Thesis

Aardvark Therapeutics trades at $4.07. Cash per share is $4.45. Enterprise value is negative $8.2 million. The market is pricing the only clinical program for Prader-Willi Syndrome hyperphagia — a disease where children eat themselves to death and nothing is approved — as worth less than nothing.

The implied probability of program viability: 8%.

On February 27, the company paused all trials after "reversible cardiac observations" in a healthy volunteer study. No details. Stock down 67% in a month. Reasonable reaction to an information vacuum.

On March 23 — yesterday — the 8-K landed with the actual data. The cardiac finding is QRS prolongation. Not QTc. Not structural damage. Not arrhythmia risk. A dose-dependent, reversible conduction effect that showed up at supra-therapeutic doses administered without the dose escalation the trial actually uses. The efficacy dose (200 mg BID) sits below the safety threshold. The pause is voluntary — no FDA clinical hold.

Eight percent viability is wrong. I think it's closer to 60%. Even at 30% — deeply skeptical, "probably dead but let's see" — the stock is worth $5.86. At 20%, it's $5.05. You only need to believe viability is above 8% to make money. That bar is on the floor, and the filing cleared it by a wide margin.

Long AARD at cash floor. 1-2% starter. Q2 2026 FDA guidance collapses the uncertainty in weeks, not months.

What QRS Prolongation Is and Isn't

This is the entire thesis. If you understand this distinction, you understand the trade.

QRS measures conduction velocity — how fast the electrical signal moves through the ventricles. QTc measures repolarization — how long it takes the heart to reset after each beat. QTc prolongation causes torsades de pointes, a potentially fatal arrhythmia that has killed drug programs dead. QRS prolongation does not. They are different measurements with different risk profiles, different regulatory paths, and different survival rates for the drugs that cause them.

The market spent three weeks trading on "cardiac observations." That phrase is compatible with QTc prolongation, structural damage, myocarditis, heart failure — any of which would justify pricing the program at zero. It is also compatible with QRS prolongation, which is navigable. The February sell-off priced the worst case. The March 23 filing revealed something closer to the best case within the "cardiac finding" category.

The data, specifically:

At 1,600 mg BID (twice the trial dose, no escalation): 2/8 had QRS increase >25% from baseline. At 800 mg BID (target trial dose, no escalation): 1/23. All reversible. No SAEs. No serious cardiac symptoms. No medical intervention required.

Two details elevate this from "manageable" to "probably solvable." First, the healthy volunteer study bypassed the stepwise dose escalation the HERO trial uses (200 → 400 → 800 mg over three weeks). Volunteers went straight to full dose, creating higher peak plasma concentrations than trial patients would ever see. The study that caused the panic used harsher conditions than the actual trial design.

Second, the CEO states 200 mg BID produces plasma concentrations "substantially below the threshold where QRS effects are observed." That same 200 mg dose was just published in Molecular Metabolism (March 2026, peer-reviewed) showing hunger reduction versus placebo. The safe dose and the effective dose appear to converge at 200 mg. If that holds in PWS, this problem is solved by reducing the target dose — not by killing the program.

Why the Market Is Still Wrong

The 8-K is one day old. Analyst updates haven't landed. The stock gapped down on the February pause and has been drifting in a vacuum since. Nobody is accumulating: volume is 0.7x the three-month average, options open interest is negligible. This isn't smart money saying "we've seen the data and it's bad." This is nobody having read the data yet.

Some context on who IS positioned:

Cormorant Asset Management — $3.9 billion AUM, biotech specialist, run by Bihua Chen — holds a $13.3 million position from the IPO. They haven't filed a sale. These are people who read cardiac safety data for a living. If the QRS finding were a program killer, Cormorant would know it before we would.

CEO Tien-Li Lee bought $360K in open market stock between September and December 2025. CFO Nelson Sun bought $92K. At $8-16 per share, prices 2-4x where the stock sits today. The healthy volunteer study hadn't fully read out when they bought, but they knew the program's risk profile better than anyone. They didn't sell after the pause.

