ANNX — Q2 2026 Financial Report Analysis

Annexon, a drug developer with no products on the market yet, lost $55.3 million in the second quarter of 2026, 12.6% more than a year earlier, as it paid to move manufacturing of its eye drug vonaprument to a commercial-scale plant and stepped up spending on corporate and advisory work ahead of two big regulatory steps. Loss per share still fell to $0.28 from $0.34, but only because the share count rose by more than a third. The quarter matters less for its numbers than for what it sets up: a pivotal Phase 3 result in geographic atrophy and a U.S. approval filing for its Guillain-Barré syndrome drug, both guided for the fourth quarter of 2026, plus a new loan of up to $200 million that pushed the company's cash-runway guidance from the second half of 2027 to "into 2028."

At a glance

• $209.2 million in cash and short-term investments at June 30, down from $238.3 million at the start of the year. It would have fallen much further without $63.8 million raised by selling new shares on the open market during the first half.
• A $50 million loan drawn on July 30 (part of a facility of up to $200 million from Oxford Finance) brings available funds to roughly $259 million. It is Annexon's first debt of this kind, and it is secured by substantially all of the company's assets.
• Contract manufacturing spending rose 87% to $12.9 million, the single biggest driver of higher costs. This is spending to make vonaprument at a commercial-ready facility, which only makes sense if the company expects the drug to succeed.

The quarter in numbers

Annexon has no revenue, so the useful figures are what it spends, what it loses and how much cash it has left. All figures are from the company's Form 10-Q and earnings release for the quarter ended June 30, 2026.

For the first six months, the picture runs the other way: net loss was $99.5 million versus $103.5 million a year earlier, because R&D fell 11% to $82.4 million after the costly work to prepare the European filing for tanruprubart was finished. Spending dipped in the first quarter (R&D of $35.8 million) and came back up in the second ($46.6 million, about 30% higher quarter over quarter).

Where the money went

R&D (the cost of running trials and making drug supply) rose $2.4 million. The 10-Q attributes the increase "primarily" to a $6.0 million rise in contract manufacturing "for the manufacturing technology transfer of vonaprument to a commercial-ready facility." Other R&D lines fell: consulting and professional services dropped $1.4 million after the European filing for tanruprubart went in in January 2026; staff costs fell $1.3 million on lower headcount; and direct trial costs fell $1.2 million as the large ARCHER II eye trial moved past enrollment, partly offset by higher costs for the FORWARD study of tanruprubart in the U.S. and Europe.

G&A (overhead: finance, legal, corporate affairs) rose $3.1 million, or 41%. The filing points to a $2.3 million increase in consulting and professional services "for corporate affairs activities across our portfolio," with staff costs also higher, mainly from stock-based compensation (pay in shares rather than cash).

Interest income fell 27% to $1.9 million because the company held less cash on average than a year ago.

What the headline numbers hide

• The drop in loss per share is entirely from more shares, not from a smaller loss. The net loss grew $6.2 million, but it was spread over 201.0 million weighted shares instead of 148.3 million, after a November 2025 stock offering (about $80.5 million net) and at-the-market sales this year. An "at-the-market" (ATM) program lets a company sell new shares into the market a little at a time; Annexon sold about 11.8 million shares this way in the first half for $63.8 million net, and $117.8 million of capacity remains on its 2026 program. Each sale reduces existing holders' slice of the company.
• The prior-year per-share figure includes a one-off. The Q2 2025 loss per share of $0.34 was calculated on a loss of $51.0 million, which includes a $1.9 million non-cash "deemed dividend" from modifying stock warrants. There was no such item this year. (The press release describes the 2025 loss attributable to shareholders as $49.2 million; the income statement table in the same release shows $51.0 million, with $49.2 million being the net loss before that charge.)
• Cash burn is close to the reported loss. Operating activities used $94.8 million in the first half against a $99.5 million net loss; the gap is mostly $9.0 million of stock-based pay, which is a real cost to shareholders but not a cash outflow. Cash used was 7.6% higher than the $88.1 million used in the first half of 2025, even though the reported loss was smaller, because the company paid down accrued bills.
• The longer runway is mostly borrowed and raised money, not lower spending. In May, Annexon guided that $225.0 million would last "into the second half of 2027." In August it guided "into 2028," with the press release crediting cash plus "funds from the credit facility." The loan is not cheap: it carries interest at the greater of one-month SOFR (a benchmark short-term rate) plus 4.6%, or 7.6%, which means at least about $3.8 million a year on the first $50 million. It is interest-only until September 2029, and lenders can require minimum cash or minimum revenue levels if certain events occur. A further $100 million is available only if specified milestones and conditions are met, and the last $50 million needs the lenders' agreement.
• A rough check on the runway. At the first-half pace of about $47 million of operating cash a quarter, roughly $259 million (June cash plus the July loan) covers a little over five quarters, which reaches around the end of 2027. The 10-Q also says management expects R&D and G&A to increase as it prepares for approval and commercial launch. "Into 2028" therefore likely depends on some combination of further ATM sales, later loan tranches or spending choices the company has not detailed. This is our arithmetic, not company guidance.

