Another banger
$ABVX Scorpion called Abivax a fraud. To believe that requires the reader to completely dismiss the very large clinical dataset and/or believe this is the greatest conspiracy in the history of medicine by an order of magnitude. Show the falsified records. Show the cancer excess at the dose you’re attacking. Explain why blinded readers saw healthier colons on both scopes and biopsies. The evidence Scorpion made public falls well short of its headline, and they themselves tell you that they've already covered their short position by the time you read through the report. scorpioncapital.com/
Start with 50 mg dose.
In June, we estimated 0.59 cancers per 100 patient-years at 50 mg, excluding non-melanoma skin cancer. We kept the doses separate because 50 mg was the dose under scrutiny. Abivax’s July deck put the updated number at 0.64 across 944 patient-years at 50 mg, within its historical ulcerative colitis range of 0.30–0.70. Non-melanoma skin cancer also came in at 0.64, below the company’s 0.70–1.40 benchmark. Those are the 50 mg numbers. teapawt.net/deathtouch2k/status/20… · ir.abivax.com/static-files/b…
Yes, the narrower Phase 3 maintenance analysis came in higher. Four cases over 438 patient-years gave you 0.91. The added Part 2 data had two cases over 288 patient-years, or 0.69. One patient had a platelet count over 900,000 at baseline and was later diagnosed with a blood-cell cancer. The other was diagnosed with prostate cancer after roughly three months on the drug. Investigators assessed both as unrelated. Then you get to the full 50 mg dataset and the rate is 0.64. If your case depends on the smaller slice, explain why the rest of the exposure should count for less rather than ignoring it because it's inconvenient to your short report credibility. sec.gov/Archives/edgar/data/…
Scorpion’s UVA experiment is a lab measurement of light absorption. The public trial protocol already disclosed animal phototoxicity for obefazimod’s main breakdown product and required sun precautions and skin monitoring. The observed 50 mg skin cancer rate across Phase 2 and 3 still came in below the historical benchmark. That comparison leaves room for a rare or delayed risk, but again no evidence. It makes the sweeping cancer claim a lot harder to sell AND physicians are already screening for skin cancer risk due to OTHER STANDARD OF CARE DRUGS THAT CAUSE CANCER IN THESE PATIENTS AND CAN CONFOUND THE SKIN CANCER FINDINGS IN EXPERIMENTAL DRUG TRIALS. scorpionreports.s3.us-east-2… · cdn.clinicaltrials.gov/large… · ir.abivax.com/static-files/b…
Now explain the biopsies.
Two blinded Phase 3 trials randomized 1,275 patients. At 50 mg, eight-week remission was 21.7% vs 2.5% on placebo in one trial and 19.8% vs 6.3% in the other. The 25 mg dose missed the main endpoint in the second trial. The 50 mg dose hit it in both. sec.gov/Archives/edgar/data/…
Then 580 patients who had responded to obefazimod were randomized again. After 44 more weeks, 51% were in remission on either dose versus 10% after switching to placebo. A normal-looking colon on scope? 42% at 25 mg, 48% at 50 mg, 10% on placebo. Improvement confirmed by both scope and biopsy? 50%, 57%, 10%. If widespread gut-cell damage is your explanation for how this drug works, explain why the tissue looked healthier under the microscope. abivax.com/wp-content/upload…
And the evidence goes back further. All three doses improved the main disease-activity score in the 254-patient Phase 2b trial. In the open-label follow-up, 114 of 217 were in remission at two years, with missing results counted as failures. In a separate, selected group that stepped down to 25 mg, 88 of 130 were in remission three years later. That’s a substantial record of follow-up on top of the three blinded Phase 3 trials. sciencedirect.com/science/ar… · academic.oup.com/ecco-jcc/ar… · sec.gov/Archives/edgar/data/…
“Foreign trials” is doing a lot of work here.
