The trade is not the +138% rerating that already happened. It's what the September options expiration doesn't cover.

Puma Biotechnology (PBYI) sells one approved drug — NERLYNX (neratinib) for HER2+ extended-adjuvant breast. Q1 2026 changed two things. The company posted the first two consecutive year-over-year demand increases in NERLYNX launch history, breaking a decade-long melting-ice-cube narrative. And a Phase II oncology asset (alisertib, Aurora-A kinase inhibitor) generated a biomarker-defined signal in 3rd-line ER+/HER2- breast cancer with the next readout due Q4 2026. Listed options bracket Q2 earnings (Aug 6) and the related CELC PDUFA (Jul 17) but not the H2 alisertib update — Dec '26 expiry has 7 OI total. Equity is the only liquid way to hold that binary.

What's priced

NERLYNX-only SOTP: 2026 guide of $202-206M revenue, $16-19M net income, debt-free with $101.5M cash. At 12-15x forward earnings plus cash, that's $5.20-$6.90/share. Alisertib option, probability-weighted across outcomes (5% home run, 20% Phase III initiates, 45% mixed, 30% fails or attenuates): ≈$1.95/share. Sum: $7.15-$8.85 versus current price.

The stock is at conservative fair value. The visible earnings inflection has been priced; the alisertib option is at floor.

The market has one sell-side analyst (HC Wainwright, HOLD, $5 target, unchanged since 2024). Listed options imply ≈40% probability of $10+ by Sept 18, ≈18% probability of <$5. Both expire before the H2 alisertib readout.

Where the edge sits

The slow piece — NERLYNX rerating. For a decade NERLYNX was priced as terminal decline. Q1 metrics show inflection: NRx +25% QoQ and +11% YoY, sales call activity +44% YoY, two consecutive YoY demand increases for the first time ever. A cross-ticker scan of the HER2+ space (DSNKY, AZN, RHHBY, PFE/Seagen) found no displacing catalyst in the next 12-18 months. With one analyst covering, multiple expansion happens slowly as coverage refreshes.

The fast piece — alisertib in PIK3CA-WT/ESR1-mut 3rd-line breast. Phase II ALISCA-Breast1 50mg arm: median PFS not reached, zero progressions at interim. The n is undisclosed and likely single digits — that caveat matters more than the headline. What partially survives the small-n problem is the mechanism: the PIK3CA-WT/ESR1-mut subgroup is a clinical proxy for c-MYC positivity (alisertib's target), and ESR1+PIK3CA testing is already standard-of-care post-CDK4/6. No companion diagnostic development is required.

This is the resolution to the strongest bear point. Aurora kinase inhibitors have zero Phase III oncology approvals across roughly two decades and multiple drugs. Alisertib itself failed Phase III in PTCL. But the prior failures used unselected populations or the wrong biomarkers — none used a clinically validated c-MYC proxy that SOC testing already provides. The biomarker mechanism is what differentiates this trial from the AKI graveyard. It does not guarantee Phase III success; it changes the comparison.

P_you vs P_market

Listed options are roughly efficient on visible events. The 40% probability on the H2 alisertib update is in no derivative.

Risks (ranked)

1. AKI class base rate. 0 Phase III oncology approvals across ≈20 years. Biomarker mechanism is the resolution above; if the H2 update shows progressions, the option dies and equity reverts toward NERLYNX-only fair value (≈$5-7).

2. CELC PDUFA July 17. Gedatolisib + fulvestrant in 2L PIK3CA-WT showed mPFS 9.3mo vs 2.0mo (HR=0.24). Approval (≈78% probability) raises the SOC bar that PBYI must clear in any breast Phase III design.

3. Subgroup n undisclosed. Zero progressions at n=5 is not the same as at n=20. The signal can attenuate as enrollment expands.

4. Insider P-code absence. Twelve months show no open-market purchases by CEO/CFO/directors at any price level — including the 2024-2025 trough. P is the only bullish Form 4 code; absence at trough is rhetoric over conviction.

5. Reflexivity overshoot. PBYI ran $30 → $230 → $50 in 2014-2017 on neratinib. Same risk on alisertib if the H2 update fails to confirm.

Catalysts

Sizing logic

Continuous-hold Sharpe estimates ≈0.23 (below the 0.5 investability threshold). Path-conditional Sharpe estimates 0.5-0.7. The trade requires path management, not size at entry. Start small; scale only on H2 alisertib update with disclosed n>10 still showing zero or near-zero progressions. Trim aggressively if H2 shows progressions.

What would change our mind

Bullish: Q2 print at top of guide; H2 alisertib disclosed n>15 with zero or near-zero progressions; FDA Type B confirming Phase III path; insider P-code purchase at any size; sell-side coverage initiation.

Bearish: Q2 miss; H2 alisertib shows progressions in subgroup; G3+ neutropenia or discontinuations rise at higher dose; CELC PDUFA approval creates a SOC bar PBYI's design must beat by a larger margin than realistic.

Why LR 1.3, not higher

The +138% YTD move means the obvious rerating has been captured. Stock is at conservative fair value, not at mispriced entry. The remaining edge is concentrated in one binary outcome six months out, which the AKI base rate makes asymmetric in both directions. The memo is bullish in direction and modest in magnitude — claims of conviction higher than this would be confusing narrative with novel information.

Evidence