Ionis Pays to License TTR Drug

When it comes to reaping the financial benefits of its efforts, Ionis ranks at the bottom of the industry and last week provided a new low point in this ongoing saga.

Licensing drugs for commercialization purposes is normal in the biotech space, especially when a smaller company lacks the resources to do so.  In return, the licensor typically receives an upfront fee and other milestones in addition to a royalty on sales.

Ionis Pharmaceuticals has just broken with this sacred tradition.  In fact, it ended up giving $200M to Akcea Therapeutics for it to market TTR amyloidosis antisense drug candidate Inotersen for which regulatory approvals are expected this summer.  The two companies will share the profit/loss from the upcoming commercialization of Inotersen and the GalNAc-conjugated follow-on compound in early-stage development.

To put it in simple terms, Ionis is transferring billions of (stock) market value (à Alnylam’s ~$15 billions market cap largely rests on its TTR franchise) to Akcea in return for Akcea's recently established sales infrastructure for which it might have spent $50M.  As I’ve been saying all along: building commercialization capabilities does not involve magic and for orphan drugs certainly don’t require Big Pharma footprints.  All it requires is the will to just do it.

Unfortunately, Stan and his longtime followers at Ionis only feel comfortable playing in their early-stage sandbox and don't seem to really care about creating shareholder value.

No other takers?

  

Stating that they have just transferred billions of stock market value may also be partly wishful thinking. 

Last August, when GSK declined to license Inotersen, Ionis said that pharmaceutical companies had instantaneously started to line up to license the drug.  Then after nothing happened in the coming months, Ionis changed to wanting to keep the US to themselves and licensing rest of world.

And now this: it is ‘licensing’ the drug to its own spin-off company to which it already controlled more than 2/3 of the shares in addition to important veto powers regarding Akcea’s corporate development.

Ionis says that they were forced to give the nod to Akcea because the drugs were racing towards approval and other companies wouldn’t have been able to ready Inotersen for commercialization in time.  This, of course, doesn’t make any sense since why was Akcea more ready to do so?  Couldn’t Ionis have sent its TTR commercial team which it is transferring now to Akcea to just about any other company as well?   

Clearly, nobody was substantially interested in Inotersen and my guess is that this is not due to Alnylam’s Patisiran believed to have much better commercial prospects than Inotersen.  Instead, it is Alnylam's RNAi GalNAc compound which greatly limits the absolute value of both Patisiran and Inotersen as it looks like a vastly superior TTR knockdown drug (~quarterly subcutaneous dosing, much greater knockdown) and may be approved within the next 2 years already, much earlier than Ionis' GalNAc follow-on.  

  

Dicerna Supremely Confident About Clinical Pharmacology of its GalNAc Platform

Dicerna has just entered its GalNAc-RNAi conjugates into the clinic and by the looks of it, seems extremely confident about being able to forecast its performance in humans.  

Single-dose study

At least this is what I am forced to interpret into their remarkable decision to merely conduct a single-dose phase I/II study with DCR-PHXC for primary hyperoxaluria before going straight into a planned (multi-dose) pivotal study in 2019.

There is, of course, precedent from Alnylam’s ample experience with a related GalNAc-RNAi conjugate format and how animal studies have translated into humans and how repeat-dosing in man have increased and extended the knockdown compared to single dose administrations.

 

Dicerna thus believes it will be able to predict the optimal dosing schedule for the all-important registrational study based on single-dose monkey-to-human translation and the effect of repeat dosing in monkeys.  

You’ll probably be scratching your head already how accurate such modeling by triangulation can be.  Complicating matters, the primary aim is not to achieve a predetermined level of target knockdown (lactate dehydrogenase A/LDHA), but in fact oxalate knockdown which is downstream from LDHA.

So a lot of moving parts between a single-dose gene knockdown and therapeutic lowering of a toxic metabolite following prolonged and pronounced multi-dose knockdown.  And don’t get me started on the impact of RNAi trigger formats, the nature of their chemical modification as well as the target gene identity on gene silencing duration…and what about the small issue called 'safety'?

