RXi Provides Disappointing Clinical Update for Scarring Drug

At the Rodman & Renshaw investor conference today, the CEO of RXiPharmaceuticals dropped a little bombshell in the form of a disappointing clinical update on their lead clinical candidate, RXI-109 for the treatment or prevention of dermal scarring.  In that interim look for efficacy, RXI-109 and placebo were not really distinguishable in scar severity sending the stock down 30-40% in the middle of today’s trading session.

1301 study design

The 1301 study is the first of three phase IIa studies evaluating 109 in a number of different scar settings.  In this case, RXI-109 was administered following scar revision surgery on the lower abdomen.  Part of the same scar received either 3 injections of the self-delivering RNAi compound, another part placebo solution.  Half of the subjects (50% of the study/16 subjects have enrolled as of today), received 109 on days 1, 8, and 15 following surgery (cohort 1), the other half received injections on days 14, 21, and 28 following surgery, the latter apparently inspired by clinical design trends observed for competitor antisense drug from Pfizer/Excaliard.

I find that blindly adjusting your clinical design based on such competitive intelligence is a worrisome sign of lack of confidence.

1301 study results

The first interim look for efficacy took place at 1 month post-surgery.  Results were based on the blinded visual assessment of scar severity on a scale from 1 (good fine-line scar) to 10 (worst scar imaginable). 

Unfortunately, the close to 50% knockdown of target CTGF observed in a similar 3-dose phase I study, did not translate to an obvious improvement in scar severity: for the immediate treatment group, the VAS score was 2.0 for both the 109 and the placebo side of the scar; for the delayed treatment group, the VAS score was slightly better in the 109 side (2.0) than on the placebo side (2.5) even reaching statistical significance.

Before you get excited and buy into the biological rationalizations by the CEO for why it makes sense that delayed, but not immediate would exhibit such a benefit, note that a VAS difference of 0.5 on a scale from 1 to 10 would appear to be clinically meaningless despite the statistical significance.  Moreover, looking at the VAS scores across the board, it seems that the placebo side in the delayed treatment cohort is a statistical outlier as it should not have performed any different than the placebo cohort in the immediate treatment group.    

On the other hand, given that the scars have not had time to fully develop and the VAS scores were still so low, it is possible that real differences will emerge at later time-points such as month 3 when the next interim look will take place and it may thus be premature to declare the study or drug for that matter a failure.

RXi needs a plan B and investors patience

In light of the disappointing trial update, it may not be a coincidence that Geert Cauwenbergh today took the opportunity to talk more about their other, preclinical pipeline candidates in the dermatology and ophthalmology space in much more detail than had been the case.   

I am less excited about the prospect of RXi expanding their dermatology footprint given the relatively modest gene knockdowns observed, limited tissue penetration from the site of injeciton, and the cosmeceutical nature of their current line-up (acne, depigmentation etc).  By contrast, I am much more excited about  the prospect and value of self-delivering RNAi triggers in the ophthalmology space given the great unmet medical needs there and the highly encouraging tissue penetration/biodistribution data for self-delivering RNAi in that organ.

There will be more to talk about that in the future.  As stock market investors, however, one has got to wonder whether management is aware of that value and knows how to best exploit it (pro tip: VEGF is a no-no for RNAi in the eye).  Even more concerning is the fact that the majority shareholder (Tang Capital) still holds close to half of the company and is in the process of unloading it thereby putting constant pressure on the stock. And with only $10M in the bank, you know what the trip to the Rodman & Renshaw conference was all about.

So no more than a small starter position for me despite the steep sell-off today.

Antisense Comparison Provides Hope for RXi’s Dermal Scarring Drug Candidate

In early December, RXi reported phase I data from new cohorts that were added to the multi-dose study of RXI-109 in dermal scarring.  Accordingly, by increasing the dose to 10mg self-delivering RNAi trigger from previously 7.5mg per injection site, the company was now able to achieve a 50% target CTGF gene knockdown (5 and 7.5mgs: 43%). 

Based on the meager 30-40% knockdowns and apparent efficacy that were observed in comparable clinical studies targeting the same CTGF with a phosphorotioated antisense compound at 5mg per cm scarline_­, the prospects of RXI-109 would suddenly appear to be much brighter.  Of course, the ISIS compound was spun out into Excaliard which was then acquired by Pfizer in late 2011 for the apparently promising data observed with the dermal scarring candidate EXC001.  EXC001 is now in late-stage clinical development.

