NRIX — Q3 2026 Financial Report Analysis

A $700 million Roche deal turned a cash-burning biotech profitable for one quarter

Nurix Therapeutics is a clinical-stage drug developer: it has no approved medicines and sells nothing, so it normally loses money every quarter and lives on cash raised from investors and payments from larger drug companies. In its fiscal third quarter (the three months to August 31, 2026; Nurix's fiscal year ends in November), it reported net income of $549.8 million, or $4.73 per diluted share, against a $86.4 million loss a year earlier. Almost all of that came from one event. Roche paid Nurix $700.0 million upfront for a worldwide license to bexobrutideg (NX-5948), Nurix's lead drug for a blood cancer called chronic lymphocytic leukemia (CLL). The deal took effect in July 2026, and Nurix booked $653.2 million of it as revenue in this quarter. Take that payment out and the business looks much as it did before: it lost $93.7 million from operations, slightly more than last year.

At a glance

• $1.06 billion in cash and investments at August 31, up from $593 million at the end of November 2025. The Roche upfront is why, and it is the most important number in this filing.
• $653.2 million of one-time license revenue produced the profit. Without it, the quarter's operating loss was $93.7 million, compared with $91.4 million a year ago.
• Roche now pays 60% of bexobrutideg's global development costs. That cut this quarter's research spending by $4.5 million, the first sign of a lower cost base for Nurix's most expensive program.

The numbers

Money figures are in millions of US dollars. "Diluted EPS" means profit or loss per share, counting shares that could be created from stock options and warrants.

"n/m" means the percentage isn't meaningful because the figure swung from a loss to a profit. The 83.1% operating margin (the share of revenue left after running costs) is a mechanical result of one large license payment, not a measure of how the business normally runs.

What happened in the quarter

The Roche license. Roche gets exclusive worldwide rights to bexobrutideg. The two companies will develop and sell it together in the United States and split US profits and losses 50/50; Roche sells it everywhere else and pays Nurix royalties "ranging from the low teens to the high teens" percent on those sales. Development costs are shared 40% Nurix, 60% Roche. On top of the $700.0 million upfront, Nurix can earn up to about $1.6 billion more: up to $290 million for development milestones, $280 million for regulatory milestones, $300 million for launches outside the US and $710 million for sales milestones outside the US. None of those contingent amounts is counted in this quarter's results.

Of the $700.0 million, Nurix recorded $653.2 million as revenue immediately, because under accounting rules the license itself was handed to Roche the day the deal took effect. The other $46.8 million is set aside to be recognized as Nurix performs development work for the partnership; $1.2 million of it was used this quarter, and $45.6 million is still sitting on the balance sheet as "deferred collaboration funding."

Collaboration revenue more than doubled, for a one-off reason. Revenue from Nurix's drug-discovery partnerships rose to $19.9 million from $7.9 million. The filing attributes the increase "primarily" to a development milestone from Sanofi: Sanofi contributed $10.0 million this quarter, versus nothing a year ago. Pfizer revenue rose to $9.7 million from $5.8 million. Of that, $5.2 million related to work satisfied in earlier periods, and a Pfizer research milestone was achieved in August 2026. Gilead fell to $0.2 million from $2.1 million as an extended research term ended. Over nine months, collaboration revenue was actually down 13%, to $35.2 million from $40.4 million, because research terms with Gilead and Sanofi wound down.

R&D fell, but only because Roche started paying its share. R&D expense dropped 8.0% to $79.3 million. The filing says this was due to "the cost sharing offset related to the Roche Collaboration Agreement and a decrease in contract manufacturing due to timing." The Roche offsets total $4.5 million ($3.3 million cost-sharing plus $1.2 million from the deferred upfront), so before them R&D would have been about $83.8 million. Outside spending on bexobrutideg itself was $20.0 million, down from $25.1 million, but over nine months it was up 21% to $72.1 million as patient enrollment accelerated. Nine-month R&D overall rose 7.3% to $251.0 million.

G&A jumped on deal costs. General and administrative expense rose $21.2 million to $34.4 million. The filing attributes the increase to "one-time expenses associated with the Roche transaction for consulting and business development activities."

