Blog Listing

Seres Therapeutics: A Matter Of Details

  • A “surprise” fail on the P2 SER-109 trial caused the stock to plummet 70%.
  • Insider selling began in March – right when enrollment completed for P2.
  • There is one key difference between P1 and P2.

Seres Therapeutics (NASDAQ:MCRB) Postulation

I awoke today to something I absolutely did not expect – the report of Seres Therapeutics down over 70% on bad “surprise” data. It didn’t make sense.

How can you have Phase I data that showed 96.7% (29 of 30) of the patients achieving clinical resolution by week 8 and Phase II data that showed no difference between placebo:

Based on 8-week data, CDI recurrence occurred in 44% of subjects (26 of 59) who received SER-109, compared to 53% of subjects (16 of 30) who received placebo.

Statistically, this just screams impossible. Something must be different in their methodology, otherwise this is a statistical anomaly.

So I dug. The Phase I trial design is not on, but you can find ithere. There are some interesting things in the methods, but let me highlight these points:

  • Two days prior to dosing, patients discontinued antibiotics for CDI.
  • One day prior to dosing, patients underwent a bowel preparation (to minimize residual antibiotic in the gastrointestinal tract), followed by overnight fasting.
  • Two sites used a regimen of 300 mL of magnesium-citrate (one with Dulcolax), and 2 sites used polyethylene glycol.

Summary: Before treating with SER-109, all patients were discontinued from antibiotics, and their gut ‘flushed’ out to remove as much organic matter as possible, then a fairly large influx of SER-109 mix of bacteria:

15 patients who each received 30 capsules of SER-109 (observed dose of 15 capsules on day 0 and day 1). The dose of spores varied between 3 × 107 and 2 × 1010


1 × 108 spores (approximately 17-fold lower than the geometric mean dose administered in part 1 and 3-fold above the minimum dose shown to be effective). Depending on spore content, patients received an observed dose of 1-12 capsules on day 0.

Wiped clean, then given the drug and results = incredible.

Phase II: The trial design is not publicly available like the Phase I is and the site for the SER-109 Phase II trial has nothing on bowel preparation (but does indicate some things about antibiotic use).

The S-1 SEC filing does not address bowel preparation, but does explain something:

  • In contrast, enrolled patients in our Phase 1b/2 clinical study were permitted to be on long-term antibiotic therapy.

Secondly their CDI definition changed, and in fact, makes it even harder to fail in the Phase II:

In this study, an episode of recurrent CDI will be defined as three or more unformed stools per day over two days with a positive C. difficile stool test and requiring antibiotic treatment. By comparison, our Phase 1b/2 clinical study defined an episode of recurrent CDI as three unformed stools over one day with a positive C. difficilestool test and did not require antibiotic treatment.

So not only did they miss their efficacy data, they actually made it easier to beat by defining recurrent CDI in the Phase II trial as an even worse outcome than in the Phase I. The odds are stacked that something is seriouslydifferent between these two trials.

Officially, the company does not have any statement on the bowel preparation in any statement, including their CC following the results. Roger does not answer the question of bowel prep around the 13:00 minute mark.

The bowel preparation is so important to the mechanism of action of their drug – since their drug is meant to replace the microbiome, not kill it or alter it.

This crucial evidence can only currently be found in a Cowen (Ritu Baral) report shown here:

This isn’t official or verified, but the company should have told us if this bowel cleanse/preparation was different. So if the above statement by Cowen is true, this is what our timeline looks like:

So here, if this is true, which everything I’ve read points to this being true, there is a day here where the patients aren’t being dosed and are just… hanging out. Perhaps Cowen means Day 1 = Day 0 of the original publication definition. Or perhaps the above timeline is correct.

There seems to be something here – whether its the bowel preparation or the timeline. With that said, it seems that management probably knew the market was overreacting (overly optimistic) about this Phase II trial…

On May 2nd, 2016, Seres Announced they finished enrollment of their Phase II SER-109 clinical trial.

This is approximately 1 month after all the insider selling begins:

You can see it begins March 28th, and picks up in April (Source: There is not a single buy, and nearly everyone on the management has decided to sell. Hindsight is always 20/20, but things are starting to make sense.

So what’s the thesis here?

Seres Therapeutics knew that the probability for their efficacy data from their Phase II trial would not be good.

And they knew because there was something different in their Phase II trial that would allude to very interesting and important data that could help solidify trial design for 262. Here is our thesis going forward, and we explain the company’s perspective afterwards:

  • If Cowen (Ritu Baral) is correct about using the bowel preparation AND correct about the timeline (Day 1 =/= Day 0):

Extremely bullish: The single day of no dosing, no antibiotics, and no laxative allows for the problem – C. diff. – to bloom in the gut again, it likely would have over 24 hours (night of Day -1 to morning of Day 1) to infect the patients, before SER-109 has a chance to be released in the guy. This highlights the problem of C. Diff. in the first place – how persistent it is, and that it takes advantage when the microbiome is thinned or weakened. Thus, we can expect the difference from Phase I to Phase II to be held within that 24+ hours of “C. Diff. incubation”. This is an easy fix for efficacy in future trials.

  • If Ritu Baral is incorrect on the day definitions (Day 1 = Day 0):

Extremely bearish: Then we have no substantial difference other than the preparation of the SER-109, the fact it was double-blind, and other small details. I believe Roger and management when they explain that the new preparation for SER-109 is more pure and concentrated, and I believe that the difference is not accounted for here. Thus, the change from P1 to P2 results is unclear, and the addition of blinding lead to a loss of their product’s efficacy.

  • If Ritu Baral is incorrect on the bowel preparation (which would be a catastrophic difference for the company to not mention):

Extremely bullish: The company overlooked (and negligently didn’t state) possibly the most important aspect of their pre-treatment, and the results can be easily explained. It’s highly unlikely the company would do this, but if they did, there’s a very obvious reason for it.

Here is their hypothetical perspective if they changed the timeline or the bowel preparation in Phase II (and it’s very clear if you listen to the conference call):

The company did this not for efficacy, but rather to simply study the microbiome from a very academic perspective. Seres have 109, but they have 262, which is easier to produce, manufacture and source and is more controlled and reliable than 109.

So, instead of attempting to show that 109 is efficacious (and lose a lot of information about the microbiome by flushing it out and installing the SER-109 microbiome community), why not see what happens (in humans – not mice this time) when the present microbiome is only slightly perturbed, and SER-109 is simply tossed into the gut to fend for itself?

Why not learn something new, rather than just confirm what the company likely already knows?

After researching this and listening to the conference call, what I’m seeing here is that the company is very aware that the run-up to this Phase II trial was overbought, and they also realized that the market was likely anticipating identical Phase I data results.

In their situation, with a passion to actually understand the microbiome, and the realization that your company is extremely well funded and basically sitting on a gold mine, why not throw caution to the wind and provide your team of scientists with some seriously juicy data? After all, if it fails, you still have SER-262.

With all that said, it will be hard for me to paint a pretty picture for investors when the management said the results were “unexpected” – I don’t believe that is the case at all. Rather, I think the scientific data provided in these results was placed on a higher pedestal than that of efficacy (and shareholder value). Sorry guys!

Finally, you will hear chatter about the change in product formulation/preparation from their P1 to P2 trial. My take is this has nothing to do with the results, without knowing more details, however, it is an open possibility – but I doubt it.

So it all comes down to timeline and bowel preparation. These facts will literally make or break this company.

Disclosure: I/we have no positions in any stocks mentioned, but may initiate a long position in MCRB over the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Leave a Reply