Alnylam Once Again Clutches at IP Straws to Support Valuation Gap (with correction)

(21Aug14) Yesterday, I mistakenly stated that Alnylam wrongfully concluded that Dicerna was infringing on a newly issued Tuschl patent.  Following comments in the comment section below, it came to my attention that indeed there was a claim that I missed, claim 81 (and some contingent claims), that covers RNAi triggers of 25 base-pairs as follows:

81. An isolated double-stranded RNA molecule, comprising:

(i) a sense strand and an antisense strand that form a double-stranded region of up to 25 
base pairs, said sense strand having an identity in the double-stranded region of at least 85 
percent to a target RNA molecule; and

(ii) at least one strand having a single-stranded 3’-overhang, wherein said 3’-overhang 
has been stabilized against degradation; and

(iii) at least one nucleotide analogue, 


wherein said RNA molecule is capable of target-specific RNA interference. 

Note that Dicerna's RNAi triggers make use of the 2'-O-methyl modification which sometimes is found in the 3' overhang and can also have stabilizing activity.  Taken together, this claim indeed questions Dicerna's RNAi triggers, and although I would expect vigorous debate around whether 25 base-pairs are covered by the patent's description requirements should it come to a patent litigation, the assumption is that Alnylam's new patent rightfully questions many, if not most of the RNAi triggers used by Dicerna currently.  

Since I'm at it, the new patent also comes awfully close to the asymmetric RNAi trigger designs by RXi Pharmaceuticals and others (asiRNAs).  RXi e.g. uses dsRNA lengths of below 15bp with the guide strand having a long 3' overhang.  I am a bit surprised that Alnylam got just enough extension both below and above their traditional 19-23bp stronghold to start overlapping with some asiRNA and Dicer-substrate designs.

Regardless, I stand by my point that Alnylam has re-invigorated their patent-related press releases in order to explain the valuation gap to its peers in the public markets.  The original blog entry follows here:



This morning, Alnylam greeted the competition with another IP-related press release.  It wrongly claims that a patent it just obtained covers competing technologies.  This suggests that it either lacks an understanding of RNA technology basics or that it is afraid that the market will come to understand that the valuation difference to its peers has no basis in either a commercially more attractive clinical pipeline, a superior patent estate, or simply better technology.

Dicerna’s Dicer-substrate technology not in 14-24bp range

Today’s press release concerns US patent application 13/725262 which is part of the Tuschl patent estate covering certain RNAi triggers with 3’ overhangs.  Although the patent has not finally been published, based on the latest submitted claim set, the RNAi trigger covered by the main claim should comprise the following features:

a)      a dsRNA length of 14-24 base-pair; 

b)      at least 1 3’ overhang;

c)       at least one ‘nucleotide analogue’;

d)      and the dsRNA is non-enzymatically processed.

Clearly, in citing the Rose et al. and another paper by Dicerna (actually their scientific founders from the Rossi lab at the City of Hope) as proof of Dicerna’s infringement, Alnylam hopes that its investor and business development audience does not actually read scientific papers. 

‘Specifically, the newly allowed patent application broadly covers small interfering RNA ("siRNA") molecules of various designs, including so-called "dicer substrate" RNAi triggers (Amarzguioui et al., Nat Protoc.2006;1(2):508-17; Rose et al., Nucleic Acids Res. 2005 Jul 26;33(13):4140-56)…’

Otherwise, it would quickly become apparent that Dicerna’s version of RNAi triggers have a dsRNA length of 25 base-pairs and, well, are enzymatically processed: Dicer substrates!

[Note: in the original entry I mistakenly said Dicerna's triggers were 27 base-pairs; to be precise, they are 25/27 designs with 25 base-pairs and a 2 nucleotide 3' overhang on the guide.]

So as the actual clinical pipelines of Arrowhead Research and Tekmira are about to look more attractive in terms of commercial value (HBV alone), look forward to more Alnylam patent-related press releases to help the market understand why Alnylam has a market cap of $5 Billion and its competition only about 1/10^th of that.