Dicerna: if this rushed design is an effort to catch up with the competitive program by Alnylam to better compete for patients for the pivotal trial, it’s probably not worth it.  Either you are greatly increasing the risk of phase III failure or, more likely, you'll be sent back into earlier-stage multi-dose studies by regulators (similar to what happened to the Alport program by Regulus Therapeutics).

Dicerna talking down value of mystery program

Following DCR-PHXC, Dicerna plans to bring two additional drugs into the clinic in the near future: one against HBV and another one for an undisclosed orphan disease.

Aside from the fact that I believe that there is little point in keeping the target identity a secret since they have pretty much given it away by characterizing it as one with >100,000 patients in the US alone (à alpha-1-antitrypsin-related liver disease), it is remarkable that the CEO has said that they are looking for a ‘risk-sharing’ partner at this early stage already.

Even more surprising was the comment last week that Dicerna will even wait to partner the program before entering it into clinical development!  

This truly is unheard of for a company in the red-hot genetic biotechnology space where funding is relatively easy to come by these days for a company of a profile like Dicerna (clinically ready genetic technology, $700M market cap).  Here, the universally accepted name of the game is to get at least a couple of drug candidates into the clinic with minimal ownership dilution (e.g. by partnering) early on.  Developing an orphan candidate to clinical readiness and then idling it sends out a clear signal that the candidate is deemed to have disappointing prospects.

 

It would be easy for Dicerna, with the stock up 100% since the recent offering, to do another ~$100M raise to comfortably navigate three programs through proof-of-concept.

With these two unexplainable apparent unforced major errors, it got me thinking: can it be explained with the uncertainty around the trade secret litigation with Alnylam? According to the Q4 2017 conference call, a trial date has been set for April.

Commonly Used RNAi Trigger Modification Can Integrate into Genome and Cellular Transcripts

To endow RNAi molecules with drug-like properties, they need to be modified (for stabilization, immuno-silence, RISC incorporation).  Modifications commonly used these days include 2’-O-methyl and 2’-fluoro (2’-F) modification of the ribose sugar ring and phosphorothioation of the phosphate connecting the constituent nucleotide monomers.

A study by Ionis and AstraZeneca scientists now shows (Saleh et al 2018) that 2’-F nucleosides are readily incorporated into RNA polymerase transcripts and the genome in tissue culture cells. By contrast, 2’-MOE nucleosides, a modification that Ionis chiefly uses for its antisense oligos, was highly refractory to such incorporation under the same conditions.

So regardless of the political motivation behind this publication- Ionis likes to paint the 2’-F modification used by competitor Alnylam as dangerous, whereas Alnylam likes to say same about Ionis’ phosphorothioates- the fact that turnover products from RNAi trigger degradation may be used in this way raises genotoxicity concerns that need to be taken seriously.

Even if minor degrees of 2’-F incorporations into transcripts and genomic and mitochondrial DNA turned out to be harmless, not undertaking the appropriate studies could catch companies in the space on the backfoot when regulators suddenly demand them.

It is possible that RNAi bellwether Alnylam indeed has responded to this concern as they have taken to minimizing the 2’-F content in their latest generation of GalNAc-conjugates while increasing 2’-O-methylation.  Although Alnylam justified this change with wanting to further increase the stability and thus longevity of the gene silencing, in light of twice annual administrations already possible with the old format (see inclisiran for PCSK9 lowering) and increasing 2'-O-methyl content making it harder to find intrinsically potent molecules, this move had me wondering whether it had actually to do with toxicity concerns instead.  This paper would support this notion.

Antisense Technology Produces Huntington’s Disease Breakthrough

Huntington’s Disease (HD) is one of the most prevalent severe, ultimately fatal genetic neurological diseases of our times for which impactful treatments are desperately lacking.  Based on results from a clinical study aiming to knockdown the disease-causing gene using RNaseH antisense technology, this is about to change over the next 3 years.