Of course, there are a number of caveats with this reasoning.  For one, although the study protocols appear very similar, it is possible that the tissue biopsies taken to obtain the CTGF knockdown measures were of dissimilar sizes.  Given that CTGF knockdown can be expected to wane quickly away from the injection site, such differences could have a material effect on the apparent knockdown efficacy. 

On the darker side, when you consider a nice visual therapeutic effect (see picture) in light of a 30-40% knockdown, you start to wonder whether the actual effect on scarring was less due to blunting of CTGF expression and more due to some non-specific immune-related effect of the phosphorothioate backbone in the antisense compound.  The RXI-109 compound may not ‘benefit’ from such an effect.

Nevertheless, the 10mg dose results are a step forward and form a useful basis for the phase II studies that RXi Pharmaceuticals will be rolling out this year in lower abdominal scar revision (already initiated), keloid scar revision, and scar revision following cosmetic breast surgery.  

Unfortunately, prospects would have been even brighter had the company employed a more potent RNAi trigger design.  And when the CEO, in 2014, still shows a slide with Kreutzer-Limmer controlling dsRNA lengths of 15bp and over (to justify the use of dsRNA < 15bp), I would suggest they update their presentation slides.  Similarly, it would be honest to not talk about a market cap of $50M, but to present their more meaningful capital structure, including preferreds and the like.

RXi Reports 43% CTGF Knockdown in Multi-Dose Dermal Anti-Scarring Trial

Today, RXi Pharmaceuticals announced the results from its multi-dose phase I study of RXI-109, the company’s self-delivering RNAi compound for dermal anti-scarring.  Importantly, in the two highest of the three dose cohorts a credible 43% (average) gene knockdown was observed three days after the last (=third) intradermal injection of RXI-109.  It is the first time that a knockdown was reported for the so-called ‘self-delivering’ class of RNAi triggers.

The results followed those from a single-dose study lastmonth where dose-dependent gene knockdowns were claimed three months after the single injection (note the difference in the time points).  Turns out that this was a slightly misleading conclusion as in my book a numerical 15% target reduction does not constitute a clinically meaningful knockdown for the vast majority of target genes and indications, and I'm not even discussing the precision of gene expression measurements.

Whether a 43% knockdown of CTGF is clinically meaningful also remains to be seen as no data were presented on the actual impact of RXI-109 on scar formation.   Pfizer, following its acquisition of dermal scarring antisense company Excaliard, would probably know best what type of knockdown was required.

   

In a broader sense, the 43% number also raises the question of whether self-delivering RNAi triggers by RXi Pharmaceuticals will be a class of gene silencing agents that will struggle to achieve 50% gene silencing, instead of 70, 80, 90% and more that might be required for most indications.

Overall, mediocre results and it stands to reason that the future of RXi Pharmaceuticals will be in ocular indications and not in dermal anti-scarring.

Comment on Alnylam's $3B market cap

I, like many of you, have watched with wide open eyes Alnylam reaching a $3B market cap today. In less than two weeks, this company added over $1B in valuation based on highlighting in their press releases the best single datapoints from individual patients (e.g. 'over 80% knockdown' for ALN-TTRsc), instead of average knockdowns, area under the curves, and dosages.

While that does not entirely surprise me as a veteran of reading between 'topline data', a more intriguing question is what the company will do with such a low cost of capital.  Remember, the situation was similar about 5 years ago when Alnylam failed to either raise capital and/or acquire Tekmira to avoid the litigation.  I expect the company to act this time on its share price, a view supported by constant analysts upgrade on any news piece the company throws in front of them (--> fees for investment banking business).

RXi Reports Dose-Related Knockdown Three Months Following Single Injection

RXi Pharmaceuticals today reported top-line results from the first of two phase I studies with RXi-109 in dermal scarring.  Intriguingly, the company claims to observe target gene knockdown three months following a single intradermal injection of their self-deliverable RNAi trigger (p=0.02) in a manner that was apparently dose-related. 

Such a drug-dependent and dose-related knockdown would exceed my expectations from this trial as stated in a recent preview here.  The reason why I merely expected to see a correlation between CTGF levels and phenotypic effect on dermal scarring being reported today is that I did not have high confidence that the tissue residence time of RXI109 would be prolonged enough to observe a bona fide RNAi knockdown. 

To wit, the tissue biopsy on which this data rests was obtained during a tummy tuck three months after the intradermal injection of the RNAi.  To assess whether there was an RNAi knockdown in such a single-dose study, I would have thought that an early time-point such as two weeks after injection would have been more appropriate, also because it is likely that some of the CTGF-producing cells might be proliferating in this setting (RNAi duration inversely correlated to proliferation status).

