ANAB is three bets in one wrapper: a GSK royalty stream, a Phase 3 clinical decision, and an active lawsuit. The 10-K shows all three clearly. None of them offer informational edge at this coverage level.

The Setup

AnaptysBio at $55 ($1.5B market cap) is splitting into two public companies in Q2 2026: a Royalty Co that collects Jemperli royalties from GSK, and a Clinical Co built around rosnilimab, a novel PD-1 agonist for autoimmune disease. The stock is up 231% in twelve months. Twelve analysts cover it, all but one rate it a buy.

The 10-K filed March 3 confirms the thesis the market already priced. Jemperli hit $1.1B in worldwide sales (+89% YoY per GSK's Q4 call). ANAB collected $224.9M from GSK in FY2025. Rosnilimab Phase 2b hit its primary endpoint at all three doses with durable off-drug responses. Cash is $311.6M. Management bought back $68.6M in stock at ≈$20/share average — the stock is now $55.

What the 10-K also shows, buried in the contract terms and litigation disclosures, is why 38.6% of the float is short.

Factor Decomposition

Six factors drive ANAB, but they collapse into three independent bets.

Bet 1: The Royalty Machine (≈40% of value)

ANAB collects tiered royalties on Jemperli sales. The tiers matter: 8% below $1B, 12% on $1B-$1.5B, 20% on $1.5B-$2.5B, 25% above $2.5B. These are the Amendment No. 3 rates from October 2020 — not the original 4-8% that sometimes gets cited.

The nonlinear math: at $1.1B sales, ANAB earns ≈$92M in royalties. If Jemperli hits $2B (it grew 89% last year and GSK is pursuing label expansion into rectal cancer), royalties jump to ≈$240M. The rate INCREASES as sales grow through tiers.

The Sagard complication: ANAB sold future royalties to Sagard Healthcare for $300M. Sagard collects everything until they've received $600M total (≈$250M accrued, ≈$350M remaining). Once hit, ANAB recaptures the full stream. At $240M/yr flowing to Sagard, the cap is reached mid-2027 — not the "4-8 years" you'd calculate with the wrong royalty rates.

The accounting distortion: Sagard creates $79.9M in non-cash interest expense (up from $50.1M because Jemperli growing faster accelerates the effective interest rate). This makes the GAAP P&L show a $13.2M net loss when operating income was actually +$47.9M. The worse the non-cash interest looks, the faster the Sagard cap approaches. It's a good problem dressed as a bad number.

Edge assessment: None. GSK is a $150B company with 30+ analysts. We cannot out-forecast Jemperli sales. The Sagard math is in the 10-K and has been since 2020. Every analyst modeling ANAB has this.

Bet 2: The Clinical Decision (≈30% of value)

Rosnilimab is a PD-1 agonist — it activates PD-1 to induce immune tolerance, the opposite of cancer checkpoint inhibitors. The Phase 2b data in rheumatoid arthritis:

Hit primary endpoint (DAS28-CRP) at all three doses. Responses deepened through six months. Durable off-drug for three or more months — suggesting quarterly or less frequent dosing is possible. Works in patients who failed other DMARDs.

The Phase 3 decision is imminent (H1 2026). The 10-K language is hedged: "We are assessing advancement to Phase 3." Management may condition the go-ahead on finding a pharma partner to share the $300-500M Phase 3 cost.

Behind rosnilimab: ANB033 (CD122 antagonist) posted clean Phase 1a data in October 2025 — 70-75% reduction in target CD8 T cells, safety clean, now in Phase 1b for celiac disease and eosinophilic esophagitis. No FDA-approved biologics exist for celiac. ANB101 (BDCA2 modulator) is in Phase 1a with ramping spend.

Edge assessment: None. RA is a $30B+ market analyzed by every immunology-focused biotech analyst. We cannot assess PD-1 agonist Phase 3 probability better than 12 specialist analysts.