Short interest: 12.7% of float, 17.1 days to cover. That's a coiled spring. If FDA says "resume at lower dose with cardiac monitoring," shorts need seventeen days of average volume just to exit. The squeeze potential doesn't create the thesis, but it shapes the magnitude of any re-rating.

The Scenario Math

Four outcomes, probability-weighted:

EV: $8.28. Return: +103%.

But the more revealing table:

I'm at 60%. I could be wrong about the FDA path. I could be wrong about 200mg in PWS. I cannot be wrong that this finding — dose-dependent, reversible QRS prolongation at supra-therapeutic doses — justifies an 8% viability assessment. That number was set in a three-week information vacuum. The vacuum ended yesterday.

Where I Could Be Wrong

The honest part. Three things keep this at a starter position instead of a full conviction bet.

The 200mg efficacy question in PWS. This is the one that matters most and the one where I have the least insight. The Molecular Metabolism data is from an obesity trial. PWS hyperphagia is a different beast — hypothalamic dysfunction, not normal hunger signaling. The HERO trial was designed at 800mg for a reason. Maybe that reason is "200mg doesn't move the needle in PWS." If the effective dose for the actual indication is above the safety threshold, the entire dose-pivot narrative collapses. No public data resolves this. It's genuinely unknown, and I'm assigning 45% probability that 200mg is sufficient in PWS, which is really just an informed guess.

FDA opacity. The filing says "voluntary pause" and "working closely with FDA." That's what companies say. FDA could require a Thorough QT/QRS study under ICH E14 — six to twelve months and $10-20M. FDA could impose a formal clinical hold despite the company's voluntary framing. The 2026 political environment has reduced FDA staffing, which slows everything regardless of merit. I think a clinical hold is 15% likely, but I'm pricing FDA risk off historical base rates for QRS findings, not specific knowledge of this interaction.

Nobody else sees this yet. That's either alpha or delusion. The volume is dead. No institutional options positioning. The stock is at the 52-week low ($4.01). When you're the only one who thinks something is cheap, you're either early or wrong, and you only know which one after the fact. The edge — reading the QRS data faster than the market — decays within days as analysts update. If this hasn't moved by mid-April, the speed interpretation was wrong and the market's 8% might reflect information I don't have.

The Trade

Entry: $4.00-4.50. Negative enterprise value. Pipeline for free. If you're going to buy a biotech with a paused program, buy it when the cash exceeds the market cap and the filing just told you the finding is manageable.

Size: 1-2% starter. This is a doorway state — two interpretations fit, can't tell which yet. Size for surviving the wrong one. Not the time for Kelly.

Catalyst: Q2 2026 FDA path-forward guidance. One to ten weeks out. After that, the stock re-prices and the cash-floor entry is gone — in either direction. This is a pre-catalyst trade. You get in now or you pay up for confirmation later.

Scale: FDA confirms resumption path → 3-4%. PWS efficacy data at 200mg → 5%+. Insider buying at $4 → add. FDA clinical hold → cut immediately. No insiders buying by June → reassess.

Conclusion

AARD at $4.07 is a mispricing. The market priced the worst case during an information vacuum, and the information that ended the vacuum says the finding is manageable. QRS, not QTc. Dose-dependent. Reversible. No SAEs. Safe dose appears to equal the effective dose. Voluntary pause, not clinical hold. Cash exceeds the market cap.

The risk is real: 200mg efficacy in PWS is genuinely unknown, FDA is a black box, and nobody else is positioned for recovery. Those uncertainties keep this at 1-2%, not 5%. But they don't justify 8% viability. Nothing in the filing justifies 8% viability.

I'd rather own this at $4 and be wrong about the dose than watch it from the sideline at $8 because I wanted one more datapoint. The filing is twenty-four hours old, the catalyst window is weeks away, and the pipeline costs negative eight million dollars.

Evidence