The pipeline: what the spending is buying

• Tanruprubart for Guillain-Barré syndrome (GBS), a sudden immune attack on nerves that can cause paralysis. A one-time infusion; in the company's placebo-controlled Phase 3 trial, about 90% of treated patients improved by week 1. The European Medicines Agency has been reviewing the marketing application since January 2026. On August 6 the company reported that all of the first 10 U.S. and European patients in the open-label FORWARD study (everyone gets the drug, no placebo group) showed meaningful strength gains within four days. Because the Phase 3 trial was run outside the U.S. and Europe, FORWARD is meant to show the FDA that results carry over to Western patients. The company plans to discuss this data with the FDA and submit a U.S. approval application (a BLA) in Q4 2026.
• Vonaprument for geographic atrophy (GA), an advanced form of dry age-related macular degeneration that destroys central vision. The Phase 3 ARCHER II trial (659 patients, injections into the eye versus a sham procedure) measures the share of patients who lose 15 or more letters on an eye chart. All eligible patients have now had at least 12 months of treatment, with fewer than 10% dropping out. In August the company added a second primary endpoint at Month 24 alongside the existing Month 15 one, so the trial can succeed at either timepoint. The Month 15 readout stays on track for Q4 2026; the full trial, including Month 24, is expected to finish in Q3 2027. An open-label extension study was also launched.
• ANX1502, an oral pill blocking an early step of the same immune pathway, is in a small proof-of-concept study in cold agglutinin disease (a rare condition where the immune system destroys red blood cells). Dosing is complete; an update is expected in fall 2026.

Takeaway: Annexon's Q2 numbers are a holding pattern; what matters is that the company now has about $259 million (including a new loan) to reach two binary events in Q4 2026, the ARCHER II Month 15 readout and the tanruprubart BLA. The new Month 24 endpoint gives the eye trial a second chance to succeed, but it also means a Month 15 miss would not settle the question until late 2027, adding about nine months of spending before a final answer.

Outlook

Management's milestones for the next six months are the BLA submission for tanruprubart (Q4 2026), the ARCHER II Month 15 topline result (Q4 2026), an ANX1502 update (fall 2026) and a decision from European regulators on tanruprubart, for which no date has been given. Cash runway guidance is "into 2028."

Our read: spending is now shifting from trial costs toward getting ready to launch (commercial-scale manufacturing, corporate affairs consulting, a board appointment the company tied to preparing for commercialization), which is the right order only if one of the two late-stage drugs reaches the market. The Q3 report should show whether contract manufacturing and G&A keep climbing, and how much more stock was sold through the ATM program. The share count rose 35% in a year, so how the company funds itself matters almost as much to shareholders as the trial results. The things to watch, in order: whether the Month 15 ARCHER II data shows a clear benefit on vision loss; whether the FDA agrees the FORWARD data are enough to file the BLA on schedule; and whether milestone-based loan tranches become available, which would reduce the need to sell more shares.