The Phase 3 trials enrolled across 36 countries. Eastern Europe accounted for 38% of enrollment. In Western Europe, 50 mg remission was 19.0% vs 5.6% on placebo. In a smaller exploratory North American group, it was 8 of 62 vs 0 of 36, with scopes also favoring treatment. The protocol called for independent central readers who were blinded to treatment and visit order. If Scorpion has evidence of misconduct, show the sites, the records and what happens to the results when those sites are removed. sec.gov/Archives/edgar/data/… · academic.oup.com/ecco-jcc/ar… · abivax.com/wp-content/upload… · cdn.clinicaltrials.gov/large…
The “150-fold” experiment used zero cells. ZERO CELLS. Why are we relying on experiments that don't even include living cells when we have thousands of patient years worth of safety data?
Scorpion measured about one-fifth of the nominal drug concentration in a cell-free reconstruction, then applied that result to cell data from 2015 to get the 150-fold headline. That raises a question about an old assay. The DDT comparison is based on fat solubility. In people given 50 mg, the parent drug’s plasma half-life was about one hour and its main breakdown product’s was about four days. If the claim is dangerous accumulation in human tissue, show the tissue exposure. scorpionreports.s3.us-east-2… · scorpionreports.s3.us-east-2… · journals.asm.org/doi/10.1128…
And the supposedly fictitious biology was measured in patients. Both doses raised miR-124 in blood and colon tissue versus placebo in Phase 3, following the same finding in Phase 2. The Phase 3 analysis included 673 paired sigmoid biopsies, alongside paired rectal biopsies and blood samples. Scorpion can question the old binding assay. It still has to deal with the separate structural work and the biological change measured in actual patients. scorpionreports.s3.us-east-2… · academic.oup.com/ecco-jcc/ar… · pmc.ncbi.nlm.nih.gov/article… · abivax.com/wp-content/upload…
The cancer cases deserved a serious look in June. We did the math then, and the expanded 50 mg data landed right within our estimates of background cancer rates in ulcerative colitis patients. Scorpion has a set of lab hypotheses and a fraud headline and completely ignores the fact the drug has show ZERO EVIDENCE OF ANYTHING THEY CLAIM IN THOUSANDS OF HUMAN BEINGS AND THEREFORE IS MOST LIKELY COMPLETELY WRONG.
Abivax has two positive 50 mg induction trials, a positive randomized maintenance trial at both doses, and biopsies that show the colon getting healthier. If you want to overturn that record, bring CLINICAL EVIDENCE that can explain it.
1
6
1,007
Again will have more later but able to listen to most of the call on way to hospital, and just great data from $PEPG
10mg / kg looks to be a best-in-disease drug, regardless of what happens at 12.5mg / kg. Functional markers all improving, the CLCN1 data was great data to see
As noted previously albuminuria is a known benign feature of ASO class due to reabsorption competition with the megalin receptor, amongst others, in the proximal tubule. If you don’t see an SAE hypomagnesemia or eGFRcys decline it shouldn’t be a concern.
Also, evidence dating back to NHP studies suggests if a problem is going to appear it appears early in dosing
See below for rough metrics of competition value from this summer (with Avidity now having failed its Phase III, best splicing improvement they ever showed was 17% and best dyne splicing improvement they ever showed was 27%).
$PEPG is currently trading well below this. In fact, it’s still trading below its peak after 15 mg / kg SAD!
Replying to @WaveStrumLS
Select Comps:
$RNA: $4.2Bn allocable to DM1
$DYN: $2.1Bn allocable to DM1
$RNA will have a control premium baked in that $DYN does not. However, given $PEPG 's single asset nature and institutional backers, I expect a buyout if they show best-in-disease data in the higher MAD cohorts, the way they did in the SAD cohorts.
$PEPG's FD market cap is currently ~$136M.
Avidity will read out its Phase III trial in 2H 2026.
Dyne will read out its registration trials in Q1 2027 with a confirmatory to follow.