Pipeline changes

• Bexobrutideg (NX-5948) degrades BTK, a protein that leukemia cells rely on to survive. Rather than just blocking the protein, as existing BTK drugs do, it tags it for the cell to destroy. A Phase 2 study in CLL patients who have already failed three types of drugs, designed as a potential basis for faster ("accelerated") US approval, has been running since October 2025. With Roche, a Phase 3 trial (DAYBreak CLL-306) enrolled its first patient in August 2026. It compares bexobrutideg head to head with pirtobrutinib, the current standard treatment for patients whose disease has progressed after a first-generation BTK drug. The trial has two primary measures: how many patients respond, and how long they live without the disease getting worse.
• Zelebrudomide (NX-2127) was discontinued in August 2026 "as part of a strategic prioritization" of Nurix's wholly owned cancer drugs. It was a small cost line ($0.7 million of outside spending this quarter), so the financial effect is minor. It does leave Nurix with one fewer drug in the clinic.
• NX-1607, a drug meant to strengthen the immune system's attack on tumors, dosed its first patients in September 2026 in Phase 1b expansion groups covering colorectal, lung, prostate cancer and melanoma.
• Gilead extended the research term on one partnered target by two years in August 2026, for a $10.0 million fee that Nurix expects to receive. Together with the Sanofi milestone, that explains the $20.0 million in accounts receivable (money owed to Nurix) on the balance sheet.

Takeaway: This quarter's profit is a one-time event. The important change is to Nurix's balance sheet and cost structure: it now has $1.06 billion in cash and a partner paying 60% of the development costs of a drug that has entered a head-to-head Phase 3 trial. The trade-off is that Nurix gave up half of the US profits and all non-US sales, in exchange for royalties, on what is now its only late-stage drug.

What the headline numbers hide

• The profit says nothing about whether the business can sustain itself. Without the $653.2 million license revenue, pre-tax results would have been a loss of about $89.7 million, compared with $86.4 million a year ago. The underlying spending rate hasn't changed.
• Cash flow tells the same story. Operating cash flow (cash actually coming in minus cash going out from running the business) was positive $447.1 million over nine months. Subtract the $700.0 million upfront and that becomes an outflow of about $253 million, or roughly $84 million a quarter. That is our own calculation, and it is only approximate: it also includes $20.0 million of milestone money earned but not yet collected, and about $22.0 million of deal-related bills not yet paid.
• G&A is inflated by one-time deal costs. The $21.2 million jump in G&A is attributed to Roche transaction expenses, so G&A should fall back next quarter. Adjusting for it would make the underlying operating loss smaller, not larger.
• R&D's decline is a cost transfer, not a cutback. Roche's 60% share now reduces Nurix's reported R&D. As the Phase 3 trial enrolls, total program costs should rise, but Nurix will pay only 40% of them.
• The tax bill was small for a $563 million pre-tax profit. The $13.6 million provision works out to an effective rate of about 2.4%. Nurix had federal tax-loss carryforwards of about $323.4 million at the end of fiscal 2025 (past losses that can offset future taxable profit), which mostly shields this one-time gain.
• Share count. Shares outstanding rose 7.3% since November, to 109.2 million. Most of the increase came from about 4.7 million pre-funded warrants that were exercised in August 2026. These warrants were already counted in per-share figures, so the exercise did little real diluting. The rest came from a January 2026 sale of 1.0 million shares under Nurix's "at-the-market" program (which lets the company sell stock gradually on the open market) and from employee stock awards. $300.0 million of stock remains available to sell under that program. Diluted EPS uses a higher share count (116.2 million) than basic EPS, because once there is a profit, stock options count as potential shares.

Outlook

Nurix does not give financial guidance. The 10-Q states only that existing cash is enough to fund operations "for at least the next 12 months," which is standard minimum wording rather than an estimate of how long the money will last. Our rough arithmetic: $1.06 billion divided by about $84 million of quarterly underlying cash burn is roughly 12 quarters, or three years. That figure is ours, not the company's, and real costs should rise as the Phase 3 trial and NX-1607 expansion cohorts enroll patients, so treat it as an upper bound.

What to watch next:

1. Bexobrutideg trial news. Enrollment pace in DAYBreak CLL-306, and any data from the potentially pivotal Phase 2 study, which could support an accelerated-approval filing. Development milestones from Roche (up to $290 million) depend on this progress.
2. The new cost base. The fiscal fourth quarter will be the first full quarter with Roche paying its share and without one-time deal costs. That gives the cleanest view yet of Nurix's ongoing spending.
3. Partnership income. The $10.0 million Gilead fee and further Pfizer and Sanofi milestones can make collaboration revenue jump around from quarter to quarter. Don't read one strong quarter as a trend.

The next filing is the annual report (10-K) for the fiscal year ending November 30, 2026. Last year's was filed on January 28, so the next one should arrive in late January 2027.