What Alnylam Doesn't Want You To Know

Recently Alnylam (NASDAQ:ALNY) put out a press release titled "Alnylam Retains Key Claims for Kinesin Spindle Protein RNAi Patent in Interference Proceedings." On the surface this seems like great news. Protiva, A Tekmira (TKMR) subsidiary, had challenged Alnylam's patent that surrounds EG5. This patent covers the enabling technologies of ALN-VSP minus its method of delivery which is manufactured and owned by Tekmira. It is important to note that this is separate issue from the main litigation between the two companies surrounding LNP delivery. This patent interference is just another example of the stress their relationship is under. To be quite honest, it was not on the minds of most investors of Tekmira or Alnyalm until the recent PR war.

A segment of Alnylam's press release reads, "United States Patent and Trademark Office (USPTO) has maintained key claims in the company's KSP RNAi patent (US Patent No. 7,718,629) in interference proceedings." With the absence of any comment from Tekmira, it seemed as though they had a won a clear victory at this preliminary motions phase of the interference proceedings.

It wasn't upon reading Tekmira's press release the following day where I was able to get additional clarity on the issue. There is a huge difference between the tone of the two separate releases. A segment of Tekmira's release sates, "At this stage of the proceedings, the BPAI denied or deferred all four of Alnylam's motions and granted two of our three motions, including our motion that Alnylam's broad claims are unpatentable due to lack of adequate written description support."

Clearly the two sides interpreted the results of the USPTO's orders differently. I expected nothing less. Alnylam has a history of declairing victory in their PR releases when reality might be different. Its similar to watching two fighters raise their hands up in the air in victory when a fight goes to a decision. Most of the time, the audience who has watched the fight knows the result despite both fighters claiming a win.

In keeping with my analogy, the problem here is that we did not get the chance to watch the fight and we just get to see both fighters raising their hands at the end. This uncertainty existed until Tekmira released the USPTO's documents along with their press release. With these documents, we are able to watch PPV on replay. We have the chance to dig deep into what actually happened and make a judgment for ourselves. Alnylam didn't want you to see it. The requested Dr. Rossi's deposition, who testified for Protiva, be stricken from the record. After reading it, I can see why. Throughout all of the motions, his arguments are sound and embarrassing for Alnylam. Don't worry, I will provide one of the many examples.

Let me give you a summary of the whole interference in plain English. Both Alnylam and Protiva have patents relating to the silencing of EG5. Alnylam is the senior party as they were the first to file but Protiva says they invented it first. The united states is on a "First to Invent" patent system, so whoever wins that phase of the interference is the ultimate victor in this matter.

The following is the text of Claim 1 in Alnylam's EG5 patent in which Protiva attempts to undermine. I post this quote knowing full well how dry a patents text is. It is important for the understanding of the issues of this interference.

"A double-stranded ribonucleic acid (dsRNA) for inhibiting the expression16 of a human kinesin family member 11 (Eg5) gene in a cell, wherein said17 dsRNA comprises a sense strand comprising a first sequence and an 18 antisense strand comprising a second sequence complementary to SEQ ID19 NO:1311, wherein said first sequence is complementary to said second 20 sequence and where in said dsRNA is between 15 and 30 base pairs in 21 length."

I am going to center this article around Protiva's Claim 2. This is the claim where both parties are claiming victory and the most important claim in the document in terms of invalidating Alnylam's patent. The other claims are pretty cut and dry and made no alteration to what already existed in Protiva and Alnylam's patents.

In English, claim 2 basically says that Alnylam's broad range of base pairs from 15-30 in length is crazy. Anybody would have known that at the time 19 base pairs was the only useful way to silence EG5. Furthermore, that the documentation that Anylam provided in their patent only show benefit of silencing EG5 was only with 19 base pairs. They provided no documentation on anything related to anything other than 19 base pairs.

Dr. Russi made and supported his arguments using citations in Alnylam's own patent! This is good stuff. Take a look at Russi's own words:

"Alnylam's '629 patent cites Elbashir et al. as a reference in support of the length of the 30 claimed dsRNA molecules. However, Elbashir et al. teaches that varying siRNA 1 duplex length can alter silencing ability and that such effects are not necessarily predictable.

Elbashir reported that siRNA duplexes of 20, 22 and 23 Nucleotides (i.e., 18, 20, 21 base pairs) in length were more than 8 fold less efficient in silencing compared to a 21 nucleotide (i.e., 19 base pair) siRNA duplex, and siRNA duplexes of 24-25 nucleotides (i.e., 22-23 base pairs) in length did not result in any detectable interference.

As a general matter, during the relevant time period, 21 nucleotide with 19 base pairs duplexes were generally viewed as the optimal length, while duplexes substantially longer or shorter in length, absent data to the contrary, were expected to produce sub-optimal silencing compared to a corresponding 21mer, or possibly no silencing at all.

Other than in the claims, it is my opinion that Alnylam's 629 patent does not reasonably convey to a person of ordinary skill in the art that Alnylam had possession of the full scope of claim 1; i.e., dsRNA molecules 'wherein said dsRNA is between 15 and 30 base pairs in length'"

Using the citation provided in Alnylams own patent by Elbashir was pretty elegant. In circumstances like this, Alnylam can use just the 19 base pairs as justification for the whole genus (15-30) IF the results of the other members of the genus are predictable. This is why Dr. Russi emphasized that the results were indeed not predictable in Elbashirs own words. In addition, Elbashir validates that other lengths are not particularly useful.