This is a fantastically stupid take on $MDGL.
Merck characterized Vioxx's launch in its 1999 annual report as: “Vioxx: Our biggest, fastest and best launch ever.”
How did that turn out?
The speed of a commercial launch says effectively nothing about latent toxicity.
I am out of $MDGL as of this morning. Maybe the whistleblower claim proves to be without merit. However, it should be taken extremely seriously until such time as $MDGL provides evidence for that.
Biography of the whistleblower below, btw.
Anyone attempting to claim the complaint is bullshit a priori has absolutely zero basis for doing so until more information comes out.
Lastly, if $IVA hits its Phase III on the heels of this, $MDGL will be pummeled. Whatever you think the fair value of $MDGL is, pure trading dynamics alone will drive it into the ground.
When I looked at $MDGL earnings, I saw there was over 55,000 patients on the drug up about 100% from last year. If there was any safety issue with this drug, we would all know about it by now. This is just fearmongering to create headlines.
$ABOS
Further evidence that the beta-amyloid hypothesis as the cause for Alzheimer's Disease is bogus.
Statistically significant reduction in plaque burden at weeks 36 and 72, and yet a total failure across all endpoints. Plus, 10% of treated patients developing ARIA-E's.
No more beta-amyloid drugs. Waste of time and money. Never going to fundamentally change what it means to have Alzheimer's.
It remains very unclear whether donanemab and lecanemab have any benefit at all over the long-term (pubmed.ncbi.nlm.nih.gov/4260…).
Remember, the Phase III trial of donanemab simply showed an iADRS decline of -10.2 vs. -13.1 for placebo on the 144 point scale, with the -2.9 delta being far below the 9 point threshold that is considered clinically meaningful in early-stage Alzheimer's Disease. This was masked by the PR's claiming there was a 22% slowing, which might sound like okay at least it's something, until you understand the actual numerical difference.
If there is any minuscule benefit to these beta-amyloid therapies, it's likely going to be highly transient, confined to those therapies that can target beta-amyloid aggregates, and merely a function of the fact that neurons rely on very complex spatial-temporal molecular dynamics and if you have a clumps of protein occupying physical space they shouldn't, it's obviously going to interfere with those spatial-temporal molecular dynamics. Contribution to the pathophysiology of AD is extremely small.
Time to think big and take moonshots on new ideas.
Time for a portfolio update.
TL;DR $ABVX and $NKTR seriously tanked the portfolio since the last one ($TENX was also an L but that was a small position so was quite a small contribution). $PEPG, $QURE, and $IVA are really critical readouts coming up in the next few weeks that are likely to move the portfolio substantially, one way or another. Wanted to put out where things sit ahead of then for accountability
As noted previously, majority of capital was committed many months after each portfolio's start date, so start dates are somewhat misleading in that regard (e.g. for WSLS II about 40% of capital...a lot of it $ABVX post-cancer freakout... committed from May 2026 onwards). Probably going to combine them into one and have a better display layout when I have time, don't really want to do a more classic time weighted approach here bc I think that would paint a rosier view of performance than is warranted.
_______________________
Re $ABVX, my brilliant strategy of loading up time and again since Fall 2025 with more and more non-biotech allocated capital because hey the market will totally figure it out this time and you know I think I'll be able to flip that into other positions with readouts further out, plus its really derisked and probably getting bought anyways, and where else can you find an opportunity like that in the market has all worked out about as well as you would think...pretty terribly. WSLS II in particular is really living or dying by the VAX at the moment.
However, I have continued to hold basically my full position, with the exception of selling a small silver to get a little tax loss harvesting motion going (cope) and free up a little capital for some diversification in an ugly market and to play a few nearer term binaries.
I made a judgment call on $NKTR pricing dynamics that there was a reasonable probability there would be little buying support after it nose dived following off treatment data based on the broader market conditions, the apparent lack of confidence in management across the Street and the lack of near-term catalysts, and closed out a big part of the position.