PS: the claim that usiRNAs infringe on this and other patents by Alnylam largely depends on the definition of ‘nucleoside analogue’ and ‘modified nucleotides’.

PPS: this patent does not change Alnylam's position as very similar ones related to 3' overhangs have already issued.  However, by slicing and dicing a patent application, it is possible to get issued a set of highly similar patents which, of course, is great fodder for the PR department.

A Very Obvious Method of Generating Tuschl-type siRNAs

John Leavitt over at the RNAi Litigation blog just posted ‘Alnylam’s’ defense against the accusation of the University of Utah that Utah had been deprived of ownership over the 3’ overhang feature of the Tuschl-type siRNAs. Setting aside the underlying merits of the case, the Motion to Dismiss the Second Amended Complaint by the University of Utah highlights one thorny issue with the granted Tuschl II (T-II) patents in the US, namely that they, so far, do not claim the 3’ overhangs themselves, but a method of generating 3’ overhang siRNAs. This method stipulates, in a first step, the synthesis of the individual strands, and then, in a second step, hybridizing (‘combining’) them to form the 3’ overhung siRNAs. The motion by the Defense consequently argues that Utah is missing the point in its suit by alleging ownership over the 3’ overhang feature, but not the method of generating 3’ overhung siRNAs subject of the US patents.

This to me is a) misleading since the Tuschl II IP estate to which Utah lays claim includes the European T-II patent which expressly claims the overhangs and Alnylam/Max Planck are obviously hoping to get similar composition of matter claims issued in the US, and b) the issued US T-II could only be considered novel by the USPTO based on the novelty and utility of the composition of matter that results from this method, i.e. 3’ overhang siRNAs for inducing gene silencing.

That the USPTO issued these claims in the first place is quite puzzling. Synthesizing and combining small RNAs is the most obvious method of generating 3’ overhung siRNAs. I would argue that even the average highschool student can come up with this method after a basic lesson on nucleic acid structure, not to speak of the ‘person having ordinary skill in the art’ which is considered the standard for obviousness. Unless I have missed an important exception in US patent law, similar to the Swiss-style claim construction in Europe to which the US T-II claims are reminiscent of, the US T-II claims seem very tenuous to me and probably should not have been granted.

It is amazing that both Plaintiffs and Defense are spending all this energy (=time and legal fees) skirting around the main issues (namely that Tuschl was probably motivated to test the 3' overhang feature based on Bass' speculations, and that Bass cannot be named a (co-)inventor since there was not even a semi-formal collaboration between Bass and Tuschl and Utah never bothered to file a patent). The answer by Utah to this motion is predictable, and so it will go on and on...

Bass versus Tuschl: Patentability of 3’ Overhangs Could Depend on Meaning of ‘Person of Ordinary Skill in the Art’

2000-2001 was one of the most exciting, fast-moving periods in RNAi history. All the evidence from the various fields converged to unravel the central mechanisms of RNAi resulting, amongst others, in the discovery of how RNAi can be triggered in human cells. Plant scientists discovered that small RNAs were involved, a group in Cold Spring Harbor found that the 3’ overhang-generating enzyme Dicer played a role, fly biochemists, led by Tom Tuschl, found that synthetic siRNAs could induce RNAi first in fly extracts, and then also human cells. Eureka!

The discovery of RNAi in human cells would not have been possible without the contributions by so many other bright minds. In addition to the research publications, the scientific discourse that happens at conferences and the hypotheses put forward in review articles are a vital part of the scientific process. Everybody stands on the shoulders of others, but it is also acknowledged that the peer recognition belongs to the person/group that publishes first. It is this peer recognition more so than monetary compensation that is the satisfaction of many a scientist. If I’m honest, yours truly probably wasn’t all that different.