Bet 3: The Lawsuit (≈30% of value)

ANAB filed against GSK/Tesaro in Delaware Chancery Court on November 20, 2025, alleging material breach of the Collaboration Agreement. Same day, Tesaro counter-filed, alleging ANAB breached first and seeking an injunction preventing ANAB from terminating the agreement.

The 10-K does not disclose what the breach was. The counter-suit seeking to prevent termination tells you ANAB was trying to exit — whatever Tesaro did was serious enough for ANAB to want out of the collaboration entirely.

The timing — the same month as the separation announcement — is notable. Either ANAB needs to resolve collaboration rights before splitting, or the litigation is designed to maximize value pre-separation, or it's coincidence. The 10-K can't tell you which.

Here's the detail that makes the litigation even more dangerous: buried in the contract terms (lines 700-703), GSK can terminate the collaboration at any time with 90 days' notice. No cause required. ANAB can only terminate for uncured material breach. The asymmetry is extreme. GSK doesn't need to win the lawsuit — they can walk away unilaterally.

Why GSK probably won't: Jemperli is their fastest-growing oncology asset, up 89%, with ASO-1 rectal cancer label expansion data expected later this year. Terminating would be mutually destructive. But the option exists, and in a deteriorating relationship, options get exercised.

Edge assessment: None — and this is the critical factor. The 38.6% short float (12.5 days to cover, institutional, not meme) is almost certainly driven by intelligence on the breach specifics that we don't have. If you could price the litigation correctly, you'd know whether ANAB is cheap or expensive at $55. We can't.

The Independence Problem

The three bets aren't fully independent. If litigation goes badly (Bet 3), the royalty stream is impaired (Bet 1) and the Sagard recapture timeline extends or breaks (also Bet 1). The separation may be delayed or blocked (part of Bet 3). Only the clinical decision (Bet 2) is truly orthogonal.

F1 (Jemperli) <-> F2 (Sagard)     [derivative]
      ^
F4 (Litigation) --> F5 (Separation)  [blocks on downside]

F3 (Phase 3)      [independent]
F6 (Pipeline)     [independent]

The bearish correlation structure means the downside scenarios cluster. If ANAB loses the lawsuit, you don't lose 20% (the litigation factor weight) — you lose 60% because royalties, Sagard, and separation all degrade together.

The Short Case Deserves Respect

38.6% short float on a $1.5B biotech with 12 analyst buy ratings is not normal. These are institutional shorts with 12.5 days to cover. They have a thesis. The most likely thesis centers on the litigation — the one factor where public filings reveal almost nothing.

Other short arguments: Clinical Co needs $300-500M for Phase 3 with no obvious funding source post-separation. The $68.6M buyback at $20 was the bid under the stock — with the authorization expiring March 31, 2026, that support disappears. The taxable spinoff structure is unusual and creates $20-50M in friction.

Conclusion

ANAB is an interesting company executing a complex corporate restructuring with a genuine royalty machine, novel clinical assets, and a material lawsuit — all at the same time. The bull case (sum of parts: Royalty Co at $800M-$1.1B + Clinical Co at $500M+ = $1.3-1.6B vs current $1.5B, with Phase 3 upside to $3B+) is real. The bear case (litigation impairs royalty stream, Clinical Co can't fund Phase 3, separation delays) is also real.

We have no edge on any factor. The royalty is GSK commercial beta. The clinical decision is immunology science. The litigation outcome requires legal intelligence we don't have. At $1.5B market cap with 12 analysts, the discovery phase is complete — this is a known, debated, and heavily-shorted name.

The one analytical detail worth noting: the corrected royalty tiers (8/12/20/25%) compress the Sagard recapture to mid-2027 at current growth, which is more bullish than casual analysis suggests. But every analyst with the 10-K has the same numbers.

No position. Monitor for Phase 3 announcement (H1 2026) and separation completion (Q2 2026 target).

Predictions

Evidence