I took that capital and did the following:
I took a position in $ZURA on the basis of Adam's really thoughtful work. All of the arguments made sense to me for why IL-17A was the critical driver of Bimzelx's superior efficacy. However, I became concerned about tibulizumab's IL-17A target engagement in vivo. Briefly, my concern arises from the fact that a) tibulizumab's binding affinity is 14pM to ixekizumab's 1.8pm in vitro b) I think there is a risk that the very high concentrations of BAFF in vivo relative to IL-17A further interfere with target engagement w/ IL-17A, perhaps as may have happened with sonelokimab and IL17-F, recognizing $ZURA's bispecific has two ends so this view is predicated on things like aggregation dynamics, membrane bound BAFF etc and c) the only in vivo metric of target engagement w/ IL-17A I have been able to find across the IL-17A antibodies / small molecules is an increase in Total - IL17A and cosentyx and tibulizumab show a similar ~50x ish increase (see 2020 article in Dermatology) (investors.zurabio.com/static…, onlinelibrary.wiley.com/doi/…).
Increase in Total IL17-A is a really imperfect metric, and is highly susceptible to measurement error at baseline, differing clearance rates, etc. which could all seriously mess with results.
However, collectively that represented too much risk for me to want to take on, especially given the important upcoming binaries with 10mg / kg MAD for $PEPG, $QURE's BLA filing decision, and $IVA (more on that later). So, I sold the position for a tiny gain. Depending on how $ZURA reads out, might be back in for SSc.
I went back into $MAZE in reasonable size.
I opened a position in $IVA (some brief commentary here: teapawt.net/WaveStrumLS/status/211…)
I went back into $PCVX after it dumped into the 60s following data.
I opened a position in $MDGL.
The thought process with $PCVX and $MDGL was basically these late stage companies got torched from their highs in a manner somewhat akin to $ABVX. If you believe this reflects short term pricing dynamics that don't reflect underlying value (degrossing, etc.) but have hit quality names really hard, at some point that will turn around and so you’ll get more of a return than you should for a company at this stage, while not exposed to drug failure / trial risks the same way. For $MDGL, I also used it as somewhat of a hedge against $IVA. If $IVA fails and sentiment swings back in favor, I expect $MDGL will rise and hopefully offset some of the $IVA loss. If $IVA succeeds, my hope is that it will have relatively minimal impact next to the $IVA gains.
I further increased my $QURE position after the 4 year data. I have written about this elsewhere but the likelihood that this is having disease modifying effect is near certain in my opinion based on the totality of data. If that's true, I think it should be approved.
As mentioned, if $PEPG, $IVA and $QURE do what I hope they do in the next few weeks, we hopefully will be back on a good growth path with the port. If not, that will suck.
I do continue to believe for all the many reasons previously discussed that $PRME has the kind of upside potential to be a true marquee VC style investment wherein everything else could get crushed and, if it reaches its full potential, the portfolio still delivers beats the market over the long-term. (And as always, I think its very derisked at this price sans a black swan safety event given the $VERV acquisition at the depths of the bear market for 1 - $1.3Bn, the actual platform nature of the technology, the evolving FDA approval environment that favors said platform, etc.)
I was in and then out of $DFTX, decided I liked other opportunities better
Long $IVA, $IRD, $MDGL, $PCVX, more $MAZE and more $QURE over the past couple of weeks. Sold a bunch of $NKTR to fund. Will post portfolio update over the weekend
Re $IVA ..
Dr. Ralph DeFronzo (his work was a big part of why I was skeptical of $WVE's obesity play) has discussed in a number of public forums and written a lot on the ppar-y agonist pioglitazone as a standalone therapy and as part of a combo therapy (which includes an old school GLP-1 exenatide on top of metformin) in diabetes (65ish% of T2D patients have MASH). Worth a listen, as liver dysfunction is a key part of insulin resistance / diabetes and I think provides good insights into how ppar-y interfaces with both the liver and the multiplicity of systems that effect the liver.