IONIS-HTTRx study results presented at CHDI

At the major annual HD gathering this week, lead investigator Tabrizi from the University College London presented data from a phase I-II study of antisense drug candidate IONIS-HTTRx in patients with early symptoms of the disease.

IONIS-HTTRx (aka RG6042), developed by Ionis and partnered with Roche, targets both wild-type and mutant huntingtin protein to provide all patients with the most potent and safest sequence possible.  While mutant allele-specific approaches have been discussed and are being pursued by antisense rival Wave Life Sciences, the concern that suppressing wildtype huntingtin in adults seems to me a far-fetched theoretical discussion, if not wishful thinking given that antisense is unlikely to completely eliminate wildtype huntingtin and that the drug is being given to adults.  

To wit, the concern is based on genetic mouse models where complete huntingtin knockout causes embryonic lethality. Importantly, this is not recapitulated in mouse model where knockout is delayed until adulthood.

Call me cynical, but the choice of an allele-specific approach by Wave Life Sciences has likely been primarily motivated by allowing for some follower differentiation.  Ironically, by limiting themselves to a much reduced sequence space around co-segregating single nucleotide polymorphisms, the sequences may not only lack the potency, but also the apparent safety of IONS-HTTRx.

Good safety profile maintained until highest dose tested

In this study, IONIS-HTTRx was given monthly over 3 months.  Speaking to the safety of the approach, apart from transient side effects due to the intrathecal mode of administration, no adverse events of note were reported and all 46 subjects have subsequently rolled over into the open-label extension study.

Importantly, doses as high as 120mg monthly were well tolerated at which dose the knockdown effect seems to have plateaued.  This not only bodes well for IONIS-HTTRx, but for the entire neurological disease franchise of Ionis Pharmaceuticals. 

The human safety and tolerability results are consistent with a range of small and large animal models (including dogs, pigs, non-human primates), some of which were a year or longer in duration.

Robust huntingtin lowering

It is widely thought that mutant huntingtin protein is harmful to the cell expressing it, ultimately resulting in neuronal death throughout the brain.  This can be visualized by considerable, 30%-level atrophy of the brain over the course of the disease.

Although it is not possible to assess huntingtin levels directly in brain tissue in humans from biopsies, the experience in the preclinical animal models and human data from spinal muscular atrophy drug SPINRAZA allows one to model the correlation between ASO-dependent protein changes in the cerebrospinal fluid (CSF) and the various areas of the CNS quite well.

Since SPINRAZA is based on chemistry that is highly similar to that of IONIS-HTTRx, they behave identically in terms of biodistribution for all intents and purposes.

In the two highest cohorts, 90mg and 120mg, mean 40% mutant huntingtin reductions in the CSF were observed shortly after the last dose with further reductions (50%+) seen thereafter.  Such levels are predicted to reflect 55 to up to 85% reductions in the cortex and 20-50% in the caudate.

This degree of knockdown exceeds that necessary to see changes in HD animal models, some of which with manifest disease symptoms at treatment onset.

The future is bright

The study results therefore raise hopes that  IONIS-HTTRx may halt or even reverse disease in manifest HD patients.  As exciting and medically likely with the biggest bang for the buck is the prospect that the drug candidate may prevent disease symptoms to emerge in the first place when given early to those at risk of developing the disease (~1 in 2 with a parent having HD, 200,000 in the US alone). 

The companies and clinical investigators are now busy designing a pivotal outcomes study which I would guess will ultimately involve 150-250 early-stage patients over a period of 16-24 months.  The size and extent of the trial will depend on which endpoints are chosen.  A pre-planned interim look at ~12 months may also make sense to look at brain size as a potential way to obtain accelerated approval.  Accelerated approval under the new FDA may also be possible based on functional analyses the investigators are currently conducting using data from the present study and the open-label extension with dosing well beyond 3 months..

In drug development, sometimes you can tell early on whether a drug candidate is destined to breeze through to approval. In this regard, IONIS-HTTRx certainly feels like Spinraza.