BLTE — Q2 2026 Financial Report Analysis

Spending to launch a drug the FDA hasn't approved yet

Belite Bio's second-quarter 2026 net loss widened to $28.4 million from $16.3 million a year earlier. The company has no revenue. The loss grew because it is building a sales and operations team for tinlarebant, its oral pill for Stargardt disease type 1 (STGD1), ahead of an FDA decision due February 12, 2027. The FDA accepted the application with Priority Review, a shorter six-month review clock (announced August 11). With $780.0 million in cash and US Treasuries and liquid funds down only $18.6 million in the quarter, money is not the constraint. The open questions are regulatory and scientific.

At a glance

• $780.0M in cash and Treasuries at June 30 ($279.9M cash plus $500.1M bills and notes), down from $798.6M at March 31. That is about $18.6M used in the quarter, after counting interest earned.
• Non-GAAP selling, general and administrative (SG&A) costs rose to $10.9M from $1.3M a year ago, almost double Q1's $5.7M. Non-GAAP means excluding stock-based pay, which costs no cash. The jump is the commercial build-out showing up in costs before any sales exist.
• $6.6M of interest income offset about 19% of the quarter's $34.9M in operating expenses, up from $1.3M a year ago, thanks to the cash raised in late 2025.

The quarter in numbers

*The release gives balance-sheet figures only for June 30, 2026 and December 31, 2025. Each American Depositary Share (ADS), the form in which the stock trades on Nasdaq, represents one ordinary share, so the per-share loss is also the per-ADS loss.

For the first half, the net loss was $55.4 million versus $30.6 million, with R&D of $33.9 million (vs $20.4 million) and SG&A of $33.7 million (vs $12.7 million).

Where the money went

R&D (+$7.2M YoY). Management attributes the Q2 increase "primarily" to a royalty payment owed for reaching another milestone under the license agreement. Belite licenses the patents covering tinlarebant from Columbia University. Its FY2025 annual report (Form 20-F) says Belite owes Columbia up to $20 million for development and regulatory milestones, and expected to pay $6 million "in the near term" as development milestones were reached. The release doesn't give the Q2 payment's size. Non-GAAP R&D rose from $13.8M in Q1 to $17.2M in Q2, which suggests a payment of a few million dollars. For the half year, the company also cites higher spending on drug manufacturing, both the active ingredient (API) and the finished drug product, plus consultant fees. That is the inventory and supply chain a launch needs.

SG&A (+$10.1M YoY). Management attributes this to professional service fees and wages from "team expansion". The Q1 release said all key commercial leadership hires were complete. Excluding stock-based pay, SG&A went from $1.3M a year ago to $5.7M in Q1 and $10.9M in Q2. That is the clearest sign in this filing that Belite is already spending like a company preparing to sell a product.

Interest income. Other income of $6.6M, up from $1.3M, comes from interest on bank deposits and US Treasury bills and notes. Net cash from financing was $663.2 million in 2025 (per the 20-F), mostly from share offerings late in the year, so the balance earning interest is much larger than a year ago.

The pipeline: what the spending is for

• STGD1, United States: The NDA (New Drug Application, the formal request to sell the drug) rests on the Phase 3 DRAGON trial: 104 patients aged 12 to 20, tracked for 24 months. Tinlarebant slowed the growth of damaged retinal areas (atrophic lesions) by 35.7% versus placebo. At the American Society of Retina Specialists meeting, Belite also showed a secondary measure. Quantitative autofluorescence (qAF), a reading of how much toxic vitamin-A byproduct builds up in the retina, fell about 2% from baseline on the drug and rose about 20% on placebo by month 25. The FDA decision date is February 12, 2027. Belite puts the US STGD1 population at about 53,000 people, and no treatment is approved.
• STGD1, Japan: An NDA was filed with Japan's health ministry on September 8 under Sakigake, an expedited review pathway. Belite estimates about 9,500 patients in Japan. DRAGON II, a 73-patient Phase 2/3 trial that includes 15 Japanese patients, has finished enrolling.
• Geographic atrophy (GA): GA is the advanced form of dry age-related macular degeneration, a much larger market than STGD1. The Phase 3 PHOENIX trial has enrolled 530 patients, and the company "expects to conduct an interim analysis". No date is given.

Takeaway: The loss nearly doubled because Belite is paying for a launch before it has a product to sell. Non-GAAP SG&A rose about 8x year over year and nearly doubled from Q1. With $780 million in liquid funds and Q2 liquid funds down only $18.6 million, financing is not the near-term risk. The share price will move on the FDA's February 12, 2027 decision and the timing and outcome of the PHOENIX interim analysis.

What the headline numbers hide

• The cash used is smaller than the loss. Stock-based compensation is a real cost to shareholders but not a cash outlay. It was $6.8M in Q2 and $20.0M in the first half, about 36% of the half-year net loss. Liquid funds fell $18.6M in Q2 against a $28.4M GAAP loss. Note that the company's own non-GAAP figures are not cash-flow measures, as the release states, and the release includes no cash-flow statement.
• Liquid funds rose in the first half despite a $55.4M loss. Cash plus Treasuries went from $772.6M at December 31 to $780.0M at June 30, with the increase in Q1 ($798.6M at March 31). Shares outstanding rose by about 0.93 million (39.34M to 40.27M), and shareholders' equity rose $6.2M instead of falling. Add back the loss and stock-based pay, and roughly $41.5M of new equity came in during the half. The release doesn't say from where. The 20-F lists warrants and an at-the-market share-sales program (selling stock gradually through brokers at market prices) as past funding sources. Either way, part of the half-year's spending was covered by new shares, not existing cash.
• Dilution drives part of the per-share figure. The weighted share count rose 23% year over year, mainly from 2025's offerings. That is why the loss per share grew 40% while the total loss grew 74%.
• One-off items: The Columbia milestone payment inflated Q2 R&D, and the release gives no amount. Milestone payments come with regulatory progress, so more are likely around an approval decision. There were no one-offs in the prior-year quarter.
• No going-concern doubt. There is no going-concern language, meaning no warning about whether the company can keep operating. The FY2025 20-F says existing funds are sufficient for at least the next 12 months. At Q2's pace that $780M lasts more than a decade. Even if quarterly cash use tripled to about $56M after launch, it would cover about 3.5 years. That is unusual for a company still waiting on its first approval.

Did management deliver on what it said last quarter?

This is our first report on Belite Bio, so there is no earlier report of ours to check against. The company's own Q1 2026 release said the rolling NDA submission would be completed in the second quarter of 2026. It was: the FDA's acceptance with Priority Review was announced on August 11. The Q1 release also said key commercial leadership hires were done. The step-up in Q2 SG&A matches that.

Outlook

The company gives no financial guidance. The dated catalysts are the February 12, 2027 FDA decision for STGD1 and Japan's Sakigake review of the September 8 filing. The PHOENIX interim analysis in GA is planned but has no date. Expect operating costs to keep rising through year-end: Q2 non-GAAP operating expenses were $28.1M, up from $19.5M in Q1, as hiring and manufacturing ahead of launch continue. Expect more Columbia milestone payments as regulatory steps are reached.

Our read: the STGD1 decision is the near-term binary event. A first-ever approved treatment for a disease with no options, backed by a positive Phase 3 result, a Breakthrough Therapy designation and Priority Review, makes a delay more likely than an outright rejection. Still, an FDA decision is never guaranteed. The size of Belite's opportunity depends on GA, a far larger population. That makes the PHOENIX interim the bigger long-term swing factor. Until then, the burn rate matters less than the timing: $780 million buys Belite years of runway, enough not to need new financing for its launch.