The notion that a correlation between CTGF levels and wound healing would be observed was based on CTGF reflecting tissue inflammation.  So regardless of whether there was an RNAi effect or not, you might expect to see such a correlation.  It is difficult, however, to explain a drug-dependent and dose-related target gene knockdown with this notion, except for by a rare coincidence.

These results then bode well for the multi-dose phase I studies from which results will be reported in time for the Investor and Analyst Symposium on July 12.  While safety was the primary focus of the first study (no adverse event on the early wound healing process confirmed), the effect of RXi-109 on wound healing will be the focus of the second study.  It should be added though that given the small size of the trials and the patient population which is not predisposed to scarring, spotting the difference will be tough.

Disclosure: long RXII.

Can RXi Pharmaceuticals Spot the Difference?

The imminent announcement of phase I results for RXI109 will be a clinical highlight of RNAi Therapeutics in 2013.  RXI109 is the self-delivering RNAi trigger against dermal scarring and is developed by RXi Pharmaceuticals.  Needless to say, as the company is committing 90% of its resources to this trial and indication, the results should cause major volatility in the stock.

While the dermal anti-scarring landscape is complex, RXI109 for the present indication can safely be categorized as a cosmeceutical.  According to RXi Pharmaceuticals, already $100M are spent each year in the US on non-FDA approved ointments against dermal scarring.  The interest in RXI109 is thus for its commercial potential and the clinical results are a milestone in the development of so-called ‘self-delivering RNAi triggers’.

Self-delivering RNAi triggers are a concept coined first by Dharmacon (although one could argue that was largely a branding achievement as it essentially involved known cholesterol conjugation), but is getting more widely adopted these days.  Beyond its local indications for which they self-delivering RNAi triggers were initially developed, I expect the concept to also be applied to certain systemic delivery strategies.  I could imagine that in an effort to render GalNAc-siRNAs more potent, self-delivering chemistries will be useful.

Phase I studies

RXi has conducted two phase I studies.  In both studies, volunteers got multiple surgical incisions symmetrically on both sides of the abdomen.  For each pair of incisions, one side either received RXI109 or placebo by intradermal injection.  In the first study, RXI109 was given just once before incision, from 1mg to 10mg per 2cm incision (similar range as in the Excaliard antisense trials).  In the second study, RXI109 was given three times within two weeks from 2.5mg to 7.5mg per 2 cm incision.

In addition to safety and tolerability, the important endpoints will be a visual assessment of scarring and then, based on a biopsy obtained from a tummy tuck at Day 84, important biomarker data in the form of CTGF levels (the target gene) and a histological evaluation of the scar tissue.

Although this is a blinded study, the company has discussed blinded results in extenso.  On the safety front, there seems to be little cause for concern, and adverse events are consistent with what you would expect from an incision.  Management appears to be very bullish on the therapeutic outcomes since in many cases left and right sides look different.  So if they are different, the side that looks better should have been given RXI109, right? 

Unfortunately, you have to look very hard to spot the differences.  In one example shown, the ‘average differences’ in scar tissue area were 31%.  Since they will put their best foot forward with this example, the largest effect size that we can expect is 31%.  And this assumes that in each case, it is the drug-treated side that outperforms the placebo-treated side. 

It is thus difficult for me to be optimistic that this trial allows for a therapeutic effect to be demonstrated.  For this, the natural wound healing variability would have to be really small (I admittedly don't know what this is).  On the other hand, it is with this symmetrical, intra-patient control design that such small differences might be teased out.  Regardless, I expect enough data to be collected from the studies that it will make for a nice headline and narrative about how RXI109 had improved wound healing and the correlation with CTGF (I bet there will be a correlation, whether due to knockdown or not).

So while I think that RXI109 is a decent RNAi Therapeutics (not the best possible one given the short dsRNA length), it will be important to conduct future studies in patient populations more prone to scarring to increase signal to noise.  This could be for example in the scar-revision setting or in Asian populations.

   

Trading the event

As I expect major volatility and have some confidence in the science behind RXI109, I have taken a long position ahead of the event.  It is not clear whether the results from the two studies will be presented separately or together.  I suspect the latter given that the CEO of RXi in February/March guided the results from the first study to be forthcoming in April.  Since it is almost June already, it is likely that the company expects the biggest bang from presenting the results together.  This should happen before July.  Once again, given that there is so much potential for data-mining, I expect positive headlines- justified or not.