Tom Tuschl clearly deserves the scientific recognition. Being the first to demonstrate in a robust manner RNAi in a range of human cells in such a competitive field is no minor feat, nor did it come out of nowhere. Tuschl, partly in collaboration with Zamore, Bartel, and Sharp, did a considerable part of the groundwork leading up to the breakthrough. One should not forget that Brenda Bass, who (or is it really the University of Utah) contests the intellectual ownership over the 3’ overhang feature of the ‘Tuschl siRNA’, was in fact commenting on a paper involving Tuschl during his time as a post-doc in Cambridge, Mass, when she speculated about the possibility of 3’ overhangs in the Cell Review. The other discussions that she claims to have had with the Tuschl siRNA inventors and Zamore were also likely strongly influenced by that research…although I cannot say that for sure. In that sense Tuschl stood on the shoulders of many RNAi giants, Bass on top of Tuschl, and then (maybe) Tuschl again on top of Bass. You get the picture.

Whether having earned the uncontested peer recognition will be followed by the issuance of strong patents, however, is an entirely different question, as I believe that there is merit to Brenda Bass’ contention that Tuschl was at least in part inspired by her work and ideas. On the other hand, I should repeat that because Bass chose to publicly disclose her ideas without filing for patent protection, her ideas became a free-for-all and part of the prior art so that trying to get her name on the patent would seem like inequitable conduct in broad daylight. Therefore, the consequences of the decision whether Bass was instrumental for Tuschl to choose the 3' overhangs in his siRNA design would seem one of patentability of 3’ overhangs, not ownership.

Why do I believe the University of Utah has a point with regard to Tuschl having been inspired by its very own Bass? Some evidence can be found in the commentary of the early papers themselves which I re-read with the question in mind of what gave Tuschl the actual impetus to choose the 3’ overhang feature.

Chapter 1: Long double-strand RNA is processed into small RNAs which direct RNAi in fly lysates (Zamore, Tuschl et al. Cell March 31, 2000)

In this paper, which is a central component of the Tuschl I patent application, Zamore and Tuschl studied the processing of long double-stranded RNAs in fly lysates and found that 21-23 nucleotide small RNAs were generated. They subsequently mapped the cleavage sites in the target mRNA and found that it was cleaved at 21-23 nucleotide intervals. Consequently, the authors suggest that the 21-23 nucleotide RNAs direct RNAi target cleavage- a hypothesis that would survive scientific scrutiny.

Of note, their proposed model of RNAi in Figure 7 shows siRNA without any evidence of 3’ overhangs.

Chapter 2: Bass proposes that small RNAs observed in Zamore et al. paper form 3’ overhangs (Cell April 28, 2000)

In this review in Cell, which shortly followed the Zamore et al. paper and which is a key exhibit in the Utah case, Bass proposes, for the first time, that since only RNase IIIs were known to be able to cleave double-stranded RNAs at specific sites, the small RNAs observed in the Zamore et al. studies ought to form 3’ overhangs as the typical feature of RNase III cleavage.

Importantly, she further speculates that the predicted instability of these 3’ single-stranded overhangs explains why Zamore et al. observed a mixture of small RNAs between 21 and 23 nucleotides, instead of small RNAs of a single size (i.e. 23 nucleotides) as one might have expected from the involvement of a single enzyme, since the single-stranded RNA 3’ overhangs of the initially generated siRNAs would be susceptible to degradation in the fly lysate. As we will see, this idea may have been picked up by Tuschl in his subsequent studies, providing additional support that he and his group indeed read this review very closely. Is Bass’ speculation in fact the reason why half the RNAi Therapeutics world and research/reagent providers still cling to the (actually largely useless, and sometimes indeed harmful) 3’ dTdT feature?

