If there ever was the impression that the pharmaceutical industry could afford not to pursue RNAi Therapeutics with urgency, this could not be further from the truth- at least not if you consider Roche a bellwether of the industry. Severin Schwan, who once served as the leader of the Roche Diagnostics division before becoming Roche's CEO, set the tone of this year’s Roche Investor Day by stating that the challenge the industry faces is to capitalize on our exponentially increasing understanding of the molecular basis of disease by being able to target more than the 100 out of the roughly 2,000,000 proteins in our bodies that mankind was able to develop in all its history.
This begs the question of where exactly the bottleneck lies. Is it true that small molecules and monoclonal antibodies alone will not be able to significantly expand on the 100 targets? If the answer is at least a partial ‘yes’, then the industry cannot afford to not invest in RNA(i) Therapeutics as the technology that virtually opens up the entire genome as drug targets.
The limitation in target space is quite obvious for monoclonal antibodies. While they have proven a great new class of therapeutics, they cannot address the majority of proteins that happen to reside inside cells. To be clear, there is a lot of innovation going on in monoclonal antibodies, including the development of ever more optimized and also ‘armed’ antibodies, but a lot of this has to do more with life-cycle management of existing MAbs than exploring new target space. Sure, with more insights into disease, new extracellularly accessible targets will emerge, but I do not see this alone sufficient to sustain the pipelines in the next 20 years. Of note, being in his early 40’s means that Dr. Schwan is one of the few CEOs that may actually be measured during his tenure against his long-term vision of drug development, rather than merely against his ability to squeeze short and mid-term value out of the current pipeline.
The limited target space of MAbs is actually also something that drove Genentech’s decision (now fully owned by Roche) to go against industry trends and adopt small molecules as a technology platform. By focusing on only a few select, but important cellular pathways rather than working superficially on many, a strategy that has really borne fruits, new attractive genetically defined targets have emerged that were not within the grasp of MAbs.
Certainly, there is room for novel exciting small molecule approaches for therapy, but their ability to expand on the druggable space is fundamentally limited by their cross-reactivity. To be fair, RNAi Therapeutics also have their off-target challenges, but I would submit that by off-targeting a random set of genes, which hopefully will teeter out in genomic noise, rather than structurally related, and functionally cross-talking GPCRs, kinases, phosphatases, proteases etc, there is less risk for confounding phenotypes, especially if you can weed out the most obvious unacceptable off-target profiles of RNAi Therapeutics early on with genomic profiling technologies.
So while safer and better MAbs, the exploitation of new major disease pathways also with small molecules, and the rapid growth of the emerging markets should drive the growth of Roche for the next 5-10 years, they are well aware that in order to substitute this growth they need new platform technologies, especially RNAi Therapeutics, and given the timelines, the time to invest is now, just as they invested into MAbs and PCR in the early 90’s when the precise commercial timelines of those technologies were still very much uncertain. And when it comes to developing a new platform, the quality of the science is paramount, and it is therefore not a coincidence that Roche has chosen Alnylam and Tekmira as its closest outside collaborators in that effort.
What is unique to Roche is that they have a very substantial foothold in (technology agnostic) diagnostics which allows them to lead the personalized medicine effort that aims to maximize the therapeutic benefit of a drug, thereby lowering clinical trial failure rates and justifying premium pricing in front of payers. If you had wondered what ever happened of the personalized medicine ‘hype’, it (not the hype) is already a reality. It seems that for virtually every drug in pre-clinical/early clinical research, Roche Pharma and Diagnostics are now collaborating to develop drug response markers and companion diagnostics. It is therefore only a matter of time until they come out on the other end of the drug approval process. Importantly, by seeking to understand the molecular basis of a ‘disease’ in a particular patient, this trend directly plays into the strength of RNAi Therapeutics as the key players in a disease do not care whether they are druggable or not. In fact, I find it ironic that there are actually some pharmaceutical companies, even some major biotechnology companies that use RNAi as a tool for target discovery that would limit their screening libraries to siRNAs that target only the so called druggable space. Not only are they missing out on a more complete understanding of a disease process in general which would also feed back positively on small molecule and MAb-based drug development, they will also miss out on many, if not most important targets. Once again, they risk being left behind by a company that sees the Big Picture and not only preaches innovation but actually lives it…Roche.
