesson not to be blinded by company brands, but closely look at who actually has the know-how and partner with them instead.
These frustrations first came to the fore in comments by Alnylam’s David Bumcrot in the summer of 2007 in which he proclaimed that the MIT lipidoids that Alnylam helped fund are much superior to Tekmira’s SNALP technology and will be used for their upcoming clinical programs. To boot, he added that Tekmira’s ApoB was a terrible target. But, of course, despite of Alnylam’s efforts already in 2007 to marginalize Protiva/Tekmira, both Roche and Alnylam had to admit defeat and come back to Tekmira which allowed Alnylam to file their IND for liver cancer.
All this made it clear to Roche that it needed alternative delivery options. Roche went out to look for alternative delivery technologies and in 2008 bought Mirus Bio for the Dynamic Polyconjugates (DPCs) for $125M (almost 4x Tekmira’s current market cap). This number just illustrates the value of delivery in any of Alnylam’s deals, and if you consider that Alnylam most likely passed on DPCs and is betting the house on SNALP instead, the $1B figure that Tekmira demands is very well anchored in what actual money has changed hands.
In addition to Novartis crossing Roche’s RNAi plans, another reason for Roche unhappiness could therefore be that Alnylam over-promised them in terms of what Alnylam had to offer in delivery. As Roche interacted with Tekmira, it may also have noticed that Tekmira’s views on what SNALP know-how and IP access could be gotten from Alnylam and what Roche had to come to Tekmira for strongly differed with what Alnylam might have told them. As a result, on top of the ~$300M upfront they had given to Alnylam, they now had to close a separate deal with Tekmira to actually be able to do something with that RNAi trigger IP.
Obviously, as SNALP delivery solidified its position as the industry's leading systemic delivery technology with not that many other advanced options (maybe Silence's lipoplex comes closest, but for endothelial indications), it turned out to be a bad deal for Roche.
Coming back to Novartis and SNALP delivery. In early 2009, Novartis surprisingly bailed out Marina Biotech, under the name of mdRNA then, by giving them $7.25M for kicking the tires on their ‘DiLa2’ liposomal delivery technology. This came a bit surprising to me, because I had considered Tekmira to be leading in systemic liposomal delivery. This to me indicated that Alnylam indeed is trying to keep Novartis’ RNAi ambitions in check by restricting access to delivery and this is why Novartis, knowing of the potential of liposomal delivery, took a chance with Marina.
You can easily see what pressures Alnylam might be facing from its current and former partners. It has taken over half a billion in license fees, a lot of that presumably driven by Alnylam’s claims about the maturity of and, importantly, control over SNALP delivery. The question is: Do the partners want their money back or are they insisting on their promised access to SNALP delivery?
One of these partners not mentioned so far in this story is Takeda. In 2008, Takeda paid Alnylam $100M for non-exclusive access to the metabolic and cancer fields, plus $50M in technology transfer payments, the latter of which has apparently been completed just in the nick of time to meet the guidance provided by Alnylam. In other words, Takeda paid Alnylam $150M for SNALP technology. In previous conference calls, Tekmira noted that it was working together with Takeda to sort out SNALP delivery (as Takeda and Alnylam were working on the ‘tech transfer’). Despite of Takeda’s complementary comments in yesterday’s press release, I would not be surprised if Takeda, like Roche, is quite frustrated by now over the uncertainties over SNALP access and the Novartis exclusive target picks, and you’ve got to wonder what role Takeda will play in Tekmira’s case against Alnylam.
And lest we forget, there is, of course, Merck. Once spurned by Alnylam and having invested probably over $1.5B in RNAi Therapeutics development already with hundreds working on the technology, Merck has every reason to turn the table on Alnylam. Merck's publication record also makes it clear that they harbor strong ambitions in SNALP delivery. While the details of the Tuschl II settlement under which Merck gained access to Tuschl II IP are unknown to the public, it is quite possible that Merck is satisfied with that outcome and is now focusing their attention on gaining control over the most advanced systemic delivery technology. Should it view the Tekmira litigation against Alnylam an opportunity to gain leverage over Alnylam, I would bet that they will indeed grasp at it.
Considering this tangled web of interest and the central importance of Tekmira’s technology, it is no wonder that Alnylam is scrambling to stand on its own SNALP feet. Alnylam- setting aside personal issues for a moment, the decision can’t be that hard, can it? In fact, I don’t think there is even a choice.