Chapter 3: First demonstration, by the Tuschl lab, that synthetic siRNAs, especially those with 3’ overhangs, can trigger RNAi (fly lysates; Elbashir et al. Genes and Development January 15, 2001)

Before the Tuschl group famously published in Nature on the finding that synthetic siRNAs with 3’ overhangs can trigger efficient RNAi in human cells, they published initial results with such siRNAs in fly lysates in Genes & Development. This paper is critical as it is the first use by the Tuschl group of siRNAs with 3’ overhangs. How would they explain the adoption of this feature for their RNAi trigger design?

In addition to mapping the RNAi cleavage sites in the target RNA, Elbashir and colleagues cloned the ~21-23nt small RNAs themselves. These were then analyzed for the modifications at the 5’ and 3’ ends, upon which they found that the small RNAs contained 5’ phosphates and 3’ hydroxyl groups. They then speculated that these termini indicate that they had been processed by RNase III as such ends are characteristic of cleavage by this enzyme. They then go on to show that synthetic siRNAs with 3’ overhangs were more efficient triggers of RNAi in fly lysates than the corresponding (blunt) RNAi triggers without them. The explanation for the choice of the 3’ overhangs was the RNase III speculation.

Importantly, Tuschl and colleagues acknowledge in their model at the end of the paper that Bass, in her 2000 Cell review, already proposed the involvement of an RNase III. I guess this is pretty good evidence that there was at least some inspiration, although they also suggested earlier in their description of the experimental results that this conclusion could also been drawn independently from the small RNA end modifications that they determined. Here, I would disagree with Tuschl and colleagues as there are other RNases that leave 5' phosphates and 3' hydroxyls, and it is really the understanding of a dsRNA expert like Bass that long double-stranded RNAs get processed by RNase III. That an RNase III was involved in RNAi was shown around the same time by the Hannon group in Cold Spring Harbor. But even if we assume that the conclusion of the RNase III involvement could have been drawn independently, the Bass review was part of the prior art at this point.

I was surprised that it was the RNase III speculation and not direct experimental evidence that led Tuschl and colleagues choose the 3' overhangs in this research. Without re-visiting this issue, I had been under the impression that it was the small RNA cloning results that allowed Elbashir et al. to reconstruct the 3' overhang siRNAs.

Final Chapter 4: Synthetic 3’ overhang siRNAs trigger RNAi in human cells (Elbashir et al. Nature May 24, 2001)

This is the famous paper where Elbashir and colleagues from the Tuschl lab showed for the first time RNAi gene silencing in human cells, using synthetic 3’ overhang siRNAs. Unlike the preceding Genes&Development paper, this one discloses the exact nature of the overhangs. With the exception of one siRNA, all of them contained a dTdT in the overhang (deoxy-ribose, not ribose). This immediate adoption of deoxy would seem strange initially, because the siRNAs that are generated from long dsRNAs would have ribonucleotides in their overhangs and why risk a loss of silencing activity by substituting them with deoxy? The cited reason was that it would reduce the cost of synthesis, and probably also stabilize them from degradation. The synthesis cost argument clearly is not a valid reason for using deoxy for a basic science study such as this one, so it makes sense that it was actually instability that was on their mind, an instability first suggested by Bass in her review.

In light of the evidence in the above publications alone, I believe that it will be very difficult to make the argument that Bass did not influence the 3’ overhang feature in Tuschl’s siRNA design. Nor do I see much evidence that Tuschl made an effort to hide this and quite fairly referenced the Bass review in the Genes & Development paper. I have seen much worse examples of groups deciding not to cite the work of others because it may take away from the novelty of their research. And furthermore, if the Bass review was prior art, why didn’t others jump on the idea and published before Tuschl? Was it because he had an unusual understanding of what was going on in the field and the Bass review, presentations, and discussions only gave an extraordinary person like Tuschl the right insights, or was it ‘just’ because he was already set up to do the experiments and that there would have been many others that were inspired by Bass and would have come up with the 3’ overhang siRNAs if Tuschl had gotten into a bad accident.

