September 3, 2026
Bonus Content: Ultragenyx Falls Nearly 47% on Failed Angelman Trial. Its Next FDA Decision Is 16 Days Away.
Dear Reader,
Sovereign wealth funds — the largest, most patient, most sophisticated pools of capital on Earth — have begun making substantial, concentrated allocations to artificial intelligence.
These are not hedge funds chasing momentum. These are nations deploying generational capital.
Abu Dhabi’s sovereign wealth ecosystem — worth $2.3 trillion — has embedded AI infrastructure into the financial system itself.
Norway’s $2.1 trillion Government Pension Fund Global — the world’s largest sovereign wealth fund — now uses large language models to screen every portfolio company daily.
CEO Nicolai Tangen wrote: “Artificial intelligence is changing how we work as an investor.”
Kuwait Investment Authority… Qatar Investment Authority… Mubadala… Each has recorded its largest-ever digital and AI commitments.
The Trump administration issued an executive order to establish a U.S. sovereign wealth fund — with AI-related technology investment as a stated priority.
And the Stargate Project — a joint venture between OpenAI, SoftBank, Oracle, and UAE-backed MGX — plans to deploy $500 billion over four years building AI infrastructure in the U.S. President Trump described it as “the largest AI infrastructure project by far in history.”
Combined, sovereign wealth funds now manage a record $15 trillion in assets globally — with AI as a core strategic allocation priority.
This matters, because the historical pattern of state-level capital entering a sector is well documented.
It played out in oil in the 1970s…
Telecommunications in the 1990s…
Technology equities in the 2000s…
When sovereign capital establishes a position, the assets they are buying tend to reprice accordingly.
See the full breakdown of where sovereign funds are allocating here.
Good investing,
Alexander Green
Chief Investment Strategist, The Oxford Club
Ultragenyx Falls Nearly 47% on Failed Angelman Trial. Its Next FDA Decision Is 16 Days Away.
Binary outcomes in biotech do not arrive with warning labels. On Wednesday afternoon, September 2, Ultragenyx Pharmaceutical disclosed that its Phase 3 ASPIRE trial of apazunersen in Angelman syndrome had missed every endpoint it was built to hit. The study did not achieve the primary endpoint of change from baseline in Bayley-4 cognitive raw score, nor the key secondary endpoint of net response in the Multidomain Responder Index. The market response was immediate and severe. RARE closed at $26.53 on September 2, then traded around $14 in premarket on September 3, a decline of roughly 47%.
This is not a story about a drug that almost worked. No efficacy differences were observed between treated and control groups on Bayley cognition scores or MDRI, including the five individual MDRI components. Jefferies was blunt: the failure of the drug in all five sub-domains of the MDRI indicates a clear signal failure and not a design failure. Phase 1/2 data had shown genuine early promise, with April 2024 interim results showing rapid and clinically significant improvement in cognition on Bayley-4 relative to natural history data. That signal did not translate. The controlled setting exposed what the open-label cohort concealed.
What the Numbers Show
The success of apazunersen was a key driver of Ultragenyx’s value, according to William Blair, and the asset’s failure robs the biotech of any meaningful near-term milestones from that program. Management moved quickly to contain the damage. The company said it would assess its planned operations, pivot to its growing commercial business, and roll out significant expense reductions, though the specifics remain unspecified. The commercial anchor is real: Ultragenyx reaffirmed its 2026 total revenue outlook of $730 million to $760 million and held $436 million in cash, cash equivalents, and marketable securities as of June 30, 2026. Q2 2026 EPS came in at -$0.90, beating the consensus estimate by a wide margin. The balance sheet survives the trial failure. The pipeline story does not, at least not as it was written before Wednesday.
The Next Catalyst: 16 Days
The FDA set a PDUFA action date of September 19, 2026 for UX111, the company’s AAV9 gene therapy for Sanfilippo syndrome Type A. If approved, UX111 would be the first approved therapy for Sanfilippo syndrome Type A, a rare disease affecting young children that leads to progressive, irreversible neurodegeneration and early death. This is a resubmission: the FDA issued a complete response letter on the original application, then accepted the resubmitted BLA after Ultragenyx added long-term data.
The regulatory tension is structural. Bears will focus on the fact that the developmental benefit is derived from comparison with external natural history rather than a randomized placebo-controlled trial, leaving the FDA room to question patient comparability and residual confounding. That is the same translation problem that destroyed apazunersen. Bulls counter that, during its prior late-cycle review, the FDA indicated that the neurodevelopmental outcome data were robust and that the biomarker data provided additional supportive evidence.
Options Market: Volatility Has Adjusted
RARE entered the ASPIRE readout as a clinical-stage binary. During the September 2 session, shares reached a daily high of $26.80. Implied volatility on RARE has spiked following the collapse and the proximity of the September 19 PDUFA date. With the stock near $15, September and October options are pricing the next move at over 30% in either direction. IV rank is elevated, reflecting that current volatility sits near the top of its 52-week range. That environment favors defined-risk structures over outright directional bets.
Structured Trade Framework
Bull case: If you believe the FDA accepts UX111’s biomarker and long-term natural history framework as sufficient for accelerated approval, a defined-risk long call spread in October expiration captures upside with contained premium. A $15/$20 call spread limits loss to the net debit while participating in a recovery toward pre-collapse levels.
Bear case: A second rejection on UX111 at an already-distressed stock price would likely pressure shares toward the $10 range. A long put or bear put spread with a September 19 expiration captures that outcome. With IV elevated, the cost of protection is high; a spread structure reduces that drag.
Neutral / strangle: For traders who have no conviction on the UX111 binary but expect a large move either direction, a long strangle using out-of-the-money calls and puts in September expiration isolates the volatility event itself. The risk: if the FDA delays or the stock barely moves, elevated IV crushes both legs.
Risk Factors
Angelman readthrough is real. Analysts have raised concerns about potential readthrough from the ASPIRE results to other ongoing Angelman studies involving apazunersen. Regulatory risk on UX111 is asymmetric: a second FDA rejection on the same asset, after a prior complete response letter, would likely be a terminal event for that program. Liquidity risk also matters here. Trading volume for RARE spiked well above its recent average on September 2, which can widen bid-ask spreads in the options market and increase slippage on any defined-risk structure.
Action Checklist
- Confirm the September 19, 2026 PDUFA date for UX111 remains unchanged before sizing any position.
- Check current bid-ask spreads on RARE options before entry; elevated volume does not equal tight markets in a name this size.
- Size any structure to the defined maximum loss, not the expected value. A second binary miss is not an outlier scenario.
- Monitor for any FDA correspondence, advisory committee convening, or information request on UX111 between now and September 19.
- Treat the $436 million cash position as a floor check: the August 19, 2026 FDA approval of GENGLYCOS provides a commercial anchor that makes outright bankruptcy unlikely, but does not cap downside from pipeline failure.