14 comments:
Dirk-
Thanks for your comments of the Roche take on this space.
Which RNAi therapeutics technologies have the best chance of creating marketed drugs? Which technologies have the least amount of off-target effects? How will the off-target effects be eliminated in the future?
Thanks, Steven. I assume that it must have been a similar perception of drug development trends that caused you to adopt RNAi Therapeutics as a platform.
All good questions. I invite you to browse the blog to find my opinions on some of them. As to off-target effects...it is unlikely that you will eliminate them completely, although I would expect that a profile of maybe 5 off-targets that reduce targets by 50% while the on-target has an efficacy of about 80% would be a good benchmark.
I am even debating with myself whether it would be worth the expense to progress 2 equipotent candidates in which only the siRNA sequence differs (for the same target) into early clinical development almost in parallel, and then go with the one that shows a better off-target profile in Man. A lot of the expenses would be shared and you would not lose 1-2 years going through the same IND-enabling cycle again.
I got into RNAi because it was catalytic (not stoichiometric like MAb and antisense), potent (pM IC50s), and drug design 'lite' (24 hr design vs months/years/never for small molecules). First analyses comparing in vitro IC50 based on the amount of drug 'above' a cell culture with amount of drug needed to saturate the extracellular space to the same level indicated microgram/kg dosing would work. In truth, my estimate of the concentration gradient between blood and extravascular/extracellular space is about 1/200th (not what I expected). So much higher doses needed. Delivery is being solved, slowly, but the future is an RNAi one. I am biased (MDRNA shareholder) but I like oral delivery option for this technology...the only marketing weakness of RNAi is the needle and now MDRNA has thrown that away.
BTW, I have never seen a commercial candidate with the poor level of discrimination you have described (50% vs. 80%). The seco-analogues MDRNA uses really break the partial binding of incomplete sequence homology and give wider discrimination.
2 RNAi per clinical program is a big company luxury IMO but a great idea.
One more comment.
Santaris has used locked nucleic acids to form 'second generation' antisense molecules that have superior pharmacological properties to the ISIS first generation molecules. They also licensed the diagnostic uses of locked nucleic acid as superior gene probes to Exiqon which is a $200 MM plus Danish company in its own right.
When MDRNA acquired from Valient the only bridged nucleic acid analogue patents in the world NOT owned by Santaris (Santaris had gone to Japan to scoop up the other LNA patents) MDRNA simultaneously became the first RNAi company that can offer either a potent, second generation single strand antisense option or a double strand RNAi option. I am not aware of another company that can play both sides of the street like that. Not ISIS, Alnylam, Rxi, etc.
And by selective use of their bridged- and seco- analogues they can now, by design, dial in a Tm change at any given base pair of from minus 10 C to plus 8 C. Changing a BP interaction by 18 C is an immense energy change in a very small three dimensional space.
Do you understand the ability to tune the biophysics of a dsRNA at this level?
FWI, my world view is that when it comes to RNA/DNA and their role in life, God is a biophysicist and Tm is his tool.
You might look deeper into this on your own and report your findings (full disclosure: I am MDRNA 'founder' and current shareholder).
Do either of you think that Santaris will be more effective at inhibiting microRNAs than Regulus?
How does Santaris technology stack up against the blockmir technology of Mirrx Therapeutics?
Which company has the best technology for microRNA mimics?
Nice post, well said.
Beyond expanding the number of druggable targets, the inherent scalability of RNAi therapeutic development is also extremely powerful. By scalability I mean, how much the investment in developing the siRNA therapeutic platform translates to the development of single drugs. Small-molecules are astonishingly non-scalable. Although we know an incredible amount about what makes a 'drug-like' molecule, achieving specificity and sensitivity of each small-molecule therapeutics requires tremendous labor, cost, and risk. Therapeutic mABs are somewhat more scalable, where the platform technology is mostly applicable across different individual mAB candidates. However, individual mAB development is still very costly and time consuming. siRNA therapeutics have the potential to be immensely scalable, where, following the development of the platform, developing an siRNA against an new target is becoming trivial as compared to small-molecule antagonists. In addition, many of the delivery platforms used for siRNA and largely independent of the siRNA sequence being delivered, further allowing for plug-and-play therapeutic development.