The decision whether a person in the art of ordinary skill would have been able to exploit Bass’ speculations to come up with the 3’ overhang siRNAs, similar to how it apparently influenced Tuschl and colleagues, is one of the critical questions challenging the patentability of the 3’ overhang feature, possibly the only RNAi trigger feature of relatively broad scope and value that Alnylam can still hope to control.

The Importance of 3’ Overhangs in RNAi Therapeutics

[Note: some of what follows, has already been discussed in the entry ‘On the Importance of Being Tuschl’]

There has been a flurry of activity surrounding the Tuschl patents. It appears that at stake is no less than the Tuschl II patent which claims the use of 3’ overhangs in siRNAs. The uncertainty comes from some of the owners of the Tuschl patent estates prosecuting the Tuschl I patent series such that it uses scientific data underlying the Tuschl II patents. Although I cannot see that 3’ overhangs are explicitly claimed in Tuschl I, the presence of such overhang data in Tuschl I could invalidate the claims in Tuschl II as their prior use would suggest that overhangs were already practiced in the art and/or trigger double-patenting/interference issues.

It has always been a surprise to me that virtually out of nowhere, Tuschl II data, generated at the Max Planck in Gottingen after his time at the MIT, would appear in Tuschl I, which is based on work at the Whitehead/MIT and UMass. The work underlying Tuschl I identified that short siRNAs are generated from longer dsRNAs (in Drosophila extracts) and that the short siRNAs are the likely mediators of RNAi. No mention of overhangs, and no reason to anticipate that overhangs would confer an advantage. The work by Elbashir, Lendeckel, and Tuschl that underlies Tuschl II characterized the siRNAs processed in Drosophila extracts and noted that they contained 3’ overhangs. Synthetic versions of these siRNAs were then found to mediate RNAi in Drosophila extracts and, in the famous Nature paper, human cells. The 3’ overhangs were not just functionally irrelevant consequences of RNase III processing, but found to confer a distinct advantage in RNAi silencing efficacy. This advantage has been borne out in many studies since by labs throughout the world, and the reason seems to be that the overhangs allow for efficient Argonaute loading of the guide strand. For a while, 3’ overhangs appeared to have an additional advantage over blunt-end siRNAs in that they would avoid some of the innate immune responses, although this is an area in which the jury is still out. Another finding that has been borne out by numerous studies is that 19-21 bp lengths work better than smaller or larger siRNAs (not discussing here Dicer-substrates though).

Tuschl I claims double-stranded RNAs of 21 to 23 nucleotides. Note that it says nucleotides, and not base pairs, so in theory it could be dsRNAs that do not have to be blunt ended. But again, at the time of the invention, nobody in the field would have been able to anticipate the presence, let alone an advantage of the overhangs. Tuschl II claims dsRNAs with individual strands of 19-23nt in length, with at least one of the ends having a 3’ overhang.

I do not want to comment here too much on the specifics of the case, and Doug Macron from ‘RNAi News’ has done a great job in following the story- except to say that it is clear that Max Planck (and Alnylam) would never have agreed to such use of Tuschl II data, and even if there had been some procedural issues (e.g. Tuschl ‘swore’ etc, when in reality most academic scientists trust that their patent agents would act in their best interest, and therefore just sign off patent documents that come across their desk/bench) I believe that at some point fiduciary duty has been violated by those hired and paid, also by Max Planck and Alnylam, to prosecute the Tuschl patents.

While I consider it possible that some fancy workarounds of the Alnylam IP estate may give certain freedom-to-operate to some companies, scientifically the sweet spot of siRNAs are those 19-21bp in length with at least one 3’ overhang end and the drug development economics would dictate these to be used and licensed. This area is very well covered by Alnylam through Kreutzer-Limmer, Tuschl I and II, although an unexpected outcome of the Tuschl proceedings could change the picture. On the other hand, one cannot dismiss the possibility that some non-Tuschl-like structure in combination with a certain modification pattern similar to what Silence Therapeutics claims e.g. actually works and is advantageous contrary to expectation, that is non-obvious, and I'd be happy to look at data that systematically demonstrates this.