Alnylam had Dr. Tuschl testify on their behalf. Tuschl's counter arguments were asserting that base pairs other than 19 can be useful for silencing, and even Dr. Russi's own work at the time of the filing said so! This is an equally crafty defense and both Tuschl and Rossi had some pretty entertaining arguments.

"There were publications, including patent applications and scientific literature by Dr. Rossi himself, that were available at the time of filing the applications from which the '629 patent claims benefit , which suggested and even evidenced that both shorter and longer siRNA duplexes could be as potent, and even more potent, than the dsRNA comprising 19 nucleotides.

Kim et al (Nature Biotechnology 23(2):222-226; Exhibit 1031) . . . shows that 'synthetic RNA duplexes 25-30 nts in length can be up to 100 fold more potent than corresponding conventional 21-mer small interfering RNAs (siRNAs)"

Dr. Tuschl went on to make a point that Dr. Rossi was a co-author of the Kim paper.

My first reaction to this argument indicated to me that maybe this was why Protiva might be especially interested invalidating (or at least limiting the scope of) the patent in this manner. If they believed that duplexes of 25-30 nucleotides in length were 100 fold more potent than what Alnylam was using, that is pretty valuable stuff to go after. Perhaps they are interested in it.

Now, what did the USPTO say after all the arguments were heard? Basically, it was 5 and a half pages of bad news for Alnylam. If you are interested enough to be reading this article, read page 21-26 labeled at the bottom of the actual document itself. If you can explain to me how that is a victory for Alnylam, id be surprised. I will cherry pick a more succinct narrative of the findings.

"We find that a preponderance of the evidence supports Protiva's position that Alnylam's '629 patent lacks written description support for the involved Alnylam claims...

...We find Dr. Rossi's testimony credible evidence that a skilled worker would not interpret the disclosure of the '629 patent to show possession of the claimed genus of dsRNAs that are between 15 and 30 base pairs in length and inhibit expression of Eg5. Although the '629 patent states generally that the disclosed dsRNAs can have a length anywhere from 15 to 30 bp in length, it provides no examples of dsRNAs with a duplex length other than 19 bp that actually function to inhibit Eg5 expression, and Dr. Rossi has cited evidence that those skilled in the art recognized that lengths other than 19 bp had been shown to produce much less silencing or no silencing at all. The evidence thus supports a finding that a skilled worker would not have recognized the description provided in the '629 patent to show possession of the full genus of functional dsRNAs defined by Alnylam's claim...

Dr. Rossi has testified that those skilled in the art would recognize all of the specific dsRNAs described in the '629 patent to include a 19-bp duplex and 2-nucleotide overhanging ends, and Alnylam has pointed to no evidence to the contrary. Thus, regardless of how many examples of 19-bp duplex dsRNAs are provided in the '629 patent, Alnylam has provided no reasonable basis on which to doubt Dr. Rossi's testimony that the '629 patent provides no examples of dsRNAs that inhibit Eg5 and form a duplex with a length of 15, 16, 17, 18, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 bp.

Dr. Tuschl identified U.S. Patent Publication 2005/0244858 and the Kim reference as evidence that siRNA duplexes longer than 19 bp were effective in gene silencing. Both of these references, however, refer to dsRNAs that serve as substrates for the enzyme Dicer, which cleaves them into smaller segments. Alnylam has not provided adequate evidence to support a conclusion that those skilled in the art would have recognized the '629 patent to describe dsRNAs that are substrates for Dicer"

...The preponderance of the evidence of record therefore supports finding that those skilled in the art would not have recognized in the specification of the '629 patent a written description of the claimed compounds that is adequate to show possession of the full scope of the claimed genus."

What does this all mean in context of who won? Well, you need to understand first that this is a preliminary motion phase of the interference and what that means. In this phase, the parties raise issues that will be contested throughout the interference. If a party can raise an issue during the motion period and fails to do so, that party will not be permitted to raise it later. During the motion period, both parties try to obtain a position that will be advantageous in later stages of the interference. Often, the outcome of the interference will be determined by positioning that takes place during the motion phase. Please view this great resource on where preliminary motions fit into the scope of the interference.

This granted motion leaves Alnylam position significantly weakened. Despite Alnylam's inference to the contrary, The USPTO did not uphold Alnylam's 19 BP part of the patent, it invalidated its clams to the entire genus (15-30). Alnylam's claim that the USPTO upheld key technology in ALN-VSP (19 BP) is misleading. Since this was the only motion that had any bearing on either Protiva or Alnylam's patents, Protiva and Tekmira have had a clear cut victory at this stage.

What will ultimately happen in this manner is unknown to me, but it frustrates me to see deceptive tactics being used. There is a difference between focusing on the positive, and manipulating the information. In my opinion, actions like their latest press release are just another piece in a growing puzzle of Alnylam planting seeds of distrust when it comes to litigation related issues. Alnylam is an otherwise great company with many talented people working for it, but I am growing tired of leadership of John Maraganore. He has brought the company good fortune as a result of his connections, PR related efforts, and good science but I feel that the chickens may be coming home to roost some time in the not to distant future ...

Time will tell.

Disclosure: I am long TKMR.