The below papers speak to the likely impact of GLP-1's on the minor weight gain that was seen in previous trials of ppar-y agonist pioglitazone
Standalone Therapy: In ACT NOW pioglitazone demonstrated a significant average weight gain of 3.9kg over .77kg placebo. (pubmed.ncbi.nlm.nih.gov/2142…)
Combo Therapy: The EDICT Trial which combined pioglitazone, metformin and exenatide demonstrated a 1.3kg weight gain over .6kg placebo at 3 years (39% lost weight).
(pmc.ncbi.nlm.nih.gov/article…)
Different patient populations and cross trial comparison caveats. However, given how much stronger the current GLP-1's are than exenatide, I would expect any associated weight gain to be entirely mitigated in current GLP-1 users (who have been allowed to participate in the trial and will have a sub-group analysis).
This sits on top of a few other constructive facts:
- 51% of patients in the $IVA PhaseIIb trial having stable weight or weight loss with standalone therapy
- $IVA's drug shows meaningfully less weight gain than historical ppar-y agonists like pioglitazone
- SGLT2i's in combo w/ ppar-y agonists have historically shown weight loss and the LEGEND study confirmed SGLT2i's mitigate weight gain with $IVA's drug as well, I think any worries about weight gain have a pretty low-risk of being a commercial concern. (pubmed.ncbi.nlm.nih.gov/3855…)
- Any reported weight gain is really more of a weight redistribution, less visceral fat, etc, combined with fluid retention anyways
3
37
3,522
Long $IVA, $IRD, $MDGL, $PCVX, more $MAZE and more $QURE over the past couple of weeks. Sold a bunch of $NKTR to fund. Will post portfolio update over the weekend
Re $IVA ..
Dr. Ralph DeFronzo (his work was a big part of why I was skeptical of $WVE's obesity play) has discussed in a number of public forums and written a lot on the ppar-y agonist pioglitazone as a standalone therapy and as part of a combo therapy (which includes an old school GLP-1 exenatide on top of metformin) in diabetes (65ish% of T2D patients have MASH). Worth a listen, as liver dysfunction is a key part of insulin resistance / diabetes and I think provides good insights into how ppar-y interfaces with both the liver and the multiplicity of systems that effect the liver.
The below papers speak to the likely impact of GLP-1's on the minor weight gain that was seen in previous trials of ppar-y agonist pioglitazone
Standalone Therapy: In ACT NOW pioglitazone demonstrated a significant average weight gain of 3.9kg over .77kg placebo. (pubmed.ncbi.nlm.nih.gov/2142…)
Combo Therapy: The EDICT Trial which combined pioglitazone, metformin and exenatide demonstrated a 1.3kg weight gain over .6kg placebo at 3 years (39% lost weight).
(pmc.ncbi.nlm.nih.gov/article…)
Different patient populations and cross trial comparison caveats. However, given how much stronger the current GLP-1's are than exenatide, I would expect any associated weight gain to be entirely mitigated in current GLP-1 users (who have been allowed to participate in the trial and will have a sub-group analysis).
This sits on top of a few other constructive facts:
- 51% of patients in the $IVA PhaseIIb trial having stable weight or weight loss with standalone therapy
- $IVA's drug shows meaningfully less weight gain than historical ppar-y agonists like pioglitazone
- SGLT2i's in combo w/ ppar-y agonists have historically shown weight loss and the LEGEND study confirmed SGLT2i's mitigate weight gain with $IVA's drug as well, I think any worries about weight gain have a pretty low-risk of being a commercial concern. (pubmed.ncbi.nlm.nih.gov/3855…)
- Any reported weight gain is really more of a weight redistribution, less visceral fat, etc, combined with fluid retention anyways