Recent Alnylam Patents Blur Distinction between Blunt-end dsRNAs and those with 3’ Overhangs

As many of you will be aware, the battle for RNAi IP is heated and particularly centers on whether Alnylam may also dominate over blunt-ended double-stranded RNAs between 22 and 24 base-pairs in length. Longer dsRNAs are covered by Fire and Mello and relatively easily accessible, 21bp and shorter dsRNAs are the domain of Kreutzer-Limmer and exclusive to Alnylam as are dsRNAs up to 25bp with 3’ overhangs (Tuschl II).

Although, even in the absence of applying obviousness criteria, I doubt that if Kreutzer-Limmer ultimately failed to be applied to 22-24bp RNAs, any other patent application would be able to do so based on prior art, this still raises the question whether 22-24bp RNAs will fall into a free-for-all grey zone.

While the seminal Elbashir et al. publication underlying Tuschl II provided compelling evidence that in most cases 3’ overhangs will enhance RNAi gene silencing, at least in tissue culture, recent patent application by Alnylam itself make me rethink the value of blunt-ends, particularly for in vivo applications.

At the end of January, Alnylam issued a press release on the issuance of the Woppmann patents in the UK (UK 2417727). Although I would not, by any stretch of the imagination, consider it to be an umbrella patent of the stature of Tuschl II, it is still a quite curious patent with relatively early 2003-4 priority in that it describes siRNAs with one blunt-end and one 3’ overhang end (often just one nucleotide) to have gene silencing advantages over the classical Tuschl design features (two 3’ overhangs, best if 2 nucleotides long).

I know that the following passage from the PR has caused some confusion among some of the readers of this blog, and admittedly also myself, as without carefully reading the claims of the patent it could be misunderstood as also covering dsRNAs with two blunt-ends, instead of dsRNAs combining the two features in just one molecule: ”The claims cover siRNA molecules of any length that contain "overhang" and "blunt end" design features, including siRNAs containing chemical modifications and certain novel motifs.”

Does this mean that Alnylam is back-tracking here on the value of 3’ overhangs, i.e. Tuschl II, which could have serious repercussions for the whole RNAi Therapeutics IP space and give companies like Silence Therapeutics or RXi some room to breath? The issue therefore is whether the one 1-nucleotide overhang is really that advantageous or just serves as a fig-leaf designed to disguise the value of blunt-end siRNAs. I therefore found the following passage from the description of another recent Alnylam patent application on the targeting of the Huntingtin gene by RNAi (USPTO application no. 11/588,674) of interest:

“In one embodiment, at least one end of the dsRNA has a single-stranded nucleotide overhang of 1 to 4, preferably 1 or 2 nucleotides. dsRNAs having at least one nucleotide overhang have unexpectedly superior inhibitory properties than their blunt-ended counterparts. Moreover, the present inventors have discovered that the presence of only one nucleotide overhang strengthens the interference activity of the dsRNA… dsRNA having only one overhang has proven particularly stable and effective in vivo, as well as in a variety of cells, cell culture mediums, blood, and serum.”

Maybe it should not come as a surprise that with intense research, the design features of RNAi triggers will evolve over time, somewhat reminiscent of what had happened with monoclonal antibodies before and this can only be a blessing for the realization of RNAi Therapeutics. However, from a business point of view, this raises the question what will be considered sufficiently novel or merely an improvement of a fundamental design. Clearly, at a time when many dsRNAs and also single-stranded oligonucleotides are found to be able to accomplish gene silencing via RNAi, the systematic analysis of particularly dsRNAs between ~19-25bp in length with various types of overhangs or no overhangs should be of value here. I’m almost sure this has been done already and is ongoing in other parts of the industry, but it certainly would be nice to see an unbiased publication on that very subject.