Invion, where is the data?

On August 17, Invion (ASX : IVX) provided an update on their INV103 program (recombinant Cpn10/Hsp10) in Systemic Lupus Erythematosus (SLE). Unfortunately, I got side-tracked by other things and haven’t had the bandwidth to comment on this announcement before now. I have previously provided detailed commentary on this company and will not be reviewing the technology in depth in this post.

As always, we can expect – irrespective of the quality of the data or the meaning of the outcome – a positive and enthusiastic soundbite for the retail investor. You know, the ASX disclosure equivalent of a little shareholder cuddle, just to let you know that everything is fine, even though the data is quite evidently not “fine”:

Executive Vice President R&D and Chief Medical Officer, Dr Mitchell Glass, said, “We are pleased that the data from this trial support the dose escalation that has underpinned the strategy for ala-Cpn10.

In a way this statement is true, in the sense that the company did not report any major adverse events from ala-Cpn10 (INV103) and so it would appear that the higher dose regimen was probably tolerated. Given what we know about innate serum levels of hsp10 in different disease states, this is a useful piece of knowledge for the program, if it is indeed true. It should be noted that there is a very different risk profile in a chaperone like hsp10 administered at elevated levels for 4 weeks, compared to long-term use (that may have an increased associated cancer risk), which the company somewhat acknowledges. Clearly, however, we don’t develop drugs just for safety, there has to be some efficacy too.

There are three major concerns for the retail investor in IVX’s announcement:

Where is the fourth cohort?

Consistent with the clinicaltrials.gov entry (NCT01838694), the announcement refers to four (4) cohorts (second paragraph). However the data presented in the press release stops at 100mg which is not the fourth cohort, it’s the 3rd cohort. As far as I can see no data has been disclosed for the fourth cohort, which seem strange to me. Although the clinicaltrials.gov entry is one of the most poorly descriptive trial summaries I have ever seen, it does imply 4 cohorts of 10 patients each, which would (back of the envelope) be the minimum number of patients you would need to demonstrate statistics for placebo-controlled study like this. Either the company stopped at 7 patients in each cohort, which is an odd decision, or the company abandoned the last part of the dose escalation. Neither is a particularly good sign given the expense and difficulty of recruiting patients into a study like this.

To me, the original scope of the study as articulated would be the minimum amount of data you’d need to get interest from a potential co-development or licensee, assuming that the results were encouraging. Why, for the sake of a handful of patients (and given the sunk cost in the bioprocess development, expense of setting up the trial infrastructure for a study like this, etc.) would the company choose to do only part of the job? Moreover, it was a blinded study so the company would have only (possibly) been able to see pooled statistics. Either you start to see clear clustering (hard to imagine with such a small number of patients per cohort) or you don’t see any significance and you pull the plug. In this respect, the key words are:

Serum biomarkers of vascular inflammation were too variable in all cohorts to draw absolute conclusions about biological effect.

Why are you reporting only two time points?

This was supposed to be a study to capture the pharmacokinetics and it is my understanding that there were 9 patient visits over the course of the treatment. Presumably patient blood workup would have been done as part of those visits in order to measure serum levels of INV103 (from the previous dose), as well as take peripheral blood mononuclear cells (PBMCs) for measurement of the various inflammatory biomarkers used as a readout of drug response. With a multi-dosing study like this, to make the statistics robust, you want to measure the PBMC biomarkers as a function of each dose to see whether or not the patient is experiencing a sustained effect or not. Yet the company only choose to comment on baseline and the final dose. There is a huge amount of critical information missing in the middle.

Although there is scant data on the serum half-life of INV103, various papers compare the serum half-life of a variety of N-terminus modified recombinant Hsp10s and report a range from 4-7 days, which supports the twice-weekly dosing plan for this Phase II study. However in order for IVX to have improved their understanding of the kinetics, they would have to take those intermediate blood samples and I would be stunned and amazed if the PBMC workup wasn’t done at the same time. If it was, it isn’t reported, and if it isn’t reported you should be asking why.

What about all the other biomarkers?

In company presentations as recently as July and August, a panel of biomarkers are discussed as readouts of clinical efficacy. These biomarkers included IL-6, TNF-α, ICAM-1, MCP-1, VCAM-1 and urinary protein. The adhesion molecule biomarkers (particularly ICAM-1) are a pretty robust read-out of immune suppression in PBMCs in this disease model, as is IL-6 and TNF-α, but this data is not reported in the announcement. Again, you have to question why such a selective few biomarkers were reported? Also now that the data has clearly been unblinded, why is the company only reporting mean level changes (for example of IL-6) for each cohort? We want to know what the spread is and how it compares to the range seen (overlap) with the placebo arm, because SLE is a highly variable disease and we want to see that variability in both the treatment and control arms.

To conclude…

Whatever the reason, the company cut this study short, and it wasn’t because the data was a slam-dunk. There is more missing from this announcement, including apparently a whole patient cohort, then there is actually reported and what is reported doesn’t really support the dove-ish ideas of bigger studies or partnering. I also feel that with only 70% of the patients recruited, and selective and ambiguous results reported, IVX has done a poor job of demonstrating accountability and outcome to shareholders. I would argue, based on the information presented, that this asset is in fact not even remotely close to being suitable for partnering and I don’t understand how this statement can be made:

we are pleased at the consistency of the responses

Consistently bad? Consistently ambiguous? Consistently inconsistent? Consistently…. what? Please tell us.

The truth is, the only “investor” logic you need to apply to this company is as follows. This is a company that has about a ~$15m market cap. It just finished a “Phase II” (really a Phase IIa) study in a disease with a major unmet need for good drugs, and with a biologic that was truly non-trivial to produce (the bioprocess is pretty impressive). If the results were really exciting, you would have seen the company jumping from the rooftops. They aren’t.

2 thoughts on “Invion, where is the data?

  1. Lol Chris – I think you should frame the IVX results announcement. It really is a classic example of how important it is to be able to read between the lines. And read quickly.

    Yep the trial went pear shaped.

    And when put in context of the overall position it means the company is well and truly in the badlands.

    And lo and behold like clockwork a couple of days after the announcement a chunk of the company is given away at rock bottom prices to raise a trivial amount of money and the share price tumbles.

    Well spotted.

    Like

  2. Pingback: Ouch, Invion… | The Long Tail

Say something useful (or at least interesting)...

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s