These issues become extremely important as we move away from developing single-target therapeutics to more multi-targeted, systems medicine, approaches. The current small molecule / protein therapeutics approaches likely won't cut it. Although I'm probably preaching to the choir...
Steven,
I agree with you that regulating the strength of base pairing during RNAi (within the siRNA duplex and when bound to target mRNA) is a very important element in dealing with RNAi Therapeutics specificity. Having both UNA and BNA available in the tech tool box should certainly increase the requisite capabilities. It would be good though to see confirmation through peer-reviewed publications of mdRNA’s ability to thus dramatically reduce off-targeting on a consistent basis. What I generally like about mdRNA is their sincere enthusiasm of what they are doing and getting things done with a can-do attitude, but at the same time there is a fine line to disappointing some of the resulting expectations if there is no follow-up validation showing sufficient depth in the various areas…at least this is what I’m strongly missing about the company. It used to be that ‘we can’t publish’ because of all the undisclosed inventions, but other companies have been able to do so.
On the topic of bridged nucleic acid analogues (also touching on a question by ‘Anonymous’)… I am surprised that you say that mdRNA has obtained access to the only BNA patent outside LNAs, because I’ve seen e.g. a BNA paper by ISIS, technology of which should also be available to Regulus, on a novel 2’-4’ constrained nucleic acid analogue (http://pubs.acs.org/doi/abs/10.1021/jm801294h), and it is my general impression that research in this area is still unfolding. On the question of Santaris vs Regulus, based on the publication record, Santaris seems to be ahead in terms of potency. But considering that ISIS is a nucleic acid chemistry powerhouse and they are apparently working on BNA alternatives, this may change and Santaris would probably do well in vigorously continuing their nucleic acid chemistry efforts if they are not doing so already.
Santaris vs Mirrx Rx…I believe that while they both aim at microRNA inhibition, the difference is more a philosophical one than a technical one: inhibit one particular microRNA-target interaction vs inhibiting all of the functions of a given microRNA which may involve 100s of target interactions. Which one is preferable in principle probably depends on the application. Mirrx, however, has a bit of catching up to do in terms of technical validation.
Jamie, Thanks for your comments. Scalability, of course, is an important attraction of RNAi Therapeutics. There are generally a bewildering range of drug technologies, and some of them look very promising for certain diseases and can be valuable partnering candidates for late-stage drug candidates. For the internal capabilities of a Big Pharma, however, technology investments in platforms which provide for scalability/amortization benefits will be particularly important.
Dirk-
The ISIS "novel" analogues begin with an LNA and then substitute one of the bridge -H bonds with an -R, where R is the 'signature' ISIS O-methyoxy-ethyl residue. I actually believe are they are anticipated by the Wengel issued patents and therefore not even patentable. But if somehow patentable to ISIS, they still will have 'species' claims to Wengel's 'genus' claim for LNA. Unfortunately ISIS own language in the publication where they admit to using locked nucleic acids will make it impossible for them to later claim that this is somehow different chemistry.
Ergo: ISIS will not be able to commercialize without a license from Wengel estate.
So I am back to the conclusion that only Santaris and MDRNA can use bridged nucleic acid analogues.
The ISIS paper does show the potency of 14-mers without any delivery chemistry which is what Santaris has been saying for a couple of years.
In Europe, the Wengel patent has been revoked in opposition proceedings.
The Wengel patent estate consists of some 53 patent families according to Santaris Pharma. Can you be more specific about which patent you are speaking of?
It is the fundamental LNA patent by Wengel:
EP 1015469 B1, BI- AND TRI-CYCLIC NUCLEOSIDE, NUCLEOTIDE AND OLIGONUCLEOIDE ANALOGUES
Santaris 53 patent families (mostly applications) are mostly directed to specific modification patterns, uses, sequences, i.e. more narrow patents.
And many of these are having a tough time at the patent offices.
Interesting! Thanks for the heads up. I will take a look if I ever get free time from current project:
http://www.sec.gov/Archives/edgar/data/1488039/000114420410016530/0001144204-10-016530-index.htm
Post a Comment